Controversy regarding Farm Size and Productivity

In this article we will discuss about the controversy regarding farm size and productivity.

The genesis of the problem can be traced to the chronic food shortage which India has been facing ever since the separation of Burma. The situation, aggravated by the Second World War set in motion a thinking about the need for a policy for agricultural development. Though some programmes had been launched even earlier, serious discussions started in the post- independence era only.

The discussion relation to the strategy for agricultural development brought into prominence two schools of thought- one based on technological approach and the other on institutional reforms.

The latter school addressed itself, among other things, to the question of farm organization. The major objective of this approach was to effect area-organization of the units of production (i.e., farms) to achieve high level of productivity and efficiency through appropriate land reforms.

It is necessary to define the terms “productivity “and “efficiency”— the terms used in the proceeding paragraphs. By productivity, we mean the gross output per unit of land. Efficiency on the other hand, is denoted by the surplus of value of output over all costs (including the imputed value of inputs contributed by the farmer himself and his family).

Productivity, for the present purpose is measured per unit of net area rather than that of gross cropped area. The former automatically takes care of the effect of intensity of cultivation of land associated with different size classes of farms.

The initial debate on farm size policy and institutional reforms was based on a priori reasoning. The emergence of “Studies in the Economics of Farm Management” in the mid-fifties, however, provided economists with a statistical base to make valid generalizations.

The main issues involved in the debate on farm size and productivity are nicely summed up in the following three observations made by Amartya K. Sen (1962):

(1) “When family labour employed in agriculture is given an imputed value in terms of the ruling wage rate, much of Indian agriculture seems un-remunerative.”

(2) “By and large, the profitability of agriculture increases with the size of holding, ‘profitability’ being measured by the surplus (or deficit) of output over costs including the imputed value of labour.”

(3) “By and large, productivity per acre decreases with the size of holding.” In other words, there exists an inverse” relationship between farm size and productivity.

It is clear that the first of the three observations concerns itself with opportunity cost of (family) labour and the relevance of market wage rate for the valuation of family labour. The second and the third of Sen’s observations provide contradictory evidence as far as productivity and efficiency are concerned. While economic efficiency criterion lends support to large sized farms, the consideration of productivity per acre points in the opposite direction.

Khusro (1964) subjected the Farm Management data to statistical analysis and confirmed the existence of an inverse relationship between farm size and productivity. But when he corrected acreage for differences in soil fertility by using land revenue as a correction factor, the inverse relationship yielded place to constant relationship.

Khusro’s findings, however, cannot be accepted because of:

(i) The aggregation bias,

(ii) The weak statistical significance of correlation coefficients and

(iii) The limitations of land revenue as an index of soil fertility.

Sen (1962, 1964) expressed doubts about the statistical validity of the observations which he made on ‘a priori’ reasoning, using aggregate size class data as published in the Farm Management Studies. Ashok Rudra (1968) too, challenged the statistical validity of the inverse relationship that was revealed to exist on the basis of aggregated data.

He preferred to estimate rank correlation coefficients between farm size and productivity. Out of the 17 coefficients, 15 had a negative sign (indicating the existence of inverse relationship) and 9 of these were statistically significant too. Rudra’s analysis too, suffers from aggregation bias. His measure of productivity based on gross-cropped area also imposes serious limitation on accepting his results as relevant and valid in the present context.

Saini (1969-1971) using disaggregated data for 9 States over a period of time ranging from two to three years provided firm evidence. Out of the 25 coefficients estimated by Saini, 22 had a negative sign and 18 of these were statistically significant also. In terms of confidence limits, the cases with positive signs also did not rule out the existence of inverse relationship between farm size and productivity.

Thus, according to Saini, the inverse relationship turned out to be a confirmed phenomenon in the pre-green revolution in Indian agriculture. These conclusions were also confirmed by Deepak Mazumdar who points out “As the size of the farm decreases, output per acre increases.” C.H. Hanumantha Rao (1966) also supported this thesis.

We may here point out that some other research workers, analyzing the same Farm Management data collected during the fifties arrived at yet another conclusion. According to these research workers who included Krishna Bhardwaj (1974), A.P. Rao (1967) and Usha Rani (1971) there was no evidence of a significant change in production per acre as the size of the farm changed in any direction.

Broadly speaking, all of them were of the view that yield per acre did not change as the size of the farm changed. Chattopadhyay and Rudra (1976) also expressed their reservation about the inverse relationship about the inverse relationship between farm size and productivity.

Thus, what emerged iron the debate concerning size of the farm and productivity in the pre-green revolution agriculture was quite confusing and conflicting. Analysis of the same data yield different conclusions. This was mainly due to the different statistical techniques employed by these research workers.