FAO Investment Centre
Socio-economic and Production Systems Studies

  Le Centre d'investissement
de la FAO
tudes des systmes socio-conomiques et productifs

 
El Centro de Inversiones de la FAO
Estudios de los sistemas socioeconmicos y productivos


 

INDONESIA

SOUTH KALIMANTAN AGRICULTURE AREA DEVELOPMENT PROJECT

 

Objectives of the Project

The Province of South Kalimantan (KALSEL) is the smallest of Kalimantan's four provinces with some 2.9 million inhabitants living in 37,660 km2. The incidence of poverty in the province is 18.6% compared with a national average of 13.7%.

The overall objectives of the project are to: (a) alleviate poverty through rural development in upland agricultural areas in some of the province's poorest villages; (b) generate economic growth in these areas; and (c) improve the conservation of natural resources by assisting the intensification and diversification of agricultural systems.

This study was carried out by the FAO Investment Centre. The field work was undertaken in the 5 kabupatens: Tabalong, Hulu Sungai Utara (HSU), Hulu Sungai Tengah (HST), Hulu Sungai Selatan (HSS) and Tapin for four weeks. The social assessment study covered 20 villages, purposively selected as representative of the range of agro-ecological zones, socio-economic conditions and farming systems present in the project area. Field work used a modified version of the Farming Systems Diagnosis Method. The core team split into two sub-teams, each of 4-5 members, to which each kabupaten added 4-5 trainees, selected from the technical staff of agricultural planning, food crops, estate crops, livestock and fisheries.

Each sub-team visited a total of 10 villages, spending 2 days per village. In total, the study teams spent 5 days in each kabupaten, undertaking Participatory Rural Appraisal (PRA) in four villages. In each kabupaten, the field work was followed by cross-cutting analysis and presentation of findings at a debriefing held with members of the kabupaten project working group (pokja tingkat II). After completing the field work, the results of a cross-cutting analysis of data from the 20 villages, 5 kabupatens and 140 household interviews was presented to the members of the Provincial Project Task Force (Pokja Tingkat I).

 

PROJECT AREA BACKGROUND

Population and Administrative Units

Administratively, South Kalimantan province is sub-divided into 9 rural districts (kabupatens), 109 sub-districts (kecamatans) and 2,169 villages (desas), within which Government has designated upland areas of 5 adjacent kabupatens as the priority project area: Tapin, Hulu Sungai Selatan (HSS), Hulu Sungai Tengah (HST), Hulu Sungai Utara (HSU) and Tabalong. The five project kabupatens have a population of nearly one million living in an area of 11,300 km2 (See Map).

The project area forms a coherent geographic and social unit united by a common history. The first Banjarese settlement began in the Hulu Sungai (literally "upper river") and gradually extended northward to Tabalong and south to Tapin in response to population pressure. The 5 upland districts contrast sharply with the four lowland districts of the province, which are sparsely populated but have experienced rapid population growth as a result of transmigration. Most projects in recent years such as transmigration schemes, plantations and forest concessions have focused on the lowland districts, leaving the upland districts in a state of relative neglect.

Hulu Sungai Tengah has the largest population, accounting for 34% of the total, and Tabalong the smallest, with 11% of the total. Population density varies widely between agro-ecological zones, ranging from a high of over 200 persons per km2 in parts of HSS and HST to a low of 12 persons per km2 in Tabalong.

Kabupaten

Population

% of Total

Tapin

60161

20%

Hulu Sungai Selatan (HSS)

60507

20%

Hulu Sungai Tengah (HST)

103546

34%

Hulu Sungai Utara (HSU)

45947

15%

Tabalong

32315

11%

All Project Kecamatans

302476

100%

Each of the 435 administrative villages in the 20 kecamatans is composed of a number of smaller natural villages. The number of hamlets per desa ranges from one or two in the densely-settled sawah areas to five or more in the mountains. There are, on average, 4,000 households and 15,124 persons per kecamatan and 184 households (695 persons) per village Between 1980 and 1990, population grew at a rate of 1.13% per annum.

The sex ratio of the population of remote villages shows an excess of females over males of working age, suggesting out-migration. Conversely, in urban areas there is an excess of males aged 20-45. Historically, since most communications were by boat, the main settlements grew up along the two main rivers (Barito and Martapura). With the construction of a tarmac road running south to north through the project area, dense settlements have also grown up along the road. Spontaneous settlements in the hills and mountains tend to be on the edge of fertile valley bottoms.

Social Background

The project area has a number of important characteristics that set it apart from the rest of Kalimantan:

Social Indicators

Rural literacy rates (84%) in South Kalimantan are well above the national average (75%). However, there is still a major gender gap (17% of the females are illiterate compared with only 8% of the males). The tendency for females to be less literate than males may be connected with the strong Islamic traditionalism of the local Banjarese society.

Over 97% of the rural population is Moslem, 1% is Catholic, 1% is Buddhist and the remaining 1% of "other" refers to the indigenous Dayak Kaharingan religion. Average age of women at first marriage is relatively low: 5% are married by age 13, 15% by age fourteen, 27% by age fifteen, 37% by age sixteen, 48% by age 17 and 67% by age nineteen. The average woman aged 50 and above has given birth to 5-6 children, but birth rates are slowing down in the younger generation. Average household size in rural areas is four members and is less than four in some of the more densely-populated kecamatans.

Infrastructure

In addition to the main highway, there is a dense network of kecamatan and desa roads. Nearly all villages are now accessible by road, although some remote upland Dayak villages are still accessible only by footpath or by motorbike. In these areas, many new roads were constructed in early 1997.

Most villages in the lowlands and foothills have been electrified, whereas in the mountains, the remote villages still have no electricity. Private generators have been installed by wealthy individuals and sometimes there is a collectively-owned generator belonging to the village. Even in electrified villages, few poor households have electricity because they cannot afford it.

Land Tenure Status

The last agricultural census was done in 1983 and provides data on size of holdings only at provincial level. In the Province as a whole, 36-37% of the holdings are under 0.5 ha, 27% are between 0.5 and 1 ha, 30% are between 1 and 2 ha and 8% are above 2 ha. The bottom 37% of the farms controls only 7% of the land area whereas the top 8% controls 52% of the land.

Income and Expenditure

According to the latest Susenas household income and expenditure surveys, in 1996 average income per household in project kabupatens ranged from Rp3.2 million in HSU to Rp4.3 million in Tapin. Expenditure ranged from Rp2.7 million per annum in HSU to Rp3.8 million in Tabalong. These figures give a false impression of affluence because the top decile raises the average.

Average Income and Expenditure in Project Kabupatens, 1996

Rp per Household

Tapin

HSS

HST

HSU

Tabalong

Income: per month

361524

300704

322436

263328

315804

per annum

4338288

3608448

3869232

3159936

3789648

Expenditure: per month

312832

276460

288916

225368

317900

per annum

3753984

3317520

3466992

2704416

3814800

Source: elaborated from per capita monthly figures in Susenas: Laboran Ringkas Susenas Modul Konsumsi 1996 Kalimantan Selatan, June 1997.

Geographic Distribution of Poverty

The incidence of poverty in South Kalimantan is 18.6% compared with a national average of 13.7%. The official poverty line for South Kalimantan is currently Rp415.260 per household per annum for rural areas. This corresponds to the cost of 2,100 calories per person per day based on a basket of 26 consumer goods. The incidence of poverty is highest in HSU (28%) and lowest in HST (8%). In HST and HSS, the percentage of population below the poverty line is below the national average whereas in HSU and Tabalong it is well above both the national and provincial average.

Poverty Line 1996 (Rp)

Tapin

HSS

HST

HSU

Tabalong

Total

% below poverty line

15.50%

11.74%

8.06%

27.79%

21.00%

17.46%

Population below poverty line

19922

21981

18026

77574

32232

169736

HHs below poverty line

4981

5495

4507

19393

8058

42434

Source: elaborated from per capita monthly figures in Susenas: Laboran Ringkas Susenas Modul Konsumsi 1996 Kalimantan Selatan, June 1997.

 

FARMING SYSTEMS BY AGRO-ECOLOGICAL ZONE

A number of farming systems were identified in the project area, by agro-ecological zone, starting from uplands to lowlands.

Zone and Farming System

Villages

Dayak upland swidden system: upland rice intercropped with banana, maize, squash, cucumber and cassava with jungle rubber on fallow

Aniungan, Ulang, Kindingan RT4, Pipitak Jaya

Banjarese upland systems: upland rice/groundnut rotation + rubber

Padang Sari RT1 3 & 4, Kindingan RT 1-3 Pipitak Jaya Rt1, Puyun, Auh, Batu Panggung, Santu'un, Rantau Bujur RT 1-3, Hamak Utara

Transitional Javanese Semi-intensive Systems: sawah + ladang + cattle

Rantau Bujur RT 4 & 5, K Habang Lama, Padang Sari RT 1 & 2

Transitional Extensive System (Banjarese): sawah + ladang + rubber

Tambak Pipi'i, Garagata

Lowland Sawah/Rubber System (Banjarese)

Jimamun, Mungkur Palawan, Buruk Balik, Batu Kijang, Jarau

Lowland Sawah/Surjan System (Banjarese)

Baru

Five of these are distinct farming systems, each with its own problems, priorities and development potential. The transitional Banjarese extensive paddy-upland rice-rubber system is not a distinct farming system, but a combination of elements of the Banjarese upland rice/rubber/groundnut farming system with those of the paddy/rubber system in the sawah-zone. The upland Dayak swidden (slash and burn) system is found on secondary forest on the slopes of the Meratus mountains and is practiced by an indigenous ethnic group. Labour requirements are very demanding and often done by large groups of people using rotating exchange labour. The upland Banjarese system is based on the farming of alang-alang grasslands in the foothills of the Meratus mountains. All farm operations are done manually, usually with family labour, and little use of exchange labour. The transitional zone includes both upland and valley bottoms, combining elements of upland and sawah-based systems; the paddy/rubber system is found in the flat areas along the main roads and rivers; and the sawah/surjan system is found on the edge of the inland swamps.

The unifying factor in five of the six systems is that rubber is the main cash income source and rice is grown for food security. The lowland paddy/surjan farming system in Labuan Amas Selatan (LAS) of Hulu Sungai Tengah kabupaten differs from all the others in that no rubber is grown. This area - although poor - may not be compatible with the World Bank definition of the project area (focused on upland farming systems) because the main opportunities lie in citrus, rehabilitation of larger irrigation schemes and harnessing of inland swamps for opportunistic cropping of palawija crops in the dry season.

Summary Typification of Farming Systems

Parameter

Upland Systems

Transitional Systems

Sawah Systems

 

Dayak

Banjarese

Banjarese

Javanese

Paddy/rubber

Surjan

Intensivity

extensive

extensive

Extensive

Intensive

semi-intensive

Intensive

Animal traction

nil

nil

hired in

major

nil

Nil

Manure use

nil

nil

Nil

slight

nil

Nil

Crop residue use

nil

nil

Poultry

cattle

ducks

Poultry

Cropping

1 year in 7

2 years in 6

1 yr. in 2-3

Continuous

continuous

Continuous

Double cropping

nil

groundnuts

Groundnuts

rice

rice 5%

rice 10%

Fertilizer use

nil

5% of HHs

Some

high

moderate

Moderate

Off-farm

nil

some

Major

some

major

Major

Role of rubber

moderate

major

Major

Immature

major

Absent

Livelihood Systems by Farming System

The livelihood analysis is based on information obtained from over 140 household interviews undertaken in villages by members of the study team. The information presented below focuses on the main similarities and differences in livelihood strategies by farming system and by socio-economic stratum (wealthy, moderate and poor).

Upland Dayak Farming System

The Dayak farming system is land rich but poor in cash and consumer goods. In Dayak villages, one of the main features that distinguish wealthy households from the poor is cash earnings from bananas, which account for nearly 50% of the total income earned by wealthier households. For moderate and poorer households, rubber is the main source of cash income, either through own production or share tapping, whereas the rubber income earned by wealthier households comes from rubber trading. Wealthier households also differ from the lower stratum in terms of the amount of land owned, the number and type of off-farm activities (high percentage of income from salaries, trading and rice milling), type of housing and ownership of other assets such as a motorbike. Chickens are kept in small numbers by most households except the poorest.

In terms of wealth distribution, the average upper stratum household, who represents just 6% of the population, earns over Rp5 million per year. Moderate households, who constitute less than 30% of all households, earn less than Rp1 million per year. Poorer households, who are the majority (nearly 70%), earn less than Rp500,000 per year, which puts them at the GOI poverty line.

Upland Banjarese Farming System

In the Upland Banjarese system, most farmers depend on rubber as their primary source of cash income. For poorer households, 80% of total income comes from rubber share tapping or own production. Although the upper stratum households are more diversified and earn part of their income from groundnut sales, coffee or off-farm activities, rubber still constitutes 60% of the income of moderate households and 40% of that of wealthier households. As in the Dayak system, it is the wealthier households who are the rubber traders. Nearly all households own chickens, although in small numbers due to the high incidence of disease.

Wealthier households in the Upland Banjarese system make up just 10% of the population but earn over Rp3 million per year. Moderate households, who represent around 30% of the population, earn between Rp1 million and Rp3 million per year. Poorer households, who constitute over half of all households in this farming system, earn less than Rp500,000 per year. Most poorer households are food deficit. Households in the upper stratum have access to electricity and own a television set, radio and bicycle. Over half of the households in the middle stratum received loans compared with only one third of the households in the upper and lower stratum.

Transitional Banjarese Farming System

The livelihood strategies for Transitional Banjarese households are based on a combination of those for Banjarese households in the Uplands and the Lowlands. The major difference among the systems is the presence of PIR smallholder/nucleus estate rubber outgrower schemes in the transitional zone. Although income from this scheme accrues only to wealthier households, all households in this system have jungle rubber as their main source of livelihood.

Transitional Javanese Farming System

Income sources of households in the Transitional Javanese System are much more diversified than other systems. Although rubber is an important source of cash income for the moderate and lower strata, and is mainly earned through share tapping, it contributes just 20% to total income for both types of households. This is primarily due to the absence of mature rubber trees, and is likely to change once the young rubber trees recently planted come into production. Several moderate and upper stratum households have income from clone rubber nurseries. For all strata of Javanese households, cattle are an integral part of the farming system, and most also keep chickens and maybe a few ducks. Surplus production of maize and other vegetables is sold on the local market as are legumes such as soybean, groundnuts and mungbean.

The average income of the upper stratum, who constitute just 5% of all households in this system, is around Rp5 million per year. Moderate households, who represent 25%, earn around Rp1 million per year while poorer households (70% of all households) earn about Rp500,000. In terms of assets, the upper stratum have access to electricity and own a television and motorbike, while only a few households in the lower stratum would own even a radio or bicycle. Of those with loans, the majority have been from moderate and poorer households.

Lowland Banjarese Farming System

In Lowland Banjarese System jungle rubber is the main source of livelihood, accounting for nearly 70% of total income for both poor and moderate households, and over 50% for wealthier households. While the moderate and wealthier strata sell paddy on the local market and are more diversified into off-farm income-earning activities, the poorer tend to be food deficit and are more dependent on share cropping and casual labour. Nearly all households keep chickens and some ducks, both for own consumption and for local sale.

The upper stratum, who constitute 10% of all households in this farming system, earn over Rp3 million per year; moderate households, who make up 30% of the population, earn about Rp1 5 million per year; and poorer households, who are 60% of all households, earn less than Rp500,000 per year. Ownership of assets is similar to that of other systems. Most moderate households have received at least one loan, while the about half of the households in the upper and lower strata have participated in the programme.

Lowland Banjarese Sawah/Surjan Farming System

This farming system is the only one in the project area that does not have rubber as a source of income. Instead, padi sawah is the main source of livelihood. In the upper stratum, the income sources include paddy sales (30% of total income), own paddy consumption (17%), citrus (16%), poultry (7%) and various off-farm activities such as salaried employment, shopkeeping and trading (30%). Among moderate households, paddy sales account for only 5% of income, citrus sales account for 35%, salaries for 35% and own paddy consumption for 25%. It is primarily wealthier households who own chickens or ducks, averaging over 25 heads. Of all the farming systems, the Lowland Sawah/Surjan offers the fewest income-earning opportunities for poorer households, with 65% of their income coming from casual labour and 35% from share cropping paddy.

In this farming system, all categories of households are less well off than their counterparts in other systems. Wealthier households, for instance, who make up less than 10% of all households, earn under Rp3 million per year. Moderate households, 35% of all households, earn under Rp1 million per year and poor households (55%), earn less than Rp500,000. Nearly all of the households in this farming system, however, have received a loan.


DYNAMICS OF SOCIO-ECONOMIC CHANGE

Socio-Economic Differentiation

Sources of socio-economic differentiation common to all farming systems are: (i) access to paddy land; (ii) the size and productivity of rubber holdings, both owned and sharetapped; and (iii) off-farm income. Other factors vary by farming system:

In the upland Dayak swidden system, socio-economic differentiation is slight: all households have access to land for swidden fields and the size of the field matches the size of the household labour force. Differentiation is determined mainly by access to cash from bananas, rubber and off-farm sources;

In the upland Banjarese system, differentiation is more pronounced: the wealthy control large areas of rubber and upland rice, which they work with share tappers and hired labour; in spite of widespread access to fields for upland crops, households with no rubber of their own tend to be poor;

In the transitional Banjarese system, upland is in short supply and differentiation is greater than in the upland systems; access and control over rubber and sawah land are the determining factors;

In the transitional Javanese system, the determining factors are on-farm capital and know-how: capital includes cattle, ploughs, carts and human capital such as knowledge of intensive cropping or clone rubber grafting techniques;

In the lowland Banjarese sawah/rubber farming system, socio-economic differentiation is greater than elsewhere and is determined mainly by access and control over rubber trees and sawah land; this system has the highest proportion of landless and the greatest dependence on share tapping and casual labour;

In the lowland Banjarese sawah/surjan system, wealth is determined by ownership of paddy land, and among the owners of paddy land, by numbers and productivity of citrus trees.

Gender Aspects

Within the project area, women are heavily involved in agricultural activities. They play a major role in padi production, they tap, process and market rubber, they often have primary responsibility for fruit and vegetables as well as for chickens and ducks, and in the poorer households they gather wild foods in forest areas to help meet household food and nutrition requirements. Most women also control the household budget, and make important decisions regarding the use of household income. As members of local rotating savings groups (arisans), they have a strong influence in determining household investments, including those for the purchase of land and/or improved agricultural inputs.

Gender Division of Labour and Control of Resources

M Palawan Village (Tabalong) (Lowland Banjarese Farming System)

Resource

Who Does the Work

Who Controls the Resource

or Activity

Women

Men

Women

Men

Padi Sawah

   

X

X

 

Ploughing

 

X

   
 

seed preparation

X

     
 

Transplanting

X

     
 

Weeding

X

     
 

Harvesting

X

X

   
 

Threshing

X

X

   

Rubber

 

 

X

X

 

Tapping

X

X

   
 

Processing

X

X

   
 

Marketing

X

X

   

Garden

       
 

Vegetables

X

 

X

 
 

Fruit

X

 

X

 

Livestock

       
 

Chickens

X

 

X

 

Forest & Wild Products

       
 

wild foods

X

 

X

X

 

Firewood

 

X

X

X

Women and men have similar labour schedules in regards to agricultural activities, but women do all the cooking, cleaning and other household maintenance activities as well. Women also have primary responsibility for taking care of children, the elderly and other household members when they fall ill. In addition, women play an important role in community affairs, and spend a lot of time preparing for and participating in the various social and religious festivals. Easing women's workload can have a high payoff in terms of increasing their productivity as well as their opportunities for remunerated employment and income. As outlined in the Table below, some women in the Upland Dayak Farming System start their day at 2:00 am in order to have enough time to earn a cash income through rubber tapping activities and to ensure household food security through own production of rice. Activities that do not simultaneously seek to reduce the amount of time women spend on tasks will either fail, or will negatively impact the amount of time women are able to allocate to household food and nutrition security and overall well-being.

Despite their contribution to agricultural production, women often lack access to information and advice on improved agricultural practices. Although this situation holds true for most farmers in the project area, women, and particularly female-headed households, are at an even greater disadvantage due to the tendency of government extension workers to address mainly male household members. Village women's organization activities have also focused on training women to be better housewives and mothers. Neglect of women's role in agricultural production denies them access to the benefits that come with investment in agriculture and rural development and undermines the sustainability of project interventions.

Women's Time Use

Pipitak Jaya Village (Tapin) (Upland Dayak Farming System)

Time

Households with Ladang

Households with Rubber + Ladang

2 am

Sleep

wake, go for water, boil water

2:30 am

 

walk to rubber garden and start tapping

4-5 am

 

rest in rubber garden

5 am

wake, pray, clean house, wash

Collect latex, process into slab

6 am

Go to ladang field, weed upland rice and intercrops until 11:30

go home, pray, cook, eat breakfast, wash dishes, clean house, rest

9 am

 

go to ladang field to weed crops

11:30 am

Go home, pray, cook, eat, clean up, rest

go home, pray, cook, eat, clean up, rest

14:00

Go back to ladang field, work some more

go back to ladang field, work some more

16:00

Cut wood for cooking and carry it home

cut wood for cooking and carry it home

16:30

Go to the river, bathe, wash clothes, pray, cook, eat, wash dishes

go to the river, bathe, wash clothes, pray, cook, eat, wash dishes

19:00

Rest

Rest

19:30

Sleep

Sleep

Although not all female-headed households are poor, they do tend to be over-represented in this category given the proportion of female-headed households to total households in a village, as illustrated below.

Distribution of Female-headed Households by Socio-Economic Stratum, Pipitak Jaya

Stratum

Total HHs

Female Headed Households

% Female Headed HHs

Upper stratum

13

0

0%

Middle stratum

33

3

9%

Lower stratum

34

2

6%

Poorest

39

9

23%

In the absence of a spouse, the women's labour is often the main resource available, and they share crop paddy or share tap rubber to support a large number of dependants. In cases where household child care is lacking, mothers are often found tapping rubber or in the rice fields with children on their back. The labour constraints of female-headed households, combined with their limited access to government services, often perpetuates their situation, making the climb out of poverty a difficult struggle.

Female-headed Household - Jimamun Village (HSU)

(Lowland Banjarese Farming System)

Household members: Female-head (aged 30); 4 children (aged 14, 9, 8, 3); mother (aged 70) = 6 total

Labour force: 1

House condition: Very old, thatch roof, wood walls and floor, no latrine

Land holding: None

Livestock: None

Other Assets: None

Livelihood:

Share crops padi sawah, only during harvest time (2.5 months own consumption)

Share taps rubber 5 days per week for 5 months out of the year (Rp330,000)

Off-farm - roof making for 3 months out of the year (Rp30,000)

 


Total income including food: Rp580,000

Expenditures:

Rice (Rp350,000/year)

Other food (Rp250,000/year)

Medical (Rp5,000/year)

School (Rp100,000/year)

Clothing (Rp40,000/year)

Social (Rp5,000/year)

Total expenditures: Rp750,000

Loan recipient (Rp500,000) to buy 100 ducks, half died, other half given to son

Problems:
No money, especially after husband died 2 years ago

Arisan member, Rp1000 per week (every 6 weeks receives Rp8,000)

Priorities:
Money to purchase fishing equipment and a rice field

Social and Economic Change

The situation is dynamic and is evolving rapidly, especially in the upland areas, as is illustrated by the timeline from Auh village. The first chainsaws, TV and motorbikes date from the 1970s. The most recent changes have been electricity, road access, public taxi services, parabola antennas and village nurses. Agriculture is also in rapid evolution, as swidden farming gives way to permanent cropping and ladang fields are converted to rubber. However the full impact of recent projects has not yet been felt.

In the lowlands, agricultural development is relatively stagnant, whereas off-farm activities are in rapid evolution. Most rubber was planted 50 years ago and there is no new planting or replanting. The only farm activity that seems to be evolving is paddy production: through irrigation, IRRI varieties and fertilizer application. Migrants are returning from Saudi Arabia and building houses, mosques and Koranic universities and investing in shops and rice mills.

Downward Mobility and Process of Impoverishment

Causes of poverty include physical isolation from markets (as in the case of Dayak villages), poor health, widespread landlessness in the sawah villages and lack of off-farm income-earning opportunities. Downward mobility occurs mainly due to deaths and illness in the family, not to natural calamities. Crop failure is a contributing factor but is usually not a major cause of impoverishment, because most households have income from rubber tapping. Rubber has a stabilizing effect, providing a secure source of weekly income that households can count on for at least 6 months per year. The sawah/rubber system is more vulnerable than other systems, because in wet years, paddy suffers from uncontrolled flooding and rubber tapping is poor. In upland systems, rubber tapping and upland crops tend to compensate for one another, as one does well in dry years and the other in wet years.

A typical case of downward mobility is that of a woman whose husband died of cholera from drinking unboiled water while he was hunting in the mountains, leaving her with 5 young children to support. She has a severe labour constraint for upland rice production. So she is concentrating on tapping her husband's rubber trees, which provide a small but secure weekly income that she uses to buy rice. Her older children have dropped out of school to assist her on the farm She has not yet had to sell or mortgage her assets.

Poverty tends to breed more poverty: poor widows are more likely than others to have disabled children (lame, blind, mentally retarded) who will never be self-supporting. An example is a poor woman in Auh whose son became blind because an eye infection was not treated on time. In Aniungan, the son of the founder of the village is poor. He has access to a lot of land, but he is old, a widower, and his son has left the village. This shows that in the Dayak system, wealth is largely determined by the size of the family labour force and stage in the household cycle rather than inheritance.

Problems and Priorities as Expressed by Villagers

The main problems of the Dayak swidden system are: (i) poor road and market access, with consequent low cash incomes and the need to grow all their own food; (ii) high labour demands for upland crops that conflict with rubber tapping; (iii) infestation of the main crops - upland rice, bananas and rubber - by insect pests (walang sangit in rice, fusarium in bananas (pisang talas) and termites (rayap) in rubber); (iv) destruction of maize, groundnuts and rubber seedlings by wild pigs and monkeys; (v) decreasing area of land available for new ladang (swidden) fields due to the practice of planting rubber seedlings after rice on fallow fields; and (vi) health problems stemming from poor sanitation and long distances (20-30 km) on foot to the nearest rural health care facility. Their priorities are: (i) opening up of road and market access, (ii) a solution to the critical pest problems (walang sangit, fusarium and termites), (iii) new planting of rubber (preferable clone) on communal (adat) land and (iv) a village nurse. Since social stratification is not very pronounced within this farming system, the priorities of the upper, middle and lower stratum are similar.

The main problems in the upland Banjarese system are: (i) low income, especially during the rainy season when crops are still in the fields and little money is coming in from rubber tapping; (ii) low price of rubber compared with 1995-96; (iii) declining rubber production due to the old age of existing trees and severe termite infestation - resulting in loss of 20% of trees; (iv) decreasing land availability for new ladang fields due to gradual conversion of fallow land to rubber; (v) declining soil fertility due to shortening of the fallow cycle; (vi) low productivity of upland rice due to declining soil fertility and pest damage (walang sangit); (vii) low returns to labour for upland rice; (viii) destruction of groundnuts by pigs, monkeys and mice; (ix) severe problems of dry-season water supply and sanitation when seasonal streams dry up; and (x) consequent flare-up of oral-faecal diseases such as cholera, hepatitis and diarrhoea. In a number of upland villages, there is open conflict over land rights between villagers and state-owned forest enterprises, rubber and oil palm estates, transmigration schemes, as well as private forest concessions and mining enterprises (especially coal).

The main priorities expressed by villagers are: (i) new planting of clone rubber on alang-alang grasslands; (ii) solutions to pest problems (termites = #1 problem, then walang sangit); (iii) clean and reliable dry-season water supply (e g through hand pumps); and (iv) health education through a village nurse and posyandu2). The new planting of rubber is already taking place spontaneously with local varieties; farmers have heard about clone rubber and want to plant it, but have rarely seen it in production and cannot get hold of a supply. Since the upland Banjarese farming system has rather pronounced socio-economic stratification, the problem of low income is compounded by landlessness (around 30%), skewed ownership of rubber, and dependency of over 50% of poor households on share tapping as their main source of livelihood.

The problems and priorities of the transitional Banjarese system are similar to those of the upland and the lowland sawah zone systems, except that population density is higher, land holdings are smaller and there is little or no communal land available for new planting of rubber. But radical cutting and replanting of existing rubber would hurt the poor, who because of skewed land ownership are highly dependent on share tapping for their livelihood. In addition, due to greater proximity of migrants from Java and Madura who use animal traction on uplands and intensive inputs on sawah, the Banjarese upper stratum perceives the "shortage of capital" for inputs and hired labour as a constraint for increased production.

The lowland Banjarese sawah/rubber farming system has similar problems of low income, especially in the rainy season, when rubber tapping is low. As everywhere, farmers protest that rubber prices have fallen. Moreover, production is gradually declining because rubber holdings are old. Termite infestation is severe and affects young trees (5 years old) in addition to old ones. In wet years, rainfed sawah land is affected by excessive flooding, with consequent crop loss. In dry years, the paddy crop has been severely affected by the stemborer (wereng), especially the improved varieties grown under irrigation. In the sawah zone there is little or no unused land available for new rubber planting.

Farmers in the sawah/surjan farming system also have a problem of low income because the main source of income in the other systems - rubber - does not grow there. The pattern of income distribution is determined largely by access to paddy land, most of which is single cropped due to lack of irrigation or water shortages. Only a few households (upper 30% of the population) have income from citrus and most owners of orange trees mortgage their fruit to local traders (ijon system) several months before the harvest. In addition, there is a severe shortage of drinking water for six months a year and an acute sanitation problem, compounded by pollution of irrigation canals, which - according to villagers - is caused by discharges from a rubber factory.

1 In four villages, one farming system is practiced in certain RTs (hamlets) and another in other hamlets.

2 An informal, village-level centre where mothers monitor their infants' growth with the nurse's assistance.