Economic analysis of leader compensation in communal irrigation systems of northern Thailand

Abstract

Communal irrigation systems differ in how irrigation leaders and their assistants are compensated by system users. In Northern Thailand, compensation is generally on either an exemption or fee basis. No satisfactory explanation for this has been given. This aspect was chosen for study partly because it was thought likely to bear on water distribution and hence on the economic efficiency of such systems, and partly to test the applicability of an economic approach, particularly principal-agent theory, which has been used successfully to explain other agricultural contracts. This study develops and tests two theories to explain the different compensations. The first theory uses a "principal-agent" approach, the thesis being that the existence of one or the other type of pay has an efficiency rationale in terms of net gain associated with the principal-agent relationship between irrigation committee and users of a system: Choice of agent compensation is argued to be largely determined by comparative cost advantage relative to the benefit of internalizing the potential "externality" of overappropriation of water by users at the head of a system. Factors considered to affect this choice are value of scarce water, distance between head and tail users along the main canal, number of users, costsharing benefits of head users, number of villages, tenancy rate, and political vote power of head users relative to tail users. The second theory argues in terms of "more pay for more work" subject to practical constraints to the exclusive use of one type of pay. The theories may be combined into one, since each can accommodate the other. The results of the hypothesis testing reveal that using both theories leads to better explanation of the results than does either theory alone. The most successful explanatory variables were value of scarce water, tenancy rate, and leader workload. There was no evidence to suggest "political failure" due to head-user majority voting power being used to institute a "weak" leader compensation that might adversely affect water distribution to tail users. While the results give some support to both theories, further empirical and theoretical work appear called for.