Advice to Invest Less in U.S. Bonds

Lawrence H. Summers presided this month over his last commencement as president of Harvard. He has been known for making impolitic remarks.
Lawrence H. Summers presided this month over his last commencement as president of Harvard. He has been known for making impolitic remarks. (By Chitose Suzuki -- Associated Press)
By Paul Blustein
Washington Post Staff Writer
Thursday, June 22, 2006

A former U.S. Treasury secretary is advising some of the world's biggest holders of U.S. Treasury bonds that they ought to find much better ways to invest their money.

Lawrence H. Summers, who headed the Treasury in the last 18 months of the Clinton administration, has argued in recent speeches that developing countries in Asia, Eastern Europe, Latin America and Africa should put much of their excess funds into stocks. Too often, he contends, the central banks of those countries invest their hoards of foreign securities -- now totaling several trillion dollars -- in safe but low-yielding U.S. Treasurys.

The return "will be zero" on those Treasurys after inflation and currency changes are factored in, Summers said in a lecture last week at the Center for Global Development, a Washington think tank. Meanwhile, he said, the developing countries are passing up much more lucrative investments -- "this, in societies where hundreds of millions of people are still desperately poor." In another speech, this one in Bombay a few weeks ago, he said, "It is striking to estimate the cost to developing countries" of their Treasury-heavy portfolios.

Given Summers's previous job as the official with chief responsibility for financing the U.S. government, there is an obvious paradox in his suggesting that any central bank disinvest in U.S. securities. In a telephone interview yesterday, he said he saw no incongruity in his position, because he is not urging wholesale dumping. He said central banks around the world must keep "large volumes" of their money in super-safe assets such as Treasury bills. And "any diversification" into riskier investments such as stocks "is not likely to be rapid, in ways that would affect" the Treasury's ability to borrow at affordable interest rates, he said.

Summers, who announced in February that he will step down as president of Harvard University, is known for his impolitic remarks -- a famous example being his suggestion that "intrinsic aptitude" could explain why fewer women than men have excelled in science and math. So his argument against Treasurys might seem to be just another case of his penchant for flippancy.

But Summers has developed an elaborate rationale for his proposal, and he still commands attention in policymaking circles. The World Bank, where Summers once was chief economist, has shown considerable interest in the idea.

Summers's proposal is based on fundamental shifts in the global financial system that arose well after he left office -- in particular, an immense buildup in the reserves of foreign exchange held by developing countries' central banks. Those reserves have grown as the United States, with its burgeoning trade deficit, imports goods from abroad.

The dollars Americans spend on foreign products eventually end up in the hands of central banks overseas, and the central banks invest the proceeds largely in U.S. government securities. They do so in part because they want to protect themselves against financial crises of the sort that struck Thailand, Indonesia, South Korea, Russia, Brazil and Argentina a few years ago.

For a developing country, accumulating a big war chest of dollars can help discourage speculators from trying to drive down the value of its currency.

But the upshot, in Summers's view, is "the central, global financial irony of our times": Countries that need capital to finance rapid development are shipping more money to the United States than is flowing in the opposite direction -- and it is their official policies to do so.

Suppose, he said in his lecture, "you were on Mars, and you had not seen planet Earth, but you had studied economics, and someone said there are these substantial number of countries that are growing at 4, 5, 6, 8, 10 percent a year that are relatively poor, and there are these other countries that are rich, aging, growing at 2 percent a year, 3 percent a year, 4 percent a year perhaps, with slowly growing populations."

Although such a Martian economist would assume otherwise, "the flow of capital is actually very substantially from poor countries to rich countries, and in particular it is from poor countries to the world's richest and most powerful nations -- on a scale never before contemplated or seen."

Moreover, Summers said, the chief explanation is not "capital flight," in which wealthy foreigners in unstable countries park their savings in the safe haven of a U.S. bank account. Rather, it is the collective decisions of central bankers as dollars roll into their reserves from the export of goods to the U.S. market.

But the reserves that have piled up, Summers said, are "far in excess of what is necessary" to defend against a crisis -- probably more than $2 trillion too much, he estimated, based on the most commonly used guidelines. Moreover, "no one could suppose that these are going to be high-return investments," because the interest rates on U.S. Treasurys are about 2 percent after inflation, and for developing countries even that paltry yield would be wiped out if -- as many analysts expect -- the dollar declines against other currencies.

The phenomenon is occurring not only in countries such as China that are well known for amassing vast quantities of Treasurys, but also in many poor countries that, while holding much smaller amounts than China, still hold sums that are sizable relative to their gross domestic products. Algeria, for example, has about $50 billion in what Summers calls "excess reserves," which is about half of its GDP. A few countries, such as Singapore, have already begun using their reserves much more creatively, Summers noted.

Summers acknowledged several problems with his proposal, such as the danger that central bankers would start gambling with their reserves by putting them into even more risky investments than stocks.

But he pointed out that at an annual return of 5 percent -- a conservative estimate for the long-run yield on a sensible stock portfolio -- the $2 trillion in excess reserves could produce average annual yields of about $100 billion. That is more than all the rich countries in the world spend on foreign aid to poorer countries each year.

Therein lies potential, he said, for realizing economists' fondest wish: "an almost free lunch."

© 2006 The Washington Post Company