50 MULTIPLE PURPOSE RIVER DEVELOPMENT 



our form of representative government may be the proper subject 

 for a study in social justice, but it lies outside our immediate 

 purview. 



In a purely analytical sense, however, the issue of the distribution 

 of income may arise to plague us unless it is clarified. Our efficiency 

 criteria were derived analytically by taking as given the prevailing 

 distribution of income. Changes in the distribution of income will 

 affect the efficiency of a given allocation of resources. Since the 

 decision to invest or not to invest — or a decision to employ untradi- 

 tional approaches to water resources development — may result in a 

 redistribution of income, it is necessary to contend with both the 

 distribution existing before the event and that reflecting its conse- 

 quences. It is conceivable that what would be an efficient solution 

 in terms of the prevailing income distribution may be less efficient 

 following a reshuffling of costs and gains among the members of 

 the community. While this is a valid theoretical objection to 

 carrying out studies of comparative efficiency, its practical signifi- 

 cance will be negligible for the problems encountered in this study. 

 The income redistribution would affect our efficiency criteria only 

 through the influence on the constellation of relative prices.^* This 

 structure of prices arises out of the distribution of total income 

 among individuals of differing preferences. (We take technology 

 and resource endowment as given for this purpose.) The effect on 

 the distribution of income of the magnitudes we will be treating, 

 however, can be inferred from a numerical illustration. Assume a 

 proposed river basin development program would involve a total 

 investment of a billion dollars over a twenty-year period. The 

 income redistributive consequences of alternative means of under- 

 taking the development may approximate half the total. ^^ If we 

 assume an average annual national income of $500 billion, this will 

 aggregate to ten trillion dollars over the time span considered. A 

 redistribution of income amounting to as much as half the total 

 investment funds committed to the program would represent only 

 five-thousandths of one per cent of total national income. As a 



" We postpone the discussion of a somewhat related problem until Chapter 

 III. 



"Some insight into the quantitative relationships may he inferred from the 

 material presented in Chapters vii and viii, where we treat income redistributive 

 consequences explicitly. 



