MacroScan - Agriculture's Role in Contemporary Development

The evidence of an acceleration in GDP growth in India increases, aided on occasion by periodic revisions in the base year used for constructing the estimates. While the factors accounting for this acceleration are still being debated, another unusual feature of this growth since the 1980s has received less attention: the growing disproportionality between agricultural and non-agricultural growth. Using GDP figures with 1993-94 as base, S. Sivasubramonian estimates that the rate of growth of GDP in the agricultural and allied sectors rose from 1.5 per cent during 1970/71-80/81 to 3.43 during 1980/81-90/91, and then declined to 2.97 per cent during 1990/91-99/00. Over the same periods the rate of growth of non-agricultural GDP accelerated from 4.38 per cent to 6.37 per cent and 7.14 per cent respectively (Chart 1). (The figures for output in the form of ''services of dwelling'' are separated out from that of the non-agricultural sector for these calculations.)
Chart 1 >>

What is particularly remarkable is that the acceleration of non-agricultural growth during the 1990s was accompanied by a decline in the rate of agricultural growth. The disproporationality is visible even when the comparison is restricted to industrial and agricultural growth (Table 1). Figures for the initial years of this decade indicate that this disproportionality in growth has only increased since.
Table 1  >>

These sectoral trends suggest that domestic agricultural growth is now not a constraint on the growth of the non-agricultural sector. This does mark a structural shift in the pattern of growth, when compared with the first three decades of post-Independence development, when the agricultural bottleneck was seen as an important factor responsible for the failure of the strategy of development based on the Mahalanobis model. The argument was that the Mahalanobis strategy underestimated the agricultural constraint by treating agriculture as a bargain sector in which output growth could be accelerated without much investment, by making suitable institutional adjustments.

There were three forms of intersectoral linkages between the agricultural and non-agricultural sectors that were seen as important. First, with the agricultural sector accounting for 61 per cent of non-residential GDP in 1950/51 (at constant 1993-94 prices) and 76.2 per cent of employment, demand from the agricultural sector was seen as crucial to sustaining the demand for non-agricultural products and services, especially manufactured products. Second, since agricultural commodities constituted a significant share of input costs in some industries and of the wage basket in most, increases in agricultural prices were variously analysed as affecting industrial production. In particular, if an industry was agro-based or was characterised by a tendency for money wages to rise with increases in the prices of wage goods, it would experience an increase in costs that may not be neutralised by an increase in final product prices. In the event, profits could be squeezed and manufacturing investment affected adversely. Thirdly, increases in agricultural prices would constrain the growth of demand in the manufacturing sector, since consumers would allocate a larger share of their incomes to food consumption and a smaller share to manufactures demand and the government may reduce public expenditure to reduce absorption and dampen price increases. This constraint on demand growth would also adversely affect the ability of firms in industries producing mass consumption goods to raise prices in order to cover higher costs.

These different ways in which agricultural performance was expected to affect non-agricultural growth were predicated on the operation of two transmission mechanisms: first, increases in non-agricultural growth were expected to result in increases in the direct (inputs) and indirect (wage goods) demand for agricultural products. Second, since, agricultural growth was seen as constrained from the supply side, any disproportionality in industrial and agricultural growth was expected to result in an abnormal increase in the prices of agricultural goods, since those prices were largely determined by the relative levels of supply and demand.

In the aftermath of the agricultural crisis of the mid-1960s, this problem was compounded by the fact that the provision of support to agricultural production in the form of cost-plus remunerative prices, offered a floor price that encouraged speculation. This was because if speculative hoarding was not followed by the expected increase in prices, stocks could be disposed at the cost-plus support price, which reduces the risk of large losses. As a result, increases in demand relative to supply inevitably raised prices, whereas increases in supply in years of a good harvest did not result in any significant decline in market prices.

It needs to be noted that these mechanisms are operative only if there are limits on altering domestic supply with imports. If foreign exchange can be accessed easily to finance such imports, the structure of domestic supply need not be largely determined by the structure of domestic production. Commodities in whose case domestic demand exceeds supply based on domestic production could be imported to hold down the price level. During the 1950s and early 1960s, India faced a binding balance of payments constraint, since access to foreign exchange was limited too export revenues, limited FDI inflows and flows of capital through the bilateral and multilateral aid network. Yet, the economy witnessed rapid non-inflationary growth in manufacturing even when agricultural growth was moderate because of access to food imports through the P.L. 480 route, which enhanced supplies and helped dampen price increases. It was when access to such imports tapered off that the agricultural constraint proved binding, leading to the deceleration of manufacturing growth during the late 1960s and 1970s.

It is in this background that we need to assess the changed circumstances of the 1980s and 1990s, especially the latter decade, when the disproportionality in non-agricultural and agricultural growth widened considerably, without triggering inflation and limiting non-agricultural growth on account of an inflationary barrier. In fact, changes in the environment and pattern of growth triggered tendencies that prevented the realisation of the denouement expected based on the late-1960s and 1970s experience.

The relevant change in the environment was the transformation of the world of international finance that, for the first time, provided ''emerging markets'' like India access to private international finance. It is now widely held that the Indian government exploited that opportunity during the 1980s, to overcome the development impasse of the 1970s. Deficit-financed expenditure was used to accelerate non-agricultural growth, and the resulting disproportionality between non-agricultural and agricultural growth was managed by using imports financed largely with external debt to change the structure of domestic supplies and dampen inflation. And as Chart 2 indicates this was even more true of the 1990s than was true in the 1980s.

A second factor allowing for growing disproportionality between agricultural and industrial growth is a change in the pattern of demand and production, involving a reduction in the direct agricultural-input dependence of the non agricultural sector. As Sastry e/. al. (p. 2392) have shown, the available input-output tables for the Indian economy indicate that: ''In 1968-69 one unit of rise in industrial output was likely to enhance demand from agriculture by 0.247 units, which was reduced to 0.087 by 1993-94. On the other hand, in 1968-69, one unit rise in industry was to cause 0.237 units demand from the services sector, which increased to 0.457 units in 1993-94.'' (Table 2).
Chart 2  >>

This reduction in agricultural input dependence of the non-agricultural sector would be greater once we take account of the growing share of service in non-agricultural GDP. While services accounted for 43 and 48 per cent respectively on the increment of GDP at current prices in the 1970s and 1980s, the figure rose to 58 per cent and 62 per cent respectively during the 1990s and the years 2000-01 to 2004-05. Given the much lower agricultural input dependence of services, this would have strengthened the tendency noted above.
Table 2  >>

Thirdly, growth in both the agricultural and non-agricultural sectors has been such that the employment elasticity of output growth has been falling over time. This means that employment growth has been increasingly short of economic growth and output per worker has risen significantly in the non-agricultural sector where output growth has been particularly high. While a part of this rise in output per worker may have meant an increase in the wages of sections of the already employed, it would principally mean an increase in income inequality because of an increase in managerial salaries and profits. Both these tendencies imply that the indirect demand for agricultural wages goods would grow at a much lower rate than output partly because of the slower growth in employment and partly because increases in per capita incomes accrue to those whose demand for food is satiated.

What is more, an ongoing study by Abhijit Sen based on recent NSS suggests that even among the relatively poor the share of income allotted to food consumption is being squeezed by the growing requirements set by expenditures on health, fuel, transportation and education. The collapse of public provision in some of these areas, requiring purchases from private suppliers, and the increase in prices in others, is responsible for the enforced shift away from food consumption in the household budget.

The net result of all this is that agriculture is increasingly faced with a growing demand constraint at a time when input costs are rising. This is a reversal of the situation prevalent till the 1980s when the agricultural supply constraint constituted a barrier to rapid non-agricultural growth. As a result, as Chart 3 indicates, the input-output price parity in agriculture, which moved in favour of agricultural producers during the 1980s, has stagnated and moved against agricultural production during the liberalisation years since the early 1990s.
Chart 3  >>

The consequence of these recent trends is that the Indian economy can record the observed creditable rates of growth of aggregate GDP even when the agricultural sector languishes. A feature of the growth process in a more open and liberalised environment is that the peasantry has a much smaller a role in sustaining economic growth and can thus be partially excluded from development. This is partly reflected in the fact that agriculture accounted for just 21 per cent of GDP in 2004-05. But neither the peasantry nor the landless labourers dependent on agriculture shrink as fast, given the pattern of agriculture growth. Employment in the agricultural sector amounted to as much as 60 per cent in 1999-2000, a decline of just 16 percentage points since 1950-51. It bears emphasising that these outcomes of the patterns of growth underlie the agricultural crisis and agrarian distress being reported from different parts of the country, at a time when the non-agricultural economy is on a roll and GDP is rising rapidly.