Page 1 of 17

RAP Publication: 1996/19

Low External Input Sustainable Agriculture (LEISA)

in Selected Countries of Asia:

Regional Workshop Report and Case Studies

Food and Agriculture Organization of the United Nations

Regional Office for Asia and the Pacific

August 1996

VIETNAM

Duong Ngoc Thanh et al

For many years in the Asia-Pacific Region, the production of cereals had been met by expansion of area.
However, with decreasing land availability, this is no longer possible. Increased production has to depend more
on increasing productivity through increased use of package of technologies, consisting of external inputs such
as chemical fertilizer, pesticides/-herbicides, better varieties and use of irrigation water. These technologies
benefited more the larger, and resource-endowed farmers, and adoption has not spread rapidly among the
small resource-poor farmers. Low adoption has been blamed on ineffective extension systems, lack of credit,
market, etc. But more important is the inappropriateness of the technology to the needs of small farmers, high
cost of these external inputs and non-availability due to physical isolation. Further, the sustainability of
agriculture for resource-poor farmers in the region has become a matter of increasing importance for
practitioners, researchers, policy-makers and farmers.

Sustainability has been defined, for the purpose of including resource-poor as well as resource-rich farmers, as
managing agricultural productivity while maintaining or improving the resource base. This means that
agriculture must be environmentally solid, economically feasible, socially scrupulous, and flexible (for future
needs). One of the most promising paradigms that has emerged for the benefit of small scale resource-poor
farmers is Low External Input and Sustainable Agriculture (LEISA), which can enable such farmers to achieve
higher income and attain sustainability by:

1. Optimizing the use of locally available resources, thereby achieving a synergetic effect among the
various components of the farming system (soil, water, animals, plants, etc.) so that they complement
each other in the production of output.
2. Minimizing the use of external inputs, except where there is a serious deficiency and where the effect on
the system will be to increase recycling of nutrients.

The aim is not to maximize short-term production, but to attain an adequate and sustainable level over the
longer term.

To achieve these goals, LEISA must tap the most viable indigenous knowledge and practices and ecologically
 Page 2 of 17

friendly technologies in a given ecological and sociocultural setting, since the experience in one agro-
ecological setting may not be appropriate in other areas. There are many cases of farmers using LEISA under
different agro-ecological zones in different countries. These experiences should be documented to learn more
of the principles, constraints, and potentials in order to provide policy-makers, development workers and
farmers [with] alternative and viable strategies to develop sustainable farming systems.

In late 1989, Vietnam emerged as the third largest rice-exporting country of the world — after Thailand and the
USA — with about 1.8 to 2 million tons of milled rice per year. However, even though Vietnam exported
remarkable amounts of rice, the living standards of the rice farmers did not much improve because the rice
production costs were high, whereas rice prices remained quite low. It is believed that integrated fanning
systems can help rice farmers increase their farm income.

Recent research findings showed that through integrated fanning systems development approaches, these
agro-ecosystems can be turned to more profitable fanning (V.T. Xuan, 1991). Rice farmers shifting to
diversified agriculture can earn better incomes while maintaining sustainable use of physical, biological and
socioeconomic resources; thus they can improve their living standards.

Objectives of the case study

This study provides the database for LEISA by documenting existing cases of low external input agriculture,
use of indigenous technologies. This will enable the drawing of conclusions on the short-term and long-term
potentials of this approach, so as to guide potential users, policy-makers and development workers in its use,
constraints, potentials, with economic viability, sustainability and national food security as main considerations.

Specifically, the case study will find [out] the different traditional practices in [various) zones and evaluate
constraints, as well as potential, on the basis of secondary data. The study will also identify out the
management practices followed by the case study farmers using LEISA and evaluate economic profitability,
ecological sustainability, as well as employment generation.

The Mekong Delta: General information

Physical environment

The Mekong Delta consists of 11 provinces, with an area of the Delta is around four million hectares, and
holding about 12 percent of the total area of the country. The Delta belongs to monsoon' tropical climate, lies
104.50o to 106.50o east of longitude, and 8.40o to 10.55o north of latitude.

Average temperature is about 26-27oC. There is not much variation in temperature (the highest in the dry
season from 28-32oC, and the lowest from 22-24oC in the wet season). Relative humidity remains high
throughout the year - about 83 percent - and solar radiation is about 147 kcal/cm2/year.

Average rainfall is about 1800 mm, and the rainy season occurs from May to November, supplying 80 percent
to the rainfall. The dry season is from April to December. Rainfall distribution among areas in the region is
different: in the forest area, about 2000-2400 mm; in the central and northeast Delta, about 1400-1600 min;
and in the south, about 2300 mm.

Most of the soils in the country are acid sulfate soils which are subjected to saline water intrusion in the dry
season. The Mekong Delta has three principal soil types.

1. Alluvium soils, which are found along the Mekong and Basac rivers, covering an area of about 1.1. million
hectares (28 percent of the Delta). These are mostly slightly acidic (pH value of 4.5-6.5) and are suitable for
cultivation of high-yielding rice.

2. Sulfate soils which occupy 1.6 million hectares, mainly in Plain Reeds (Dong Thap Muoi) and Long Xuyen
Quadrangle. Sulfate soils consist of salty sulfate found in the south along the coastal line, and sulfate found in
the Plain of Reeds and Long Xuyen Quadrangle. Salty sulfate soils cover about 1.1 million hectares comprising
28 percent of the Delta. Sulfate soils occupy 0.5 million hectares (13 percent of the Delta). These soils contain
 Page 3 of 17

high concentrations of sulfate and low pH values ranging from 2.26 to 3.54; these are constraints to rice
cultivation.

3. Salty soils, which are found along the coast; they cover about 808,750 hectares, equivalent to 21 percent of
the Delta.

Floods are caused by the Mekong River in the wet season. A part of the Mekong Delta in the west bordering
Cambodia consists of Kien Giang, An Giang, Dong Thap and a part of Long Ali province, covering 1.2 million
hectares (31 percent of the Delta), is frequently flooded from August to October. The water level rises varying
from one to four meters; the biggest flood occurs in September. This area still remains for deepwater rice
farming systems, even though a large area is converted into irrigated dry land. Flood is the main constraint for
rice fields, causing some crop loss.

Socio-economic environment

Data as of 1994 indicate a population of about 15.4 million in the Mekong Delta, of whom 13 million belong to
the agricultural sector. The minimum and maximum number of family members are three and 15, respectively.
Tell percent of the families have more than tell members, with average family size at seven members. "Farming
households number 2.4 million; and population density, 395 people per square kilometer. Approximately 5.5
million (36 percent of the population) are economically active members of the agricultural labor force. They
primarily grow rice. Although hired labor, hired carabao and tractor plowing from season migration among the
Mekong Delta provinces are available, there is still shortage of labor force during the peak season.

Average farm size per household is 1.13 hectares; and average land per capita, 0.18 ha. with the range from
0.03 to 0.84 ha. The number of years in rice farming ranges from 5-50, with an average of 25 years
experience. About 24 percent of the farmer respondents had not gone to school; 21 percent had 12 years'
schooling; 24 percent, 6-9 years of education; and about 5 percent had reached university level.

The Mekong Delta is an important region of Vietnam. It holds 12 percent of the total area of the country, and
contributes about 38 percent to the country's GDP. In 1994, it provided 13-14 million tons of rice which played
an important role in rice exports. The Delta has a potential in developing the economy of Vietnam. Since the
government adopted agricultural policy reforms in 1986 and the open-door policy in 1989, the economy in the
Mekong Delta has developed rapidly.

Agricultural production

The Mekong Delta is the biggest rice-producing region in Vietnam. In 1994, the Delta held nearly 47 percent of
the rice area in the whole country. The agriculture sector in the Mekong Delta is diversified. Farming systems in
the Mekong Delta are diversified, varying according to agro-ecological zones.

Figure 1 shows various cropping patterns in the Mekong Delta. In areas where irrigation is available. farmers
can grow three rice crops per year, with at least two rice crops of high-yielding, modern varieties. In rainfed
areas, traditional photo-sensitive varieties are grown. Some areas grow medium-duration modem varieties.
Farmers now arrange their cropping sequence well in order to maximize the use of their lands.

Problems and constraints to agriculture

1. A national-level problem is that policy-makers lack knowledge and appreciation of FSRE approaches. For
production, processing, marketing and utilization aspects in diversified agriculture, policies and programs that
will make efficient use of the agricultural and human resource base should be evolved.

2. Under the orthodox Marxism regime, agricultural research and extension usually follow the top-down
approach without feedback from farmers. Farmers themselves are often unaware of their ecological
environment so that they can manage their farming activities better.

3. The idea of biodiversification in sustainable agriculture and rural development (SARAD) is also new for most
Vietnamese scientists. They often do not understand basic scientific concepts in FSRE, and perhaps due to the
previous closed-door policy, did not have opportunities to be aware of or understand advances in FSRE,
approaches and other changes in the advanced period.

4. Within institutions in Vietnam integration of interdisciplinary studies is more difficult because scientist of
different disciplines hardly ever get together. This has caused overlapping, waste of time and money, and
 Page 4 of 17

inefficient use of human resources in agricultural research. A common voice in FSRE with unique approaches
and close linkage among universities and research institutes needs to developed for agricultural research and
extension in Vietnam.

Agricultural production problems

1. Low income from mono-rice production due to high production cost.

2. Under-utilization of farm resources: bio-diversification of crop-livestock-fish-farming system did not work
very well.
3. Water pollution in the ricefields/canals due to excessive application of agro-chemicals.
4. Low shrimp and fish yields, thus causing low economic profitability.

Research Highlights

Rice-fish systems

Rice-fish systems have been in practice at various sites: the state farms of Co Do and Song Hau, Cai Be, and
Long Phu, Vinh Long by Cuu Long In-Service Training College. The most common species are common carp
(Cyprinus carpio), Nile tilapia (Oreochromis niloticus), silver barb (Puntius gonionotus), silver carp
(Hypophthalmychthis molitrix), and rohu (Labeo rohita). In addition, the rice-wild fish systems (snakehead fish,
snakeskin gouramy, walking catfish) in Thoi Binh and Tran Van Thoi, Minh Hai by Can Tho University, and the
pig-fish systems in Song Hay State Farm showed an Increase in life production and net farm income as
compared to rice monoculture. They also help in keeping the environment clean by reducing application of
chemicals in the ricefield, in combination with integrate pest management (IMP) practices.

Field surveys were conducted at the Song Hay State Farm to evaluate technical and economic aspects of the
rice-fish mixed culture which was developed very fast at the state farm: from ten households iii 199 1, to 69
families in 1992, and more than 300 farms in 1993.

Average fish production is 117 kg/ha in six to ten months with average-net income from fish of US$79 in total
net farm income of US$442 per household.

Dang Kieu Nhan (1994) showed a high increase in total farm household income in 1994 in rice-fish systems
(US$1429) at Hau My Bac, Cai Be, Tien Giang, compared to US$872 in 1991. Incomes front, rice seem to be
the same while those from fish, fruits and other crops increased significantly, Integration of fish into rice farming
can make better use of farm resources, thus increasing farm income. Per capita income in rice-fish households
is also higher than that in monorice households.

The rice-wild fish systems were also investigated at four places in the Mekong Delta: Song Trem (Minh Hai),
Thanh Tri (Soc Trang), Phuong Ninh, and Song Hau (both in Can Tho). Generally, wild fish resource are low
and more limited from 1991 to 1994 (1947 kg/ha). Results also showed that wild fish production decreased at
Song Trem, Thanh Tri, and Phuong Ninh but increased at Song Hau State Farm due to better management on
dikes and trenches, and its accessibility to floodwater from the Mekong River.

Nguyen Anh Tuan and Bui Minh Tam (1994) analyzed data obtained in 1991 and 1992 from 54 farmers in the
U Minh areas, and found that the rice-wild fish systems in the region are generally extensive with low inputs
and low economic profitability. Due to saline water intrusion, farmers can grow only one rainfed rice crop.
Traditional local rice varieties with long growth duration usually give low yields (1.8-2.2 t/ha). Some kinds of
wild fish, especially snake-skin gouramy (Trichogaster Pectoralis) arc stocked into the ricefields, which yields
between 83 and 90 kg/ha. Marginal rates of return (MRR) of 1.8-2.3 show the low economic profitability of the
systems.

Integrated livestock-fish systems

Le Nhu Xuan and Nguyen Anh Tuan (1994) conducted experiments in 1992 and 1993 in the farm households
in O Mon, Can Tho. Five ponds averaging 1200 square meters each and with separate sluice gates and
pigpens were chosen for the experiment. A polyculture with stocking density of seven fish/sq m was applied
with Nile tilapia occupying from 41 to 60 percent. The rest were silver carp, rohu, common carp, silver barb and
catfish (Pangasius micronemus). The fish were fed on pig excreta and waste feeds which were washed out
daily onto the ponds. After 300 days fish gave high elides of 14.2-15,8 kg t/ha, of which marketable sizes were
 Page 5 of 17

283, 225 and 1340 g for Nile tilapia, silver and catfish. Silver carp, common carp, and rohu reached 353 g, 373
g, and 366 g, respectively. Total pig production was 1673-2493 kg after ten months of culture.

Rice-freshwater prawn (Macrobrachium rosenbergii) systems

The rice-freshwater prawns systems were established mostly in low-lying areas of the Mekong Delta: Phung
Hiep, Can Tho, Tam Binh, Vinh Long, where high value shrimps can significantly increase farm incomes.

At Dai Thanh (Phung Hiep, Can Tho) Le Canh Dung (1994) worked with 101 farmers cooperators, later
reduced to 48 farmers in 1993. Major rice varieties planted were IR 19660, MTL58, and MTL 107. Rice yields
were stable during the period of study (4.5-4.9 t/ha in the dry season, and 4.2-4.8 t/ha in the wet season)
except in the dry season of 1993 when yield was low due to brown plant hopper and rice ragged stunt disease.
Shrimp yields were moderately low: from 107 kg/ha in 1991 to 85 kg/ha in 1993, and particularly very low in
1992 (64 kg/ha) due to water pollution.

Although the return above variable cost (RAVC) and the benefit-cost ratio (BCR) were lower in 1993 as
compared to 1991, increased total farm income was recorded in 1993 (US$954) in rice- freshwater prawn
systems compared to 1991 (US$670). Income from rice was reduced but that from fruits, cash crops, and
livestock increased remarkably.

At Phy My, Long Phu (Tam Binh, Vinh Long) a combination of rice-fish-freshwater giant prawn was set up,
using marble (sand) goby (Oxeleotris marmoralus) and the new rice varieties: MTL110 and MTL119 which
gave 40-50pereent higher yields than the other varieties.

Integrated pest management (IPM) was also practiced in rice-fish culture.

Rice-fish-fruit trees

In 1993-94 with 11 farmer cooperators, Nguyen The Huy worked on two cropping models: first, rice-rice-
freshwater prawn-fruit trees; and second, rice-rice-marble goby-fruit trees.

Results showed that new rice varieties gave higher yields than concurrent ones like IR19660, 2001, etc.
Freshwater giant prawn and marble goby were integrated in the rice systems, and gave (farmers] better
incomes: in 1994 the rice-rice-marble goby systems gave the highest RAVC (US$865/ha) but its BCR was
lower than the rice-rice-prawn systems (1.11 vs 1.39).

Rice-saline water prawn systems in coastal areas

Rice-saline water prawn systems in coastal areas [were practiced] even in combination with traditional rice in
the wet season in coastal areas where tiger prawn (Penaeus monodon) and banana prawn (Penaeus
merguiensis) were raised in saline and brackishwater in the dry season: Vinh Chau and My Xuyen, Gia Rai,
Tra Cu.

Results from field surveys showed that a number of farmers practiced the systems on a total of 5,000 ha;
average shrimp production was about 244 kg/ha/crop. Net income was US$2200/ha/year of which two-thirds
was from shrimp farming -- higher than that of salt-shrimp (US$70), and extensive shrimp culture (US$145).
However, it was lower than that for semi-intensive shrimp monoculture (US$700).

Le Canh Dung (1994) conducted a study in 1993 in two districts -- My Xuyen and Vinh Chau [Soc Trang]) using
participatory rural appraisal (PRA) to understand the main constraints in traditional rice production in the rice-
saline water shrimp systems. There is a trend towards reducing traditional rice area in the systems due to low
productivity and gross return. The shrimp component is highly profitable. In Minh Chau with mostly traditional
rice and natural shrimp, RAVC/ha from shrimp in 1993 was more than four times compared to that from rice
(US$773 vs US$173 per ha) and BCR of shrimp was six times higher (31.5 vs 5.4). In My Xuyen with modern
medium rice and tiger prawn, RAVC from shrimp was about 50pereent hi-her than rice (US$618 vs US$390
per ha), but BCR seemed the same (3.8 vs 3.5).

However, as shrimp from natural sources seems to be more and more limited, it is necessary to maintain the
sustainability of rice-shrimp systems in Vinh Chau.

Nguyen The Huy et al (1994) worked with 13 farmers in 1993 at Cay Cong hamlet, in the tidal wetland areas of
Dong Xuan Village. Environmental factor such as hydrophysical and chemical characteristics in the ponds were
 Page 6 of 17

recorded in the dry season. Economic indicators were also obtained and computerized. During two continuous
growing seasons preliminary results seemed promising. Survival percentage and average weight of tiger prawn
were high. Higher prawn yields of 1.1-1.9 tons/ha were recorded at stocking rates of 4-7 prawns/m2 and high
feeding rates. Average RAVC was about US$5450/ha/year. Shortened growing period (3.5-4 months) can
enhance [yield as] farmers can grow tiger prawn twice in the dry season, as [they raise] artificial postharvest in
nursery ponds instead of collecting juveniles from rivers and canals.

Modern rice varieties like MTL119, MTL136 which have moderately high yields (4.5-5.0 tons/ha) are now
replacing traditional local varieties which yield only 1.2-1.8 tons/ha of paddy. Besides rice and prawn, farmers
now grow sapodilla on surrounding dikes.

Growing tiger prawn can help farmers increase their incline. However, the BCR is still low compared to the rice-
freshwater prawn system. Thus the tiger prawn system needs more capital investment. Results show that
traditional local rice gave very low yields, especially without shrimp culture. Traditional rice alone gave negative
RAVC while those integrated with shrimp gave better RAVC. The new MTL136 gave the highest yield and
RAVC (5 tons/ha paddy and US$564/ha). Traditional rice varieties need to be replaced by appropriate modern
varieties to get higher yields. Moreover, a combination of rice and shrimp culture can help farmers earn more
incomes than monoculture rice.

Deepwater rice farming systems (DWR)

Deepwater rice research has been done under collaboration between Can Tho University and IRRI since 1990.
Data from three field surveys and a number of experiments showed that low productivity in deepwater rice
areas can be turned to more profitable cropping patterns by growing other upland crops after deepwater rice,
or changing to double rice systems with embankments to prevent floodwater from [entering] the Mekong River
systems. Maize is most grown in DWR areas, followed by watermelon, mungbean, cowpea, sesame, melon,
etc.

In newly-irrigated areas, modem rice (ILR) has now replaced DWR. The two modem rice crops give a gross
return of around US$900/ha, nearly four times higher than DWR alone ($242). However, their gross margin is
only thrice as much (US$148 vs US$54). On the contrary, labor productivity in DWR is higher than in modem
rice (US$2.4 vs US$1.66 per manday).

Investment on irrigation for the production of ILR is beneficial to society. Operation and maintenance costs of
canals and irrigation systems are estimated at 0.59 tons of paddy per ha of land, and the increase in crop
return added is estimated at 2.35 tons of paddy per ha, about four times of investment costs; thus the high
rates of return.

The impact of the change in rice production is very clear when farmers get more than 10 t/ha of paddy as
compared to 2.0-2.5 t/ha of DWR. However, gross return in irrigated lowland rice (ILR) areas is only 50 percent
higher, because the price of DWR is 20 percent higher, and farmers can earn incomes from upland crops after
DWR, accounting for about US$367/ha in 1993. Results show that labor and capital productivity in ILR areas
are lower than those in DWR areas. However, gross margin and family income are 26 percent and 27 percent,
respectively, higher than in DWR. The rate of return on capital in ILR is 0.20; and in DWR, 0.25. Analysis
however shows that technical efficiency in the use of land, labor and capital is 16 percent higher in ILR
compared to the DWR ecosystem.

The change also has a positive effect on the rural non-farm sector. The higher the amount of market surplus
from rice, the more generated employment and incomes in rural trade, transport, and other services.

Average household income from off-farm and non-farm activities is about 67 percent higher in ILR than in DWR
areas.

Rainfed lowland rice (RF)

Nguyen Ngoc De et al (1994) reported that rice farming in rainfed areas still have several problems. Moreover,
farmers' living standards in those areas have not improved. Surveys were dong in April 1993 in the provinces
of Long An, Ben Tre, Tra Vinh, Soc Trang Kien Giang, and Minh Hai to understand existing rice farming
systems, and identify the major problems in rice production in those areas, using participatory rural appraisal
(PRA) techniques. A total of 298 farmers in nine different agro-ecosystems were involved in the study.

Traditional rice is the dominant crops in the rainfed areas of the Mekong Delta, the crop is characterized by low
yields and susceptibility to insects/pests. In areas where maximum water level is less than 40cm, some
 Page 7 of 17

medium rice varieties like IR42, MTL83 are beginning to replace traditional varieties. Improved cultivation
practices (such as direct dry seeding, raised beds with shallow drains, polder to store fresh water) can help
farmers to grow one more rice crop early in the wet season, or one upland crop. In some areas, farmers can
grow rice integrated with tiger prawn in the dry season, and even with freshwater prawn in the wet season.
Major problems are lack of fresh water in dry season, excess water in the rainy season, lack of capital for
investment, lack of advanced technologies. New rice varieties appropriated to each agro-ecosystem, in
combination with advanced techniques, credit, and efficient agricultural extension, are necessary in the rainfed
areas of the Mekong Delta.

Field survey in rainfed areas of the Mekong Delta

About 39 percent of the rice area in the Mekong Delta are rainfed. In recent years, the increase of agricultural
products has been partly attributed to diversification of these rainfed rice farming systems. Field surveys were
carried out (with 181 farmer respondents) in two ecosystems: (1) shallow rainfed (0-30 cm) and (2)
intermediate rainfed (31-60 cm).

Four major farming systems were identified: (1) one monorice crop, (2) rice-shrimp. (3) rice-upland crops, and
(4) rice-fish. The rice-shrimp and rice-upland crops systems are prevalent ill shallow, water areas, while tile
rice-fish system is practiced in intermediate areas. Shrimp, fish and upland crops are major sources of income
for farmers. Income from rice is usually lower than other components like shrimp, fish and other upland crops,
with income from watermelon noted as the highest (US$1320/ha). Farmers growing rice-fish have the highest
household Income (US$1700/year), while those with one monorice get only US$255 year.

Production function and optimized inputs for rice, shrimp, fish, and upland crops are also studied. For optimum
inputs, results indicated increasing seed rates. N, and P, with. However, lower rates of nitrogen application in
rice-upland crops. For shrimp and fish, the current inputs of fries and juveniles, feeds, arid others are
underutilized. It is advisable to increase N arid P for mungbean and watermelon, and decrease chemicals arid
labor inputs.

Other rice-based farming systems

Experiments conducted oil the improvement of cropping systems (Nguyen Xuan Lai, 1994) in Vinh Chau and
Soc Trang showed OM269 and IR64 as two promising rice varieties for the areas. Rice can replace sweet
potato and pachyrrhizus in the wet season due to its technical feasibility with low material and labor inputs.
Rice can give higher economic return than sweet potato arid pachyrrhizus. In addition, the two varieties
mentioned call leave considerable amount of rice straw (equivalent to 500 kg of paddy) that farmers have to
buy for mulching [in growing] Chinese onion.

Ta Quoc Tuan (1994) established five rice-upland crop models at Thanh Thang, Thot Not, Can Tho in 1991-92.
Other [studies dwelt oil] yield trials of new rice varieties. NPK effects oil peanuts and economic analysis of farm
households. (It was found that] of upland crops after rice (of intercropping among upland crops) give higher
production costs and gross returned their tile double rice systems. The peanut (spring-summer)-kenaf
(summer-autumn) cropping gives the highest RAVC (US$347/ha). Using 100 kg N/ha call increase significantly
yields of shell peanut up to three ha. The high-yielding varieties OM90-2 and OM90-9 (which are resistant to
brown leathopper) can perhaps replace the old rice variety CL7.

Average annual farm household gross return is around US$1545 of which agriculture accounts for 80 percent,
with 55 percent from rice production.

Agroforestry and agroforestry-fish system

Agroforestry and agroforestry-fish systems were set up at An Bien experiment station, Kien Giang, Hoa An
experiment station, Can Tho. The models introduced a combination of agriculture and forestry in order to
increase farm household incomes. Nguyen Duy Can (1993) reported that at An Bien and Hoa An stations,
Melaleuca leucadendron and Eucalyptus grow very well. Local fish like snakehead (Channa striata), snakeskin
grouramy (Trichogaster Pectoralis), walking catfish (Clarias sp.) are harvested after the flooding period, and
can give additional incomes to farmers. By these systems reforestation can be done, with the recovery of some
kinds of natural resources, meanwhile utilizing farm resources in better ways.

Case Studies
 Page 8 of 17

An Minh is located in the rainfed areas of Kien Giang province; almost all farmers cultivate by traditional
practice. Meanwhile, Phu Tan is located in irrigated areas of An Giang province, with farmers practicing
modern techniques.

The farm household size is seven for traditional and six for modern farms. Available labor force is four per
household, both for two locations. In the traditional-practice area, the a-e of the household head is 52 years;
and 45 in modern practice area; years in school is three for traditional and four for modern farms.

Farm size is 1.8 hectares for rice. 1.2 hectares for other crops in An Minh, and 0.95 hectare for rice in Phu Tan.

The farming systems used in An Minh and Phu Tan are indicated below.

An Minh Traditional rice + Sugarcane + cowpea + cucumber + fish + ducks, chicken +

rarabao, pigs

Phu Tan Wet and dry season rice (modern variety) + ducks, chicken, pigs

Under the traditional practice, farmers use transplanting method; while in the modern practice farmers use
direct seeding. Both farmers use mechanical means for land preparation, and mechanical thresher irrespective
of the system. The major difference between the two ecosystems is in the use of agrochemical (fertilizer,
pesticides, herbicide, etc.)

Total rice production per hectare per year is 10.8 tons in the modern practice (wet season, 4.9 tons and dry,
season, 5.9 tons), compared to 2.8 tons in traditional practice and 8.7 tons of paddy of traditional practice and
other crops equivalent to rice.

Farmers in the irrigated ecosystem use chemical fertilizer and pesticides — many times more in irrigated
modern rice than in the, cultivation of traditional rice.

Two rice crops for modern practice give more farm income than traditional rice monoculture. However, to
obtain such a high farm income, farmers had to incur input costs about such times more than in traditional
practice. This additional investment may not be easy for small rice firms as access to institutional credit is not
adequate for poor farmers who may want to use the modern practice. It is evident that poor farmers usually get
low yield for irrigated lowland rice due to lack of capital for material inputs. They usually borrow cash from
private moneylenders or private dealers at very high interest rates, so they may gain less from the chance in
the system than the relatively high income farmers.

Of the modern practices in irrigated lowland rice ecosystem, the adoption of modern varieties is associated
with increased use of labor and capital in rice production. Thus, the shift from traditional to modern practice
with modern varieties in the deepwater areas significantly increases income-earning opportunities, [especially]
with the use of the relatively abundant resource — human labor. Although the share of hired labor income is
higher in villages that widely adopt modern varieties, family inputs — mostly for land preparation, care of crops,
and other fan-n management activities — have also increased.

The change has also had an effect on the rural non-farm and off-farm sectors. The higher amount of marketed
surplus from rice, and the expansion of the market for non-farm goods and services have generated additional
employment and incomes in rural trade, transport and services sectors. Average household income from non-
farm and off-farm activities is higher in modern practice than in traditional practice.

However, economic profitability of households with one to two modem rice crops is not as good as that with
sole traditional rice. Due to lack of freshwater supply for rice in the dry season, not all the traditional rice areas
in the Mekong Delta can be converted to modem rice practice with high input. In fact, some areas have low rice
yields because of adverse soil conditions. In those areas farmers can practice some kind of agro-forestry
farming systems, or they can integrate aquaculture. This system is much more sustainable than the high input
intensive double rice crop culture practiced in the irrigated areas.

Sustainability of the System

The integrated fanning system practiced has been evolved by perfecting several [systems] followed by trial-
and-error. The greater change the farmer has brought about in the farm is the elimination of-chemicals..
Another noteworthy practice in the integration of livestock into the system so that could get more manure, value
addition to biomass production, and above all, more profit from the system.
 Page 9 of 17

Ecological sustainability

The land use system in the homestead and the rice area close to the farm is intensive. The practices are highly
eco-friendly and economically very stable. The successful rice-fish culture practice attests to the claim.

Ecosystem health indicators

The modern practice of intensive use of fertilizer and pesticides for rice has brought a great change in the
farming system. The farmer in the traditional practice has intensified is compost production in the farm by using
all carabao dung, goat droppings, dry leaves and farm wastes. The green manure collection is an advantage to
the farmer.

The introduction of ducks also helps keep the, pest population under check. Another method followed is to
leave a patch of rice in the center of the farm during harvesting. According to the farmer, all the pests gather on
this patch, which patch is burnt later, killing the pests. Ecosystem health indicators such as spiders and
predators are promoted in the farm.

Economic sustainability

The farm's economic sustainability is encouraging. The farmer has good market contacts and all the produce is
sold to his advantage. Since family labor alone is engaged, cost of production is very low. That they are not
applying for any farm loan is an indication of a healthy economic condition.

The sale of fingerlings, duck meat and eggs, and goats is done in such a manner that it generates good
income. It is time for the farmer to reduce his physical strain on farm jobs. Even now, they collect green
manure, plough the fields without a break, and manage every activity. Such involvement forces them to keep
the new dry lands unused or underused. The investment made on the new land may not give enough return
and now remains unproductive.

Social sustainability

The social acceptance received by the systems is commendable. Farmers in villages and adjoining areas treat
the systems as bold experiments, serving as their guidelines. They provide not only the details of their farming
methods, but also share their inputs, as well as their trade secrets (the "secrets of their success") with fellow
farmers. They invite local authorities and development agencies to bring to their notice their local needs. The
skills-sharing experience has brought great demand for their produce and they find the demand more than they
could satisfy. The patronage from the community is encouraging so that farmers exert greater efforts to
improve their farming methods.

The farmer's responsibilities to his family, especially his school-going children, will increase sharply, and
providing one's children with opportunities for good education is part of his targets.

There are great possibilities for expanding farming to the newly acquired lands. Farmers should also introduce
agro-forestry so that at a minimum investment and management, reasonable returns can be obtained in the
long run.

Conclusions

1. Farmer's preferences for low-input systems vary considerably, depending upon the phase of crop
production.
2. There is considerable interest among farmers in low-input approaches for most phases of crop
production; 40-80 percent of conventional farmers expressed interest in low-input practices.
3. Although overall levels of interest in low-input methods vary by phase of crop production, preferences
for six of eight items were sufficiently inter-correlated so that one can identify an underlying, uni-
dimensional construct of preference for low-input practices.
4. The major antecedent of preference for low-input practices was the farmers' level of concern about
pollution, followed by support for agricultural research.
5. Preference for low-input practices did not bear a major relationship to general or agriculturally-related
political ideology, attitudes toward the farm crisis or federal commodity programs, and profit orientation.
6. There was no strong pattern for small or part-time farmers to be a disproportionately large constituency
 Page 10 of 17

for low-input practices or for large farmers to be disinterested in low-input approaches.

7. The pattern of results was consistent with established agro-ecological knowledge to indicate that the
production practice preference items have prima facie construct validity.

Findings show that the change has led to a decline in productivity of labor and capital. fit rainfed lowland
ecosystems, the farmers income, net of non-labor cost, was US$2 per day of labor, while in the irrigated
ecosystems, US$1.8 per day of labor.

Research towards the development of component technologies and cropping patterns (including water
management, fertilizer use, control of insects and diseases) must be strengthened. Questions are varied about
the sustainability of intensive rice culture in irrigated systems, although in the short run they are profitable
compared to existing alternatives available to farmers. It may be useful to establish farm trials on
representative land types to examine long-term soil fertility and pest management of intensified rice-based
farming systems, to address the issue of sustainable agriculture.

Table 1. Rice production in the Mekong Delta, Vietnam

1975 1980 1985 1990 1992 1994

Area (1000 ha)

Deepwater rice(*) 1256 1407 1099 863 724 655

Modem rice (dry season) 156 424 542 839 1055 1452

Modem rice (wet season) 297 457 571 852 938 1342

Total 1709 2288 2212 2554 2717 3449

Average yield (t/ha)

Deepwater rice(*) 2.03 2.14 2.47 2.78 2.14 2.85

Modem rice (dry season) 2.87 2.41 4.05 4.62 4.93 5.15

Modem rice (wet season) 2.34 2.21 3.54 3.57 3.44 3.85

Average 2.41 2.25 3.35 3.66 3.50 3.95

Production (1000 tons)

Deepwater rice(*) 2550 3011 2715 2399 1549 1867

Modem rice (dry season) 448 1022 2195 3876 5201 7477

Modem rice (wet season) 694 1010 2021 3041 3227 5167

TOTAL 3692 5043 6931 9377 9977 14511

Source: Rice production statistics

* Includes areas of floating rice.

Table 2. Socio-economic characteristics of selected DWR

and ILR systems in the Mekong Delta, Vietnam, 1993.

Item Deepwater Irrigated

No of sample farms 250 300

Farm size (ha) 1.8

Tenure (percent)

Owner-operator 96 95

Leaseholder 2 2
 Page 11 of 17

Other 2 3

Household size 7

Main labor/household 5 4

Farm characteristics

Age (household head) 52 50

Schooling 4 5

Years in farming 30 25

Rice cropping intensity (percent) 97 98

Total cropping intensity 148 199

Source: Mekong Delta Farming System R&D Center, Can Tho University and
South Institute of Agricultural Planning.

Table 3. Adoption of biological, mechanical and other labor-saving technologies in selected DWR and
ILR systems in the Mekong Delta of Vietnam, 1993.

Item Deepwater Irrigated

Yield (t/ha) 2.7 4.8

Fertilizer use

percent area

N (kg/ha) 10 120

P (kg/ha) 8 62

K (kg/ha) 0 20

NPK (kg/ha) 18 202

US$/ha 7 90

Pesticides use

percent area 30 99

US$/ha 2 30

Herbicides use

percent area 25 80

US$/ha 1 6

Engine pump use (percent area) 1 100

Electric pump use (percent area) 0 0.8

Tractor adoption (percent area) 75 80

Thresher adoption use (percent area) 98 99

Direct seedling (Percent area) 40 98

Source: Mekong Delta Farming System R&D Center,

Can Tho University and South Institute of Agriculture Planning.

Table 4: Labor use (manday/ha) in selected DWR and ILR systems

in the Mekong Delta of Vietnam, 1993
 Page 12 of 17

Item Deepwater Irrigated

No of farms 250 300

Pre-harvest

Repair and cleaning of dikes 2 7

Plowing and harrowing 2 2

Levelling 1 2

Puddling 1 2

Broadcasting and transplanting 5 8

Fertilizer, herbicide and pesticide 1 7

application

Hand-weeding 6 25

Irrigation and draining 1 5

Supervising 1 2

Harvest and post-harvest

Harvesting 20 26

Threshing an hauling 5 7

Drying an storing 3 10

TOTAL 48 103

percent hired labor 53 40

percent female labor 35 42

Source: Mekong Delta FAMN System R&D Center,

Can Tho University and South Institute of Agriculture Planning

Table 5. Benefit-cost analysis of rice-fish systems at Co Do state farm, Can Tho, 1991-1993.

Patterns by year Total Cost Gross return Net return Net (in US$) Benefit-cost
ratio

(in thousand VND per ha)

DMR-WMR

1991 2794 7362 4568 351 1.63

1992 4288 9680 5392 490 1.25

1993 5013 9034 4021 383 0.80

DMR-WMR with
fish

1991 2949 8296 4347 334 1.47

1992 4346 9224 4878 443 1.12

1993 5360 9879 4519 430 0.84

Source: Mekong Delta FSRD Centre's Rice-Fish Project

Exchange rules:
1991 - 1 USD = 13,000 VND;
1992 - 1 USD = 11,000 VND;
1993 - 1 USD = 10,500 VND;
 Page 13 of 17

Table 6. Structure of household income in rice-fish farming compared to monorice at Hau My Bac, Tien
Giang (USD)

Activities Rice-Fish Monorice

1990-91 1993-94 1990-91 1993-94

N percent N percent N percent N percent

On-Farm 784 90 1356 95 367 68 682 85

Rice 496 57 521 38 325 62 467 58

Fish 113 13 549 38 35 4

Animal 105 12 89 6 21 4 26 3
husbandry

Fruits and 70 8 197 14 21 4 154 19

others

Off-farm 78 9 45 3 119 23 19 2

Non-farm 10 1 28 2 40 9 98 13

Total income 872 1429 526 799

Per capital income 145 238 137 160

Table 7. Costs and returns analysis or rice and wild and stocked fish at 4
locations in the Mekong Delta (1 000 VND).

Component Song Trem Thanh Tri Phuong Ninh Song Hau

Stocked fish:

Total cost 654

Gross return 1283

Wild fish:

Total cost 50 50 50

Gross return 161 365 382 373

Rice

Total cost 1009 3766 2689 5605

Gross return 1512 7571 5539 9120

Tax 121 140 160 399

Total costs 1180 3956 2899 6658

Gross returns 1673 7936 5921 10776

Net return 493 3980 3022 4117

BCR 0.42 1.01 1.04 0.62

Table 8. Structure of farm household income

in rice-shrimp farming systems, 1991-93 at Dai Thanh Village, Can Tho.

Source of income 1991 1993

 Page 14 of 17

Amount (USD) percent Amount (USD) percent

On farm:

Rice cultivation 220.8 33.0 141.7 15

Shrimp 275.0 41.0 344.1 36

Fruit and cash crops 74.4 11.0 258.7 27

Animal husbandry 37.1 6.0 151.1 16

Off-farm 63.4 9.0 58.9 6

Total income 670.7 954.5

Source: Le Canh Dung (1994): Rice-freshwater

prawn farming system in Phung Hiep, Can Tho.

Table 9. Cost and returns in rice farming in An Giang, 1992-93 (per ha of land)

Item Unit Floating rice Irrigated rice

Wet season Dry season

Yield tons 2.4 4.9 5.9

Labor manday 48 84 98

Fertilizer NPK kg 27 195 219

Cash expenses US$ 103.4 240.5 249.1

Material input 27.6 124.2 145.6

Hired equipment 23.9 57.4 48.4

Hired labor 34.2 40.8 37.0

Land tax 17.7 18.1 18.1

Non-cash expenses US$ 85.1 123.2 134.2

Family labor 22.8 53.1 61.4

Family equipment 0.9 7.0 7.4

Interest charges 61.4 63.1 65.4

Total Cost US$ 188.5 363.7 383.3

Gross return US$ 242.5 406.0 488.9

Gross margin US$ 54.0 42.3 105.6

Family income US$ 139.1 165.5 239.8

Production cost US$ 79 74 65

Labor productivity US$ 2.30 1.62 1.70

Source: Can Tho University-IRRI farm households survey (1994).

Table 10. Structure of household incomes in deepwater and irrigated ecosystems,

Mekong Delta, 1990-91 and 1992-93 (US$).
 Page 15 of 17

Source of income Deepwater ecosystem Irrigated ecosystem

1990-91 1992-93 1991-92 1992-93

On-farm 409 452 521 504

Cultivation of rice 181 145 459 445

Cultivation of other crops 151 199 3 5

Fisheries and livestock 77 108 59 54

Off-farm 42 39 64 70

Hiring out of labor 36 29 52 56

Hiring out of capital and equipment 6 10 12 14

Non-farm activities 68 74 121 118

Total household income 519 565 706 692

Household size 7 7 7 7

Per capital income 74 81 101 99

Source: Can Tho University-IRRI farm households survey (1994).

Table 11. Cost and returns analysis of tested models at

An Bien, Kien Giang, 1991 -1992 ('000/ha VND)

Model Gross returns Total cost RAVC BCR

Mono traditional rice 1020 815 205 0.25

Traditional rice + fish 2655 905 1750 1.93

Traditional rice + fish +duck 2925 1106 1819 1.65

Sugarcane + fish 3322 1265 2057 1.63

Table 12 Annual farm budget for traditional Practices in the

Mekong Delta of Vietnam

Item Value (US$)

Per farm Per hectare

Receipts

Total cash receipts 1566 800

Total non-cash receipts 630 350

Total receipts 3196 1220

Expenses

Cash costs

Seeds bought 25 14

Fertilizer 20 11

Pesticides 11 6

Herbicides

Hired labor 44 43

Hired tractor 45 25
 Page 16 of 17

Food for laborer 18 10

Fuel and oil 5 3

Irrigation fee

Tax 61 34

Land preparation 16 9

Interest on loans 18 10

Total cash costs 297 165

Non-cash costs

Depreciation 72 40

Unpaid family labor 216 120

Services paid in kind 144 80

Total non-cash costs 432 240

Total costs 729 405

Returns above cash costs 1890 1050

Returns above non-cash costs 1764 980

Net farm income 1467 815

Table 13. Annual farm budget for modern practices in the

Mekong Delta of Vietnam, 1995

Item Value (US$)

Per farm Per hectare

Receipts

Total cash receipts 1767 1860

Total non-cash receipts 190 200

Total receipts 1957 2060

Expenses

Cash costs

Seeds bought 33 35

Fertilizer 141 148

Pesticides 83 87

Herbicides 14 15

Hired labor 142 149

Hired tractor 28 30

Food for laborer 21 22

Fuel and oil 11 12

Irrigation fee 65 68

Tax 90 95

Land preparation 24 25
 Page 17 of 17

Interest on loans 29 30

Total cash costs 681 716

Non-cash costs

Depreciation 95 100

Unpaid family labor 123 130

Services paid in kind 171 180

Total non-cash costs 389 410

Total costs 1070 1126

Returns above cash costs 1277 1344

Returns above non-cash costs 1482 1650

Net farm income 887 934