Social Welfare Underpinnings of Urban Bias and Unemployment
构建模型论证，由于农村公共服务（如教育、医疗、电力）成本高于城市，从社会福利角度看存在歧视农村居民的激励，这为城市偏向和失业问题提供了新的解释。
The reallocation of labour from rural to urban areas, although an historical concomitant of growth in per capita incomes, is widely viewed to be a troublesome process. It is a general belief that poor countries are somehow over-urbanised and migration excess'ive, that urban unemployment or underemployment is a problem, and that the investments to accommodate an expanding urban population are unjustifiably burdensome. Many governments have reacted to migration by attempting to control it via internal passport systems and the forced removal of migrants to rural areas. Mozambique is a recent case in point. These concerns contrast sharply with the conclusions of models of the migration process. For instance, Bhagwati and Srinivasan (I974) argue that input and output tax-subsidy schemes can solve all policy problems raised by migration, without recourse to direct interference with the migration decision.' In this paper we develop a model suggesting greater difficulties in designing policy toward migration and the allocation of labour between rural and urban areas than those identified in the existing literature on this problem. The critical factor responsible for our conclusions is that some goods, notably education, medical care and electricity, cost relatively more in the rural areas than in the urban areas of LDCs. While these cost differentials are not well documented (Linn, I982), we believe that they are substantial and help to account for the relative dearth of theses goods and services in the countryside. For instance, Turvey and Anderson (I 977, p. I 62) consider that the cost of providing electricity in rural areas is typically 2-5 to 4 times higher than in the cities. This differential-cost phenomenon has implications for the efficient allocation of resources in LDCs and for the rural-urban distribution ofincome. It bears on a variety of decisions: unemployment policy, shadow wage pricing of labour, migration policy, taxation of inputs and outputs, and the provision of public services. In effect, agriculturalists are inefficient or high-cost consumers as a result of their location. As a consequence, there is an incentive, from the social welfare viewpoint, to discriminate against rural residents if feasible. This general type of conclusion in a situation of an asymmetric utilityT possibilities frontier has been noted by other researchers: see Mirrlees (I 972) and Arnott and Riley (I 977). It has been dubbed 'the unequal (ex post) treatment of (ex ante) equals' in the first-best optimum. We develop these ideas in the context of the dual economy with differential costs of consumption in Section II.