Aggregate Analysis of Research and 

 Productivity in Agriculture 



The study reported in this chapter is an aggregate analysis of pro- 

 ductivity in thirty-six countries (see table 5.2). The data base uti- 

 lized is similar to that used in an important earlier study by 

 Hayami and Ruttan (1971). We have added production data for a 

 recent year and have research data unavailable to them. In view 

 of the significance of their work, we have attempted to maintain a 

 degree of comparability. 



Specification 



The framework for the analysis is a production function frame- 

 work. The specification incorporates the possibility of intercoun- 

 try productivity ("level") differences and within countries over- 

 time C'trend'') productivity changes. The production functions of 

 two countries are depicted in figure 5.1. The observed input-out- 

 put combinations are the points a^.a--,, h^, b-^. The shifts from a-^ to 

 Oy in country A and from b^ to b-y in B reflect both changes in in- 

 puts (conventional and others) and a time-shift of the production 

 functions. The figure shows country B's productivity level as 

 being higher than that of A (the productivity differences are 

 measured from sample period's averages). 

 The algebraic form of the production function is:' 



y^,=f(X)e^j'yj' (5.1) 



1. See Mundlak (1970) for a similar specificaiion. 



78 



