The lost secret of aid efficiency

By Simon Scott, Counsellor, OECD Statistics Directorate

Wilson Schmidt
Wilson Schmidt (1927-81). Photo courtesy of Lisa Hill Corley and George Mason University

In 1963 John Pincus of the RAND Corporation suggested redefining aid to reduce all forms of aid to their value as grant or subsidy, and in 2014 the OECD’s Development Assistance Committee (DAC) agreed to his suggestion. (See an earlier post about this evolution.)

Of course, in the meantime, DAC members had not just been sitting around for 51 years. In 1969 they used Pincus’ “grant equivalent” method in a Recommendation to soften the terms of aid, and in 1972 they used it again to decide which loans were soft enough to count as official development assistance. The World Bank and the IMF also used the method in measures to keep the lid on developing countries’ debt, and the Paris Club used it to equalise creditors’ efforts under different debt relief options.

Yet one potential use of grant equivalents has been thoroughly neglected. Ironically it was the very application of the method that was most discussed 50 years ago, namely its potential to ensure the most efficient use of aid funds.

The idea originated with a brilliant young American economist, Wilson Schmidt. Schmidt proceeded from the realisation that the grant-equivalent of a loan was not the same for the lender as for the borrower. The lender’s sacrifice is the difference between his expected repayments and what he could have earned by investing elsewhere. But the borrower’s benefit is the difference between his expected repayments and what he will earn from the project he finances with the money.

Take the example of a loan from A to B at 2% interest, which both A and B expect to be repaid. The lender, A, could have invested safely at 6%. So A’s sacrifice is 4% a year. But the borrower, B, will use the money on a project with a 10% return. So B’s benefit is 8% a year, and the loan is worth twice as much to him as it costs A.

Eric Schmidt and James Buchanan
Eric Schmidt and James Buchanan before Wilson Schmidt’s photograph at the Center for Study of Public Choice in 2012. Photo courtesy of George Mason University.

The loan’s duration also affects relative grant equivalents. Say this loan was for 6 years. A’s sacrifice is thus 4% for 6 years, or 24% of the nominal value of the loan. B’s benefit is 8% a year, or 48% of the nominal value of the loan. B could get the same nominal value from a loan for 12 years with a 4% annual benefit, i.e. at 6% interest. This would reduce A’s sacrifice to zero, since his alternative investment only pays 6%.

Schmidt saw the opportunity to drive aid dollars further through longer loans at higher interest rates. But the conclusion was somewhat veiled by his formal presentation of all possible scenarios. In fact, he put more emphasis on the relatively unlikely instance of returns being higher for the donor than the recipient. In this case, it would be more efficient for the donor to give a grant, since he could not lend at the borrower’s rate of return without taking a loss. Pincus was initially inclined to dispute this point, but cried uncle on it in 1967.

The lesson from Schmidt’s work emerged more clearly in 1965 when Richard Cooper, working independently, zeroed in on the need to compare donor and recipient grant equivalents to identify “optimum terms for foreign lending.” He then did the job himself and concluded that “[t]he terms of foreign assistance…could very likely be modified, by lengthening maturities and raising yields, to the mutual benefit of donors and recipients alike.”

Yet even after Cooper’s crystal-clear presentation, the insight never caught on with aid donors. Many of them gave up lending entirely in the 70s or 80s. Only a faint echo of the idea persisted in development banks’ requiring minimum rates of expected return to approve project loans.

Perhaps the neglect arose because the protagonists swiftly moved on to other topics, and to public service. In 1975, the then (and now in 2017!) California Governor Jerry Brown appointed Pincus to the State Board of Education, and Presidents Johnson and Carter both appointed Cooper to top jobs in the State Department.

Schmidt’s career also flourished. In 1966 he became head of the Economics Department at Virginia Tech, where he was instrumental in advancing public choice research through his appointments of Charles Goetz, Gordon Tullock and James Buchanan. In 1970 President Nixon appointed him Deputy Assistant Secretary of the Treasury, and in 1981 President Reagan nominated him to be the United States’ Executive Director at the World Bank, where he could have given effect to his insights on aid lending.

But it was not to be. While awaiting Senate confirmation of his appointment in the summer of 1981, Schmidt was staying at the prestigious Cosmos Club near Dupont Circle in Washington, D.C. On the night of Sunday, July 19, a fire broke out in his room, apparently started by a discarded cigarette in an overstuffed chair. He was found unconscious from smoke inhalation in his bathroom and taken to the hospital, where he died two days later.

While Schmidt’s life and career were tragically cut short, his brilliance and entrepreneurial flair lived on in his son Eric Schmidt, who became the chief developer of Java and later CEO of Google. (Eric Schmidt is also Chairman Emeritus of New America, with whom my colleague William Hynes and I recently tried to put the SDGs in order.)