Investment in Agricultural Research and Extension: A Survey 27 



Improvement Center (CIMMYT) in Mexico, and the Interna- 

 tional Rice Research Institute (IRRI) in the Philippines, have 

 been in existence long enough to have had a significant impact on 

 production. The high-yielding Mexican wheat and the "miracle" 

 rice varieties were developed in these centers. Undoubtedly, the 

 work done in these centers had a very significant impact"* though, 

 as we will show in later chapters, they have by no means domi- 

 nated the discovery of new technology in developing countries.^ 



The advantages of the centers are that they can attract out- 

 standing scientists and provide them with scientific equipment, 

 experimental plots, technical assistance, and an "environment" 

 that cannot readily be provided in the national systems. They are 

 not subject to the administrative and political constraints that 

 may inhibit the productivity of the national organizations. 

 Perhaps the most important advantage of these centers lies in the 

 scale of their operations. A theoretical model is presented in 

 chapter 8, in which research work is viewed as a search for 

 superior technologies. In terms of this model, the manifestation 

 of the larger scale of operation of the international centers is that 

 they have much larger sample sizes and higher variance popula- 

 tions to search than the national systems, and thus they can be 

 expected to be more productive. 



The major disadvantage of the international centers is that they 

 are very expensive. Expenditure per senior scientist in 1973 ap- 

 proximated $100,000 in some of the centers. This generally ex- 

 ceeds the comparable expenditure for even the most expensive 

 developed country systems. In the United States, expenditures 

 per scientist man-year were approximately $70,000 in the animal 

 sciences and $50,000 in the crop sciences in 1970 (Evenson and 

 Welch 1973). The international centers would be roughly compa- 

 rable to the most costly U.S. stations. The same research budget 

 expended in national research systems in less developed coun- 

 tries would purchase at least twice as many scientist man-years at 

 the international centers do. 



4. See, for example, Dalrymple (1972) who summarizes the spread of the high- 

 yielding varieties created in the centers. 



5. The effect of the national systems is one of the main questions studied in this 

 volume. Chapter 5 reports some results incorporating the contribution of the in- 

 ternational centers. Chapter 6 addresses the issue in a developing country— India. 



