This paper examines the impact of two multisectoral highland development projects on ethnic minorities in northern Thailand. While social sector interventions were relatively successfully delivered and locally appreciated, strategies for agricultural development failed to take account of local cultural and production systems, and were fraught with a range of design and implementation difficulties. These uneven outcomes are traced to the assumptions underlying project design, contradictions in policy (especially relating to land tenure), and the institutional cultures of implementing agencies.
The northern highlands are home to most of Thailand's 800,000 ethnic minority hill peoples (chao khao in Thai, generally rendered as “hill tribes” in English). Policies towards highland minorities have been shaped by territorial and security concerns, the eradication of opium cultivation, and the control of shifting cultivation. A policy of integration towards the hill peoples was adopted beginning in 1976. During the 1980s and 1990s, 5 integrated rural development projects were implemented, coordinated by the Office of the Narcotics Control Board (ONCB), and supported by the United Nations Drug Control Programme (UNDCP). Their objectives were to replace opium cultivation with alternative sources of income and improve social services. This paper examines 2 of these projects.
The two projects
Pae Por is a dissected mountain range near the border with Myanmar in Chiang Mai and Tak provinces, inhabited by Karen of the Scaw sub-group (Figure 1). Wieng Pha is situated on the border of Chiang Mai and Chiang Rai provinces, and has an ethnically diverse population, including Lahu (55%), Lisu (24%), Karen (13%), Akha (5%), Meo (1%), local Thai (1%), and Haw Chinese (1%). Features of the 2 project areas are summarized in Table 1.
Rice was the main crop in both locations, cultivated on the uplands and as irrigated paddy. Most households relied on their own labor. Many households were not self-sufficient in food, meeting the shortfall through casual labor or sales of forest products. Increasing population pressure had resulted in declining fallow periods and falling yields in both areas. Physical infrastructure was poor, especially in Pae Por, where only one village could be reached by a paved road. Before the project, health and education services were very limited, especially in Pae Por, where only 1% had any formal education. Health indicators were very poor, especially for women. Substantial areas of opium poppy were cultivated, and levels of addiction were quite high in both project areas.
Complex and costly management structures
Both projects aimed to raise the living standards of hill tribe populations by “improving socioeconomic position and self-reliance and progressive integration into the mainstream of Thai society,” as they declared in their project documents. Pae Por Highland Development Project (HDP) was executed by the Department of Local Administration (DOLA) and managed by the 2 District Officers into whose jurisdictions the project fell. At the field level, implementation was centered on 15 “key villages,” each staffed with a 3-person team consisting of a non-formal education teacher, a community health worker (CHW), and an agricultural agent. Management of Wieng Pha, though formally the responsibility of a multi-agency Project Management Committee, in practice fell to the Project Director in the Royal Forestry Department (RFD), the executing agency. The heads of 4 RFD Watershed Development Units served as field managers, each responsible for 2 to 4 project agroforestry extension officers.
The project designs were complex (Wieng Pha had no less than 27 objectives in 7 sectors) and both projects involved a multiplicity of institutions, including DOLA, RFD, the Ministries of Interior, Public Health, Education, and the Departments of Public Welfare, Agricultural Extension, Community Development, and Technical and Economic Cooperation (DTEC), some of which were involved at national, provincial, and district levels. In addition, 4 donor organizations supported the projects: UNDCP, UNDP, UNICEF, and UNFPA. This complexity led to intricate management arrangements and high administration costs.
Positive and negative project impacts
Some of the main outputs and outcomes of the 2 projects are summarized in Table 1 above, and discussed below.
Education and health services
A substantial number of health facilities and non-formal education centers were established under both projects. In Pae Por HDP, well-equipped schools provided a conducive learning environment for 1500 children—about half of the eligible population—while the number of health centers (Figure 2) was doubled, and staff with hill tribe backgrounds provided services such as vaccination and family planning.
Opium addiction and detoxification
Baseline addiction rates stood at 8% in Pae Por HDP, and 20% in Wieng Pha, both exceeding the national estimate for hill peoples (6.8%). The two projects treated 766 addicts altogether, in community-based programs in villages or in district health centers. Rates of non-recidivism were estimated as 15–20%, not dissimilar to such programs elsewhere. The cursory nature of rehabilitation and follow-up limited impact, and in both areas there was evidence of an increase in the number of addicts during the project. Thus opium addiction remained a massive problem.
In Pae Por HDP, the main strategy was one of agricultural commercialization, mainly through the introduction of coffee and other cash crops. This strategy had several flaws. Most critical, given the predominance of subsistence objectives and the lack of infrastructure and markets, was the heavy dependence on inputs. Second, implementation lacked coherence and sustainability: while seeds and fertilizer were free, distribution was haphazard and inequitable, and no provision was made to continue supplies after the end of the project. The emphasis on coffee proved unfortunate, as prices fell steeply and failed even to cover production costs. Further, the proposed marketing arrangements were never provided. Only 13% of the target of 560 ha of coffee was established.
These limitations were the result of several factors. First, since no research or consultation was involved in the design of the component, it took no account of local production systems or their social and cultural underpinnings (Figure 3)). Second, no attention was given to the development of sustainable alternatives to shifting cultivation. Third, the one-month's training provided to the 18 Public Welfare Department staff was insufficient to make them effective extension workers.
Extension in Wieng Pha was provided by 12 agroforestry extension agents, each covering 4–5 villages. These were temporary project employees, and all were male, ethnic Thais, none speaking any hills language. A range of seeds and some complementary inputs and breeding livestock were distributed, and some temporary increases in production seem likely to have occurred. However, farmers claimed that distribution was arbitrary and inadequate. While new paddy land was to be developed, only 3% of the target of 9000 rai (1440 ha) was met because of limitations of landform and water availability. Although cropping intensities in Wieng Pha were rising and yields falling, little was done to address the need for alternative systems of land management, as the facility's intention to provide research, training and demonstration was never established. The “environmental protection” component in practice consisted of the RFD's reforestation of 1600 ha with Pinus khesia, with villagers participating as laborers. The RFD's policing of land use continued, forcing farmers into practices that they knew to be unsustainable.
Impact on opium poppy cultivation
There was little change in the area of poppy cultivated and both areas continued to export opium (Figure 4). In Wieng Pha, an early decline was followed by almost as marked a resurgence. Pae Por continued to account for a significant proportion of the declining national production of 3200–4800 ha.
Citizenship and land tenure
Thai citizenship and secure land tenure are central to hill peoples' security, livelihoods, and sense of belonging to the broader society. However, residential and land security depend on multiple and sometimes inconsistent criteria and policies. These include the legal status of villages under the 1914 Local Administration Act, the RFD's 1985 classification of watersheds, the 1964 National Forest Reserve Act, and the classification of communities under the 1988 Second Master Plan for Highland Development and Drug Abuse Control. The inconsistency of these rules means that most hill communities' occupation of their land is irregular according to at least one criterion, and they live under the constant threat of eviction, a threat which from time to time is realized.
In Pae Por, Thai citizenship was granted to over 12,000 persons (53% of the population). Beneficiaries regarded this as the project's main achievement. However, none of the land use permits envisaged were allocated, as most of the project area was in Reserve Forest, meaning that cabinet approval would have been required. Land use rights thus remained insecure in Pae Por, providing little incentive to invest in sustainable methods of land management.
In Wieng Pha, there was no systematic attempt to provide citizenship. Nor were any land use permits issued, as this was against RFD policy in watershed protection areas. Fully 39 of the 51 villages in Wieng Pha project area were classified “informal villages without potential” under the master plan, with the ominous implication that the government intended to relocate them. Since none of the villages were legalized during the project, there was little advance in the security of residents.
Comparing the 2 projects and their impacts
Substantial sustainable impact in the social sectors in both projects was not matched by effectiveness in either reducing opium cultivation and use, or in promoting economically and environmentally sustainable agriculture.
There were substantial historical, social, and environmental differences between the two areas, Pae Por being more remote and ethnically homogeneous than Wieng Pha. The histories, cultures, and interests of the implementing agencies also differed markedly. DOLA is the agency at the head of district administration, responsible for grouping communities into villages and registering, overseeing, and taxing them: its role was characterized by a logic of incorporation. The domain of the RFD, in contrast, has historically been unoccupied land and its concern with containment and exclusion. DOLA, while willing to grant secure land titles, was unable to do so. The RFD, although able, was unwilling.
Both viewed the hills minorities through a lens of national integrity tinged by ethnic unease. However, DOLA's aim was the political and cultural inclusion of minority populations, the incorporation of which would allow it to advance its own territorial interests: hence the emphasis in Pae Por on the granting of citizenship, and other initiatives such as “training to create consciousness of being Thai,” the marshaling of “key villages” into numbered streets and houses, and the daily broadcasts of Thai-language radio from loudspeakers set on poles. The RFD's officers directed the project from the fastness of the watershed offices high in the forest rather than from the district headquarters where DOLA and other agencies were located. For them, the officially perceived threat to environmental integrity from the hill tribes' shifting cultivation reinforced their legitimacy as guardians of the forest—which, in the face of pervasive deforestation, was becoming tenuous to the point of irony. While legislation and regulations governing land occupation and use were inconsistent, the RFD's authority against these “enemies of the forest” was ultimately founded on the threat of the consequences for almost any community or household, were the letter of the law to be enforced.
Despite these differences, both projects exemplify a model of development as the spatial extension of state agency staff and infrastructure. The premise that beneficiary populations had little to contribute was reflected in the lack of meaningful consultation, despite the project requirement that “the active involvement of the villagers be solicited in the selection, planning and implementation of project activities,” according to the Wieng Pha HDP.
The centralization of decision-making and the disregard of local cultures and production systems were partly a reflection of the dominance of the vertical, clientelistic culture of both implementing agencies. This explains the lack of flexibility, neglect of the social context, and the high overhead costs that characterized both projects. The priority of outputs over outcomes based on locally defined needs, and the vertical reporting system transmitting information upwards only in terms of predefined indicators—whether or not these were relevant or even accurate—left little incentive or scope for responding to problems identified in the field. Even when shortcomings became apparent, no adjustment to the programs followed.
In the case of the health and education programs, this top-down ethos was not disabling. Nonformal education curricula and health services had been developed for the hills in other areas, and these were delivered by well-managed and relatively responsive sectoral institutions.
For agriculture and land management, locally specific ecological and social factors were much more critical, and here the inflexible culture of implementation combined with inappropriate design assumptions proved fatally asphyxiating. In Pae Por, it was wrongly assumed that profitable replacement crops were extendible, while the importance of land management and tenure security was overlooked (Figure 5). In Wieng Pha, environmental problems were recognized in the abstract, but contradictory policies and administrative sclerosis rendered the chosen strategy non-viable on several counts.
Land rights could not be issued to stabilize land use; and no sustainable land management technologies were identified for extension because poor interagency collaboration hobbled the research component that was to develop them. The RFD fell back upon more familiar, coercive means of regulating land use by suppressing shifting cultivation. The gap between expectations raised and achievements registered was not lost on the inhabitants, one of whom remarked pithily: “Where there is thunder, there should be rain.”
Pae Por and Wieng Pha: key characteristics of project areas, outputs and outcomes. (Sources: Manu 1987; PPHDP 1986; WPHDP 1987, 1989; UNDCP 1991, 1992)