This study is based on field survey of households involved in Swidden in Orissa, India. Conventionally, swidden agricultural system has been interpreted as inefficient (economically), destructive (ecologically) and an inflexible static form (institutionally) of agriculture. The economic efficiency in swidden, in particular poses an analytical problem, largely due to the institutional structures and multiple crop situations involved in this system. Conventionally, efficiency of farm performance is understood through the ratio of output and inputs wherein productivity per unit of input is the undercurrent of efficiency. In this context, the paper attempts to clarify some of the concepts relating to economic efficiency in swidden agricultural system, measure technical efficiency and identify the key factors affecting efficiency of swidden agricultural system practiced in the region. Results suggest that swidden in particular is not an inefficient form of agriculture, particularly from energy use point of view.

Conventionally, the efficiency of farm performance is understood through the ratio of output and input wherein the productivity per unit of input is the undercurrent of efficiency. Therefore, it is also computed through total factor productivity. But this differs across the farming systems. Economic efficiency of swidden is an analytically complex issue to handle because of the institutional structures and multiple crop situations in this system. It is argued that swidden is a suboptimal method of cultivation where the output is low, and the labor use intensity is very high (Mahapatra and Debi, 1973; Hinton, 1978; and Komkris, 1978). At the core of the problem lies the argument of economic inefficiency of this form of agriculture. There are also views that support swidden as an economically efficient system (Ramakrishnan, 1992). Since swidden involves a rotation of fields, forest regeneration becomes an integral part of the system. This recycling of land use is a form of conservation practice which prevents the wastage of land. It is also observed by many researchers that swiddening can be a stable system with minimal cutting of new forests (Fox et al., 1999; and Xu et al., 1999).

Although researchers have arrived at different estimates in determining the self-sufficiency among swidden cultivators, the general conclusion is that they are able to feed themselves adequately (Chin, 1989). Ramakrishnan (1992) also studied the average yield of grains and seeds, leaves and fruits and tuber and rhizomes under different shifting cycles in the northeastern part of India. The results reveal that the production is higher than the needs of an average household size, and can be produced in a short cycle of five years. But average yield alone does not reveal the economic efficiency of the swidden system, unless the output-input ratio is very high (both in energy as well as in monetary terms) when compared to the alternative systems of agriculture.