THERE SEEMS little room for doubt that the United States could support for economic security one and a half to two billion more a year than were spent in the years from 1933 to 1937. We could provide about four billion for these purposes instead of the two and a half billion that we did provide. This conclusion was reached on the basis of our estimates of potential tax capacity and our comparisons of tax collections with the national income. Moreover, the economic effects of raising money for security need not be injurious to business if more progressive taxes are used to obtain the funds. Such a program will not necessarily restrict business activity, reduce investment, or cause inflation. It is more likely to create the purchasing power needed for genuine recovery. Under such conditions security expenditures may be good for business, if they are accompanied by some governmental program which will stimulate private investment. Having arrived at the conclusion that adequate funds may be collected for security without undue injury to business, we should then ask: how should these funds be distributed? Must they be spent without any significant relationship to need? Did the FERA compromise with the ability principle because it is impossible to work out a system to distribute security funds according to need? If not, how should we measure need in any system set up to distribute funds on this basis?