Untitled

Politics in an Era of Divided Government
POLITICS AND POLICY IN AMERICAN

INSTITUTIONS
Steven A. Shull, Series Editor
Presidents as Candidates
Inside the White House
for the Presidential Campaign
Kathryn Dunn Tenpas
Reasonable Disagreement
Two U.S. Senators and
the Choices They Make
Karl A. Lamb
Legislative Learning
The 104th Republican Freshmen in the House
Timothy J. Barnett
Elections and Governance in the Second Clinton Administration
Edited Harvey L. Schantz
POLITICS IN AN ERA OF DIVIDED
GOVERNMENT
ELECTIONS AND GOVERNANCE IN THE SECOND
CLINTON ADMINISTRATION
Edited by
Harvey L. Schantz
Routledge
New York • London
Published in 2001 by
Routledge
29 West 35th Street
New York, NY 10001
Published in Great Britain by

Routledge
11 New Fetter Lane
London EC4P 4EE
Routledge is an imprint of the Taylor & Francis Group.
This edition published in the Taylor & Francis e-Library, 2002.
Copyright © 2001 by Routledge
All rights reserved. No part of this book may be reprinted or reproduced or uti-
lized in any form or by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying and recording, or in any information
storage or retrieval system, without written permission from the publishers.
Library of Congress Cataloging-in-Publication Data
Politics in an era of divided government : elections and governance in the

second Clinton administration / edited by Harvey L. Schantz
p. cm. — (Politics and policy in American institutions)
Included biliographical references and index.
ISBN: 0-8153-3583-0 (hardcover)
1. Presidents—United States—Election—1996. 2. United States. Congress—
Elections, 1996. 3. Elections—United States. 4. United States—Politics and
government—1993–2001. I. Schantz, Harvey L., 1951– II. Garland reference
library of social science. Politics and policy in American institutions
JK526 1996L
324.973’0929—dc21 2001023028
ISBN 0-203-90315-3 Master e-book ISBN

ISBN 0-203-90319-6 (Glassbook Format)
To Our Families
Contents
List of Tables ix
Acknowledgments xi
Abbreviations Used in Notes xiii
Series Editor Foreword
Steven A. Shull xv
Introduction
Harvey L. Schantz xvii
Chapter 1 Some Things Are Predictable: Nominating

Dole, Clinton, and Perot
Emmett H. Buell Jr. 1
Chapter 2 Congressional Nominations in 1996:

Procedures, Candidates, and Electoral Patterns
Harvey L. Schantz 41
Chapter 3 The Presidential Campaign and Vote in 1996:

Job Ratings of Presidents—and Success or
Failure at the Polls
Milton C. Cummings Jr. 63
Chapter 4 Strategic Partisan Decisions and Blunted

National Outcomes: The 1996 Senate Election
Campaign and Vote
Douglas B. Harris 85
vii
Chapter 5 Sideshows and Strategic Separations: The Impact
of Presidential Year Politics on Congressional
Elections
Garrison Nelson 105
Chapter 6 Clinton’s Second Transition: Historic

Aspirations Amidst Divided Government
Margaret Jane Wyszomirski 129
Chapter 7 The Irony of the 105th Congress and Its Legacy

Roger H. Davidson and Colton C. Campbell 155
Epilogue
List of Contributors 183

Index 185
List of Tables
1.1 1996 Invisible-Primary Finances of Declared

Republican Candidates 7
1.2 1994–1996 Republican Presidential Straw Polls 10
2.1 Nominating Methods for the U.S. House, 1996 42

2.2 Nominating Methods for the U.S. Senate, 1996 42
2.3 Candidate Pool for Congressional Seats, 1996 50
3.1 President Clinton’s Job Rating, 1993–1996 66

3.2 Voter Support for Clinton, Dole, and Perot:
The Gallup Poll’s Three-Way Trial Heats–Between
February 1995 and Election Eve 1996 70
3.3 Presidents Seeking Reelection; Their Job Ratings;
and the Outcome of the Presidential Election:
Presidential Approval Ratings in an Election Year
and the President’s Share of the November
Presidential Vote, 1956–1996 81
4.1 States with 1996 Senate Elections by Presidential

Performance in 1992 and 1996 89
5.1 Fifty-Plus House Seat Election Gains and House

Party Control in the Twentieth Century 107
ix
5.2 House and Gubernatorial Pre-Election
Presidential Cues, 1892–1996 110
5.3 Congresses With Divided Government By

Era, 1789–2001 113
6.1 Clinton Cabinets I and II 135
7.1 Congressional Workload Comparisons,

1995 and 1997 163
7.2 Committee and Subcommittee Seats:
105th Congress (1997–1998) 168
Acknowledgments
I must thank all of the authors for contributing to the intellectual breadth
of this project. This book would not be possible without their work and
specialized knowledge. In addition to his own chapter, Milton Cummings
provided wise counsel and a critique of each chapter.
At Routledge, I am in debt to social science editors Amy B. Shipper and
Maria C. Zamora for their support and suggestions. I also benefited great-
ly from the insights of Professor Steven A. Shull, editor of the Routledge
series on “Politics and Policy in American Institutions.” I would also like
to thank production editor Jeanne Shu and her team for helping to turn the
manuscript into a book.
xi
Abbreviations Used In Notes
AJPS American Journal of Political Science

AP Associated Press
APQ American Politics Quarterly
APSR American Political Science Review
CD Columbus (Ohio) Dispatch
CQWR Congressional Quarterly Weekly Report
CSM Christian Science Monitor
DMR Des Moines Register
JOP Journal of Politics
NJ National Journal
NYT New York Times
RC Roll Call
USA USA Today
WP Washington Post
WSJ Wall Street Journal
xiii
Series Editor Foreword
The Routledge series, “Politics and Policy in American Institutions” strives

to show the interaction of American political institutions within the con-
text of public policy-making. A public policy approach often by definition
is all-encompassing. Admittedly, my own interests focus on national poli-
cy-making, but the series will also include works on all levels of govern-
ment. Indeed, I do not want my own specialties to define the series. There-
fore, we seek solid scholarship incorporating a wide range of actors,
including those outside the usual definition of government actors. The pol-
icy concerns, too, are potentially quite broad, with special interests in the
policy process and such substantive issue areas as foreign and defense pol-
icy, economic and budget policy, health care, social welfare, racial politics,
and the environment. The series will publish a considerable range of works,
ranging from upper division texts to scholarly monographs, including both
hard and soft cover editions.
Some might argue that we do not need another edited book on the topic
of divided government. However, Politics in an Era of Divided Govern-
ment concentrates on elections and governance during the second Clinton
administration. In this valuable collection of original essays, editor Harvey
L. Schantz focuses heavily on the results and the aftermath of the 1996
elections of Congress and the president. The volume covers party nomina-
tions, the general election, and subsequent government organization, argu-
ing that these three processes are highly interrelated. The individual
authors describe events but also make provocative arguments about the
implications of elections for government structure and policy-making.
Perhaps not surprisingly, Politics in an Era of Divided Government finds
a blurred boundary between elections and governance. At the end of the
volume, Schantz sets the stage for the 2000 congressional and presidential
elections. He contends that “impeachment and electoral politics have over-
shadowed policy during the second Clinton administration.” As the first
xv
xvi Series Editor Foreword
edited volume in this series, Politics in an Era of Divided Government pro-

vides an assessment of the important connection between elections and
governance. Those linkages are crucial in understanding the relationship
between American political institutions and public policy, the primary
focus of this Routledge series.
Steven A. Shull
Introduction
HARVEY L. SCHANTZ
The Constitution of the United States directs that the American electorate
choose a president every four years; a House of Representatives every two
years; and one-third of the Senate every two years. In 1996, U.S. political
leaders and voters complied with this mandate, as the electorate chose a
president for the fifty-third time and also elected the members of the 105th
Congress. The Clinton presidency, along with the 106th Congress elected
in the 1998 midterm election, will complete 212 years of governance under
the Constitution of the United States.
The national election of 1996 and its aftermath occurred in an era of
divided government. In the eight presidential administrations since 1969,
there has been divided partisan control of the Congress and president for
26 of 32 years. In 1995, though, Bill Clinton became the first Democratic
president since Harry Truman in 1947-1948 to share control of the gov-
ernment with a Republican Congress. The 1996 and 1998 elections main-
tained this political lineup.
This book describes, explains, and reflects upon the 1996 presidential
and congressional elections, devoting equal coverage to three phases of the
political process: the major party nominations, the general election, and the
subsequent government organization. In so doing, this study links elections
and governance.
THE PARTY NOMINATIONS FOR THE PRESIDENT

AND CONGRESS
The first two chapters of this book address the party nominations for the
presidency and Congress. Although not called for in the Constitution, the
major party nomination process is the crucial first step for any candidate
seeking the presidency or a seat in the U.S. House or Senate. All presidents,
aside from George Washington, have won the presidential general election
as the nominee of a major political party. Throughout U.S. history, and
xvii
xviii Introduction
especially in the twentieth century, the number of third party and indepen-
dent candidates winning seats in Congress has been minimal.
To formally gain their presidential nomination, the Democratic and
Republican parties require a candidate to win a majority of the delegate
support at their national party conventions, both of which are held in the
summer of election years. The delegates to the national convention are
mostly selected between February and June of the election year in a local
caucus–state convention process or by presidential primaries in each of the
fifty states, the District of Columbia, and those territories allocated dele-
gates by the national party committees.
Emmett H. Buell Jr., in his chapter, “Some Things Are Predictable:
Nominating Dole, Clinton, and Perot,” provides a detailed description of
how the 1996 presidential nominating process unfolded, with special atten-
tion to the Republican nomination contest won by former U.S. Senator
Robert Dole. Buell’s account of the Republican nomination contest covers
all the major phases of the process but emphasizes the importance of the
invisible primary—the extended period before formal delegate selection
begins. During the invisible primary, the field of candidates is finalized.
Candidates lay the groundwork for their campaigns: raising funds, orga-
nizing their staff, and clarifying their message. The role of the media is
most important at this stage, since the public does not yet know all of the
individual candidates. Through polls and candidate coverage, the press
judges the viability of the various candidates.
The invisible primary period can last as long as five years. By contrast,
the visible part of the nomination process, in part because of the front-
loading of delegate selection caucuses and primaries, often proceeds quick-
ly and dramatically. In 1996, for example, delegate selection began with the
Louisiana caucuses on February 6. By March 19, 1996—only about six
weeks after the state caucuses and primaries began—Dole commanded a
majority of the delegates to the Republican National Convention and thus
could lay claim to the presidential nomination.
Among the Democrats, the pace was even quicker: President Bill
Clinton’s only challenger withdrew in April 1995. Clinton’s renomination
campaign was thus much easier than that of the last Democratic president,
Jimmy Carter, in 1980, or that of Republican president George Bush in
1992. And the Reform Party nomination, as Buell indicates, was a foregone
conclusion: “the party of Ross Perot” unsurprisingly nominated Ross Perot
as its presidential candidate. In all three cases, it was the work that took
place before the first primaries and caucuses that allowed the 1996 presi-
dential nomination process to proceed as predictably as it did.
Virtually all congressional nominations in each of the fifty states are
made by direct primary elections, a method of nomination that developed
and spread throughout most of the United States in the first two decades of
the twentieth century. However, there are differences in the primary elec-
Harvey L. Schantz xix
tion systems used by the states. These differences, as described by Harvey

L. Schantz, include state variations in voter eligibility, in the role of
party organizations in endorsing candidates, and in the margin needed
for victory.
In his chapter on congressional nominations, Schantz also looks at pri-
mary candidates in 1996. He finds that incumbent representatives and
senators did very well in primaries, and were rarely challenged by other
current office-holders. Current office-holders gravitated to open districts,
since they are viable candidates in the absence of a congressional incum-
bent. The weakest candidates in primaries were non-office-holders, who
generally only won nominations when their party was not expected to win
in November. A few non-office-holders did, however, win a congressional
seat in 1996; included in this group were celebrity candidates, who had
earned high name recognition outside of politics.
A third focus of the chapter on congressional nominations is electoral
patterns in the primaries. Schantz finds that in 1996, margins of victory
were almost always substantial when incumbents sought renomination. In
the absence of an incumbent, however, there was frequently a close contest.
Voter turnout in congressional primaries was very low, almost always less
than half of the turnout in the ensuing general election.
THE GENERAL ELECTIONS FOR THE PRESIDENT

AND CONGRESS
The middle three chapters of this book focus on the general elections for
the national executive and legislature. Presidential elections are indirect:
voters at the polls do not vote directly for their preferred candidate but,
rather, pull the lever for a party’s slate of electors. These electors are
pledged—but not legally bound—to support the candidate preferred by the
voters in their state. Every state is allocated electoral votes equal to its rep-
resentation in Congress (the House and Senate). To be elected president, a
candidate must receive a majority of electoral votes.
Milton C. Cummings Jr., in his chapter, “The Presidential Campaign and
Vote in 1996: Job Ratings of Presidents—and Success or Failure at the
Polls,” provides a succinct description of the political setting and campaign
leading up to the presidential election. Cummings devotes much attention
to the pattern of the popular vote and the electoral vote verdict. An impor-
tant social factor in the voting pattern in 1996 was once again the gender
gap, whereby women were more likely than men to support Clinton. Both
major parties once again shared all the electoral votes, as Clinton won 379,
over 100 votes more than the necessary 270. Perhaps the most important
aspect of the 1996 electoral verdict was that it continued divided govern-
ment in Washington, with a Democratic president and a Republican
Congress.
xx Introduction
Cummings views the 1996 election as a referendum on the incumbent,

and thus to explain the outcome of the election, Cummings consults popu-
lar approval of presidential job performance. According to Cummings,
presidential approval ratings are a reliable predictor of presidential election
outcomes. Therefore, Clinton’s reelection is predicted by his 54 percent job
approval rating. Finally, using an important typology of presidential elec-
tions, Cummings classifies the 1996 election as a “reaffirmation of support
by a vote of confidence” in the Clinton administration.
The Constitution of the United States initially provided for indirect elec-
tion of U.S. senators. But since 1914, due to the seventeenth amendment,
senators have been directly elected. For electoral purposes, senate seats are
divided into three classes, and every two years one of these classes is up for
reelection. Thus, the senate seats up for election in 1996 were last filled in
1990.
Douglas B. Harris, in his chapter, “Strategic Partisan Decisions and
Blunted National Outcomes: The 1996 Senate Election Campaign and
Vote,” describes the backdrop for the senate elections and the outcome on
election day. In the 1996 senate elections, the Republicans, who were
defending 19 seats—four more than the Democrats were trying to retain—
were most successful in senate elections, extending their majority from 53
to 55.
Yet according to Harris, these numbers do not tell the whole story, for
two reasons. First, turnover usually occurs when an incumbent retires. In
this regard, the Republicans had the upper hand in 1996, since they had to
defend only five open seats, while eight Democratic senators were retiring.
Moreover, despite Clinton’s 31-state victory, the senate races were evenly
divided between states won by Dole and states won by Clinton. Thus, in
Harris’s view, rather than accomplishing a surprising feat by extending
their majority in 1996, the Republicans did less well than might have been
expected before the government shutdowns of November 1995–January
1996 turned public opinion against them.
Altogether, four senate seats changed parties in 1996. In the Plains
states, Republicans lost in South Dakota but gained in Nebraska. In the
South, Republicans gained two seats, Alabama and Arkansas. This
turnover continued a trend of Republican gains in the South, and, as Harris
points out, these gains are an integral part of the Republican senate major-
ity.
The Republican senate majority, along with the reelection of President
Clinton, continued one of the major trends in U.S. government in the sec-
ond half of the twentieth century: the increased frequency of divided party
control of the president and one or both chambers of Congress. In his chap-
ter, “Sideshows and Strategic Separations: The Impact of Presidential Year
Politics on Congressional Elections,” Garrison Nelson examines congres-
sional elections, especially those for the U.S. House, from the vantage point
Harvey L. Schantz xxi
of divided government. As he notes, most explanations for the rise of divid-

ed government point to changes in electoral behavior. But Nelson persua-
sively contends that the increased frequency of divided government is,
rather, the result of strategic decisions made by recent presidents, especial-
ly Richard M. Nixon and Bill Clinton, and recent Speakers of the House of
Representatives, especially Thomas P. “Tip” O’Neill and Newt Gingrich.
Most of the recent academic discussion about the results of divided
party government has centered upon its impact on the amount, quality, and
responsibility for legislation. But Nelson suggests that divided government
has other effects: most particularly, under divided government, presidents
usually enjoy greater public approval, benefiting from the public’s negative
perception of Congress.
In addition, divided party government allows political leaders in both
branches to better realize their electoral goals. Leaders are able to run
against each other and not have to shoulder the responsibility for the per-
formance of the government. As divided control in Washington, D.C. has
become the typical political lineup, according to Nelson, politicians have
more frequently followed the lead of President Harry Truman in 1948, and
have used divided government as a reelection tool. In 1996, the chief ben-
eficiaries of this strategy were Speaker Newt Gingrich and President
Clinton; both seem to have benefited from a learning curve in this regard.
ORGANIZING THE NEW ADMINISTRATION AND CONGRESS

As soon as the elections are over, political leaders and the media turn their
attention to the tasks of organizing the new Congress and presidency. The
headline in the New York Times on Thursday, November 7, 1996, just two
days after election day was “Clinton Preparing for 2D Term With Shuffle
of Top Officials.” In Congress, also, attention quickly shifts from the elec-
tion to organization.
The organizing for a new presidential administration takes place during
the transition, which may begin even before election day and stretches past
the inauguration. In this crucial period, the new or reelected president
nominates and appoints cabinet-level officials and the White House staff,
and also should develop a policy agenda for the new term of office. The
importance of presidential transitions was formally recognized by the
Congress in the Presidential Transition Act of 1963, which provides fund-
ing for transitions involving a new president, to both the incoming and out-
going administrations.
The presidential transition of 1996–1997 did not bring major change
because it was an inter-term transition, involving the same president.
Nevertheless, as Margaret Jane Wyszomirski shows in her chapter,
“Clinton’s Second Transition: Historic Aspirations Amidst Divided
Government,” crucial decisions of personnel and policy were made during
the presidential transition period. These personnel decisions involved
xxii Introduction
replacing seven of the fourteen Department Secretaries, as well as filling

key positions in the White House staff, including selection of a new Chief
of Staff. Special care was taken by the president to assemble effective staff
teams in the areas of national security and economic policy.
Of special interest to Wyszomirski is the process by which Clinton set
his second term policy agenda. Wyszomirski shows that Clinton’s agenda
emerged from events that transpired during the election campaign, partic-
ularly the Democratic party platform, as well as his 1997 inaugural and
State of the Union addresses. During the first 100 days of his new term,
Clinton’s policy agenda, aided by growing government revenues stimulat-
ed by economic growth, set the stage for bipartisan cooperation in con-
gressional policy-making.
Discussions of presidential administrations often suggest that the first
100 days or six months are the most propitious time for presidential leg-
islative success with Congress. But Wyszomirski argues that the dynamics
of policy-making may be different in divided government. In these circum-
stances, she suggests, the first six months of a new administration may have
to be dedicated to working out a bipartisan agenda with the Congress.
Thus, rather than an initial outburst of policy followed by a decline, law-
making in divided government may take longer and most likely will be
achieved in a more piecemeal fashion.
The 105th Congress, elected in the November 1996 election, convened
on January 3, 1997. But much of the organizing for the new Congress took
place in the months between election day and the new year. In this period,
the four congressional parties held their conferences—meetings of all the
legislative members of the party—in which they decided basic organiza-
tional matters such as the selection of party leaders and committee chairs
and assignments. These conference decisions were then ratified by the full
chambers, along a party line vote, when the Congress convened.
At the outset of the 105th Congress, in both the House and Senate, there
was stability in the standing committee system and the party leadership.
The standing committees of Congress are permanent legislative committees
that prepare legislation for the full chamber. In the 105th Congress, the
House retained the same 19 and the Senate the same 16 standing commit-
tees employed in the previous Congress. Similarly, all four congressional
parties retained the same leaders: in the House, Republican Speaker Newt
Gingrich—despite ethics charges which threatened his position—as well as
Republican Majority Leader Richard Armey and Democratic Minority
Leader Richard A. Gephardt; in the Senate, Republican Majority Leader
Trent Lott and Democratic Minority Leader Tom Daschle.
However, every Congress brings some change to Washington, D.C. In
their chapter, “The Irony of the 105th Congress and Its Legacy,” Roger H.
Davidson and Colton C. Campbell examine the new Congress and also
contrast it with its predecessor. The 104th Congress, elected in 1994,
Harvey L. Schantz xxiii
brought revolutionary change to Congress, as the Republicans won control

of the House for the first time since Dwight Eisenhower was swept into
office in 1952. The Republicans also captured the Senate, after losing con-
trol of this chamber in the second Reagan midterm, 1986. Compared to its
predecessor, the 105th Congress brought a minimal amount of change to
Washington, as the Republicans retained control of both chambers.
According to Davidson and Campbell, the 1996 elections set the stage
for the Congress which was to follow. A number of trends in House elec-
tions and campaigns made it more difficult for the Republican party to leg-
islate on its own. Chief among these was the net loss of Republican seats,
leaving the GOP with the narrowest House majority since Republican con-
trol of the first Eisenhower-era Congress. Between the elections of 1994
and 1996, the Republican majority declined from 230 to 227, but the loss
of seats in 1996 was actually nine, for in 1995 Republicans had gained five
Democratic defectors and won an additional seat in a 1995 special election.
The policy environment of the 105th Congress differed greatly from the
previous Congress. After the 1994 election congressional Republicans,
especially in the House, aggressively sought to pass the items contained in
the “Contract with America.” But as Davidson and Campbell point out,
the Republican congressional majority in 1997 had not been elected on the
basis of an agreed-upon policy agenda, and thus Republicans did not move
quickly to pass mandated legislation. Rather, the first months of the new
Congress were spent developing party priorities and formulating a bal-
anced budget agreement with Democratic President Clinton.
In Congress, there is frequently a tension between the power of the party
leadership and the committees. In the 104th Congress, the House leader-
ship was ascendant, dominating the policy process. But in the 105th
Congress, according to Davidson and Campbell, committee leaders began
to exert themselves. Moderate House Republicans also gained in power, as
the Republican leadership needed all available votes in order to produce
winning coalitions.
By comparison, the Senate was “an island of relative tranquillity.”
Although Republicans increased their party advantage to 55-45, the rules
of the Senate, which carefully protect the rights of the minority and indi-
vidual members, leave it less open to party government.
By early 1998, Congress devoted most of its attention to the Clinton
impeachment. These proceedings, according to Davidson and Campbell,
were aided by two basic principles of congressional politics—party and
committee. That is to say, the autonomy of the extremely partisan House
Committee on the Judiciary provided a congenial setting for the beginning
of the effort to impeach President Clinton. Although the impeachment
effort ultimately failed, these proceedings continued the partisan era on
Capitol Hill. Ironically, the 105th Congress turned out to be even more
partisan that the revolutionary 104th Congress.
xxiv Introduction
Although this book is organized around three distinct steps of the polit-
ical process—nominations, general elections, and transitions—the chapters
demonstrate that these steps are highly interconnected. Most directly, the
nominations set the stage for the general elections, and the elections are the
prelude to organizing the new Congress and administration. The decisions
of political leaders will in turn influence the next nomination and electoral
cycle. And of course, there are reciprocal influences between all three
stages. More generally, though, this book once again reminds us that there
is not a clear boundary between elections and governance.
CHAPTER 1
Some Things Are Predictable

Nominating Dole, Clinton, and Perot
EMMETT H. BUELL JR.
This is an account of how the Republican, Democratic, and Reform parties

chose their respective presidential nominees in 1996. As memories of the
last nominating cycle fade, we tend to forget how slim the odds of a sec-
ond Clinton term once seemed or how much H. Ross Perot frightened both
major parties. Ever adaptable, President Clinton reinvented himself after
disaster struck his party in the 1994 midterm elections and he eventually
swept to renomination unopposed. No event astonished election analysts
less than Perot claiming the nomination of a party of his own making.
Most of what follows is about the Republicans who experienced the
only seriously contested nomination of the three. I argue that the Republi-
can choice was highly likely if not wholly predictable owing to what the
nominating process has become. Presumably a front-runner can still self-
destruct in such a system, but, if 1996 is any guide, process-related advan-
tages will cancel out short-term factors like candidate style or weak mes-
sage. Much depends on who else is running, of course, but that, too, is
increasingly determined by the process. In 1996, unprecedented “front-
loading” of the primary calendar, combined with the associated lack of
viable rivals, extraordinary media exposure, and related leads in the polls
and fund-raising set Dole up as the inevitable nominee.
NOMINATING STAGES AND FRONT-LOADING

Put simply, the increased resort of state parties to front-loading has
changed the dynamics of presidential nominating politics in both of the
major parties. The proliferation of February and March primaries now
impels aspirants to start campaigning years in advance of the first votes to
select convention delegates. Long familiar to pundits as the “invisible pri-
mary,” this lengthy run-up to the first caucuses and primaries truly has
become the most critical phase of the whole process.1 This is the time when
presidential hopefuls must raise huge sums, recruit activist supporters,
1
2 Some Things Are Predictable
build campaign organizations in key primary states, articulate a message,

and impress the news media. More often than not, aspirants drop out after
discovering that they lack the will, the resources, the stamina, or some
other essential reason to continue campaigning.
Owing to the degree of front-loading now practiced by state parties, a
seemingly interminable run-up to the first act of delegate-selection is fol-
lowed by an incredible rush to judgment in a clutch of primaries. Rather
than sort these primaries into multiple phases on the assumption that each
importantly bears on the choice of the nominee,2 it makes more sense
nowadays to distinguish between “de facto” and “ex post facto” primaries.
De facto contests occur only weeks into the presidential election year and
decide the nomination in six weeks’ time or less. Voters go to the polls in
ex post facto primary states after one candidate has captured enough dele-
gates to claim the nomination.
Front-loading also has wreaked havoc with the traditional distinction
between the nominating and electing stages of presidential selection. Unen-
cumbered by a primary challenge to Clinton in 1996, the Democrats began
their general election campaign more than a year before the San Diego con-
vention nominated Dole. The sooner a candidate captures the de facto
nomination, the earlier the general election campaign begins. Thanks in
great measure to reforms of delegate-selection rules after the 1968 Demo-
cratic convention (which indirectly but importantly affected the GOP as
well), 1972 marks the beginning of a dramatic rise in the total number of
Republican primaries.3 Republican front-loading also picked up, so that, by
1988, the GOP held more presidential primaries before April 1 than after-
wards.4
Today most of the convention delegates needed for formal nomination
are selected by the end of March. The early yield of delegates increased dra-
matically when big states changed their primary dates from May and April
to March. In 1980, for example, only 40 percent of the Republican dele-
gates were chosen before April 1. That figure soared to 61 percent in 1988
before subsiding to 51 percent in 1992.5 Front-loading reached a new high
in 1996, when New York, Ohio, and California switched to March. In the
end, 29 primary states (including eight of the twelve most populous) select-
ed roughly 64 percent of all 1,990 Republican delegates before April 1. By
March 26, Dole had accumulated 1,126 delegates, or 127 more than the
simple majority needed for nomination.6
A LONG INVISIBLE PRIMARY FOR THE GOP

It is impossible to date the exact start of an invisible primary, but as good
a marker as any is when the national press starts paying attention to likely
candidates in the next round of presidential nominating contests. Major
newspapers began speculating on the GOP’s 1996 lineup more than a year
before the 1992 New Hampshire primary.7 By December 1991, the media
list of presidential aspirants included General Colin Powell, Secretary of

Defense Dick Cheney, Vice President Dan Quayle, Secretary of State James
Baker, Secretary of Housing and Urban Development Jack Kemp, pundit
Pat Buchanan (who only twelve days earlier had entered the 1992 nomi-
nating race), Governor Pete Wilson of California, Senator Phil Gramm of
Texas, and former Tennessee governor Lamar Alexander.8 Ostensibly called
to renominate Bush, the Houston convention also showcased Quayle,
Buchanan, Baker, Cheney, Gramm, Kemp, Wilson, and still other 1996
possibilities.9
Although generally omitted from the earliest lists, Dole’s stock rose
when he became the party’s principal spokeman after the 1992 presidential
election.10 In February 1993, Dole signaled his interest in a third try for his
party’s nomination (he had run unsuccessfully in 1980 and 1988) by accel-
erating the fund-raising of Campaign America, his own political action
committee. He also set up The Better America Foundation, yet another
vehicle for raising pre-presidential money beyond the reach of campaign
finance limits.11 Dole boosted his standing among Republicans by leading
the fight against the Clinton stimulus package in the 103d Congress.
August 1993 found him “vacationing” in New Hampshire.12 By year’s end
he had campaigned for Republican candidates in 39 states.13 Campaign
America also dispatched Dole organizers to Iowa and New Hampshire by
early 1994. Dole began mapping out a general strategy shortly after the
1994 elections.14 On January 12, 1995, he registered as a candidate with
the Federal Election Commission (FEC).15
QUAYLE, POWELL, AND OTHERS OPT OR DROP OUT

Consistent with Arthur Hadley’s disclosure that as many or more aspirants
opt or drop out of the invisible primary as stay on to compete for dele-
gates,16 House speaker Newt Gingrich and five former members of the Bush
administration—Baker, Cheney, Quayle, Kemp, and labor secretary Lynn
Martin—assessed their 1996 chances before opting out. Quayle looked
almost certain to run after undertaking a national book-promotion tour in
June 1994. Slow to organize in key states, beset by health problems, and
evidently discouraged by initial efforts at fund-raising, the former vice pres-
ident departed the race on February 9, 1995.17 Several governors—Carroll
Campbell of South Carolina, Tommy Thompson of Wisconsin, and
William Weld of Massachusetts—also seriously considered running before
deciding against it. Two others, Governor Pete Wilson of California and
Senator Arlen Specter, actually declared candidacy but withdrew in the last
quarter of 1995 after failing to rise in the polls or raise the sums necessary
to wage serious campaigns.
All of these early departures aided Dole enormously, but none helped
more than Colin Powell’s. A few conservatives in the GOP, most notably
Kemp and William Bennett, former education secretary and drug czar,
openly welcomed a Powell candidacy. Ralph Reed of the Christian Coali-

tion, Speaker Gingrich, Republican National Committee (RNC) Chairman
Haley Barbour, and talk-show host Rush Limbaugh carefully avoided crit-
icizing him. Dole extolled Powell’s accomplishments but warned that the
party would not remake itself to suit Powell; Gramm and Buchanan
charged that he was a Democrat in all but name.18
Powell’s most outspoken opposition issued from conservative pundits
and leaders of groups constituting the Republican right. To support some-
one so bereft of basic beliefs, George Will wrote, was to embrace a “watery
Caesarism—trust the leader, well, because he is the leader.” Paul Weyrich
of the Free Congress Foundation declared that, if Powell got the nomina-
tion, “it would be as if Ronald Reagan never lived and Nelson Rockefeller
never died.” Gary Bauer of the Family Research Council blasted Powell as
“Bill Clinton with ribbons,” an opponent of smaller government, and an
advocate of failed social programs. David Keene of the American Conser-
vative Union castigated Kemp and other Powell supporters in the GOP
willing to “sacrifice the work of decades on the altar of celebrity.” Oliver
North declaimed that “this guy is not a Republican.” Carol Long of the
National Right to Life Committee vowed to oppose a Powell candidacy.19
How much this reaction influenced Powell’s decision not to run may
never be known. How much he would have hurt Dole in the primaries is
also a matter of considerable speculation. Powell looked strong in some
polls but weak in others. Gallup put him even with Dole in September
1995, and one percentage point behind in early November.20 One October
poll showed him besting Dole among likely Republican primary voters.
The same poll found Powell winning and Dole losing in respective trial
heats against Clinton.21 Still another sampling commissioned by Clinton
consultant Dick Morris showed Powell losing to Dole in a two-man race.22
Powell cut Dole’s support in half in New Hampshire, but consistently
trailed in polls of likely Iowa caucus participants.23 Powell in any event
opted out of the race on November 8, admitting that the political life
required a calling that he did not yet hear.24
On balance it appears that Powell mania helped as much as it hurt Dole.
On the one hand it confirmed that Dole’s support was soft. On the other
hand it deprived the other candidates of press attention and activist sup-
port. Contributors held back from giving to Alexander, Indiana Senator
Richard G. Lugar, and Specter as long as a Powell candidacy looked like-
ly.25
THE FINAL FIELD: INSIDERS AND OUTSIDERS

By September 1995, all eight of the challengers who would formally con-
test Dole’s nomination at least through the Iowa caucuses had thrown their
hats into the ring. In addition to Gramm, Alexander, Buchanan, and Lugar,
they included Congressman Robert Dornan of California, former ambas-
sador to the Economic and Social Council of the United Nations Alan
Keyes, and two multimillionaires with little or no experience whatever in
public office—Steve Forbes and Morry Taylor. Dole and Lugar won respect
in most party circles as accomplished legislators.
Recent “outsiders” typically have not held public office at the point of
starting their presidential bids, although some had previously served in
government or enjoyed considerable access to high public officials.26 Posing
as outsiders naturally obliged the 1996 outsiders to paint Dole and Gramm
as charter members of the party establishment. Lugar escaped most of these
attacks because of his low standing in the polls.
Some outsiders looked more convincing than others. Heir to a vast for-
tune, Forbes arguably belonged to the GOP “establishment” despite lack
of experience as an office-holder. Alexander had worked for Senator
Howard Baker and Richard Nixon before serving as governor of Ten-
nessee, president of the University of Tennessee, and Bush’s education sec-
retary. His approach was to attack incumbents in offices he had not yet
held. Buchanan had spent nearly all of his life inside the Beltway, achieving
notoriety as a conservative activist, Nixon aide, syndicated columnist, and
television personality. Buchanan came across as a more authentic outsider
than Alexander or Forbes, however, because of his 1992 bid to deny Pres-
ident Bush renomination and his passionate criticism of orthodox Repub-
lican views on trade. Although a House member since 1977(except for
1983-84), Dornan nonetheless was widely regarded as no less a fringe can-
didate than Keyes and Taylor.27
The Republican finalists often differed strongly on policy matters, even
though each billed himself as a conservative.28 Though much mentioned
during the campaign, the outsider-insider distinction had remarkably little
connection with candidate stands on the issues. True, most outsiders
favored term limits while all of the insiders opposed them. Still, no outsider
except Forbes offered any tax plan as radical as Lugar’s proposal to sup-
plant income taxes with a national sales tax. Alexander’s call for massive
transfers of federal power to state and local governments elicited no more
support from outsiders than insiders. Alexander, Buchanan, and Taylor
joined Dole and Gramm in attacking the Forbes flat tax. Alexander and
Forbes supported free trade along with Dole, Gramm, and Lugar. Every
candidate wanted to dismantle the Department of Education. With the pos-
sible exceptions of Dornan and Keyes, every candidate favored Senate pas-
sage of a balanced budget resolution. All favored voluntary prayer in pub-
lic schools and rejected race quotas. True, Lugar did not go much beyond
quotas when speaking out on affirmative action, but neither did Forbes.
Meanwhile, insider Gramm joined outsider Buchanan in calling for an end
to affirmative action. (Dole sometimes made this argument too.) The out-
sider-insider distinction also had little if anything to do with which candi-
dates pledged to make life tougher on imprisoned felons, oppose gun con-
trol, or resist gay rights. Outsiders Forbes and Taylor wrote off the Chris-
tian right while outsiders Buchanan and Keyes identified with it.
THE MONEY CHASE AND EARLY SPENDING

No activity of the invisible primary affects the rest of the nominating
process more importantly than fund-raising, and this is especially true
when front-loading reaches 1996 proportions. Conventional wisdom
maintained that the 1996 aspirants had to raise at least $20 million by the
end of 1995 to become viable.29 Any candidate in the contemporary system
who counts on taxpayer money must abide by Federal Election Campaign
Act (FECA) contribution limits, including a cap of $1,000 per individual,
up to $250 of which is matched when the contribution is made by check.
These caps have not changed since the FECA took effect in 1976, forcing
candidates to raise huge sums from many individuals in small amounts
capped by limits oblivious to inflation.
Every Republican candidate except Forbes and Taylor filed for matching
funds and agreed to abide by FECA caps on contributions, total spending,
and state spending. By not filing, Forbes and Taylor freed themselves to
spend whatever amounts they wished. For all other Republican candidates,
state spending limits varied by population rather than by importance in the
process. Theoretically, for example, no federally financed candidate could
spend more than $618,200 on the first primary in New Hampshire, exact-
ly the same limit set for Alaska and a dozen other small states.30
From 1976 (the first race regulated by the FECA) through 1992, the gen-
eral rule for contested Democratic nominations was that the candidate who
raised the most money during the invisible primary went on to claim the
nomination. Republican history suggests a slightly different conclusion:
The race cannot be won without amassing a great deal of money, but rais-
ing more than any rival is neither necessary nor sufficient to win the nom-
ination.31
Dole largely owed his nomination to highly successful fund-raising dur-
ing the invisible primary. Without $25 million in seed money, his campaign
could not have gone all out to compete in several straw votes, build state
organizations, commission polls, conduct focus groups, and counter
Forbe’s massive advertising assault in Iowa and other early-bird states.
Table 1.1 shows that Dole topped the list in individual contributions
reported through the end of January 1996, surpassing Gramm by nearly
$10 million, Alexander by almost $15 million, and all others by $17 mil-
lion or more.32 He also led in matching funds, outpacing Gramm, his clos-
est competitor, by more than $1.5 million.
Table 1.1: 1996 Invisible-Primary Finances of Declared Republican Candidates
Candidate Individual Matching Funds Adjusted Adjusted Spending Cash on Hand Campaign Debt
Contributions Receipts Spending Subject to
Limit
Alexander $10,294,257 $1,933,475 $12,535,861 $12,050,711 $9,245,489 $425,806 $94,070
Buchanan $7,540,444 $2,383,252 $10,731,288 $10,630,098 $10,630,098 $101,192 $1,398,234
Dole $25,113,989 $5,552,297 $31,988,345 $27,152,687 $21,660,153 $4,835,659 $5,447,981
Dornan $245,444 $0 $288,444 $285,951 $0 $2,491 $173,640
Forbes $2,074,564 $0 $25,440,564 $25,136,300 $0 $304,264 $23,650,979
Gramm $15,648,123 $3,987,412 $25,715,538 $24,119,921 $16,342,012 $1,515,050 $1,781,363
Keyes $2,073,531 $0 $2,083,545 $1,841,638 $0 $204,428 $511,747
Lugar $4,503,489 $1,363,342 $7,367,800 $6,805,507 $5,672,510 $539,666 $1,381,905
Specter $2,283,651 $592,651 $3,203,441 $3,188,768 $3,188,769 $29,121 $311,366
Taylor $36,236 $0 $5,354,351 $5,340,807 $0 $3,244 $5,442,408
Wilson $5,123,841 $953,654 $6,362,258 $6,051,190 $2,448,884 $290,369 $1,503,680
Totals $74,937,569 $16,766,083 $131,071,435 $122,603,578 $69,187,915 $8,251,290 $41,697,373
Source: Data kindly supplied by Robert Biersack of the Federal Election Commission.
Note: Data from inception of each campaign through January 31, 1996.
For all of its undeniable importance, if early money determined every-

thing, a close correlation should always show up between how much
money the candidate raised or spent and how well that candidate did. Polit-
ical reality belies such simplicity.33 Gramm was the first casualty of the
Louisiana and Iowa caucuses, despite having raised more from individual
contributors, received more in matching funds, and reported more cash-on-
hand than any other candidate abiding by the FECA except Dole. The com-
paratively impecunious Buchanan actively contested Dole’s nomination
longer than anyone else, including the biggest spender of them all.
Forbes withdrew before Buchanan, but not before setting new spending
records for the invisible primary. A candidate only since September 22,
Forbes nonetheless laid out $14 million in the final quarter of 1995, most
of which went into radio and television advertising in Iowa, New Hamp-
shire, and Arizona. By mid-December, he had spent more on ads in the Des
Moines media market than had all of the Republican and Democratic can-
didates at the same point in the 1988 nominating cycle.34
NATIONAL POLL STANDINGS

Another barometer of likely candidate success is standing in the national
polls. Writing about the 1976 invisible primary, Arthur Hadley maintained
that the last Gallup poll released before the first voting in New Hampshire
always predicted the Democratic and Republican nominees.35 Although
wrong about the Democrats in 1976, Hadley’s law has stood up remark-
ably well for every contested nomination since. Substituting Iowa for New
Hampshire, every Democratic and Republican front-runner in the final
Gallup poll before Iowa has gone on to win his party’s nomination. Still, as
I have argued elsewhere, it is unwise to regard the last poll of the invisible
primary as representative of all polls preceding it. Democratic front-run-
ners typically emerge quite late in the invisible primary, while eventual
Republican nominees typically lead in the polls for most if not all of the
invisible primary.36 Dole’s standing in the national polls adhered closely to
the Republican pattern. The unannounced Powell, as previously noted,
posed the only threat to Dole’s otherwise impressive lead in every Gallup
poll taken between March 1994 and January 1996. In the polls with only
announced candidates, Dole always received between 45 and 55 percent of
all the support.37
TELEVISION NEWS EXPOSURE

Coverage in the news media constitutes yet another measure of candidate
success during the invisible primary. Publicity closely follows the polls,
with most of it going to front-runners, runners-up, and potential upstarts,
while candidates written off as hopeless receive remarkably little notice.38
Seldom, if ever, has an early front-runner not already in the White House
enjoyed more of a media advantage than Dole, who exploited his position
as Senate Republican leader to the hilt. In 1993 alone, he appeared on 31
weekend television news and talk shows, including seven interviews on
CNN’s Newsmaker/Late Edition and six on Larry King Live; in 1994 Dole
marked his 50th appearance on NBC’s Meet the Press.39
At a point when poll standings signified little more than name recogni-
tion, no other candidate came close to matching Dole in television expo-
sure. The number of stories naming him on all three networks more than
doubled the total for all other aspirants during the early phase of the cam-
paign.40 He also predominated during the latter part, albeit less impressive-
ly. Altogether, he showed up in a total of 765 stories over a period of 109
weeks, averaging seven stories per week. Gramm, his closest rival in the
polls, appeared in only 227 stories, or only two per week.41
Dole’s coverage looked all the more impressive for its variety. During the
latter phase of the invisible primary, for example, the networks turned to
him more often than all of his rivals combined for pronouncements on the
103d Congress and the 1994 midterm elections. He dominated in coverage
of the budget battle and associated government shutdowns (100 stories or
ten times the coverage of all other Republican candidates on this issue), the
balanced budget amendment (15 stories to Gramm’s 2), welfare reform (27
Dole stories compared to 14 mentioning Gramm, Buchanan, or Wilson),
and health care (59 stories compared to 23 for Gramm and Specter). All
three networks aired a total of 61 nightly news stories mentioning his
Bosnia views, compared to 25 paying similar attention to his rivals. Even
in the case of the U.S. takeover of Haiti, which Powell helped broker, Dole
got the most coverage.42
STRAW POLLS: “IS ANYBODY HERE FROM IOWA?”

Though yielding no delegates, the invisible primary indicators mentioned
thus far do forecast who eventually wins the nomination. The same cannot
be said of straw polls. Of the 18 Republican and Democratic straw votes
that took place during the invisible primaries of 1976–92, in which 1,000
voters or more participated, eventual nominees won only nine.43 The
Republicans staged a record 11 of these contests during the 1996 invisible
primary. Dole won only three, including the crucial Presidency III show-
down in Florida. Gramm won five and tied Dole in the much publicized
Ames, Iowa poll. Buchanan prevailed in two others. Table 1.2 summarizes
the outcome of each straw vote.
Table 1.2: 1994–1996 Republican Presidential Straw Polls
Date Event Winners and Losers Particulars
June 24, 1994 Iowa GOP convention Dole 27%, Alexander Voting limited to hold-
in Des Moines 15%, Gramm 15% of ers of $25 tickets.
1,349 votes cast
January 7, 1995 Louisiana GOP conven- Gramm 72%, Gramm launched a

tion Buchanan 12%, and major effort to contact
Alexander 5% of every delgate; Dole
1,247 votes cast did not compete.
March 3, 1995 South Carolina GOP Gramm 35%, Voting limited to

“Silver Elephant” Alexander 25%, ticket-holders; blocs of
fundraiser Dole 21% of 1,200 tickets purchased by
votes cast campaigns and distrib-
uted to supporters.
April 8, 1995 Oklahoma GOP state Gramm 50%, Keyes Each participant paid
convention 22%, Dole 15% of $10 “delegate fee” to
1,439 votes cast vote.
June 17, 1995 Virginia Republicans Buchanan 59%, Participants paid $25
Keyes 11%, Gramm each to vote at a
8%, Lugar 8%, Dole fundraiser in Tyson’s
7% in vote of 1,083 Corner.
participants
August 19, 1995 Iowa GOP convention Dole and Gramm Each participant paid
in Ames tied with 24% each $25 to vote; news
of 10,598 votes cast; reports of extensive
Buchanan got 18%, voting by activists
Alexander 11%, and from other states;
Wilson only 1% widely interpreted as
Dole setback.
September 17, 1995 National Federation of Gramm 35%, New York Times
Republican Women con- Alexander 17%, described delegates as
vention meeting in Albu- Dole 17% (1 vote a “fairly good cross
querque less than Alexander), section of the Republi-
Wilson 15%, Keyes can party;” interpreted
8%, Lugar 5%, outcome as a plus for
Buchanan 2% of Gramm and a setback
about 1,200 votes for Dole.
cast
Table 1.2 continued

Date Event Winners and Losers Particulars
November 4, 1995 Maine GOP Gramm 42%, Lugar Participants paid $15
21%, Dole 10%, to vote.
Specter 8% with
1,500 votes cast
November 18, 1995 Florida’s Presidency III Dole 33%, Gramm Voting limited to reg-
26%, Alexander 23% istered Florida Repub-
of 3,325 votes cast licans, mostly chosen
by lot or party office
as delegates to state
convention.
January 30, 1996 Alaska GOP Buchanan 33%, Buchanan waged an

Forbes 31%, Dole extensive television
17%, Keyes 10%, campaign in contest
Gramm 9% of about open to all registered
10,000 votes cast voters.
February 11, 1996 California GOP mail Dole 36%, Buchanan Participants paid $25
poll 25%, Forbes 18%, per ballot; of 390,000
Gramm 9% of invitations mailed out
21,329 votes cast to Republicans,
20,000 were returned
along with fee; more
than 100 delegates to
the GOP state conven-
tion also voted.
Sources: “Dole Wins Early Nod in Race for President,” Des Moines Register, June 25,1994,
IA; “Texas Senator Has a Victory in Louisiana,” New York Times, January, 1995, 12A; “As
South Carolina GOP Uses ‘Straw Poll’ Dinner to Raise Cash, Hungry ’96 Hopefuls Crowd
Table,” Wall Street Journal, March 3, 1995, 12A; “Gramm is Winner in South Carolina GOP
Straw Ballot,” Washington Post, March 5, 1995, 7A; “Buchanan Big Winner in Va. Poll,”
Washington Post, June 18, 1995, 1B; “Freewheeling Iowa Straw Poll Even Has Out-of State
Voters,” Washington Post, August 19, 1995, 1A; “Dole or Gramm? Iowa GOP Says Yes,”
Washington Post, August 20, 1995, 9A; “Maine GOP Backs Gramm in Straw Poll,” Wash-
ington Post, November 5, 1995, 18A; “Dole Narrowly Wins Florida GOP Straw Poll,” Wash-
ington Post, November 19, 1995, 1A; “Buchanan, Forbes Top Dole in Alaska GOP Straw
Poll,” Washington Post, January 31, 1996, 4A; “Dole Wins Calif. Straw Poll,” Associated
Press release, February 11, 1996; author’s telephone interview with Oklahoma GOP staff.
At least seven of these limited voting to purchasers of tickets costing

between $10 and $25. Voting early and often was a common practice. In
Ames, for example, anyone from anywhere could cast as many votes as he
had tickets.44 The Alexander, Gramm, and Taylor campaigns brought in
voters from out of state by the bus- and plane-load.45 Small wonder that
master of ceremonies Dan Quayle opened the gathering by asking, “Is any-
body here from Iowa?”46
No candidate took these pseudo-events more seriously than Gramm.
Like the early losers of past nominating races, Gramm hoped to parlay a
string of straw poll wins into a firm grip on second place. Accordingly, the
Gramm campaign invested heavily in the Louisiana straw vote of January
1995.47 Gramm competed with Alexander to buy up tickets for the “silver
elephant” poll in South Carolina; Gramm bought more than Alexander
and won.48 His organization in Iowa admitted to having spent at least
$200,000 on the Ames vote alone, although half a million is a more likely
figure.49 Gramm likewise bought the vote in the Maine straw poll.50
Any benefit accruing to Gramm from these synthetic victories proved
evanescent. The vast sums spent on contesting straw polls probably exceed-
ed whatever new money they attracted. Moreover, his inability to rise in the
real polls gave pundits additional reason not to take his ersatz wins seri-
ously.51
The book on straw polls should not be closed before making note of
Presidency III, a conspicuous exception to the other contests just
described.52 Participation in this straw poll was restricted to Republican
delegates registered to vote in Florida who had been chosen in county cau-
cuses and lotteries. GOP office-holders and state party officials also got to
vote, as did contributors of $5,000 or more.53 Dole went all out to win this
November showdown by flying in 21 out-of-state politicians, wooing del-
egates, and lavishing money on social events. Like Gramm, Alexander, and
Buchanan, he took time off from Iowa and New Hampshire to meet with
the delegates. After extravagant spending by all four campaigns,54 Dole
won with 33 percent. Gramm finished second, Alexander third.
EARLY CAUCUSES AND DE FACTO PRIMARIES

For all his early advantages in fund-raising and the polls, Dole suffered
greatly during the invisible primary. Much of the problem was of his own
making. At no point during the campaign did Dole make a substantive
argument for nominating him. His utterances raised doubts that he pos-
sessed any vision whatever. When asked what he would do first if elected,
for example, Dole answered, “If I get elected, at my age, you know, I’m not
going anywhere. It’s not an agenda. I’m just going to serve my country.”55
Changing stands on key issues increased this concern. Having built a
reputation in the Senate as a consummate deal-maker and dedicated prag-
matist, Dole now tried to pass himself off as no less ardent a conservative
than Gramm, the rival he feared most during the invisible primary. Dole
endorsed Oliver North for the U.S. Senate, signed the same no-new-taxes
pledge in New Hampshire that he had spurned in 1988, began opposing
affirmative action, decried welfare for unwed mothers in their teens, blast-
ed multiculturalism, and assailed the gratuitous violence and explicit sex in
films and recordings.56
Incredibly, for all of his experience with the news media, Dole repeated-
ly gave the press openings to point out contradictions and inconsistencies
in his rhetoric. It soon came to light that he had not seen the films he had
criticized and that he had avoided attacking Hollywood moguls who had
contributed to his campaign.57 Another flap erupted when his campaign
aides first solicited and then returned a contribution from the Log Cabin
Republicans, a gay caucus in the GOP. Dole contradicted his own aides
before lapsing into an uneasy silence on the matter. By that time, however,
his “flip flopping” had drawn criticism from all parts of the GOP.58 Dole
also angered the religious right by campaigning for the right-to-life vote
while refusing to take an absolute stand against abortion.59
As front-runner, Dole naturally drew fire from his rivals. Alexander,
Buchanan, and Gramm repeatedly disputed his credibility as a conservative
and frequently called attention to his advanced age (72 in 1995). Dole gave
Republicans fresh basis for concern on both counts when he delivered the
GOP’s response to Clinton’s 1996 State of the Union address. Almost
everyone agreed that he looked old and sounded tired compared with Clin-
ton. Many Republicans saw the speech as yet another demonstration of his
infirmities as a candidate, a sentiment his rivals pounced upon. “Senator
Dole,” Alexander assured an Omaha audience, “is too decent to fake a
vision he does not have.”60
Dole endured much worse from the ceaseless barrage of Forbes attacks
in Iowa, New Hampshire, and Arizona. In the final week of January 1996,
for example, Forbes aired 526 spots in Iowa and another 630 in New
Hampshire, most of which flayed Dole as a Washington insider, a toady of
the special interests, and a tax-hiker.61 Before Forbes got into the race, Dole
had enjoyed a comfortable lead of 40 percent in Iowa. Scarcely six months
later, Dole had dropped to 28 percent and Forbes had moved up to second
place with 16 percent. Forbes passed Dole in some New Hampshire polls
and in Arizona.62
On January 12, Dole effectively declared total war against Forbes. In
Alabama to address the Southern Republican Leadership Conference, Dole
pointed out that Forbes had held only one minor post in government. A
Dole ad attacking Forbes’s inexperience and “risky ideas” aired the same
day in Iowa. Another grossly distorted Forbes’s stand on the imprisonment
of violent criminals. Yet another branded him as “just too liberal on wel-
fare,” “untested,” and “more liberal than you think.” One week before the
Iowa caucus vote, the Dole campaign bought 138 half-minute spots on Des
Moines television.63
Dole settled on the flat tax as the riskiest Forbes idea, charging that it
would burden ordinary Americans while making moguls like Forbes even
richer.64 He released his tax returns of the last 29 years and repeatedly chal-
lenged Forbes to do the same. Dole even floated a flat tax of his own with
deductions for home mortgages and charitable contributions. He ridiculed
Forbes for prescribing the flat tax as a remedy for almost every ailment: “If
you’ve a got a headache, the cure is the flat tax. If your feet hurt, the cure
is the flat tax. If you don’t want to pay any taxes, the cure is the flat tax.”
Beginning February 1, the Dole campaign ran an effective if inaccurate ad
featuring Governor Stephen Merrill of New Hampshire warning that the
Forbes tax would cost the typical New Hampshire household $2,000 more
in taxes. A similar Dole spot featuring Senator Charles Grassley aired in
Iowa.65
Dole attacked on other fronts as well. He repeatedly accused Forbes of
trying to buy the nomination and branded him a liberal for taking on the
Christian Coalition. The Dole campaign sent flyers to 40 percent of all reg-
istered Republicans in Iowa and New Hampshire denouncing Forbes’s sup-
port of welfare for illegal immigrants and distorting his position on impris-
oning violent criminals.66 Extensive “push polling” supplemented these
attacks.67 The consequences of all this could be seen in the Iowa polls. In
two months’ time, Forbes’s negative ratings among likely caucus partici-
pants soared from 26 percent to 46 percent while Dole’s rose from 9 per-
cent to 29 percent.68
Gramm despaired of the polls even more than Dole. Forbes had dis-
placed him as Iowa runner-up in November, and by January Gramm had
fallen to fifth place in the final poll of likely caucus participants. He
remained frozen at 5 or 6 percent in the New Hampshire polls. Arizona
and California also looked grim for Gramm.69
Gramm hoped to turn things around by winning the Louisiana caucus-
es on February 6. Though boycotted by most candidates out of deference
to Iowa, Louisiana nonetheless marked the actual start of 1996 delegate-
selection. Buchanan saw these rogue caucuses as a golden opportunity to
knock Gramm out of the race and thereby become the conservative alter-
native to Dole. Gramm, however, had built a strong organization there and
fully expected Louisiana conservatives to back a Texas conservative. Inex-
plicably, however, he spent most of the final week elsewhere while
Buchanan worked Louisiana hard and won a key endorsement from Gov-
ernor Mike Foster.70
Returns from Louisiana revealed that Gramm had badly underestimat-
ed Buchanan, who trounced him in the preference vote and ended up with
13 of the 21 delegates. The Voter News Service (VNS) poll of caucus par-
ticipants disclosed that about 90 percent identified themselves as “conser-
vative,” and that most of these claimed to be “very conservative.”

Buchanan won big among conservatives of every description; he also
appealed to elderly, less educated, low-income, and anti-abortion
activists.71 In a single stroke Buchanan changed the dynamics of the Repub-
lican race.
When the Iowa precinct caucuses took place six days later, only 96,451
of a projected 130,000 turned out, a dropoff some attributed to all the neg-
ative campaigning.72 Dole “won” with 26 percent in a straw poll taken just
before the actual selection of delegates to county conventions. Buchanan
finished a strong second at 23 percent, and Alexander rounded out the top
three with 18 percent.73 Forbes came in a dismal fourth, after spending
$407 per vote.74 Gramm quit two days later.
The Iowa results must have alarmed Dole, for he had barely bested
Alexander among self-identified moderates and fared poorly with the con-
servatives.75 Moreover, he had been judged less electable than Alexander in
the VNS poll. The poll also showed that half of his early supporters had
backed somebody else on caucus day. Less than one-third of the majority
of caucus attenders who did not identify with the religious right had sup-
ported him. R.W. Apple compared Dole’s 1996 vote to the 36 percent he
won in the 1988 caucuses and pronounced it “dangerously narrow.”76
“Narrow” likewise described the support for Buchanan and Forbes.
Buchanan owed his strong showing to an intense but overlapping follow-
ing consisting of the very conservative, those who identified with the reli-
gious right, those who wanted a nominee with conservative convictions,
those who wanted the GOP to oppose abortion, and those who had not
supported Dole at any point in the race. Forbes fared well only with those
most opposed to nominating a career politician and those most concerned
about tax plans. Perceived electability surfaced as Alexander’s most note-
worthy asset in the VNS poll, probably the result of his endlessly reiterat-
ed claim that only he could beat Clinton.
All attention was now riveted on the final eight days of the New
Hampshire campaign. The daily tracking polls showed Dole peaking at 32
percent six days before the primary vote.77 He fell steadily thereafter to 23
percent on the eve of the primary. Buchanan rose from 19 percent to 25
percent where he remained until he gained a point in the final sampling.
Forbes’s support dropped from 25 percent to 13 percent between February
11 and 16. Meanwhile, Alexander rocketed from 11 percent to 23 percent
on February 16. With this kind of momentum, he looked almost certain to
finish second if not first.
Facing the prospect of finishing third in New Hampshire, which almost
certainly would have driven him from the race, Dole shifted his fire from
Buchanan to Alexander in a desperate bid to finish no worse than second.
He filled the airwaves with ads branding Alexander a liberal on taxes,
crime, and spending. Massive push-polling embellished these charges. The
Alexander campaign had expected something of this sort, for everyone now
appreciated that whoever beat Buchanan in New Hampshire would in all
probability become the nominee.78
Dole’s change of primary targets paid off. Alexander dropped by three
crucial points in the last three days of the New Hampshire campaign. Dole
ended up in second place, with 26 percent to Buchanan’s 27 percent, a dif-
ference of only 2,090 votes out of 208,993 cast. Alexander got 23 percent
while Forbes finished a dismal fourth. Buchanan picked up six delegates to
four for Dole and two for Alexander.79
The VNS exit poll showed Buchanan scoring impressive majorities
among the very conservative, the religious right, those who had voted for
him in the 1992 primary, pro-life plank advocates, those who based their
vote mainly on abortion, and those who wanted a strong and principled
conservative as nominee.80 Buchanan also captured a majority of the small
number most concerned about foreign trade.
Dole garnered a majority from the 10 percent who viewed Washington
experience as a plus and from those mentioning the deficit as the most
important issue in their vote. Nearly half of all voters most concerned
about nominating an electable candidate backed him, a big improvement
over the Iowa poll numbers. On the debit side, only 10 percent who felt
strongly that the nominee should have a vision of the future voted for him,
as did 15 percent of those wanting a champion of conservative values.
Alexander fared well among liberals and moderates, those wanting a
candidate with a vision for the future, and those most offended by negative
advertising. He even carried a majority of those voters wanting someone
“not too extreme” to win the nomination, an interesting comment on
Dole’s attempt to brand Buchanan as “too extreme” to carry the Republi-
can standard. Almost half of those citing education as the biggest factor in
their choice of candidates voted for Alexander. As in Iowa, he lost out to
Forbes among voters most desirous of an outsider nominee. And, in a near
perfect reversal of the Iowa entrance poll, Alexander lagged well behind
Dole among those looking for a winner against Clinton. Forbes found lit-
tle to celebrate except his majority among the voters wanting an outsider
as nominee and his plurality among those who regarded taxes as the most
important issue.
Many in the GOP took fright at Buchanan’s apparent strengths and
Dole’s demonstrated weaknesses.81 Celebrating in New Hampshire,
Buchanan rightly anticipated that party leaders would “come after this
campaign with everything they’ve got.”82 Indeed, within 48 hours of his
victory, Buchanan had been attacked by Newt Gingrich, New York Mayor
Rudolph Giuliani, House GOP conference chairman John Boehner, Colin
Powell, George Will, Senator Olympia Snowe, Rush Limbaugh, Senator
John McCain, and Senator Alfonse D’Amato.83
Playing to Republican fears of another debacle on the scale of Barry

Goldwater’s nomination in 1964, the winless Alexander asked Dole to
withdraw so that he could unify the party against Buchanan and Clinton,
a proposal that Dole promptly spurned.84 Lugar rebuffed Dole’s entreaties
to get out.85 At this point, of course, the process precluded the entry of new
candidates. Even in defeat, Dole looked like the only viable alternative to
Buchanan. Alexander’s only hope of altering this perception was to win
something soon. For his part, Dole announced that “the real Bob Dole”
would now make his appearance to save the GOP from the “extremist”
Buchanan.86
Four primaries now loomed in the immediate future: Delaware on Feb-
ruary 24; Arizona, North Dakota, and South Dakota on February 27. Dole
should not have been surprised by the Delaware outcome, for he and
almost every other candidate had agreed not to campaign there out of
respect for New Hampshire. (Delaware Republicans scheduled their con-
test to occur four days after the New Hampshire vote, thereby violating the
tradition of allowing New Hampshire an entire week unto itself before the
next primary. The New Hampshire party retaliated by pressuring the can-
didates to ignore Delaware. Forbes and Keyes refused.87) Forbes made a
major effort in Delaware, capturing all 12 delegates under winner-take all
rules. Dole came in second.
Although declared an easy winner in North and South Dakota on Feb-
ruary 27, Dole suffered through a long night of network predictions that
he would lose to both Forbes and Buchanan in Arizona.88 In the final tab-
ulation, however, Dole edged out Buchanan to finish second. He now
vowed to win the South Carolina primary on March 2.89 A new Gallup poll
of registered Republicans nationwide showed that he was still the front-
runner.90
Arizona hurt Buchanan much more than it did Dole. Buchanan had
spent nearly all of his time there since winning New Hampshire. A win in
Arizona, Buchanan told the press, “would give us such a propulsion it
would be very, very difficult to stop us from winning the nomination.”
After receiving false reports of victory, Buchanan told ecstatic followers
that he was going “all the way to the White House.”91 The actual outcome,
however, suggested that he was going nowhere.
Up to this point, Alexander had raised substantial sums by selling the
scenario that Dole would collapse just as he had in 1980 and 1988.92 This
time, however, front-loading dictated the outcome almost as much as the
pace. In the week following New Hampshire, Alexander finished fourth in
Delaware, fourth in Arizona, fourth in South Dakota, and fifth (behind
even the departed Gramm) in North Dakota. Indeed, having failed to win
10 percent of the preference vote in any of these primaries, Alexander now
risked losing his federal matching funds.93 Polls in South Carolina and the
Super Tuesday South held out little hope of his recovery.94
Reestablished as a contender, Forbes concentrated on the South Caroli-

na primary, on the New England states voting in the “Junior Tuesday” pri-
maries of March 5, and on New York, where he won a protracted battle to
get his name on the March 7 ballot. Though still pushing the flat tax and
advertising on a lavish scale, Forbes let up on the attacks and supplement-
ed his message with foreign policy themes.95
Thanks largely to the overwhelming support of Republican officials and
party regulars, Dole won big in the South Carolina primary. He picked up
45 percent of the preference vote, compared to 29 percent for Buchanan,
13 percent for Forbes, and 10 percent for Alexander. According to the VNS
exit poll, he won outright majorities among senior citizens (a 15-percent
improvement over his New Hampshire showing), deficit hawks (a 30-per-
cent improvement), those most concerned about nominating a winner in
the general election (a 34-percent gain), those who prized Washington
experience (97 percent for Dole, a gain of 7 percent), and free traders (no
comparable question asked in New Hampshire). Dole also made inroads
into Buchanan’s base. Among South Carolina voters identifying with the
religious right, he captured 40 percent of the vote against Buchanan’s 43
percent.96
South Carolina sent a clear signal to Buchanan that he could not win a
two- or even three-man race against Dole. Compared to the New Hamp-
shire vote, his support among voters wanting an avowedly conservative
nominee dropped by 18 percent. The Palmetto primary also cast a pall over
Alexander and Forbes. Alexander got enough votes to become eligible
again for federal money, but in every other respect he suffered a devastat-
ing loss.
Finally on the fast track, Dole captured outright majorities of the pref-
erence vote in Connecticut, Maryland, and Rhode Island and won a plu-
rality in all of the other Junior Tuesday primaries on March 5.97 Two days
later Dole picked up all 93 of the delegates at stake in the New York pri-
mary, even though 45 percent of those sampled in an exit poll expressed
dissatisfaction with the choice of candidates, and 57 percent complained
that he lacked new ideas.98 By this time, Alexander and Lugar had dropped
out and endorsed him. Dole romped through the Super Tuesday primaries
on March 12, forced Forbes out of the race on March 14, and trounced
Buchanan in a Midwestern regional primary on March 19. Dole may well
have gone over the top in delegates at this point, but he did not claim the
prize until sweeping the California Tuesday primaries on March 26. This
meant that primary voters in Pennsylvania, Indiana, North Carolina, the
District of Columbia, Nebraska, West Virginia, Arkansas, Idaho, Kentucky,
Alabama, Montana, New Mexico, and New Jersey would play no mean-
ingful part in deciding the 1996 Republican nomination. Dole racked up
huge majorities in all of these ex post facto primaries, garnering more than
70 percent of the vote in nine of the thirteen.
By that time, however, even Buchanan had given up, suspending his can-
didacy six days before the Pennsylvania primary.99 He had lost all of the pri-
maries since New Hampshire, prevailing only in the little-noted caucuses in
Missouri on March 9.
THE UNCONTESTED DEMOCRATIC NOMINATION

Analysts on the Republican “Tsunami” of 1994 typically counted Clinton’s
presidency as part of the debris. Rightly or wrongly, much of the blame for
this electoral debacle fell on Clinton. In 1992 he had posed as a centrist
“new Democrat” who would end chronic welfare dependency, make com-
munities safe, and promote the interests of the great middle class. Once in
office, however, he made common cause with liberal Democrats in the Con-
gress to push a tax-and-spend program of economic recovery. At the same
time he backtracked on promises to open the armed forces to gays and to
appoint Lani Guinier Assistant Attorney General for Civil Rights. Early
accounts of his administration reported a White House awash in consul-
tants and consumed by interminable and inconclusive meetings.100
More problems followed when he set welfare aside to concentrate on a
massive overhaul of the nation’s health-care system. The end product of a
task force headed by wife Hillary Rodham Clinton was so lengthy and
complicated that it bewildered even members of his own administration.101
Congressional Democrats haggled over approaches while congressional
Republicans and allied interest groups pounced on the plan’s ambiguities.
Clinton suffered a stinging defeat and Clinton health care became a night-
marish caricature of what liberal Democrats like Hillary had in mind for
the nation.102
Accusations of marital infidelity, fraudulent dealings, misuse of FBI files,
and concealing evidence in the suicide of a key aide all added to the Presi-
dent’s woes, as did the killing of American soldiers in Somalia. Clear signs
of economic recovery helped his job approval rating remarkably little. Less
than 40 percent of the public approved of his performance at one point
during the summer of 1993. Some improvement could be seen at year’s end,
but, for most of 1994, Clinton remained below 50 percent.103
Massive Democratic defeat in 1994 generated a conventional wisdom
that Clinton would not be reelected. Prominent Democrats talked openly
of dumping him. Liberal columnist Jack Newfield scourged him as an
unprincipled liar and called on him to bow out. Pundit Jack Germond
described him as “politically naked” for want of a following in the Demo-
cratic party. One poll found that 66 percent of the Democrats interviewed
did not want Clinton to escape a primary challenge. Matched against a
generic Republican and an unidentified independent in this poll, Clinton
lost to the Republican, 33 to 40 percent.104
Although not immediately apparent, the electoral setback of 1994 did
have a positive side for Clinton. It undermined any primary challenge from
the left. With liberal Democrats in disrepute, potential rivals like Jesse Jack-
son saw little point to fighting him for a worthless nomination. Indeed,
even as Democrats bewailed Clinton’s failings, pundit Fred Barnes noted
the curious absence of primary challengers. “Is there anybody organizing?”
Barnes asked. “Is there anybody talking about organizing?”105 Meanwhile
Clinton embarked on a strategy of “triangulation” that moved him back to
the center and distanced him from liberal Democrats as well as conserva-
tive Republicans in Congress.106
The closest thing to a challenge issued from an old enemy, Robert Casey,
the former governor of Pennsylvania. Casey had gained national attention
for his opposition to abortion and attacks on “left-wing ideologues” in his
party. He had also undergone massive heart and liver transplant surgery
and lacked both the stamina and resources to make an all-out effort. Less
than a month after forming an exploratory committee, Casey admitted that
he could not keep up the pace of fund-raising and speechmaking.107 His
withdrawal in April 1995 marked the end of organized Democratic oppo-
sition to Clinton’s renomination.
Clinton by this point already had achieved an astounding comeback,
surpassing Dole in most polls. His resurrection began with an effective
show of presidential leadership immediately after the Oklahoma City
bombing in April 1995. Two months later, Clinton stunned liberal Democ-
rats and stole some of the Republicans’ thunder by proposing to balance
the budget within ten years.108 The State of the Union address in January
1996, with its reference to ending “the era of big government,” further
strengthened his credibility as a centrist. (By this time Clinton had raised
more prenomination money than Dole.109) Shortly before the Republican
convention in San Diego Clinton let it be known that he would sign a com-
promise version of the Republican welfare-reform bill. Dole had hoped to
make welfare reform an issue in the fall campaign.110
Little need be said about 33 uncontested primaries in which Clinton
won 87 percent of nearly 11 million votes cast. His support ranged from
76 percent in his home state of Arkansas to nearly 100 percent in Georgia
and Indiana.111
THE PRE-CONVENTION PHASE OF THE GENERAL ELECTION

CAMPAIGN
Usually the winner of an early nomination has cause to celebrate. More
time exists to reunite the party, refine the message, scrutinize potential run-
ning-mates, and script the convention—all essentials of success in the fall
campaign. Unfortunately for Dole, the long interval between the actual and
the official nomination became an ordeal of straitened finances and Repub-
lican infighting, exacerbated by unanswered Democratic attacks, frustra-
tion in the Senate, and his own blunders.
The most immediate consequence of Dole’s costly wars against Forbes,

Buchanan, and Alexander came to light in April disclosures that his cam-
paign had spent all but $2 million allowed under the FECA limit. More-
over, at least a quarter of what still could be legally spent was reserved for
the San Diego convention nearly four months distant, at which point Dole
would receive his federal grant for the general election.112 Although party
coffers could be tapped for so-called “generic” campaigning in the interim,
Dole recognized the depths of his predicament. Clinton, he lamented on
CNN, “has $25 million in the bank and we’re broke.”113
Democratic money was already taking a toll on Dole support in Cali-
fornia, Ohio, Michigan, Illinois, Florida, and other key electoral states.
There Democratic ads continually tarred Dole as opposed to the welfare of
senior citizens, students, and the environment. From October 1995 up to
the Chicago convention, the Democratic National Committee (DNC) spent
$42.4 million on these so-called “issue advocacy” ads. Dole depended
heavily on Republican National Committee (RNC) money to answer these
attacks. The RNC did respond in eight of a dozen states targeted by the
Democrats, but only in California did it come close to matching the
Democrats dollar for dollar.114
With the race basically decided by Super Tuesday, Dole returned to the
Senate in hopes of showcasing his leadership ability. After achieving initial
success on giving the next president an item veto and on raising the debt
ceiling, Dole learned that his interests as presidential nominee did not
always coincide with congressional Republicans. This became apparent
when House Republicans embarrassed him by passing a repeal of the
assault weapons ban. Dole angered them and the gun lobby by refusing to
schedule a Senate vote. Five Senate Republicans helped the Democrats kill
a medical savings account bill important to Dole’s presidential campaign.
Incredibly, given his own time in Congress, Dole got bogged down in a bat-
tle over term limits that few senators wanted to fight. Democrats took full
advantage of Senate rules to thwart his initiatives, and they forced him
repeatedly to block a popular bill to revise the minimum wage.115
Although accounts differ on Dole’s reluctance to resign from the Sen-
ate,116 they agree that his strategy of campaigning for president on the Sen-
ate floor had not worked. Rather, it had inspired more Republicans in and
out of Congress to express misgivings about their nominee. Falling in the
polls, Dole announced on May 15 that he would give up his seat and “seek
the presidency with nothing to fall back on but the judgment of the people
and nowhere to go but the White House or home.”117
Back on the campaign trail Dole got into fights that distracted him and
dismayed the GOP. One such flap erupted after Dole revealed his doubts
that cigarettes were addictive and that the tobacco industry should be more
regulated. The Democrats carried on as if their candidates had never taken
tobacco money or fought for tobacco in Congress.118
Dole stumbled again when declining an invitation to speak to the

National Association for the Advancement of Colored People (NAACP).
President Bush had passed up a similar opportunity in 1992, and nobody
denied that an NAACP convention was less than an ideal venue for con-
servative views on affirmative action. Dole, however, made headlines by
bluntly accusing chairman Kweisi Mfume of trying to set him up.119
Although his share of the anti-abortion vote had increased since New
Hampshire, the issue itself had not gone away. Ralph Reed warned Repub-
licans just before the March 19 primaries that “you cannot, you should
not, and you must not retreat from the pro-life and pro-family stand[s] that
have won you that majority in the first place.”120 When Reed hinted in May
that the Christian Coalition might support a change in the GOP platform
to permit an exemption for mothers in danger of dying in childbirth, Bay
Buchanan, sister and campaign manager of Pat Buchanan, avowed that “no
words can be changed, no words can be added. There is no compromise,
there is no negotiating, there is no appeasement.”121
Moderate Republicans seized this opportunity to call for an open debate
on abortion in the platform committee. Senator Olympia Snowe main-
tained that “the more we suppress debate on this issue, the more it hurts
the Republican party. It’s not the abortion issue in and of itself, but what
the Republican party says about women—less government is better, except
in the case of women.” Congressman James Greenwood of Pennsylvania
argued that the platform should not even mention abortion.122 Dole spoke
out on June 6. “I expect to run for president with the existing pro-life lan-
guage from our 1992 platform,” he said, “and with the declaration of tol-
erance for different points of view on such issues as abortion.”123
On June 11, Dole argued that tolerance language belonged in the pro-
life plank itself because abortion was a moral issue, not to be equated with
the likes of taxes and trade.124 This statement dismayed Congressman
Henry Hyde, the platform committee chairman, who had been led to
believe that general language in the preamble would be acceptable to all.
“There are enough contentious issues,” Hyde warned Dole, “when you put
a major one to rest, you like to see that it stays there.”125
By early August a floor fight looked likely when moderates repeatedly
failed to win any concession from the platform committee. Rebuffed on the
plank itself, Dole supporters tried to insert tolerance language elsewhere in
the platform. Fearing more adverse publicity if the issue went to the whole
convention, Dole settled for a general statement of tolerance and an appen-
dix listing all of the proposals defeated in the platform committee. The
right-to-life language remained unchanged from 1992. “The unborn child
has a fundamental individual right to life which cannot be infringed. We
support a human life amendment to the Constitution and we endorse leg-
islation to make clear that the Fourteenth Amendment’s protections apply
to unborn children.”126
Although hints of a flat tax in Dole’s economic package turned up in the

press as early as March,127 the Dole team appeared to make little progress
over the next several months. Dole headquarters finally released the long-
awaited plan on August 4. It contained major concessions to Forbes and
accordingly contradicted Dole’s career as a deficit hawk. Rather than adopt
a single tax rate as advocated by Forbes, the plan called for $548 billion in
tax cuts—achieved principally by a 15-percent slash for every tax bracket
and a reduction in the capital gains from 28 percent to 14 percent. The plan
also called for a $500-per-child tax credit and repeal of the 1993 increase
in Social Security taxes on higher-income Americans. Little was said about
balancing the budget.128 The plan came under Democratic attack even
before Dole first revealed its particulars, in a speech to the Chicago Cham-
ber of Commerce.129
THE NATIONAL NOMINATION CONVENTIONS

The most vestigial of presidential nominating institutions is the national
convention. Presidential nominations in a front-loaded system are decided
long before the delegates are called to order. Typically the nominee picks a
running-mate shortly before the convention begins. Contemporary conven-
tions have been likened to operas, scripted and scored mainly for benefit of
the audience.130 Certainly Byron Shafer’s description of the convention as
“bifurcated” remains apt in the era of massive front-loading.131 Indeed, the
televised version now amounts to little more than prime-time coverage of
selected speeches (when not interrupted by pundits interviewing them-
selves). Only those in attendance or watching C-SPAN experience the real
convention.
Convention platforms, however, still retain value as guides to party prin-
ciples and priorities. The Democratic platform adopted August 27 in
Chicago echoed Clinton themes of limited government offset by rejection
of “the misguided call to leave our citizens to fend for themselves.” Amer-
icans, according to Democrats, wanted a “course that is reasonable, help
that is realistic, and solutions that can be delivered—a moderate, achiev-
able, common sense agenda that will improve people’s daily lives and not
increase the size of government.”132 Most of the Democratic plank on eco-
nomic growth lauded Clinton’s version of a balanced budget. The fact that
the Democrats had embraced the principles of smaller government, a bal-
anced budget, tax relief, and even welfare reform spoke volumes about lib-
eral disarray and how much Clinton’s stock had improved since 1994.
This is not to say that voters lacked a choice in 1996. The Republican
and Democratic platforms clashed on what to do about Medicare, funding
of the arts, and support for National Public Radio and Legal Services. The
Republicans’ manifesto called for constitutional amendments requiring a
balanced budget, recognizing the unborn child’s right to life, denying auto-
matic benefits to children of illegal immigrants, and enhancing the rights of
crime victims. It also proposed to terminate the Departments of Commerce,

Education, Energy, and Housing and Urban Development. In contrast, the
Democrats talked about “reinventing government” and saving programs
from the Republican ax. The parties also differed on gun control and other
“hot button” issues. Taking aim at the Republican right-to-life plank, the
Democrats lauded “the right of every woman to choose, consistent with
Roe v. Wade, and regardless of ability to pay.” No reference was made to
partial-birth abortions, prominently mentioned in the Republican plat-
form. Responding to the Republican plank on illegal immigration, the
Democrats decried “those who used the need to stop illegal immigration as
a pretext for discrimination.” After professing support for the ideal of
equal treatment under law, the two parties offered opposing views of how
the law should regard race, gender, and sexual orientation in the workplace
and universities. In these and other respects the platforms pointed up
important differences between a predominantly liberal and a preponder-
antly conservative party.
Previous research has shown that convention delegates are more ideo-
logically intense and unified than are registered voters of the same party or
the electorate as a whole.133 As documented in earlier studies, in 1996,
Republican and Democratic delegates positioned themselves at opposite
ends of the ideological continuum while registered voters in the aggregate
occupied the center.134 On issues of policy and ideology, Republican dele-
gates more often identified themselves as conservative than registered
Republican voters, and Democratic delegates were always more liberal
than registered voters of their party.
Interviewed before they learned that Kemp was the vice-presidential
pick, barely half of the San Diego delegates offered positive assessments of
Dole’s campaign. By putting Kemp on the ticket Dole finally electrified a
convention desperate for something to celebrate. Spirits lifted as conven-
tion orators extolled the economic plan, remembered Ronald Reagan,
plugged family values, and ridiculed Clinton. In Chicago the euphoric
Democrats shrugged off news of Dick Morris’s sex scandal, danced in the
aisles, and renominated their ticket by acclamation.
THE PEROT PARTY NOMINATES PEROT

Nothing aids prediction better than repetition, and, in key respects, the
Perot candidacy of 1996 resembled the Perot candidacy of 1992. Both
times Perot sought to endow his candidacy with the appearance of popular
demand. Perot launched his first campaign during a guest appearance on
CNN’s Larry King Live. He turned again to the King show to make his
1996 announcement. In 1992 Perot ran as an independent and transformed
his United We Stand organization into an electoral machine. Four years
later he chose to run as a third-party candidate and thereby needed a party
to nominate him. Perot looked almost certain to run in 1996 after having
won nearly 19 percent of the popular vote in 1992 and having helped the
Republicans recapture the House in 1994.
The tip-off to what Perot had in mind for 1996 came when he sum-
moned the Republican aspirants and Clinton surrogates to address the
1995 convention of United We Stand. Ostensibly the point of these invita-
tions was to promote discussion of key issues with Perot followers. Nearly
every Republican spoke, as did Thomas McLarty III of the White House
and DNC chairman Christopher Dodd. True to form, Perot laid down an
impossible ultimatum at the end of the proceedings: unless the Republicans
and Democrats balanced the budget, solved the problems of Medicare and
Social Security, curbed the special interests, and imposed term limits on
themselves—all of this to be accomplished by Christmas—he might well
take up the torch.135
Perot waited only until September before launching the Reform party,
the principal purpose of which was to nominate a presidential standard-
bearer. Its platform would champion campaign finance reform and propose
solutions to other problems discussed in Dallas. Perot promised to find
“world-class” bidders for the presidential nomination, and, when asked,
pointedly refused to rule out his own candadicy.136
By this time a major effort was already underway to get his party on the
1996 ballot in California. State law gave Perot little more than three weeks
to acquire the necessary names of “party members” for a line on the pres-
idential ballot. Paid and volunteer workers collected more than the 89,007
needed. Eventually, under one name or another, Perot or the Reform party
got on the ballot in 44 states. Perot ran as an independent, rather than as
a third-party candidate in Alaska, Mississippi, South Dakota, Tennessee,
Texas, and Wyoming.137
Gordon Black, Perot’s pollster, summed up the conventional wisdom
about who would get the nomination: “It has to be Perot.” Perot said as
much himself. “Let’s assume the dust clears, and that’s what the members
of this party want,” he said in March 1996. “Then certainly, I would give
it everything I have because probably there’s not a luckier person alive in
this country today.” Days later, after decrying the evils of deficit spending
and high national debt, Perot declared: “You and I are going to have to
climb back in the ring again and make sure that it gets done, otherwise
we’re leaving a mess to our children.”138
Nearly every potential candidate mentioned in the news wanted no part
of a Perot party. Former Connecticut governor Lowell Weicker, former
Oklahoma senator David Boren, and former Minnesota congressman Tim
Penny all declined the honor. Reportedly, Jack Kemp considered running
but decided against it after receiving little encouragement from Perot.
Richard Lamm, a Democrat and former Colorado governor, expressed
interest if Perot would only disavow an interest in running. Perot gave him
no such assurance, but Lamm announced anyway on July 9, 1996. After
inveighing against “big money, big influence, and narrow elite interests” in
his announcement speech, Lamm acknowledged that he would have no
credibility if Perot funded his campaign.139
Perot waited only a day before declaring his own candidacy in yet anoth-
er appearance on Larry King Live. “If anybody should do this,” he
avowed, “I should do it.” He reiterated this line on ABC’s Good Morning
America before adding: “And I will do it, and I’m in a unique position to
do it.”140
As worked out by Perot, the nomination would be decided in several
stages. First, everyone who signed a petition to get the Reform party on a
state ballot, or who applied to become a party member, would be sent a
preference form. Respondents could propose anyone, but only persons
named in at least 10 percent of these responses would get on the ballot. The
finalists would speak at the first of two conventions in Long Beach, Cali-
fornia, on August 11. A second vote limited to these finalists would occur
by mail, telephone, or other means. Elaborate precautions would be taken
to count the ballots and prevent fraud. The winner would be announced
August 18 at a second convention in Valley Forge, Pennsylvania, where the
nominee would deliver his acceptance speech.141
The problematic nature of this process first became apparent when 10
percent of the 979,882 preference forms proved undeliverable, and only
43,135 completed forms were filled out and returned. Nearly two-thirds of
these respondents backed Perot, but Lamm also qualified as Perot’s only
rival on the ballot.142 By this time an exasperated Lamm was complaining
to reporters about the low response rate in the first round of voting, Perot’s
unwillingness to share his mailing list, the sending of more than one ballot
to some individuals, apparent favoritism for some states over others,
putting Perot’s picture on the ballot, Perot’s initial reluctance to divulge the
names of the firms hired to distribute the ballots and tabulate the vote, and
Perot’s unwillingness to debate him in Long Beach.143
The Long Beach convention turned out to be little more than a rally for
Perot, and he claimed the nomination one week later after winning 65 per-
cent of 49,266 votes cast in all 50 states and the District of Columbia. The
total amounted to a meager return of 4 percent on more than a million bal-
lots mailed out, or about half of the Republican turnout for the 1996 Iowa
precinct caucuses. The electorate exceeded 1,000 in only 12 states and fell
below 100 in five states and the District of Columbia. California alone cast
35 percent of the total vote, more than the combined returns from New
York, Texas, Ohio, Michigan, and Pennsylvania. Perot won by more than
60 percent in 41 states while losing to Lamm in Alaska, Colorado, Min-
nesota, and the District of Columbia.144
If this process was meant to legitimate Perot’s candidacy, it had quite the
opposite effect. Lamm’s well-publicized carping coincided with a big jump
in Perot’s polling negatives. Perot also suffered in the trial-heat compar-
isons with Dole and Clinton, falling from an average of 16 percent in the
earliest polls to only 6 percent in mid-August. Subsequent polls showed
him stuck at 7 percent. His efforts to find a running-mate met with refusal
until author Pat Choate accepted.145
CONCLUSIONS AND CODA

How much did Dole owe his nomination to the front-loaded process
described above? True, the primary calendar is only one part of the overall
situation, but it does figure importantly in who runs and who does not.
Clearly the massive front-loading of the 1996 primaries largely accounts
for the paucity of strong Republican candidates at a time when it looked
like almost any Republican could defeat Clinton. Significant back-loading
would have reduced the criticality of the invisible primary by allowing can-
didates time to raise money and to make strategic adjustments after the
actual primary season had started. Dole would not have become the pro-
hibitive favorite by default, nor would he have remained the only realistic
choice after losing the New Hampshire, Delaware, and Arizona primaries.
Significant back-loading in any event is not in the immediate future of
presidential nominating politics. Although the GOP has approved a plan to
encourage late contests,146 the dates of 2000 primaries and caucuses are
about as front-loaded as 1996. More than the simple majority needed for
nomination will have been chosen by March 14, and more than two-thirds
of all 2,061 delegates will be corralled by March 21. Unless the front-run-
ner stumbles badly, the Republican race should effectively end with the
March 7 primaries. Republican winner-take-all rules of delegate selection
should help in expediting an early decision.
As in 1996, the invisible primary is proving to be the most critical stage
of the process in both parties. The run-up to 2000 has already winnowed
Republicans Senator Fred Thompson, Governor Pete Wilson, Senator John
Ashcroft, Senator Robert Smith, Congressman John Kasich, and Lamar
Alexander. Aschroft, Smith, Kasich, and Alexander waged extensive cam-
paigns before departing. Smith departed the GOP as well. Still other
Republican withdrawals appear likely before 2000. On the Democratic
side, a once-crowded field has narrowed to Vice President Al Gore and for-
mer senator Bill Bradley. Congressman Richard Gephardt, Senator Bob
Kerrey, Senator John Kerry, and Senator Paul Wellstone ventured out on
the campaign trail before withdrawing.
Even more so than in 1996, a prohibitive favorite dominates the Repub-
lican version of the invisible primary. Gallup polls taken in the first six
months of 1999 showed that Texas governor George W. Bush increased his
support from 42 percent in January to 59 percent in June. Elizabeth Dole’s
standings over this same period dropped from 22 percent to 8 percent.
None of the other aspirants—including Pat Buchanan, Steve Forbes, Sena-
tor John McCain, and Dan Quayle—got more than 9 percent in any of
these polls.147 Moreover, in marked contrast to Dole at the same point in

the 1996 race, Bush hit upon an attractive theme (“Compassionate Con-
servatism”) and exploited a near consensus among Republicans to pick a
winner.
Bush also set a new record for fund-raising. On July 15, 1999, his cam-
paign reported receipts of $37.3 million from 74,000 contributors in only
six months, $5 million more than the total amount that Bob Dole took in
during his entire campaign. Forbes ranked second with $9.4 million ($6.6
million of which was his own money) while McCain placed third with $6.3
million. Receipts for the remaining active candidates ranged from $3.5 mil-
lion for Elizabeth Dole to $1.9 million for Alan Keyes.148
Clearly mindful of what happened to Bob Dole, Bush announced that
his campaign would not accept federal matching funds. This freed him
from the FECA spending limits that hobbled Dole in 1996. “You’ve got
limits that constrain a candidate,” he said with Forbes clearly in mind. “I
want to be in a position to respond.”149 Taking a page from the Clinton
playbook, Bush has also raised so-called soft money for an early start to the
general election, as well as for his nominating campaign. Worried by the
early polls, Clinton, Gore, and the DNC have launched a soft-money drive
of their own.150
Several other parallels from the Republican process of 1996 deserve
passing mention. It is a safe bet that Bush has gotten far more coverage in
the news media than any and perhaps all of his rivals. Alexander,
Buchanan, Forbes, and Keyes again cast themselves as outsiders, as did
Quayle and family-values advocate Gary Bauer. Not surprisingly, front-
runner Bush has come under attack as the establishment candidate, albeit
for raising so much money and getting so many endorsements rather than
for government service. Much of the debate centers on competing notions
of conservatism since all of the candidates claim to be conservative. Abor-
tion is central to this discussion, trade evidently less so. As of June 1999,
Forbes had spent nearly $3 million on advertising in four early-voting
states.151 At least one straw poll took on great importance in the 2000
Republican race when Alexander decided to quit after a poor showing in
Ames.
Thanks to Perot’s decision to accept a federal grant and abide by spend-
ing limits in the 1996 general election, the Reform party is entitled to $12
million of public money for its 2000 convention and fall campaign. Near-
ly 70 and discredited by his comparatively weak showing in 1996, Perot’s
future in the Reform party appears to have dimmed with his political star.
Attempts to loosen his hold on the party began only two weeks after the
1996 election and have continued since.152 Perot acquired a formidable rival
for control of the party in 1998, when Reform candidate Jesse Ventura
won election as governor of Minnesota. A former Navy Seal, professional
wrestler, and actor, the flamboyant Ventura has since eclipsed Perot in
media exposure. He disavows any interest in claiming the presidential nom-

ination for himself and disapproves of another Perot candidacy. Perot, of
course, has not taken himself out of contention and still commands a loyal
if diminished following. “As long as I am helpful to the organization,”
Perot proclaimed at the 1999 Reform party convention, “I am certainly
happy to help and participate in any constructive way.”153 The next day
Perot lost out to Ventura in the Reform party’s choice of a new national
chairman.
NOTES
ACKNOWLEDGMENTS: I researched and wrote initial drafts of this chapter while on

sabbatical at the Institute of Governmental Studies (IGS), University of California,
Berkeley. I am indebted to Nelson W. Polsby for admitting me to so productive a
workplace and am deeply grateful to Nelson, the talented library staff, and many
other friends at IGS for making my time there so productive and enjoyable. Robert
Biersack of the Federal Election Commission, Linda Fowler of Dartmouth College,
and Mark Acton and Catherine Tyrell of the Republican National Committee kind-
ly provided essential data. None of the above is in any way responsible for any
errors or omissions on my part.
1. Arthur Hadley, The Invisible Primary (Englewood Cliffs, NJ: Prentice-

Hall,1976).
2. See, for example, John H. Kessel, Presidential Campaign Politics, 4th ed.
(Pacific Grove, CA: Brooks/Cole, 1992).
3. See Nicol C. Rae, The Decline and Fall of the Liberal Republicans: From
1952 to the Present (New York: Oxford University Press, 1989), 127-56. William
G. Mayer provides useful data on how the number of primaries on both sides
increased in tandem. See The Divided Democrats: Ideological Unity, Party Reform,
and Presidential Elections (Boulder, CO: Westview Press, 1996), 12-13.
4. The same held true for the Democrats. See Emmett H. Buell Jr., “The Invis-
ible Primary,” In Pursuit of the White House, ed. William G. Mayer (Chatham, NJ:
Chatham House, 1996), 7.
5. Elaine Ciulla Kamarck and Kenneth M. Goldstein, “The Rules Do Matter:
Post-Reform Presidential Nominating Politics,” in The Parties Respond: Changes in
American Politics, 2d ed., ed. L. Sandy Maisel, (Boulder, CO: Westview Press,
1994), 183.
6. “The National Tally,” CQWR, 1996, 577, 650, 727, 826, 907, 1183, 1329,
and 1704.
7. Maureen Dowd, “Stars of War Room Are Auditioning for the Presidential
Battles to Come,” NYT, February 11, 1991, 7A; “Washington Wire: Gramm in
‘96,” WSJ, September 20, 1991, 1A.
8. Andrew Rosenthal, “Weary of ’92 Campaign? Now Try Pondering ’96,”
NYT, December 22, 1991, 10A.
9. Jack Anderson and Michael Binstein, “Buchanan Aims Right Toward

1996,” WP, August 3, 1992, 16C; Clara Germani, “Jockeying for ’96 Is in Full
Swing,” CSM, August 21, 1992, 7; Michael McQueen and John Harwood, “With
‘92 Presidential Race Just Begun, GOP Hopefuls Line Up at the ’96 Gate,” WSJ,
August 21, 1992, 8A. Other potentials for 1996 included South Carolina Governor
Carroll Campbell, former drug czar and education secretary William Bennett, Wis-
consin Governor Tommy Thompson, Massachusetts Governor William Weld, and
Elizabeth Dole of the Red Cross and wife of Senator Robert Dole.
10. John Dillin, “Front-Runners Emerge in ’96 GOP Free-for-All,” CSM,
November 20, 1992, 2.
11. James A. Barnes, “Whoosh! There Goes Bob Dole!” NJ, February 2, 1993,
351.
12. See Ann Devroy, “Not ‘Fishing’—Just Testing New Hampshire Waters,”
WP, August 20, 1993, 8A.
13. AP, “Bob Dole’s Everywhere as GOP Point Man,” CD, January 2, 1994,
4A.
14. Bob Woodward, The Choice (New York: Simon & Schuster, 1996), 27-39.
15. Evan Thomas, et al., Back From the Dead: How Clinton Survived the
Republican Revolution (New York: The Atlantic Monthly Press, 1997), 48.
16. Buell, “The Invisible Primary,” 3 for a brief discussion of dropouts in the
1988 and 1992 invisible primaries.
17. Richard L. Berke, “Citing Financing, Quayle Drops Out of ’96 Campaign,”
NYT, February 9, 1995, A1.
18. AP, “Powell Finds Fault With GOP Leaders,” CD, September 18, 1995, 8A.
19. George Will, “Powell Couldn’t Run for President on Character Alone,”
CD, April 16, 1995, 3B and “If Powell Runs, He’ll Face Difficult Questions—Such
as These,” CD, October 29, 1995, 3B; R. W. Apple Jr., “Powell Stirs Waves on
Republican Right,” NYT, October 3, 1995, 11A; Berke, “Powell Sounds Out
Friends About a White House Run,” NYT, October 27, 1995, 8A; Berke, “Right
Tries to Mobilize Against Powell Candidacy,” NYT, November 2, 1995, 11A; and
Berke, “Powell Record is Criticized by Conservatives in GOP,” NYT, November 3,
1995, 9A.
20. Data from Gallup Organization Newsletter Archive, Vol. 60, October 7,
1995, 21; and William G. Mayer, The Presidential Nominations,” in The Election
of 1996, ed. Pomper (Chatham, NJ: Chatham House, 1997), 33.
21. Berke, “Poll Finds GOP Primary Voters Are Hardly Monolithic,” NYT,
October 30, 1995, 10A.
22. Dick Morris, Behind the Oval Office: Winning the Presidency in the
Nineties (New York: Random House, 1997), 156.
23. Darmouth College news releases on polls of October 1–4 and October
22–25, 1995, kindly supplied by Professor Linda Fowler. Iowa data come from
Mayer, “The Presidential Nominations,” in The Election of 1996 (Chatham, NJ:
Chatham House, 1997) 33.
24. Francis X. Clines, “Powell Rules Out ’96 Race; Cites Concerns for Family
and His Lack of A Calling,” NYT, November 9, 1995, 1A.
25. Berke, “GOP Candidates Feel The Gen. Powell Blues,” NYT, September 24,
1995, A18.
26. See James W. Ceaser and Andrew E. Busch, Upside Down and Inside Out:
The 1992 Elections and American Politics (Lanham, MD: Rowman & Littlefield,
1993). For an application of this scheme to the 1996 GOP candidates, see Ceaser
and Busch, Losing to Win: The 1996 Elections and American Politics (Lanham,
MD: Rowman & Littlefield, 1997), 57–87. Elsewhere Busch lists 16 outsider can-
didates since 1952, of which six sought the Democratic or Republican presidential
nomination while serving in public office. Since 1984, however, only one of the nine
outsiders listed by Busch held public office at the point he entered the presidential
nominating race. Busch, Outsiders And Openness in The Presidential Nominating
System (Pittsburgh, PA: University of Pittsburgh Press, 1997), 24–25.
27. B. Drummond Ayres Jr., “For Dornan, Keyes, and Taylor, Low Polls Can’t
Dash High Hopes,” NYT, December 30, 1995, 1A.
28. For example, see the symposium, “My Guy: Why My Presidential Candi-
date is Mr. Right,” Policy Review 75 (Summer 1995): 6–17.
29. See discussion in Buell, “The Invisible Primary,” 12.
30. “1996 State-by-State Expenditure Limits For Presidential Candidates,” FEC
document kindly provided the author by Robert Biersack.
31. Buell, “The Invisible Primary,” 14–16.
32. “Hard money” comparisons hardly constitute the whole picture of pre-
nomination campaign finance in 1996. See, for example, Mary Jacoby, “Dole’s PAC
Aided Presidential Bid,” RC, November 2, 1995, 1; AP, “Alexander Discloses Con-
tributors’ Names,” NYT, December 31, 1995, 10A; and Ruth Marcus and Charles
Babcock, “When is a Candidate Not a Candidate?” WP National Weekly Edition,
January 8–14, 1996, 18.
33. See Buell, “The Invisible Primary,” 11–16.
34. Anthony Corrado, “Financing the 1996 Elections,” in The Election of
1996, ed. Pomper (Chatham, NJ: Chatham House, 1997), 144–5.
35. Hadley, The Invisible Primary, 2.
36. Buell, “ The Invisible Primary,” 16–18.
37. Data from Gallup Organization Newsletter Archive, Vol. 60, October 7,
1995; February, June 30, August 17, September 14, October 7; also February 1,
1996 at www.gallup.com; data for November 1995, January 5–7, 1996, and Janu-
ary 12–15, 1996 from Table 1.3 of William G. Mayer, “The Presidential Nomina-
tions,” in The Election of 1996, ed. Pomper, 31–32.
38. Buell, “‘Locals’ and ‘Cosmopolitans’: National, Regional, and State News-
paper Coverage of the New Hampshire Primary,” in Media and Momentum: The
New Hampshire Primary and Nomination Politics ed. Gary R. Orren and Nelson
W. Polsby, (Chatham, NJ: Chatham House, 1987), 60–103; Buell, “Meeting Expec-
tations? Major Newspaper Coverage of Candidates During the 1988 Exhibition
Season,” in Nominating the President, ed. Buell and Lee Sigelman, 150–95; Buell,
“The Invisible Primary,” 25–28.
39. Tim Curran, “Handicapping the Three Senate Republicans Gearing Up to
Run for the White House in 1996,” RC, November 21, 1994, 10.
40. These data cover the period October 1, 1993, to February 6, 1996, and they
are based on the Vanderbilt Television News Abstracts. The early stage dates
through June 30, 1994, and the late stage begins on July 1, 1994.
41. See Buell, “The Invisible Primary,” and “The Invisible Primary Revisited,”
paper presented at the Southern Political Science Association meeting in Atlanta,
November 9, 1996, for more extensive comparisons of Republican candidate cov-
erage.
42. See Buell, “The ‘Invisible Primary’ Revisited” for more detailed compar-
isons of topical coverage. Not all of this coverage helped Dole, of course, since the
flood of stories about government shutdowns linked him to the unpopular Gin-
grich.
43. Buell, “The Invisible Primary,” 22–23.
44. Paul Taylor, “Freewheeling Iowa Straw Poll Even Has Out-of-State Voters,”
WP, August 19, 1995, 1A.
45. See Thomas, et al., Back From the Dead, 52. For more on behind-the-scenes
maneuvering at the Ames poll, see Woodward, The Choice, 240–7.
46. Taylor, “Dole or Gramm? Iowa GOP Says Yes,” WP, August 20, 1995, 9A.
47. Dan Balz, “Gramm Wins, Opponents Minimize Louisiana Presidential
Straw Poll,” WP, January 8, 1995, 4A.
48. Balz and Eric Pianin, “Gramm is Winner in South Carolina GOP Straw Bal-
lot,” WP, March 5, 1995, 7A.
49. Berke, “Surprising Straw Poll Gives Dole a Glimpse of the Battles Ahead,”
NYT, August 21, 1995, 1A.
50. Steve Campbell, “Straw Polls Aren’t Everything They’re Made Out to Be,”
Maine Sunday Telegram, October 29, 1995, 3C.
51. See, for example, Berke, “Surprising Straw Poll”; Charles Cook, “What the
Iowa Straw Poll Says, and What it Doesn’t Say,” RC, September 4, 1995, 8; Mor-
ton Kondracke, “Straw Polls Aside, Dole’s on Track for GOP Nod,” RC, Septem-
ber 4, 1996, 6; Thomas Edsall and Balz, “Straw Poll Winner Still Falls Short,” WP,
November 20, 1995, 4A.
52. Florida has the best overall record of straw-vote victors winning the state’s
presidential primary and eventually capturing the nomination. See Buell, “The
Invisible Primary,” 22–23.
53. William March, “Florida is Place to Be for GOP,” Tampa Tribune, July 16,
1995, 1A.
54. Michael Murphy, a key Alexander operative, later conceded that his cam-
paign had spent about half a million on Presidency III, “money we didn’t have” that
should been invested in New Hampshire. Remarks at a George Washington Uni-
versity symposium on the 1996 presidential election, April 20, 1996.
55. AP, “Dole Willing to be Another Reagan,” CD, July 16, 1995, 9A; Charles
Cook, “Rivals Chip Away at Granite State’s Primary Electorate,” RC, September
17, 1995, 6.
56. Berke, “Now Officially Dole Is Making a Run for ’96,” NYT, April 11,
1995, 1A; AP, “Dole Launches Presidential Bid, Vows No Tax Hike,” CD, April 11,
1995, 3A; “Bob Dole: Where He Stands,” from Dole campaign website.
57. Bernard Weinraub, “Violent Movies and Records Undercut Nation, Dole
Says,” NYT, May 31, 1995, 1A; Kevin Merida, “Dole Cites Murder in New Attack
on Hollywood,” WP, June 28, 1995, 4A.
58. Berke, “Gay Congressman of His Own Party Brings Fire on Dole,” NYT,
September 7, 1995, 1A; Berke, “Dole in Switch Says Aides Erred in Refunding Gay
Gift, “ NYT, October 18, 1995, 1A; Steven A. Holmes, “Reversal on Gay Dona-
tions Embroils Dole,” NYT, October 19, 1995, 11A.
59. Berke, “Christian Right Issues Threat to the GOP,” NYT, February 11,
1995, 1A; Berke “Politicians Woo Christian Group,” NYT, September 9, 1995, 1A;
Berke “Christian Coalition Ends Convention With A Dual Identity,” NYT, Sep-
tember 10, 1995, 11A.
60. Michael J. Wines, “Dole’s Response Gets a Response of Its Own, and It’s
Fairly Underwhelming,” NYT, January 25, 1996, 8A.
61. Gerald F. Seib, “Dole’s Twin Secrets: Old Style Organizing and a Newfound
Cool,” WSJ, March 11, 1996, 1A; Berke, “Wealthy Newcomer Gains His Rivals’
Attention,” NYT, January 13, 1996, 1A; and Berke, “Forbes, A Newcomer in Perot
Clothing,” NYT, January 15, 1996, 8A.
62. Mayer, “The Presidential Nominations,” 39, 40–42; AP, “Polls Show
Forbes Remains Close, But Must Count on Independents’ Turnout,” CD, January
30, 1996, 4A; Judy Keen, “A Two-Man Contest By Any Poll,” USA, January 30,
1996, 4A; Berke, “Wealthy Newcomer Gains His Rivals’ Attention.”
63. Berke, “Wealthy Newcomer”; Elizabeth Kolbert, “GOP Candidates Strug-
gle To Stand Out in Blur of Ads,” NYT, January 20, 1996, 8A; Martha Moore and
Judi Hasson, “Campaign ‘96: Dole Buying ‘Awesome Amount’ of Ads in Iowa,”
USA, February 7, 1996, 4A.
64. The flat tax plan as outlined by Forbes in his announcement speech of Sep-
tember 22, 1995, called for abolishing the federal income tax in favor of a 17 per-
cent rate across the board. Certain exemptions provided that the head of a family
of four would pay no taxes whatever on the first $36,000 of income. No taxes
would be collected on Social Security, pensions, personal savings, or capital gains.
“A New Conservative Vision,” speech by Steve Forbes at the National Press Club
in Washington, D.C., obtained from the Forbes campaign.
65. David E. Rosenbaum, “Panel Calls for a Flat Tax, But Fails to Specify the
Rate,” NYT, January 18, 1996, 10A; Katharine M. Seelye, “Dole, in Iowa,
Unleashes Tough New Attacks,” NYT January 26, 1996, 10A; Rosenbaum, “For
Dole and Forbes, the Fight Is on the Air,” NYT, February 7, 1996, 10A; Moore and
Hasson, “Dole Buying ‘Awesome Amount’ of Ads.”
66. Holmes, “Courtship of Iowans Intensifies,” NYT, January 25, 1996, 8A;
AP, “Dole Turns Up Heat on Forbes as Early Primaries Approach,” NYT, January
28, 1996, 8A.
67. Ernest Tollerson, “Forbes Steps Up Defense Against Attack Ads,” NYT,
February 10, 1996, 8A; and Larry J. Sabato, ed., Toward the Millennium: The Elec-
tions of 1996 (Boston: Allyn and Bacon, 1997), 74–76.
68. Elizabeth Kolbert, “Pointing Up Contrast, Lugar Points to His Campaign
Ads,” NYT, February 12, 1996, 9A.
69. Mayer, tables 1.4 and 1.5 in “The Presidential Nominations”; Roger K.
Lowe, “Dole Opens Final Push for Iowa Hearts, Votes,” CD, February 11, 1996,
1A; Berke, “Wealthy Newcomer Gains his Rivals’ Attention”; AP, “New Poll Shows
Forbes Gaining in California,” NYT, February 3, 1996, 8A.
70. Kevin Sack, “In Louisiana, Church Pews Are Trenches in GOP War,” NYT,
January 24, 1996, 12A; Berke, “Who Knew? Louisiana Is Beginning to Choose,”
NYT, February 6, 1996, 10A.
71. AP, “Buchanan Deals Gramm Sharp Blow,” CD, February 7, 1996, 1A; AP,
“Buchanan Gets Overwhelming Support From Religious Right,” CD, February 7,
1996, 2A; Keen, “Stunning Setback for Gramm,” USA, February 7, 1996, 4A;
Apple, “Louisiana Makes it Harder on Gramm,” NYT, February 8, 1996, 13A;
James Bennet, “Disappointed Gramm Says Top 3 Iowa Finish Is a Must,” NYT,
February 8, 1996, 13A.
72. See Keen, “Buchanan, Alexander Show Power for N.H.,” USA, February
13, 1996, 1A for projected turnout; Table 2.1 of Harold W. Stanley, “The Nomi-
nations: Republican Doldrums, Democratic Revival,” in The Elections of 1996, ed.
Michael Nelson, (Washington, DC: CQ Press, 1997), 22. Also see Darrell M. West,
Air Wars: Television Adverting in Election Campaigns, 1952–1996, 2nd ed. (Wash-
ington, DC: CQ Press, 1997), 63–64.
73. Although no delegates would be officially allocated until June, when the
other steps in Iowa’s caucus-convention process had been completed, the AP
released a preliminary count, in which Dole received 8, Buchanan 6, and Alexan-
der 5 out of 25 total. “Iowa Delegates, National Total,” USA, February 13, 1996,
3A.
74. Berke, “Dole Tops the Field in Iowa Caucuses,” NYT, February 13, 1996,
1A; Kolbert, “Campaign Spending Per Vote: Who Got His Money’s Worth,” NYT,
February 14, 1996, 10A. According to this tabulation the per-vote cost for Gramm
was $122, Alexander $47, Dole $35, and Buchanan $27.
75. The Iowa VNS poll is available at www.cnn.com/ALLPOLITICS/politics/
1996/polls.
76. Apple, “Dole Ends the Race in Iowa As a Scalded Front-Runner,” NYT,
77. Buell, “The Changing Face of The New Hampshire Primary,” in In Pursuit
of the White House 2000, ed. Mayer (New York: Chatham House/Seven Bridges,
1999), 89.
78. Woodward, The Choice, 385–6; Lowe, “Dole Stays Bright as Polls Dim,”
CD, February 16, 1996, 1A; Sabato, Toward The Millennium, 43–45; Robert D.
Novak, “Dole Can’t Blame Anyone But Himself,” CD, February 22, 1996, 11A;
Murphy presentation at George Washington University.
79. Rhodes Cook, “GOP Faces Uncharted Terrain In Wake of Buchanan
Upset,” CQWR, February 24, 1996, 438–442; updated with official returns as pub-
lished in “1996 Republican Primary Results,” CQWR, ugust 3, 1996, 63.
80. The New Hampshire VNS poll is available at www.cnn.com/ALLPOLI-
TICS/politics/1996/polls.
81. Richard Wolf, “GOP Steps In to Help Dole,” USA, February 22, 1996, 1A;
Berke, “Buchanan Victory Stirs Opposition Within the GOP,” NYT, February 22,
1996, 1A.
82. “Buchanan Victory Speech—New Hampshire,” text obtained from
Buchanan campaign.
83. AP, “Buchanan Edges Dole,” CD, February 21, 1996, 1A; Steve Lee Myers,
“Buchanan’s Policies Assailed by Giuliani as Peril to U.S.,” NYT, February 22,
1996, 12A; Wolf, “GOP Steps in To Help Dole;” Will, “Republican’s Mission:
Thwart Buchanan,” CD, February 22, 1996, 11A; Berke, “Buchanan Victory Stirs
Opposition Within the GOP,” NYT, February 22, 1996, 1A; Robin Toner, “Radio
Host Fears for Conservatism’s Fate,” NYT, February 23, 1996, 13A; Ayres, “A
Rare Chance for State Republicans in Tuesday’s Vote,” NYT, February 25, 1996,
13A; Bennet, “D’Amato Attacks Vigorously On ‘Extremism’ of Buchanan,” NYT,
March 3, 1996, 11A.
84. Keen, “Dole Vows Fight for GOP ‘Heart and Soul,’” USA, February 22,
1996, 6A.
85. Seelye, “A Vow to Unleash ‘the Real Bob Dole,’” NYT, February 22, 1996,
10A; Seelye, “Dole Adopts a New Persona: Savior of the Grand Old Party,” NYT,
86. Seelye, “Dole Adopts A New Persona.”
87. Apple, “Delaware Vote to Test Forbes’s Viability,” NYT, February 23,
1996, 12A. For more on the New Hampshire primary, see Buell, “The Changing
Face of the New Hampshire Primary,” 88–143.
88. Bill Carter, “3 Networks Admit Error in Arizona Race Reports,” NYT, Feb-
ruary 29, 1996, 9A.
89. Hasson, “GOP Race Deeper in Disarray,” USA, February 28, 1996, 4A;
Jerry Gray, “Dole Counts On a Surge Of Support in the South,” NYT, February 28,
1996, 11A.
90. Richard Benedetto, “Dole Capturing Poll Percentages, Unlike Primaries,”
USA, February 27, 1996, 2A.
91. AP, “Buchanan Says Arizona Race is Key,” CD, February 27, 1996, 1A;
“Campaign ’96: Late Arizona Votes Dull Buchanan’s Enthusiasm,” USA, February
28, 1996, 4A.
92. Jessica Lee, “Alexander Aide Lives on Bottom Line,” USA, February 27,
1996, 4A; Stephen Labaton, “When Big Money Fails to Win Some Delegates,”
NYT, March 1, 1996, 11A.
93. The FECA makes candidates for a major party nominations ineligible to
receive matching funds for 30 days. Failure to attain 10 percent within the 30 days
terminates matching funds unless a candidate requalifies by winning 20 percent in
a subsequent primary, in which case they receive the matching funds retroactively.
“Campaign ’96: Alexander’s Goals,” USA, February 29, 1996, 3A.
94. Keen, “Scrambling Starts for Primary-Heavy 2 Weeks,” USA, February 23,
1996, 4A; Rachel L. Swarns, “Southerner Places His Hopes Down Home,” NYT,
February 27, 1996, 13A.
95. Elaine Sciolino, “Forbes Offers View of His Foreign Policy Stands,” NYT,
March 1, 1996, 11A.
96. The analysis of South Carolina exit poll findings in this chapter relies on
information obtained from the Hotline by Institute of Governmental Studies librar-
ian Terry Dean.
97. Results of primaries are available at www.fec.gov.pubrec/presprim.htm.
98. Jonathan D. Salant, “One Man Corners the Ballot in N.Y.,” CQWR,
March 9, 1996, 646. The remaining nine New York delegates were slated in June.
99. Bennet, “With Cursory Nod to Dole, Buchanan Backs Off Race,” NYT,
April 18, 1996, 12A.
100. Elizabeth Drew, On the Edge: The Clinton Presidency (New York: Simon
& Schuster, 1994); Woodward, The Agenda: Inside the Clinton White House (New
York: Simon & Schuster, 1994).
101. Robert B. Reich, Locked in the Cabinet (New York: Vintage Books, 1998),
168.
102. Theda Skocpol maintains that Clinton’s health care plan opened the
Democrats to telling if inaccurate Republican attacks. See Boomerang: Health Care
Reform and the Turn Against Government (New York: W.W. Norton, 1997). For a
somewhat different view of the effectiveness of these attacks on public opinion, see
West and Burdett A. Loomis, The Sound of Money: How Political Interests Get
What They Want (New York: W.W. Norton, 1999.)
103. Paul R. Abramson, John H. Aldrich, and David W. Rohde, Change and
Continuity in the 1992 Elections, rev. ed. (Washington, DC: CQ Press, 1995),
323–25.
104. Jack Newfield, “It’s Time to Dump Bill,” New York Post, November 21,
1994, 1A; Howard Kurtz, “Talk Grows of Dumping Clinton ’96 Ticket,” CD,
November 25, 1994, 4A; Apple, “Clinton’s Grip on ’96 Ticket Not So Sure,” NYT,
November 20, 1994, 1A; Scripps Howard News Service, “Democrats Support
Challenge to Clinton,” CD, December 8, 1994, 4A.
105. “Democrats Support Challenge to Clinton.”
106. Dick Morris immodestly claims the credit for this strategy in Behind the
Oval Office; see also Thomas, et al., Back From the Dead, and Walter Dean Burn-
ham, “Bill Clinton: Riding the Tiger,” in The Election of 1996, ed. Pomper, 13.
107. AP, “Ex-Governor Explores a Bid Against Clinton,” NYT, March 25, 1995,
8A; David Yepsen, “Casey Seen as No Threat to Clinton,” DMR, April 1, 1995,
5M; Catherine Manegold, “Ex-Governor Ends ’96 Challenge to Clinton,” NYT,
April 19, 1995, 10A.
108. See Woodward, The Choice, 206–10.
109. Corrado, “Financing the 1996 Election,” 143.
110. Morris, Behind the Oval Office, 291–305; Ceaser and Busch, Losing to
Win, 97–100; Drew, Whatever It Takes (New York: Viking, 1997), 97–100.
111. ”1996 Democratic Primary Results,” CQWR, August 17, 1996, 79–80.
Clinton did not enter non-binding primaries in North Dakota and Michigan.
112. Ruth Marcus, “Dole Campaign Spent $3.5 Million in March,” WP, April
20, 1996, 9A.
113. Moore, “Dems’ Barrage of Ads Gets Little GOP Response,” USA, April 18,
1996, 7A; Labaton, “Dole Campaign Nears Spending Limit for Primaries,” NYT,
March 22, 1996, 11A; Jane Fritsch, “In Dole’s Race, Party’s Money Now a Life-
line,” NYT, May 22, 1996, 1A.
114. Brooks Jackson, “Financing the 1996 Campaign: The Law of the Jungle,”
in Toward the Millennium, 237–8; Corrado, “Financing the 1996 Election,” 148.
115. Kolbert, “Gingrich and Dole Aides Try New Unified Party Message,” NYT,
March 20, 1996, 1A; David S. Cloud, “Tough Campaign Challenges For the Sen-
ate Leader,” CQWR, March 30, 1996, 861; Helen Dewar, “Striving to Lead, Dole
Stumbles Into a Week of Setbacks,” WP, April 21, 1996, 6A; Thomas, et al., Back
From the Dead, 74; Jackie Koszczuk, “Dole Leaves Senate Behind To Hit Cam-
paign Trail,” CQWR, May 18, 1996, 1359.
116. Drew, Whatever It Takes, 90; Woodward, The Choice, 421–8; Thomas, et
al., Back From the Dead, 72–78.
117. Koszczuk, “Dole Leaves Senate Behind,” 1360.
118. Thomas, et al., Back From the Dead, 106–7; Fritsch, “Democrats as Well
as GOP Profit from Tobacco,” NYT, July 6, 1996, 1A.
119. Keen and Gary Fields, “Dole Says He Was ‘Set Up’ by NAACP,” USA, July
12–14, 1996, 1A.
120. Bennet, “Abortion Foes Warn Dole Not to Shift On Platform,” NYT,
March 17, 1996, 17A.
121. Bennet, “Top Conservative Would Back Shift on Abortion Issue,” NYT,
May 4, 1996, 1A: Bennet, “Leader of Christian Coalition Denies Shifting on Abor-
tion,” NYT, May 5, 1996, 1A.
122. Clines, “Abortion-Rights Supporters Fight for Their Say in GOP,” NYT,
May 6, 1996, 1A.
123. “Text of Dole’s Statement On His Abortion Stance,” (Provided by Dole
campaign office) NYT, June 7, 1996, 10A.
124. Berke, “Dole, Ignoring His Advisers, Lashes Out at Abortion Foe,” NYT,
June 12, 1996, 1A.
125. Berke, “Battle Over the Abortion Plank Jolts the Republicans Yet Again,”
NYT, June 14, 1996, 1A; Berke, “In Many States, Abortion Feud Splits GOP,”
NYT, June 20,1996, 1A.
126. Restoring the American Dream: The Republican Platform of 1996 (Wash-
ington, DC: Republican National Committee, 1996), 34–35. Seelye, “Moderates in
GOP Vow Fight on Platform Abortion Language,” NYT, August 7, 1996, 1A;
Rosenbaum, “Accord Satisfies GOP Moderates on Abortion Issue,” NYT, August
8, 1996, 1A. The general statement of tolerance declared: “While our party remains
steadfast in its commitment to advancing its historic principles and ideals, we also
recognize that members of our party have deeply held and sometimes differing
views. We view this diversity of views as a source of strength, not as a sign of weak-
ness, and we welcome to our ranks all Americans who may hold differing positions.
We are committed to resolving our differences in a spirit of civility, hope, and mutu-
al respect.” Restoring the American Dream, 32
127. David E. Sanger, “Recycled Ideas Echo Forbes and Buchanan,” NYT,
March 25, 1996, 13A.
128. Adam Nagourney, “Dole to Advocate 15% Cut in Taxes, His Campaign
Says,” NYT, August 5, 1996, 1A.
129. Seelye, “Dole Offers Economic Plan Calling for Broad Tax Cut Aimed at
Spurring Growth,” NYT, August 6, 1996, 1A.
130. Larry David Smith and Dan Nimmo, Cordial Concurrence: Orchestrating
National Party Conventions in the Telepolitical Age (Westport, CT: Praeger, 1991).
131. Byron E. Shafer, Bifurcated Politics: Evolution and Reform in the National
Party Convention (Cambridge, MA: Harvard University Press, 1988).
132. Today’s Democratic Party: Meeting America’s Challenges, Protecting Amer-
ica’s Values: The 1996 Democratic National Platform (Washington, DC: Democra-
tic National Committee, 1997), 1.
133. For a review of the landmark literature and a detailed comparison of 1988
convention delegates, see Buell and John S. Jackson III, “The National Conven-
tions: Diminished but Still Important in a Primary-Dominated Process,” in Nomi-
nating the President, 228–35.
134. Random Washington Post-ABC News Telephone polls of 505 Republican
delegates, 508 Democratic delegates, and 1,514 adults nationwide, as described in
national convention supplements to the WP National Weekly Edition, August
12–18 and August 26–September 1, 1996.
135. Berke, “Perot Calls Meeting for ‘96 Contenders to Address Issues,” NYT,
June 3, 1996, 1A; Ayres, “Perot Leaves Door Open for ‘96 Presidential Run,” NYT,
August 14, 1995, 6A; Seib, “Perot’s Followers Remain Frustrated, Directionless
After Weekend of Wooing,” WSJ, August 14, 1995, 14A.
136. AP, “Perot Changes Mind, Launches Third Party,” CD, September 26,
1995, 2A.
137. Seelye, “In Quest for a Third Party, First Hurdle Is the Highest,” NYT, Sep-
tember 28, 1995, 11A; Ayres, “Perot Claims Victory in Effort to Qualify Party in
California,” NYT, October 25, 1995, 12A; Tollerson, “Perot’s Party Gains 2 States,
New York and Arizona, in Its Campaign for Presidential Ballot,” NYT, June 28,
1996, 11A; Sam Howe Verhovek, “Perot is in the Contest for President, Unless,”
NYT, March 19, 1996, 12A; author’s interview with Richard Winger of Ballot
Access News, September 9, 1997.
138. Verhovek, “Perot as a Political Presence: 1992 All Over Again?” NYT, Jan-
uary 23, 1996, 6A; Mini Hall, “Perot: I’d ‘Give It Everything’ If Asked to Run,”
USA, March 20, 1996, 4A; Nagourney, “When Perot Talks, It’s Like a Campaign,”
NYT, March 25, 1996, 17A.
139. Sabato, Toward the Millennium, 85–8; Berke, “Perot’s ‘96 Strategy in Set-
ting Up Third Party Is Called Masterful, Even by Detractors,” NYT, September 27,
1995, 1A; Tollerson, “A Third Party in the Wings Waits for a Leader to Arrive,”
NYT, June 3, 1996, 12A; Hall, “Lamm Edges Closer to ‘96 Race,” USA, June 11,
1996, 7A; Hall, “Experts Put Perot Atop Reform Ticket,” USA, June 24, 1996, 8A;
Hilary Stout, “Lamm to Seek Presidential Bid of Perot’s Party,” WSJ, July 10, 1996,
18A; Tollerson, “Lamm, Ex-Governor of Colorado, Seeks Reform Party’s Nomina-

tion,” NYT, July 10, 1996, 1A.
140. Hall, “Perot Sets Up Fight for Reform Party Nomination,” USA, July 11,
1996, 3A; Berke, “Perot Declares He Will Seek His Party’s Presidential Nod,” NYT,
July 12, 1996, 1A.
141. Tollerson, “Reform Party Awaits the Mail to Determine Who the Candi-
dates Will Be,” NYT, July 9, 1996, 9A; “Candidate Picked by Mail Survey,” USA,
July 12–14, 1996, 6A.
142. Ballot Access News, August 12, 1996, 6.
143. James Brooke, “Perot Rival Protests Foul-Ups in Party’s Nationwide Sur-
vey,” NYT, August 2, 1996, 12A; and Ayres, “Reform Party’s Split Is Widening as
Members Head to Convention,” NYT, August 10, 1996, 1A.
144. Ballot Access News, September 9, 1996, 5.
145. David Moore, “Perot Candidacy Hurt By Major Party Conventions,” The
Gallup Poll Monthly, 372 (September 1996), 4–5; Thomas, et al., Back From the
Dead, 159–60.
146. The 1996 GOP convention approved a plan to reward late-voting states
with bonus delegates. States holding their primaries between March 15 and April
14, 2000 get a 5-percent increase in the size of their national convention delegation.
The bonus for states voting between April 15 and May 14 is 7.5 percent. States
selecting their delegates on or after May 15 get a 10-percent boost. Alan Greenblatt
and Rhodes Cook, “Nominating Process Rules Change,” CQWR, August 17,
1996, 2299. According to the “Preliminary 2000 Delegate Allocation” breakdown
released to the author on July 22, 1999, American Samoa and 20 late-voting states
received a total of 56 bonus delegates.
147. Poll data accessed from http://www.cloakroom.com/members/polltrack, the
Hotline.
148. Salant, “White House Hopefuls Report New Fund-Raising Numbers,” July
15, 1999, http://www.cnn.com/allpolitics; Federal Election Commission, “Financial
Activity of 1999–2000 Presidential Campaigns Through June 30, 1999,”
http://www.fec.gov/finance/prsq.
149. Don Van Natta Jr., “Bush Foregoes Federal Funds and Has No Spending
Limit,” NYT, July 16, 1999, A1.
150. Neil A. Lewis, “Flush Bush Turns to Soft Money,” NYT, July 17, 1999, A9;
Van Natta, “Democrats Are Trying to Double Unregulated Donations for 2000,”
NYT, July 25, 1999, A1.
151. “Forbes Leading Spending Race,” July 20, 1999, www.cnn.com/allpolitics.
152. Tollerson, “Breather for Reform Party Is Giving Way to Discord,” NYT,
November 24, 1996, 20A; Bill Dedman, “Perot Dissidents Form A New Party,”
NYT, October 5, 1997, 17A.
153. AP, “With Ventura as Their Star, Reform Party Gathers to Chart Its
Future,” July 23, 1999, www.cnn.com/allpolitics; AP, “Ventura Again Says He Will
Not Seek Presidency in 2000,” NYT, July 25, 1999, A17; “Perot Doesn’t Hint of
2000 Plans in Speech to Reform Leaders,” July 24, 1999, www.cnn.com/allpolitics.
CHAPTER 2
Congressional Nominations in 1996

Procedures, Candidates, and Electoral Patterns
HARVEY L. SCHANTZ
Congressional election returns consistently underscore a central character-

istic of United States elections and the Congress: two-party hegemony. Vir-
tually all members of Congress are elected as Democrats or Republicans.
Thus, although it is not called for in the Constitution, major party nomi-
nations are the crucial first step in congressional selection. This chapter is
an examination of the congressional nomination process of 1996, with an
emphasis on procedures, candidates, and electoral patterns in primaries.
CONGRESSIONAL NOMINATING METHODS

Popularly elected government has long called into existence the prior nom-
ination of candidates.1 In the United States, the predominant method of
nomination for offices other than the presidency changed from party con-
ventions to direct primaries in the early years of the twentieth century.2 The
direct primary opened up the selection of candidates to more people, for it
places the formal selection of party candidates in the hands of the elec-
torate rather than in those of party officials or members.
The 1996 congressional nominations followed the pattern of most of the
twentieth century, as the overwhelming number of U.S. House and Senate
3
party nominations were made in direct primaries (tables 2.1 and 2.2). One
major difference between House and Senate primaries, though, was that
most House primaries were not contested whereas most Senate primaries
were contested. There were also a few nominations decided by party con-
ventions, party committees, and write-ins. The Democrats and Republicans
nominated candidates for all of the Senate seats, but both parties failed to
nominate in a few House districts.
41
42 Congressional Nominations in 1996
Table 2.1: Nominating Methods for the U.S. House, 1996

Nominating System Democrats Republicans
N % N %
Primary Electiona
Contested 151 35.5 156 36.5
Uncontested 235 55.3 232 54.3
386 90.8 388 90.9
Nonprimary Nominationsa
Party Convention 19 4.5 17 4.0
Party Committee 1 0.2 3 0.7
Write-In 2 0.5 0 0.0
Self-Nominated 17 4.0 19 4.4
39 9.2 39 9.1
Total Nominations 425 100.0 427 100.0
No Candidateb 10 2.3 8 1.8
House Elections 435 100.0 435 100.0
a
These are the number and percentage of total nominations made by designated method.
b
These are the number and percentage of House elections in which designated party did not
offer a candidate.
Table 2.2: Nominating Methods for the U.S. Senate, 1996

Nominating System Democrats Republicans
N % N %
Primary Electiona
Contested 20 60.6 24 72.7
Uncontested 11 33.3 7 21.2
31 93.9 31 93.9
Nonprimary Nominationsa
Party Convention 1 3.0 1 3.0
Party Committee 0 0.0 0 0.0
Self-Nominated 1 3.0 1 3.0
Total Nominations 33 100.0 33 100.0

No Candidateb 0 0.0 0 0.0
Senate Electionsc 33 100.0 33 100.0
a
These are the number and percentage of total nominations made by designated method.
b
These are the numbers and percentage of Senate elections in which designated party did not
offer a candidate.
Excludes twin primaries for short-term Kansas seat.
c
CONTEMPORARY PRIMARY SYSTEMS

Primary systems—which are regulated by state laws—vary along a number
of dimensions. Three of the most crucial are party organizational control
over the ballot, voter eligibility, and margin needed for victory.
Party and Ballot Access
In seven states the party organizations are granted a measure of legal con-
trol over primary ballot access, which is exercised through a preprimary
convention.4 In Utah and Connecticut a threshold of support at the party
convention is necessary for inclusion on the primary ballot. In Utah party
conventions nominate two candidates for the primary ballot, but there is
automatic nomination of a candidate receiving 70 percent of the delegate
support at the Republican or 60 percent support at the Democratic prepri-
mary convention. In Utah’s three congressional districts in 1996, there were
three primaries and three convention nominations. In the second district
there were two primaries for the seat of retiring Republican Enid Greene.
Republican Merrill Cook, a perennial candidate, placed second at the party
convention, but won the primary and general election.
In Connecticut nominations are made by party conventions with the
provision that any unsuccessful candidate receiving 15 percent (recently
reduced from 20 percent) or more of the delegate support on any conven-
tion ballot is eligible to file for a “challenge” primary. There was one chal-
lenge primary in 1996, and in this contest the endorsed candidate won. In
both Connecticut and Utah there was not a Senate seat up for election in
1996.
The congressional nominating systems of Colorado and New Mexico, as
well as the senatorial nominating system of New York, grant party organi-
zations less control over candidate access to the primary ballot. In New
Mexico candidates need 20 percent, in New York 25 percent, and in Col-
orado 30 percent (recently changed from 20 percent) of the preprimary
convention or party committee support for automatic inclusion on the pri-
mary ballot. However, candidates not receiving this minimum level of sup-
port may appear on the primary ballot by fulfilling a petition requirement.
In these systems, as well, the endorsed candidate does not always win the
subsequent primary. In 1996, for example, Gene Nichol won the Colorado
Democratic state assembly with 53 percent, but he lost the U.S. senatorial
primary to Tom Strickland, who had received 45 percent support at the
state assembly. Meanwhile, in the fourth district, Don Ament narrowly
won the party endorsement, and with it the top spot on the ballot, but he
lost the Republican primary to Bob Schaffer.
In North Dakota and Rhode Island the political parties legally endorse
congressional candidates but are not empowered to exclude others from
the ballot. “During the 1990s,” in Rhode Island, according to Maureen
Moakley, “in a series of high visibility races for governor and federal office,
unendorsed candidates trounced their party-backed opponents.”5 One such

race took place in the second district in 1996, where Democratic Lt. Gov-
ernor Bob Weygand defeated the endorsed candidate in the primary and
won the U.S. House seat in November.
In addition to these seven states, in a few other states, most notably
Minnesota, the political parties provide “extra-legal” endorsements.6 In
1996 though, Republicans declined to endorse a senatorial candidate
because their convention was deadlocked after 14 ballots.
The direct primary has lessened the grip of party organizations over the
nomination process. The preprimary nominating or endorsing convention
is an attempt to insure some organizational control over the primary. But
as we have seen, in 1996, at least, the preprimary endorsing convention is
an unsure method of maintaining party organizational control over the
nomination.
Voter Eligibility
In an open primary system registered voters may participate in the party
primary of their choice. In states with a closed primary voters may only
participate in the primary of the party to which they are registered. In
1996, 21 states were open primary states.7 Sixteen states had closed pri-
maries. Another nine states were modified-closed states, allowing nonaffil-
iated voters to choose the party primary of their choice. Additionally, in
West Virginia the Democratic primary was closed, but the Republican pri-
mary was open to nonaffiliated voters. According to the Supreme Court,
states cannot prevent state parties from opening their primaries to nonaf-
filiated voters; such a restriction was deemed by the Court an infringement
upon the associational rights of the party and its members.8 Three remain-
ing states, discussed below, used a unified primary system in 1996.
Another crucial dimension is the period of time before primary day that
a voter must be registered with a political party in order to be eligible for
the party primary. In twenty-five states, including four closed or modified-
closed states, voters may decide the party primary in which they wish to
participate as late as primary day.9 In seven states the deadline for switch-
ing partisan affiliation is 21 days or less before the primary. This deadline
is about a month in nine states. There are more extended deadlines in five
states, ranging from about three to six months. And, finally, New York
State requires a voter to have changed partisan affiliation 30 days before
the previous general election—a period of about eleven months before the
September congressional primary.
The Supreme Court examined lead times in two 1973 cases.10 In Rosario
v. Rockefeller, the Supreme Court allowed the extended New York State
registration deadline, reasoning that this law does not prevent voters from
participating in primaries, even in years in which they switch party affilia-
tion. However, in Kusper v. Pontikes, the Supreme Court voided an Illinois
law which prevented a voter from participating in a party primary within
23 months of voting in another party’s primary. The Court felt that the Illi-
nois procedure limited a voter’s right of political association by preventing
participation in a party primary.
Party organizations tend to view the open primary with trepidation,
fearing loss of control and pointing to the possibility of “cross-over voting”
by nonpartisans in an effort to select a weak candidate for the general elec-
tion. The U.S. Supreme Court saw these concerns of parties as legitimate in
Rosario.
The Unified Primary States
Alaska, Washington, and (since 1998) California operate under a blanket
primary system, in which voters may switch back and forth from party to
party to select their favorite nominee for each office. Nominations are
granted for each office to the leading candidate from each party.
In recent years the Republican party of Alaska has resisted the blanket
primary, arguing that, in light of Tashjian v. Republican Party of Con-
necticut, a political party has the right to decide whether or not to allow
independents into its primary. But in August 1996 Alaska held a blanket
primary, as the Alaska Supreme Court and the U.S. Supreme Court upheld
this system.11
In March 1996, the unified or consolidated primary system was adopt-
ed in California when voters approved Proposition 198 and this system
was used for the 1998 congressional primaries. This change was opposed
by the state’s Democratic and Republican party leaders.12 In 1996 Califor-
nia operated under a fairly closed primary, not allowing voters to change
party registration within 29 days of the primary.
Louisiana since 1978 has operated with a unified primary. In Louisiana
all candidates for an office are on one ballot. If a candidate receives a
majority of the votes cast, that candidate is declared the winner and a gen-
eral election is not held. If no candidate receives a majority, the two lead-
ers—regardless of party affiliation—compete in the general election ballot.
The law, strongly advocated by Governor Edwin Edwards (D.,
1973–1980), was intended to prevent general election contests between
Democratic candidates who had survived a difficult primary and run-off
and Republicans nominated without opposition.13
In 1996, six of seven Louisiana congressional seats were won in the pri-
mary. The one runoff, in congressional district 7, was between two Democ-
rats. This pattern is consistent with earlier years. Between 1978 and
1994—nine congressional election years—70 regular House elections took
place in Louisiana. Of these 70, 60 were decided in the primary. There were
ten runoffs on election day: five were Democrats versus Republicans, four
were between two Democrats, and one involved two Republicans.
In 1996, a Louisiana U.S. Senate seat was decided in a runoff on elec-
tion day, as Democrat Mary Landrieu beat Republican Louis “Woody”
Jenkins, who had led in the first primary. However, only one of six U.S.
senate races between 1978 and 1992 was decided on election day: a 1986
runoff won by a Democrat over a Republican. The other five contests were
won with a majority of the vote in the unified primary held in September.
On December 2, 1997, the U.S. Supreme Court struck down the
Louisiana statute that set up the unified primary system. The Court’s deci-
sion in Foster v. Love upheld a federal appeals court, finding that due to
the lack of election day runoffs, the Louisiana system is contrary to an
1872 federal law which set a national election day for members of Con-
gress.14 In 1998, Louisiana held its congressional primary on election day,
along with provision for a (not needed) later runoff.
Margin Needed for Victory
In the United States, general and primary elections usually require a simple
plurality for victory. Two of the exceptions are southern primaries and pri-
maries in Iowa and South Dakota. In Iowa and South Dakota a candidate
must receive at least 35 percent of the total primary vote in order to win
party nomination. In Iowa, a party nominating convention is held to select
a nominee if no candidate garnered the required minimum.15 In South
Dakota a runoff primary is held between the two leading candidates. These
infrequently employed procedures were not used in 1996.
Seven southern states and one border state—Alabama, Arkansas, Flori-
da, Georgia, Mississippi, Oklahoma, South Carolina, and Texas—current-
ly require a majority for victory in primaries.16 A ninth state, North Car-
olina, since 1989 has required 40 percent of the total vote, rather than 50
percent. If no candidate receives at least one-half (or 40 percent in North
Carolina) of the votes cast in the primary, a runoff primary is held, if the
second candidate files, four weeks or so later. These states adopted the
runoff between 1902 and 1939. V. O. Key Jr. and Cortez A. M. Ewing, in
their era two of the closest students of southern politics, pointed to the
ideal of majority rule rather than factional or personal advantage as the
prime motivation for adoption of the runoff.17 Democratic hegemony in the
South required a runoff to insure a form of majority decision because of the
frequent splintering of the vote in the southern Democratic primaries in
these years.
In 1996 U.S. House contests, there were 15 runoff primaries held for
Democratic nominations and 12 runoff primaries for Republican nomina-
tions. There was a reversal in ten of these runoffs. In U.S. House primaries
from 1970 to 1986, the primary leader lost 28.6 percent of the subsequent
runoffs.18
Only one incumbent in 1996 was involved in a runoff: Texas Republi-
can Greg Laughlin, who led in the primary but lost the runoff primary.
Incumbents are rarely in runoffs, but incumbents generally run greater risks
in a runoff than in the first primary because opposition is no longer frag-
mented between more than one candidate.19
Runoff primaries are sometimes criticized for serving as an obstacle to

women and minority candidates. In 1996, runoff primaries placed a heavy
burden upon female candidates. No women were nominated in runoff pri-
maries, and four Democratic female candidates and one Republican female
candidate lost the runoff to a male candidate after leading in the initial pri-
mary. From 1970 to 1986, five of eight women who led in the House pri-
mary won the runoff.20 It is difficult to identify the race of all congression-
al candidates, but at least one runoff resulted in the elimination of a minor-
ity candidate—a native American woman—by a white male in the seventh
district of North Carolina Democratic runoff.
In Senate contests there were five runoff primaries held to determine
party nominees: Alabama (both parties); Arkansas (Democrats); Georgia
(Republicans); and Texas (Democrats). In all five of these nominations, the
winner of the first primary was victorious in the runoff. All of these runoffs
involved two white males, except for the Texas contest where a Hispanic
male led in the primary and won the runoff.
NONPRIMARY NOMINATIONS
In every election year a number of congressional nominations by both
major political parties are made by means other than the direct primary. In
1996, four Senate and 78 House major party candidates were nominated
by nonprimary methods, including self-nomination in unified primaries in
Texas and Louisiana (tables 2.1 and 2.2).
Convention Nominations
By the second decade of the twentieth century, nominating conventions had
given way to the direct primary as the most common way of nominating
party candidates. In a few states and among southern Republicans, though,
the party convention persisted.21 However, the party convention was aban-
doned by southern Republicans during the 1960s and it was forsaken for
statewide nominations by New York, Delaware, and Indiana in 1968,
1970, and 1974, respectively.
Today, “pure” convention nominations for Congress are made only in
Virginia. In Virginia, responsibility for the selection of a nominating
method is shared by party committees and elected officials. District and
state party officials have, for House and Senate nominations respectively,
the option of choosing a direct primary or other nominating method. How-
ever, an incumbent office holder nominated by, or filing papers for, a direct
primary in the previous election year may veto the selection of a nominat-
ing method other than a primary. In 1996, all but two party nominations
for the U.S. House and Senate, following the recent pattern, were made by
convention. The two exceptions were Republican primaries held to renom-
inate Senator John Warner and Representative Herbert H. Bateman. Both
legislators chose a primary in order to avoid what promised to be difficult
Republican conventions dominated by delegates more conservative than

themselves. Both Warner and Bateman easily won their primaries and gen-
eral elections.22
In Connecticut all six Democratic House candidates and five Republi-
cans were nominated by party convention. In Utah three of the six con-
gressional nominations, including the nominations of both running incum-
bents, were by convention.
In Arkansas there were three party conventions held to nominate sub-
stitutes for candidates previously nominated by primary. On June 15,
Arkansas Republicans nominated Representative Tim Hutchinson for the
U.S. Senate, in place of Lt. Governor Mike Huckabee, who had resigned
the nomination in anticipation of becoming governor. In November, Tim
Hutchinson became the first Republican Senator from Arkansas since
Reconstruction. At a more contentious gathering, third district Republi-
cans selected Asa Hutchinson to succeed his brother in the U.S. House.23
Democrats in the third district also held a convention to nominate a sub-
stitute candidate. In November, Asa Hutchinson defeated Democrat Ann
Henry.
Party Committee
Nomination of major party candidates by party committee is infrequent,
and is most often used to replace a candidate selected in the primary who
has subsequently withdrawn or died. A committee is also used for original
nomination when the primary has not nominated a qualified candidate.
Generally, candidates chosen to replace incumbents do well in Novem-
ber. The one such case in 1996 was highly unusual, however. In Oregon,
first-term Representative Wes Cooley won the Republican primary in the
second district. But after the filing deadline for the primary, it became
apparent that Cooley had lied about his military record and wedding date.
Republican leaders—national, state, and local—urged Cooley to resign the
nomination. Reluctantly he did, and local political leaders then chose Coo-
ley’s predecessor, Robert F. Smith, as the Republican nominee.24 Smith won
comfortably in November.
Candidates chosen by party committee to substitute for a House aspi-
rant or to fill a nomination left vacant in the primary usually fare poorly in
the general election. This was true in 1996 as the three original party com-
mittee nominees received between 21.6 percent and 30.0 percent of the
two-party vote.
Write-in Nominees
A write-in nominee is a candidate who wins the party primary but does not
appear on the ballot. Rarely are major party congressional candidates nom-
inated as write-ins. In 1996, there were two such nominations for Democ-
ratic House candidates, in the first district of Arizona and the eighth dis-
trict of Illinois. Write-in candidates may be selected with exceedingly few
votes. The Arizona nominee received 1,942 write-ins and the Illinois nom-
inee but 75. Generally, write-in nominees do poorly in November. In 1996,
the Arizona nominee received 36.7 percent, and the Illinois nominee 39.8
percent, of the two-party vote.
Self-nominated Candidates: Louisiana and Texas
Candidates entered in the Louisiana unified primary, as described above,
are actually the functional equivalent of self-nominated candidates in a
general election. In 1996, self-nominated Democrats entered five district
primaries and self-nominated Republicans entered six district primaries.
Outside of Louisiana, self-nomination has not been an ordinary method
of major party nomination. However, in 1996, due to judicial invalidation
of the regular primary in 13 Texas congressional districts, Democratic and
Republican candidates in the November 5 special election were “self-nom-
inated.” Candidates were free to file for this general election up to August
30.25
DISTRICTS CONCEDED BY THE PARTIES

In 1996, both major parties nominated candidates for all of the Senate elec-
tions. But in House races, Democrats did not field a candidate for ten seats
and Republicans were without a candidate in eight districts. For the second
congressional election year in a row, Democrats conceded more seats than
Republicans. This is a major reversal in party coverage of congressional
districts. Democrats had long offered more candidates than the Republi-
cans.26
One consistency, though, is that most conceded seats continue to be in
the South. In 1996, the ten districts without a Democratic candidate were
in Florida (3 seats), Kentucky (1), Louisiana (2), South Carolina (2), Texas
(1), and Virginia (1). Republicans did not field candidates in districts in
Georgia, Louisiana, Illinois (2), Massachusetts, Ohio, and West Virginia
(2).
The recent surge in Republican coverage of the congressional districts
must surely have contributed to their 1994 and 1996 electoral success. The
current situation, however, contrasts with the 1896–1930 era. During this
earlier period of Republican ascendance in Congress, the Democratic party
offered more complete candidate coverage of congressional districts.27
CONGRESSIONAL CANDIDATES
We now examine the number and career status of candidates in congres-
sional primaries. The career status of the candidates is classified as incum-
bent, current office-holder, or non-office-holder.28 This scheme builds on
earlier studies.29 Less systematically, we highlight celebrity candidates. For
each party we look at seats with and without an incumbent. For seats with
running incumbents, we also contrast the incumbent party primary and the
challenging party primary.
SENATE SEATS WITH AN ENTERED INCUMBENT

The number of candidates running for an occupied U.S. Senate seat is quite
small, and the number running in the party of the incumbent is further
attenuated. In 1996, the forty major party nominations for these 20 seats
attracted a total of 86 candidates, 44 Democrats and 42 Republicans (table
2.3). In the party primaries with an incumbent, there were 35 candidates,
including the 20 incumbents. There were 51 candidates in the primaries of
the challenging major party.
Numbers alone do not constitute a formidable challenger pool. In Alas-
ka, seven Democrats entered the primary to select a challenger to Republi-
can Senator Ted Stevens. Yet, Joelle Hall, the State Democratic Party Exec-
utive Director, called the field of candidates “disappointing.”30 Hall turned
out to be prescient, for the eventual party nominee received only 11.9 per-
cent of the two-party vote. For this reason, we turn our attention to the
career status of the candidates.
Table 2.3: Candidate Pool for Congressional Seats, 1996a
Seats with an Incumbent
Number of Candidates in Primary of:
Type of Incumbent Number of Seats Incumbent Party Challenging Party
Democratic
Senator 7 11 (0)b 18 (5)
Republican
Senator 13 24 (1) 33 (4)
Democratic
Representative 149 209 (8) 282 (21)
Republican
Seats without Incumbent
Number of Candidates in Primary of:
Outgoing Incumbent Number of Seats Incumbent Party Challenging Party
Democratic
Senator 7 19 (10) 27 (11)
Republican
Senator 5 23 (9) 15 (4)
Democratic
Republican
a
The Senate analysis excludes the unified primary in Louisiana and the short-term contest in Kansas. Con-
vention nominations are tabulated as one-candidate primaries. The House analysis excludes districts in
Louisiana and Texas with unified primaries; districts in which one or more nominations were decided by a
method other than a primary; and the Vermont seat which has a Socialist incumbent.
b
The figures in parentheses are the number of office-holding candidates other than the incumbent.
Only ten office-holders sought incumbent senate seats, and but one of
these ten challengers, a state representative from South Carolina, was a
candidate in the incumbent senator’s own party primary. Republican Strom
Thurmond beat this challenger by better than two to one.
The Iowa senate seat retained by Democrat Tom Harkin attracted three
Republicans, a state senator, and a state representative, and U.S. Represen-
tative Jim Ross Lightfoot, who won the primary. The Texas seat retained
by Republican Phil Gramm attracted four Democrats, including two mem-
bers of the U.S. House. However, in a story which made the front page of
the New York Times, both representatives were eventually defeated by Vic-
tor M. Morales, a $36,000 a year civics teacher.31
In South Dakota, Democratic U.S. Representative Tim Johnson won an
uncontested primary and then defeated Republican Senator Larry Pressler
in the general election. In Massachusetts, Republican Governor William
Weld was unopposed in the primary, but he lost to Democratic Senator
John Kerry in November. In Montana, Lt. Governor Dennis Rehberg
defeated two candidates in the Republican primary, but he lost the general
election to Senator Max Baucus.
The remaining office-holder seeking an incumbent seat was Don
McCorkell, a Democratic state representative in Oklahoma. McCorkell
lost the primary to Professor Jim Boren, who enjoyed wide name recogni-
tion because his cousin David Boren had held the contested Senate seat
from 1979 until his resignation in 1994.32 In November, Jim Boren lost to
Republican incumbent James M. Inhofe.
HOUSE SEATS WITH AN ENTERED INCUMBENT

Many of the patterns concerning candidate numbers and career status in
Senate primaries are evident in House contests. In 1996, there were 341
congressional districts with an incumbent running for another term in
which a primary was the only means of nomination used. These potential
682 major party nominations drew a total of 1,065 candidates, 523
Democrats and 542 Republicans (table 2.3). In the party of the incumbent
there was a total of 469 candidates (including 341 incumbents), and in the
primaries of the challenging party there were 596 candidates.
There were very few current office-holders among the challengers. This
was particularly so in the party primary of the incumbent member of Con-
gress. Republican incumbents drew only two office-holders as opponents,
and Democratic incumbents drew only eight office-holding opponents in
their own primaries, including three in a safe Democratic district in Detroit
to oppose losing incumbent Barbara-Rose Collins. The five other Democ-
ratic incumbents with an office-holding challenger were successful. In the
most covered of these campaigns, in Georgia, Representative Cynthia
McKinney, an African-American, defeated three white candidates, includ-
ing a state senator, in a district with a new white majority due to recent
redistricting.33
In challenging party primaries there were 41 Democratic and 21 Repub-
lican office-holders, including 33 state legislators. Office-holding candi-
dates won 30 Democratic and 13 Republican nominations, and in Novem-
ber eight of them, six Democrats and two Republicans, beat the opposing
party incumbent. The most familiar of these successful nominees was
Democrat Dennis J. Kucinich, a state senator in 1996 when he defeated
Republican Martin R. Hoke, but best known for his stint as mayor of
Cleveland in the late 1970s.
In challenging party primaries, non-office-holders were by far the largest
group of candidates, and they won 156 Democratic nominations to oppose
Republican incumbents and 129 Republican nominations to oppose Demo-
cratic incumbents. In November, 10 of these Democrats, but none of these
Republicans, won. In these districts, no challengers entered the remaining
7 Republican and 6 Democratic primaries, and nominees were not select-
ed.
Democratic Representative Carolyn McCarthy, a nurse when elected to
Congress in 1996, was the least politically experienced of the elected non-
office-holders. Her political career was instigated by the murder of her hus-
band and maiming of her son in a shooting spree on a Long Island Rail-
road commuter train. After the tragedy, McCarthy advocated gun control
and she was very disappointed with her representative’s views on this issue.
When McCarthy decided to run for Congress, local Republican leaders did
not support her plans, but she was eagerly courted by local and national
Democrats. In September she won an uncontested primary and in Novem-
ber she defeated Representative Dan Frisa.34
SENATE SEATS WITHOUT AN ENTERED INCUMBENT

In 1996, 13 Senators retired and this resulted in a like number of senate
seats without an entered incumbent. In the twelve states with party pri-
maries, there were 84 candidates for these seats, 34 Democrats and 50
Republicans.35 The 84 candidates were evenly divided between the defend-
ing and challenging parties (table 2.3). The average number of major party
candidates running for an open senate seat was 7.0, and it was 4.3 for seats
with an entered incumbent.
A larger contrast between open senate seats and those with an incum-
bent, however, is the career status of the candidates. In the senate races
without an entered incumbent, 34 of the 84 entered candidates, 40.5 per-
cent, were current office-holders. By contrast, in the 20 senate races with
an incumbent only 10 of 66 challengers, 15.2 percent, were in office. The
largest gap was in incumbent parties; in seats with an incumbent senator,
only 1 of 15 challengers was an office-holder, but in seats with a retiring
incumbent, 19 of 42 candidates held office.
Elected officials dominated in the open senate primaries, winning nine

Republican and eight Democratic nominations. These nominees were U.S.
Representatives (7), state senators (3), state attorney generals (2), state
treasurers (2), a governor, a state secretary of state, and a state representa-
tive. In November, these nominees won 10 of the 12 open senate seats.
Non-office-holders won seven open senate seat nominations, but only
two of this group won the general election: Republicans Susan Collins of
Maine and Chuck Hagel of Nebraska. Both of these new senators had
extensive political experience, but neither had been elected before to pub-
lic office.
HOUSE SEATS WITHOUT AN ENTERED INCUMBENT

In the major party primaries for the 47 open seats exclusively using the pri-
mary, there were 377 candidates (table 2.3). These candidates were almost
evenly divided between the Democratic party (193) and the Republican
party (184). There were 232 candidates in the primaries of the incumbent
party, and 145 candidates in the challenging party primaries. Whereas the
open House seats averaged 8.0 candidates, only 3.1 candidates entered
major party primaries in districts with an incumbent.
Both political parties had willing candidates for all open congressional
seats, and the field of candidates swelled for a dominant party primary. In
each of the 94 party primaries in the open districts there was at least one
candidate, and in 81 primaries more than one. But a few districts skewed
the average number of candidates. In an overwhelmingly Democratic dis-
trict centered on west Baltimore, a record 27 Democratic hopefuls vied in
the Democratic primary.36 A traditionally Republican district, the seventh
district of Indiana, attracted the largest GOP field, 15 candidates.
There were ample office-holders among the congressional candidates in
the open districts and they tended to concentrate in the party of the outgo-
ing incumbent. As with the senate races, the contrast between the incum-
bent and open seats was greater with respect to the number of office-hold-
ing candidates than it was to the total number of candidates. In open
House seats, office-holders composed 25.2 percent, 95 of 377, of the can-
didate pool. By contrast, in the 341 incumbent House seats, only 72 of 724
challengers, 9.9 percent, were office-holders.
In the open races, the contrast between the candidate pools of the
incumbent and challenging party primaries, however, was much greater in
House races than it was for the Senate. In the open House races the pool
of office-holders was much larger in the incumbent party, probably because
there are many more elected officials in locally dominant parties. In the 29
open Democratic seats, there were 47 Democratic and ten Republican
office-holders, and in the 18 open Republican seats, there were 29 Repub-
lican and nine Democratic officials. These 95 office-holders were from a
variety of state, county, and municipal posts, and included 52 state legisla-
tors. There were office-holders in both party primaries in only 11 of the 47

open districts.
In the 29 Democratic open seats, the 47 Democratic office-holders won
19 nominations and 16 general elections, and the 102 non-office-holders
won 10 nominations and five general elections. The ten Republican office-
holders won six nominations and three general elections, while the 91 non-
office-holding Republicans won 23 nominations and five general elections.
In the 18 Republican open seats, the 29 Republican office-holders won
ten nominations and eight general elections, and Republican non-office-
holders won eight nominations and six general elections. One of these win-
ning non-office-holders was Jim Ryun of Kansas, who in the 1960s was the
world record-holder for the one-mile run and an Olympic athlete. The nine
Democratic officials won six nominations and two general elections. The
35 non-office-holding Democrats won twelve nominations and one gener-
al election. Independent candidate Jo Ann Emerson, the widow of Missouri
Republican Representative Bill Emerson, won the remaining Republican
district. Emerson was forced to run as an independent because her husband
died after the filing deadline, but she joined the new Congress as a Repub-
lican.
Overall, in both Senate and House primaries, incumbents did very well
and only infrequently were they challenged by office-holders in their pri-
mary or in the opposition party primary. Office-holders were much more
successful in primaries and general elections than non-office-holders. In
open districts, party success increased with previous control of the district
and candidate quality.
ELECTORAL PATTERNS IN CONGRESSIONAL PRIMARIES

Competition, turnout, and incumbency performance in the 1996 congres-
sional primaries are the concern of this section of the chapter.
CONTESTED AND UNCONTESTED PRIMARIES
U.S. House
In 1996, only 39.7 percent, 307 of 774, of the House primaries were con-
tested. Specifically, the proportion of House primaries contested, control-
ling for party and incumbency, was:
Democratic Republican
Without incumbent 46.8% 110 56.8% 109
235 192
With one incumbent 27.2% 41 24.0% 47
151 196
Incumbency considerably diminished the frequency of contests in House

primaries, especially in Republican primaries. The electoral strength of
incumbents deterred many potential challengers from entering into a pri-
mary fray.
In each of the 18 election years between 1956 and 1990, Democratic pri-
maries, both with and without an incumbent, were contested more often
than Republican primaries.37 However, in 1992 and 1994, in primaries
without an incumbent, Republican contests were slightly more frequent. In
1996, this pattern continued, and the gap between the parties widened con-
siderably, a difference caused by the large drop in Democratic contests.
There has been a marked trend toward contests in southern nonincumbent
Republican primaries. In 1996, 69.0 percent of these primaries were con-
tested, compared to only 27.2 percent between 1956 and 1974.38 By con-
trast, in 1996, 49.2 percent of the southern Democratic nonincumbent pri-
maries were contested, a decline from the 66.7 percent contested rate for
1956–1974.
Another contributing factor to the rise of contested Republican pri-
maries is that party’s recently enhanced electoral prospects. In both Demo-
cratic and Republican primaries—when there is not an entered incum-
bent—the likelihood of a public contest increased with a rise in the party’s
November prospects, measured as the party general election vote. Thus
there were 19 districts in which a nonincumbent of either party won with
60 percent or more of the two-party vote, and in all these districts the win-
ner had survived a contested primary. By contrast, party primaries were
contested less than 30 percent of the time in districts in which the party
received less than 30 percent of the two-party vote. In all regions of the
United States, enhanced electoral prospects led to more frequent contests.
Among primaries with an incumbent, there was a contest about one-
fourth of the time. The presence of an incumbent eliminated the positive
relationship between November prospects and primary contests.
The traditional party battle within each of the sections is of some impor-
tance in explaining the frequency of contests.39 Rarely were incumbents of
a weak sectional party challenged in the primary—southern Republicans (4
of 51 contested); Plains states Democrats (0 of 7); and Mountain Democ-
rats (0 of 3). The Atlantic, which traditionally has few contests, was the
only region in 1996 with more frequent contests in primaries with an
incumbent. And the Border states and Midwest were the only sections
above the national contesting rate for all four types of House primaries.
Surprisingly there were no challenges to any of the ten Republican incum-
bents in the Plains.
U.S. Senate
In 1996, 71.0 percent, 44 of 62, U.S. Senate primaries were contested. This
is a marked contrast with the House figure of 39.7 percent, and is to be
expected given the longer term of office for senators and their more diverse
and larger constituencies. (In addition there were contents in both pri-
maries for the remaining two years of Robert Dole’s vacated Senate seat.)
The proportion of each type of U.S. Senate primary contested was:
Democratic Republican
Without incumbent 75.0% 18 88.9% 16
24 18
With incumbent 28.6% 2 61.5% 8
7 13
Overall, the Senate primaries followed patterns similar to the House pri-
maries: more contests in primaries without an incumbent, more contests in
Republican primaries after controlling for incumbency, and a recent spurt
in southern Republican contests. In these basic respects Senate primaries
were consistent with—but always with more frequent contests than—the
House primaries.
The overwhelming number, 34 of 42, of Senate primaries without an
incumbent were contested. The distribution of the uncontested primaries
reflected traditional party strengths and the stature of the eventual nomi-
nees. The two uncontested Republican primaries were in strongly Democ-
ratic West Virginia and Massachusetts, where the Republican nominee was
Governor William Weld. Three of the six uncontested Democratic pri-
maries were in the traditionally Republican Plains states of Kansas,
Nebraska, and South Dakota. Furthermore, the Democratic nominee in
Nebraska was Governor Ben Nelson, and the South Dakota Democratic
nominee was Tim Johnson, elected to the at-large congressional district
since 1986. Another uncontested Democratic primary was in strongly
Republican Idaho. The sixth uncontested primary was in New Jersey,
where Democratic Representative Robert G. Torricelli was the nominee.
Primaries with a senate incumbent were less likely to be contested. Only
2 of 7 Democratic senators, John D. Rockefeller IV of West Virginia and
Paul Wellstone of Minnesota, were challenged in the primary. Among
Republicans, though, eight of thirteen incumbents were challenged. In the
southern and border states, seven of eight Republican senators had prima-
ry competition. Only Jesse Helms of North Carolina was not challenged.
However, outside of the South only one senator, Ted Stevens of Alaska, was
challenged, while four were renominated in an uncontested primary. (The
newly appointed Republican Senator from Kansas, Sheila Frahm, was also
challenged in the primary.)
VOTER TURNOUT
An Overview of Participation Levels

In order to measure voter turnout in the 1996 congressional primaries, we
selected the 74 congressional districts and 15 states (not including the
short-term Kansas senate race) with simultaneously contested Democratic
and Republican primaries. The numbers voting were then compared to the
voting age population and the number of general election voters in these
districts and states.
The central finding of this analysis is that turnout levels in congression-
al primaries were quite low. In more than one-half of the House districts
and states the turnout for congressional primaries was less than 20 percent
of the voting age population. In no district and in only one state did pri-
mary turnout exceed 35 percent of the voting age population.
Turnout in congressional primaries was also low in comparison to voter
participation in subsequent general elections. In 59 of 74 House districts
with two contested primaries, turnout in the primaries was less than half
of the general election figure. For the senate seats, turnout in the primaries
attracted more than half of the general election vote in only two of 15
states.
The Distribution of Voters Between Party Primaries
The nomination contests that we are studying take place, of course, with-
in the confines of a political battlefield defined by the skirmishes of the two
major parties.40 A party’s congressional nomination gains importance
directly with the stature of the party among the electorate as a whole. The
stronger a party is among the voters in a district, the more that party’s pri-
mary becomes the real focus of electoral decision. Popular interest, it fol-
lows, should thus center upon the primary of the stronger political party.
This speculation is confirmed by an analysis of the 74 House districts
and 15 states with a pair of contested primaries in 1996. Voter participa-
tion in the Democratic or Republican primaries resembled the November
vote split. As the November percentage of the Democratic party rose, so
did the percentage of primary voters participating in the Democratic pri-
mary. Thus the Democrats received between 30.0 percent and 39.9 percent
of the November vote in 11 districts, and in those districts the average
Democratic percentage of the primary vote was 34.4. percent. In four
House districts, the Democrats received 70.0 percent to 79.9 percent of the
November vote, and in those districts they received an average of 76.3 per-
cent of the primary vote.
In senate races, also, there is often a correspondence between the
distribution of voters on primary day and election day. In Alaska, for
example, the Democrats received 10.3 percent of the two–party primary
turnout, and in November they received 11.9 percent of the two-party
vote. Meanwhile, in Rhode Island, the Democrats received 82.1 percent of

the primary vote, and 64.4 percent of the two–party November vote.
The distribution of voters on primary day, however, is more likely than
the November election to reflect the traditional political pattern of a state
or district. This is evident in the just–noted Rhode Island senate race in
which the traditionally strong Democratic party drew a higher propor-
tion of primary voters than November voters. This proposition is also
illustrated by the voting pattern contrasts in the first and ninth congres-
sional districts of Tennessee. The first district in the eastern end of the
state has traditionally been a “one–party” or “safe” Republican con-
stituency. In 1996, the Democrats received 33.3 percent of the two-party
vote, but only 14.5 percent of the primary voters. On the other had, the
ninth district in the southwest corner of the state, encompassing Mem-
phis, has traditionally been a Democratic stronghold. In 1996, Democrats
received 62.1 percent of the two-party vote, but 89.6 percent of the pri-
mary voters. In both districts, the dominant party had a stronger hold on
voters on primary day than on election day. These districts thus support V.
O. Key Jr.’s proposition that there is a “tendency for popular interest to
concentrate in the primary of the stronger party.”41
MARGINS OF VICTORY
Primaries with an Incumbent

Competitive levels in congressional primaries with an incumbent are
extremely low. Most of these primaries are uncontested, and those with a
public contest are rarely decided by a close margin. In 1996, no senate
incumbent won with less than 60 percent of the leading two-candidate
vote, and six of the ten contested Senate primaries were won with 80 per-
cent or more of this vote. (The short-term Kansas primary is excluded from
these figures.)
Of the 347 House incumbents in party primaries, only two lost, and
only six more won with less than 60 percent of the leading two-candidate
vote. Forty-six of 88 contested House primaries were won by the incum-
bent with 80 percent or more of the two-candidate vote.
An examination of the losing and closely contested incumbent races sug-
gests some of the factors that lead to incumbency weakness in congres-
sional primaries. In Detroit, Democratic Representative Barbara-Rose
Collins, weakened by ethical problems, largely stemming from campaign
finance, lost to state representative Carolyn Cheeks Kilpatrick, a well-
respected challenger.42 Three of the six representatives winning by a small
margin—Republicans Joseph McDade (53.3 percent) and Bill Goodling
(54.9 percent) of Pennsylvania, and Republican Jay Kim (58.2 percent) of
California—were also hurt by charges of wrongdoing.43 In addition, and in
contrast to the California challenger, the Pennsylvania challengers in these

two primaries were credited with strong campaigns.44
In Texas, Republican Representative Greg Laughlin—who switched
from the Democrats to the Republican party in June 1995—failed to win
the nomination of his new party. Laughlin lost to Ron Paul, a Republican
U.S. representative in the 1970s and 1980s from a partially overlapping
district, who enjoyed high name recognition and “ran an aggressive and
well-funded campaign.” Paul charged that Laughlin was not a true Repub-
lican or conservative.45
Political ideology also played a role in three of the close incumbent wins.
In Albany, New York Democratic Representative Michael R. McNulty was
held to 57.0 percent of the vote by a challenger who argued that McNulty
was too conservative, pointing out that McNulty supported the Republican
Contract with America 63 percent of the time. In the Hudson Valley of
New York, Republican incumbent Sue W. Kelly won 55.8 percent of the
two-leading candidate vote, as a former Representative from an adjoining
district, Joseph J. DioGuardi, attacked her for being too liberal on social
issues, especially for being too permissive on abortion. In San Diego,
Democratic Representative Bob Filner was held to 54.5 percent of the vote,
as he was attacked for being too liberal. In these three cases, also, the chal-
lenger campaigns were high profile.46
Only rarely do congressional incumbents lose primaries. One exception
to this general proposition is the poor fate of appointed U.S. Senators.47 In
1996, the hazards facing appointed senators continued as Senator Sheila
Frahm of Kansas, appointed to the Senate to replace Robert Dole on May
24, 1996, was defeated in the Republican primary on August 6 by U.S.
Representative Sam Brownback, who then went on to win the remaining
two years of Dole’s term.
Primaries Without an Incumbent
Levels of competition were higher in primaries without an incumbent than
they were in primaries with an incumbent. Primaries without an incumbent
were more often contested; and contested primaries were more often won
with a smaller percentage of the vote.
In the absence of an incumbent, many of the contested primaries were
won by a small margin. In House primaries, the winning margin was less
than 60 percent in 40 of 110 Democratic contests and 42 of 109 Republi-
can contests. In Senate primaries, five of 18 Democratic contests and six of
16 Republican contests were won with less than 60 percent of the vote.
House primaries without an incumbent were rarely won with more than
80 percent of the two-leading candidate vote. In House primaries only 15
of 110 Democratic contests and 13 of 109 Republican contests were won
this overwhelmingly. Senate outcomes were more frequently one-sided, as
eight of 34 contested Senate primaries were won with at least 80 percent.
CONCLUSION
The major party nominating process is the crucial first step in the election
of members of Congress. Once the party nominations have been decided,
the quest for office and the realistic alternatives presented to voters are
ordinarily limited to the two major party nominees. In the American polit-
ical system, to paraphrase V. O. Key Jr., a good deal of politics is intraparty,
rather than interparty.48 Congress is a national legislature with local roots,
and the constituency-based nomination process is an integral link between
the two Congresses: the one in Washington, D.C. and the one in the dis-
tricts. 49
NOTES
1. Frederick W. Dallinger, Nominations for Elective Office in the United
States (New York: Longmans, Green, 1897), 3.
2. V. O. Key Jr., Politics, Parties, and Pressure Groups, 5th ed. (New York:
Crowell, 1964), 375.
3. The major source for 1996 nomination methods and final results in pri-
maries is Richard M. Scammon, Alice V. McGillivray, and Rhodes Cook, America
Votes 22: 1996 (Washington, DC: Congressional Quarterly Inc., 1998). This hand-
book was supplemented by CQWR, 1996; Philip D. Duncan and Christine C.
Lawrence, Politics in America 1998 (Washington, DC: Congressional Quarterly
Inc., 1997); and Michael Barone and Grant Ujifusa, The Almanac of American Pol-
itics 1998 (Washington, DC: National Journal Inc., 1997).
4. The Book of the States, 1996–1997 (Lexington, KY: Council of State Gov-
ernments, 1996), 157–158; and Karen M. Markin, Ballot Access 2: For Congres-
sional Candidates (Washington, DC: National Clearinghouse on Election Adminis-
tration, Federal Election Commission, 1995).
5. Maureen Moakley, “Political Parties in Rhode Island: Back to the Future,”
in Parties and Politics in the New England States, ed. Jerome M. Mileur (Amherst,
MA: Polity Publications, 1997), 95–112, quote 109.
6. Joseph A. Kunkel III, “Party Endorsement and Incumbency in Minnesota
Legislative Nominations,” Legislative Studies Quarterly 13 (May 1988): 211–223.
7. This classification of state primary systems is from the state reports in
America Votes 22: 1996.
8. Tashjian v. Republican Party of Connecticut 107 S. Ct. 544 (1986).
9. League of Women Voters Education Fund, Vote! The First Steps (Wash-
ington, DC, 1996). This paragraph does not include the three unified primary
states.
10. James A. Palmer, Edward D. Feigenbaum, and David T. Skelton, Election
Case Law 97 (Washington, DC: Federal Election Commission, 1997), 82–83,
99–100, 234.
11. Ballot Access News, May 4, 1995; April 3, 1996; May 28, 1996; July 20,
1996; and August 12, 1996; Ronald D. Elving, “Court Lets Decision Stand on Alas-
ka’s Open Primary,” CQWR, May 17, 1997, 1156.
12. Marc Birtel, “Ballot: ‘Jungle Primaries’ Adopted... Reclassifying Lions

Rejected,” CQWR, March 30, 1996, 902–903; Michael Ross, “California Voters
Adopt an Anti-party ‘Open’ Primary,” Party Developments, June 1996, 5–6.
13. Michael Barone and Grant Ujifusa, The Almanac of American Politics
1982 (Washington, DC: Barone and Co., 1981), 426; Jack Bass and Walter DeVries,
The Transformation of Southern Politics (New York: Meridian, 1976, 1977), 181.
14. Gregory L. Giroux, “Louisiana: Primary Statute Struck Down,” CQWR,
December 6, 1997, 3041; “States Cannot Change November Election Date for Fed-
eral Offices,” Party Developments, February 1998, 24.
15. Book of the States, 157–158; Markin, Ballot Access 2.
16. Warden Moxley, “Southern Primaries,” in Guide to U.S. Elections,
1789–1974 (Washington, DC: Congressional Quarterly Inc., 1975), 889–993; and
Charles S. Bullock III and Loch K. Johnson, Runoff Elections in the United States
(Chapel Hill: University of North Carolina Press, 1992), 1–8.
17. V. O. Key Jr., Southern Politics in State and Nation (New York: Knopf,
1949), 422; Cortez A. M. Ewing, Primary Elections in the South: A Study in Uni-
party Politics (Norman: University of Oklahoma Press, 1953), 6–7.
18. Bullock and Johnson, Runoff Elections, 37.
19. Bullock and Johnson, 50.
20. Bullock and Johnson, 66.
21. Key, Southern Politics, 438–442.
22. Alan Greenblatt, “It Will Be Warner vs. Warner as Incumbent Beats
Miller,” CQWR, June 15, 1996, 1707; Almanac of American Politics 1998, 1440,
1446.
23. Almanac of American Politics 1998, 114, 122.
24. Robert Marshall Wells, “Smith May Try to Reclaim 2nd as Cooley Ends
His Bid,” CQWR, August 10, 1996, 2271; Almanac of American Politics 1998,
1185–1186.
25. Juliana Gruenwald, “Panel Redraws Texas Map, Sets Election Date,”
CQWR, August 10, 1996, 2270; and idem, “New Districts to Stay in Place;
Tougher Races for Candidates,” CQWR, September 7, 1996, 2542.
26. Also see Peverill Squire, “Competition and Uncontested Seats in U.S.
House Elections,” Legislative Studies Quarterly 14 (May 1989): 281–295; and
Gary C. Jacobson, The Electoral Origins of Divided Government: Competition in
U.S. House Elections, 1946–1988 (Boulder, CO: Westview Press, 1990), 46–50.
27. Cortez A. M. Ewing, Congressional Elections, 1896–1944: The Sectional
Basis of Political Democracy in the House of Representatives (Norman: University
of Oklahoma Press, 1947), 55–57.
28. The career status of candidates is based on the listings and campaign nar-
ratives in CQWR, 1996; Almanac of American Politics 1998; and Politics in Amer-
ica 1998.
29. The methodology used here is an adaptation of analyses in David T.
Canon, Actors, Athletes, and Astronauts: Political Amateurs in the United States
Congress (Chicago: University of Chicago Press, 1990), 74–87; Gary C. Jacobson,
The Politics of Congressional Elections, 4th ed. (New York: Longman, 1997),
34–37; and Gary C. Jacobson and Samuel Kernell, Strategy and Choice in Con-
gressional Elections (New Haven: Yale University Press, 1981), 19–34.
30. CQWR, June 1, 1996, 1554.
31. Sam Howe Verhovek, “Running on Dare, Teacher Wins Senate Primary,”
NYT, April 11, 1996, Al, B8.
32. Alan Greenblatt, “Republicans Inhofe, Stevens Easily Win Renomina-
tion,” CQWR, August 31, 1996, 2482.
33. Deborah Kalb, “Redistricting Leaves McKinney Fighting the Pack in the
4th,” CQWR, June 29, 1996, 1815–1816.
34. Politics in America 1998, 972–973; Almanac of American Politics 1998,
979–980.
35. The thirteenth open senate seat was in Louisiana, where a unified party
primary attracted 15 candidates: six Republicans, four Democrats, and five inde-
pendents.
36. Deborah Kalb, “32 Candidates Try to be Heard in Race for Maryland
7th,” CQWR, February 24, 1996, 451–453.
37. For presentation of the 1956 to 1974 data, see Harvey L. Schantz, “Con-
tested and Uncontested Primaries for the U.S. House,” Legislative Studies Quarter-
ly 5 (November 1980): 545–562.
38. Schantz, 552–553.
39. Schantz, 551–552.
40. V. O. Key Jr., American States Politics: An Introduction (New York:
Knopf, 1956), 97–100.
41. Key, American State Politics, 100.
42. Almanac of American Politics, 1998, 766; Robert Marshall Wells,
“Collins Loses to Kilpatrick,” CQWR, August 10, 1996, 2264.
43. In 1998, Kim, confined to two months of “home detention,” for ten cam-
paign finance misdemeanors, lost his primary. See Marc Birtel, “Rep. Kim Must
Remain East for Most of Campaign,” CQWR, April 4, 1998, 902.
44. Politics in America 1998, 206, 1247, 1273.
45. Politics in America 1998, 1401; Juliana Gruenwald, “Bryant, Laughlin
Foiled by Unlikely Opponents,”CQWR, April 13, 1996, 999–1000, quote 1000.
46. Almanac of American Politics 1998, 1018, 1023; Politics in America
1998, 231–232, 1018, 1023–24; and Ronald D. Elving, “Incumbent Filner Battling
Challenger in Primary,” CQWR, March 23, 1996, 814–816.
47. William D. Morris and Roger H. Marz, “Treadmill to Oblivion: The Fate
of Appointed Senators,” Publius 11 (Winter 1981): 65–80.
48. Key, Politics, Parties, and Pressure Groups, 435.
49. Roger H. Davidson and Walter J. Oleszek, Congress and Its Members, 7th
ed. (Washington, DC: CQ Press, 2000).
CHAPTER 3
The Presidential Campaign

and Vote in 1996
Job Ratings of Presidents—and Success or Failure at the Polls
MILTON C. CUMMINGS JR.
In an election year that some commentators asserted produced the dullest

and least exciting presidential campaign in many decades, there were
nonetheless a number of noteworthy features in the final voting returns.
President Bill Clinton, the Democratic nominee, won a decisive victory over
former Senator Bob Dole, his Republican opponent, and Ross Perot, the
candidate of the Reform Party. Clinton carried thirty-one of the fifty states
and the District of Columbia, and won 379 electoral votes to 159 for Dole.
Clinton’s victory marked the first time since 1936 that a Democratic pres-
ident had been elected to a second full term. And he was only the fourth
Democratic president in history—along with Franklin Roosevelt,
Woodrow Wilson, and Andrew Jackson—to win two consecutive presi-
dential terms.1
Clinton’s vote went up between 1992 and 1996; but the vote for the
other presidential candidate who ran in both years, Ross Perot, dropped
sharply, from 18.9 percent in 1992 to 8.4 percent four years later. Even so,
Perot’s 1996 presidential showing was the sixth largest vote percentage
polled by a third-party or independent presidential candidate since the Civil
War. In addition, though little noted, there was another sign in 1996 that
many voters were not wedded firmly to the two major parties. Between
1992 and 1996, the vote for minor-party candidates for president other
than Perot more than doubled.2 Those “other” minor-party tallies included
close to 700,000 votes for Ralph Nader on the Green Party ticket, and
nearly half a million votes for Harry Browne on the Libertarian ticket.
The 1996 voting for Congress also produced an outcome that would
have a powerful impact on relations between the president and Congress
for at least the next two years. The Republican party suffered a moderate
net loss of seats in the House of Representatives and gained strength in the
Senate. But the election left the Republicans with clear majorities in both
63
64 The Presidential Campaign and Vote in 1996
houses of Congress, and Clinton continued to face a Congress controlled

by the opposition party after his second inauguration.
The presidential election results were:
Popular Vote Electoral Vote Popular Vote

Percentage
Bill Clinton (D) 47,402,357 379 49.2%

Bob Dole (R) 39,198,755 159 40.7%
Ross Perot (RP) 8,085,402 0 8.4%
Others 1,591,358 0 1.7%
Totals 96,277,872 538 100.0%
THE AMERICAN ELECTORATE OF 1996

The size of the electorate that turned out to vote dropped sharply between
1992 and 1996. About 96.3 million people went to the polls in 1996—
eight million fewer than had voted four years earlier. Moreover, since the
number of Americans who were of voting age increased between 1992 and
1996, this drop in the actual vote also produced the largest four-year drop
in the percentage turnout since the wartime election of 1944. The 1996
turnout—just 49.0 percent—was the nation’s lowest since 1924.3
This decline in the number of Americans who voted for president took
place in every major region of the country and in almost every one of the
fifty states. (Only two states with small populations, Idaho and Wyoming,
were exceptions.) In addition, the actual drop in the number of voting
Americans in some of the large states was enormous. In New York the total
vote was down by 610,000; in Pennsylvania the drop was 453,000; and in
Illinois it was 749,000. In California alone, 1,112,000 fewer people voted
for president in 1996 than had voted in 1992.
In some respects, however, the men and women who voted in 1996 did
not differ greatly from the active electorates in other recent presidential
elections. By their own report, the ratios between Protestants and Catholics
(54 percent to 29 percent) and between Democrats and Republicans (39
percent to 35 percent) remained about what they had been four years ear-
lier.4 A solid majority (66 percent) had family incomes of $30,000 per year
or more. And more than half (52 percent) of those who went to the polls
were women—a trend that first began to assert itself in the 1960s.
One sign of change in the active electorate that is likely to continue in
the near future did turn up in 1996. The ethnic and racial composition of
America’s voters became somewhat more diverse in 1996. In 1996, as
before, a very large majority of those who voted (83 percent) were white—
compared to 10 percent who were African-American and 5 percent who
identified themselves as Hispanic. Nevertheless, compared with 1992, the
white share of the electorate was down by four percentage points in 1996,
while the African-American and the Hispanic percentages both increased.

If present population trends hold, Hispanic Americans are destined to
become the nation’s largest ethnic minority group in the twenty-first cen-
tury.
Two other changes in the active electorate between 1992 and 1996 also
should be noted. There was a noticeable jump (from 19 percent to 23 per-
cent) in the share of the total vote that was cast by members of union
households in 1996—a trend that worked to President Clinton’s advantage.
And, in 1996, the percentage of the voters who were under 30 went down
by five points, from 22 percent to 17 percent, while the proportion of the
electorate who had reached or passed age 60 went up from 16 percent to
24 percent. Compared with 1992, it was a markedly older group of voters
who went to the polls in 1996.
PRELUDE TO 1996—1993, 1994, AND 1995

President Clinton began his presidency in 1993 with the broad approval
that the American public usually gives a new president at the start of the
first term.5 After his first few days in office, the Gallup poll reported that
58 percent of the public approved of the job Clinton was doing as presi-
dent, and only 20 percent disapproved (See table 3.1). Over the next nine-
teen months, however, his popularity plummeted, so that by September
1994—shortly after his major proposal for health-care legislation had been
rejected by Congress—only 39 percent of the public approved of Clinton’s
job performance; a full 54 percent disapproved. Two months after that
came the Republican electoral sweep in the congressional elections of 1994,
which produced GOP majorities in both the U.S. House and the U.S.
Senate.
The Democratic defeat that year was especially dramatic in the House
of Representatives, which the Democrats had controlled for forty consecu-
tive years. No other elected institution of the national government—not the
presidency, not the U.S. Senate—had ever been controlled that long by one
party. Now the Republicans and the new Speaker of the House, the out-
spoken Newt Gingrich of Georgia, were firmly in control.
Among Democrats, Clinton himself received much of the blame for their
party’s electoral catastrophe. As Gerald Pomper has written, in addition to
losing the midterm elections, the Clinton record of his first two years
included “the defeat of ambitious programs such as health-care reform, a
cascade of investigations and allegations about his personal conduct, for-
eign policy embarrassments, and an often maladroit administration of the
executive branch.”6
During the winter of 1994–1995, there was widespread speculation
among politicians and in the press that one or more prominent congres-
sional Democrats might challenge Clinton for the Democratic nomination
in 1996, just as President Jimmy Carter had been challenged in 1980.
Table 3.1: President Clinton’s Job Rating, 1993–1996*

Date of Interviews Approve (%) Disapprove (%) No Opinion (%)
Clinton Elected with 43% of the Vote, November 1992
Clinton Inaugurated, January 1993
January 24–26, 1993 58 20 22
April 22–24, 1993 55 37 8
“Travelgate,” “$200 haircut,” several presidential

nominations encounter difficulties with Senate,
David Gergen joins White House, May 1993
June 5–6, 1993 37 49 14
September 13–15, 1993 46 43 11
Clinton announces health care proposal in

speech to Congress, September 1993
September 24–26, 1993 56 36 8
November 19–21, 1993 48 43 9
Clinton’s second State of the Union address, January 1994

January 28–30, 1994 58 35 7
April 22–24, 1994 48 44 8
Health care reform dies, September 1994

September 6–7, 1994 39 54 7
Midterm elections—Republicans capture the House and Senate,

November 1994
December 28–30, 1994 40 52 8
Oklahoma City bombing, April 1995

April 21–24, 1995 51 39 10
September 14–17, 1995 44 44 12
November 17–18, 1995 53 38 9
January 5–7, 1996 42 49 9
Government shutdown ends, Clinton’s fourth

State of the Union address, January 1996
January 26–29, 1996 52 42 6
April 25–28, 1996 56 37 7
July 18–21, 1996 57 35 8
Republican and Democratic Conventions, August 1996

September 7–9, 1996 60 31 9
October 1–2, 1996 58 34 8
October 26–29, 1996 54 36 10
________________________________________________________________
*Responses to the question: “Do you approve or disapprove of the way Bill Clinton is han-
dling his job as president?”
Source: Data provided by the Gallup poll.

Public attention focused on the efforts of Speaker Gingrich and the fresh-
man Republicans to push their ambitious and conservative program, the
Contract with America, through Congress. At one point President Clinton
even felt the need to remind journalists that he was still relevant to the gov-
erning process: “The Constitution gives me relevance, the power of our
ideas gives me relevance.”7
Amid the Democrats’ gloom, however, there remained two factors that
should have given Clinton’s partisans some hope for the future. The presi-
dent’s public approval ratings were often low during his first two years in
office, and certainly they were too low to serve as a basis for reelection to
a second term. Yet even so, he nearly always retained the support of a solid
core of about 40 percent of the electorate (table 3.1). Even during his worst
periods in 1993 and 1994, Clinton’s job ratings in the Gallup poll never
went as low as did those of his three immediate predecessors—Jimmy
Carter, George Bush, and even that formidable vote-getter, Ronald Reagan.
Carter’s approval rating fell to 21 percent in July 1980, as he was
preparing for his ultimately unsuccessful fall reelection campaign. Bush
reached a low point of 29 percent in early August 1992, again just a few
weeks before the fall presidential campaign. And Reagan’s approval ratings
stood at 35 percent in January, 1983, shortly after the 1982 recession
reached bottom. In that January 1983 Gallup poll, 56 percent of the pub-
lic “disapproved” of “the way Ronald Reagan is handling his job as
President.”8
A second factor that some Democratic electoral strategists noted is that
despite the Republicans’ great midterm election victory in 1994, there was
no net realignment of the nation’s voters in terms of partisan identification.
The ratio between Democrats and Republicans in the electorate at large
remained unchanged. According to the Gallup Poll, the distribution of the
electorate in 1995 was Democrats 38 percent; Republicans 29 percent; and
independents 33 percent.9 This meant that among prospective voters,
Democratic identifiers continued to outnumber Republican identifiers by a
moderate margin going into the 1996 election year; and in fact, the exit
polls suggest, about four million more Democrats than Republicans voted
in November 1996. The Republican party made large gains in terms of
party identification in the mid-1980s, sharply narrowing their disadvan-
tage in numbers compared with the Democrats. But the GOP did not con-
tinue to gain ground in the party identification figures in the first half of
the 1990s.
Moreover, even before his party’s drubbing in the 1994 elections,
Clinton began to take steps to try to improve both the perceived perfor-
mance and the actual performance of his administration. He started by
making key changes in his White House staff. In June 1994, the president
announced that his longtime friend Thomas F. (Mack) McLarty would be
replaced as the White House Chief of Staff by Leon E. Panetta, director of
the Office of Management and Budget and a former member of Congress

from California. Panetta had served in the House of Representatives for
sixteen years and—unlike McLarty—had broad knowledge of the
Washington political world. After Panetta took charge as chief of staff, the
morale and performance of the White House staff appeared to improve.
Clinton also brought in State Department spokesman Michael McCurry to
be his White House press secretary, replacing Dee Dee Myers. McCurry
proved skillful in handling the daily White House press briefings and in
dealing with the news media.
Slowly, the president’s job-approval ratings began to rise. When a feder-
al office building was bombed in Oklahoma City in April 1995, killing 169
people, the nation was shocked and looked to Washington for reassurance.
Clinton and the federal government responded decisively to the tragedy. In
the aftermath, Clinton’s standing in the polls went up. As John Tierney has
noted, some of Clinton’s post-1994 foreign policy actions also seemed to
help his popularity. He also appeared to gain greater approval by taking
more moderate positions on a broad range of issues. The Democratic pres-
ident was moving to “the center.”10
Much of the national political debate in 1995 focused on the attempts
of Republicans in Congress to pass their legislative program, and Clinton’s
efforts to block or modify the Republican proposals. During Clinton’s first
two years in office, when the Democrats controlled Congress, the president
did not veto a single bill. In Clinton’s second two years in office, with a
Republican Congress, he exercised his veto fifteen times.11
In December 1995, Republican leaders tried to force Clinton to approve
their budget bill by withholding funds to run the federal government. The
result was a government shutdown. In November, and again in
December–January, many government agencies were closed for days. It was
a defining event in Clinton’s first term. The public appeared to blame the
Republicans in Congress more than they blamed the president for the
impasse; when the confrontation was over, the president’s job-approval rat-
ing had increased to 52 percent in the Gallup poll taken at the end of
January 1996. During the rest of the year, through the November election,
President Clinton’s job rating never dropped below 52 percent (table 3.1).
This surge in the president’s popularity had other effects on the 1996
campaign. Clinton was able to avoid a potentially divisive fight within his
own party for the nomination; he was the first incumbent Democratic pres-
ident since Franklin Roosevelt not to have a significant challenge for
renomination. And by early 1996 Clinton had moved ahead of all of his
potential Republican opponents in the polls. It was a lead that the presi-
dent never relinquished.
1996: THE GENERAL ELECTION CAMPAIGN

August 1996 was the month during which the three most visible presiden-
tial candidates were formally nominated. Dole was formally chosen by the
Republicans on August 14; Perot easily won his party’s nomination by a
margin of two to one over the former governor of Colorado, Richard
Lamm, on August 17; and Clinton was renominated by the Democratic
National Convention on August 28.
One other minor-party candidate received a fair amount of media atten-
tion—Ralph Nader of the Green party. In the spring of 1996, polls showed
that about five percent of the public said that they would vote for Nader, a
well–known consumer advocate. Many Democratic political leaders ini-
tially worried about the Nader candidacy, because they feared that most of
his support would come from Democratic voters who would otherwise
vote for Bill Clinton. During the summer and fall of 1996, however,
Nader’s support faded.
Both the Republicans and the Democrats appeared to gain a surge in
popular appeal from their national conventions—the so-called “convention
bounce.” Dole’s choice of Jack Kemp as his candidate for vice president
also was initially well received both by the public and by Republican
activists. Nevertheless, after the Democratic convention, a new Gallup poll
reported that President Clinton and his running-mate, Vice President Al
Gore, held a 55 percent to 34 percent advantage over Dole and Kemp going
into the final two months of the campaign (See table 3.2).
As the last phase of the campaign got underway, it was clear that
Clinton was continuing to benefit from good news about the American
economy. Price inflation and unemployment were remarkably low: infla-
tion was running at 3.0 percent and unemployment was 5.4 percent.12
In many voters’ minds, however, there were continuing questions about
the Whitewater affair and the controversy over Clinton’s business dealings
in the 1980s in Arkansas. Nevertheless, the main drama during September
centered around plans for the upcoming televised presidential and vice-
presidential debates.
On September 10, Ross Perot selected the author and economist, Pat
Choate, as his vice-presidential running-mate. On September 17, though,
the bipartisan Commission on Presidential Debates decided to exclude
Perot and Choate from the televised debates, on the grounds that only
Clinton and Dole had a realistic chance of winning the election. Perot filed
a lawsuit against the debates commission in the District of Columbia
Federal District Court, but about one week before the first debate, Judge
Thomas F. Hogan ruled against him.
The Clinton and Dole campaigns agreed to schedule two presidential
debates, without Perot, on October 6 and October 16, and one vice-presi-
dential debate on October 9. Millions of Americans watched the debates,
although the audiences were smaller than they had been four years earlier.
Table 3.2: Voter Support for Clinton, Dole, and Perot: The Gallup Poll’s
Three-Way Trial Heats Between February 1995 and Election Eve 1996
Date of Interviews For Clinton % For Dole % For Perot % For Others,
No Opinion %
February 3–5, 1995* 45 51 4
April 17–19, 1995 40 37 18 5
August 4–7, 1995 39 35 23 3
Government shutdown ends, January 6

January 12–15, 1996 43 39 16 2
Clinton’s fourth State of the Union address, January 23

January 26–29, 1996* 54 42 4
March 8–10, 1996 47 34 17 2
Dole clinches Republican nomination, March 26

April 9–10, 1996 49 35 15 1
July 25–28, 1996 50 35 10 5
Republican Convention, August 12–15

August 16–18, 1996 48 41 7 4
Democratic Convention, August 26–29

September 9–11, 1996 55 34 5 6
October 3–4, 1996 51 39 5 5
October 5–6, 1996 55 35 5 5
Presidential and vice-presidential debates, October 6–16

October 17–18, 1996 55 32 8 5
October 20–21, 1996 54 35 6 5
October 30–31, 1996 52 34 10 4
Final Poll 52 41 7 0
Election Results 49 41 8 2
*Perot’s name not included in poll questions.
Source: Data provided by the Gallup poll.

In their first debate, in Hartford, Connecticut, Clinton and Dole sparred

over the economy, education, Medicare, and tax cuts. Dole accused Clinton
of being “a liberal,” but Clinton retorted, “That’s what their party always
drags out when they get in a tight race.”13 Polls indicated that viewers pre-
ferred Clinton’s performance—51 percent thought that Clinton had “won”
the debate, compared with 32 percent who said that Dole had won.14
In another relatively low-key debate on October 9, the vice-presidential
candidates, Gore and Kemp, faced off. Once again the Democrats scored a
debate victory. Polls showed that the voters thought that Gore had “won”
the debate, 57 percent to 28 percent.15
The days leading into the final presidential debate brought good news
and bad news for the Democratic camp. On October 13, the Dow Jones
industrials stock market index broke the 6,000 barrier, an event that the
Democratic party seized upon as another sign that the economy was doing
well. But the Clinton administration’s moments of success often seemed to
be followed by troubles, and this time the difficulty involved campaign
finance. The Republicans began to attack Clinton and the Democrats for
accepting contributions of $485,000 to the Democratic National
Committee from the members of an Indonesian banking family and for tak-
ing other contributions from foreign sources.
The final presidential debate took place on October 16 in San Diego,
California; and this time Dole sharply attacked Clinton over what Dole
said were the administration’s ethical problems and “scandals.” For the
most part Clinton ignored his rival’s attacks. Once again, he led in the post-
debate polls, 59 percent to 29 percent.16 Dole had been counting on the
debates to enable him to catch up with Clinton. But the debates had failed
to narrow the gap.
There were now just nineteen days left before election day. On October
23, Dole sent his campaign manager Scott Reed to Dallas to ask Ross Perot
to drop his presidential bid and endorse the GOP ticket. However, Perot
declared that he was in the race to the finish.
In the final days of the campaign, Clinton urged people to turn out and
vote. He also made campaign stops that he hoped would benefit
Democratic congressional candidates. Dole launched a dramatic last-
minute push to mobilize support—a seventeen-stop, ninety-six-hour sprint
during which he campaigned almost around the clock. Perot’s campaign
finale, on election eve, consisted of four thirty-minute “infomercials” in
which he sharply attacked the Clinton presidency. Perot suggested that
Clinton, if reelected, would spend much of his second term answering
charges of corruption and scandal in his administration. The final polls
indicated that there had been some narrowing of Clinton’s lead. But all of
the major national polls still had Clinton ahead.
On election day, it was sunny or partly sunny over much of the nation,
with high temperatures in the 60s as far north as Nebraska, Illinois, Ohio,
and Maryland. There were showers, however, in the states around the
Great Lakes. Despite the generally favorable weather, the turnout was low.
For the first time in seventy-two years, there were more nonvoters than vot-
ers in a presidential election. When the votes were counted, Clinton had
won by a margin of 8.5 percentage points over Dole and by about 8.2 mil-
lion popular votes.
NOVEMBER 5, 1996—THE PATTERN OF THE VOTE

On election day, 47.4 million people cast their ballots for Clinton and
almost 39.2 million voted for Dole. Perot polled nearly 8.1 million votes, a
sharp drop from the 19.7 million votes he received in 1992, but still an
unusually strong showing for an independent candidate. In the electoral
college, because Perot did not win the popular vote in any state, all of the
electoral votes went to Clinton and Dole. The Democratic nominee
received 379 of the nation’s 538 electoral college votes; and Dole won 159.
Because the presidential popular vote was split three ways, in twenty-six
states no presidential candidate won a majority of the total vote. Of these
states that were won by popular-vote pluralities, Dole carried thirteen, and
Clinton carried thirteen. On the other hand, in the twenty-four states that
were won with more than 50 percent of the total popular vote, Clinton was
the decisive winner. Dole won clear majorities in six states, most with small
populations—Alabama, Alaska, Idaho, Kansas, Nebraska, and Utah.
These states gave the former Kansas senator 32 electoral votes. Clinton, by
contrast, carried eighteen states with a popular-vote majority. These states,
and the District of Columbia, gave Clinton a total of 230 electoral votes—
just 40 short of the number needed for an electoral college majority. Aside
from this overall verdict, there were many noteworthy features in the
November 5, 1996 vote.
THE NATIONAL VOTE

1. About 96.3 million voters went to the polls. As has been pointed
out, the number of people who voted for president was down by
more than eight million from 1992, and the percentage turnout—
49.0 percent—was the nation’s lowest since 1924.
2. President Clinton’s share of the total popular vote—49.2 percent—
was substantially higher than the 43.0 percent that he received in
1992. But he remained one of eleven American presidents to be
elected with less than 50 percent of the vote. He was also one of
only three presidents to win two terms while receiving less than
half the vote each time. The other two presidents who won two
terms without polling at least 50 percent of the vote were Grover
Cleveland, in 1884 and 1892, and Woodrow Wilson, in 1912 and
1916.
3. In the total popular vote for president, Clinton ran 8.5 percentage
points ahead of Dole. However, in the voting dynamics that deter-
mined the crucial outcome in the electoral college, the popular vote
for the two major-party candidates, Clinton defeated Dole by 9.4
percentage points—54.7 percent to 45.3 percent. Clinton’s 9.4
point lead in the two-party vote came close to Franklin Roosevelt’s
two-party vote lead over Wendell Willkie in 1940 (55.0 percent to
45.0 percent). It also somewhat bettered the two-party vote lead
that George Bush had over Michael Dukakis in 1988—53.9 per-
cent to 46.1 percent.
THE PEROT FACTOR

1. As has been noted, the 8.1 million votes (8.4 percent) polled by
Ross Perot was the sixth largest percentage received by a minor-
party or independent presidential candidate since the Civil War.
Even so, in raw numbers, Perot’s 1996 vote was down by 11.6 mil-
lion from what it had been in 1992.
2. It is probable that many 1992 Perot voters did not vote in 1996; in
the 1996 exit polls, only 12 percent of the 1996 voters said that
they had supported Perot in 1992, but Perot’s actual share of the
1992 vote was almost 19 percent.17
3. The exit polls suggest that more than seven million 1992 Perot vot-
ers did go to the polls again in 1996. Among these former Perot
voters, about a third (33 percent) voted for Perot a second time.
But about two-thirds of the 1992 Perot supporters voted for one of
the two major-party candidates in 1996. Furthermore, this sizable
group of voters went for Dole over Clinton by a margin of two to
one, 44 percent to 22 percent.
4. Despite their drop in numbers from 1992, the eight million Perot
supporters in 1996 remained a huge bloc of voters who could be
“up for grabs” in the next presidential election, in the year 2000.
Much depends on whether Perot chooses to run again in 2000, and
whether the Reform party runs another presidential candidate if
Perot drops out.
SECTIONAL PATTERNS
1. In the presidential vote, as in 1992, the most Democratic region
was the Northeast, where Clinton outpolled Dole by 56 percent to
35 percent. Along the East Coast, for example, Clinton carried
every state from Maryland to the Canadian border—a feat that not
even Franklin D. Roosevelt had been able to achieve. Clinton, in
fact, swept the eastern seaboard—from the District of Columbia to
Maine—twice, in 1996 and in 1992.
2. In the Midwest, the outcome was closer. There, Clinton ran ahead
of Dole 48 percent to 41 percent—very close to the division of the
vote in the country as a whole.18
3. Next to the Northeast, the Democrats’ most solid regional strong-
hold was the group of Pacific Coast states of California, Oregon,
and Washington with their seventy–two electoral votes. In this sec-
tion of the country, Clinton led Dole by a decisive margin, 53 per-
cent to 34 percent.
4. With Clinton and Gore as their nominees, the Democrats in 1996,
as in 1992, were running the first all–southern major–party presi-
dential ticket since Andrew Jackson and John C. Calhoun ran
together in 1828. In 1996, the Clinton–Gore ticket again made
inroads into previous Republican strength in presidential voting in
the South. By the narrowest of margins—about 24,200 votes out
of more than 26 million votes cast—Clinton and Gore won the
popular vote in the South.19 It was the first time since 1976, when
the Democrats nominated Jimmy Carter of Georgia for president,
that the Democrats had held their own or ran ahead in the popu-
lar vote in the South. The Democrats’ popular vote in the South
allowed them to carry four states—Arkansas, Florida, Louisiana,
and Tennessee—of the former Confederacy.
5. Nevertheless, the South remained a very important sectional base
for the Republican party in presidential voting. In the eleven states
of the former Confederacy, Dole won a solid majority of the elec-
toral votes and, as noted, ran almost even with Clinton in the pop-
ular vote—46.1 percent to 46.2 percent. In the rest of the country,
Clinton led Dole by a sizable margin—50.4 percent to 38.7 per-
cent.
6. In the Rocky Mountain states, a region where Ronald Reagan car-
ried most of the states by landslide margins in the 1980s, Dole ran
ahead of Clinton in 1996—by 47 percent to 43 percent. Perot won
9.6 percent of the vote in the Rocky Mountain states, a big drop
from his showing in 1992 when he polled nearly one vote in every
four in the region. Nevertheless, among the eight states in the
Rocky Mountain region Clinton was the winner in three—New
Mexico, Nevada, and Arizona. No Democrat had carried Arizona
since Harry Truman won there in 1948.
7. The Clinton tide ran strongly in the East, Midwest, and the Pacific
Coast region. It also enabled Clinton to carry at least some states
in most major sections of the country. However, there were impor-
tant centers of Dole strength in four areas—most of the Southeast,
Indiana, a string of six states in the center of the country running
from Texas to the Canadian border, and five of the eight Rocky
Mountain states.
8. In the popular vote New England was the strongest Democratic

section and the Rocky Mountain region was the best area for the
Republicans. There were greater sectional differences in the popu-
lar vote and popular vote swing in 1996 than in 1992. This rise in
sectionalism, according to Harvey Schantz, “is mainly a reflection
of the increased individuality of the Middle Atlantic states and
especially New England. The Mountain states were also quite dis-
tant from the national vote distribution in 1996.” 20
SOCIAL GROUPS
1. As in 1992, Clinton did well with some traditionally Democratic
groups among which Reagan had made heavy inroads in the
1980s. Among members of union households, Clinton led 59 per-
cent to 30 percent.21 Clinton also did well among two groups that
had remained Democratic in the 1980s and in 1992: Clinton was
backed by 84 percent of African-American voters and by 78 per-
cent of Jewish voters.
2. Hispanic Americans, who voted 62 percent to 24 percent for
Clinton over Bush in 1992, were even more pro-Clinton four years
later as they backed Clinton over Dole by more than three to one—
72 percent to 21 percent. Hispanic Americans were also one of the
few major groups in the United States whose voting turnout
increased in 1996. Close to five million Hispanic Americans went
to the polls in 1996 (compared with four million in 1992); and vot-
ing shifts toward Clinton in Hispanic American neighborhoods
were a key factor in President Clinton’s victories in Florida and
Arizona.22
3. The contribution made by African-American voters to the
Democratic presidential victory was particularly noteworthy. One
in every ten Americans who went to the polls was African-
American. This large group voted 84 percent for Clinton, and only
12 percent for Dole. The exit polls suggested that African
Americans gave Clinton a margin of about 6.9 million votes, or
most of his 8.2 million margin of victory.
4. The state of the economy was strongly reflected in the returns. One
third of the voters (33 percent) reported that their family’s finan-
cial situation was better in 1996 than it had been in 1992. Among
that large group of voters, Clinton led Dole by 66 percent to 26
percent. There was a smaller group of voters (20 percent), howev-
er, who said that their family’s financial situation was worse in
1996 than in 1992. Those voters opted for Dole over Clinton, 57
percent to 27 percent.
5. Clinton made gains in 1996 in one group that had been a key com-
ponent of the Roosevelt New Deal coalition but which had been
less strongly Democratic in recent years—Roman Catholics.

Nearly three voters in every ten (29 percent) were Roman Catholic;
and they supported Clinton over Dole 53 percent to 37 percent.
Among Protestants, by contrast, Dole had a moderate lead over
Clinton, 47 percent to 43 percent. One source of Clinton’s stronger
showing among Catholics in 1996 was undoubtedly the surge of
support the president received from Hispanic American voters.
6. Another aspect of the voters’ religious identifications left a deep
imprint on the returns. In 1996 voters who were interviewed as
they left the polls were asked whether they considered themselves
members of the “Religious Right.” Seventeen percent said “yes”;
80 percent said “no.” Among the self-identified members of the
Religious Right, Dole led Clinton by a decisive margin—65 percent
to 26 percent. Among the larger group who did not consider them-
selves adherents of the Religious Right, Clinton ran well ahead of
Dole, by a margin of 54 percent to 35 percent.
7. In discussions of American politics in recent years, the voting pref-
erences of members of the “Religious Right” have, quite correctly,
received a great deal of attention. The 1996 exit polls suggest,
however, that when examining the voters’ religious preferences,
there are two other groups in American society that should be
looked at for their impact on American presidential elections. In
1996, some 87 percent of the voters said that they were
Protestants, Catholics, or Jews. But another six percent said that
they belonged to “other religions,” and seven percent said that
they had no religion. These latter two groups together accounted
for 13 percent of the American electorate—and they were a cluster
of voters three-fourths the size of the self-identified Religious
Right. In 1996, adherents of other religions and those professing
no religion voted heavily for Clinton, giving the president close to
60 percent of their votes compared with 23 percent for Dole.
8. Perhaps the most striking feature of all in the 1996 election was the
“gender gap”: the tendency of men and women to vote differently
in the presidential race. Among male voters, Clinton and Dole
polled an almost equal share of the vote. Men voted 44 percent for
Dole; 43 percent for Clinton; and 10 percent for Perot. Among
women, however, the president led Dole by sixteen percentage
points—54 percent to 38 percent; with 7 percent for Perot. If all
the voters had voted the way that American women cast their bal-
lots in 1996, Clinton would have won the election by more than
15 million votes, instead of by 8.2 million.
PARTY LOYALTY AND DEFECTION

1. Despite the Clinton victory, the major change in party identifica-
tion that took place in the mid-1980s was once again strongly evi-
dent in the returns. Among those who voted, self-described
Democrats outnumbered Republicans by only a moderate mar-
gin—39 percent to 35 percent. This moderate margin was certain-
ly helpful to Clinton, because it meant that there were probably
about four million more Democrats than Republicans at the polls.
But it was a major change from the large advantage the
Democratic party used to have in the number of party identifiers.
As recently as 1980 (and in spite of Ronald Reagan’s presidential
victory that year), the exit polls suggested that 43 percent of the
voters considered themselves Democrats, compared with 28 per-
cent who said they were Republicans.23 By the second half of the
1980s, however, the Democratic party had clearly lost the large
lead in party identification that it had enjoyed for more than four
decades.
2. Nevertheless, although the number of Republican identifiers had
grown substantially since the early 1980s, Dole suffered some sig-
nificant defections among Republican voters in 1996. Eighty per-
cent of the Republicans supported their party’s presidential ticket
in 1996; but 13 percent voted for Clinton and 6 percent defected
to Perot.24 Clinton did slightly better than Dole in holding his
party’s identifiers, winning 84 percent of the votes of Democrats,
while losing 10 percent to Dole and 5 percent to Perot. It was only
the second time since the Johnson landslide over Goldwater in
1964 that the Democrats mobilized a higher percentage of their
partisans than did the Republicans. Clinton also had greater suc-
cess with Democratic voters than Bush had in mobilizing the
Republicans in 1992.25
3. Clinton also out-polled the Republican ticket among voters who
considered themselves independents, as he had in 1992. His lead
over Dole among independents was eight points (43 percent to 35
percent)—about what his lead was among the electorate as a
whole.
4. In 1996 Ross Perot showed some strength among all three types of
voters—Democrats, Republicans, and independents. He won the
support of 5 percent of the Democrats and 6 percent of the
Republicans. Among independents, the Perot share rose to 17 per-
cent.
CONTROL OF GOVERNMENT
1. Despite the Democratic presidential victory for Bill Clinton, in
many of the contests for public offices other than the presidency,
Republican candidates did well in 1996. The election left the
Republican party in control of both the U. S. Senate and the House
of Representatives, and it also left the GOP with a strong lead in
the nation’s governorships.
2. In a close, hard-fought battle, the Republicans won the important
contest for control of the House of Representatives. The
Democrats made a modest net gain of seats to bring the House
membership to 207 Democrats and 227 Republicans, and one
independent member. The Republicans’ overall majority in the
House was the narrowest House majority held by either party since
1954, and it narrowed further in the midterm election of 1998, as
the partisan split in the new 106th Congress was 223 Republicans,
211 Democrats, and one independent.
3. In the Senate, the Republicans gained two seats, increasing their
margin of control. The new Senate had 55 Republicans and 45
Democrats. The election left the new Senate Majority Leader Trent
Lott of Mississippi, former Senator Dole’s successor, with the
largest Republican Senate majority since 1929. The partisan divi-
sion of seats was not changed in the 1998 midterm election. In
2000, 19 Republican seats and 14 Democratic seats will be up for
election.
4. Perhaps the most important point of all about the 1996 election
was that for two years, at least, there would continue to be divid-
ed government in Washington. The Democratic party was in con-
trol of the presidency; the Republicans controlled both houses of
Congress. The situation reminded some observers of the electoral
outcome of 1972, when President Richard Nixon and an opposi-
tion Democratic Congress were both reelected just as the
Watergate scandal began to unfold. At the very least, the 1996
electoral verdict meant that during a period when an almost
unprecedented series of investigations of the president were under-
way, the chairs of congressional committees who might have to act
on the results of those investigations were members of the opposi-
tion party. The 1996 election results were thus an invitation to
struggle between the executive and the legislative branches of the
government. On the other hand, given the closeness of the balance
of political forces left by the returns, the outcome also gave lead-
ers in both parties reasons for caution, and even for a measure of
cooperation. How the Democratic president and the Republican
congressional leaders shared their power would go a long way toward

determining how their parties would fare in future elections.
EXPLAINING THE PRESIDENTIAL ELECTION OUTCOME OF

1996: “IT’S THE PRESIDENT’S JOB RATING. . . .”
Throughout nearly all of the 1996 campaign year—from March 1996 to
the end of October—President Clinton maintained a lead of at least 12 per-
centage points over his Republican opponent, Robert Dole, in the Gallup
poll. The only exception occurred in a single poll taken in August, imme-
diately following the Republican National Convention. In that one poll,
Dole cut the Clinton lead to seven percent (table 3.2).
Journalists who were following the presidential campaign found it a
frustrating contest to cover because they thought it was so difficult to make
it interesting for their readers. Republican activists who worked in the Dole
campaign were also frustrated because they could never seem to close the
gap between their presidential ticket and the Democrats. As the chairman
of the Republican National Committee, Haley Barbour, put it after the elec-
tion was over, “We started (presidential) polling in April, and we never
measured Dole above 41 percent in a single survey, even during the [GOP]
convention.26 Two political scientists who studied the 1996 campaign, Paul
S. Herrnson and Clyde Wilcox, called their analysis “A Tale of a Campaign
That Didn’t Seem to Matter.”27 Until the last few days of the campaign, the
incessant poll results hardly seemed to move at all. Then, in the final week,
Clinton’s lead over Dole narrowed a bit, but not enough to make the elec-
tion close.
After the election, there was some discussion of whether, with different
candidates and different campaign strategies, the Republican party could
have made the outcome different. At the very least, it is fair to say, for the
Republicans to have won in 1996 would have been very, very hard. The
fact is that Clinton probably won the November 1996 election sometime
between early 1995 and the spring of 1996. It was during this period that
the way the American people viewed Clinton’s job performance as presi-
dent moved from clearly negative to clearly positive.
This movement of opinion can be traced in table 3.1, which sets forth
the responses of the American public to the question: “Do you approve or
disapprove of the way Bill Clinton is handling his job as president?” A sim-
ilar question has been asked concerning the president’s job performance for
every president since Franklin Roosevelt. As the poll numbers in table 3.1
make clear, Clinton’s job rating at the end of 1994 was not good—40 per-
cent of the American public approved of the job he was doing, 52 percent
disapproved.
In April 1995, following the administration’s firm response to the
Oklahoma City bombing tragedy, the president’s poll ratings moved into
positive territory (51 percent approving, versus 39 percent disapproving).
During the next nine months, the president’s approval ratings sometimes
sagged again, most notably at the beginning of January 1996, in the mid-
dle of the government shutdown confrontation with the Republican
Congress. However, other polls taken at the time suggest that more of the
public blamed Republican congressional leaders than blamed Clinton for
that impasse.28 And when the confrontation over the government shutdown
was over, President Clinton soon began to enjoy the highest job ratings he
had had in two years. Moreover, although his job rating slipped a little just
before the November election, his approval ratings remained strongly tilt-
ed toward the positive side. Looking at President Clinton’s job ratings dur-
ing his first four years as president, the overall pattern is clear. It was
almost as though the collective judgment of the American public was: first
two years, job performance “poor”; last two years, job performance “pret-
ty good.”
As the final set of poll numbers in table 3.1 indicates, President Clinton
went into the last week of the 1996 election campaign with an approval
rating of 54 percent. In an election in which the vote for minor-party pres-
idential candidates reached 10 percent, Clinton did not get 54 percent of
the total vote. (His share of the total was 49.2 percent.) Clinton did, how-
ever, get 54.7 percent of the two-party vote.
For other presidents, also, there has been a fairly close relationship
between the way the American public rated the president’s job performance
and the verdict of the voters when the president sought reelection. During
the second half of the twentieth century, eight American presidents have
fought election campaigns as an incumbent attempting to be returned to
the White House. Two of those presidential candidates, Lyndon Johnson
and Gerald Ford, had ascended to the presidency through the death or res-
ignation of the previous president. The other six presidents seeking reelec-
tion—Eisenhower, Nixon, Carter, Reagan, Bush, and Clinton—had been
elected to a first term.
In table 3.3, the job approval ratings of these eight presidents as they
approached election day are set forth. The percentage of the total vote
these presidents received and the election outcome are also listed. Five pres-
idents—Eisenhower, Johnson, Nixon, Reagan, and Clinton—had positive
approval ratings (above 50 percent) as they attempted to win another term
in the White House. All five were returned to office. On the other hand,
three incumbent presidents approaching election day clearly were in trou-
ble with the voters, with mediocre job ratings (Ford, 45 percent approval),
or very low ratings (Carter, 37 percent; and Bush, 36 percent). All three of
those presidents were defeated for reelection.
Table 3.3: Presidents Seeking Reelection; Their Job Ratings; and the Outcome of the Presidential Election: Presidential Approval Ratings in an Election Year
and the President’s Share of the November Presidential Vote, 1956–1996
Election President Seeking Date of Job Rating President’s Job Rating* President’s % of Total Outcome of
Year Reelection Interviews % Who % Who Vote in November the Election
Approve Disapprove
1956 Eisenhower (08/03–08/56) 67% 20% 57.4% PRESIDENT
REELECTED
1964 Johnson (06/04–09/64) 74 13 61.1 PRESIDENT
REELECTED
1972 Nixon (mid–June, 72) 59 30 60.7 PRESIDENT
REELECTED
1976 Ford (06/11–14/76) 45 40 48.0 PRESIDENT
DEFEATED
1980 Carter (Sept., 80) 37 55 41.0 PRESIDENT
DEFEATED
1984 Reagan (10/26–29/84) 58 33 58.8 PRESIDENT
REELECTED
1992 Bush (09/21–24/92) 36 54 37.4 PRESIDENT
DEFEATED
1996 Clinton (10/26–29/96) 54 36 49.2** PRESIDENT
REELECTED
*Responses were to the question: “Do you approve or disapprove of the way ___________ is handling his job as president?”
**In 1996 Clinton polled 54.7% of the two-party vote.
Source: The Gallup poll presidential approval ratings are drawn from the following sources: George H. Gallup, The Gallup Poll: Public Opinion, 1935-1971
(New York: Random House, 1972), 1441, 1885 (data for 1956 and 1964); Gallup Opinion Index, October-November, Report No. 182, 19 (data for 1972);
Gallup Opinion Index, July, 1976, Report No. 132, 2 (data for 1976); Gallup Opinion Index, October-November, Report No. 182, 13 (data for 1980);
Gallup Report, December, 1984, Report No. 231, 11 (data for 1984); The Gallup Poll Monthly, September, 1992, 39 (data for 1992); and the Gallup Web
Site, on May 8, 1997, http://www.gallup.com/ratings/(data for 1996).
In 1996, President Clinton’s overall favorable job rating undoubtedly

stemmed from a complex blend of factors and public motivations—from
favorable evaluations of the state of the economy, to positive voter respons-
es to Clinton’s skills as a campaigner. But a president who had 54 percent
of the American public who approved of the job he was doing as chief exec-
utive was going to be extremely hard to beat.
CONCLUSION
The preceding discussion may help to explain why President Clinton
coasted to a relatively easy reelection victory in 1996. But it still leaves
unanswered two major questions. What was the fundamental nature of the
verdict handed down by the American electorate in 1996? And, amid the
large number of elections that make up the total “population of elec-
tions,”29 how is the 1996 election to be classified?
One famous typology of elections that might shed light on the nature of
the electoral verdict in 1996 was put forward many years ago. V. O. Key
Jr. once classified presidential elections into three broad types: (1) Votes of
Lack of Confidence; (2) Reaffirmations of Support by Votes of Confidence;
and (3) Realignments. The 1996 presidential election appears to be a reaf-
firmation of support by a vote of confidence. In this type of election, as
defined by Key, “substantially the same coalition of voters prevails as pro-
vided the majority in the preceding election. Such an election may be
regarded as a vote of confidence in the general course of action the
Administration has followed.”30
As we have seen, there were interesting nuances in the vote patterns of
1996 that reflected some change from 1992 to 1996 in the social and geo-
graphical groupings that produced a victory for Clinton in both election
years. On the other hand, the most striking aspect of Clinton’s two elec-
tions was the similarity of the voting coalitions that put him into power
two times. In 1996, the fundamental character of the electoral verdict was
a reaffirmation of support, however qualified, for the Clinton administra-
tion. The election returned Clinton and the Democratic party to power in
the executive branch, and it gave them four more years to take on the risks
of governing. Four years later, to be sure, Clinton would not be running
again. But how well his administration was perceived as managing the gov-
ernment would go a considerable way toward determining whether the
Democratic party would receive another reaffirmation of support at the
presidential level—or a “Vote of No Confidence”—in the election of 2000.
NOTES
1. Gerald M. Pomper, “The Presidential Election,” in The Election of 1996:
Reports and Interpretations, ed. Pomper (Chatham, NJ: Chatham House, 1997),
173.
2. All 1996 election returns are from Richard M. Scammon, Alice V.
McGillivray, and Rhodes Cook, America Votes 22 (Washington, DC: Congressional
Quarterly Inc., 1998).
3. Richard M. Scammon and Alice V. McGillivray, America Votes 20
(Washington, DC: Congressional Quarterly Inc., 1993), 9; and Scammon,
McGillivray, and Cook, America Votes 22, 1.
4. The composition of the electorate is based on the Voter News Service exit
poll available at www.cnn.com/ALLPOLITICS/1996 polls. For a comparison to
the electorate of 1992, see Milton C. Cummings Jr., “Political Change Since the
New Deal: The l992 Presidential Election in Historical Perspective,” in American
Presidential Elections: Process, Policy, and Political Change, ed. Harvey L. Schantz
(Albany: State University of New York Press, 1996), 51–53.
5. This narrative account of Clinton’s first three years in office and the 1996
campaign and vote draws heavily on Milton C. Cummings Jr. and David Wise,
Democracy Under Pressure: An Introduction to the American Political System, 8th
ed. (Fort Worth: Harcourt Brace College Publishers, 1997), 405–420.
6. Pomper, “The Presidential Election,” 173.
7. George J. Church, “The Democrats: The Learning Curve,” Time,
September 2, 1996, 33.
8. Responses to the question: “Do you approve or disapprove of the way
__________ is handling his job as President?” Gallup Opinion Index, Report No.
180, August 1980, 26; Gallup Report, December 1983, No. 219, 18; and data pro-
vided for 1992 by the Gallup Poll.
9. Data supplied by the Gallup Poll.
10. See John T. Tierney, “The Context: Policies and Politics, 1993–1996,” in
America’s Choice: The Election of 1996, ed. William Crotty and Jerome M. Mileur,
(Guilford, CT: Dushkin/McGraw-Hill, 1997), 24–25.
11. Church, “The Democrats,” 32.
12. U.S. Bureau of Labor Statistics.
13. “A Closer Look at the Debate,” USA, October 7, 1996, A13.
14. Susan Page, “Debaters Clash Over Vision, Rivals Cite Fundamental
Differences,” USA, October 7, 1996, A1.
15. www.allpolitics.com, October 9, 1996.
16. Bill Nichols, “Dole on Offense in Debate,” USA, October 17, 1996, A1.
17. This section is based on the Voter News Service Exit Poll available at
www.cnn.com/ALLPOLITICS/1996polls.
18. The Midwest includes the five Great Lakes states of Illinois, Indiana,
Michigan, Ohio, and Wisconsin, the Prairie states of Kansas, Nebraska, South
Dakota, and North Dakota, along with Minnesota, Iowa, and Missouri.
19. The South, as here defined, includes the eleven states of the former
Confederacy—Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, North
Carolina, South Carolina, Tennessee, Texas, and Virginia.
20. Harvey L. Schantz, “An Update on Sectional Voting: The 1996
Presidential Election,” Party Developments, September 1997, 21–22.
21. This section is based on the Voter News Service exit poll.
22. B. Drummond Ayres Jr., “The Expanding Hispanic Vote Shakes
Republican Strongholds,” NYT, November 10, 1996, Section 1, 27.
23. The party identification data cited for 1980 came from exit polls of
12,782 voters conducted by CBS News and the NYT; reported in NJ, November 8,
1980, 1878.
24. This section is based on the VNS exit poll.
25. See Cummings, “Political Change Since the New Deal,” 78.
26. Quoted in Larry J. Sabato, “The November Vote—A Status Quo
Election,” in Toward the Millennium: The Elections of 1996, ed. Larry J. Sabato,
(Boston: Allyn and Bacon, 1997), 155.
27. Paul S. Herrnson and Clyde Wilcox, “The 1996 Presidential Election: A
Tale of a Campaign That Didn’t Seem to Matter,” in Toward the Millennium,
121–142.
28. George Hager, “Republicans Throw in Towel on Seven-Year Deal,”
CQWR, January 27, 1996, 213–216, esp. 215.
29. The phrase “population of elections” is V. O. Key’s. See Key, “The
Politically Relevant in Surveys,” Public Opinion Quarterly 24 (Spring 1960):
54–61, at 55.
30. V. O. Key Jr., Politics, Parties, and Pressure Groups, 5th ed. (New York:
Crowell, 1964), 522–536, quote 526.
CHAPTER 4
Strategic Partisan Decisions and Blunted

National Outcomes
The 1996 Senate Election Campaign and Vote
DOUGLAS B. HARRIS
INTRODUCTION
The elections for Senate seemed to be the bright spot for the Republican
party in the otherwise troubled 1996 elections. Where Republican presi-
dential nominee Bob Dole lost decisively to President Bill Clinton and
where Republicans lost seats in the House of Representatives, Republicans
actually extended the number of Senate seats they held. Senate Republicans
were able to capitalize on an extraordinary number of opportunities in an
otherwise unfavorable political climate, picking up two Senate seats. Put in
the context of the other national election outcomes, the Republican gains
in the Senate are impressive indeed.
However, when one views the great many opportunities Senate
Republicans squandered and how short they fell of their earlier stated hope
of a filibuster-proof majority, it seems Republicans were lucky to have
extended their Senate majority at all. A tide of anti-Republican sentiment—
blunted only in the last weeks of the campaign by growing evidence of cam-
paign finance irregularities by the Democratic party and some backlash
against the activities of organized labor on behalf of Democratic candi-
dates—kept Republicans from capitalizing as well as they might have on a
host of remarkable opportunities for gains in Senate seats.
THE OUTCOMES OF THE 1996 SENATE ELECTIONS

Thirty-five Senate seats were determined during the 1995–1996 election
cycle. Thirty-three of those seats were scheduled to be up for election in
1996 and the resignations of Senators Bob Packwood (R-OR) and Bob
Dole (R-KS) increased that number to thirty-five. Packwood’s Oregon
Senate seat was filled in a January 1996 special election. The net outcome
of these races was a one seat increase for Senate Republicans. In the
November elections (excluding the race to succeed Packwood),
85
86 Strategic Partisan Decisions and Blunted National Outcomes
Republicans achieved a net gain of two seats. Republicans endured the

defeat of one Senate incumbent, but were able to pick up three seats previ-
ously held by Democrats in open Senate races.
THE NATIONAL VOTE

In the thirty-three states that had Senate races, 51,291,071 votes were cast.
Over 25.3 million (49.4 percent) of those votes were for Republicans while
just over 24.5 million (47.8 percent) were for Democrats.1 Despite the nar-
row vote margin in terms of votes cast, Republicans won nearly two-thirds
of the Senate races. Of course, the apportionment of Senate seats based on
states rather than population explains a great deal of the disparity between
votes and seats in Senate elections.
THE RACES
The relatively little partisan turnover in the 1996 elections that saw
Republicans extending their Senate majority from 53 to 55 seats masks the
fact that with few exceptions the 1996 Senate elections were quite compet-
itive. There were a handful of races, however, in which returning Senate
incumbents fended off challengers handily. Incumbent Senators Ted Stevens
(R-AK), Joseph Biden (D-DE), Thad Cochran (R-MS), Pete Domenici (R-
NM), Fred Thompson (R-TN), and John D. Rockefeller IV (D-WV) all
won reelection with over 60 percent of the two–party vote. And several
other Senate incumbents—Larry Craig (R-ID), Mitch McConnell (R-KY),
Carl Levin (D-MI), and James Inhofe (R-OK)—had little trouble winning
reelection.
Despite these convincing victories by incumbents, the 1996 Senate elec-
tions were remarkably competitive. In Georgia, Secretary of State Max
Cleland defeated Republican Guy Millner by 30,024 votes, 1.3 percent of
the total vote. In Louisiana, Democratic State Treasurer Mary Landrieu
secured victory over Republican Louis Jenkins with a margin of only 5,788
votes, or 0.3 percent of the total vote. And in New Hampshire, former
Democratic Representative Dick Swett failed to unseat incumbent
Republican Senator Robert Smith; Smith bested Swett by less than 15,000
votes, little more than 3.0 percent of the total vote. In fact, the New
Hampshire race was so close that on election night several news organiza-
tions as well as the Democratic Senatorial Campaign Committee (DSCC)
projected that Swett had unseated Smith only to have to retract hours later,
announcing Smith’s narrow reelection victory. A DSCC fax sent out on
election night proclaimed: “For the first time in 20 years, New Hampshire
will be sending a Democrat to the United States Senate with former Rep.
Dick Swett’s victory ... over GOP Sen. Bob Smith ... Smith was unable to
defend his record as an ineffective legislator who did not represent the
interests of average New Hampshire citizens.”2 In all, 26 of the 34 Senate
elections were won with less than 60 percent of the vote. And of those 26,
sixteen were won with less than 55 percent of the vote. In 1996, over three-
fourths of the races were competitive.
REPUBLICAN GAINS
The net outcome of these largely close Senate races was that Senate
Republicans, who had won a majority in the 1994 elections, were able to
extend that majority by two seats. As in recent years, in 1996 the South
was of particular importance in Republican gains. Of the three Republican
victories for Senate seats previously held by Democrats, two were in south-
ern states. Four key retirements by southern Democratic Senators J.
Bennett Johnston (D-LA), Howell Heflin (D-AL), David Pryor (D-AR), and
Sam Nunn (D-GA) forced Democrats to defend these vacated seats in the
increasingly Republican South. Democrats lost the seats vacated by Heflin
and Pryor to Alabama Republican Jeff Sessions and Arkansas Republican
Tim Hutchinson. Moreover, Democrats were barely able to hold onto the
Senate seats vacated by Johnston and Nunn, as Democratic candidates
Mary Landrieu of Louisiana and Max Cleland of Georgia won the two
closest 1996 Senate races. Landrieu won with 50.2 percent of the vote
while Cleland won with only 50.7 percent of the two-party vote. The close-
ness of their races should not diminish the significant accomplishments of
Landrieu and Cleland, who became only the second and third non-incum-
bent Democratic Senate candidates since 1978 to have succeeded another
southern Democrat to the Senate (the other was Louisiana Democratic
Senator John B. Breaux, who won a 1986 election to succeed retiring
Senator Russell B. Long to the Senate).3
Nevertheless, had it not been for Republican gains of two southern
Senate seats previously held by Democrats, there would have been no net
change in Senate seats, as the Republican gain in Nebraska was cancelled
by the Democratic gain in South Dakota. Ultimately, the South was crucial
in the extension of the Republican Senate majority.
THE SETTING: STRATEGIC DECISIONS AND THE

ESTABLISHMENT OF OPPORTUNITIES
In elections, the public does not make choices that translate unencumbered
into electoral victories or governing mandates. Different institutional struc-
tures—from the Constitutional requirements of equal apportionment of
Senate seats to states and staggered terms, to the strategic decisions of
incumbents (to seek reelection or retire) and potential challengers (to run
or not to run)—channel the effects voters have on individual Senate races
and the national verdict.
88 Strategic Partisan Decisions
THE LUCK OF THE DRAW: WHICH STATES HAVE SENATE RACES IN 1996?
In their study of Senate elections, Alan I. Abramowitz and Jeffrey A. Segal
wrote: “The more seats the president’s party must defend of the thirty-three
or thirty-four Senate seats at stake in a given year, the more seats it can
expect to lose.”4 Strictly speaking, Republicans had more to lose in the
Senate races of 1996. Of the thirty-four Senate seats to be filled in the
November 1996 elections, nineteen were held by Republicans and fifteen
were held by Democrats. This, however, is the only point at which
Democrats had an advantage over Republicans.
First of all, the fact that Republicans had 19 seats up indicates that many
of those states electing senators were Republican states more generally. In
1996, Republican presidential candidate Robert Dole won 19 states while
President Bill Clinton won 31 states in his victorious reelection effort.
Moreover, Clinton won 49.2 percent of the total popular vote to Dole’s
40.7 percent. But for many reasons, Clinton’s strength at the top of the
ticket did not translate into Democratic strength for other offices. Weak
political parties and candidate-centered elections lead the list of possible
explanations for a lack of presidential coattails. Another explanation, how-
ever, is just as important in identifying why Republicans ran so well in
Senate elections despite the overall weakness of Dole’s showing in the pres-
idential race: of the 33 states that had Senate races in 1996, a dispropor-
tionate number were states in which Dole won. Of the 19 states that Dole
won, 16 of them were states that had a total of 17 Senate races (table 4.1).
And of the 31 states won by President Clinton, only 17 had Senate races in
1996. Thus, although Clinton won 31 states, the 34 Senate races in
November 1996 were split evenly between states won by Clinton and states
won by Dole.
Moreover, a sizable proportion of the states that had Senate races in
1996 were never won by Clinton (table 4.1). Thirteen states with fourteen
Senate races in 1996 voted for President George Bush in 1992 and Dole in
1996. Only three states that had been won by both Bush and Dole had no
Senate race in 1996. Where there were 29 states that President Clinton won
in both 1992 and 1996, twelve of those 29 had no Senate seats up in 1996.
And Democrats were disadvantaged in 1996 among the five states that
voted for Clinton in only one of his two presidential elections. There were
only two states Clinton won in 1996 that he had failed to capture in 1992;
neither of those states had a Senate seat up in 1996. There were three states
(Colorado, Georgia, and Montana) that Clinton won in 1992 but lost in
1996; all three elected senators in 1996. Bill Clinton’s national electoral
strength did not translate well into Senate races because, in part, Clinton
was not particularly strong in half of the states which had Senate races.
Table 4.1: States with 1996 Senate Elections by Presidential Performance in

1992 and 1996
States won by
Bush 1992 and Clinton 1992 and Bush 1992 and Clinton 1992 and
Dole 1996 Dole 1996 Clinton 1996 Clinton 1996
1996 Senate Race
14 (13 states) 3 states 0 states 17 states
Alabama, Alaska, Colorado, Arkansas, Delaware,

Idaho, Kansas (2), Georgia, Illinois, Iowa,
Mississippi, Nebraska, Montana Kentucky, Louisiana,
North Carolina, Maine, Massachusetts,
Oklahoma, South Michigan, Minnesota,
Carolina, South New Hampshire, New
Dakota, Texas, Jersey, New Mexico,
Virginia, Wyoming Oregon, Rhode Island,
Tennessee, West
Virginia
No Senate Race
3 states 0 states 2 states 12 states
Indiana, Arizona, California, Connecticut,

North Dakota, Florida Hawaii, Maryland,
Utah Missouri, Nevada,
New York, Ohio,
Pennsylvania, Vermont,
Washington, Wisconsin
THE VALUE OF INCUMBENCY
Despite a handful of incumbent losses in House races, 1996, like many pre-
vious national election years, might well have been dubbed “the year of the
incumbent.” Where incumbents typically fare better in House races than in
Senate races, 1996 proved a good year for incumbent senators as well as
House members. Of the twenty incumbent senators seeking reelection,
nineteen were successful. Thus, the percentage of incumbents winning the
general election was 95.0 percent.5 Only twice in post-war Senate elections
has the Senate incumbent reelection rate been higher; in 1960 when 96.6
percent of incumbents seeking reelection won, and 1990 when 96.9 percent
of incumbents seeking reelection were successful.6
The only incumbent to lose the general election was Republican Senator
Larry Pressler of South Dakota who was unseated by that state’s only
House member for the last ten years, Democrat Tim Johnson. Pressler,
Chairman of the Senate Committee on Commerce, Science and
Transportation, had hoped to capitalize on a significant fundraising advan-
tage. Pressler’s committee had overseen the overhaul of a telecommunica-
tions law that attracted the interest of many telecommunications corpora-
tions that contributed significantly to members of Congress, particularly
members of the Committee on Commerce, Science and Transportation. In

the 1995–1996 election cycle, Pressler raised and spent $1.49 for every dol-
lar raised and spent by Johnson.
But where Pressler’s position helped him secure a considerable fundrais-
ing advantage over his formidable challenger, it also opened the incumbent
to criticism. In particular, Johnson questioned Pressler’s sources of cam-
paign funds by tying donations to the telecommunications bill; Johnson
said, “He’s taken over $400,000 in contributions from telecommunications
companies. You don’t have to be a hopeless cynic to think there’s a tie-in
here.”7 Also, Pressler’s support of the elimination of federal funding for
public broadcasting was an issue Johnson and other Democrats raised to
arouse the party faithful and persuade moderate voters. In March 1996,
DSCC Chairman Kerrey predicted that Pressler’s criticism of public televi-
sion would cost him in the November elections; Kerrey said, “The long-
standing critic of Big Bird is going to get the big bird.”8
Although Pressler enjoyed many of the advantages of incumbency,
Johnson’s victory should not have been a great surprise. As the only House
member representing South Dakota, Johnson had represented the same
constituency as Senator Pressler for the last ten years. And Johnson had
typically garnered a higher percentage of the South Dakota electorate in his
statewide races than had Pressler in his Senate races. Of course, this may
partly be due to the fact that—as senator—Pressler consistently had drawn
better challengers. Nevertheless, Johnson had proven himself popular with
the constituency he shared with Pressler. In fact, in 1990—the one election
in which they ran simultaneously—Johnson received 67.6 percent of the
vote, while Pressler received only 53.8 percent of the vote.
The case of Johnson versus Pressler is very much an example of an
exception that proves a rule. Even in the one race in which a Senate incum-
bent was defeated in pursuit of reelection, that senator lost to another
incumbent—a House member—who represented the same constituency.
Where Abramowitz and Segal’s general point—that the more seats a presi-
dent’s party must defend the more it will likely lose—is sound, evidence
from 1996 and some general lessons on the value of incumbency in con-
gressional elections suggest that we should take into account the number of
each party’s incumbents seeking reelection that year as well. Where
Republicans had 19 seats at stake and Democrats had only 15, that advan-
tage was quickly negated by the fact that Republicans had 13 of 19 incum-
bent Senators running for reelection while Democrats had only 7 of 15
incumbent Senators returning. This left Republicans the task of protecting
only six open Senate seats while Democrats had to protect eight open seats.
Democrats were able to defend all of their incumbents, but only five of
their eight open seats. Republicans, on the other hand, were successful in
defending all six of their open seats and twelve of their thirteen returning
incumbent Senators. The value of incumbency aided Republicans in pro-
tecting the lion’s share of their seats in play, while Democrats had to defend
more than half of their seats up in 1996 without a returning incumbent.
EARLY EXPECTATIONS AND CANDIDATES’ STRATEGIC DECISIONS: WHAT A

DIFFERENCE A YEAR MAKES
In an era in which congressional elections are largely candidate-centered, a
political party’s electoral prospects are governed primarily by its ability to
recruit strong candidates. Although such recruitment activity is weak in the
United States compared to similar activity in other industrialized democra-
cies, the electoral prospects of Democrats and Republicans at a given time
can do much to encourage strong candidates to run under the party’s label.
Gary Jacobson argued that “when the partisan outlook is gloomy, shrewd
and ambitious politicians figure that the normally long odds against defeat-
ing an incumbent are even worse than usual and wait for a better day. . . .
Politicians of the other party, sensing that electoral tides are moving in their
direction, view the chances of winning as better than usual, so more and
better candidates compete for the nomination to challenge incumbents.”9
And party prospects also affect whether or not a party’s incumbents decide
to retire or run for another term in the legislature. Evidence from House
elections is telling in this regard. Michael K. Moore and John R. Hibbing,
studying retirements from the House from the 1960s through 1992, found
that “retirements generally occur on the basis of rational calculation con-
cerning advancement potential, chances of securing reelection, and perhaps
financial considerations.”10 And if, as Jacobson suggested, challengers and
other elites make their strategic decisions based in part on partisan and
national factors, then it is likely that congressional incumbents take these
into account when assessing their reelection prospects, and consequently
when making their decisions whether or not to retire.
The strategic decisions of candidates to emerge or wait, to seek reelec-
tion or retire, establish parameters for party success in elections. And—
given candidate registration deadlines, the need to raise campaign funds,
and the dictates of campaigning more generally—those strategic decisions
are necessarily made generally about a year before the election. Hence, the
national and partisan conditions that affect politicians’ strategic decisions
in 1995 are likely to affect electoral outcomes in 1996. And in contempo-
rary politics, what a difference a year can make.
The 1996 election cycle saw more incumbent senators retire than any
post-war election. Eight incumbent Democratic senators and five incum-
bent Republican senators announced that they would not seek reelection in
1996. The previous post-war high was the 1978 figure of ten retirements.11
Coupled with the summer 1995 resignation of Senator Bob Packwood
(R-OR) and the midterm retirement of Republican presidential nominee
Bob Dole, the 1996 elections would necessarily produce a large freshman
class in the Senate. Again, many of the decisions that would subsequently
affect the outcomes of the 1996 elections were made in 1995. In fact, all
of the thirteen Senate retirements were announced by January 1996.
Moreover, Senate retirements in the 1995–1996 election cycle seem to sup-
port the hypothesis that retirements from Congress are influenced by a can-
didate’s assessment of his or her party’s prospects in the ensuing election.
In fact, in the first six months after the Republican victories in the 1994
elections, five Democratic senators announced their retirement from the
chamber; and the remaining three retiring Democratic senators announced
between August and October 1995. But as political conditions changed and
public sentiment toward the Republican revolution became more negative
in late 1995, partially as a result of the government shutdowns,
Republican—not Democratic—senators began announcing that they would
not seek reelection. Although Senator Hank Brown (R-CO) had announced
his retirement in December of 1994, the remaining four Republican sena-
tors who retired in 1996 all announced their plans between November
1995 and January 1996.
The l3 retirements, along with an additional open seat in Kansas caused
by an incumbent primary loss, contributed significantly to the overall com-
petitiveness of the 1996 Senate elections. According to reporter David E.
Rosenbaum, the fourteen open seats in the November elections were “by
far the highest since the Constitution was changed early in this century and
states began electing senators by popular vote rather than by votes of the
state legislature. The previous record was 10 open seats in 1978. As a con-
sequence, the Senate contests included some of the most expensive and bit-
ter ones ever.”12
ELECTORAL OPPORTUNITIES: MANY FOR REPUBLICANS, FEW FOR DEMOCRATS

Thus, although Republicans had more seats to defend in the 1996 elections
than did Democrats, securing the return of most of their incumbents sig-
nificantly mitigated Republican exposure. Republicans had to protect two
open seats in states that Clinton won in 1992 and would win in 1996 as
well (Maine and Oregon). Democrats also had to defend two open seats in
states that voted Republican in the presidential elections of both 1992 and
1996 (Alabama and Nebraska). Republicans were successful in protecting
both, while Democrats lost both that they had to protect. And of the four
seats Republicans had to defend in states won by Clinton in both elections
(Kentucky, New Hampshire, New Mexico, and Tennessee), all four were
won by popular Republican Senate incumbents. Except for the total num-
ber of seats to defend, Republicans enjoyed advantages in almost all of the
critical institutional factors that would influence the national verdict for
Senate seats in 1996. Thus, even if Republicans had a difficult time in the
presidential race, they were confident that the opportunities open to them
would compensate for any unexpected losses. National Republican
Senatorial Committee (NRSC) Chairman Senator Alfonse D’Amato said,

“Notwithstanding the top of the ticket, we are going to hold the majority.
Even if we lose one or two Republican seats . . . On balance, we have more
Democratic seats in play than Republican seats.”13
CAPITALIZING ON OPPORTUNITIES: PARTIES, CANDIDATES,

AND CAMPAIGNS
With all of their opportunities, early in the 1995–1996 election cycle
Republicans had been discussing the potential of the 1996 Senate elections
to produce a filibuster-proof majority in the Senate. The sixty seats needed
for such a majority seemed plausible in November 1995. With the 1994
election still in recent memory and with the political effects of the federal
government shutdowns still in doubt, Republicans were confident in their
prospects in the 1996 Senate elections. This feeling coupled with the
not–coincidental record number of Democratic senators retiring made it
seem likely that Republicans would take full advantage of their consider-
able opportunities. Republican pollster Ed Goeas did not question whether
Republicans would maintain their Senate majority but, rather, he pondered
by how much they might extend that majority: “Are we going to come in
at 58, 60, 61?”14
EARLY RETURNS: OREGON REPLACES PACKWOOD—WITH A DEMOCRAT

The first test of Republican strength was a special election in Oregon to
replace Republican Senator Robert Packwood, who had resigned in the
wake of scandal. Political observers and politicians often look to special
elections as predictors of success in the November elections to follow. In
January 1996, voters in Oregon participated in both an electoral experi-
ment and the selection of their senator. In both the December primary elec-
tions for the Democratic and Republican nominations for Senate as well as
the subsequent general election, Oregon voters made their selections by
mail.
On December 4, Gordon Smith, the president of the Oregon Senate,
won the Republican nomination while Representative Ron Wyden beat
U.S. House colleague Peter DeFazio for the Democratic nomination. On
the night that each won his party’s nomination, Smith and Wyden invoked
both national and statewide issues in setting the tone for the general elec-
tion campaign. Smith painted Wyden as a tax-and-spend liberal, saying,
“It’s hard to think of a tax Ron hasn’t voted for.”15 In addition, Smith
poked fun at a misstep Wyden made in his primary campaign. When asked
in a televised interview if he could quote the prices of milk and bread in
Oregon, Wyden failed. On primary election night—when it was clear that
he would face off against Wyden—Smith said, “Real Oregonians have real
problems and buy real products. And they know what they cost.”16
For his part, Wyden hoped to tie Smith to an increasingly unpopular

national Republican party. Wyden linked Smith to the “Newt Gingrich
Congressional majority,” saying that “the choice in this election . . . is
between mainstream values and extreme values.”17 Wyden, like many
Democrats in both House and Senate campaigns in 1996, persisted in this
throughout the race. In one televised debate Smith bristled at what was
likely an effective campaign strategy; Smith told Wyden, “If you want to
run against Newt Gingrich, you’d better move to Georgia.”18 Wyden’s
strategy of nationalizing the Oregon election was successful, as his cam-
paign was likely the beneficiary of good timing and strong voter sentiment
against national Republicans as a result of the federal budget standoff and
the government shutdowns.19 A poll of Oregon voters conducted in late
January revealed that Oregon voters clearly reacted more positively to
national Democrats than Republicans in the wake of the government shut-
downs. And, in the end, Wyden won by less than 20,000 votes.
Parties successful in a special election are likely to suggest that that par-
ticular election is an indicator of how the party will do in the following
November elections. By contrast, officials and activists of the losing party
will likely emphasize the unique aspects of the special election and discount
it as a bellwether for November. Despite the fact that Wyden won by only
one percent, probably aided by the high levels of turnout fostered by the
mail-in balloting system, Democrats were quick to point to the possibility
of a national trend. One Oregon Democratic pollster said, “The Oregon
election shows the bloom is off the rose [for Republicans]. . . . It does not
mean the Democrats are going to sweep the races, but it does mean
Republicans will have to make more of a statement about the environment
than (just saying) we’re overregulated.”20 And House minority leader
Richard Gephardt (D-MO) said of Wyden’s victory: “To say that this elec-
tion doesn’t have national implications is like saying that an elephant does-
n’t have ears.”21
A signal of a national Democratic trend or not, the Oregon race—by
virtue of its mail-in character—did provide some hope for those who
believe that more voter participation and more candidate and campaign
contact with voters, by means other than television, are good for represen-
tative democracy. Record high turnout and some return to traditional cam-
paigning marked this new voting method.22
The election of Ron Wyden to the Senate in the January 1996 Oregon
special election made the Republican’s hope for a filibuster-proof Senate
majority more distant by one critical seat. The significance of Wyden’s cap-
turing the seat previously held by Republican Packwood was not lost on
Republican officials and campaign operatives. Republican pollster Ed
Goeas said, “I think the wind went out of that sail [the filibuster-proof
majority] when the special election in Oregon went by the wayside.”23
PARTY ACTIVITIES AND PARTY STRATEGIES
Candidate Recruitment and Quality

Although political parties have little control over the selection of candi-
dates who will run under the party’s label in contemporary candidate-cen-
tered elections, they can engage in some activities that can encourage can-
didates to run. Party recruitment efforts were likely of unusual importance
in the 1996 Senate elections given the record number of open seats. Because
Democrats had considerably less money than Republicans and they had to
field so many non-incumbent candidates, the DSCC hoped to recruit many
candidates who would be able to get along without much financial assis-
tance from the party. To meet this objective, DSCC Chairman Senator Bob
Kerrey (D-NE) recruited many Democratic candidates with business back-
grounds and considerable personal wealth. Dubbed “Kerrey’s Merry
Millionaires,” these candidates included several entrepreneurs and business
executives (Mark Warner of Virginia, Tom Bruggere of Oregon, Walt
Minnick of Idaho, and Elliot Close of South Carolina) and a millionaire
physician and pharmaceutical executive (Charlie Sanders of North
Carolina).24 Note, however, that none of these candidates was victorious in
his Senate bid; Sanders of North Carolina did not even make it to the gen-
eral election. Democratic recruitment efforts were more successful with
professional politicians as candidates for the Senate. In Illinois, DSCC
Chairman Kerrey unsuccessfully tried to entice William Daley, lawyer and
brother of Chicago’s Mayor Daley, to succeed Senator Paul Simon as the
Democratic candidate for Senate in Illinois. Instead, the Democrats slated
Representative Richard Durbin for the Senate and he won in November.
On the Republican side, the NRSC had considerably more resources
than the DSCC and thus did not have to rely so heavily on wealthy politi-
cal novices. In fact, when it came to those crucial and highly competitive
open seats, neither party had much trouble recruiting professional politi-
cians to take a shot at the Senate race. In the fourteen open seats,
Republicans ran twelve candidates with prior political experience, includ-
ing five House members or former House members, four state legislators,
an attorney general, a state treasurer, and a state cabinet official. In those
same fourteen open seat races, Democrats ran eleven candidates with prior
political experience, including four House members or former House mem-
bers (including one who had also been governor), one (additional) gover-
nor, one state legislator, an attorney general, two state secretaries of state,
and two state treasurers.
Party Money
Contemporary parties are also active in fundraising and contributions to
candidates. The DSCC and NRSC spent nearly $100 million in the
1995–1996 election cycle with the NRSC’s $66.1 million more than dou-
bling DSCC disbursements of $30.8 million.25 Direct contributions to can-

didates remained only a small portion of party spending while more and
more dollars were allocated to party development and overhead expenses.
Although Democrats languished behind Republicans in fundraising, in the
last weeks of the 1996 election cycle, some corporate and trade group
donors increased giving to Democrats as it became increasingly possible
that Democrats could regain control of the House and possibly the Senate.
Steve Jarding, the DSCC communications director, said, “They’re coming
back because it looks like we can win some of these seats. They’re hedging
their bets.”26 Despite the Democrats’ fundraising disadvantage, both
national parties remained active and strong in the 1996 Senate elections.
Moreover, a Supreme Court ruling in the summer of 1996 had an impact
on party spending in the 1996 election cycle as well. On June 26, 1996, the
U.S. Supreme Court, in Colorado Republican Federal Campaign
Committee v. Federal Election Commission, opened up a new area of polit-
ical party campaign spending, allowing parties to now make “independent
expenditures” in elections for national office.27 In 1996, all independent
expenditures (IEs) made by the national parties were made by the two sen-
atorial campaign committees; neither national committee nor neither
House campaign committee made any. Moreover, the Republicans’ consid-
erable fundraising advantage made this new avenue of party spending par-
ticularly helpful to the GOP effort to extend their majority in the Senate.
In 1996, the Democratic senatorial campaign committee had independent
expenditures of $1,452,507 and the Republican committee’s expenditures
were $9,438,331.28
When the Court handed down the Colorado Republican decision, there
was considerable confusion among both politicians and commentators
regarding the impact the decision would have on the 1996 elections. Not
surprisingly, the parties’ respective responses to the decision were made in
light of their perceptions of the impact the decision would have on their
competitiveness. The day after the Court announced its decision, the NRSC
began planning for a new subdivision in order to both coordinate party
independent expenditures as well as to demonstrate the “independence” of
the new IE unit from ongoing NRSC activities that had already been “coor-
dinated” with candidates. The IE subdivision “was located in a building
separate from the NRSC headquarters, and was staffed by employees rela-
tively far removed from prior NRSC coordinated activity with particular
Senate candidates. While funding the independent expenditures unit, the
NRSC steadfastly denied any control or advance knowledge of the unit’s
spending decisions.”29 Democrats—who had supported IEs in principle by
submitting an amicus brief in the Colorado Republican case—tried to
diminish Republicans’ considerable advantage in 1996 IEs by asking the
Federal Election Commission (FEC) to issue an advisory opinion on
whether parties could, as a practical matter, engage in independent expen-
ditures given the fact that prior to the Court’s decision parties and candi-
dates had been coordinating their activities. This was less of an attempt by
Democrats to clarify the practical application of the new decision than it
was an attempt to have the FEC rule on Republican IEs.30 The FEC General
Counsel proposed an advisory opinion that would have halted Republican
and Democratic IE activity in 1996, but the FEC failed to pass it.
Left unregulated, both parties engaged in IE spending in Senate races. As
expected and noted, Republicans outspent Democrats in IEs by over a six
to one margin. In fact, the over $1.4 million in IEs Republicans made in the
Louisiana Senate race nearly equaled all of the Democratic IEs combined.
Over 55 percent of Republican IEs, over $5.2 million, were made in the six
open seats they had to defend and the three open seats picked up from
Democrats and “81.1% of all Republican IEs were spent in the 14 open
Senate races” of 1996.31
NRSC independent expenditures were an important component in
Republican success in the 1996 Senate elections. If the goal of the NRSC
and Republicans generally was to win the battle of the open seats and to
protect incumbents in trouble, IEs were an important resource on both
fronts. “Republican IEs in open seats and in. . .three close races where
Republicans sought to protect their incumbents [Smith in New Hampshire;
Helms in North Carolina; and Pressler in South Dakota] amounted to
95.0% of all Republican independent expenditures.”32 Both parties, though
Republicans far more than Democrats, exploited this new avenue of cam-
paign spending in strategic ways, as IEs were added to the increasingly sig-
nificant arsenal of national party spending opportunities.
CANDIDATE-CENTERED SENATE ELECTIONS

Despite these national party activities, congressional races, particularly
Senate races, are elections that have very individualistic characteristics.
Though parties can help themselves through candidate recruitment and
financial assistance, most Senate races have distinctive aspects. In 1996, for
example:
*In Massachusetts, popular Republican Governor William
Weld challenged incumbent Democratic Senator John Kerry in
a race that was characterized by voluntary spending limits and
primarily (if not exclusively) genteel, intellectual debates on
issues.
*In Louisiana, State Treasurer Mary Landrieu had to overcome
a significant rift with several African-American Louisiana
Democratic leaders in her successful quest to defeat Republican
state legislator Woody Jenkins.
*In Minnesota, former Republican Senator Rudy Boschwitz

unsuccessfully attempted to avenge his 1990 defeat at the hands
of Paul Wellstone.
*Nebraska’s popular Democratic Governor Ben Nelson was
upset by a late media blitz by the campaign of his Republican
opponent Chuck Hagel. Hagel spent over $210,000 in personal
money as well as almost $250,000 in campaign money to fund
the media blitz. The NRSC also chipped in over $325,000.33
*And in New Jersey, Republican Representative Dick Zimmer
lost to Democratic Representative Robert Torricelli in what was
described by many as one of the least civil campaigns in memo-
ry.
Altogether, major party candidates for the Senate raised about $226.7
million in the 1995–1996 election cycle. In both parties, incumbent candi-
dates raised the most money. On average, Democratic incumbents raised
$5,015,685 and Republican incumbents had receipts of $3,536,845. In
open seats, Democratic candidates averaged $2,839,688 and Republican
candidates took in $3,442,813. In both parties, challengers had the lowest
average receipts, $2,515,975 for the Democrats and $2,992,036 for the
Republicans.34 Winning candidates had considerable financial advantages
over their opponents. Only five losers raised more money than their win-
ning opponents in the 1995–1996 election cycle.35 Losers, on average, spent
79 cents for every dollar spent by their victorious opponents. And when the
five losers who spent more than their opposition are removed from the
analysis, the average loser spent 58 cents for every dollar spent by the win-
ner in that race.
INCUMBENT-CENTEREDNESS AND PARTISAN TRENDS IN OPEN SEATS

Despite the individual nature of many of these elections, the national out-
comes of Senate elections, if not always the individual races themselves, are
best understood from a perspective that places primary emphasis on party
activities and partisanship. This is mainly due to the influence of national
parties and political conditions on state candidates and campaigns.
National Themes
Candidates from both parties invoked national themes in their individual
races. Democrats did their best to tie their Republican opponents, incum-
bents and non-incumbents alike, to some of the perceived ideological
excesses of the Republican congressional majority. Democrats made
charges of “extremism” on Medicare, the government shutdowns, and
other issues. And they attempted to tie even Senate candidates, some of
whom were not even in Washington for the 104th Congress, to House
Speaker Newt Gingrich. This tactic was used even in Massachusetts where
Senator John Kerry highlighted a long-standing association between his

opponent, the very moderate Republican Governor Weld, and Gingrich.
Kerry called Weld “one of Newt’s lieutenants,” and recalled for voters
Weld’s 1995 reference to Gingrich as an “ideological soulmate.” Only days
before the election, Weld conceded that Kerry’s strategy could be effective;
he said, “If it’s me and [Kerry], I win. If it’s a nationalized race, if it’s Kerry
versus Gingrich in people’s minds, then I’m a goner.”36
Republicans, on the other hand, made frequent use of the word “liber-
al” to describe their Democratic opponents. The relatively consistent use of
this strategy was not an accident. A chief architect of this strategy, consul-
tant Arthur J. Finkelstein, worked not only as a strategist for the NRSC,
but Republican Senate candidates were encouraged by NRSC Chairman
D’Amato to hire Finkelstein personally as well; six Republican Senate can-
didates did hire Finkelstein (Rudy Boschwitz, MN; Nancy Mayer, RI;
Raymond Clatworthy, DE; Dick Zimmer, NJ; Robert Smith, NH; and
Larry Pressler, SD). Finkelstein’s strategy was to call all Democrats “liber-
als.” According to Democratic strategist Mandy Grunwald: “Essentially he
has dictated the message strategy for the Republican party. I don’t know a
Senate race in the country where the Republican message isn’t charging lib-
eral, liberal, liberal.”37 In some races, “liberal” became part of the
Democratic candidate’s name:
*Rhode Island: “Liberal Jack Reed opposes the balanced bud-
get amendment.”
*Minnesota: “Ultraliberal Paul Wellstone voted for the largest
tax increase in American history.”
*New Jersey: “Tell liberal Bob Torricelli to stop raising your
taxes.”38
In the races where that was not the case, “liberal” was used nevertheless.
In Ray Clatworthy’s attempt to unseat Delaware Senator Joseph Biden, the
NRSC issued a press release which claimed that “Ray Clatworthy is clos-
ing the gap on Joe Biden because he is . . . focusing on the differences
between his commonsense conservatism and Joe Biden’s extremely liberal
voting record.”39 And in New Hampshire, Senator Robert Smith ran ads
which called former Representative Dick Swett “too liberal for New
Hampshire.”
The day after the election, the DSCC quickly pointed out that
Finkelstein’s strategy had been unsuccessful, as the only one of Finkelstein’s
six candidates to win was New Hampshire’s Smith who only narrowly
avoided defeat in his reelection bid. And Senator Larry Pressler,
Finkelstein’s other incumbent client, was not as lucky as Smith. Pressler’s
campaign had called Representative Tim Johnson “surprisingly liberal.”
For his part, Johnson claimed that South Dakotans, who had elected him
to the House five times, knew that he was “Just solid, like South Dakota.”
Presidential Vote and Senate Vote

The large number of open seats in 1996 likely enhanced the importance of
partisanship in the national outcomes of the Senate elections. Studies of
House elections are useful in explaining the importance of national and
state-specific factors in the 1996 Senate elections. Abramowitz and Segal’s
suggestion (that the more Senate seats a president’s party must defend in an
election, the more seats the president’s party will lose) has its correlate in
the study of House elections in the “exposure model” which predicts the
net partisan change in outcomes of House races on the basis of how many
seats a party already has in the House. If a party has many seats above an
equilibrium point, it is likely to lose seats. If it currently has fewer seats
than its equilibrium, it will likely gain seats.40
Although the exposure thesis predicts the broad contours of outcomes
in House elections, it is improved considerably when one factors in the
decisions of incumbents to run for reelection or retire.41 Not only does the
exposure model fare better, but by implication it also seems that national
factors better explain seat swings when there are many open seats. In fact,
a study of presidential coattails in House elections finds that “presidential
coattails are, on average, nearly two times stronger in open than in defend-
ed seats.”42
Additionally, it should be noted that scholars James Campbell and Joe
Sumners have found that “given the prominence of Senate elections, . . .
presidential coattails for Senate candidates are about half as long as those
provided to House candidates.”43 Nevertheless, Campbell and Sumners
found that “even in the recent period of weakened mass partisanship and
this rise of more highly financed Senate contests, presidential candidates
have coattails that affect the Senate vote.”44 Moreover, it is likely that other
(non-presidential) national and partisan factors affect the Senate vote as
well. This is probably even more the case when the intervening variable of
an incumbent running for reelection is removed.
Given these scholarly findings, we would expect national factors—coat-
tails, national partisanship, and national issues—to be more pronounced in
open seats than in seats with an entered incumbent. This proposition is
supported by simple correlations of a state’s vote for Clinton with the
state’s vote for the Democratic Senate candidate. There are significant dif-
ferences in the relationship between the presidential vote and Senate vote
in open seats and in seats with incumbents running. The correlation of the
presidential vote to the Senate vote was much stronger in open seat races
(r=.84, n=14), than in elections where an incumbent was running (r=.59,
n=20). When no incumbent was running, the Democratic vote for President
and Senate were strongly correlated, but when an incumbent was running,
the strength of that relationship diminished.
Thus, a party can mitigate the potential losses in a difficult election year
by persuading its incumbents to run for reelection, thus blunting the effects
of national tides on the outcome in that individual race. This likely bene-
fited Republicans in 1996. In many of the states where Bill Clinton was
popular and Republicans were vulnerable, Republicans had popular
incumbents in place seeking reelection. Democrats, on the other hand, had
a difficult time retaining incumbents in Alabama and Nebraska, for exam-
ple. Republican presidential candidates carried both of those states in 1992
and 1996, but incumbents Howell Heflin and Jim Exon did not seek reelec-
tion and therefore were not poised to blunt the Republican trends in their
states.
CONCLUSIONS
Scholarly attention to Senate elections—like attention to the Senate more
generally—is scant compared to studies of House elections. Comprehensive
evaluations of the national outcomes of Senate elections are rare indeed.
But Senate elections have important national and partisan consequences,
both on election night and for the subsequent six years of governance. In
the 1996 Senate elections, the advantages Democrats enjoyed in November
1996 were mitigated and the Republicans’ disadvantages alleviated by the
strategic decisions candidates of both parties made as early as 1995.
Republicans’ advantages in the seats to be protected, incumbents returning,
fundraising, as well as strategic disbursements, allowed Republicans to
blunt the effects of an albeit small Democratic tide in 1996.
Democratic failure to gain Senate seats in an otherwise Democratic year
continues a pattern whereby Republicans have made significant gains in the
Senate since the 1960s. From the 1958 to the 1968 elections, Democrats
held at least 60 Senate seats (they held as many as 68 after the 1964 elec-
tions). Throughout the 1970s, Democrats won smaller Senate majorities.
But in l980 Republicans gained twelve Senate seats and won a majority for
the first time since the election of 1952. In 1986, however, Democrats once
again gained control of the Senate. Republicans gained a Senate majority
once again, though, in 1994.
Since 1960, Democratic strength has eroded in the South and the West.45
Immediately after the election of 1960, Democrats held all twenty-two
southern Senate seats and twenty of the West’s twenty-six Senate seats.
Twelve years later, by contrast, in the aftermath of the 1972 election, the
Democrats held only fifteen of those southern seats and fifteen of those
western seats. And in 1980, when Democrats lost control of the Senate, it
was in large part due to Republican gains in the South and West.
Democrats only had twelve of the South’s twenty-two seats in 1980; and
they held onto only nine of the West’s twenty-six seats. While Democratic
losses in the West seem to have leveled, the 1996 election showed further
erosion of Democratic strength in the South: Democrats held only seven of
the South’s twenty-two Senate seats in 1997–1998. The two seats
Republicans picked up in 1996 equal the two they picked up in Alabama
and Arkansas. From 1980 to 1996, there was stability in the partisan com-
petition for Senate seats in the Northeast, the Border states, and the West,
and Democrats gained in the Central states. However, these trends do not
offset Republican gains in the South. There are forty-eight Senate seats in
the twenty-four states of the South and West. Democrats controlled forty-
two of those seats in 1961 (and that represented nearly two-thirds of the
Senate Democratic party at that time). But after the 1996 election,
Democrats held only seventeen of those seats (a little over a third of the
Senate Democratic party).
Democrats could conceivably regain a Senate majority without recoup-
ing southern and western losses, but that would require winning more seats
in one or all of the three other regions than they have in more than thirty
years. Overall trends currently favor Republicans; they are now the party
of the South and the West by nearly two-to-one. And nearly half the Senate
seats are in the South and West. In 1998, though, the Democrats gained an
eighth southern seat in North Carolina, as Democrat John Edwards defeat-
ed Republican incumbent Lauch Faircloth. However, there was no net par-
tisan change in 1998, as the GOP held 55 seats and the Democrats 45 seats
during 1999–2000.
NOTES
1. Election returns throughout this chapter are from Richard M. Scammon,
Alice V. McGillivray, and Rhodes Cook, America Votes 22 (Washington, DC:
Congressional Quarterly Inc., 1998).
2. Quoted in “Elections ’96; Washington Insight,” Los Angeles Times,
November 7, 1996, A16.
3. Alan Greenblatt and Robert Marshall Wells, “Parties Aim to Dominate
Senate, But Big Gains Look Unlikely,” CQWR, September 21, 1996, 2682–92,
2683.
4. Alan I. Abramowitz and Jeffrey A. Segal, Senate Elections (Ann Arbor:
University of Michigan Press, 1992), 94.
5. This figure drops to 90.5 percent if one counts Senator Sheila Frahm’s
(R–KS) unsuccessful attempt to win the Republican nomination. Frahm had been
appointed to fill the vacancy left by Senator Robert Dole’s resignation from the
Senate; she was defeated by Representative Sam Brownback in the Republican pri-
mary.
6. For 1946–1994 data, see Harold W. Stanley and Richard G. Niemi, Vital
Statistics on American Politics, 5th ed. (Washington, DC: CQ Press, 1995),
190–191.
7. William M. Welch, “Two Incumbents Vie in South Dakota,” USA,
November 1, 1996, A11.
8. Quoted in Chris Black, “Kerrey Speaks Mind on Republican Rivals,”
Boston Globe, March 9, 1996, 3.
9. Gary C. Jacobson, The Politics of Congressional Elections, 3rd ed. (New

York: HarperCollins, 1992), 164; also see Jacobson and Samuel Kernell, Strategy
and Choice in Congressional Elections, 2nd ed. (New Haven: Yale University Press,
1983).
10. Michael K. Moore and John R. Hibbing, “Is Serving in Congress Fun
Again? Voluntary Retirements from the House Since the 1970s,” AJPS 36 (August
1992): 824–8, 827. Emphasis added.
11. Stanley and Niemi, Vital Statistics, 190–191.
12. David E. Rosenbaum, “Two Parties Battling for Control of Senate in Most
Competitive Lineup of Races in Years,” NYT, November 6, 1996, B5.
13. Quoted in Elaine S. Povich, “On the Edge of their Seats: In Senate, Power
Hangs in Balance,” Newsday, September 23, 1996, A4.
14. Quoted in Juliana Gruenwald, “Sizing Up the Senate; Parties Aim at
Magic Numbers in Senate Elections,” Minneapolis Star Tribune, December 26,
1995, 7A.
15. Quoted in “Two Advance in Race to Succeed Packwood,” NYT,
December 7, 1995, B16.
16. “Two Advance in Race to Succeed Packwood,” B16.
17. Quoted in “Two Advance in Race to Succeed Packwood,” B16.
18. William Claiborne, “Democrats Win Oregon Senate Race; Observers Say
Result Could Be Bellwether,” WP, January 21, 1996.
19. “Oregon Voters in Poll Blame GOP for Budget Standoff,” USA, January
31, 1996.
20. Quoted in Lynne K. Varner, “Wyden Win in Oregon Heartens
Democrats—Is Victory Start of a Trend or a One-Time Thing?” The Seattle Times,
February 5, 1996.
21. Quoted in Timothy Egan, “Oregon’s Mail-In Election Brings Cheer to
Democrats,” NYT, February 1, 1996, A1, A19.
22. Robert Marshall Wells, “Wyden, Smith in Tight Race for Packwood
Seat,” CQWR, January 20, 1996, 155–56.
23. Quoted in Greenblatt and Wells, “Parties Aim to Dominate Senate,”,
2682.
24. See “Kerrey Senate Campaign Strategy Didn’t Pan Out on Election Day,”
Omaha World Herald, November 15, 1996, 24. See also Robert Marshall Wells
and Jonathan D. Salant, “Wealthy Democrats Are Tapped to Challenge GOP
Senators,” CQWR, February 24, 1996, 443–7.
25. Federal Election Commission.
26. Quoted in Ruth Marcus, “GOP Keeps Fund-Raising Lead Despite Trade
Groups’ Shift in Giving,” WP, November 3, 1996, A31.
27. See John P. Forren and Douglas B. Harris, “Misconceptions of Party in
Colorado Republican Federal Campaign Committee v. FEC,” paper prepared for
delivery at the 1997 Annual Meeting of the Midwest Political Science Association,
Chicago, Ill. Much of the discussion below is taken from this paper.
28. Federal Election Commission.
29. Forren and Harris, “Misconceptions of Party,” 23. See also Ruth Marcus,
“Reinterpreting the Rules,” WP, October 26, 1996.
30. “Democrats Seek to Foil Republican Spending Plans with FEC Advisory
Opinion,” Political Finance and Lobby Reporter, July 24, 1996.
31. Forren and Harris, “Misconceptions of Party,” 27–28.
32. Forren and Harris, “Misconceptions of Party,” 28.
33. Jonathan D. Salant, “Late Spending Won Senate Seat for Hagel,” CQWR,
December 21, 1996, 3451.
34. Federal Election Commission. The FEC document had both Republican
Sam Brownback and Democrat Jill Docking listed as “challengers” in their Kansas
senate race. As no incumbent was running in the general election, their receipts
have been subtracted from the “challenger” pools and added to the open seat cat-
egories.
35. They include Republicans Millner (GA) and Pressler (SD), and Democrats
Strickland (CO) and Mark Warner (VA). The fifth case is somewhat misleading. In
North Carolina, Democrat Harvey Gantt raised $300,000 more than Senator Jesse
Helms in the 1995–96 election cycle, but Helms had raised and spent over $6.7 mil-
lion in the previous two election cycles although he was not up for reelection in
either.
36. See R. W. Apple Jr., “Kerry vs. Weld: An ‘Elegant Hammering’ of a Race
Remains a Tossup,” NYT, November 3, 1996, Section I, 35; and Frank Phillips,
“Weld Ranks 2d Only to Gingrich in Help from GOPAC,” Boston Globe, August
23, 1996.
37. Quoted in Howard Kurtz, “GOP Consultant’s Strategy: Label Opponents
Liberally,” WP, October 22, 1996, A1.
38. Kurtz, “GOP Consultant’s Strategy.”
39. NRSC press release, October 8, 1996.
40. See Richard W. Waterman, Bruce I. Oppenheimer, and James A. Stimson,
“Sequence and Equilibrium in Congressional Elections: An Integrated Approach,”
JOP 53 (May 1991): 372–93.
41. Ronald Keith Gaddie, “Congressional Seat Swings: Revisiting Exposure in
House Elections,” Political Research Quarterly 50 (September 1997): 699–710.
42. Jeffrey J. Mondak, “Presidential Coattails and Open Seats: The District-
Level Impact of Heuristic Processing,” APQ 21 (July 1993): 307–19, 314.
43. James E. Campbell and Joe A. Sumners, “Presidential Coattails in Senate
Elections,” APSR 84 (June 1990): 513–24, 519.
44. Campbell and Sumners, “Presidential Coattails,” 520.
45. For our purposes, the South is defined as Alabama, Arkansas, Florida,
Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas,
and Virginia. For later discussion, the West is defined as Alaska, Arizona,
California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon,
Utah, Washington, and Wyoming.
CHAPTER 5
Sideshows and Strategic Separations

The Impact of Presidential Year Politics on
Congressional Elections
GARRISON NELSON
INTRODUCTION: BENEATH THE STATUS QUO

So identical were the electoral college maps of the 1992 and 1996
presidential elections that analysts were quick to decree that 1996 was a
“status quo election.”1After all, only five states had shifted their party alle-
giance between the two elections. President Bill Clinton had picked up two
states—Florida and Arizona—from his previous showing and lost three
others. Both Florida and Arizona were heavily populated with retirees. It
was widely asserted in press accounts that Clinton had benefited from the
Democratic party’s efforts to raise anxiety among senior citizens about the
Medicare and social security cuts contemplated by the Republican Con-
gress.2 Republicans called it a tactic of “Mediscare” and it seemed to have
worked in these two states. Three other states that had cast electoral votes
for Clinton in 1992—Colorado, Montana, and Georgia—returned to the
Republican fold in 1996. But forty-five states and the District of Columbia
retained their electoral affiliations from the previous election.
In the elections for Congress as well, continuity seemed to triumph over
change. In the Senate, there was a net shift of only two seats as the Repub-
lican majority grew from 53 to 55, and only one of twenty Senate incum-
bents lost, Larry Pressler (Rep-S.D.). In the House, the shift was only nine
seats, and the incumbent success rate was 94.4 percent. There is thus little
surprise that the 1996 election was dubbed “the status quo election.”
That continuity can triumph in a time of national peace and economic
prosperity is no surprise. That the election continued divided government
in place is perhaps the real story of 1996. Why for the first time in Ameri-
can history was a Democratic president elected simultaneously with a
Republican Congress?
It is the major contention of this chapter that the historic linkage
between presidential and congressional elections was broken in the 1970s
and 1980s by the conscious actions of two key figures—President Richard
105
106 Sideshows and Strategic Separations
M. Nixon and Speaker of the House Thomas P. “Tip” O’Neill Jr. These
two men sought to protect their party’s control over their respective insti-
tutions and believed that the continuing linkage between presidential and
congressional elections would jeopardize that control.
THE CONTEXT OF THE 1996 HOUSE ELECTIONS

In the House elections of 1994, history was made. For forty years
(1955–1994), the Democrats had controlled the House, the longest single-
party control of an elective national institution in American history. The
Republicans had become a “permanent minority.”3In spite of the Republi-
can presidential victories of 1956, 1968, 1972, 1980, 1984, and 1988,
House Democrats had held on. In 1972 and 1984, House Democrats
retained control of the chamber in the face of 49–state landslide victories
by Presidents Nixon and Reagan. Scandals and mismanagement seemed to
have little or no impact upon Democratic control of the House. But Con-
gress began to slide precipitously in the estimation of the public. In Decem-
ber of 1965, at the close of the first session of the 89th Congress—the
Great Society Congress—62 percent of people polled by Louis Harris and
Associates said that Congress was doing a “good job.”4 During the Water-
gate crisis, the Gallup Poll began its annual “Confidence in Institutions”
ratings. Congress scored a 42 percent confidence rating in that initial poll
but it slipped steadily to an approval rating of 18 percent in the 1991
through 1994 ratings.5Even the long-noted tendency of citizens to differen-
tiate between the positive qualities of their own members of Congress and
the negative ones of the institution seemed to collapse. Incumbency that
had protected House Democrats fell before the 1994 tide. Thirty-four
incumbents lost in 1994 and all were Democrats.6
What had happened in 1994 was the nationalization of a congressional
election. For years, the prevailing wisdom of House elections was con-
tained in former Speaker Thomas P. “Tip” O’Neill’s (Dem-Mass.) pithy
aphorism, “All politics is local.”7 Protect your constituents and bring home
the bacon and your reelection was guaranteed.
Newt Gingrich (Rep-Ga.) changed that perception. Gingrich, the osten-
sible leader of the House Republicans in the summer of 1994, worked with
a savvy conservative pollster, Frank Luntz, to identify positions that were
both popular with the public and which could create a philosophical agen-
da uniting most House Republicans. Ten proposals were identified and
labeled with public-pleasing phrases and named the “Contract with Amer-
ica.”8 Democrats scoffed, calling it the “Contract on America.”
Gingrich was triumphant. By nationalizing the election, Gingrich had
crafted a strategy to overturn the Democrats’ forty-year dynasty. When the
results rolled in that evening, Gingrich was decreed to be a genius. Demo-
cratic incumbents fell throughout the nation, including scandal-weakened
Ways and Means Chair Dan Rostenkowski (Dem-Ill.), and Speaker of the
Garrison Nelson 107
House Thomas Foley (Dem-Wash.), the first Speaker to lose an election

since 1862.
Meanwhile, not a single incumbent Republican governor or member of
Congress lost that evening. The pundits were stunned and one of the
descriptive terms gaining currency in the weeks following the 1994 election
was “tsunami”—a giant tidal wave.9So dramatic was the perception of the
impact of the 1994 election and new Speaker Gingrich’s agenda on Amer-
ican politics that Newt Gingrich was named “Man of the Year” by Time
magazine for 1995.10
Table 5.1: Fifty-Plus House Seat Election Gains and House Party Control
in the Twentieth Century
Election Prior Election Net Seat Impact on House
Year Majority Minority Change Party Control
1910 R 219 D 172 D +56 Dems gained control
1912 D 228 R 161 D +63 Dems expanded control
1914 D 291 R 127 R +59 Dems retained control
1920 R 240 D 190 R +61 Reps expanded control
1922 R 301 D 131 D +74 Reps retained control
1938 D 331 R 89 R +75 Dems retained control
1946 D 218 R 208 R +55 Reps gained control
1994 D 258 R 176 R +54 Reps gained control
Source: Calculated by author from U.S. Bureau of the Census, Historical Abstracts of the
United States, Colonial Times to 1970, Bicentennial Edition, Part I (Washington, DC: U.S.
Government Printing Office, 1975), 1083-1084; and U.S. Bureau of the Census, Statistical
Abstract of the United States: 1998 (118th ed.) (Washington, DC: U.S. Government Printing
Office, 1998), 287, Table no. 466.
In historic terms, the change was less dramatic than it first appeared.
This was the thirteenth time in the twentieth century that a fifty-seat or
more House gain had taken place (table 5.1). On eight occasions, Democ-
rats had gained fifty-plus seats, while in five years, Republicans benefited.
Only five times had party control of the House been affected by a fifty-plus
seat shift: three for the Democrats, 1910, 1930, and 1948; and two for the
Republicans, 1946 and 1994. On three occasions, the party in power lost
fifty-plus seats but retained control of the House: 1914 and 1938 for the
Democrats and 1922 for the Republicans. And in five cases, the House
majority party expanded its numbers by more than fifty seats: 1912, 1932,
1958, and 1974 for the Democrats and 1920 for the Republicans. In three
cases, the magnitude of these expansions presented the gaining party with
90-plus House seat turnarounds within a two-election cycle. The Democ-
rats gained 119 seats in the 1910–12 cycle; the Republicans gained 91 in
the 1918–20 cycle; and the Democrats gained 146 in the 1930–32 cycle.
These gains are of the magnitude that creates new House majority parties.
This was the goal of newly elected Speaker of the House Newt Gingrich: to
follow up 1994’s gain with a massive seat expansion in 1996 that would
consolidate Republican power in the House through the end of the twenti-
eth century and well into the next one.
But a presidential election awaited. Money and talent that might have
made the House seat expansion possible would likely be siphoned off for
the presidential race. The 1996 presidential contest could be potentially
disruptive for the Republican congressional majority.
THE SIDESHOW PROBLEM

American politics are defined by presidential elections. For the most part,
midterm congressional elections are treated as referenda on presidential
performance and as harbingers of trends for the upcoming presidential
elections.11 Unless midterm elections result in a massive philosophical shift
such as the Republican takeover of 1994 or the Democratic Watergate
babies of 1974, they tend not to be very memorable in their own right.
Congressional elections in presidential years are even more relegated to
the back pages of history, and are often viewed primarily as reflections of
presidential politics. For many years, political pundits calculated the size of
the coattail effect of successful presidential candidates by measuring the
percentage of the president’s vote and the percentage of House seats won
by his party.12It was generally inferred that longer coattails would help the
president’s legislative agenda and that shorter coattails would limit his leg-
islative effectiveness.
In an important 1966 study, Milton Cummings compared the district-
by-district results to assess the separation of the presidential and congres-
sional electoral systems.13 In the 1970s, Walter Dean Burnham amplified the
theme of party decomposition.14Presidential and congressional election sys-
tems are now somewhat independent entities, and this trend is revealed by
examining midterm elections as predictors of presidential elections, split
district results for the U.S. House and president, and the incidence of split
party control in Washington.
MIDTERM ELECTIONS AS PRESIDENTIAL ELECTION

PREDICTORS
For most of a century, the party controlling the House would win the
upcoming presidential election. From 1856 through 1944, there were twen-
ty-two presidential elections. The party controlling the House gained the
presidential popular vote 21 times (95.5 percent) and the electoral vote 19
times (86.4 percent). In 21 of 22 elections, the popular vote leader in the
Garrison Nelson 109
presidential balloting had been foreshadowed by his party’s victory in the

midterm House election.
The only genuine exception to this predictive link between midterm
House elections and presidential contests in the 1856–1944 era occurred in
the 1880 election of Republican James A. Garfield. Apart from Garfield,
the other two exceptions to the linkage between the midterm and presi-
dential elections—the Hayes–Tilden contest of 1876 and the first Cleve-
land–Harrison battle of 1888—proved the rule. The House Democrats had
won both the midterm elections of 1874 and 1886 preceding those presi-
dential contests. It was the machinations of the fifteen-member Electoral
Commission and its 8–7 Republican split that deprived Governor Samuel
Tilden (Dem-N.Y.) of his victory. By counting only the Republican electors
from three southern states, the Commission gave former Ohio Governor
Rutherford B. Hayes a one-vote win in the electoral college.15 In the 1888
case, it was the switch of fifty-one electoral votes in the states of New York
and Indiana from the Democratic column in 1884 to the Republican one
four years later that prevented President Grover Cleveland (Dem-N.Y.)
from having his popular vote plurality reflected in an electoral college
majority.16 For almost ninety years, the midterm elections were the best pre-
dictor of presidential politics. And then this relationship ended.
THE BEGINNING OF THE DISCONNECTION

In 1946, a war-weary nation disappointed by unfulfilled promises in the
aftermath of World War II turned on the Democratic party and ended their
control of Congress. For the first time since the election of 1928, Republi-
cans won both the House and the Senate. At last, the Republicans would
end the Democrats’ sixteen-year grip on the White House. After all, the
party that controlled the House following the midterm had gone on to win
almost all of the subsequent presidential contests.
Yet when the votes were tallied, President Harry Truman retained the
White House and the 1946 House midterm election did not predict the
1948 presidential election. Nor did the 1950 midterm election predict the
1952 Republican victory. The failure of midterm House elections to predict
subsequent presidential contests recurred in 1956, 1968, 1972, 1980,
1984, 1988, and 1996. In the thirteen elections from 1948 through 1996,
the party that won control of the House in the midterm election has won
the subsequent presidential contest only four times—1960, 1964, 1976,
and 1992—for a success rate of 30.8 percent.
A THREE-VARIABLE PRESIDENTIAL ELECTION PREDICTIVE MODEL

One simple dimension is obviously insufficient to explain American poli-
tics. By adding more variables, preferably those that the key political actors
know, predictability may be increased. In table 5.2, a simple three-variable
predictive model is used to forecast the presidential elections from 1892

through 1996.17 Using three readily available cues from the political envi-
ronment in place at the time of the election, presidential contests were once
quite predictable. The three cues were:
1) partisan control of the House following the midterm election;
2) the relative size of the House seat loss for the president’s party in
the midterm election; and
3) the proportion of governors held by the president’s party at the
time of the election.
Table 5.2: House and Gubernatorial Pre-Election Presidential Cues,

1892–1996
Midterm Number of In-Party Governors
House Cues Over 55.0 percent 45.0 to 54.9 percent Under 45.0 percent
In-Party T.ROOSEVELT, 1904 McKINLEY, 1900
Holds House TAFT, 1908 Stevenson, 1952
HOOVER, 1928
Seat-swing F.ROOSEVELT, 1936
Under 10 % JOHNSON, 1964
Carter, 1980
________________________________________________________________________________
In-Party WILSON, 1916 F.ROOSEVELT, 1944 COOLIDGE, 1924
Holds House F.ROOSEVELT, 1940 Humphrey, 1968
Seat-swing
Over 10 %
Out-Party BUSH, 1988 Cox, 1920
Holds House EISENHOWER,
1956, NIXON, 1972
Seat-swing Bush, 1992
Under 10 %
Out-Party TRUMAN, 1948 Harrison, 1892
Holds House Bryan, 1896
Taft, 1912
Seat-swing Hoover, 1932
Over 10 % Nixon, 1960
Ford, 1976
REAGAN, 1984
CLINTON, 1996a
Note: Presidential electoral vote winners are CAPITALIZED. Exceptions are in italics.
a
1996 Prospects: Out-Party Holds House (-)
Seat-swing of 21.0 % Over 10 % (-)
In-Party Governors of 38% Under 45% (-)
In-Party Nominee an Incumbent (+)
Source: Adapted and updated from W. Ross Brewer and Garrison Nelson, “Election Expec-
tations and Outcomes: A Theory of Nominating Convention Conflict, 1896-1976,” in Public
Policy and Public Choice, ed. Douglas W. Rae and Theodore J. Eismeier, (Beverly Hills, Cal.:
Sage Publications, 1979), 151–207.
Garrison Nelson 111
The expectation was simple. If all three cues were positive (the presi-
dent’s party held the House at the midterm; the seat loss was under 10 per-
cent; and 55 percent or more of the governors at election time belonged to
the president’s party), then the party holding the White House would retain
it, regardless of the status of the nominee, be it elected incumbent, a Vice
President-successor, or a new nominee. Conversely, if all three cues were
negative (the president’s party lost the House in the midterm; the House
seat loss exceeded 10 percent; and less than 45 percent of the governors
belonged to the president’s party), then the party holding the White House
would be voted out of office, regardless of the status of the nominee, be it
elected incumbent, a Vice President-successor, or a new nominee. And if the
cues were mixed, then incumbent presidents would continue in office and
non-incumbent nominees of the in-party would be defeated.
As table 5.2 indicates, this three-variable model worked perfectly in
twenty-two elections from 1892 to 1976. When all three cues were posi-
tive, the five pre-1980 in-party nominees were elected: one elected incum-
bent (Franklin D. Roosevelt, 1936); two successor-Vice Presidents
Theodore Roosevelt, 1904, and Lyndon B. Johnson, 1964); and two first-
time nominees (William Howard Taft, 1908, and Herbert Hoover, 1928).
When all three cues were negative, the six pre-1980 in-party nominees were
defeated: three elected incumbents (Benjamin Harrison, 1892; William
Howard Taft, 1912; and Herbert Hoover, 1932); one successor-Vice Presi-
dent (Gerald Ford, 1976) and two first-time nominees (William Jennings
Bryan, 1896, and Richard M. Nixon, 1960). When the cues were mixed,
only the eight White House incumbent candidates were successful: William
McKinley, 1900; Woodrow Wilson, 1916; Calvin Coolidge, 1924; Franklin
D. Roosevelt, 1940 and 1944; Harry Truman, 1948; Dwight Eisenhower,
1956; and Richard Nixon, 1972. And all three non-incumbent nominees of
the president’s party met defeat in mixed circumstances: James Cox, 1920;
Adlai Stevenson, 1952; and Hubert Humphrey, 1968.
Here was a simple presidential predictive model based only upon three
political cues that worked in twenty-two consecutive elections. Since 1980,
however, this model has failed to predict the White House victors in five
consecutive elections. President Jimmy Carter’s 1980 reelection bid had all
three cues in his favor yet he was defeated by 9.7 percent of the vote. The
size of Carter’s defeat enabled him to pass Martin Van Buren and gain the
dubious distinction of the nation’s worst defeated Democratic incumbent.
In 1984, all three cues were negative, yet President Ronald Reagan swept
to a 49-state landslide in spite of them. Four years later, Vice President
George H. W. Bush was elected President in spite of the fact that he was a
non-incumbent confronting mixed cues. To further confound the model,
President Bush lost as an incumbent in 1992 while also facing mixed pre-
dictive cues. And in 1996, President Bill Clinton faced a bleak political
landscape. Clinton’s party did not hold the House; the seat loss for his
party in the midterm election was 21 percent; and Democrats only had 19
governors. Yet he won reelection. A predictive model that fails to predict
successfully five consecutive times is a model that should be discarded.
SPLIT PARTISAN RESULTS IN CONGRESSIONAL DISTRICTS

In the twelve presidential elections from 1900 through 1944, the mean pro-
portion of congressional districts splitting their partisan results between
their own House candidate and presidential nominees was only 11.3 per-
cent. But in the presidential elections from 1948 through 1996, the district
differential jumped to 30.5 percent.18
The 44.1 percent presidential–House seat differential in 1972 set a high
standard for ticket-splitting that was toppled in 1984 by President Reagan’s
49-state landslide over former Vice President Walter Mondale. In that year,
the presidential–House seat differential rose to 45.0 percent.
Only the nomination of southern presidential candidates by the Democ-
rats—Lyndon Johnson of Texas in 1964; Jimmy Carter of Georgia in 1976
and 1980; and Bill Clinton of Arkansas in 1992 and 1996—has kept the
differential relatively low. Four of the five greatest presidential–House seat
party differentials were recorded in those years with non-southern Democ-
ratic nominees: Hubert Humphrey of Minnesota in 1968 (32.0 percent);
George McGovern of South Dakota in 1972 (44.1 percent); Walter Mon-
dale of Minnesota in 1984 (45.0 percent); and Michael Dukakis of Mass-
achusetts in 1988 (34.0 percent). The three southern–born Democratic
nominees averaged a 28.9 percent district party differential in their five
campaigns, while the four northern–born Democratic nominees averaged a
38.8 percent differential. Not coincidentally, the five southern nominations
resulted in four presidential victories for the Democrats while the four
northern nominations resulted in none. But the Democrats won control of
the House eight times and the Senate six times in those contests regardless
of the regional origins of their presidential nominee.
THE GROWTH OF DIVIDED GOVERNMENT

Divided government is neither new nor uncommon: it has occurred in 37
of the 106 Congresses convened in the United States—34.9 percent (table
5.3). Defined simply as one of the three elective national institutions—the
presidency, the Senate, and the House of Representatives—organized by a
party majority dissimilar from at least one of the other institutions, divid-
ed government has been a regular feature of American government. How-
ever, the incidence of divided government has varied between political eras.
Before 1947, divided government only occurred about one-fourth of the
time. However, in the last fifty years, divided government occurred more
than 60 percent of the time.
Garrison Nelson 113
Table 5.3: Congresses With Divided Government by Era, 1789–2001

Eras
1789–1857 1857–1947 1947–2001 Totals
N percent N percent N percent N percent
United government 26 76.5% 33 73.3% 10 37.0% 69 65.1%
President isolated 3 8.8 3 6.7 14 51.9 20 18.9

House isolated 4 11.8 7 15.6 3 11.1 14 13.2
Senate isolated 1 2.9 2 4.4 0 0.0 3 2.8
Totals 34 100.0% 45 100.0% 27 100.0% 106 100.0%
United Government, 1789–2001 69/106 65.1 %

Divided Government, 1789–2001 37/106 34.9 %
________________________________________________________________________________
Source: See Table 5.1.
It has been within this most recent half-century of American politics that
divided government has become the rule and not the exception. However,
analysts were slow to discern this change in political life. With the Demo-
cratic capture of Congress in 1948, analysts concluded that the two-year
period of divided government, was a post-war anomaly. The next time of
divided government, under Republican President Dwight Eisenhower, last-
ed for six years (1955–61) and was explained away with the concept of the
deviating election.19 With Eisenhower retired from public life, the Democ-
ratic party with its party-identifier advantage resumed control of American
political life. Thus, the election of Senator John F. Kennedy (Dem-Mass.)
was seen to be a “reinstating” one.20
The four instances of divided outcomes in the eleven elections between
1946 and 1966 (36.4 percent) were now safely explained. But analysts
were not prepared for the extraordinary eruption of divided government
that appeared in 13 of the 16 Congresses (81.2 percent) elected between
1968 and 1998.
Once again, events have overtaken explanations. The 1968 circumstance
was explained away with the George Wallace vote that enabled Nixon to
gain an electoral vote majority while white southerners could vote for Wal-
lace’s presidential candidacy yet retain Democrats in Congress. Nixon’s
1972 victory was contemporaneously attributed to his Southern Strategy
and his own non-partisan Committee to Re-elect the President, which de-
emphasized his Republican affiliation, thus permitting Democrats to vote
for the reelection of Nixon and their own Democratic members of Con-
gress. But in 1988, George Bush won 40 states and the Democrats still
retained both houses of Congress. Bush’s victory was the fifth for Republi-
cans in the previous six presidential contests, but in none of those elections
were they able to end Democratic hegemony over the House. The new
explanation was that Americans trust Republican presidents in foreign pol-
icy, but wish Democrats to control domestic policy in the House of Repre-
sentatives.21
Once again, anomalous events have taken command. In 1996, President
Bill Clinton became the first Democratic presidential nominee ever to win
election and have his party gain control of neither chamber of Congress.
Back to the drawing board.
SEPARATION AND SURVIVAL I: THE PRESIDENTIAL

DISENGAGEMENT OF RICHARD NIXON
The midterm election of 1970 marked an important turning point in Amer-
ican history. Eager to remove liberal Democrats from the U.S. Senate, Pres-
ident Nixon and Vice President Spiro Agnew campaigned from coast to
coast.22 Republicans gained two senate seats, but they also suffered a net
loss of eleven governors, including seven incumbents.23 This was the largest
loss of governors for any party since 1938. Nixon and his campaign oper-
atives were fearful that Nixon’s fragile electoral coalition would not hold
in 1972.
Entering the 1972 election year, Nixon’s Republicans confronted a bleak
political landscape. The Vietnam War continued to claim American lives.
Democrats held a ten-seat margin in the Senate and a 74-seat margin in the
House. Nixon’s Gallup Poll approval rating was a very modest 49 percent
and polls indicated that Democratic party identifiers outnumbered Repub-
lican party identifiers 41 percent to 23 percent.24 To hold the presidency, the
Nixon White House developed a three-part strategy.25 First, run Nixon as
the incumbent president, avoid mentioning the Republican party, and do
little to help other Republican candidates. Second, appeal to the support-
ers of Alabama’s Governor George Wallace by further soft-pedaling the
Republican party’s traditional commitment to civil rights and by finding
prominent southern Democrats such as Texas Governor John Connally to
endorse President Nixon, the so-called Southern Strategy.
Third, use the tactic of divide and conquer among the supporters of the
Democrats’ three leading moderate contenders: Senators Edmund Muskie
of Maine, Hubert Humphrey of Minnesota, and Henry M. “Scoop” Jack-
son of Washington.26 This tactic would hurt all three and pave the way for
the nomination of a Democratic candidate who was likely to have a polit-
ical base strong enough to obtain the nomination but too narrow to win
election. The likeliest of these candidates was Senator George S. McGovern
(Dem-S.D.), an anti-war liberal who had rewritten the party’s delegate
selection rules. This part of the 1972 strategy eventually led to the Water-
gate crisis.
In the Congress, Nixon’s 1972 strategy of electoral separation had won
him few friends. Not many Republican members felt obligated to Nixon
for their victories, and to some, he had squandered a great opportunity to
gain party control of Congress in pursuit of a personal victory. This lack of
Garrison Nelson 115
Republican obligation would manifest itself in the House committee votes

on impeachment in July 1974.
Nixon’s 1972 experiment in consciously separating presidential elec-
tions from congressional ones seemed to be bad politics and even worse
government. However, it would be tried again. The next time, it would be
Democratic congressional leaders who would separate the elections to pro-
tect themselves from an unpopular incumbent of their own party.
SEPARATION AND SURVIVAL II: THE CONGRESSIONAL

DISENGAGEMENT OF TIP O’NEILL
Congressional Democratic leaders welcomed the 1976 election of Jimmy
Carter. Following the election, Democrats held sixty-one seats and a 23-
seat margin in the Senate, and 292 seats and a 149-seat margin in the
House. Disillusionment with President Carter was slow to develop, but
once it did, it was difficult to repair. It became clear that Carter’s executive
style was to confront the legislature when it failed to deliver on his pro-
gram. His style had some limited success in one-party Georgia, but it failed
miserably in two-party Washington.27
President Carter’s approval ratings dropped from a high of 75 percent in
1977 to 28 percent in 1979. In July 1980, popular approval of the Carter
presidency dropped to 21 percent, the lowest approval rating ever scored
by a president in the years since the Gallup Poll had first posed the ques-
tion in 1938. No previous president had fallen so low, so fast, and so com-
pletely.28
In an extraordinarily candid poll conducted among 243 members of
Congress by U.S. News and World Report in 1979, only 10.2 percent of
Democrats said that they would be helped by President Carter’s “presence
at the head of the Democratic national ticket” in 1980. Almost two-fifths
of the Democrats (36.7 percent) said that Carter’s presence would hurt
their campaigns.29
Congressional Democrats launched two hardball strategies to get Carter
off the 1980 ticket. In the spring of 1979, a six-month investigation was
launched to force Carter to answer questions concerning the financing of
the 1976 presidential campaign and a number of congressional Democrats
urged Massachusetts Senator Ted Kennedy to challenge Carter for the 1980
nomination.30 But the Kennedy challenge weakened as waves of patriotism
induced by the Iran hostage crisis pushed Carter’s popularity from 31 per-
cent in October 1979 to 61 percent in December.31
Carter’s conflicts with Congress worsened as he permitted the Justice
Department and the FBI to collaborate on an elaborate sting operation
involving agents posing as Arab businesspersons who tried to bribe mem-
bers of Congress. This operation, known as “Abscam,” netted one U.S.
Senator and six U.S. Representatives; all but one were Democrats.32 One
month before the convention, the Democratic Senate launched an investi-
gation of Carter’s brother, Billy, and his lobbying contract with the Libyan
government of Muammar Qaddafi. Despite this investigation, Carter was
able to gain renomination with only 64 percent of the delegates, the lowest
delegate percentage ever recorded for a renominated Democratic incum-
bent.33
The fears of congressional Democrats that Carter’s 1980 candidacy
would be costly were partially confirmed as the Senate fell to the Republi-
cans for the first time since the election of 1952. Republicans picked up
twelve seats in the Senate and 35 in the House, reducing the Democrats’
margin from 119 seats to 51.34 But the Democrats held onto the House
in 1980 because during the 1970s, the House Democrats developed
incumbent protection strategies by funding the expansion of constituen-
cy operations with more district offices and larger staffs to accommo-
date the growing casework demands of the voters.35 The “all politics is
local” mindset had sunk in.
The Democrats controlled the House in 1981, but their smaller
majority permitted a group of conservative southern Democrats to
become the key swing vote in the House. These conservative Democrats
provided President Reagan with a number of legislative successes. In
fact, Ronald Reagan’s success rate with the Congress reached 82 percent
for 1981 according to Congressional Quarterly.36 It was the highest suc-
cess rating for a president with Congress since 1965, the first session of
Lyndon Johnson’s Great Society 89th Congress.
Reagan’s success with the Congress heightened the importance of the
1982 congressional elections. In California, which gained two seats in the
1982 reapportionment, U.S. Representative Phil Burton was instrumental
in a successful Democratic gerrymander. As a result, the party split among
the California House delegation shifted from 22 Democrats–21 Republi-
cans to 28 Democrats–17 Republicans—a net gain of ten seats.37
Tony Coelho was a second California Democrat whose assistance
allowed House Democrats to elect majorities in the Reagan years. In 1981,
Coelho became chair of the Democratic Congressional Campaign Com-
mittee (DCCC), the House party’s key fund-raising operation. In this posi-
tion, Coelho was able to move vast sums of PAC (Political Action Com-
mittee) money into the campaign coffers of incumbent House Democrats.38
To journalist Brooks Jackson, Coelho’s policy was “honest graft.”39 But it
was successful and the House Democrats survived both the Reagan 49-
state landslide of 1984 and the Bush 40-state near-landslide of 1988.
Incumbent–focused PAC contributions not only contributed to Democrat-
ic control of the House, but it led to overall incumbent reelection rates that
averaged 94.8 percent in the six congressional elections between 1980 and
1990.40
A third person who helped the House Democrats separate from and sur-
vive presidential politics was Christopher Matthews, a speechwriter in the
Garrison Nelson 117
Carter White House who joined the staff of Tony Coelho in 1981 as a
“media consultant.”41 Within months, Matthews was working for Speaker
O’Neill. O’Neill was a master of “retail politics” back home in his Massa-
chusetts congressional district and on the floor of the House in dealing with
individual members. But in 1981, Speaker O’Neill faced President Ronald
Reagan, a master of wholesale politics.
With Matthews’s assistance, O’Neill was made over from an out-of-
touch Boston Irish pol to the one person standing between Reagan’s insen-
sitive Republicans and those Americans who were poor, elderly, unem-
ployed, and of color. House Speaker Thomas P. O’Neill Jr. became simply
“Tip” to most Americans and was able to thwart successfully the Reagan
agenda. And before O’Neill and Matthews were through, it was Reagan
who was seen as out of touch.42The Democrats were now able to withstand
most of the legislative thrusts of the Reagan White House, and according
to Congressional Quarterly, the president’s success rate with the Congress
dropped steadily throughout his term in office.43
The presidential–House electoral disconnect of the 1980s was unprece-
dented. The three Democratic presidential defeats of 1980, 1984, and 1988
had yielded a total of only 20 of 153, or 13.1 percent, of possible state vic-
tories (including the District of Columbia) and only 10.8 percent of the
total electoral votes. Not even the horrendous consecutive defeat strings
suffered by presidential Democrats in the 1896–1908 era and the
1920–1928 era were as one-sided as the 1980–1988 sequence. The Democ-
rats failed to capture a single Congress in those two previous eras. Yet in
the 1980–1988 era, the House Democrats won every congressional election
and captured 57.9 percent of the House seats in those three presidential
debacles. Their continued success was the greatest president–House elec-
toral disconnection in American history. The strategy of separation had
worked.
SEPARATION AND SURVIVAL III:

THE DISENGAGEMENT OF PRESIDENT CLINTON, SPEAKER
GINGRICH, AND THE REPUBLICAN FRESHMEN
President Clinton faced an uncertain future late in the summer of 1995.
The Cook Political Report for August 1995 counted only twelve states that
Clinton could hope for in 1996: five were listed as “Solid Democratic,” five
as “Likely Democratic,” and two in the “Lean Democratic” column.44
Taken together, these twelve states provided 105 electoral votes, or 165
fewer than necessary for President Clinton’s reelection. Senator Dole’s
totals from this same report were as follows: “Solid Republican,” 12 states
with 79 electoral votes; “Likely Republican,” 5 states with 72 electoral
votes; and “Lean Republican,” 10 states with 98 electoral votes. With 14
months to go, Dole had 27 states with 249 electoral votes, only 21 shy of
victory. There were 12 “Toss-up” states with 184 electoral votes. President
Clinton had to carry 90 percent of the electoral votes in the toss-up states
to win, while Senator Dole needed only 11 percent.
Clinton also confronted a Congress controlled by a powerful and ide-
ologically united Republican majority. Speaker Gingrich and his conser-
vative House loyalists attempted a coup de grace on the weakened Clin-
ton presidency by partially shutting down the federal government twice
between November 1995 and January 1996. The closings were intended
to make the president accept their budget proposals, which would have
altered the federal role substantially. Coming as it did when Gingrich’s
presidential trial balloons were aloft, this gambit may have been the
Speaker’s attempt to demonstrate to Washington that he was the more
powerful leader of the two. However, at the urging of longtime Gingrich
nemesis ex-Representative Leon Panetta (Dem-Cal.), the new White
House Chief of Staff, Clinton did not budge. The shutdown tactic back-
fired, and it was the Gingrich Republicans who suffered the brunt of the
criticism.45 Knowing that they were outmaneuvered, House Republicans
backed down and Speaker Gingrich put his presidential aspirations aside.
The Speaker threw his support to Dole.
But maintaining divided government for another presidential term
may not have seemed an unpleasant situation to the Speaker. Should Dole
lose and the Republicans retain control of the House, Gingrich would
guarantee his continued preeminence within the Republican party.
Should Bob Dole win the presidential election, Speaker Gingrich would
have tumbled from 1995’s “Man of the Year” to occupying a supporting
role. With the national government continuing to be divided between the
executive and legislative branches, Gingrich would not have to relinquish
that preeminence.
Rationality overcame President Clinton, as well. With his return to
office on January 20, 1997, the clock began to tick on his lame duck sta-
tus. Bill Clinton will be fifty-four years old at the end of his second term.
The youngest former president, Teddy Roosevelt, left the White House in
1909, at the age of fifty. In 1912, though, Teddy Roosevelt vigorously
sought to regain the presidency.
President Bill Clinton has sought to protect his legacy and to advance
his standing among American presidents by vesting himself heavily in the
career of Vice President Al Gore. Every major event of the second term
has occurred with Vice President Gore at the president’s side. To protect
the Gore candidacy in 2000 and his own legacy, it was in the interest of
President Clinton that the House Democrats not regain power. The ideo-
logical aftermath of the 1994 election debacle was that the House’s mod-
erate Democrats had suffered disproportionate losses, and in 1995, five
conservative southern Democrats crossed the aisle to become Republicans.
In the 104th Congress, the Democratic party was much more liberal than
Garrison Nelson 119
the president.46 And the House Democratic leader Dick Gephardt had
sought the presidency in 1988.
A House Democratic majority would have elected Dick Gephardt
Speaker and jeopardized both the Gore nomination and the Clinton lega-
cy. To limit the House Democrats and to counter the Gingrich Republicans
in his bid for reelection, President Clinton adopted the triangulation strat-
egy outlined for him by Dick Morris. This strategy enabled Clinton to “cre-
ate a new position, not just in between the old positions of the two parties
but above them as well.”47 It made wonderful short-term sense. Polariza-
tion had taken hold. The Gingrich–led House Republicans pushed their
conservative agenda to the point of closing down the federal government
twice in 1995–96 and the Gephardt–led House Democrats seemed locked
in the age–old liberal Democratic mindset of raising taxes and increasing
welfare entitlements.
Neither position seemed to address the public’s needs for positive change
while protecting the most vulnerable of the nation’s citizens. Triangulation
allowed President Clinton to call for an income–limited tax cut and mean-
ingful welfare reform. He would be above the fray and be seen as presi-
dential. Triangulation worked and Bill Clinton became the first Democrat
reelected to the presidency since Franklin D. Roosevelt’s four consecutive
successes from 1932 to 1944.
Within the Congress, much of the partisan rancor subsided in 1996 as
many of the House Republican freshmen shifted away from the Contract
with America and supported President Clinton’s shift to the legislative mid-
dle.48 It was they who had come to understand the virtues of separation and
survival. Newt Gingrich may have separated himself and his speakership
from the collapse of the Dole candidacy, but many fast-learning Republi-
can freshmen separated themselves from Speaker Gingrich to ensure their
own reelection.49
For the congressional Democrats, triangulation was an electoral disaster.
Throughout much of 1996, columnists contended that the Democrats had
an outside chance to recapture the House. But President Clinton kept his
distance from them for most of the year. Not until August in a speech
before the steelworkers did he make his first call for a Democratic Con-
gress.50 In spite of the urging of many key supporters that a Democratic
Congress would bring an end to the legislative investigations that were
plaguing him, the president’s White House operatives provided minimal
financial assistance from the campaign coffers.51
A new round of divided government had been purchased. But what
would it mean?
THE CONSEQUENCES OF DIVIDED GOVERNMENT FOR

PRESIDENTIAL POPULARITY AND SUCCESS WITH CONGRESS
In his classic book Presidential Power, Richard Neustadt argued that in
order to gain the power to persuade, presidents must successfully combine
public popularity with a solid professional reputation established inside the
Beltway of Washington, D.C.52 Legislative success with the Congress is one
way of measuring a president’s professional reputation. For without it, how
can a president convince others within Washington that he has the power
to accomplish anything? Presidential popularity and success with Congress
may easily be compared over time. Widely reported sources of these assess-
ments include the Gallup Poll, which has been asking Americans about the
job performance of presidents since the second administration of Franklin
Roosevelt, and the “presidential support scores” generated by Congres-
sional Quarterly to assess the relative legislative success of presidential
administrations since 1953.
The annual averages for the public opinion surveys represent twenty
(or so) separate estimates of public opinion from samples of randomly
selected citizens. The congressional success score is an annual measure of
the won–lost record of a president on floor votes in both the House and
Senate on which he took a position. The comparison of these data from
1953 through 1998 shows that Congress tends to be more supportive of
presidential agenda items than the public is of the president.53 For every
president from Dwight Eisenhower to Bill Clinton (1993–1998), with the
exception of George Bush, the mean annual presidential support score
exceeded the mean annual approval rating. The largest differential was
during the four–year Carter administration from 1977–1980, when the
president enjoyed 76.4 percent support in Congress but only 46.3 percent
support among the public. During the first six years of the Clinton admin-
istration, through the end of 1998, the president had an average CQ sup-
port score of 61.4 and a mean Gallup approval rating of 53.2.
These data are most clear when comparing the presidential popularity
ratings to a president’s legislative success rate in the U.S. House. Histori-
cally, most divided government circumstances pit the House against the
president and the Reagan–O’Neill and Clinton–Gingrich encounters bear
out the contention that these are the federal institutions which are each
other’s historical enemy. When a president’s year in office is added to the
equation, the uniqueness of the conflict becomes clearer. Presidential suc-
cess in the House of Representatives is relatively constant for the first five
years of a united government’s presidential administration. The first year’s
average success rate is 85.0 percent in the House, while the fifth year’s suc-
cess rate is 83.5 percent, only 1.5 percent lower. This is not true of divided
government presidencies, where success rates start lower at 63.5 percent
for the first year and decline to 47.5 percent by the fifth year.
Garrison Nelson 121
The sixth year appears to be a last gasp of legislative-executive comity.

But by the seventh year in office, presidents are more successful with the
public than they are with their legislative program. Clearly, the members of
Congress are preparing themselves for the next presidential administration.
Furthermore, the slippage among the citizenry is far greater for presi-
dents operating in united government circumstances than in divided ones.
The range in popular support from high years (first years) to low
years(fourth years) is 24.5 percent (66.3–41.8) for presidents in united cir-
cumstances. The range from high years (first years) to low years (sixth
years) is less than a third of that—7.8 percent (59.2–51.4)—in divided cir-
cumstances. This would appear to be counter-intuitive. Presidents should
fare worse with the public when they are confronting a hostile Congress
and must defend themselves and their legislative agendas on a daily basis.
But such is not the case. Ironically, presidents do better with the public
during divided governments, for Congress is a more disliked institution
than the presidency.54
Examination of the Gallup approval ratings shows that public approval
of presidents in divided years exceeds that of presidents in united years
beginning with the third year in office. It is during the year immediately fol-
lowing the first midterm election that separation begins to have its politi-
cal benefits. All four presidents in this era who have been reelected—Eisen-
hower, Nixon, Reagan, and Clinton—have had their legislative parties
rejected at the first midterm election yet have gone on to win convincing
presidential victories. In each case, they exceeded their initial victory per-
centages. A president confronting a hostile congressional majority can
become heroic.
In times of united government, presidents are held accountable for con-
gressional action (and inaction). If the Congress acts carelessly and in
haste, the president is blamed for not providing a brake on its recklessness.
And if a united government Congress fails to act, it must be due to the
inability of the president to provide effective leadership. Such are not the
worries of presidents in divided circumstances. In these cases, the president
may portray himself as urging a recalcitrant and unresponsive Congress to
act, as Harry Truman successfully did with “the do–nothing 80th Con-
gress.” Or the divided government president may be portrayed, as was
Ronald Reagan, as the protector of the larger public’s interest against the
narrow and provincial claims of a greedy and rapacious Congress eager to
spend taxpayer money to insure their own reelections.
Presidents in late–term divided government conditions often have a pos-
itive standing in the eyes of the public, comparable to their early–term
“honeymoon” ratings and higher than those for presidents in united gov-
ernment circumstances. For those presidents whose egos need such emo-
tional support, divided government may be the remedy.55
DIVIDED GOVERNMENT AND PRESIDENTIAL LEGACIES

What of the long term issues? With reelection at hand, President Clinton
fell into the preoccupation of most two-term presidents; what will be the
judgment of history on my presidency?
In late 1996, a panel of historians and presidential analysts assessed all
of the presidents once again.56 The early “greatness” line on the Clinton
presidency was “low average.” Presumably, a successful second term would
elevate him another notch. But divided government has its own dynamic.
Presidents who face Congresses with hostile majorities may benefit in the
short–term public opinion polls but they suffer in scholarly polls.
A voting public that rejects a president’s legislative majority has already
decreed that this is a person whom they do not wish to see in total control
of the national agenda. In the 1996 Schlesinger poll, four presidents were
adjudged “below average” and seven were ranked as “failures.” These
presidents held office for little more than 45 years. Their terms in office
coincided with all or some of 24 Congresses, twenty-one of which were
elected while they were atop the ticket or held the White House. In only
eleven cases were their legislative majorities approved by the American
public—52.4 percent. In ten cases of the twenty-one, the public gave these
below average and failure presidents no congressional mandates—47.6
percent. The public had spoken long before the historians had.
In the Schlesinger poll, nine presidents were adjudged “great” and
“near-great.” These presidents held office for almost 69 years. Their terms
in office coincided with all or some of 36 Congresses, thirty-four of which
were elected while they were atop the ticket or held the White House. In
twenty-nine of these thirty-four cases, their legislative majorities were rati-
fied by the American public—85.3 percent. And among the three greatest
presidents—Lincoln, Washington, and Franklin Roosevelt—thirteen of
their fourteen Congresses were controlled in both chambers by their party’s
legislative majority—a success rate of 92.9 percent. United governments
produce great presidents.
For President Clinton, the verdict may already be in. He may have to
spend the remainder of his days in the presidential purgatory of “low aver-
age” for his 1996 willingness to encourage divided government’s tradeoff
of legislative achievement for the short-term benefits of heightened person-
al popularity. Certainly, had Clinton worked harder to elect a Democratic
Congress, he would have avoided the unfortunate set of impeachment votes
cast by the Republican House majority in 1998.
Speaker Gingrich made history, becoming the first Republican Speaker
in forty years and playing a large role in the electoral upheaval that led to
the House Republican majority. He has been decreed one of the nine
“Kings of the Hill” along with Henry Clay, Tom Reed, and Sam Rayburn
in a recent book about the history of House leadership.57
Garrison Nelson 123
Like Gingrich, Speakers Reed and Rayburn thrived in periods of divid-

ed government. Despite his party’s pummeling in the press for the 1995–96
government shutdowns, Gingrich’s Republicans were able to retain a
majority in the House. It was the first time since the election of 1928 that
the Republicans had won the House in successive elections. It was a
remarkable accomplishment. Gingrich’s return to the speakership in 1997,
in spite of ethical difficulties and stirrings of revolt within his own ranks,
was also unique.
There is an irony here. Both Democratic President Bill Clinton and
Republican Speaker Newt Gingrich seemed to benefit from the extension
of divided government in 1996. But for both the gain was only short-term.
Clinton was almost impeached by the Republican–controlled 105th–106th
Congresses. And Gingrich resigned from the House after Republicans suf-
fered losses in the 1998 elections. Speaker Gingrich, however, has earned
his historical stripes. Long after the contemporaneous public opinion polls
have faded, it will be Speaker Gingrich and not President Clinton who will
be seen as having had the greater impact upon the electoral arena and the
legislative agenda as Americans closed the last decade of the twentieth cen-
tury.
CONCLUSION
In addition to the strategic calculations of politicians, a number of struc-
tural changes in the American political system have instigated the separa-
tion between presidential and congressional politics. These factors include
the development of television as the major source of news and campaign
information; the increased incumbency advantage in congressional elec-
tions; and the political realignment of the South from the Democrats to the
Republicans, which began at the presidential level and only slowly moved
to congressional voting.
The uniqueness of the 1996 election, however, is not that the results
were divided once again, but that the deliberate separation of the presi-
dential and congressional election systems took place in both parties simul-
taneously and benefited both Speaker Gingrich and President Clinton.
Speaker Newt Gingrich, who had postponed his own presidential bid in
1996, seemed determined to insulate the House Republican majority from
the potentially negative consequences of ex-Senator Dole’s dispirited cam-
paign. On the other side, it was President Clinton who appeared to run
away from the congressional Democrats lest his personal popularity be
hurt by their low public esteem. This actor-based version of the 1996 elec-
tion casts President Clinton and Speaker Gingrich as co-stars in the latest
divided government drama.
NOTES
1. Examples of this assessment include Everett C. Ladd, “The Status-Quo
Election: An Introduction,” The Public Perspective 8 (December/January 1997),
4–5; and Larry J. Sabato, “The November Vote—A Status Quo Election,” in
Toward the Millennium: The Elections of 1996, ed. Sabato, (Boston: Allyn and
Bacon, 1997), 143–161.
2. Typical of these stories was Katharine Q. Seelye, “In Blistering Attack,
Dole Says Clinton Is Using Scare Tactics,” NYT, September 27, 1996, A22; and
Spencer S. Hsu and Ellen Nakashiuna, “GOP Says Foes Using ‘Mediscare,’” WP,
October 30, 1996, B6.
3. William F. Connelly Jr. and John J. Pitney Jr., Congress’ Permanent Minor-
ity? Republicans in the U.S. House (Lanham, Md.: Rowman and Littlefield, 1994).
4. Roger H. Davidson, David M. Kovenock, and Michael K. O’Leary, Con-
gress in Crisis: Politics and Congressional Reform (Belmont, Ca.: Wadsworth,
1966), 53, Table 2.5.
5. Gallup Poll, “Confidence in Institutions,” news release, August 15, 1997,
4–5.
6. Dave Kaplan and Julianna Gruenwald, “The House: Longtime ‘Second’
Party Scores a Long List of GOP Firsts,” CQWR, November 12, 1994, 3232–3239.
7. Speaker O’Neill’s phrase, “All Politics is Local” was the title of the open-
ing chapter of his 1987 autobiography, Man of the House: The Life and Political
Memoirs of Speaker Tip O’Neill (New York: Random House, 1987). The book was
written with William Novak. He also used the phrase again in the title of another
book, Thomas P. O’Neill Jr., with Gary Hymel, All Politics is Local and Other
Rules of the Game (New York: Times Books, 1994).
8. Among some of the titles were: “The American Dream Restoration Act,”
“The Taking Back Our Streets Act,” and “The Family Reinforcement Act,” in Con-
tract With America: The Bold Plan by Rep. Newt Gingrich, Rep. Dick Armey, and
the House Republicans to Change the Nation (New York: Times Books, 1994).
9. James A. Finefrock, “The Republican Tsunami,” San Francisco Examin-
er, December 9, 1994, A–11. Among the more evocative descriptions were “Stam-
pede!,” the cover story for Time, November 21, 1994, 46–49ff; J. Weisberg, “After
the Deluge,” New York, November 14, 1994, 28ff.; Meg Greenfield, “After the Big
One,” Newsweek, November 21, 1994, 108; and R. Lacayo, “After the Revolu-
tion,” Time, November 28, 1994, 28–33.
10. “Man of the Year: Newt Gingrich,” Time, December 25, 1995, 20ff.
11. Political scientists have tried to dispel this notion; see Lyn Ragsdale, “The
Fiction of Congressional Elections as Presidential Events,” APQ 8 (October 1980):
375–398. However, political writers continue to foster the belief in the linkage. A
smattering of midterm election analyses supports this point; see L. Walczak, “How
1986 Changes the Presidential Race,” Business Week, November 17, 1986, 8; Gus
Tyler, “Straws in the American Political Winds,” The New Leader 73 (November
12–26, 1990): 10–12; M. Kelly, “Why the President is in Trouble,” Reader’s Digest,
Garrison Nelson 125
November 1994, 85–90; and J. Weisberg, “Why It’s Even Worse for Clinton Than
You Think,” New York, November 21, 1994, 41.
12. The best-known of these efforts is Malcolm Moos, Politics, Presidents and
Coattails (Baltimore: Johns Hopkins Press, 1952). Not all of Moos’s contempo-
raries shared his belief in the coattail linkage; see Cortez A. M. Ewing, Congres-
sional Elections, 1896–1944: The Sectional Basis of Political Democracy in the
House of Representatives (Norman: University of Oklahoma Press, 1947).
13. Milton C. Cummings Jr., Congressmen and the Electorate: Elections for
the U.S. House and the President, 1920–1964 (New York: The Free Press, 1966),
esp. 39–47.
14. Walter Dean Burnham, “Insulation and Responsiveness in Congressional
Elections,” Political Science Quarterly 90 (Fall 1975): 411–435.
15. See Sidney I. Pomerantz, “Election of 1876,” in History of American
Presidential Elections, 1789–1968, Vol. II, ed. Arthur M. Schlesinger, Jr. and Fred
L. Israel, (New York: Chelsea House, 1971), 1379–1487, esp. 1413–1424.
16. See Robert F. Wesser, “Election of 1888,” in Schlesinger and Israel,
1615–1700, esp. 1645–1649.
17. This model was first presented in W. Ross Brewer and Garrison Nelson,
“Election Expectations and Outcomes: A Theory of Nominating Convention Con-
flict, 1896–1976,” in Public Policy and Public Choice, ed. Douglas W. Rae and
Theodore J. Eismeier (Beverly Hills, Ca.: Sage, 1979), 151–207.
18. These data are adapted from Harold W. Stanley and Richard G. Niemi,
eds., Vital Statistics on American Politics, 2nd ed.(Washington, DC: CQ Press,
1990), 133, Table 4–8. Data for the 1992 and 1996 elections came from Clark H.
Bensen of POLIDATA, Lake Ridge, VA.
19. Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald E.
Stokes, The American Voter (New York: John Wiley, 1960), 531–538.
20. Philip E. Converse, Angus Campbell, Warren E. Miller, and Donald E.
Stokes, “Stability and Change in 1960: A Reinstating Election,” APSR 55 (June
1961): 269–280.
21. The most persuasive advocate of this view is Gary C. Jacobson, The Elec-
toral Origins of Divided Government: Competition in U.S. House Elections,
1946–1988 (Boulder, Colo.: Westview Press, 1990), 112–120.
22. Rowland Evans Jr. and Robert D. Novak, Nixon in the White House: The
Frustration of Power (New York: Random House, 1971), 303–346.
23. “Democrats take 13 Governorships from Republicans,” CQWR, Novem-
ber 6, 1970, 2748–2749, 2770. Republicans picked up two state governorships, so
the net loss was eleven.
24. George C. Edwards III and Alec M. Gallup, Presidential Approval: A
Sourcebook (Baltimore: Johns Hopkins University Press, 1990), 61; American
National Election Studies, University of Michigan.
25. Theodore H. White, The Making of the President, 1972 (New York:
Atheneum, 1973), 48–69.
26. Bob Woodward and Carl Bernstein, All the President’s Men (New York:
Simon and Schuster, 1974), 112–130.
27. See O’Neill, with Novak, Man of the House, 297. See also the observa-
tions of Christopher Matthews, who worked for both Carter and O’Neill, Hard-
ball: How Politics is Played—Told by One Who Knows the Game (New York:
Summit, 1988).
28. “President Carter’s Popularity,” The Gallup Poll (August 1979), 223–227;
“Presidential Popularity: Carter Rating Lowest of Any President Since ’38,” Gallup
Opinion Index Report (August 1980), 24–25.
29. “Congress Tells Carter How He Rates,” U.S. News and World Report,
August 13, 1979, 21–23.
30. Warden Moxley, “Kennedy Draft: Movement With Precedents,” CQWR,
September 22, 1979, 2041–2048.
31. “The Carter Presidency: Carter Praised for Personal Qualities; Few Feel
History Will Regard Him Outstanding,” The Gallup Report (January, 1981), 3–4,
56–57.
32. “Abscam Scandal Clouded Congresses Image,” CQ Almanac 1980
(Washington, DC: Congressional Quarterly Inc., 1981), 513–521.
33. Data from Richard C. Bain and Judith H. Parris, Convention Decisions
and Voting Records, 2nd ed. (Washington, D.C.: The Brookings Institution, 1973),
Appendix C, 351–417, and updated with “Key Ballots at 1980 Democratic Con-
vention,” CQWR, August 16, 1980, 2437.
34. Charles E. Jacob, “The Congressional Elections,” in The Election of 1980:
Reports and Interpretations, ed. Gerald M. Pomper (Chatham, NJ: Chatham
House Press, 1981), 119–141. See also the recollections of ex-Speaker Jim Wright,
Balance of Power: Presidents and Congress from the Era of McCarthy to the Age
of Gingrich (Atlanta: Turner Publishing, 1996), 326–340.
35. David R. Mayhew, Congress: The Electoral Connection (New Haven:
Yale University Press, 1974); Morris Fiorina, Congress: Keystone of the Washing-
ton Establishment (New Haven: Yale University Press, 1977); and Richard F. Fenno
Jr., Home Style: House Members in Their Districts (Boston: Little, Brown, 1978).
36. Bill Keller, “Voting Record of ’81 Shows the Romance and Fidelity of Rea-
gan Honeymoon on Hill,” CQWR, January 2, 1982, 18.
37. Alan Ehrenhalt, “Reapportionment and Redistricting,” in The American
Elections of 1982, ed. Thomas E. Mann and Norman J. Ornstein (Washington, DC:
American Enterprise Institute, 1983), 48–49, 71. A solid depiction of Burton’s
efforts may be found in John Jacobs, A Rage for Justice: The Passion and Politics
of Phil Burton (Berkeley: University of California Press, 1995), 425–440.
38. Burdett A. Loomis, The Contemporary Congress, 2nd ed. (New York: St.
Martin’s Press, 1998), 68–73.
39. Brooks Jackson, Honest Graft: Big Money and the American Political
Process (New York: Knopf, 1988).
40. Roger H. Davidson and Walter J. Oleszek, Congress and Its Members, 6th
ed. (Washington, DC: CQ Press, 1998), 62–63, Table 3.1.
41. Christopher Matthews, Hardball, 42.
42. See O’Neill, with Novak, Man of the House, 331.
Garrison Nelson 127
43. “Presidential Victories on Votes in Congress, 1953–1988,” Vital Statistics

on American Politics, 248–249, Table 8–11.
44. The Cook Political Report, August 4, 1995, 1.
45. Gingrich became a lightning rod for popular discontent with Congress; see
Marjorie Randon Hershey, “The Congressional Elections,” in The Election of
1996: Reports and Interpretations, ed. Gerald M. Pomper (Chatham, NJ: Chatham
House, 1997), 212–217; and Gary Jacobson, “The 105th Congress: Unprecedent-
ed and Unsurprising,” in The Elections of 1996, ed. Michael Nelson (Washington,
D.C.: CQ Press, 1997), 143–166, esp. 144–147.
46. Richard E. Cohen and William Schneider, “Voting in Unison,” NJ, Janu-
ary 27, 1996, 179–201; and Dan Carney, “As Hostilities Rage on the Hill, Parti-
san-Vote Rate Soars,” CQWR, January 27, 1996, 199–201.
47. Dick Morris, Behind the Oval Office: Winning the Presidency in the
Nineties (New York: Random House, 1997), 79–88, quote 80.
48. The 1996 congressional voting assessments differ from those of 1995; see
Richard E. Cohen and William Schneider, “Soft Center,” NJ, December 14, 1996,
2681–2699; and Rebecca Carr, “GOP’s Election-Year Worries Cooled Partisan
Rancor,” CQWR, December 21, 1996, 3432–3435.
49. Linda Killian, The Freshmen: What Happened to the Republican Revolu-
tion? (Boulder, Colo.: Westview Press, 1998), 340–353; and Jonathan D. Salant,
“Some Republicans Turned Away From Leadership,” CQWR, December 7, 1996,
3352–3354.
50. Eric Pianin, “Clinton Makes First Call for Democratic Congress: Speak-
ing to Steelworkers, He Says Control Is Needed to Protect Social Progress,” WP,
August 9, 1996, A12.
51. Elizabeth Drew, Whatever It Takes: The Real Struggle for Political Power
in America (New York: Viking, 1997), 167.
52. Richard E. Neustadt, Presidential Power: The Politics of Leadership (New
York: John Wiley and Sons, 1960), 33–107.
53. These calculations are adapted from Gallup Monthly Index, data provid-
ed by Gallup Poll, and Vital Statistics on American Politics, 248–249, as updated
with CQ Almanac, various years.
54. Frank Newport, “Small Business And Military Generate Most Confidence
in Americans,” Gallup Poll, August 15, 1997; and John R. Hibbing and Elizabeth
Theiss-Morse, Congress as Public Enemy (Cambridge: Cambridge University Press,
1995).
55. President Clinton would appear to qualify in this category; see Stanley A.
Renshon, High Hopes: The Clinton Presidency and the Politics of Ambition (New
York: New York University Press, 1996); and Michael Kelly, “A Man Who Wants
to Be Liked, and Is: William Jefferson Blythe Clinton,” NYT, November 4, 1992,
A1.
56. Arthur M. Schlesinger Jr. “The Ultimate Approval Rating,” NYT Sunday
Magazine, December 15, 1996, 46–51.
57. Richard B. Cheney and Lynne V. Cheney, “Speaker Newt,” in Kings of the
Hill: Power and Personality in the House of Representatives (New York: Touch-
stone, 1996), 190–210.
CHAPTER 6
Clinton’s Second Transition

Historic Aspirations Amidst Divided Government
MARGARET JANE WYSZOMIRSKI
Customarily, the study of presidential transitions has defined the subject as

a transfer of power and office from one individual to another that occurs
in the approximately eleven-week period between the quadrennial Novem-
ber presidential election and inauguration day the following January. In
practice, however, transitions extend much farther in time at both ends of
the process. Transition planning reaches back into the months before the
election and is affected by the tone, content, and results of the campaign.
Similarly, transition implementation extends at least through the so-called
“honeymoon period” and into the apocryphal first hundred days of the
administration. Although a hundred days is an artificial designation, it is a
convenient and popularly recognized marker that is commonly used by
journalists and scholars.
Transitions require the president-elect to address four primary tasks: (1)
assembling an administration, which involves both making personnel
appointments and organizing the administration team; (2) setting a policy
agenda; (3) determining a political strategy that takes into account both
constraints and assets; and (4) balancing continuity and change as manifest
in politics, people, and policy.
FOUR TYPES OF TRANSITIONS

The particular character of each of these four tasks varies depending not
only on which of four types of transitions is involved but also upon the spe-
cific circumstances of the moment. Regular transfers of power can involve
one of two partisan casts—an inter-party transition or an intra-party tran-
sition. Historically, inter-party transitions are seen as the archetypal transi-
tion and have attracted most scholarly attention. An inter-party transition
occurs when a Republican incumbent is succeeded by a Democratic candi-
date (for example, Clinton succeeding Bush in 1992–93) or the reverse,
when a Republican candidate is elected to succeed a Democratic incumbent
129
130 Clinton’s Second Transition
(for example, Reagan replacing Carter in 1980–81). Inter-party transitions

face wholesale personnel demands, the necessity of enunciating a new pol-
icy agenda, and the attempt to launch a strategic presidency. Change,
rather than continuity, is the emphasis, except in areas that carry a pre-
sumption of independence or bipartisanship such as regulatory and judicial
appointments or foreign affairs.
Intra-party transitions occur when a new individual is elected to succeed
an incumbent from the same party. These have been rarer, especially in the
modern presidency (for example, Bush succeeding Reagan in 1988–89).
During intra-party transitions, continuity and change are more evenly bal-
anced as the new president—particularly if he is a former vice president—
seeks to extend and revise the policies of his predecessor. In intra-party
transitions, the president-elect is less pressed to undertake wholesale per-
sonnel replacement, and begins to establish a different and independent
governance capacity within a continuing partisan context. In cases of both
inter- and intra-party transitions, planning for the transition is expected, is
connected with campaign management, and has been expanding since the
1960s.
Presidential transitions can also occur outside the normal electoral
cycle—occasioned by the death, resignation, or impeachment of an incum-
bent president. Such accidental transitions then elevate the sitting vice pres-
ident. In the modern era, most such accidental presidents go on to run for
election in their own right at the next opportunity, which may be relative-
ly close at hand (for example, Lyndon B. Johnson had about eleven months
before he stood for election in 1964) or may be at quite a distance (for
example, Harry S Truman served out the remaining three years and nine
months of FDR’s last term before running for election in 1948). In such
unexpected circumstances, the formal transition takes place quickly, while
the effective change of administrations takes much longer. In addition, the
immediate transition occurs with little pre-planning, and outside of a direct
campaign and electoral context. Initially, continuity is the byword, since
the new president lacks electoral legitimacy in his own right. Thus, conti-
nuity in personnel and policies is, in large part, expected, while much in the
political context remains unchanged in the absence of regularly scheduled
congressional elections or the legacy of a recent campaign.
A fourth type of presidential transition involves the reelection of an
incumbent president for a second term. During this inter-term transition,
the reelected president undertakes a revision and renewal of his adminis-
tration and agenda. While the three preceding types of transitions involve
a change of executives, this type of transition does not; rather, it marks the
shift between the first and second terms of a continuing president. Here, an
incumbent president has the opportunity to deal with personnel issues
more strategically than a first-term president since he now has fewer
demands, fewer vacancies, and more experience both with governance and
staff performance. In setting a policy agenda, the reelected president can be

less concerned with electoral coalition strategy and strike more of a balance
between the unfinished business of the first term and a more emphatic con-
cern with his potential historical record. Likewise, the political experience
of the campaign and election brings both new and different considerations:
fewer electoral calculations but the specter of becoming a lameduck, a new
congressional power configuration, and succession concerns within his
own party. Certainly, the school of hard knocks will have produced lessons
in policy, governance, and administration, from which the continuing
administration may renew and reposition itself for a second chance at a
“strategic presidency.”
The reelection of William Jefferson Clinton to a second term as presi-
dent of the United States in 1996 resulted in a transition of the fourth type.
The following discussion will focus on the three basic transition concerns:
politics, people, and policy as they evolved during the long transition
process that runs from the campaign into the election, through inaugura-
tion and across the first hundred days of the administration.
POLITICAL CAPITAL AND POLITICAL CONTEXT

Every president-elect comes out of the campaign and election with a set of
factors that establish the amount of political capital that he brings to the
task of governance. Perhaps most important among these factors are the
partisan majorities in each house of Congress, the president’s electoral mar-
gin of victory, and current public opinion ratings.1 To this list, James Pfiffn-
er also adds the zeitgeist of the moment. Sometimes, these factors combine
to allow an elected official to claim a mandate for his particular policy
agenda. Although the term “mandate” is often used loosely and at times
wishfully, a mandate seems to require both a clear and substantial electoral
margin (for example, of landslide proportions) as well as a clear voter
choice among differing policy choices. Campaign victors often seek to
claim a mandate in order to accord their policy preferences greater author-
ity and legitimacy by linking them to the will of the people.2 As we will see,
neither President Clinton nor the Republican congressional leadership were
in a position following the election of 1996 to claim a mandate for a par-
ticular set of policy positions; rather, it might be said that they shared a
public invitation to engage in shared governance as a mode of action.
For the reelected President Clinton, the electoral results conferred only
limited political capital. He won reelection with 49.2 percent of the popu-
lar vote in an election that—at 49 percent—had the lowest turnout of any
presidential election since 1924. Thus, he was the plurality choice of a
minority of registered voters. This outcome made Clinton the first presi-
dent since Woodrow Wilson to be elected twice without a majority of the
popular vote. And Clinton became the first Democrat ever to win a second
term without carrying Congress.3 Furthermore, Clinton only secured a plu-
rality vote for his reelection despite being an incumbent running during a
period of peace and prosperity, against an opponent who ran a weak cam-
paign.4 However, this performance bettered his record of 1992, when he
won only 43 percent of the popular vote. President Clinton ran particular-
ly strongly among Latino voters and among women voters.5 In addition,
Clinton gathered a large electoral college margin, with 379 votes—over a
hundred more than the 270 he needed to win. Thus, Clinton gained more
political capital from the election of 1996 than he had in 1992, but he was
still a plurality president with particular electoral debts to women and
minorities, and he did not have much of a coattail effect on Democratic
congressional candidates.
Meanwhile, Republican congressional leaders could not claim an elec-
toral mandate, either. Although the 105th Congress became the first con-
secutive Republican-controlled Congress since the 1920s, the Republicans’
majority in the House fell from 236 to 225. Nor was Congress the incum-
bent safe institution it had been. The 105th Congress included many new-
comers: 53 percent (or 232 members) of the House and 40 senators had
arrived since 1992.6 Furthermore, the tenor of the 105th Congress was less
revolutionary than its predecessor and was more inclined toward incre-
mentalism. Indeed, many interpreted the joint but conflicting results of the
presidential and congressional elections to be a moderating influence upon
both Democrats and Republicans.
Both President Clinton and the Congress began their new term with
strong standings in the public opinion polls. At his inauguration, the pres-
ident enjoyed a 58 percent approval rating. Meanwhile, public approval of
Congress stood at 45 percent, its highest level since 1991.7 While these high
ratings were accompanied by high hopes for bipartisan cooperation, the
public had lower expectations concerning potential progress on a number
of particular issues. Thus, the signals seemed to indicate that both the exec-
utive and the legislature were expected to avoid partisan bickering while
conducting the public’s business, but were unlikely to be held to high stan-
dards of policy productivity in a divided government situation. Hence the
prevalent zeitgeist seems to have been one that encouraged bipartisanship.
During the opening months of 1997, both the president and House
Speaker Gingrich faced investigations into questionable practices that
diminished their political capital and distracted their attention. Speaker
Gingrich faced a House ethics investigation into his use of charitable foun-
dations that violated tax laws because of their close interactions with
GOPAC, a political action committee that Gingrich used to build toward
the 1994 Republican takeover of Congress. In January, the ethics commit-
tee voted to reprimand the Speaker for his conduct and imposed a
$300,000 penalty to help offset the cost of the probe.8 The investigation
caused a number of Democrats to call for the Speaker to step down.
Although Gingrich succeeded in winning reelection as Speaker, he was a
more chastened and weaker Speaker than he had been in the previous Con-
gress. A number of other changes further depleted Gingrich’s leadership
resources: the new GOP class of 1996–97 was less beholden to the Speak-
er than the class of 1994–95; many sophomore members recognized that
the confrontational strategies of the 104th Congress had damaged the
party; and new Republican rules spread committee and subcommittee
chairmanships more broadly and vested more agenda power in the chairs
than in the previous Congress.9
Meanwhile, the president, the Democratic National Committee, and
various members of his administration faced investigations of their own
concerning fundraising practices. These included allegations that various
White House events (coffees, overnight stays in the Lincoln bedroom) were
used to solicit political donations; that Vice President Gore had engaged in
questionable fundraising practices; and that the Chinese government tried
to influence the 1996 election with $2 million of illegal campaign contri-
butions.10
Allegations and investigations of political scandals and misbehavior cost
both the Democratic administration and the Republican Congress political
capital during the first 100 days of 1997, costs that neither party could eas-
ily sustain and that put an added premium on efforts at bipartisan cooper-
ation and on identifying points of convergence between the Clinton and the
Republican congressional policy agendas. Indeed, in March, House Repub-
lican leaders tried to regain their policy focus by announcing a 13-point
agenda that had as its highest priority achieving a balanced budget by 2002
and that Gingrich referred to as “laying the base for 21st century conser-
vatism.”11 Meanwhile, President Clinton turned to using executive powers
to advance his policy agenda during a period when congressional attention
was dominated by campaign financing investigations. Thus, the President
announced that the federal government would hire 10,000 welfare recipi-
ents as a step forward in welfare reform. He also extended family leave
provisions for federal employees so that parents could attend school con-
ferences. And in the face of stalled action on campaign finance reform,
President Clinton asked the Federal Election Commission to institute a reg-
ulatory ban on one of the most flagrant abuses—soft money.12
PEOPLE: STAFFING THE SECOND CLINTON ADMINISTRATION

A second-term president approaches the task of staffing and organizing the
administration with a number of advantages he did not have the first time
around. To begin with, a second-term president generally has fewer
appointments to make, more time to make them, and a more acute appre-
ciation of both the demands of the jobs and the capabilities of the candi-
dates. In addition, the president can choose to use appointment as a tool in
coalition-building in Congress, within his own party, or among interest
groups. These coalitions may be characterized as political/electoral or pol-
icy-oriented. Thomas Weko notes that although traditionally presidents

made appointments with an eye toward political/electoral coalition build-
ing, since the Nixon presidency, there has been more of an emphasis on
using appointments to build policy coalitions.13 Thus, staffing has become
related not only to the political context but to the policy agenda.
As part of the 1992 transition, President-elect Clinton had over 5,800
appointments for which he was legally responsible, including 590 top exec-
utive appointments requiring Senate confirmation.14 Dealing with this per-
sonnel task requires a sophisticated White House personnel service and
office that for a first-term president must itself be staffed and organized.
Establishing an appointment processing system can be one of the first and
most visible signs of the shift from a campaign mode to a governance
mode. This step is often fraught with tension and jockeying for power
among the president’s adherents. Furthermore, the president-elect must
decide and communicate what criteria will be important in the selection of
appointees, how closely he will monitor appointment decisions, and to
what extent he will consider the preferences of secretaries-designate. Then
specific appointees need to be selected, cleared, and confirmed.
In 1996, as in 1992, President Clinton wanted his administration to
reflect the diversity of America, and he preferred announcing appointments
in teams and groups rather than individually. He also wanted to avoid
some of the missteps of the first term—notably the confusion in vetting
potential nominees that had led to controversies, particularly those around
the Attorney General position, as well as the slowness of the appointment
process and an ambivalence about commissioning an effective White
House chief of staff.15 After running through three personnel directors in
the first half of his first term, Clinton finally settled in February 1995 on
Robert J. Nash, a long-time Clinton supporter who had been deputy per-
sonnel chief for the Clinton I transition. Thus, Nash approached the Clin-
ton II transition with experience and an established and running personnel
office, a close working relationship with the president, familiarity with
many of the incumbent appointees, and a smaller, more manageable task.16
The fact that Nash was also an African–American seemed to underline the
President’s commitment to diversity among his appointees.
Just before Christmas 1996, President Clinton announced the full set of
second-term cabinet selections (see table 6.1). Once again, the president
underscored his concern with diversity. He said, “I believe that one of my
jobs at this moment in history is to demonstrate by the team I put togeth-
er that no group of people should be excluded from service to our country
and that all people are capable of serving. . . .”17 Given the strength of their
electoral support, President Clinton was particularly mindful of securing
representation for women, African–Americans, and Hispanics. The official
14-member cabinet included four women, three African–Americans, and
one Latino. In a gesture to bipartisanship, the cabinet also included a
Republican, former Maine Senator William S. Cohen, as Secretary of

Defense. New appointees were selected to head the following departments:
State (Madeleine K. Albright), Defense (William S. Cohen), Commerce
(William M. Daley), Labor (Alexis M. Herman), Housing/Urban Affairs
(Andrew M. Cuomo), and Transportation (Rodney Slater). Thus, the cab-
inet was about an even mix of new faces and continuing loyalists, with one
Secretary changing his seat from Transportation to Energy (Federico F.
Pena). Except for Senator Cohen at Defense, all of the other new faces may
have been new to the cabinet, but were experienced administrators else-
where in the first Clinton administration. With the departure of Henry Cis-
neros (HUD) and Robert Reich (Labor), two of the most liberal members
of the cabinet left the administration.
Table 6.1: Clinton Cabinets I and II

Clinton II Appointee Cabinet Office Clinton I Predecessor
MADELEINE K. ALBRIGHT State Warren M. Christopher
ROBERT E. RUBINa Treasury Robert E. Rubin since 1/95
WILLIAM S. COHEN Defense William J. Perry since 2/94
JANET RENO Attorney-General Janet Reno
BRUCE BABBITT Interior Bruce Babbitt
DAN GLICKMAN Agriculture Dan Glickman since 3/95
WILLIAM M. DALEY Commerce Mickey Kantor since 4/96
ALEXIS M. HERMAN Labor Robert B. Reich
DONNA E. SHALALA HHS Donna E. Shalala
ANDREW M. CUOMO HUD Henry G. Cisneros
RODNEY SLATER Transportation Federico F. Pena
RICHARD W. RILEY Education Richard W. Riley
FEDERICO F. PENAb Energy Hazel R. O’Leary
JESSE BROWNc Veterans Affairs Jesse Brown
a
Succeeded by Lawrence H. Summers (confirmed by Senate July 1, 1999).
b
Succeeded by Bill Richardson (confirmed by Senate July 31, 1998).
c
Succeeded by Togo G. West Jr. (confirmed by Senate April 28, 1998).
Clinton also made an effort to reshuffle the White House staff at about
the same time, in contrast to the first term, when the staff had followed the
cabinet selection process and seemed to suffer from relative neglect as a
consequence.18 Almost immediately after the election, it was announced
that former deputy White House chief of staff Erskine Bowles would return
to the administration to succeed Leon Panetta as chief of staff.19 Much of
the rest of the senior staff, including the national security adviser (Samuel
Berger), the National Economic Council director (Gene Sperling), and the
U.S. Trade Representative (Charlene Barshefsky) were veterans who had
worked their way up during the first term and were “marked less by any
stamp of ideology or personality than they are by tempered, tested alle-
giance to the boss.”20 With the departures of Harold M. Ickes and George
Stephanopoulos from the White House staff, the administration lost two
more of its most prominent liberal members. Although these cabinet and
staff changes gave Clinton an inner circle that was “loyal, tested, trusted,
and collegial,” some management experts worried that the group might
prove too homogenous, with a tendency toward group-think.21
Cross-cutting the cabinet and the White House staff were two policy
teams: national security and economic policy. By mid-December, the eco-
nomic team had taken shape. It seemed to put aside the internal battles of
the first term between liberal populists who supported social programs and
non-ideological pragmatists with an eye on the financial markets. In the
second term, three pragmatic investment bankers—Treasury Secretary
Robert Rubin, OMB Director Franklin Raines, and White House Chief of
Staff Erskine Bowles—assumed the lead. Others in the economic team
seemed to be characterized by loyalty, pragmatism, familiarity, and a dedi-
cation to consensus. The president said his main criteria in selecting team
members was that “he wanted people he had confidence could do the job
. . . [and] he wanted people who could work together as a team.”22 The
newest member of the economic team, and the last to be appointed, was
Janet Yellen, a Berkeley economist who had been a governor of the Feder-
al Reserve for three years. Yellen was selected to chair the Council of Eco-
nomic Advisers.23
One of the president’s appointment priorities was reformulating his
national security team for a second term. The incumbent Secretary of State,
Warren Christopher, as well as the current Secretary of Defense, William
Perry, had both indicated that they would be leaving the administration.
The first-term National Security Adviser, Anthony Lake, was interested in
moving on—most likely to the Central Intelligence Agency. Women’s
groups were pressing the president to appoint a woman as part of his top
national security team.24 President Clinton was again interested in forming
a team whose members would complement each other, work together, be
suited to the task of helping him define foreign policy in the post-cold-war
era, and fit what would probably be his own more activist approach dur-
ing a second term.25 When the president announced his new national secu-
rity team on December 6, it was notable for the symbolism of including the
first female Secretary of State in history (Madeleine Albright) and the first
prominent Republican named to a post in his administration (William S.
Cohen). As with other White House positions, the deputy national securi-
ty adviser, Samuel Berger, was promoted to succeed outgoing National
Security Adviser Anthony Lake, who was nominated as Director of the
CIA.26 Later, Representative Bill Richardson (D-N. Mex) would be nomi-
nated for Ambassador to the United Nations, replacing Madeleine
Albright.27 Thus, except for Defense Secretary-designate Cohen, the rest of
the national security team was typical of the general second-term appoint-
ment pattern: diverse, tested, team players.
Thus, in many ways, the president succeeded in managing his
second–term personnel tasks better than he had the first term. He main-
tained his commitment to putting together a diverse administration, while
acknowledging key constituencies with representative appointments. He

retained his emphasis on assembling institutional and policy teams and
moved relatively quickly to get cabinet and senior White House staff in
position. Like many second-term presidents, he looked inside his adminis-
tration to find new blood and rewarded tested service with promotions and
recalls. In an effort to cultivate better bipartisan working relations with the
Republican congressional leadership, he appointed a Republican to an
inner cabinet spot and tempered the liberal composition of his advisory cir-
cle.
However, not all of his nominees had smooth confirmation processes:
after months of rancor, Anthony Lake withdrew his name from considera-
tion to become Director of the CIA, while Alexis Herman, before winning
confirmation as Secretary of Labor, endured months of controversy and
delay, mostly over her role in White House fundraising activities. Similar-
ly, although the most senior administration positions were filled in a more
orderly and timely manner than during his first term, the process of filling
subcabinet and ambassadorial appointments was slow. As of mid-April,
only 18 names had been sent to the Senate for 130 vacant positions that
required confirmation.28 Seven months into the term, it was reported that
30 percent of the top political jobs in the administration remained unfilled,
while another 67 ambassadorships were also vacant.29 While some of the
delay is attributable to a lengthy clearance process and to Senate handling
of nominations, in other cases, vacancies seemed to be clustered in depart-
ments which had new secretaries and thus were in more of a transition than
others. Since this was a second term, the administration was fortunate in
that it had deputies and others in place who served in an acting capacity
while the appointment process proceeded at its unhurried pace.
POLICY: SETTING A SECOND-TERM AGENDA

Each president-elect moves from the campaign and election through the
transition and into governance with a set of ideas and issues that he thinks
are important and merit official attention and action. During the campaign,
many issues arise and candidates take positions and make pledges on many
more concerns than they can possibly address upon taking office. Elective
politicians also carefully monitor the public opinion polls, both to see what
is important to the citizenry and to assess potential public response to spe-
cific policy options. The party platform is a virtual compendium of possi-
ble policy concerns for an incoming administration. Events such as the
Oklahoma City bombing occur unexpectedly and put additional issues—in
this case, domestic terrorism—before the public, eliciting responses from
policymakers. Increasingly over the past thirty years, think tanks—both
institutional and informal—have sifted through current issues to offer
agenda advice to the president-elect.30 Interest groups formulate their own
policy concerns into position papers, seeking promises from candidates as
part of the campaign process. In the 1990s, the parties through their con-
gressional leaders have offered policy agendas such as the Republican
“Contract with America” in 1994 and the Democrats’ “Families First”
agenda for 1996.31 In these various ways, issues become part of the sys-
temic agenda.
During the transition, this broad systemic agenda undergoes a process of
winnowing, focusing, and planning as the president-elect seeks to trans-
form a campaign discussion agenda into an institutional action agenda con-
ducive to strategic governance. Deciding upon this action agenda is an
important element of launching a “strategic presidency.”32
Ben Heineman has argued that a president’s “first order issues” agenda
in domestic affairs should be limited to five or six items.33 Paul Light’s
analysis of the domestic agendas of Presidents Kennedy through Carter
revealed that each initially focused on only three to seven major issues.34
Strategic agenda-setting is equally important to second-term presidents for
the following four reasons: (1) they may have more latitude to take on
issues free from the calculus of reelection concerns; (2) the campaign and
current conditions may have given rise to new issues; (3) they are likely to
continue to be concerned with issues that they feel were not adequately
addressed in the first term; and (4) they are more explicitly concerned with
establishing their place in history. In addition, reelected presidents may
want to demonstrate that their administrations still retain vigor and vision
despite their presumed lameduck status.
It is generally argued that second-term policy successes are most likely
in the initial two years, before midterm elections make it difficult for the
president to keep control of the agenda.35 The twin cycles of decreasing pol-
icy influence and increasing policy competence, identified by Paul Light,
place a premium on the first hundred days of both first-term and second-
term presidents. Furthermore, a second-term strategic agenda presents an
opportunity to build on his first-term performance.
DEVELOPING A PRIORITY AGENDA

Selection of priority agenda items reflects the interplay of many factors.
The content of a strategic agenda evolves out of the campaign, reflects the
legacies of the preceding administration and the preceding Congress, and
responds to current events. As such, setting a strategic agenda is an exer-
cise in issue and option definition rather than in issue discovery. Particu-
larly for a second-term president, the issues of a strategic policy agenda are
likely to be embedded in his party’s platform, in part because a sitting pres-
ident plays a crucial, even decisive role in its drafting. In addition, key
issues may have been incubated in congressional hearings and perhaps pro-
posed but unrealized legislation. Often potential priority issues surface in
public opinion polls throughout the campaign and election period.
As the transition continues, the president has three closely sequenced

opportunities to enunciate his policy agenda. These opportunities include
the inaugural address, the State of the Union address, and the submission
of his proposed budget. Thus we can review these five sources to discern
the basic outlines of President Clinton’s second-term policy agenda: (1) the
party platform, (2) public opinion polls, (3) the inaugural address, (4) the
State of the Union message, and (5) the fiscal year 1998 budget proposal.
Campaign documents and polls help define initial policy goals, but they do
not dictate agenda priorities once an administration takes office. A cross-
reading of all of these indicators reveals which issues appear repeatedly and
prominently as well as the macro-themes and core values that weave
through the issue agenda.
The Democratic Platform of 1996
Although the policy significance of party platforms is sometimes dismissed,
Gerald Pomper has demonstrated that platforms are important compila-
tions of party positions and promises that indicate issues that are likely to
see policy action after a successful election campaign.36 As Jeff Fishel has
noted, campaign platforms and candidate pledges “provide a foundation
from which an administration’s early policy initiatives emanate.”37 In 1996,
this was clearly the case, as an incumbent president seeking reelection,
Clinton, exerted considerable influence over platform content. The pro-
ceedings of the 1996 Democratic convention emanated a “wave of togeth-
erness,” and the platform sailed smoothly through the drafting process and
was adopted without debate.38 Many of the platform’s ideas and emphases
resurfaced in Clinton’s 1996 acceptance speech.39
Three of the four major sections of the Democratic platform were orga-
nized around the core values emphasized by President Clinton on the cam-
paign trail—opportunity, responsibility, and community.40 “Opportunity”
emphasized economic growth and balancing the budget, education, and
what was called economic security (but might, in other years, have been
termed social security and welfare). The “Responsibility” section inter-
wove a redefined role for government with a reinvigorated sense of
personal responsibility. Thus, it encompassed a range of law and order con-
cerns (crime, community policing, guns, criminal and juvenile justice,
drugs, and domestic violence) as well as personal responsibility in many
areas (immigration, welfare/work, child support, teen pregnancy, and
choice). It also addressed political responsibility with an emphasis on
reinventing government and reforming government ethics, lobbying, and
campaign finance. The section on “Security, Freedom and Peace” was
essentially a statement of national security concerns and positions. Finally,
the “Community” section encompassed a catchall set of planks ranging
from family issues, urban and housing policies, agriculture, and environ-
mental issues to discrimination and civil rights, and voluntarism.
Public Opinion Polls and Issues

As reflected in polling during the campaign of 1996, “no single issue or
problem dominated the minds of American voters.”41 The lack of promi-
nent issues during the campaign coupled with the generally favorable mood
of the electorate, especially concerning the state of the economy, were pos-
itive electoral factors for the incumbent president.42 However, these positive
factors exerted relatively few demands on the formulation of his second-
term agenda.
The 1997 Inaugural Address
Seeking to position himself as the first president of the twenty-first centu-
ry, President Clinton’s second inaugural address repeatedly referred to the
new century or the start of the twenty-first century, linked the promise of
the future with the historical legacy of the nation, and echoed the bridge
metaphor that had been a hallmark of his 1996 convention acceptance
speech.43 Whereas change had been a theme of Clinton’s first inaugural
address, the second was concerned with realizing the promise of the future
by building on strength, prosperity, and power.44 Much unfinished business
remained from the first term: reforming government, politics, and the bud-
get; educational reform and opportunity; improving racial harmony; and
assuring societal security.
The inaugural address reiterated the call for a reinvigorated sense of
“responsible citizenship” in which government is neither the problem nor
the solution but in which the American people are the solution. This “new
sense of responsibility” would be manifest in a new spirit of community,
and heightened personal responsibility for ourselves, our families, our
neighbors, and our nation.
The State of the Union Address
Although much more explicit about offering specific proposals, Clinton’s
State of the Union address returned to the macro-theme of pursuing our
historic role and common future by attending to unfinished business and
preparing Americans “for the bold new world of the 21st century.”45
Undergirding this macro-theme were the core values of opportunity,
responsibility, and community.
Unfinished business included balancing the budget, renewing democra-
cy through enacting campaign finance reform, and completing the task of
welfare reform, particularly by facilitating the move from welfare to work.
Education emerged as the president’s “No.1 priority for the next four
years.” His ten-point action plan emphasized opportunity and preparation
for the future. Highlighting the importance of education, the president
called it “a critical national security issue for our future” and one where
“politics must stop at the schoolhouse door.”
“Harness[ing] the powerful forces of science and technology to benefit

all Americans” included a smorgasbord of specific items ranging from pro-
moting connection to the Internet and “build[ing] the second generation of
the Internet,” through support for the space program and for medical
research. It also gave a scientific rather than a social welfare cast to a num-
ber of health policy proposals.
“Building Stronger Communities” entailed more attention to safer
streets, a “full-scale assault on juvenile crime,” the “largest anti-drug effort
ever,” and revival of poor urban and rural neighborhoods. The president
also emphasized the importance of service and called for a millennial cele-
bration of “the American spirit in every community, a celebration of our
common culture . . . the world’s beacon not only of liberty but of creativi-
ty. . . .”
Harking back to the Democratic platform’s section on “Security, Free-
dom, and Peace,” the president spoke of America’s world leadership role as
a balance between continuity and change while dealing with a rapidly
changing world.
“Strength Through Diversity” returned to an endorsement of the core
value of opportunity amidst diversity, calling diversity “our greatest
strength,” seeking a resolution of “our differences in our faiths, our back-
grounds, our politics,” and asking all Americans to be “repairers of the
breach.” Our leadership in the world depends upon remaining “one Amer-
ica.”
Rhetorically the theme of America’s historic role was prominent:
“America is far more than a place; it is an idea—the most powerful idea in
the history of nations. . . .” The president struck a purposeful historic par-
allel to the 1960s when he concluded with the observation that
Tomorrow there will be just over 1,000 days until the year 2000. One
thousand days to prepare our people. One thousand days to work togeth-
er. One thousand days to build a bridge to a land of new promise.
Thus, the reference to a thousand days conjured up images of John F.
Kennedy, while the “bridge” alluded to a key image of the 1996 Democra-
tic convention and campaign, and the “land of new promise” echoed Clin-
ton’s inaugural speech linkage between Martin Luther King’s “Dream”
with the American dream as a land of promise.
The Fiscal Year 1998 Budget Proposal
Shortly after the State of the Union address, President Clinton presented his
fiscal year 1998 proposed budget. Continuing the concern for bipartisan-
ship, the fiscal 1998 budget sought to address both presidential and leg-
islative agendas. As such, it was not greeted with the frosty “dead on
arrival” reactions that had become typical during years of divided govern-
ment and that had greeted President Clinton’s budget proposals in 1995
and 1996. Rather, Senate Budget Committee Chair Pete V. Domenici called
it “a very good starting point,” and Senate Majority Leader Trent Lott
expressed a hope that negotiations “might produce a balanced budget
agreement within six weeks or so.”46 The budget proposed to end deficits
by 2002 while increasing spending for education by 13 percent; for com-
bating drugs, juvenile crime, and terrorism by 5 percent; and for job train-
ing programs as part of welfare reform by 8.5 percent.47 The proposal also
juggled specific tax and spending cuts. In his emphases on issues like edu-
cation, health care for uninsured children, budget balancing, welfare
reform modifications and implementation, and fighting crime, President
Clinton provided more detail and a price tag for issues and programs that
he had sketched out in the campaign and the State of the Union address.
After the divided government results of the 1996 elections and after the
confrontational budget negotiations of 1995 and 1996, both the president
and Republican congressional leaders saw that it was in their interests to
promote a more cooperative budgetary process; in addition, they felt they
were not so far apart on goals.48
Ingredients of the Initial Second-term Agenda
During the election and transition period, six priorities, five of which were
domestic, emerged as an agenda. These were achieving a balanced budget,
continuing the process of welfare reform, a major emphasis on education
programs and tax incentives, targeted tax cuts, fighting crime, and forging
a new international leadership role for the nation. In what might be regard-
ed as an implicit priority, Clinton also sought to advance health care and
social welfare programs, including health insurance for more children and
protection for Social Security and Medicare/Medicaid. Repeatedly during
the campaign, the transition, and into the first hundred days, these policy
priorities appeared and were pushed toward policy action. Furthermore,
this agenda was pursued amidst a set of macro-themes: bipartisan consul-
tation, a view toward Clinton’s historical legacy, and an emphasis on the
core values of responsibility, opportunity, and community.
It is important to note that the five domestic priorities identified above
were essentially an institutional agenda that both the president and Con-
gress shared. The Republican leaders in Congress were also concerned with
a balanced budget, welfare reform, crime, education, and targeted tax
cuts—although the two parties certainly differed as to program specifics
and approaches, spending and cutting levels, and preferred policy targets.
Similarly, both the Republican leaders and President Clinton understood
the political utility of avoiding partisan confrontation and striving for at
least the appearance of bipartisan consultation. Both were concerned with
their historic roles: Clinton with his administration’s legacy and his role as
the first president of the twenty-first century; congressional Republicans
with establishing new historical patterns in legislative politics and seeking
to institutionalize the new conservatism of the “revolution” of 1994.
Similarly, the president and Congress could subscribe to the core values
enunciated by Clinton during the election and transition, even as they dif-
fered as to the means of realizing these values in practice and in policy.
Republicans might prefer less government responsibilities while Clinton
might emphasize more responsible and reformed government as well as a
stronger sense of responsible citizenship, but both could agree on more per-
sonal, family, and community responsibility. The concept of opportunity
might take on more of a free enterprise cast among Republicans while sug-
gesting civil rights, diversity, and other social goals for the president. Pres-
ident Clinton might see community in terms of improved racial harmony,
revitalizing poor neighborhoods, stronger environmental awareness, and
voluntarism. Congressional Republicans, in turn, might find community
more in family, family values, and morality.
Thus, despite different interpretive lenses, a governmental policy agenda
emerged during the 1996–97 transition. Though some commentators
observed that the 1996 election was not particularly concerned with a
prospective policy agenda, and even though divided government was
retained, both parties and both elected branches of government agreed on
a handful of issues that were of first importance. There was also agreement
on two cross-cutting themes—bipartisan consultation and an eye towards
history—as well as on the importance of three core values—opportunity,
responsibility, and community. Such agenda agreement, however, did not
indicate like-mindedness. Rather, it was an invitation to debate and nego-
tiate means and methods. Indeed, the initial agenda of the 1997 transition
provided an opportunity to pursue common concerns amidst distinct inter-
ests and separate visions.
POLICY AND GOVERNANCE: THE FIRST HUNDRED DAYS

As the first hundred days progressed, President Clinton kept refining and
adjusting his issue agenda as well as seeking convergence between his agen-
da and that of the Republican-controlled Congress. Following his inaugur-
al call for bipartisanship, the effort to cultivate such political cooperation
followed a bumpy road, but remained a fulcrum point of political strategy
among Democrats and Republicans, executive and legislature.49 In propos-
ing his budget, the president announced that it included items he hoped
would appeal to the Republicans, but then rankled them with a highly par-
tisan speech to the Democratic National Committee that “chastised Repub-
licans as fiscally irresponsible and racially divisive.”50
On February 11, President Clinton met with congressional leaders at
their invitation in the President’s Room of the Capitol to discuss the
prospects of a mutual agenda. The president intended the meeting to “sym-
bolize a new era of constructive engagement” in the spirit of bipartisan
interaction.51 Vice President Gore characterized the discussions as “uni-
formly excellent” in tone; Republican Senate Leader Trent Lott referred to
them as “an overall good session . . . ” with the “atmosphere the best” he
had experienced in quite some time.52 White House Press Secretary Michael
McCurry reported that President Clinton was “more than satisfied” and
“delighted with the tone.”53 As a result, national political leaders agreed to
a five-point list of issues on which they hoped to forge agreement and work
toward productive policy action. The agenda included (1) a tax credit
incentive for employers to hire current welfare recipients as part of the wel-
fare-to-work reforms; (2) exploring tax cuts for individuals within a bal-
anced budget framework; (3) juvenile justice measures; (4) aid for the Dis-
trict of Columbia as a laboratory for urban reform; and (5) education.
Clearly, there was continued resonance with the issue agenda of the cam-
paign, election, and initial transition: bipartisanship, continuing the
process of welfare reform, crime and juvenile justice, and the continued
presumption of working toward a balanced budget. Also, both branches
seemed to agree on the importance of education as a policy priority, even
though they differed considerably on what specific measures were to be
championed. Notably absent from this mutual agenda were other issues
that had been emphasized by the president during the transition: most
notably campaign finance reform, but also environmental policy (the
Superfund and toxic waste clean-up) and the extension of health care cov-
erage to more children.
By March, some House Republicans—especially sophomore members of
the class of 1994—were frustrated by the slower pace of the 105th Con-
gress and complained that Speaker Gingrich was being erratic, unfocused,
and ineffectual in setting an agenda or strategy.54 Partly in response to these
sentiments, House Republicans announced a 13-point list of their legisla-
tive priorities entitled “Creating a Better America for Ourselves and Our
Children.”55 Like the February bipartisan agenda, the House document
voiced concern with balancing the budget, cutting taxes, and community
renewal in Washington, D.C. Its concern with ensuring “the integrity of
American elections” referred to campaign finance reform in the context of
criticism of Democratic fund-raising practices and their investigation.
Although reminiscent of the 1994 “Contract with America” agenda,
“Creating a Better America” was different in notable ways. It was not a
campaign document; it was less specific; and it was not targeted to a 100-
day timetable but toward the work of an entire congress.
Within the first hundred days, the promise of drafting a bipartisan
agreement to balance the budget was realized. On May 2, 1997, President
Clinton and Republican congressional leaders announced a “historic agree-
ment” to “produce the first balanced budget in a generation—a feat that
has eluded six presidents and 14 Congresses” over nearly 30 years.56
According to a CQ analysis, congressional Republicans secured “a politi-
cal trifecta: a balanced budget in five years; significant, permanent tax cuts;
and a plan to keep Medicare solvent for another decade.”57 These were all
issues high on the Republican policy agenda, and evidence of the Republi-
can impact on changing political dynamics and policy assumptions since
the first year of the Clinton presidency.
For President Clinton, the agreement also embodied a number of impor-
tant agenda items: it was an important step toward a historic achievement
since a balanced/surplus budget had not been seen since 1969; in addition,
it included a major spending increase for education programs, restored
some welfare benefits that had been cut during the initial stage of welfare
reform, and included targeted tax cuts and tax credits. In other words,
the May 1997 bipartisan, balanced-budget agreement was a win-win sit-
uation for both the president and Republican congressional leaders—a
package that was also facilitated by unusually robust economic growth
that generated unexpected extra revenues. Thus, politics, policy, and eco-
nomics converged to produce a window of opportunity for a balanced-
budget agreement.
Similarly, legislation on juvenile crime issues got off to a propitious start.
Previously characterized by bitter partisan feuding, the House Judiciary
Subcommittee on Crime began working on a bill (HR 3) that a Justice
Department spokesman described as “a genuine effort on the part of many
people to try to work collegially and . . . try to define a comfortable mid-
dle ground.”58 By May, the House considered a juvenile crime bill that had
the makings of a grand compromise between the Clinton White House and
congressional Republicans, only to have the plan fall apart under the
weight of procedural maneuvering by House Democrats, many of whom
were unhappy with the details of the package. Thus, at the end of the first
hundred days, crime and juvenile justice policy had suffered a near miss in
the House and awaited a second round of negotiating and politicking in the
Senate.59
President Clinton’s educational plan was detailed and complex. It
employed program spending as well as tax breaks to accomplish policy
ends.60 In the 104th Congress, education policy had been a battleground,
with Republicans calling for the elimination of the Department of Educa-
tion. Now, in the 105th Congress and Clinton’s second term, Republicans
wanted to avoid being characterized as “anti-education” but continued to
differ from the president substantially with regard to program priorities
and approaches. However, in 1997, House Education and the Workforce
Committee Chair Bill Goodling (R-Pa.) did not reject the president’s edu-
cation proposals but instead hoped to refashion them. Similarly, Senator
Paul Coverdell (R-Ga.), who was developing the Senate Republican educa-
tion plan, followed up the February bipartisan agenda meeting with dis-
cussions with Education Secretary Richard Riley. In other words, the effort
at bipartisan consultation seemed to have improved the atmosphere sur-
rounding education policy negotiations.
Although much of the president’s education agenda was tied up with the
long budget process, within the first hundred days of his second term,
many of the president’s proposals were in play. By early May, bipartisan
agreement to reauthorize the Individuals with Disabilities Education Act
(IDEA) had been approved by both the House and Senate and was on its
way to a conference committee to resolve minor, technical differences
between the two versions.61 Clearly, IDEA was a piece of the president’s
education agenda that had particular resonance with his core value of
opportunity.
On both Medicare and Social Security, the parties agreed on a goal of
guaranteeing financial viability, but differed significantly on what would be
needed and preferable options to achieve this goal. With regard to Social
Security, even the Advisory Council on Social Security that reported its rec-
ommendations on January 6 was divided, displaying a three-way split
between maintaining the system and benefits, privatizing a large portion of
Social Security investments into private securities, or partially privatizing
the system by the creation of individual security accounts.62 Similarly with
regard to Medicare, Republicans called for “structural reforms” while
Clinton focused on reducing payments to providers (doctors, hospitals, and
HMOs) rather than fundamental program changes.63 Clearly, these issues
were on the governmental agenda, but it would take much more incuba-
tion time before a consensus could be forged to support action.
THE TRANSITION OF 1996–97 IN BROAD PERSPECTIVE

Although President Clinton enjoyed progress on many items of his policy
agenda, few of these moved to formal and final action. Nevertheless, posi-
tive groundwork had been laid in the first hundred days of his second term.
Also true to his call for bipartisanship, working relations between Repub-
lican leaders in Congress and the White House seemed more congenial and
less combative, perhaps in response to the split outcome of the 1996 elec-
tion. Thus, in 1997, both parties and both branches came to see bipartisan
consultation and cautious action as being in their political interests.
Together, President Clinton and the 105th Congress reached a historic
balanced-budget agreement and piloted their way through the transition of
1997. An action agenda had been selected from the broader discussion
agenda of the campaign and the unfinished work of the previous Congress
and administration. The long governance process had been engaged and
the groundwork laid for developing consensus on policy options and
specifics in a number of issue areas. Thus, this initial agenda-setting of the
transition of 1997 was largely a prelude to the intricate process of policy
formulation and enactment that lay ahead.
Measured by the standard of FDR’s first hundred days of the New Deal
in 1933 or by the standard of the first hundred days of the 104th House of
Representatives’ “Contract with America,” Clinton’s first hundred days in
1997 did not produce a wealth of newly enacted legislation. Rather, it

reopened bipartisan channels of communication and constructive consulta-
tion. That it could set a workable bipartisan tone and lay such substantial
policy groundwork in such a short period of time was in itself noteworthy.
This constructive beginning was even more of an accomplishment given the
distractions and partisanship stoked by extensive and ongoing investiga-
tions of both the president and the Speaker as well as of the fund-raising
practices of both the Democratic and Republican parties during the 1996
campaign. Indeed, in mid-May, at the end of the transition period, a New
York Times article observed that Clinton had “acted so quickly on the ini-
tiatives that his presidential campaign promoted as the central goals of a
second term that his supporters and antagonists alike are asking how he
will fill his remaining years in office.”64 Thus, it would seem that President
Clinton did indeed succeed in getting his second term off to a “strategic
presidency” start. Maintaining such momentum and realizing the fruits of
seeds planted during the transition, however, would still be subject to par-
tisan tensions.
In comparison to the first-term transition, Clinton’s second-term transi-
tion was more sure-handed. With the exception of Anthony Lake’s ill-fated
nomination to head the CIA, cabinet and White House staffing progressed
smoothly and for the most part quickly. Many of the new appointees were
not really new, but rather promotions, recalls, or relocations of people who
had amassed experience in the first term. Since the learning curve for expe-
rienced presidential personnel is generally shorter and less steep, the sec-
ond-term appointees could keep the wheels of government rolling without
the significant missteps, miscues, and overreaches of the first term.
Two final observations emerge. First, the political context of divided
government, which characterized the start of President Clinton’s second
term, was quite different than the context of party government that had
obtained during 1993. As David Mayhew has demonstrated, divided gov-
ernment does not forestall or minimize the prospect of policy-making pro-
ductivity.65 However, divided government may prompt a variation in the
tactics and expectations of achieving a strategic presidency, as well as in the
character of full-term policy-making cycles.
For example, as the second term began, the president reverted back to
his New Democrat character and emphasized the cultivation of a biparti-
san consultative style. This tack was in marked contrast to Clinton’s 1993
performance, when he moved to the left and attempted to pursue an ambi-
tious policy agenda while relying almost exclusively on Democratic support
in Congress, helping prompt a budget stalemate and the health-care reform
imbroglio. Stresses within the Democratic party occasioned intramural pol-
itics and generated significant enough general discontent to have con-
tributed to the Republican midterm gains of 1994 and the election of the
first Republican-controlled Congress in 40 years. Thus, divided govern-
ment may in some ways present a prospect for a more manageable policy
agenda than does party government, particularly for narrowly elected pres-
idents who have short electoral coattails.
In terms of policy agenda, a situation of divided government—especial-
ly in which both the president and congressional leaders operate within a
macro-theme of bipartisanship—may require that a president be explicit
and diligent in looking for points of convergence between his own values
and priorities and those of congressional leaders from another party and/or
philosophy. If so, then the president may have less latitude in determining
his own priority agenda, while being more mindful of bipartisan strategic
considerations. Agenda-setting in a situation of divided government may
also yield a different cycle of policy activity. Although initial agenda focus
and success will remain important, many policy priorities will require more
time to gestate, and more work will be needed to align values, assumptions,
and preferences as a prelude to coalition-building. Thus, it may be partic-
ularly important to outline a full agenda and secure some consensus on
basic priorities in principle, while fully expecting a long, perhaps multi-year
policy development process. Hence, rather than a steady decline in policy
influence as a term progresses, a divided-government president may build
toward a more dispersed set of policy windows of opportunity throughout
the term and may actually enjoy a series of annual strategic presidency
prospects.
If this proves to be the case, then 1997 could be seen as a year in which
some groundbreaking agreements were secured (for example, the balanced-
budget agreement and NATO expansion), and agenda status for other
issues agreed upon (for example, education, continued welfare reform, and
juvenile crime) as both executive and legislature sought to fulfill campaign
promises and meet electoral expectations. Although 1998 and 1999 held
the potential for incremental and cumulative progress on many issues, such
potential was significantly diminished by acrimonious campaign financing
hearings in Congress, petty scandals that involved various administration
officials, the special prosecutor’s investigations of the president and first
lady, and ultimately the impeachment and Senate trial of President Clinton.
Second, despite being reelected by only a plurality of the popular vote
(and with an unusually low turnout rate), working with a Congress led by
the opposing party, and experiencing significant turnover in key adminis-
trative personnel, Clinton managed to keep the policy spotlight focused on
himself (and Vice President Gore), thus generating the impression that the
governmental agenda was very much his presidential agenda. Neither cab-
inet members nor White House advisers evidenced the high visibility and
policy image that had been common in other presidencies. Rather than pol-
icy formulators and agenda setters, cabinet and White House staff mem-
bers seemed to function more in the roles of policy spokesmen, imple-
menters, and behind-the-scenes political deal-makers.
A number of elements may have helped cultivate this policy spotlight for
President Clinton (and Vice President Gore, who is literally often in the pic-
ture, if not at the president’s side). Clinton’s reputation as a “policy wonk”
who is interested in and in command of policy detail and enjoys mastering
policy substance may have initially helped position the president as the pol-
icy generator of his administration. That both Vice President Gore and
First Lady Hillary Rodham Clinton share this “policy wonk” image also
has contributed to this reputation. The visibility and degree of policy dele-
gation that has characterized Vice President Gore’s leadership of the admin-
istration’s reinventing government effort and Gore’s clear influence in tech-
nology and environmental policy issues also helps keep the spotlight on the
presidency.
Additionally, President Clinton seems to have adopted his successful and
highly personal campaign device of town meetings and policy seminars into
the governance arena. For example, in mid-April 1997, the president host-
ed a White House conference on early childhood development that spoke
to Clinton’s agenda on programs of early social welfare and making edu-
cation a top domestic priority.66 The following week, President Clinton
convened a three-day bipartisan summit in Philadelphia co-chaired by for-
mer president George Bush and politically popular Colin Powell, and fea-
turing Vice President Gore and former presidents Jimmy Carter and Gerald
Ford. Intended to inspire voluntarism nationwide, the Summit for Ameri-
ca’s Future was particularly focused on reaching at-risk youth and in
emphasizing a community service component to “the meaning of citizen-
ship in America.”67 In such cases, the president effectively demonstrates his
engagement in policy substance and issue formation, influences which
issues reach the public discussion agenda, and secures significant media
coverage.
If these propositions concerning President Clinton’s second term bear
out, then they may have broader implications for our study and under-
standing of the presidency as it enters the next century as well as for pres-
idential-congressional relations. Meanwhile, this analysis of the inter-term
transition of 1996–97 would seem to indicate that second-term transitions
merit more attention and exhibit more interesting aspects than might typi-
cally be thought when regarded simply as marking the continuation of a
previously elected incumbent.
NOTES
1. James P. Pfiffner, The Strategic Presidency: Hitting the Ground Running,
2nd ed. (Lawrence: University Press of Kansas, 1996), 111; idem, “Presidential
Transitions: Organization, People, and Policy,” paper delivered at the 1996 Annu-
al Meeting of the American Political Science Association, San Francisco, 18. Also
see Paul Light, The President’s Agenda: Domestic Policy Choice from Kennedy to
Carter (Baltimore: Johns Hopkins University Press, 1982), 26.
2. A. Lane Crothers, “Asserting Dominance: Presidential Transitions from
Out-Party to In-Party, 1932–1992,” Polity 26 (Summer 1994): 811.
3. NYT, November 11, 1996.
4. Rhodes Cook, “Even With Higher Vote, Clinton Remains Minority Pres-
ident,” CQWR, January 18, 1997, 185–188.
5. Voter News Service exit polls for 1996 in “Presidential Election Exit Poll
Results,” CNN/Time All Politics, 1996, http://allpolitics.com/elections/natl/.
exit.poll/index1 (November 8, 1996).
6. Allan Freedman, “Lawyers Take a Back Seat in the 105th Congress,”
CQWR, January 4, 1997, 27–30.
7. NYT, January 20, 1997.
8. Jackie Koszczuk, et al., “Committee Votes for Reprimand, $300,000 Fine
for Gingrich,” CQWR, January 18, 1997, 160–161.
9. Donna Cassata, “Freshmen Bring a Bit Less Fire, More Savvy to Capitol
Hill,” CQWR, January 4, 1997, 25–26.
10. Bob Gravely, “From Arms to Buddhists to Coffee: The ABCs of the Inves-
tigations,” CQWR, April 5, 1997, 797, 800–801.
11. Jackie Koszczuk, “Republicans Set the Stage: Try 104th Agenda Again,”
CQWR, March 8, 1997, 575.
12. Peter Baker and John F. Harris, “Clinton to Pursue Agenda Through Exec-
utive Powers,” WP, April 11, 1997, A1, A20; and Peter Baker, “Clinton To Seek
FEC Ban on ‘Soft Money,’” WP, June 4, 1997, A1, A9.
13. Thomas J. Weko, The Politicizing Presidency: The White House Person-
nel Office, 1948–1994 (Lawrence: University Press of Kansas, 1995), 5, 46,
109–121.
14. Pfiffner, The Strategic Presidency, 171–172.
15. On the early stages of the second Clinton transition, see Pfiffner, “Presi-
dential Transitions: Organization, People, and Policy.”
16. Al Kamen and Stephen Barr, “Presidential Personnel Chief Promises
Kinder, Smoother Transition,” WP, November 21, 1996, A23.
17. As quoted in the WP, December 21, 1996, A16.
18. John F. Harris, “Second Transition to Show if Clinton Learned Lessons,”
WP, November 10, 1996, A1, A20.
19. Peter Baker, “President Taps Bowles, Trouble-shooter and Friend, to Lead
Staff,” WP, November 9, 1996, A1, 14.
20. For more details, see John F. Harris, “Clinton Fills Out Roster for White
House Team,” WP, December 19, 1996, 25. Quotation from Todd S. Purdum,
“The Ungreening of the White House Staff,” NYT, December 22, 1996, Section 4,
10.
21. Steven Pearlstein, “Clinton’s Cabinet: Dream Team or Trouble?” WP,
December 22, 1996, H1, H2.
22. Clay Chandler, “Where Pragmatism and Process Prevail,” WP, December
14, 1996, A11.
23. Louis Uchitelle, “An Appointment That Draws No Fire,” NYT, January
7, 1997, D3.
24. John F. Harris, “Women’s Groups Seek Entry to President’s Inner Circle,”
WP, December 3, 1996, A1, A11.
25. Alison Mitchell, “Clinton Still Studying Choice for National Security
Team,” NYT, December 5, 1996, A1, A16.
26. See coverage in the WP and the NYT on December 6, 1996.
27. “Senate Confirms Richardson as Delegate to UN,” NYT, February 12,
1997.
28. Al Kamen, “Many Top Jobs Remain Open in 2nd Term,” WP, April 14,
1997, Al, A10.
29. Bill McAllister, “Critical Jobs Still Unfilled by Clinton,” WP, August 29,
1997, Al, A18.
30. Margaret Jane Wyszomirski, “Advice for a New Administration: A
Review Essay,” Public Administration Review 49 (July/August 1989): 397–401.
31. John E. Yang, “Hill Democrats Borrow a Few Themes for Center Look-
ing ‘Families First’ Agenda,” WP, June 23, 1996, A19.
32. Pfiffner, The Strategic Presidency.
33. Ben W. Heineman Jr., “Some Rules of the Game: Prescriptions for Orga-
nizing the Domestic Presidency,” in The Presidency in Transition, ed. James Pfiffn-
er and R. Gordon Hoxie (New York: Center for the Study of the Presidency, 1989),
45–53.
34. Light, The President’s Agenda.
35. See Juliana Gruenwald, “Legislative Success Elusive in Second Term,”
CQWR, January 25, 1997, 234.
36. For research on the connection between party platform promises and sub-
sequent policy activity, see Gerald M. Pomper, with Susan S. Lederman, Elections
in America (New York: Longman, 1980), 161. On the platform process, see
Stephen J. Wayne, The Road to the White House 1996 (New York: St. Martin’s
Press, 1996), 161–164.
37. Jeff Fishel, Presidents and Promises (Washington, DC: CQ Press, 1994),
38–43.
38. Deborah Kalb, “Democrats, Sensing Victory, Muffle Their Differences,”
CQWR, August 31, 1996, 2467. On the unusually harmonious process of drafting
the 1996 Democratic platform, see Deborah Kalb, “Building With Broad Planks,”
CQWR, August 17, 1996, supplement, 33–35.
39. For a text of Clinton’s acceptance speech, see CQWR, August 31, 1996,
2485–89.
40. For a text of the 1996 Democratic National Platform, see CQWR, August
17, 1996, supplement, 35–52.
41. Frank Newport, Lydia K. Saad, and David W. Moore, “The 1996 Elec-
tion: Americans Stay the Course” in Where America Stands 1997, ed. Michael
Golay (New York: John Wiley and Sons, 1997), 250.
42. Scott Keeter, “Public Opinion and the Election,” in , The Election of 1996:
Reports and Interpretations, ed. Gerald M. Pomper (Chatham, NJ: Chatham
House, 1997), 108.
43. William Jefferson Clinton, “Inaugural Address,” January 20, 1997.
Reprinted in CQWR, January 25, 1997, 252–53.
44. In discussing President Clinton’s first-term emphasis on a “new begin-
ning,” Crothers argued that Clinton interpreted his 1992 election as a mandate for
change. See “Asserting Dominance: Presidential Transition from Out-Party To In-
Party, 1932–1992,” 809.
45. William Jefferson Clinton, “State of the Union Address,” February 4,
1997, in CQWR, February 8, 1997, 380–84.
46. James Bennet, “Clinton Presents ’98 Budget, and a Goal,” NYT, February
7, 1997, A1.
47. “FY 1998: Winners and Losers,” WP, February 7, 1997, A21.
48. David E. Rosenbaum, “Even With Agreement, A Goal Remains Elusive,”
NYT, February 7, 1997, A10.
49. Just as Presidents seek to establish a “macro-theme” that underlies and
interweaves the selection of discrete issues for their policy agenda, so, too, con-
gressional leaders may set goals and select priority issues that reflect what John
Bader calls a “strategic theme.” See Taking the Initiative: Leadership Agendas in
Congress and the “Contract With America” (Washington, DC: Georgetown Uni-
versity Press, 1996), 217–8 and Chapter 6.
50. See commentary of Peter Baker, “Clinton’s Words Show Bipartisanship is
Easier Preached than Practiced,” WP, January 26, 1997, A7.
51. Peter Baker and Eric Pianin, “Clinton, Hill Leaders Agree on 5 Priori-
ties,” WP, February 12, 1997, A1, A15.
52. Adam Clymer, “Clinton and Republican Leaders Agree on Five Goals,”
NYT, February 12, 1997, A1, A12.
53. Baker and Pianin, “Clinton, Hill Leaders Agree on 5 Priorities,” A15.
54. Jackie Koszczuk, “Gingrich’s Friends Turn to Foes as Frustration Builds,”
CQWR, March 22, 1997, 679–81.
55. Dan Balz and John Yang, “Republicans Set Legislative Priorities,” WP,
March 7, 1997, A10.
56. George Hager, “Clinton, GOP Congress Strike Historic Budget Agree-
ment,” CQWR, May 3, 1997, 993, 996–997, quote 993.
57. Hager, “Clinton, GOP Congress Strike Historic Budget Agreement,”
quote 997.
58. Dan Carney, “Differences Not Standing in Way of Juvenile Crime Effort,”
CQWR, April 12, 1997, 845–849, quote 845.
59. Carney, “As Deal With Clinton Unravels, House OKs Juvenile Bill,”
CQWR, May 10, 1997, 1077–1078.
60. Jeffrey L. Katz, “GOP Steps Lightly in Response to Clinton’s Proposals,”
CQWR, February 15, 1997, 426–429.
61. Bob Gravely and Eileen Simpson, “Panels Reach Accord to Rework Pro-
gram for Disabled,” CQWR, May 10, 1997, 1079–80.
62. Jeffrey L. Katz, “Fierce Debate Looms as Congress Turns to Social Secu-
rity Rescue,” CQWR, January 11, 1997, 127–132.
63. Steve Langdon, “On Medicare, Negotiators Split Over Policy, Not Just
Figures” CQWR, February 22, 1997, 488–490.
64. Alison Mitchell, “Clinton Readies a List of Kennedyesque Challenges,”
NYT, May 18, 1997, A14.
65. David R. Mayhew, Divided We Govern (New Haven: Yale University
Press, 1991), 76.
66. The White House Bulletin, April 17, 1997; and Barbara Vobjedak,
“Experts Describe New Research on Early Learning,” WP, April 18, 1997, A3.
67. James Bennet, “At Volunteerism Rally, Leaders Paint Walls and a Picture
of Need,” NYT, April 28, 1997, A1.
CHAPTER 7
The Irony of the 105th Congress

and Its Legacy
ROGER H. DAVIDSON AND COLTON C. CAMPBELL
On January 4, 1995, the 104th Congress burst “out of the blocks at a

sprint.” For the first time in forty years, Republican majorities controlled
both chambers, and the House was intent on acting within the first one
hundred days on all ten items of the widely publicized “Contract with
America.” Lawmakers spent marathon sessions voting on such things as
constitutional amendments to balance the federal budget and impose term
limits, the line-item veto, unfunded mandates, tort reform, and even inter-
nal rules—among other things, applying national workplace laws to
Capitol Hill for the first time. Partisans filled the hours with hyperbolic
language and debate over the ethics of each other’s leaders, principally
President Clinton’s fund-raising practices and admitted ethical lapses by
then–House Speaker Newt Gingrich.
Two years after the Republican revolution, the 105th Congress
(1997–1998) promised quietude by comparison, beginning with what one
observer called a “legislative lullaby.”1 Republican leaders, plagued by nar-
row majorities, were unable to legislate effectively. In the House, skirmish-
es broke out among several factions; a rebel group of two dozen conserva-
tive sophomores variously vexed, criticized, and even publicly challenged
the Speaker’s leadership. Voicing concern that their leaders had become too
centrist, too willing to capitulate to President Clinton and the Democrats
in budget negotiations, the dissidents even met with Gingrich’s four chief
deputies in an aborted coup.2 The solid Republican party organization and
disciplined ranks that characterized the early 104th Congress were replaced
in the 105th by a disorganized, disconsolate “team.”
It was this querulous Republican team that confronted President Bill
Clinton, carried to reelection in 1996 by resurgent political fortunes, only
to be mired in scandal throughout 1998. The president was able to
strengthen his leverage in policy negotiations with Capitol Hill, according
to Elizabeth Drew, by successfully “painting the Republicans as extremist
155
156 The Irony of the 105th Congress and Its Legacy
because of their proposals for budget cuts” in Medicare and school lunch-
es—programs that help “the middle class as well as the poor—and their
efforts to roll back environmental regulations.” Additionally, according to
Nicol Rae, being on the right side of two government shutdowns helped
win back public support, decisively placing Clinton in the driver’s seat for
reelection while raising doubts regarding the Republican Congress in the
minds of voters.3
The very qualities that set the 105th Congress apart—its modest policy
goals, its narrow majorities and internal strife, and especially in the House
its weakened party leaders and ascendant committee leaders—led it to the
partisan legacy for which it will always be remembered. This was the
House’s impeachment of President Clinton in December 1998. It was a bit-
ter partisan climax for a Congress that began with a limited agenda. And
it was caused, ironically, by the House’s return to “regular order”—which
is to say, deference to its committees.
THE 1996 ELECTIONS

Elections provide defining experiences for newly-elected officials, lessons
for incumbents who survive, and models for those contemplating future
races. In 1994, Republicans were guided by a nationalized platform that
many thought hastened their historic takeover of Congress. Not one
Republican incumbent was defeated, while thirty-four Democratic repre-
sentatives and two senators lost their seats. Unhappy with either the con-
gressional process or the policy it produced,4 voters two years later nearly
deprived the GOP of its newfound majority status. The parties fought to a
virtual draw; there were no landslides, no mandates to be confidently
claimed, and no compelling national messages.
The GOP’s near-death experience in 1996 left it with the smallest House
majority since 1952. After the 1996 elections, Republicans still outnum-
bered Democrats by twenty-one (228 to 207), a narrower margin than the
twenty-five Republicans had two years earlier (230 to 205).5 Although
Democrats gained seven seats in the Northeast, Republicans continued
their march on the South (with a net gain of four seats). In the Senate the
Republicans gained two seats for a margin of fifty-five to forty-five. The
impact was greater than this gain suggests, according to Drew, “because
ideologically the new Republican senators moved the Senate further to the
right.”6 But Republicans were still five votes short of the sixty required to
shut off Democratic filibusters.
By retaining “control of the House,” wrote Drew, “Republicans may
have tacked down, but not nailed down, the realignment they were seek-
ing.”7 A simple explanation of the election outcome is elusive given the
multitude of possibilities: weak Democratic candidates; voters deliberately
splitting their tickets in order to ensure a balance of power—a Republican
Congress to check a Democratic White House; or voter backlash against an
over-ambitious 104th Congress. It is, however, important to place such

results in longer-term perspective. Each successive election is in some sense
unique: new voters are added, former voters drop out, and a few even shift
their vote from one party to another. The results turn on how many voters
fall into each category.
In the 1996 elections the electorate divided itself roughly in thirds:
Democrats (37 percent), Republicans (37 percent), and independents (26
percent).8 Often voters who claim to be independents are in fact “closet
partisans” who lean toward one party or the other. Adding these voters to
the identifiers of the two parties gave the Democrats a slight edge (47 to 42
percent) in 1996, with about 11 percent true independents. These prefer-
ences, as measured by indices of party identification, have remained rela-
tively stable over the past several elections, despite short-term fluctuations.
Party identification remains the most powerful single correlate of voting
in congressional elections. In both 1994 and 1996, 77 percent of the elec-
torate were party-line voters, supporting the candidate of the party with
which they affiliated. In both of these elections, 17 percent of voters were
party defectors, party identifiers choosing a candidate of another party. In
both elections, only 6 percent of voters claimed to have no party identifi-
cation or leaning. This distribution of voters has remained fairly stable
since the l970s.9
In 1996, admittedly a very partisan year, better than nine of every ten
Democrats and Republicans voted for their parties’ candidates; indepen-
dents’ votes split fifty-fifty. Compared with the previous congressional elec-
tion, the GOP lost support from independents (six percentage points) and
from Democrats (three points); their support from those who voted
Republican was virtually unchanged.10
“All politics islocal,” former House Speaker Thomas P. O’Neill Jr. (D-
Mass.) used to say. Candidates employ personal styles and issues to appeal
to the voters of their districts and states. At the same time local contests are
affected by current national events and affairs. In 1996 these included a
robust economy and what many considered the excesses of the Republican
104th Congress. The increasing involvement of national party committees
and allied interest groups (e.g., labor unions and environmental groups for
the Democrats, small business and the religious right for the Republicans)
has imposed national coordination upon congressional campaigns, with
the result that a mix of local and national forces converge on congression-
al elections to shape their conduct and their results.11
In order to inoculate themselves against the unpopular political idiosyn-
crasies in Washington, incumbents attempt to separate themselves from the
Washington power brokers. In 1996, many Republicans deliberately dis-
tanced themselves from the 104th Congress and its record by running their
reelection campaigns as independent voices for their districts instead of
rubber-stamping then-Speaker Gingrich’s policies. Many Republicans of
the class of 1994, for instance, “localized” their records by voting inten-
tionally against their party in procedural votes, simply to inflate their over-
all scores on opposition to the leadership.12 For example, Representative
Phil English (R-Pa.), fighting a tight race for reelection, even boasted about
receiving a thank-you note from the White House for helping raise the min-
imum wage in an ad that read, “Even President Clinton thanked Phil
English for his independent action on behalf of working families.”13
Similarly, Republican Senate veteran Mitch McConnell (R-Ky.) took credit
for helping to save the school lunch program, and also claimed that he
sometimes “goes against what his party is advocating.”
Democratic challengers especially railed against the GOP’s 104th
Congress and its leader, Gingrich. In central Massachusetts’ 3rd District,
Democrat Jim McGovern coined a winning slogan in his bid to oust sopho-
more Representative Peter I. Blute (R-Mass.): “If you wouldn’t vote for
Newt, why would you ever vote for Blute?” And in New Jersey, successful
Democratic challenger Bill Pascrell Jr.’s campaign logo showed freshman
Representative Bill Martini (R-N.J.) as a puppet on strings held by the
Speaker.
Toward the end of the 1996 election campaign, the presidential contest
took “an unexpected new turn” that may have profoundly affected control
of the House. The disclosure of questionable campaign finance practices in
soft-money fundraising by the Democratic National Committee and
President Clinton’s ceaseless efforts to raise money for his reelection
stymied the Democrats’ efforts to regain the House. Ironically, noted Drew,
voters took out their anger more on Democratic congressional candidates
than on President Clinton. In the congressional elections Drew found that
“exit polls showed that the eighty-three percent of the voters who decided
whom to vote for before the last week of the election favored the
Democrats by four percent, while those who decided in the last week
favored the Republicans by fourteen percent.”14
An alternative explanation from political scientists pointed to the
Democratic candidates themselves as culprits in the party’s failure to retake
the House. The candidate recruitment season for 1996 took place in the
shadow of the 1994 elections, when Democratic prospects were at a low
ebb. Strategically-minded Democratic potential candidates could hardly be
blamed for concluding that they should bide their time and sit out their
races. As a result, the party fielded fewer “quality challengers” than usual.15
The Democrats netted an eight-seat gain, whereas eighteen seats were need-
ed for a majority.
In the natural process of membership renewal, voters normally turn rel-
atively few members out of office. Many incumbents leave Capitol Hill vol-
untarily to retire, to run for another office, or to follow other pursuits. In
1992 and 1994, a combination of voluntary retirements and electoral
defeats of incumbent members brought in 196 new representatives and
twenty-nine new senators. And the 1996 elections added seventy-four

freshman representatives and fifteen new senators.
When the 105th Congress convened, therefore, a majority of House
members had been elected in the 1990s. High electoral turnovers in three
successive elections made the chamber significantly younger and more
junior than it was when the decade began. Less familiar with the institu-
tion, the culture, the folkways, and the history of Congress, younger mem-
bers have often shunned the intraparty or bipartisan comity needed to
grease the gears of the legislative process.16
The Senate’s membership was altered to a lesser degree by recent elec-
tions. As in the House, nearly a majority of senators serving in the 105th
Congress were elected in the 1990s; though despite respectable turnovers
in 1992 and 1994, the average age and seniority remained fairly steady. A
high number of retirements in 1996, however, foretold a changing of the
guard not unlike that already witnessed in the House.
RETURN TO NORMALCY
The dramatic 104th formed the backdrop for the Congress that followed.
Like all Congresses, the 105th flowed from the most recent elections that
produced it; indeed, the 1996 elections conveyed fresh messages that influ-
enced incumbents’ behavior as well as that of freshly elected members.
Campaigns were marked by incumbent protection and, for not a few can-
didates, a veiled appeal to the status quo. Many Republican lawmakers felt
“that if they held the House for the second time in a row,” they could retain
it for “ten to twenty years.”17 Republicans could use the incumbency
advantage “to protect their position and to extend their power—just as the
Democrats had for so long.”18 Career-oriented members found that climb-
ing a traditional internal ladder—paying their dues by working through the
committee system, earning their stripes by rising through the ranks of the
formal party structure, and developing expertise with the support of a cau-
cus of members with shared interests—aids in the pursuit of personal polit-
ical goals. The 105th emphasized a more traditional model of congression-
al power, crafting most legislation in committees, with Republican leaders
forced to bargain with the Clinton administration. Even freshman
Republican members came to Capitol Hill set on working with Democrats
to find solutions, a 180-degree reversal from the aggressive reform-minded
Class of 1994, who arrived intent on transforming Congress and the nation
on their terms.19
TAMING THE GOP AGENDA FOR THE LONG HAUL

The Republicans of the 104th Congress arrived with a sense of missionary
zeal. “They came to town in January 1995 sweeping all before them, the
conquering heroes with justified sense of satisfaction that they could truly
change America,” said Senator John McCain (R-Ariz.). “Now in 1997,”

speaking of the House, “they came back, barely keeping their majority and
with a wounded leader,” McCain said. “I don’t know,” he continued, “of
a greater example of how things can change in this town.”20
The 105th Congress thus found the Republicans in a chastened mood.
With the previous Congress’s budget deadlock with President Clinton,
internal party disagreements about how to handle pressing agenda items,
the collapse of then-Speaker Gingrich’s own public support, and the pub-
lic’s discontent with partisan bickering fresh in mind, Republicans gathered
to organize the 105th Congress with far less consensus on their agenda,
more open divisions among their coalition, tarnished leaders who were
somewhat less united than they had been, committee chairs who felt more
free to stray from the wishes of the party bosses, and significant differences
in the tactics and temperaments of the members. Although, their confi-
dence in the rightness of their cause seemed undiminished.21
Many returnees from the Class of 1994 thus tempered their conser-
vatism with a desire to fashion laws through compromise and gain more
legislative influence. “I didn’t come to Washington to burn all the buildings
down,” said Representative Roger Wicker (R-Miss.) who had served as
class president in 1995.22 Divided party government, at least for the time
being, seemed to foster a rhetoric of cooperation that reflected recognition
of the need for bipartisan policy development, especially within commit-
tees.23
Dubbed the “implementation Congress” by then-Speaker Gingrich, the
GOP had a smaller majority presaging an institution seemingly more inter-
ested in solutions than partisan agendas. “Making progress on the
Republican agenda does not require us to create fireworks,” House
Majority Leader Richard Armey (R-Tex.) said. “It requires us to simply get
down to work.”24 “I think some people are going to have to accept the con-
cept of incremental change, and realize that even if you can’t get something
this year, you’re still closer the next year,” noted Representative Ernest J.
Istook Jr. (R-Okla.), who two years earlier had initiated a project with lead-
ership support to limit the amount of private money that nonprofit orga-
nizations, even such as the Red Cross and the Girl Scouts of America, could
spend on “political advocacy.”25
Two months into the new 105th Congress, House Republicans issued a
thirteen-point list of legislative priorities entitled “Creating a Better
America for Ourselves and Our Children.” The new Republican agenda
was far less specific than the 1994 manifesto and embraced no set timetable
for implementation. Gingrich was also slow to consult with interest groups
on the agenda items.26 The priorities did not take hold in the Republican
conference, mainly because they lacked details and were not vetted by the
committee chairs empowered to execute GOP legislation.
Significantly, then-Speaker Gingrich used a special order speech to deliv-

er his agenda message to Republican colleagues. Gingrich would typically
deliver such messages “in closed-door meetings of House Republicans, bol-
stered by written material prepared by the House Republican
Conference.”27 Instead, after several months of ceding the agenda-setting
role to other GOP leaders during his ethics investigation, Gingrich went to
the well from the Speaker’s dais to speak before a near-empty chamber.
This revived a tactic he had used in the early 1980s to gain prominence in
the House, when he used C-SPAN coverage of the House proceedings in his
drive to end the GOP’s long minority status.
THE TORTOISE BEATS THE HARE: CONGRESS OF “REGULAR ORDER”

The dramatic alteration in majority and minority status was a major ingre-
dient in the legislative pace of the 104th Congress. In shock over losing
control, Democrats were slow to adjust to their newfound minority status.
At the outset of the 104th Congress, Republicans were determined to pass
their strong and aggressive agenda regardless of the degree of opposition.28
House Republican leaders put extraordinary pressure on committees to
report legislation quickly. To ensure that the committees of jurisdiction
moved with speed on the Contract, Gingrich used the leverage of his sin-
gular support among the rank-and-file, naming the new chairs of all stand-
ing committees, departing in four instances from seniority to lend his sup-
port to more activist and committed conservatives. C. Lawrence Evans and
Walter J. Oleszek reported that “after one chairman informed Gingrich
that his committee was having trouble meeting the contract timetable, the
Speaker responded, ‘If you can’t do it, I will find somebody who will.’”29
Hearings and markups were often rushed.30 Party leaders, “and task forces
on which inexperienced freshmen predominated, exercised considerable
influence on the substance of legislation in committee or through post-
committee adjustments.”31 Standing committees were frequently bypassed
not only to move legislation more rapidly, but also for other substantive
reasons such as circumventing outside committee clienteles.32
One hundred days later, House Republicans had passed all but one com-
ponent of the Contract with America. But this achievement came at a high
cost. After the Contract and the first one hundred days were successfully
negotiated, some younger lawmakers hinted that they might retire from the
House because of the late nights, frenzied pace, and stress that was placed
on their families. Others worried about the capacity of members to make
intelligent decisions when they were plagued by overwork and little sleep.
Deliberation and the quality of legislation suffered. As Barbara Sinclair
related, “many Republicans, members and staff alike concede[d] privately
that the legislation brought to the floor was sloppy at best; the careful sub-
stantive work had not been done.”33 The Speaker and other party leaders
expressed sympathy about the hectic schedule, but family-friendly schedul-
ing clearly took a backseat to the need to move an ambitious legislative

agenda.
As a routine matter, of course, Congress does not work at such a frenetic
pace. Revolutions can be exhausting and difficult to sustain given the insti-
tutionalization of certain norms—seniority, apprenticeship, and reciproci-
ty—and with an institution that is sequential, approaching solutions in
small, discrete steps, building policy from the bottom up. Elaborate writ-
ten rules and practices that govern the conduct of both the House and the
Senate make it extremely difficult to consider legislation at dizzying speed.
By the start of the 105th Congress, Speaker Gingrich agreed to a more reg-
ular order of activity: “The Contract with America was a commitment of
a legislative minority to a specific set of activities in the first 100 days. This
is the commitment of a legislative majority to two years of work. We’re
going to be pleasant. . . . We’re going to have to learn to get things done.”34
Lawmakers, especially in the House, shrank from a revolutionary agen-
da and bent to political realities. “Last Congress was a speeding train,”
asserted a Republican leadership aide. “This Congress is more of a freight
train.”35 Another GOP aide added, “We are still struggling to learn how to
be a majority, and we have not figured out how the Democrats use the car-
rot and stick successfully. Until there is change, we are in what is, quote,
the status quo.”36
The 105th was more friendly to members’ personal needs. Committees
were slow to organize themselves into subcommittees, set their agendas,
and even outline broad themes. Both chambers hardly met in January,
recessing all but four days. February was not much different; lawmakers
were in session just eleven days. Tuesdays were rescheduled for the order
of the day, and vote stacking was common. In the first six weeks of 1997,
the House was in session just forty-nine hours and the Senate just ninety-
six hours, compared to the breakneck 104th Congress, in which 183
recorded votes were held between Opening Day and March 1. From
January to March of 1997 there were sixty roll-call votes in the House,
eleven of those being on a single subject, term limits, compared to 271 in
1995. The number of bills passed dropped by 25 percent from four years
earlier—just sixty-eight bills were passed within the first three months of
the 105th Congress, compared to 111 in the 104th Congress—while the
days in session declined slightly.37 Workload comparisons of the first ses-
sions of the two Congresses are presented in table 7.1.
The moderate start produced partisan and factional snipping that the
105th Congress was doing nothing. Democrats had an obvious explana-
tion. “This is not a mechanical problem,” declared Representative Barney
Frank (D-Mass). “They have nothing to do.”38 Grumbling was heard, espe-
cially from some of the hard-core GOP members. “If I had to swear under
oath, I would say they don’t have a strategy,” Representative Jay
Table 7.1: Congressional Workload Comparisons, 1995 and 1997

104th Congress, 1st session 105th Congress, 1st session
House Senate Total House Senate Total
Days in session
167 211 378 132 153 285
Hours in session
1,525 1,839 3,364 1,003 1,093 2,096
Measures introduced
3,430 1,801 5,231 3,728 1,840 5,568
Bills
2,840 1,514 4,354 3,088 1,568 4,656
Joint resolutions
137 45 182 106 39 145
Concurrent resolutions
130 36 166 200 70 270
Simple resolutions
324 206 530 334 163 497
Measures reported
400 249 649 373 248 621
Yea/Nay votes
299 613 912 285 298 583
Total measures passed
483 346 829 544 385 929
Public bills enacted
60 28 88 93 49 142
SOURCE: Data are found in Congressional Record (104th Cong., 1st. sess.), Jan. 3, 1996,
D1535, and Congressional Record (105th Cong., 2nd sess.), Dec. 15, 1997, D1281.
Dickey (R-Ark.) complained of the new agenda. “I would say that the lead-
ership was not being as good in power as it was getting into power.”39
“Even some senior Republicans,” according to Richard Cohen, complained
that the Speaker did not send clear enough “signals about his plans. ‘There
is a perception, which I share, that the leaders have not always shot straight
with us,’ said Representative Joe L. Barton (R-Texas). ‘They have had to
resort to sleight of hand to bring some bills to the floor.’”40
Then-Speaker Gingrich conceded that Republicans had problems.
“Managing the process of change is much more difficult than maintaining
the status quo,” he said. “It is not altogether surprising that we have made
mistakes, experienced conflict or doubts, and undergone moments of
uncertainty. It is not even surprising that we have, at times, forgotten that
we are a team and lost sight of the fact that we all have the same goal.”41
In fact, the 105th Congress was not necessarily less productive than its
predecessor, at least if one compares the first sessions (see table 7.1). To be
sure, the pace was more leisurely: the two chambers spent about a third less
time in session in 1997 than they had two years earlier. And there were
fewer floor votes. But more measures were passed in the two chambers,
and more measures enacted into public law, than during the 104th
Congress’s frenetic first session.
NARROW PARTY MARGINS IN BOTH CHAMBERS

In closely divided chambers, any partisan defections can profoundly affect
the outcome of issues. Those outside the parties’ mainstream can have great
influence. This especially applies to those in the center of the political spec-
trum: conservative Democrats and moderate to liberal Republicans.
Partisan strength in voting is entrenched in what some argue is merely con-
stituency differences; that is, partisans vote together because they represent
like political and demographic areas. According to this reasoning, legisla-
tors do not follow party lines whenever they feel their party’s policies will
not benefit their constituents.
Today’s Democratic mavericks tend to be from nonminority southern
districts, whose voters typically are to the right of the party’s mainstream.
(Six of them—five representatives and one senator—decamped to the GOP
after the 1994 elections.) To hold voter loyalty in these areas, the remain-
ing Democratic officeholders must avoid their party’s mainstream course.
Fifteen of the thirty-three Democrats who voted with their party less than
70 percent of the time in 1997 were southerners.42
Republican mavericks, in contrast, are mostly from New England or the
Northeast corridor, where voters generally fall to the left of the party’s cen-
ter. In 1997, all six Republicans who gave 70 percent or less support to
their party were northeasterners.
Congress is an institution in which the majority leads. When that major-
ity is sizable, leaders can exercise discretion over what transpires on the
House and Senate floors. When the majority is slim, as in the 105th, lead-
ers must work harder to construct winning coalitions. “Majorities are built
in Congress, not elected to it,” wrote John Manley.43
Constructing winning coalitions was not easy for the 105th Congress.
Divisions and rifts in the GOP between moderates and conservatives were
legislative hurdles—growing pains that normally come with majority sta-
tus.44 Often such divisions lead to the emergence of significant and poten-
tially cohesive voting factions based on ideology, region, or particular pol-
icy positions. Gingrich and his lieutenants were forced to woo dissident
factions in their own flock as well as seek defectors from the opposition in
order to negotiate policy matters with the White House.
When political scientists began to do serious roll-call voting analysis in
the 1950s, it was obvious that party and ideology were quite distinct ingre-
dients, despite measurable differences between the parties. The major rea-
son was constituency splits within the two parties.45 In the late 1930s a con-
servative coalition of Republicans and southern Democrats emerged in a
reaction against the New Deal.46 Historically, this coalition was stronger in
the Senate than in the House, but its success rate in both chambers was for-
midable during the 1938–1964 period—no matter which party controlled
the White House or Capitol Hill. The coalition reappeared in 1981 as
President Reagan captured votes of conservative Democrats on budget, tax,
and social issues, while pulling mainstream Democrats toward more con-
servative positions on other issues.47
The sorting out of the two parties’ constituency differences helps to
explain a twenty-five year upsurge in lawmakers’ party loyalty. Rising
numbers of the Democrats’ southern flank are African–Americans. The
dwindling numbers of Democrats representing conservative districts
(including the Blue Dog Democrats) try to put distance between themselves
and their leaders. By the same token, the Republican congressional party is
more uniformly conservative than it used to be. In the South the most con-
servative areas now tend to elect Republicans, not Democrats. Elsewhere
Democrats have captured many areas once represented by GOP liberals.
The decline of archconservative Democrats and liberal Republicans, espe-
cially in the House, underlies much of the ideological cohesion within, and
the chasm between, contemporary Capitol Hill parties.48
Partisan repositioning has shrunk the ideological center in the two
chambers. “Democrats are perched on the left, Republicans on the right, in
both the House and the Senate as ideological centers of the two parties
have moved markedly apart,” wrote Sarah A. Binder.49 In other words, the
two parties are more cohesive internally, and farther apart externally, than
they were in the recent past; life on Capitol Hill has become commonly
acrimonious. The proportion of centrists—conservative Democrats and
moderate Republicans—hovered around thirty percent in the 1960s and
1970s, according to Binder.50 In the mid-1990s, only about one in ten law-
makers fell into this centrist category. The self-named Blue Dogs and the
New Democrat Coalition, each claiming twenty-five to thirty-five mem-
bers, emerged as a powerful voting bloc during the 104th Congress and
had the potential to be a powerful voice for compromise in a narrowly
divided 105th Congress. So did the moderate Republican groups.
With the 105th Congress’s narrow partisan margin, almost any cohesive
group was in fact able to wreak havoc in the GOP ranks. Such was the
effect of the abortive attempt to overthrow Speaker Gingrich in the sum-
mer of 1997. The core dissidents were eleven or so ultra-conservative mem-
bers, most from the class of 1994, who accused their leaders of abandon-
ing the revolutionary goals typified by the Contract.
The Family Caucus and Conservative Action Team (CAT), successors to
the Republican Study Group, were two conservative alliances that experi-
enced a surge in their ranks and in their clout during the 105th Congress.
Both groups promote a socially conservative, anti-tax agenda. Although
previously excluded from the leadership’s weekly strategy sessions, in the
105th they not only had a seat at that table but also sent emissaries to
meetings of the moderate Tuesday Group.51 Comprised of roughly seventy
members from all regions of the country, up from fifty in the 104th
Congress, CAT took an active role in formulating the Republican agenda,
and it enhanced its organizational infrastructure to meet that goal. The
group hired a permanent staffer to coordinate its activities and to recruit

new members, and even encouraged the creation of a Senate counterpart
(SCAT).
Moderate to liberal Republicans, such as those in the Tuesday Group
(formerly known as the Tuesday Lunch Bunch), had but forty-five to fifty
senior moderate to liberal members, mainly from northeastern, marginal
districts, and the newer Main Street Coalition, boasted only thirty mem-
bers.52 “The more decidedly conservative bent of the recent Republican
majority initially made observers dismiss” the influence of these moderate
to liberal Republicans, noted Robin Kolodny.53
Their influence enhanced by the narrow GOP majority, these centrist
factions were significant to winning coalitions. As Kolodny argued, “liber-
al Republicans exercised considerable negative influence on the congres-
sional agenda championed by the majority of the Republican conference.”
Tuesday Group members often “removed or blocked controversial propos-
als” from the common party agenda by pressuring the leadership (by
threatening to oppose procedural motions or making speeches on the floor
against the party’s position), offering motions or amendments on the floor
to counteract objectionable measures, or actually defecting from the party’s
position on several amendment votes.54
FEW STRUCTURAL OR PROCEDURAL CHANGES

The House of Representatives adopted for the 104th Congress an extensive
set of rules changes which included, among other things, revisions in the
House committee system, term limits for House leaders, alterations in floor
procedure, and a series of administrative changes designed to cut costs and
depoliticize House operations. These changes were the most dramatic alter-
ation in the House procedural environment in two decades. They were
intended to make the committee system more accountable to the
Republican Conference through the party leaders.55 On the Senate side of
the Capitol, power also was shifting, but with greater emphasis on conti-
nuity. Republican senators initially focused on a single unsuccessful reor-
ganization proposal aimed at substantially altering the filibuster rule, but
adopted a list of changes from the Mack Task Force later in the session.
THE PEOPLE’S HOUSE: CONTINUED PARTISANSHIP

The extreme centralization of the 104th Congress was a high water mark,
but was in no way inconsistent with long-term trends. By the time the
104th Congress adjourned, it was apparent that House leaders, having
overreached their powers, needed to placate committee leaders and inter-
nal factions, not to mention the Senate and the White House. As the 105th
Congress convened, many hard-core conservatives, including the firebrand
sophomores from the Class of 1994, had lowered their sights somewhat;
the badly damaged Speaker was striving to salvage some of his lost repu-
tation and power. “As more significant policy divisions emerged among
House Republicans in 1996,” noted Evans and Oleszek, “committee chairs
would retain substantial control over their own agendas.”56 At the end of
the 1997 session, Representative David M. McIntosh (R-Ind.) noted of
then-Speaker Gingrich, “People will no longer automatically follow him
blindly.”57 The Senate leadership, especially new Majority Leader Trent
Lott, pursued the classic course of seeking consensus among the
Republican Conference members.
Republicans offered only a modest package of rules changes in the 105th
Congress. The Republican Conference avoided reforming committees, opt-
ing instead for minor procedural streamlining. Committee and subcommit-
tee seats in the 105th Congress are listed in table 7.2. One change was a
requirement that committees accept testimony from witnesses only under
oath. Dubbed “truth-in-testifying,” the new House rule expanded to all
committees the practice followed by the House Government Reform and
Oversight Committee of swearing in anyone who testified at a hearing.
Aimed at nonprofit as well as profit organizations, this requirement direct-
ed witnesses to disclose the amount and source of federal grants or con-
tracts they or their organizations had received over the previous three years
(House Rule XI, clause 2 (g)).
The rules package contained other changes: a requirement that commit-
tee reports include a statement citing the constitutional authority for legis-
lation; that all committee documents be available in electronic form, as far
as feasible; stricter rules on limitation amendments to prevent spending
cuts contingent on open-ended requests for information; and a repeal of the
prohibition on committees sitting while the House is in session without
first getting special leave from the House. Members cannot hand out cam-
paign donations on the floor of the House, in the cloakrooms, or in the
Speaker’s Lobby. The GOP-inspired rule requiring three-fifths majority for
House approval to increase income tax rates was amended to two-thirds
majority. Finally, members and their staffs were required to submit to ran-
dom drug tests.
Perhaps more noteworthy were the committee and procedural changes
that did not happen. The makeup of the Ethics Committee was much
debated, and although the membership of the committee continued to
rotate, it was difficult to retain members on it (membership did remain in
place in 1996–1997 for the Gingrich ethics case). At issue was whether or
not the Committee should delegate the investigatory phase of its work to
an outside body such as a temporary, independent commission. For the
time being, the committee itself continued to shoulder this burden.
Table 7.2: Committee and Subcommittee Seats: 105th Congress (1997–1998)
SENATE HOUSE OF REPRESENTATIVES
Committee (year Members Party Sub- Total Seats Committee (year created) Members Party Sub- Total Seats
created) Ratio Units Ratio Units
Ag., Nutrition & 18 (10-8) 4 50 Agriculture (1829) 50 (27-23) 5 149
For. (1825)
Appropriations 28 (15-13) 13 151 Appropriations (1865) 60 (34-26) 13 202
(1867)
Armed Services 18 (10-8) 6 66 National Security (1822) 55 (30-25) 7 187
(1816)
Banking, Housing 18 (10-8) 5 66 Banking/Financial (1865) 53 (29-24) 5 155
& Urban Affairs
(1913)
Budget (1975) 22 (12-10) -- 22 Budget (1975) 43 (24-19) -- 43
Commerce, Sci. 20 (11-9) 7 99 Commerce (1795) 51 (28-23) 5 174
& Trans. (1816)
Energy & Nat. 20 (11-9) 4 62 Science (1958) 46 (25-21) 4 125
Resources (1816)
Enviorn. & 18 (10-8) 4 54 Resources (1805) 50 (27-23) 5 139
Public Works
(1833)
Finance (1816) 20 (11-9) 5 80 Transportation (1837) 73 (40-33) 6 212
Foreign Relations 18 (10-8) 7 71 Ways & Means (1802) 39 (23-16) 5 105
(1816)
Governmental 16 (9-7) 3 47 International Relations 47 (26-21) 5 125
Affairs (1921) (1822)
Indian Affairs 14 (8-6) -- 14 Govt. Reform/Oversight 43 (24-19) 7 124
(1977) (1927)
Judiciary (1816) 18 (10-8) 6 63 Judiciary (1813) 35 (20-15) 5 99
Labor & Hum. 18 (10-8) 4 54 Education & Work 45 (25-20) 5 124
Resources (1869) Force (1867)
Rules & 16 (9-7) -- 16 Rules (1880) 13 (9-4) 2 27
Administration
(1947)
Small Business 18 (10-8) -- 18 House Oversight (1789) 25 (5-3) -- 8
(1980)
Veterans Affairs 12 (7-5) -- 12 Small Business (1975) 35 (19-16) 4 79
(1970)
Special Aging 18 (10-8) -- 18 Veterans Affairs (1825) 29 (16-13) 3 59
(1977)
Select Ethics 6 (3-3) -- 6 Standards of Conduct 10 (5-5) ** 10
(1977) (1967)
Select Intelligence 17 (10-9) -- 19 Select Intelligence (1977) 19 (9-7) 2 32
(1977)
TOTALS 355 196- 68 988 TOTALS 801 445-356 92 2178

159
MEMBERS’ 3.5 52%- 6.3 9.9 MEMBERS’ AVG. 1.8 56%- 3.1 5.0
AVG. 48% 44%
104th 333 68 1090 104th CONGRESS 789 87 2103
CONGRESS
JOINT COMMITEES
Committee( year Members Party Sub Total * Independent Member of House counted as Democrat.
created) Ratios Units Seats ** Committee divides into fact-finding and adjudicatory sections
Economic (1946) 20 (12-8) -- 20 in dealing with specific cases of alleged wrongdoing.
Library (H/A) 10 (6-4) -- 10 Source: Complied by authors.
Printing (1846) 10 (6-4) -- 10
Taxation (1926) 10 (6-4) -- 10
Additionally, the GOP Conference rejected conservative proposals that

would have limited the latitude of committee chairs. A proposal pushed by
Rep. John Doolittle (R-Calif.) and his fellow CAT members would have
given the full Conference membership the authority to review the appoint-
ment of subcommittee chairs. (These posts are now selected by full com-
mittee chairs and are generally based on seniority.) Also dropped was a
proposal to strip chairs of the right to hire and fire subcommittee staff.
Finally, the Republican leadership abandoned, probably for good, the
more sweeping proposals for overhaul of committees that had been
bypassed two years earlier, but were still pushed by some reformers. On
November 6, 1996, a GOP reform task force, led by Representative David
Dreier (R-Calif.), offered several large-scale jurisdictional and procedural
reform plans for the House committee system. Dreier’s wide-ranging Task
Force Report recommended the elimination of House panels whose mem-
bership fell below 50 percent compared to their levels at the end of the
103rd Congress, indicating a drop in desirability. Another proposal was to
reorganize the House into eight committees under the headings of budget,
economic affairs, defense and international relations, government opera-
tions and judiciary, human resources, natural resources, physical resources,
and rules. Such a new strategy, the task force conceded, “would signifi-
cantly disrupt the institutional culture and environment of the House, and
it would take several years to assimilate to the resulting changes.”58
THE SENATE: AN ISLAND OF RELATIVE TRANQUILITY

“Hewing to tradition and its prized status as the more restrained chamber,”
wrote Michael Wines, the “Senate began its 105th incarnation” with leg-
islative ritual, the exchange “of elaborate courtesies,” and “the swearing-
in of thirty-four new and reelected members.” Referring to the chamber as
“an island of tranquillity,” Senate Majority Leader Trent Lott (R-Miss.)
laid out goals for creating a more genteel Senate, which included requiring
senators to show up for votes on time, reducing inconvenience for those
who are prompt, and preventing staff members on the floor from talking
too much.59
Elected Majority Leader in June 1996, Lott was younger, more ener-
getic, and more ideologically conservative than his predecessor, Bob Dole
(R-Kans.). Lott instituted a form of leadership similar to his days as Senate
Republican whip: congressional party leadership “as a team enterprise.”60
He relied on sharing duties with other elected leaders, attempting to pro-
vide more structure but also more openness to leadership endeavors, and
striving to include as many senators as possible in party-based activities.
Another noteworthy addition to GOP Conference rules was a six-year
term limit for chairs and for ranking members (for a maximum of twelve
years). This would have only a delayed impact because current committee
chairs are able to keep their posts until 2003. Another new rule stated that
Republicans who chair a committee could not simultaneously chair anoth-

er committee or subcommittee, with chairpersons of Appropriations sub-
committees exempted. This opened up about ten subcommittee slots to
junior senators; all eleven Republicans elected in 1994 and five of the eight
senators elected in 1996 became chairpersons in the 105th Congress.
A ten-member Reorganization Task Force was headed by Senator
Robert F. Bennett (R-Utah). He declared that the filibuster and cloture
would be off limits. “We need to separate tradition and continuity from
anachronism,” said Bennett.61 The topics explored included biennial bud-
geting and appropriations cycles, “holds,” electronic voting, broadening of
committee jurisdictions, the use of task forces, financial management, and
scheduling. Some of these items—especially altering senators’ ability to
delay or halt a bill’s progress to the floor by placing a “hold” on it—could
potentially affect the chamber in a major way. But the task force eschewed
formal hearings or research reports in favor of informal consultations with
colleagues.
LEADERSHIP VS. COMMITTEES

Decision-making processes in Congress represent an interaction of com-
mittees, parties, and the parent chambers, and the relative importance of
each in the policy-making process varies over time.62 The independent influ-
ence of committees varies inversely with the strength and influence of polit-
ical parties.63 Committee members must weigh opposing expectations of
their parent chamber, their party, and their constituents; generally, the
stronger and more disciplined the party caucuses in Congress, the less pow-
erful the role of committees. For example, if the parties’ electoral fortunes
are tied to issues and decisions, party leaders often assume decision-mak-
ing responsibilities that otherwise would fall to committees. Conversely, the
less cohesive the party caucuses, the more powerful the role of committees.
When attitudes on issues are highly fragmented within the party caucus,
policy outcomes bear little on party fortunes, and autonomous committees
are tolerated and even encouraged.
Practices adopted by the new GOP majority in 1995 continued a long-
term weakening of the position of committee chairs. This transformation
explicitly reversed the dominant 1950s portrait of autonomous and pow-
erful committees in Congress—encouraged by weak party cohesion—that
shaped policies within their jurisdictions in ways that satisfied the
parochial interests of committee members.64 Once in power, House
Republicans devised procedural reforms that counterbalanced, and in some
cases contradicted, the earlier norm of deference to committees. By giving
priority to their party’s agenda, committee members are seen to work
toward enhancing the party’s reputation and to achieving its policy goals.65
This party-dominated model is based on the premise that all legislators
seek electoral success for their party, which in turn depends partially upon
the congressional party’s legislative record. The caucuses pressure commit-

tee members to behave as agents or instruments of the party. Because they
control committee assignments, the congressional parties are able to frame
the organization and policy outlook of their committee contingents.
Additionally, committees rely on party leaders to schedule their legislation
for floor consideration, giving party leaders additional leverage with the
committees. These ties create an implicit contract between the parties and
their committee members that constrains committee behavior; as a result,
the majority party has effective control over committee decisions.66 And
because committee recommendations are constrained by the views of the
majority party, committees lack autonomy.67
In the 105th Congress, however, committee chairs recaptured important
ground in the tug and pull between leadership and committees.68 To pre-
serve his leadership, then-Speaker Gingrich agreed to changes that expand-
ed committee chairs’ power, restoring somewhat the former Democratic
majority’s model of strong, independent chairpersons and committees coor-
dinated, but not coerced, by party leaders. As a start, he disbanded the
Speaker’s Advisory Group, an inner cabinet comprised of senior leaders
and trusted colleagues, which had became the locus of long-range legisla-
tive strategy and agenda planning in the last Congress. At the same time,
the larger, more representative GOP leadership group was expanded to
include the committee chairs of Budget, Appropriations, Ways and Means,
and Rules. As a result of retirements, new full committee chairs were
named to the Government Reform and Oversight Committee, the
Agriculture Committee, and the Science Committee. These new panel lead-
ers, along with other chairs, were assured of their freedom to develop leg-
islation largely without interference from the leadership of the kind that
saw the major pieces of legislation in the 104th (such as Medicare, telecom-
munication, and appropriations bills) written in the Speaker’s “Dinosaur
Room” in the Capitol building. Committee chairs, said one congressional
aide, “can run free as long as they’re on a leash. The difference from the
last Congress is that they had a choke collar on.”69
Acting independently, committee members are freer to respond to the
demands of their home constituents or to help important clientele groups
achieve their policy aims.70 More autonomous committees can better con-
trol their agendas in the issue areas within their jurisdiction.71 Their poli-
cies are products of the preferences of committee members and their con-
stituencies, rather than something delegated by the floor, the parties, or
party leaders.72
POLICY CONSEQUENCES
Since the Republican capture of Congress in 1995, warfare between
Capitol Hill and the White House has been a constant feature of politics in
the nation’s capital. Although the 105th Congress offered the hope of
decreased conflict, it will forever be remembered for its fiery partisan cli-
max: the impeachment of President Clinton in December 1998.
By the 1996 elections, there was some reason to believe that the
Democratic president and the Republican Congress might move toward a
working, if not a peaceful, coexistence. For his part, Clinton had moved to
the political center, in the process co-opting Republican themes and issues,
and neutralizing the GOP’s earlier advantages on such issues as crime, wel-
fare reform, minimum wage, and health insurance. Congressional
Republicans, for their part, gained some of their objectives but only by
making substantial concessions to the White House. Republicans gained
political ground by, for example, forcing the president to veto a ban on
“partial birth abortions,” and to sign welfare reform and the Defense of
Marriage Act (DOMA) that prohibited federal recognition of gay unions.
Such compromises cooled the unyielding revolutionary spirit that had
marked most of the 1995 session. At the same time it enabled Clinton to
recover the initiative and, with the aid of Senate Democrats, to frame the
legislative agenda. By compromising with Clinton, Nicol Rae has argued,
Republicans in Congress were “vindicating the president’s claims to repre-
sent the center ground in American politics and depriving their presidential
nominee [Bob Dole] of valuable ammunition to use against the incum-
bent.”73
Republican leadership in the meantime had retreated from the tri-
umphalism of the 1994 takeover. In the Senate, the new majority leader,
Trent Lott, was moving cautiously. But in the House the leadership was in
open disarray. Speaker Gingrich’s ethical problems, and especially his low
public approval ratings, reduced his leverage with the rank and file mem-
bership. Moreover, the compromises reached with Clinton produced a
backlash from those members who formed the core of Gingrich’s support:
the sizable corps of junior legislators prided themselves as outsiders and
made clear their disdain for traditional models of congressional processes.
Speaking before fellow Appropriations Committee members,
Representative Mark W. Neumann (R-Wisc.) summarized the young turks’
position by saying the Committee should “place a higher priority on prin-
ciple than on passing measures to operate the government.”74
The outbreak of the scandal involving the president’s relationship with
Monica S. Lewinsky raised White House–Capitol Hill tensions to an entire-
ly new level of intensity. During the 105th Congress, the focus was almost
exclusively on the House. Would it launch a congressional inquiry into the
affair? Would articles of impeachment be the result? And would the House
act to impeach the president?
When independent counsel Kenneth W. Starr submitted his report on

September 9, 1998, the House began a course of action that led eventually
to approval of two articles of impeachment. Ironically enough, the very
adjustments that marked the 105th Congress’s return to “regular order”
served to raise the level of partisan warfare and, with it, the likelihood of
impeachment.
The most fateful decision, as it turned out, was the adoption of H. Res.
581 which referred the whole matter to the Committee on the Judiciary.75
As it became more apparent that the House would be asked to take
action—but before the Starr report surfaced—Speaker Gingrich floated a
proposal for a select panel to conduct the investigation. But his trial bal-
loon was quickly punctured by Judiciary Chair Henry J. Hyde (R-Ill.),
backed by his committee members and other committee chairs. In view of
the leadership’s earlier concessions to committees and their chairs, the
Gingrich suggestion had virtually no chance of acceptance. Although select
committees had been employed in early impeachment inquiries, Judiciary
had processed recent impeachment cases—including the 1974 inquiry of
President Richard Nixon and the Watergate affair. Hyde and his colleagues
thus had jurisdictional precedent on their side. Moreover, Hyde himself
was a figure accorded respect from both sides of the aisle (grudging from
Democrats, to be sure). Any deviation in this high-profile case would be
interpreted as a slight to a noted leader as well as a threat to committee pre-
rogatives.
Referral of the inflammatory Starr report to the Judiciary Committee,
however, was rather like handing matches to a pyromaniac. The commit-
tee had long enjoyed a reputation as the most partisan panel on Capitol
Hill. The reasons are perfectly clear, and are rooted in the parties’ recruit-
ment patterns. Republicans had staffed the committee with staunch con-
servatives committed to conservative causes within the committee’s juris-
diction: tough crime laws, permissive gun regulations, and constitutional
amendments on such subjects as abortion, school prayer, flag desecration,
and balanced budgets. Liberal Democrats gravitated to the committee in
the 1960s to fight for civil rights legislation; more recently, they have been
recruited to combat the above-mentioned conservative initiatives.
If the House was partisan, therefore, Judiciary was superpartisan.
Committee Democrats in the 105th Congress were on average ten points
more loyal to their party than the average for all Democrats (using CQ
Weekly’s party unity scores). Republicans on the committee were on aver-
age more than five points more loyal than their House colleagues as a
whole. By the same token, Democrats gave more support to President
Clinton and Republicans less support than their party colleagues outside
the committee (using CQ Weekly’s presidential support scores). As a result,
Judiciary was virtually a caricature of Sarah Binder’s portrait of the “van-
ishing middle” of the ideological spectrum. It is no exaggeration to say that
there were no moderates on the committee, at least as measured by the

most commonly used voting indices. There were, for example, none of the
party mavericks that so enlivened the Watergate proceedings a quarter cen-
tury earlier.
The partisan rancor manifested itself in ways both blatant and subtle.
Chairman Hyde’s choice of David P. Schippers as head counsel for the
inquiry, an old friend who claimed to have been a Democrat, was so bit-
terly anti-Clinton that he helped the minority Democrats circle the wagons
in the president’s defense. The decision to release Starr’s argumentative
report immediately—before members had a chance to read, much less con-
sider, the document—encouraged members to stake out their positions
before, rather than after, any deliberation or debate had taken place. The
raucous demeanor of many members, Republicans and Democrats alike,
during the televised hearings merely reinforced the growing conviction of
members and the general public that the whole process was tainted by par-
tisanship.
The special House session on impeachment took place, of course, in the
wake of the 1998 elections. Republican leadership was in visible disarray.
Chastened by the loss of five House seats (after a large gain had been
expected), Speaker Gingrich had abruptly abdicated and was exerting no
leadership on the impeachment question. His heir-apparent,
Appropriations Chair Bill Livingston (R-La.), gave every indication of
wanting the issue resolved at the earliest possible moment. Meanwhile,
Majority Whip Tom DeLay (R-Tex.), arguably the most effective member
of the leadership group, was whipping rank-and-file Republicans to sup-
port impeachment. On the other side of the aisle, Minority Leader Richard
A. Gephardt (D-Mo.) was carefully building support for the president
among Democrats. He was aided not only by the Republicans’ ill-concealed
zeal, but by the continuing public support for keeping the president in
office.
In the end, the votes on the four articles of impeachment—the last
actions taken by the 105th Congress—were mainly along partisan lines. Of
the 1,740 votes cast by House members on the four articles of impeach-
ment—two of which were adopted—a total of 92 percent followed parti-
san lines; Republicans supporting the president’s impeachment, Democrats
resisting it.76 It was a partisan conclusion to a Congress than had initially
promised a more moderate demeanor.
CONCLUSION
Unlike the partisan intensity that characterized the 104th, party cohesion
in the 105th was less pronounced and the inter-party warfare less visible,
visceral, and confrontational. Such a fluid environment offered greater
opportunities for members to be creative and entrepreneurial, but also
posed greater risk that members’ gambits would fail for their misjudg-
ments. The long-term trends—mainly strengthened party leaders with the

capacity to influence or bypass committees—seem likely to survive. As
Barbara Sinclair has pointed out, the trend toward leadership management
of legislation predated the Republican takeover and is likely to survive tem-
porary leadership setbacks.77
Much of the “unorthodox” behavior of the Republican majority in
1995 was reversed or modified in 1996, as the full impact of the American
governmental system and the two-year House election cycle became appar-
ent. These constraints on congressional change, according to Rae, are a
perfect late-twentieth-century illustration of the precepts set out by James
Madison’s celebrated Federalist essays Nos. 10 and 51.78 A new popular
majority was elected to control the branch of government closest to the
people, the House of Representatives, reflecting widespread popular
demands for change and reform. To a large extent the new majority was
able to fulfill its mandate by confirming the shift in the direction of
American public policy toward economic and social conservatism, and by
implementing significant institutional reforms of Congress. As the
Republican program of the 104th Congress began to encounter significant
popular resistance the constitutional checks provided by the Senate and the
presidency arrested the Republicans’ “revolution.”79 Further, as
Republicans began to approach reelection in 1996, they found themselves
compelled to fall back on the same incumbency protection tactics that char-
acterized congressional elections in the last two decades of Democratic
rule—accommodation, compromise, and attention to local concerns,
rather than intense agenda change and implementation.
The impeachment debates and votes in late 1998 tended to obscure this
readjustment. They highlighted the most partisan aspects of the contempo-
rary Congress, and diverted attention from the moderating influences.
Partisanship remains potent in Washington policy-making, even when both
Republicans and Democrats strive to move toward the ideological mean.
NOTES
1. Quotations from Guy Gugliotta, “In the Cradle of Republican
Revolution, a Legislative Lullaby,” WP, March 1, 1997, A4.
2. Juliet Eilperin and Jim Vande Hei, “Some Wounds Never Heal: Today’s
GOP Leadership Has Roots in ‘Guerilla’ Warriors of the 1980s,” RC, October 2,
1997, 1.
3. Elizabeth Drew, Whatever It Takes: The Real Struggle for Political Power
in America (New York: Viking Penguin, 1997), 44; and Nicol C. Rae, Conservative
Reformers: The Freshmen Republicans and the Lessons of the 104th Congress
(Armonk, NY: M.E. Sharpe, 1998), 169.
4. John R. Hibbing and Elizabeth Theiss-Morse, Congress as Public Enemy:
Public Attitudes Toward American Political Institutions (New York: Cambridge
University Press, 1995).
5. The figures include Rep. Bernard Sanders (I-Vt.), who commonly votes
with the Democrats.
6. Drew, Whatever It Takes, 237.
7. Drew, Whatever It Takes, 238; also see 240, 47.
8. Regina Dougherty, Everett C. Ladd, David Wilber, and Lynn
Zayachkinsky, eds., America at the Polls 1996 (Storrs, Conn.: Roper Center for
Public Opinion Research, 1997).
9. Norman J. Ornstein, Thomas E. Mann, and Michael J. Malbin, Vital
Statistics on Congress, 1997–1998 (Washington, DC: CQ Press, 1998), 74–75.
10. Dougherty et al., America at the Polls.
11. Thomas Edsall, “Issue Coalitions Take on Political Party Functions,” WP,
August 8, 1996, A1; and Ruth Marcus, “Outside Groups Pushing Election Laws
Into Irrelevance,” WP, August 8, 1996, A9.
12. Rae, Conservative Reformers, 207.
13. Howard Kurtz, “On the Defensive, Republicans Go Their Own Way in
Ads,” WP, October 3, 1996, A10.
14. Drew, Whatever It Takes, 196, 241.
15. Gary C. Jacobson, “The 105th Congress: Unprecedented and
Unsurprising,” in The Election of 1996, ed. Michael Nelson (Washington, DC: CQ
Press, 1997), 143–166, at 154.
16. Craig Winneker, “Forty Years of Conflict,” RC, May 15, 1995.
17. Quotations, Drew, Whatever It Takes, 4–5.
18. David R. Mayhew, Congress: The Electoral Connection (New Haven:
Yale University Press, 1974); and Morris P. Fiorina, Congress: Keystone of the
Washington Establishment (New Haven: Yale University Press, 1977).
19. Carroll J. Doherty and Jeffrey L. Katz, “Firebrand GOP Class of ‘94
Warms to Life on the Inside,” CQWR, January 24, 1998, 155–163.
20. Dan Balz, “Subdued GOP Resumes Lead with Eye to Past; Party Retained
Hold on Power, Lost Consensus on Agenda,” WP, January 8, 1997, A9.
21. Rae, Conservative Reformers.
22. Quoted in Doherty and Katz, 159.
23. Christopher J. Deering and Steven S. Smith, Committees in Congress, 3rd

ed. (Washington, DC: CQ Press, 1997).
24. Quotations from Laurie Kellman, “Armey Takes a Larger Role,”
Washington Times, February 28, 1997, A1, A13.
25. The term, according to the measure, included “carrying on propaganda,
or otherwise attempting to influence legislation or agency action, including, but not
limited to, monetary or in-kind contributions, preparation and planning activities,
research and other background work, endorsements, publicity, coordination with
such activities of others, and similar activities.” Congressional Record, August 3,
1995, H8385.
26. Laurie Kellman, “Gingrich’s New Agenda Excites Few Republicans,”
Washington Times, May 18, 1997, A1.
27. John E. Yang, “Gingrich Revisits Backbench Strategy: Speaker Makes Use
of C-SPAN, Empty Chamber to Deliver Message,” WP, March 18, 1997, A10.
28. James G. Gimpel, Fulfilling the Contract: The First 100 Days (Boston:
Allyn and Bacon, 1996).
29. Quoted in C. Lawrence Evans and Walter J. Oleszek, Congress Under
Fire: Reform Politics and the Republican Majority (Boston: Houghton Mifflin
Company, 1997), 120.
30. Barbara Sinclair, Unorthodox Lawmaking: New Legislative Processes in
the U.S. Congress (Washington, DC. CQ Press, 1997), 226.
31. Sinclair, 226.
32. Roger H. Davidson, “Building the Republican Regime: Leaders and
Committees,” in New Majority or Old Minority? The Impact of Republicans on
Congress, ed. Nicol C. Rae and Colton C. Campbell (Lanham, MD: Rowman &
Littlefield, 1999), esp. 79–80.
33. Sinclair, Unorthodox Lawmaking, 226.
34. Quoted in Dan Balz and John Yang, “Republicans Set Legislative
Priorities,” WP, March 7, 1997, A10.
35. Jennifer Bradley, “Democrats Hit GOP Work Pace,” RC, March 6,
1997, 1.
36. Jennifer Bradley, “Democrats Ready a Welcome Back Attack,” RC, April
7, 1997, 1.
37. Bradley, “Democrats Ready a Welcome Back Attack.”
38. Quoted in Bradley, “Democrats Hit GOP Work Pace.”
39. Quoted in Kellman, “Gingrich’s New Agenda Excites Few Republicans.”
40. Quoted in Richard E. Cohen, “Beginning of the End,” NJ, July 26, 1997,
1516–1517.
41. Quoted in Cohen, “Beginning of the End.”
42. Calculated from CQWR, January 3, 1998, 36–37.
43. John F. Manley, “The Conservative Coalition in Congress,” American
Behavioral Scientist 17 (December 1973): 223–247, at 224.
44. Charles E. Cook, “The Onus to Beat ‘Do-Nothing’ Rap Falls to
Chairmen,” RC, May 12, 1997, 6.
45. Julius Turner, Party and Constituency: Pressures on Congress (Baltimore:

Johns Hopkins University Press, 1951).
46. Manley, “Conservative Coalition”; Barbara Sinclair, Congressional
Realignment: 1925–1978 (Austin: University of Texas Press, 1982); and Mack C.
Shelley, The Permanent Majority: The Conservative Coalition in the United States
Congress (Tuscaloosa: University of Alabama Press, 1983).
47. Robert J. Donovan, “For America, a New Coalition?” Los Angeles Times,
July 6, 1981, A4.
48. David W. Rohde, “Electoral Forces, Political Agenda, and Partisanship in
the House and Senate,” in The Postreform Congress, ed. Roger H. Davidson (New
York: St. Martin’s Press, 1992).
49. Sarah A. Binder, “The Disappearing Political Center,” The Brookings
Review, Fall 1996, 36–39.
50. Sarah Binder’s definition of centrists in “The Disappearing Political
Center” is those members who are closer to the ideological midpoint between the
two parties than to the ideological center of their own party.
51. Juliet Eilperin and Rachel Van Dongen, “The Insider’s Guide to House
Factions,” RC, January 20, 1997, 1.
52. Other caucuses represent the hard-core partisans on both sides of the aisle:
e.g., the Mainstream Conservative Alliance, the Congressional Black Caucus, the
Progressive Caucus, and the Hispanic Caucus.
53. Robin Kolodny, “Moderate Success: Majority Status and the Changing
Nature of Factionalism in the House Republican Party,” in New Majority or Old
Minority?, 153.
54. Kolodny, “Moderate Success,” quote 154, also 158–159.
55. Evans and Oleszek, Congress Under Fire, 129.
56. C. Lawrence Evans and Walter J. Oleszek, “Procedural Features of House
Republican Rule,” in New Majority or Old Minority?, 124.
57. Alison Mitchell, “Gingrich Emerging From Self-Exile, Humbler, Wary and
Thinner,” NYT, October 27, 1997, A18.
58. Juliet Eilperin, “Committee Reform Fervor Fades as New Drier Report
Highlights Obstacles to Eliminating More House Panels,” RC, November 14,
1996, 10.
59. Michael Wines, “The 105th Congress: The Senate; Manners Will Matter,
Leaders Caution,” NYT, January 8, 1997, B8.
60. Barbara Sinclair, “Partisan Imperatives and Institutional Constraints:
Republican Party Leadership in the House and Senate,” in New Majority or Old
Minority?, 36.
61. Helen Dewar, “Will Brass Spittoons Survive Move to Modernize Senate?,”
WP, January 15, 1997, A17.
62. Deering and Smith, Committees in Congress.
63. Colton C. Campbell and Roger H. Davidson, “Coalition Building in
Congress: The Consequences of Partisan Change,” in The Interest Group
Connection: Electioneering, Lobbying, and Policymaking in Washington, ed. Paul
Herrnson, Ronald Shaiko, and Clyde Wilcox (Chatham, NJ: Chatham House,
1998), 119–123.
64. Kenneth A. Shepsle, “The Changing Textbook Congress,” in Can
Government Govern?, ed. John Chubb and Paul Peterson (Washington, DC:
Brookings, 1989).
65. D. Roderick Kiewiet and Mathew D. McCubbins, The Logic of
Delegation: Congressional Parties and the Appropriations Process (Chicago:
University of Chicago Press, 1991); Gary W. Cox and Mathew D. McCubbins,
Legislative Leviathan: Party Government in the House (Berkeley: University of
California Press, 1993); Barbara Sinclair, Legislators, Leaders, and Lawmaking
(Baltimore: Johns Hopkins University Press, 1995); Sarah A. Binder, “The Partisan
Bias of Procedural Choice: Allocating Parliamentary Rights in the House,
1789–1900,” APSR 90 (March 1996): 8–20.
67. Kiewiet and McCubbins, The Logic of Delegation; Cox and McCubbins,
Legislative Leviathan.
69. Damon Chappie, “105th Congress Gets Ready for Business: GOP
Changes Style, But Not Its Leadership,” RC, November 21, 1996, 1.
70. Richard F. Fenno Jr., Home Style: House Members in Their Districts
(Boston: Little, Brown, 1978); Fiorina, Congress: Keystone of the Washington
Establishment; Barry R. Weingast and William Marshall, “The Industrial
Organization of Congress,” Journal of Political Economy 91 (1988): 775–800.
71. Kenneth A. Shepsle, “Institutional Equilibrium and Equilibrium
Institutions,” in Political Science: The Science of Politics, ed. Herbert Weisberg
(New York: Agathon Press, 1986); Weingast and Marshall, “Industrial
Organization.”
73. Rae, Conservative Reformers, 190.
74. Cohen, “Beginning of the End,” 1517.
75. Jeffrey L. Katz, “Politically Charged Vote Sets Tone for Impeachment
Inquiry,” CQWR, October 10, 1998, 2712–2715.
76. Calculated from CQWR, December 22, 1998, 3372.
77. Sinclair, Unorthodox Lawmaking.
Epilogue
HARVEY L. SCHANTZ
The most notable political development during the second Clinton administra-
tion was the House impeachment of the president. On December 19, 1998, the
105th U.S. House voted two articles of impeachment against Bill Clinton for
grand jury perjury in testimony concerning his relations with Monica Lewinsky
and Paula Jones, and for obstruction of justice in the Jones sexual harassment
lawsuit. The 106th U.S. Senate, however, on February 12, 1999, acquitted
Clinton on both articles.
In both chambers of Congress, the pattern of the vote demonstrated the
importance of congressional political parties and the hazards of divided gov-
ernment for presidents. In the House, the grand jury perjury and obstruction of
justice articles of impeachment drew only five Democratic supporters, but was
voted for by 223 and 216 Republicans, respectively. In the Senate, neither arti-
cle of impeachment attracted a single Democratic vote, but they drew 45 and
50 Republicans, respectively. Clearly a Democratic Congress would not have
brought Clinton to the brink of removal from office.
A second significant political development was the loss of seats by House
Republicans in the 1998 midterm election. In November 1998, the Republicans
retained control of the U.S. House, but the five-seat Democratic gain was the
first pickup of seats for the president’s party at midterm since 1934, and only
the second such gain since 1862. Many congressional Republicans blamed
Speaker Newt Gingrich for the loss of seats, and three days after election day,
Gingrich declined to run for the Speakership in the newly elected Congress.
Dennis Hastert (R-Ill.) became the new Speaker on January 6, 1999, at the out-
set of the 106th Congress.
A number of politicians emerged from the 1998 election with heightened
prospects for election 2000. In New York State, First Lady Hillary Rodham
Clinton effectively campaigned for successful Democratic senate candidate
Charles Schumer, and thus positioned herself for a senate candidacy of her
own. Meanwhile, Texas Governor George W. Bush Jr., was reelected by a large
margin and quickly became the front-runner for the Republican presidential
181
182 Epilogue
nomination. Jeb Bush, a younger brother, was elected governor of Florida, fur-
ther demonstrating the appeal of the Bush name in the growing sunbelt states.
For the most part, impeachment and electoral politics have overshadowed
policy during the second Clinton administration. Nevertheless, the advent of a
balanced federal budget has greatly altered the domestic policy environment. In
the summer of 1997, the president and Congress reached a historical agreement
for a balanced federal budget by fiscal year 2002, but that goal has already
been achieved due to the growing economy. Currently the policy debate is over
how to use the federal surplus. Most Republicans would like to offer tax cuts.
President Clinton and most congressional Democrats would like to keep tax
cuts to a minimum and use the surplus money to stabilize Social Security,
Medicare, and fund other federal programs, including farm aid. A possible
solution to the policy impasse would be to use the surplus to lower the nation-
al debt.
The political stakes in the budget surplus debate are great. In 1990,
President George Bush permanently lost important credibility with the elec-
torate by agreeing to raise taxes in a budget deal with the Democratic
Congress. Both parties are wary of conceding any political ground, and both
advocate solutions most supportive of their traditional constituencies. Policy-
making in the remaining time of the Clinton administration will be tied to posi-
tion-taking for the 2000 elections.
List of Contributors
Emmett H. Buell Jr. (Ph.D., Vanderbilt University, 1972) is a professor of polit-

ical science and director of the Richard G. Lugar Program in Politics and Public
Service at Denison University. He has written numerous articles and book
chapters on presidential nominating politics and is the co-editor of Nominating
the President (1991).
Colton C. Campbell (Ph.D., University of California, Santa Barbara, 1996) is

an assistant professor of political science at Florida International University. He
has served as a Congressional Fellow of the American Political Science
Association and has received a Dirksen Congressional Research Grant.
Campbell is coeditor of New Majority or Old Minority? The Impact of the
Republicans on Congress (1999).
Milton C. Cummings Jr. (Ph.D., Harvard University, 1960) is a professor of

political science at Johns Hopkins University. Cummings has been a Rhodes
Scholar, Guggenheim Fellow, and for six years was on the staff of the
Brookings Institution. His many books include Congressmen and the
Electorate: Elections for the U.S. House and the President, 1920–1964 (1966);
The National Election of 1964 (1966); The Patron State: Government and the
Arts in Europe, North America, and Japan (1987); and Democracy Under
Pressure: An Introduction to the American Political System, 8th ed. (1997).
Roger H. Davidson (Ph.D., Columbia University, 1963) is a professor emeritus

of government and politics at the University of Maryland, College Park, and
visiting professor of political science, University of California, Santa Barbara.
From 1980 to 1988 he held the post of Senior Specialist in American National
Government and Public Administration with the Congressional Research
Service, U.S. Library of Congress. He is co-author of Congress and Its
Members, 7th ed. (2000); and coeditor of The Encyclopedia of the United
States Congress (1995).
183
184 List of Contributors
Douglas B. Harris (Ph.D., Johns Hopkins University, 1998) is an assistant pro-

fessor of government, politics, and political economy at the University of Texas
at Dallas. He is author of “The Rise of the Public Speakership,” Political
Science Quarterly (1998); and “Dwight Eisenhower and the New Deal: The
Politics of Preemption,” Presidential Studies Quarterly (1997). He was a 1995
recipient of a Dirksen Congressional Research Grant, and has been a visiting
assistant professor of political science at Colgate University and Johns Hopkins
University.
Garrison Nelson (Ph.D., University of Iowa, 1973) is a professor of political

science at the University of Vermont and senior Fellow at the McCormack
Institute, University if Massachusetts, Boston. He is the editor of the two-vol-
ume set Committees in the U.S. Congress, 1947–1992 (1993). He regularly
provides ideological rankings of all Members of Congress to National Journal
Inc. for use in their biennial editions of The Almanac of American Politics.
Harvey L. Schantz (Ph.D., Johns Hopkins University, 1978) is a professor of

political science at the State University of New York, Plattsburgh. He has
served as a Congressional Fellow of the American Political Science Association
and as a Visiting Fellow at Yale University. He is the editor of American
Presidential Elections: Process, Policy, and Political Change (1996).
Margaret Jane Wyszomirski (Ph.D., Cornell University, 1979) is Director of the

Arts Policy and Administration Program at Ohio State University and a pro-
fessor on the faculty of both the School of Public Policy and Management and
of the Art Education Department. Her most recent book is the co-edited vol-
ume The Public Life of the Arts in America (2000). She is also the co-editor of
Executive Leadership in Anglo-American Systems (1991), and co-editor of
America’s Commitment to Culture: Government and the Arts (1995).
Wyszomirski has been a past or current member of the editorial boards of
Governance, Journal of Policy Studies, Presidential Studies Quarterly, and The
Journal of Arts Management, Law, and Society.
Index
Abortion, 22, 24 Baucus, Max, 51

Abramowitz, Alan I., 88, 90, 100 Bauer, Gary, 4, 28
Abscam, 115 Bennett, Robert F., 171
African American voters, 75 Bennett, William, 3
Agnew, Spiro, 114 Berger, Samuel, 135, 136
Alabama, margin needed for victory in Better American Foundation, 3
primaries, 46 Biden, Joseph, 86, 99
Alaska, as a unified primary state, 45 Binder, Sarah A., 165, 174
Albright, Madeleine K., 135, 136 Black, Gordon, 25
Alexander, Lamar, 3, 4, 5, 27, 28 Blute, Peter I., 158
attacks on Dole, 13 Boehner, John, 16
caucuses and primaries, 15, 17, 18 Boren, David, 25, 51
Dole attacks on, 15–16 Boren, Jim, 51
fundraising, 6, 7 Boschwitz, Rudy, 98, 99
straw polls, 10, 11, 12 Bowles, Erskine, 135, 136
Ament, Don, 43 Bradley, Bill, 27
Apple, R. W., 15 Breaux, John B., 87
Arkansas Brown, Hank, 92
convention nominations, 48 Brown, Jesse, 135
margin needed for victory in Brownback, Sam, 59
primaries, 46 Browne, Harry, 63
Armey, Richard, 160 Bruggere, Tom, 95
Ashcroft, John, 27 Bryan, William Jennings, 111
Buchanan, Bay, 22
Babbitt, Bruce, 135 Buchanan, Pat, 3, 4, 5, 6, 27
Baker, James, 3 attacks on Dole, 13
Barbour, Haley, 4, 79 caucuses and de facto primaries,
Barnes, Fred, 20 14–15, 16–19
Barshefsky, Charlene, 135 fundraising, 7, 8
Barton, Joe L., 163 straw polls, 9–11
Bateman, Herbert H., 47–48 Burnham, Walter Dean, 108
185
Index 186
Craig, Larry, 86 television news coverage, 9

“Creating a Better America for Ourselves voter support between 1995–1996,
and Our Children,” 144, 160 70
Cuomo, Andrew M., 135 Domenici, Pete, 86, 142
Cummings, Milton, 108 Doolittle, John, 170
Dornan, Robert, 5, 7
Daley, William M., 95, 135 Dreier, David, 170
D’Amato, Alphonse, 16, 93, 99 Drew, Elizabeth, 155–156, 158
Debates, 69, 71 Dukakis, Michael, 73, 112
De facto primaries, 12–19 Durbin, Richard, 95
versus ex post facto primaries, 2
DeFazio, Peter, 93 Edwards, Edwin, 45
Democratic Congressional Campaign Eisenhower, Dwight, 80, 81, 111, 113
Committee (DCCC), 116 Electoral patterns in congressional
Democratic National Committee (DNC), primaries, 54–59
pre-convention phase of general Electoral votes, comparison of popular
election campaign, 21 and, 64
Democratic Senatorial Campaign Electorate, descriptions of
Committee (DSCC), 86, 95–97 of 1995, 67
Dickey, Jay, 162–163 of 1996, 64–65
DioGuardi, Joseph J., 59 Emerson, Bill, 54
Divided government Emerson, Jo Ann, 54
Clinton, Gingrich, and Republican English, Phil, 158
freshmen, 117–119 Evans, C. Lawrence, 161, 167
effects of, 120–121 Ewing, Cortez A. M., 46
growth of, 112–114 Exon, Jim, 101
Nixon, 114–115 Ex post facto primaries, de facto
O’Neill, 115–117 versus, 2
presidential legacies, 122–123
Dodd, Christopher, 25 Families First, 138
Dole, Elizabeth, 27, 28 Family Caucus, 165
Dole, Robert, 3, 5 Federal Election Campaign Act (FECA),
attacks on Alexander, 15–16 6
attacks on Forbes, 13–14 Filner, Bob, 59
caucuses and de facto primaries, Finkelstein, Arthur J., 99
12–19 Fishel, Jeff, 139
contradictions and inconsistencies, Florida, margin needed for victory in
12–13 primaries, 46
effects of front-loading, 27–29 Foley, Thomas, 107
fundraising, 6, 7 Forbes, Steve, 5, 6, 27
national poll standings, 8 caucuses and de facto primaries, 15,
nomination of, 69 17, 18
popular and electoral votes, com- Dole attacks on, 13–14
parison of, 64 fundraising, 7, 8, 28
Powell and, 4 straw polls, 11
pre-convention phase of general elec- Ford, Gerald, 80, 81, 111, 149
tion campaign, 20–23 Foster, Mike, 14
straw polls, 9–12 Foster v. Love, 46
187 Index
Burton, Phil, 116 staffing, 133–137

Bush, George H. W., 22, 67, 73, 77, 80, State of the Union address, 140–141
81, 111, 113, 149, 182 Clinton, Hillary Rodham, 19, 149,
Bush, George W., 27, 28, 182 181–182
Bush, Jeb, 182 Close, Elliot, 95
Cochran, Thad, 86
Calhoun, John C., 74 Coelho, Tony, 116
California, as a unified primary state, 45 Cohen, Richard, 163
Campaign America, 3 Cohen, William S., 135, 136
Campbell, Carroll, 3 Collins, Barbara-Rose, 51, 58
Campbell, James, 100 Collins, Susan, 53
Carter, Billy, 116 Colorado, congressional nominating
Carter, Jimmy, 65, 67, 74, 80, 81, 111, system, 43
112, 115–116, 149 Colorado Republican Federal Campaign
Casey, Robert, 20 Committee v. Federal Election
Caucuses, 12–19 Commission, 96
Cheney, Dick, 3 Congress, legacy of 105th
Choate, Pat, 27, 69 committee and subcommittee seats,
Christopher, Warren, 136 168–169
Cisneros, Henry, 135 leadership versus committees,
Clatworthy, Raymond, 99 171–172
Clay, Henry, 122 1996 elections, 156–159
Cleland, Max, 86, 87 partisanship of House, 166–170
Cleveland, Grover, 72, 109 policy consequences, 173–176
Clinton, Bill return to normalcy, 159–166
divided government, 117–119 tranquillity of Senate, 170–171
impeachment, 173–176, 181 Congressional nominations in 1996
legacy of presidency, 122–123 candidates, 49–54
performance rating, 66, 68, 79–82 districts conceded by parties, 49
popular and electoral votes, electoral patterns in primaries, 54–59
comparison of, 64 methods for nominating, 41–49
prelude to 1996, 65–68 nonprimary, 47–49
renomination of, 69 primary systems, 43–47
uncontested nomination, 19–20 Connally, John, 114
voter support between 1995–1996, Connecticut
70 congressional nominating system, 43
Clinton, Bill (second transition) convention nominations, 48
first hundred days, 143–146 Conservative Action Team (CAT),
fiscal year 1998 budget proposal, 165–166
141–142 Contract with America, 67, 106, 119,
narrow party margins, 164–166 138, 144, 147, 155, 161, 162
1997 inaugural address, 140 Convention bounce, 69
party platform, 139–140 Convention nominations, 47–48
in perspective, 146–149 Cook, Merrill, 43
policy agenda, 137–143 Cooley, Wes, 48
political capital and context, Coolidge, Calvin, 111
131–133 Coverdell, Paul, 145
public opinion polls and issues, 140 Cox, James, 111
188 Index
Frahm, Sheila, 56, 59 Hall, Joelle, 50

Frank, Barney, 162 Harkin, Tom, 51
Frisa, Dan, 52 Harrison, Benjamin, 109, 111
Front-loading, 1–2, 27–29 Hastert, Dennis, 181
Fundraising Hayes, Rutherford, 109
investigations of, 132–133 Heflin, Howell, 87, 101
presidential nominees, 6–8 Heineman, Ben, 138
Senate, 95–97 Helms, Jesse, 56
Henry, Ann, 48
Garfield, James A., 109 Herman, Alexis M., 135, 137
Gender differences in voting, 76 Herrnson, Paul S., 79
Georgia, margin needed for victory in Hibbing, John R., 91
primaries, 46 Hispanic American voters, 75
Gephardt, Richard, 27, 94, 119 Hogan, Thomas F., 69
Germond, Jack, 19 Hoke, Martin R., 52
Gingrich, Newt, 3, 4, 16, 65, 67, 99, Hoover, Herbert, 111
106, 107, 181 House elections. See U.S. House
divided government, 117–119, elections
122–123 Huckabee, Mike, 48
investigations, 132–133 Humphrey, Hubert, 111, 112, 114
105th Congress and, 161, 162, 163, Hutchinson, Asa, 48
172 Hutchinson, Tim, 48, 87
Giuliani, Rudolph, 16 Hyde, Henry, 22, 174
Glickman, Dan, 135
Goeas, Ed, 93, 94 Ickes, Harold M., 135
Goldwater, Barry, 77 Illinois, voter eligibility, 44–45
Goodling, Bill, 58, 145 Impeachment, 173–176, 181
GOP Individuals with Disabilities Education
invisible primaries for the, 2–12 Act (IDEA), 146
outcomes of 1996, 78 Inhofe, James M., 51, 86
taming of agenda, 159–161 Invisible primaries for the GOP, 2–12
Gore, Al, 27, 69, 118, 119, 133, 144, Iowa, margin needed for victory in
149 primaries, 46
Gramm, Phil, 3, 4, 5, 51 Istook, Ernest J., Jr., 160
caucuses, 14–15
fundraising, 6, 7, 8 Jackson, Andrew, 63, 74
straw polls, 9–12 Jackson, Brooks, 116
television news coverage, 9 Jackson, Henry M., 114
Grassley, Charles, 14 Jackson, Jesse, 20
Greene, Enid, 43 Jacobson, Gary, 91
Green party, 63, 69 Jarding, Steve, 96
Greenwood, James, 22 Jenkins, Louis, 45, 86, 97
Grunwald, Mandy, 99 Johnson, Lyndon B., 77, 80, 81, 111,
Guinier, Lani, 19 112, 130
Johnson, Tim, 51, 56, 89–90, 99
Hadley, Arthur, 3, 8 Johnston, J. Bennett, 87
Hagel, Chuck, 53, 98 Jones, Paula, 181
Index 189
Kasich, John, 27 McConnell, Mitch, 86, 158

Keene, David, 4 McCorkell, Don, 51
Kelly, Sue W., 59 McCurry, Michael, 68, 144
Kemp, Jack, 3, 24, 25, 69 McDade, Joseph, 58
Kennedy, John F., 113 McGovern, George S., 112, 114
Kennedy, Ted, 115 McGovern, Jim, 158
Kerrey, Bob, 27, 90, 95 McIntosh, David M., 167
Kerry, John, 27, 51, 97, 99 McKinley, William, 111
Key, V. O., Jr., 46, 58, 60, 82 McKinney, Cynthia, 51–52
Keyes, Alan, 5, 6 McLarty, Thomas F. III, 25, 67
fundraising, 7, 28 McNulty, Michael R., 59
straw polls, 10, 11 Media
Kilpatrick, Carolyn Cheeks, 58 invisible primaries and role of, 2–3
Kim, Jay, 58 television news coverage, 8–9
Kolodny, Robin, 166 Merrill, Stephen, 14
Kucinich, Dennis J., 52 Mfume, Kweisi, 22
Kusper v. Pontikes, 44–45 Millner, Guy, 86
Minnesota, congressional nominating
Lake, Anthony, 136, 137, 147 system, 44
Lamm, Richard, 25–26, 69 Minnick, Walt, 95
Landrieu, Mary, 45, 86, 87, 97 Mississippi, margin needed for victory in
Laughlin, Greg, 46, 59 primaries, 46
Levin, Carl, 86 Moakley, Maureen, 43–44
Lewinsky, Monica S., 173, 181 Mondale, Walter, 112
Libertarian party, 63 Moore, Michael K., 91
Light, Paul, 138 Morales, Victor M., 51
Lightfoot, Jim Ross, 51 Morris, Dick, 4, 24, 119
Limbaugh, Rush, 4, 16 Muskie, Edmund, 114
Log Cabin Republicans, 13 Myers, Dee Dee, 68
Long, Carol, 4
Long, Russell B., 87 NAACP, 22
Lott, Trent, 78, 142, 144, 167, 170, 173 Nader, Ralph, 63, 69
Louisiana, as a unified primary state, Nash, Robert J., 134
45–46 National nomination conventions, 23–24
Lugar, Richard G., 4, 5, 17, 18 National poll standings, 8
fundraising, 7 National Republican Senate Committee
straw polls, 10, 11 (NRSC), 95–97
Luntz, Frank, 106 Nelson, Ben, 56, 98
Neumann, Mark W., 173
Main Street Coalition, 166 Neustadt, Richard, 120
Manley, John, 164 Newfield, Jack, 19
Martin, Lynn, 3 New Mexico, congressional nominating
Martini, Bill, 158 system, 43
Matthews, Christopher, 116–117 New York
Mayer, Nancy, 99 congressional nominating system, 43
Mayhew, David, 147 voter eligibility, 44
McCain, John, 16, 27, 28, 160 Nichol, Gene, 43
McCarthy, Carolyn, 52
190 Index
Nixon, Richard M., 78, 80, 81, 105–106, campaign, description of, 69–72
111, 113, 114–115 electorate, description of, 64–65
Nonprimary nominations minor-party votes, 63
convention, 47–48 performance ratings, 79–82
party committee, 48 popular and electoral votes,
self-nominations, 49 comparison of, 64
write-ins, 48–49 prelude to, 65–68
North, Oliver, 4, 13 voting patterns, 72–79
North Carolina, margin needed for Presidential nominations in 1996
victory in primaries, 46 candidates that dropped out, 3–4
North Dakota, congressional nominating early caucuses and de facto
system, 43 primaries, 12–19
Nunn, Sam, 87 final field, 4–6
fundraising, 6–8
Oklahoma, margin needed for victory in invisible primaries for the GOP, 2–12
primaries, 46 national poll standings, 8
Oklahoma City bombing, 68, 79, 137 stages and front-loading, 1–2
Oleszek, Walter J., 161, 167 straw polls, 9–12
O’Neill, Thomas P., Jr., 106, 115–117, television news coverage, 8–9
157 uncontested Democratic, 19–20
Oregon, Senate race, 93–94 Presidential Power (Neustadt), 120
Pressler, Larry, 51, 89–90, 99, 105
Packwood, Bob, 85, 91, 93 Primaries for the GOP, invisible, 2–12
Panetta, Leon E., 67–68, 118, 135 Primary systems
Party committee, nomination by, 48 margin needed for victory, 46–47
Party loyalty and defection, 77 party and ballot access, 43–44
Pascrell, Bill, Jr., 158 runoffs, problems with, 47
Paul, Ron, 59 unified/consolidated, 45–46
Pena, Federico F., 135 voter eligibility, 44–45
Penny, Tim, 25 Pryor, David, 87
Performance ratings, 66, 68, 79–82
Perot, H. Ross, 24–27, 28–29, 63, 77 Quayle, Dan, 3, 12, 27
effects of, 73
nomination of, 69 Rae, Nicol, 156, 173
popular and electoral votes, com- Raines, Franklin, 136
parison of, 64 Rayburn, Sam, 122, 123
voter support between 1995–1996, Reagan, Ronald, 67, 77, 80, 81, 111,
70, 71 112, 116, 117
Perry, William, 136 Reed, Ralph, 4, 22
Pfiffner, James, 131 Reed, Scott, 71
Poll standings Reed, Tom, 122, 123
national, 8 Reform party, 25, 28
straw, 9–12 Regional differences in voting, 73–75
Pomper, Gerald, 65, 139 Rehberg, Dennis, 51
Popular and electoral votes, comparison Reich, Robert, 135
of, 64 Religious differences in voting, 75–76
Powell, Colin, 3–4, 8, 16, 149 Reno, Janet, 135
Presidential campaign and vote in 1996 Republican National Committee (RNC),
Index 191
pre-convention phase of general Strickland, Tom, 43

election campaign, 21 Sumners, Joe, 100
Republican party, infighting, 21–22 Swett, Dick, 86
Rhode Island, congressional nomi-
nating system, 43–44 Taft, William Howard, 111
Richardson, Bill, 136 Tashjian v. Republican Party of
Riley, Richard W., 135, 145 Connecticut, 45
Rockefeller, John D. IV, 56, 86 Tax cuts, Dole on, 23
Roosevelt, Franklin, 63, 68, 73, 111 Taylor, Morry, 5, 6, 7, 11
Roosevelt, Theodore, 111, 118 Texas, margin needed for victory in pri-
Rosario v. Rockefeller, 44 maries, 46
Rosenbaum, David E., 92 Thompson, Fred, 27, 86
Rostenkowski, Dan, 106 Thompson, Tommy, 3
Rubin, Robert E., 135, 136 Thurmond, Strom, 51
Ryun, Jim, 54 Tierney, John, 68
Tilden, Samuel, 109
Sanders, Charlie, 95 Torricelli, Robert G., 56, 98
Schaffer, Bob, 43 Transition
Schantz, Harvey, 75 See also Clinton, Bill (second tran-
Schlesinger, Arthur M., Jr., 122 sition)
Schumer, Charles, 181 accidental, 130
Segal, Jeffrey A., 88, 90, 100 inter-party, 129–130
Self-nominations, 49 inter-term, 130–131
Senate Conservative Action Team intra-party, 130
(SCAT), 166 tasks of, 129
Senate elections. See U.S. Senate Truman, Harry S, 74, 109, 111, 130
elections Tuesday Group, 165, 166
Sessions, Jeff, 87
Shafer, Byron, 23 U.S. House
Shalala, Donna E., 135 See also Congress, legacy of 105th;
Simon, Paul, 95 Congressional nominations in
Sinclair, Barbara, 161 1996
Slater, Rodney, 135 contested and uncontested primaries,
Smith, Gordon, 93–94 54–55
Smith, Robert, 27, 48, 86, 99 continuity versus change, 105
Snowe, Olympia, 16, 22 nominating methods, 42
Social group differences in voting, 75–76 outcomes of 1996 voting, 63–64, 78
South Carolina, margin needed for seats with entered incumbent, 51–52
victory in primaries, 46 seats without entered incumbent,
South Dakota, margin needed for 53–54
victory in primaries, 46 U.S. House elections
Specter, Arlen, 3, 4, 7, 11 See also Divided government
Sperling, Gene, 135 context of, 106–107
Starr, Kenneth W., 174 presidential results based on,
Stephanopoulos, George, 135 108–112
Stevens, Ted, 50, 56, 86 sideshow problem, 108
Stevenson, Adlai, 111 split partisan results, 112
Straw polls, 9–12
192 Index
three-variable predictive model, Virginia, convention nominations, 47

109–112 Voting patterns
U.S. Senate control of government, 78–79
See also Congress, legacy of 105th; national, 72–73
Congressional nominations in party loyalty and defection, 77
1996 Perot, effects of, 73
contested and uncontested primaries, sectional patterns, 73–75
55–56 social groups, 75–76
continuity versus change, 105
electoral opportunities, 92–93 Wallace, George, 113, 114
expectations and strategic decisions, Warner, John, 47–48
91–92 Warner, Mark, 95
incumbency, role of, 89–91 Washington, as a unified primary
nominating methods, 42 state, 45
outcomes of 1996 voting, 63–64, 78 Weicker, Lowell, 25
seats with entered incumbent, 50–51 Weko, Thomas, 134
seats without entered incumbent, Weld, William, 3, 51, 56, 97, 99
52–53 Wellstone, Paul, 27, 56, 98
U.S. Senate elections West Virginia, voter eligibility, 44
See also Divided government Weygand, Bob, 44
candidate recruitment and quality, 95 Weyrich, Paul, 4
distinctive characteristics, 97–98 Wicker, Roger, 160
fundraising, 95–97 Wilcox, Clyde, 79
incumbents and partisan trends in Wilkie, Wendell, 73
open seats, 98–101 Will, George, 4, 16
in Oregon, 93–94 Wilson, Pete, 3, 7, 10, 27
outcomes, 85–87 Wilson, Woodrow, 63, 72, 111
which states have races, 88–89 Wines, Michael, 170
United We Stand, 24, 25 Write-in nominations, 48–49
Utah Wyden, Ron, 93–94
congressional nominating system, 43
convention nominations, 48 Yellen, Janet, 136
Van Buren, Martin, 111 Zimmer, Dick, 98, 99

Ventura, Jesse, 28–29

Untitled

Uploaded by

Copyright:

Available Formats

Untitled

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Untitled

Uploaded by

Copyright:

Available Formats

Politics in an Era of Divided Government

POLITICS AND POLICY IN AMERICAN

Published in Great Britain by

Routledge is an imprint of the Taylor & Francis Group.

This edition published in the Taylor & Francis e-Library, 2002.

Copyright © 2001 by Routledge

Library of Congress Cataloging-in-Publication Data

Politics in an era of divided government : elections and governance in the

ISBN 0-203-90315-3 Master e-book ISBN

Chapter 1 Some Things Are Predictable: Nominating

Chapter 2 Congressional Nominations in 1996:

Chapter 3 The Presidential Campaign and Vote in 1996:

Chapter 4 Strategic Partisan Decisions and Blunted

Chapter 6 Clinton’s Second Transition: Historic

Chapter 7 The Irony of the 105th Congress and Its Legacy

List of Contributors 183

1.1 1996 Invisible-Primary Finances of Declared

2.1 Nominating Methods for the U.S. House, 1996 42

3.1 President Clinton’s Job Rating, 1993–1996 66

4.1 States with 1996 Senate Elections by Presidential

5.1 Fifty-Plus House Seat Election Gains and House

5.3 Congresses With Divided Government By

6.1 Clinton Cabinets I and II 135

7.1 Congressional Workload Comparisons,

AJPS American Journal of Political Science

The Routledge series, “Politics and Policy in American Institutions” strives

edited volume in this series, Politics in an Era of Divided Government pro-

THE PARTY NOMINATIONS FOR THE PRESIDENT

tion systems used by the states. These differences, as described by Harvey

THE GENERAL ELECTIONS FOR THE PRESIDENT

Cummings views the 1996 election as a referendum on the incumbent,

of divided government. As he notes, most explanations for the rise of divid-

ORGANIZING THE NEW ADMINISTRATION AND CONGRESS

replacing seven of the fourteen Department Secretaries, as well as filling

brought revolutionary change to Congress, as the Republicans won control

Some Things Are Predictable

EMMETT H. BUELL JR.

This is an account of how the Republican, Democratic, and Reform parties

NOMINATING STAGES AND FRONT-LOADING

build campaign organizations in key primary states, articulate a message,

A LONG INVISIBLE PRIMARY FOR THE GOP

list of presidential aspirants included General Colin Powell, Secretary of

QUAYLE, POWELL, AND OTHERS OPT OR DROP OUT

openly welcomed a Powell candidacy. Ralph Reed of the Christian Coali-

THE FINAL FIELD: INSIDERS AND OUTSIDERS

THE MONEY CHASE AND EARLY SPENDING

For all of its undeniable importance, if early money determined every-

NATIONAL POLL STANDINGS

TELEVISION NEWS EXPOSURE

STRAW POLLS: “IS ANYBODY HERE FROM IOWA?”

Table 1.2: 1994–1996 Republican Presidential Straw Polls

Date Event Winners and Losers Particulars

January 7, 1995 Louisiana GOP conven- Gramm 72%, Gramm launched a

March 3, 1995 South Carolina GOP Gramm 35%, Voting limited to

Table 1.2 continued

January 30, 1996 Alaska GOP Buchanan 33%, Buchanan waged an

At least seven of these limited voting to purchasers of tickets costing

EARLY CAUCUSES AND DE FACTO PRIMARIES

vative,” and that most of these claimed to be “very conservative.”

Playing to Republican fears of another debacle on the scale of Barry