Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
The state, the timber industry, and shifting cultivators
A problem of high visibility
The Indonesian and Malaysian governments, like many others in the tropics, are committed to eliminating shifting cultivation in all its forms. Large areas lying fallow at any one time form an essential component of the system and offend those who wish to see resources, whether of land or of timber, in commercially productive use. Yields of basic food crops, especially rice, tend to be less reliable than those in other forms of agriculture, and it is widely believed that shifting cultivators are not only inefficient but also among the poorest of the rural poor. Elimination of the system would, therefore, meet social as well as developmental goals. Persuasion has been applied for many years, but the timber boom has created a new urgency that blames shifting cultivators for eating into primary and late-secondary forest that could yield timber revenues to the state even so much so that, for every log felled, "another goes up in smoke" (Lau 1979). No government in the region has given its indigenous peoples inalienable rights to all the land they use.
The issues in Sarawak
Although collisions of interest between logging entrepreneurs, state land-settlement agencies, and indigenous shifting cultivators have also developed in Kalimantan and in Sabah, it is in Sarawak that the sharpest conflicts have arisen and have achieved by far the widest notice. The Sarawak Land Code does not recognize customary rights to land under forest, unless it was occupied and used before 1958. Even this has been applied conservatively, and only some 15 per cent of all land has registered title under systems initiated under the Brookes and codified in the colonial period (Cramb and Dixon 1988). Little of this belongs to shifting cultivators, although they claim an estimated 25 per cent, and in fact more, customary land. This compares with 19 per cent of the state estimated as used for shifting cultivation by Hatch (1982), and 28 per cent by Kavanagh, Rahim, and Hails (1989) using an estimate based on 1985 Landsat imagery by Majaran and Dimin (1989).
In fact, probably many fewer than one-half of the 570,000 indigenous people of Sarawak practice shifting cultivation in anything like its classic form. In the more densely occupied western areas, the pioneering stage made famous, or infamous, by Freeman (1955, 1970) is long past. Many communities occupy land first cleared from forest more than 300 years ago. Their forests are all secondary and contain many trees planted or conserved for their edible fruit or other useful produce. Areas of semi-permanent swamp rice supplement hill-rice swiddens (Cramb 1988; Padoch 1982a; Sather 1990). The true swiddeners, who occupy the more sparsely populated eastern parts of Sarawak, are composed of several ethnic groups, and most have a history of recent migration (Lien 1987; Padoch 1982a). They live and cultivate along and close to the rivers, penetrating only a limited way into the surrounding forests. Although their basic methods of cultivation are the same as those of the more densely settled areas, practices are less intensive, with smaller areas of selectively managed forest and much larger areas of natural regrowth. These, principally, are the people against whose "wasteful" methods generations of administrators and agriculturalists have railed, and who - together with the few thousand hunter-gatherer Penan - are now most closely affected by the timber industry.
Three main prongs of criticism address the shifting cultivators, about whom little agreement prevails. The question of poverty and insecurity of livelihood, implicit or explicit in almost every relevant planning document, may have some truth. King (1988) reports his own estimate that 40 per cent of Sarawak's population live below the poverty line, overwhelmingly in rural areas but probably largely in the west. Lian (1987) and Cramb (1989a) question the lack of self-sufficiency in basic foods, the more so when foods supplementary to rice are taken into account. Yields are low, and a great deal depends on skill in timing activities in relation to the weather, but there is often a surplus. The argument that the system is extremely destructive of natural resources has an extensive literature, spearheaded in Sarawak by Freeman (1955), and challenged by a large number of writers including Cramb (1989b), Padoch (1982a,b), and others. Much seems to depend on the area described by particular writers, for, while there are no vast tracts of Imperata cylindrica or other grassland in Sarawak, grasslands, erosion, and badly degraded forests all exist. The third argument - that shifting cultivation consumes annually large areas of primary forest - is orthodoxy (Hatch and Lim 1978), but critics have argued that Lau's (1979) estimate, mentioned above, relied on erroneous data concerning the area under shifting cultivation at any one time. They have shown that in specific cases only quite small areas of primary forest are used, most swiddens being made in secondary growth of from 7 to 20 years (Hong 1987; Lian 1987; Sather 1990). Cramb (1990) has produced data purporting to show that shifting cultivators annually increase the area cut by only 0.2 per cent - about as much as is logged each week. The debate continues, with no resolution in sight.
Lian (1987) studied a longhouse community adjacent to a timber camp, and detailed the effects of the new and temporary source of employment and income on the people. Although employed only in the lowest jobs and at low wages, the indigenous people gained important new cash income, causing them temporarily to neglect other cash crops and to reduce the area farmed to that which would principally be maintained by the women. One effect, however, was to shorten the swidden cycle. This community had received private compensation from the contractor for access over its land close to the river. Moreover, they received free electricity from the timber camp. Changing sources of income had a levelling effect on class structure. None the less, the community was looking beyond the timber period and considering future options that might include hitherto-resisted resettlement.
"Confrontations" and the environment
Confrontations between shifting cultivators and loggers supported by the Malaysian authorities became common in the 1980s, starting some years before the better-known blockades of logging roads by small groups of Penan. They are often interpreted as a desperate defence of livelihood and environment, and in this context have gained very wide publicity (e.g. World Rainforest Movement and Sahabat Alam Malaysia 1990). There are, however, those who maintain that the confrontations have been misunderstood. Lian (1993) points out that only 5 per cent of the fewer than 10,000 Penan are still hunter-gatherers and that the sago on which they rely is in areas that do not attract logging. He points out that food shortages preceded logging by some years and that land sales from Penan to other groups have exacerbated the problem. Kavanagh, Rahim, and Hails (1989), following a State Planning Department report of 1987, agree that few Penan are now wholly nomadic, but say that up to 70 per cent still spend a part of each year hunting and gathering, away from the shifting cultivation settlements made since the 1950s. Lian (personal communication) argues that the confrontations involving Penan and other peoples have had more to do with the distribution of rewards from logging than with safeguarding the rain forest or the environment generally. Hurst (1990) and Primack (1991) agree that this is true in at least some cases. Whether to do with cash rewards or with protection of food resources, the motivation would, with this interpretation, be primarily economic. The partial resolution now proposed by the government is to create some reserve areas within which Penan can obtain their traditional food supplies.
Most certainly, however, the inland people are concerned about damage to their environment, especially to their rivers (Aten Suwandi n.d.). They are resistant to resettlement, but they seek the benefits of development. They are taking advantage of the network of roads now being created by the linking up of the better-constructed logging roads. They like the resulting drop in the cost of imported goods. It is their survival as a distinctive people, not as a population, that is under threat from development. Present indications are that their response to this challenge is to take a more active role in politics, which their numbers make possible. Given the probability that Sarawak's forests will continue to be heavily logged, albeit with efforts to develop tighter management control that will support the growth of industry, further major changes must be anticipated in the coming decade. The ITTC mission strongly urged that confrontation be replaced by discussion and compromise when areas claimed under customary tenure are to be logged (ITTC 1990). Lian's (1987) evidence of mutual accommodation, from a valley in which the loggers are about equal in number to the local people, suggests that this would be possible.
The problem in Kalimantan
Kalimantan at present lacks the degree of confrontation encountered in Sarawak, even though shifting cultivators are no more considered to be legal occupiers of the land they use than are those in Malaysian Borneo. Much of their land is being cut over, and in Kalimantan there is the additional problem of transmigrant settlement. The farming systems themselves are changing. People from remote districts in East Kalimantan, sometimes under government direction, are moving closer to Samarinda and reducing fallow periods; possessing or hiring chainsaws, they clear larger areas for cash crops. This has led observers to query whether forest recovery is likely still to occur when areas cleared by several adjacent families reach as much as 300 ha, as compared with 30 ha in the more remote villages (Inoue and Lahjie 1990; Kartawinata et al. 1984).
The total area under shifting cultivation in Kalimantan is almost certainly larger than in Sarawak (about 5.5 million ha against 3.5 million ha by the best estimates). Data are very uncertain, however, and estimates of over 1 million individuals depending on the system are questionable. The RePPProT studies provide the most reliable information. In terms of land occupied, shifting cultivation is most important in West and Central Kalimantan, where it takes up 24 per cent and 22 per cent of the forest area respectively. Overall, 61 per cent of shifting cultivators are located in conversion and unclassed forest (i.e. areas already marked out for permanent clearing) but, whereas 72 per cent are on such land in East Kalimantan, in West Kalimantan the figure is only 56 per cent. The authors of the West Kalimantan study warn: "Pioneering shifting cultivation penetrates far into the... forest areas and threatens to fragment and consume all remaining non-swampland lowland forests in the short to medium term" (RePPProT 1987b, 1, 30).
It is not clear whether locals or migrants are the alleged culprits here, but such a sweeping statement necessitates fieldwork, to check both its veracity and the reasons for the situation. The view expressed is redolent of the debates about the "forest-consuming" propensities of the Iban in Sarawak, recalling that some of the Dayak groups in West Kalimantan are of Iban origin (Freeman 1955, 1970; Padoch 1982a,b). It is, however, also true that intricate agro-forestry systems have arisen in parts of West Kalimantan, even incorporating planting of the long-maturing Borneo ironwood (Eusideroxylon zwagerei), as well as rattan, rubber, durians, and a wide range of other useful trees. Particularly important are the nuts from the tengkawang trees (Shorea spp.), important sources of edible oils (Padoch and Peluso 1991). Collection of these products from managed forest, plus tapping of planted rubber, supplements and partly replaces rice swiddens as a means of livelihood.
Government policy and the shifting cultivators
The governments of both countries seek positive ways of overcoming the problems perceived as being associated with "pioneering" shifting cultivation in East Malaysia and Kalimantan. Whatever the defence of the system in its classic form, a real problem exists where numbers increase and where commercial crops are grown by swidden methods. Unfortunately, most current solutions involve resettlement and a complete change, rather than adaptation, of farming methods. With the stated aim of reducing poverty, the government of Sarawak has sought to resettle shifting cultivators in managed cash-crop blocks for over 20 years. SALCRA was set up for this purpose. The Indonesian authorities have been even more determined in moving indigenous people to what are seen as better-located sites, usually in conversion forest, where they can, supposedly, learn more intensive agriculture. East Kalimantan has been most active in promoting such resettlement schemes, which have been criticized as culturally insensitive and often economically unviable (Appell and Appell-Warren 1985; Avé and King 1986). These have usually been food-crop schemes, resembling transmigration projects. Indeed, transmigration schemes always include reserved places for locals, partly in order to dampen accusations of favouritism towards Javanese and other newcomers. After a few years, some of these resettlement projects are simply abandoned and the people drift back to their old lands; the government-provided houses are then re-used for another group of transmigrants.
More positive approaches could involve extensions of the indigenous agro-forestry systems described above from West Kalimantan, perhaps also adopting Davis's suggestion that all shifting cultivators be given tenure to a reasonably large area of land, say 20 ha (World Bank 1990). Another idea is that of community forests, which would be a kind of buffer zone around the edge of protected areas or national parks, controlled by local villagers, who would have the right to use all produce from the zone without encroaching on the clearly demarcated boundary of the park. Leighton and Peart (1990) have suggested enrichment planting of useful species, such as harvestable fruit trees and rattan, in the fallow. Other proposals call for making more effective use of traditional systems of land tenure in the management of consolidation and rehabilitation (Cramb and Wills 1990).
Lessons for the future from peninsular Malaysia
A greater and more rapid transformation
Before drawing the argument together, we digress briefly to draw lessons from the experience of Peninsular Malaysia. Much of what is now happening in Borneo has already transpired on the Peninsula. By 1989, the forested area was reduced to only 42 per cent of the land (Thang 1990). The unpublished land-use survey of 1982 showed 31 per cent of the Peninsula as already under rubber and oil-palm, and almost all this was formerly forest. Problems for the sustainability of forestry as an industry were foreseen early (Malayan Forester 1967), and the warning was frequently repeated (e.g. Kumar 1986). Peninsular Malaysia already faces a shortage of timber supply for its downstream woodworking industries, based mainly on sawmilling. Between 1973 and 1981 the value of output in the whole group of timber industries grew by 12.9 per cent per year and employment grew by 5.7 per cent (MIDA and UNIDO 1985). A much greater increase has since boosted the production of 681 large sawmills, 43 veneer and plywood mills, and more than 1,200 small woodworking plants, furniture factories, and other small wood-using enterprises (Country Profile 1990). The consequence of expansion has been an increasingly severe shortage of timber for the mills, for the building industry, for exporters of sawn timber, and for the downstream factories (Malaysian Business, 16-30 September 1989, 1-15 August 1990; Country Profile 1990).
Yet efforts at management have been stronger in the Peninsula than in Borneo, guided by what is certainly the best forestry service in the region (Burgess 1990). A National Forestry Policy was formulated in 1977, and its provisions included the creation of the PFE and also of conservation forests. Although illegal logging still occurs, management has clearly improved as resources have grown scarce. The permitted coup was reduced by 12.5 per cent between 1985 and 1990 (Malaysia 1991a). Both enrichment planting and plantation forestry are now required on a greatly increased scale to supplement improved management.
The fate of the new rural economy
Most of the land cleared from forest since 1960 has been converted into land-settlement schemes, particularly by FELDA. Mechanized forestry and land clearance have reduced the agricultural capability of the land through compaction, loss of topsoil, destruction of soil structure, and exposure to erosion (Mohd. Nor Zakaria et al. 1985). Yet, with the use of fertilizers, yields have remained satisfactory, even though increasing areas of steepland have come into agricultural use. Now, however, the expansion of agricultural land has encountered limits. The larger problems are economic and demographic rather than ecological. The cost of land preparation has risen as clearance has moved into hilly land, so that, according to two sources, the mean cost of establishing a family rose by 73 per cent between 1976 and 1985 (Drury 1988) and by 41 per cent between 1980 and 1985 (Chamhuri and Nik Hashim 1988). Depressed prices have hit rural incomes, and a shortage of labour now affects most parts of the Peninsular rural economy. The estates depend largely on illegal immigrants from Indonesia, estimates of whose total number in the Peninsula range from 100,000 to more than 1 million. In an interview, the Minister of Primary Industries has stated that if immigrant labour were to be "done away with, my whole estate sector will collapse" (Logging and Resources 1990).
With growing difficulty in attracting settlers to the raw environment of remote areas, FELDA terminated all new land development in 1991, after completion of ongoing work. All those schemes that have not recruited settlers are to be managed through the formation of plantation companies (Malaysia 1991a). This will be implemented in 222 FELDA schemes in the Peninsula, covering a total area of 300,000 ha, about 40 per cent of all FELDA land. What is being created de facto is a major extension of the estate sector. The crisis of land settlement is no greater than that in Sabah, but the reasons are different. Labour shortage in Peninsular agriculture is a product mainly of the rapid industrialization and urbanization of the national economy, and of the drift against agriculture in the terms of trade and comparative labour pricing. The apparent solution is the same as in Sabah however, involving reliance on abundant supplies of illegal immigrant labour from the less developed and more populous countries of the region.
Implications for Borneo
The Peninsular economy is increasingly dominated by urban and industrial development, and the "resource frontier" period is over. Conversion of forest to agriculture will cease even before it has reached its potential high-cost limits, some of the cut-over land will become tree plantations for higher-density timber farming, and the rest of the forest will be subject to much more stringent management controls. It may be that some cleared land, especially on the hills, will never undergo profitable development and will join considerable areas of older rubber in returning to secondary forest.
Management of Peninsular forestry has been better than that of any part of Borneo and log-export bans began sooner. Land preparation for tree-crop plantation has improved significantly since research and the major floods of 1971 revealed the heavy erosion of earlier years. This is not to say that environmental damage is fully under control, but, if anything like Peninsular management had been applied in Borneo, the situation there would now be less alarming than it is. Peninsular Malaysia is now acquiring a managed environment, both in forestry and in agriculture. But the essential difference lies in the urbanization and industrialization that have become established in the two national heartlands, in western Malaysia and in Java, leaving Borneo the acknowledged resource frontier. Lessons for management are available, but the fundamental changes in the national economies do not readily cross the South China and Java seas.
Questions of criticality in Borneo
Problems of the environment: Critical lands
We now return to some of the issues of criticality, starting with the land. The question of who is responsible for the most environmental damage, the loggers or the cultivators and settlers, is close to the heart of much of the dispute between conservationists and the authorities. Despite a dearth of hard data, there can be little doubt that erosion and sediment transport are quite substantial under natural conditions in the forest, but are increased with cultivation and increased much more with logging. In nearby Mindanao, Kellman (1969) reported an 18-fold increase in soil loss from forest after logging. An identical increase has recently been established in Sabah by an Anglo-Malaysian project (New Straits Times, 26 September 1990, 1). After rain, rivers quickly become turbid, but the opaquely high turbidity of rivers in the logging areas is a modern phenomenon. How fast the load travels downstream is unknown, but navigation difficulties are experienced on some rivers (Aten Suwandi n.d.). Douglas (personal communication), however, reports work from Sabah showing that, because logged areas very quickly revegetate, erodibility falls to modest levels in as little as a few weeks.
The same is certainly also true of shifting cultivation clearings that revert to forest rather than to grassland, which applies to most of them. Badly eroded grasslands exist in Borneo, however, and, in Kalimantan, watersheds in which soil erosion has become severe are designated "critical land" (Tanah Kritis). A recent estimate by the Worldwide Fund for Nature of 20 million ha for the whole of Indonesia has been reduced to 13.1 million by reforestation and rehabilitation work carried out, mainly in Java, by the Ministry of Forestry (MOF and FAO 1991). These new studies have, however, resulted in an upward revision of critical lands in Kalimantan to almost 3 million ha, or 23 per cent of the total area; two-thirds of these lands are in West and Central Kalimantan (Statistik Indonesia 1990). Previous estimates had always emphasized the West and South, attributing very low figures to the two less-populated provinces. Now both these are acknowledged to have an erosion problem, although no information designates its location.
In West Kalimantan the main problem area appears to lie in the hilly districts north of the Kapuas river. The RePPProT report describes "large areas of barren land" in this region. A further statement is more explicit: "Very steep ridge systems of the Western Plains and Mountains and the Middle Kapuas Basin have been degraded to grassland and scrub by over-intensive shifting cultivation. The erosion of these areas is excessive. Present reforestation activities should concentrate on these areas first" (RePPProT 1987b, 38). Though little is visible from the Pontianak-Sanggau road, extensive tracts of Imperata cylindrica grassland may be seen from the air. Since this zone lies within easy access of Pontianak and the coast, it would seem to have been under pressure for many years. There are also extensive grassland areas in the middle Melawi basin.
Similar recommendations are made for steep ridge systems in the Meratus mountain foothills of South Kalimantan that are also said to be eroding excessively as a result of shifting cultivation (RePPProT 1987a, 15). These latter areas, also under lmperata cylindrica, have largely been cleared during the past century, though some of the grassland may be older (Potter 1987). About 16 per cent of South Kalimantan has a cover of grassland, which government authorities have been attempting for many years to reforest; dry-season fires have been a recurring problem for at least a century.
Drought and fire
Since the major fires in East Kalimantan and Sabah in 1982-1983, it has been recognized that substantial evidence exists of earlier fires in parts of the east Borneo forests (Wirawan 1993). This is not surprising, since all modern reports of large forest fires to be found in the literature correspond with periods of unusual drought associated with major ENSO (El Niño-Southern Oscillation) events, and several major episodes have occurred in the past 120 years of records. Moreover, there can be no doubt that they have continued over a very long time (Brookfield and Allen 1991; Nicholls 1991). In the Malesian region, normally a principal focus of deep convection in the troposphere, ENSO events see this convection moved out into the central Pacific and its replacement with persistently descending air. The effect in these year-long events is to extend and intensify the dry season where it exists and greatly to weaken the wet season. Even in Borneo, severe drought of several months' duration can be a consequence.
Although there are reports of forest fires in 1902 and 1914, the conflagrations of 1982-1983 were almost certainly without recent historical precedent in their scale. Studies conducted by German experts in the 32,000 km² of East Kalimantan burned in 1982-1983 concluded that it was largely the impact of recent logging that led to the size of the fire, by changing the composition of the forest towards more inflammable secondary species (Schindele, Thoma, and Panzer 1989). They also pointed out that in the more severely burned areas it could be hundreds of years before some of the valuable species reappeared, for they would be suppressed by competitive vegetation. The same set of fires destroyed a further 10,000 km², 85 per cent of which had been logged, in Sabah (Woods 1987). There is some risk that such large gaps, subjected to new droughts, may never revegetate with forest and could become grassland.
Considerable new burning occurred from August to November 1991, during an ENSO event of wide scope in the western Pacific. Estimates at mid-October put the total area burned at about 90,000 ha, of which 50,000 were in Kalimantan (Kompas, 12 October 1991; Tempo, 19 October 1991). In the absence of survey, these estimates may entail more than an element of guesswork; one television broadcast around the same time put the area in Central Kalimantan alone at 76,000 ha! Large areas were covered with smoke, airplanes could not land at Palangkaraya, Pontianak, and various locations in Sumatra, and haze affected aviation as far afield as Singapore and Peninsular Malaysia. Although some of this was related to burning underground coal seams in East Kalimantan, a good deal more is attributable to deliberate clearing of local swamp forest, with small farmers taking advantage of the dry conditions to extend cultivation into swamps. The burn near Palangkaraya in Central Kalimantan was said to be started by shifting cultivators and, as it was almost entirely in swamp forest, this is probably the explanation. In South Kalimantan a line of fire was observable from the air all along the mountain foothills, the approximate location of the Acacia mangium plantations. These fires were less severe than those of 1982-1983, but they were a warning of how widespread and disruptive conflagrations can be whenever a drought occurs.
Apart from fire, other serious consequences of the 1991 drought were the widespread crop losses, the large numbers of deaths from cholera and other waterborne diseases (148 in Central Kalimantan), and the disruption to transport caused by the low volume of water in the rivers, which seriously interrupted log-rafting to plywood factories in the lower reaches of the Mahakam and Barito. Low river levels also interfered with logging in Sarawak.
Poverty and wealth
The beginning of this chapter posed the question of the effect of rapid change on human welfare. It is not simply answered, since the evidence is equivocal. If the incomes from oil and its products are included, East Kalimantan is Indonesia's richest province, providing on its own more than 5 per cent of the nation's GDP (Statistik Indonesia 1990), or a per capita income in 1988 of US$2,626, which is closely comparable to that of Selangor in the Malay Peninsula (about US$2,700 in 1990), which ranks third among the states of much wealthier Malaysia. Even excluding oil, the remaining income (US$819) was twice the national average, whereas that of West Kalimantan, the poorest of the four provinces (US$317), was below this. Within Malaysia, Sabah ranks fifth in per capita GDP, at about US$1,800 in 1990, some way below the wealthier urban states and oil-rich Terengganu, whereas Sarawak, at about US$1,550, is on a level comparable with the more developed rural states of the Peninsula (Malaysia 1991b). Small Brunei, with a per capita GDP of US$14,076 (The Far East and Australasia 1991) is among the world's richest countries.
These gross comparisons do have some meaning, and the generally higher level in Malaysian than in Indonesian Borneo is reflected in greater provision of infrastructure, especially in the urbanized areas. We have already noted, however, the extent of rural poverty in Sarawak, at least on a statistical basis. Poverty in Malaysia is measured by monthly household cash income and, on the basis of a "poverty line" that has varied somewhat through time, 1989 survey data showed 34 per cent of the population of Sabah and 21 per cent of that of Sarawak as poor, levels much higher than in the Peninsula (Malaysia 1991a). Most poverty-level households are rural in both states.
Poverty in Indonesia is largely measured by levels of household expenditure on essential foodstuffs, considered more accurate than income figures, and probably correctly so for rural families. In the most popular method, that of Sayogyo (1975), the poverty line is drawn at the cost of purchasing 20 kg of rice per person in the rural areas, 30 kg in the cities. Measured in this way, poverty has been at quite moderate levels in Kalimantan. However, much of the off-farm employment has gone to Javanese in-migrant workers, even in the large plywood factories. There has been concern that the transmigration programme will transfer poverty out of Java to rural parts of the outer islands where transmigrants are numerous (World Bank 1989). Some provisional results from the most recent National Socio-Economic Survey (SUSENAS), held in 1987, lend credence to this belief. A study carried out in South Kalimantan, using the detailed SUSENAS results (BAPPEDA/Kantor Statistik, Kalsel 1987), revealed 10.2 per cent of the urban population and 12.4 per cent of the rural population as falling below the poverty line. Yet an examination of data at the district and regency level found that some particular areas stood out as being much poorer. These were sections of the tidal swamps with many transmigrant settlements, where poverty was over 30 per cent in 7 of the 14 regencies, and parts of the east coast, where the more inaccessible regencies with large transmigrant populations showed similar poverty figures.
Pockets of poverty also exist among settled agricultural populations in the inland Hulu Sungai area of South Kalimantan. Although soils are good by Kalimantan standards, agriculture has remained risky because of problems with water control. Droughts, floods, and, more recently, low rubber prices have triggered waves of migrants seeking new land. They have travelled to Perak and Johore in Malaysia and to Riau on the east coast of Sumatra; they have also opened tidal swamps nearer home. Those away have usually retained their landholdings, which now results in high levels of sharecropping and low returns to growers, despite widespread use of high-yielding rice varieties. Multiplying occupations, rather than intensifying existing production, has been the preferred strategy, with recent migrants heading for the cities rather than seeking new rural land. West Kalimantan, with the lowest per capita incomes and household expenditures, still has 73 per cent of its labour force engaged in agriculture, the highest of the four provinces; less than 5 per cent are in manufacturing and 17 per cent in services and trade. For highly urbanized East Kalimantan, agriculture (including forestry) occupies only 43 per cent of the workforce, with 7 per cent in manufacturing and a high 39 per cent in services and trade. Income levels from farming, when not combined with other cash-earning activities, fall inevitably below the poverty line. Even though farmers in West Kalimantan, and elsewhere, may attempt to diversify, yields are low, the resource base poor, and transport undeveloped.
Wealth, on the other hand, is not conspicuous in most of Borneo, with the striking exception of Brunei. It is visible in urban areas, as in the superior housing of oil company executives in Balikpapan, Bintulu, and other oil towns, and the "palaces" of a few merchants or industrialists in Banjarmasin. Nevertheless, the timber royalties that have been available to local governments have resulted in improved urban facilities, especially in Samarinda and Palangkaraya in Kalimantan and in all the towns of Malaysian Borneo. Timber is now worth 16 per cent of Indonesia's GNP, and benefits have "trickled down" in the form of village schools and health centres. Oil incomes produce more concentrated spin-offs in the oil towns of north and east Borneo. Pangestu (1989) points out that local benefits from the oil industry have been quite substantial in East Kalimantan. The refinery at Balikpapan and the large liquefied natural gas and fertilizer plants to the north provide useful employment, and the companies have supplied schools, medical services, electricity, and water to the new communities.
There are, however, differences between the two parts of Borneo. The wealth that has flowed from the forests and the oil wells of Kalimantan has largely left the island for Jakarta, or has been transferred overseas. In Sabah and Sarawak, more of this income stays within the state. Road construction, land development, education, and other social services, and, most clearly in Sabah, the construction of glassy towers in the capital, Kota Kinabalu, all depended primarily on timber income, together with a smaller but rising income from oil and gas in the South China Sea. The Kalimantan provinces are only 4 out of the 26 that constitute Indonesia, and are subject to centralized policies emanating from Jakarta. Forces for change are nationwide, and particular regions of Indonesia, other than Java itself, are seldom at the forefront of consideration. The greater degree of economic independence in the states of Malaysian Borneo is reflected in development policies that are only in part encapsulated in the national plans, and, moreover, these plans are always careful to stress the interests of the Borneo states.
The welfare of one group of people is missing from all the statistics. These people are the illegal immigrants - almost all in the Malaysian part of Borneo and in Brunei - from Indonesia and the Philippines, with a few from other places. Where they work as agricultural or construction workers, or in other manual tasks, most are poorly paid yet they still find it advantageous to have moved. Some, however, have become tenant farmers, and still others engage in various entrepreneurial activities. There are pert-urban settlements around the towns of northern Borneo within which at least a few illegal immigrants have certainly prospered. Though they are less numerous than their counterparts in Peninsular Malaysia, the illegals constitute a significant minority, which is large in Sabah. Very little is known about them.