In other words, when a particular plot of land is occupied by small and marginal peasants working on household farms it is likely to show higher use of labour than when the same plot is taken over by a large farmer using hired labour. Not all of this may be disguised unemployment, since the fact that small/marginal cultivators also use other non-land inputs more intensively suggests that they would use this additional labour to increase per hectare productivity.

Therefore, the phenomenon that is observed at the all-India level, of increasing landlessness (even in terms of occupancy holdings) of the rural population, may also lead to less employment generation in agriculture.

In addition, households who operate less than 0.4 hectare of land can be described as just below marginal, in that their holdings are unlikely to provide subsistence for all the household members. This means that the possibilities of employing all household members on that particular plot of land are fewer, and they will be forced to seek outside employment. This proportion is also indicated in Chart 1, where it is evident that the total proportion of such households (landless plus marginal) by 1999–2000 was nearly two-thirds of the total number of rural households.

The question then remains, why has the proportion of landless population increased so substantially over these two periods? This may actually be a comment on the reduced viability of small farms, given the various changes that have occurred in the countryside during this time.

One of the more crucial changes has been the virtual collapse of rural credit in the organized sector, especially for small cultivators. The reduced availability of credit, which has been documented, has very severe implications for the functioning of small farms in particular. A number of input costs have also increased, as fertilizer subsidies were sought to be reduced and water rates and other user charges have gone up.

In addition, there is some evidence that although real wage growth slowed down substantially during the nineties, real wage rates continued to increase in most parts of the country. Since the seasonality of agricultural operations means that most cultivators, whatever the size of holding, need to hire in some labour during the peak seasons, this has the dual effect of increasing costs and raising the opportunity cost of own cultivation.

The process of trade liberalization has meant that domestic agricultural prices have less relation to domestic demand and supply conditions and are more correlated with international price movements than before. This means that even when the harvest is poor or when there is crop failure, the cultivators do not get any recompense in terms of higher prices.

There is much greater use of a range of monetized inputs, including new varieties of seed marketed by major multinational companies. Small cultivators who take on debts (often from informal credit sources at very high rates of interest) in order to pay for these cash inputs find themselves in difficulty if for some reason there is crop failure or output prices remain low.

All these could be why the proportion of rural households not operating any land increased so much over this period. This in turn would mean that are less people reporting themselves as self-employed in agriculture (which is what is observed), and a general reduction in employment generation because less people would be hired than had lost employment from own cultivation.

Is this aggregate pattern reflected across individual states? Charts 3 to 17 provide some indication of landholding changes (in terms of the proportion of landless and marginal households to the total) along with rates of growth of agricultural employment generation over these two periods, in the major states.
Chart 3 >> Chart 4 >>

In these charts, the columns describe the proportion of landless and marginal (that is, those operating holdings less than 0.4 hectare) households in the years 1987–88, 1993–94 and 1999–2000. The line describes the growth of agricultural employment in the first and second periods.

It is clear that the pattern of greater landlessness is common to all the states. A significant majority of the states also show substantially reduced agricultural employment growth. The extent of the deceleration varies from state to state but it is usually quite sharp.
Chart 5 >> Chart 6 >>

Only three states show the opposite tendency in terms of agricultural employment-Haryana, Punjab and West Bengal. In the case of Haryana (Chart 7), this is not of much significance because in both periods agricultural employment growth was negative.
Chart 7 >> Chart 8 >> Chart 9 >> Chart 10 >>
Chart 11 >> Chart 12 >>

In Punjab (Chart 13), agricultural employment moves from a negative rate of –1.14 per cent to a positive rate of 1.47 per cent, which is a reasonably large shift and the opposite of what has occurred in most other parts of the country. It should be noted, however, that in this state the proportion of landless households remained broadly constant in the latter period, suggesting that the role of changing land relations was minimal.
Chart 13 >> Chart 14 >> Chart 15 >> Chart 16 >>

West Bengal (Chart 17) presents the most interesting case. Here, there has been a significant increase in the landless population (up by 10 percentage points of total households); yet, agricultural employment appears to have accelerated over the latter period, going up from a negative rate of –0.36 per cent to a positive rate of 0.54 per cent. This process clearly deserves further examination. It is possible that cropping-pattern changes have been such as to induce greater labour-use, and there may have been other changes in the countryside which have translated into greater availability of work in agriculture.
Chart 17 >>