36 SOCIAL SCIENTIST The first approach emphasizes the importance of marketable surplus from agriculture mainly from the point of view of the urban sector which must have an adequate supply of wage goods to support its labour at as low a cost as possible. Studies of this type have stressed the need for a balanced growth of agriculture and industry and have cautioned against a policy of industrialization at the cost of agriculture. These studies have been based both on a priori deductive reasoning1 and empirical evidence derived from comparative experience of different countries.3 Industrialization creates additional purchasing power of such magnitude (particularly as it has to be fed by deficit financing in the absence of genuine savings) that an inflationary process is generated unless the supply of wage goods or marketable surplus is elastic enough. This inflationary process is distinctly regressive in its impact as it hits the poor, who spend the largest part of their income on food, the hardest. Moreover, it restricts the domestic demand for manufactured goods, reduces savings and dampens incentives to invest.3 The recognition of marketable surplus of food as a critical constraint also led a few to recommend the use of concessional imports at least marginally to ease the crisis. It is recognized now, however, that the economic and even political costs of such a solution can be prohibitive. Moreover, it is not a feasible solution for a country with a large population and for long periods of time. Generation of Marketable Surplus The obsession with marketable surplus as an overriding constraint has kd quite a few economists to insist that agriculture must first be developed enough to generate a sizable marketable surplus before any steps towards industrialization are taken. These are, however, half-truths which leave many aspects of the development process unexplained. The extremely limited perspective of such development models accounts for limited explanatory power. First of all, the models do not explain the process of the generation of marketable surplus. There is even an implicit assumption that marketable surplus can be generated independently of the development of the nonagricultural sector and that it is just a question of increasing agricultural production. We have to realize however that even agricultural production will not increase by itself unless there is a chance for the increased production to be sold. Also, we cannot have a significant and sustained jump in marketed surplus from the agricultural sector unless there is a