1 1 8 Agricultural Research and Produciivily 



evaluations that lADP has not been a primary source of real 

 growth. 



Nonetheless, it surely cannot be concluded that it has not had a 

 major impact. It clearly induced modern input adoption. The fact 

 that it did so without inducing more total factor productivity 

 growth suggests that implicit or explicit subsidies for fertilizer and 

 tractor purchases are not effective means of purchasing real eco- 

 nomic growth from a social point of view. We should note, 

 however, that the total factor productivity calculations attributed 

 perhaps too much production growth to these modern inputs, 

 since the prices used to compute the share weights were market 

 prices. If prices were actually lower to I ADP farmers, a calculation 

 based on these lower prices would have resulted in a higher 

 growth in total factor productivity in the lADP districts, since less 

 output growth would have been attributed to the modern inputs. 

 From a social point of view, however, the subsidies, to the extent 

 that they were undertaken, represent inefficiencies. In all likeli- 

 hood they also had a regressive effect on income distribution, 

 though we have not addressed ourselves to distributional effects. 

 Some ''learning from experience" associated with the modern in- 

 put use would be of economic value, but we do not see evidence 

 thai Indian farmers are unable to learn about and adopt new 

 inputs. 



In terms of economic payoff to the IAD? program, we have 

 from regression 1 , table 6.9, an estimated 2 percent higher level of 

 output for the 1960-71 period in the lADP districts. The value of 

 this output in the fifteen lADP districts is approximately 75 

 million rupees per year (1968 prices). Presumably, it is increasing 

 over lime and will continue beyond 1971 but will not be perma- 

 nent. The estimated costs of the lADP programs were from 30 to 

 40 million rupees per year in the early years, rising to 50 million 

 or so in later years. It appears from these data that the flow of 

 social returns generated by the program has been sufficient to 

 yield a reasonable rate of return. The actual rate will depend on 

 the permanence of the benefits stream in the years after 1971 and 

 the time lag between program spending and results. On the 

 assumption that the benefits flow extended to 1975, the internal 

 rate of return has been in the neighborhood of 15 percent. This 



