We're Borrowing Like Mad. Can the U.S. Pay It Back?

By Greg Ip
Sunday, January 11, 2009; B01



In its battle against the financial crisis, the U.S. government has extended its full faith and credit to an ever-growing swath of the private sector: first homeowners, then banks, now car companies. Soon, President-elect Barack Obama will put the government credit card to work with a massive fiscal boost for the economy. Necessary as these steps are, they raise a worry of their own: Can the United States pay the money back?

The notion seems absurd: Banana republics default, not the world's biggest, richest economy, right? The United States has unparalleled wealth, a stable legal tradition, responsible macroeconomic policies and a top-notch, triple-A credit rating. U.S. Treasury bonds are routinely called "risk-free," and the United States has the unique privilege of borrowing in the currency that other countries like to hold as foreign-exchange reserves.

Yes, default is unlikely. But it is no longer unthinkable. Thanks to the advent of credit derivatives -- financial contracts that allow investors to speculate on or protect against default -- we can now observe how likely global markets think it is that Uncle Sam will renege on America's mounting debts. Last week, markets pegged the probability of a U.S. default at 6 percent over the next 10 years, compared with just 1 percent a year ago. For technical reasons, this is not a precise reading of investors' views. Nonetheless, the trend is real, and it is grounded in some pretty fundamental concerns.

The most important is the coming surge in the federal debt. At the end of the last fiscal year, in September, the total public debt held by the American people (excluding debt issued to the Social Security Trust Fund or held by the Federal Reserve) stood at $5.8 trillion, or 41 percent of gross domestic product -- about what the debt-to-GDP ratio has averaged since 1956. But the Congressional Budget Office projects deficits of $1.9 trillion over the next two years. Add almost $800 billion of stimulus spending, and U.S. debt soars to 60 percent of GDP by 2010 -- the highest level since the early 1950s, when the nation was working off its World War II and Korean War debts.

The other major cause for concern is that the federal government has taken on massive "contingent liabilities" -- loans and guarantees that don't become actual costs until the borrower defaults and the federal guarantee has to be honored. For example, Washington has purchased $45 billion of preferred stock of Citigroup but has also agreed to backstop up to about $240 billion of its losses. Bianco Research, a Chicago financial research firm, puts the total of such contingent liabilities (as of Dec. 29) at more than $8 trillion. The U.S. government won't shell out anything close to that, of course; it may not pay out any money and might even turn a profit. But the worse the economy gets, the more likely it is that some of those contingent liabilities will become actual liabilities.

How unprecedented would default be? The United States has never failed to repay a debt in its history. But it has twice altered the repayment terms, notes a study by Carmen M. Reinhart of the University of Maryland and Kenneth S. Rogoff of Harvard University. In 1790, when the infant republic took over the states' colonial-era debts, it deferred some interest for 10 years. A more pertinent case occurred during the Great Depression. In 1933, President Franklin D. Roosevelt devalued the dollar by 41 percent against gold. This helped end the vicious cycle of bank failures, deflation and default that had worsened the economic downturn, but it created another dilemma. Since the Civil War, borrowers in the United States, including the government, had routinely issued bonds that allowed the holder to demand repayment in gold or its dollar equivalent, based on the price of gold when the bond was issued. Devaluation would have dramatically raised, in dollar terms, the burden of repayment. So in 1933, Congress repealed the gold clause, a decision the Supreme Court upheld in 1935.

Reinhart argues that these episodes show that there are many forms of default other than the outright failure to redeem a bond. She thinks it exceedingly unlikely that the United States would ever default on its Treasury debt but says that a default is more likely to happen on the "periphery": A state government or a federally backed entity might renege. Reinhart notes that Washington has issued a lot of guarantees recently, such as the Treasury Department's declaration that it "effectively" backs the debts of Fannie Mae and Freddie Mac. Future governments could balk at honoring those guarantees.

Now, one reason the United States should not have to default is that, in a pinch, it can print the money it needs to pay off its debt. In practice, it would not literally print dollar bills; rather, the Federal Reserve would purchase its bonds. Still, the effect is the same: The Fed, by "monetizing" government debt, would create vast new supplies of dollars to chase the same goods and services and thus also create inflation. This money-printing option is not available to countries that issue debt payable in foreign currencies or to countries that have adopted the euro, which is controlled by the independent European Central Bank.

All the same, a country may not always prefer inflation to default. Reinhart and Rogoff found numerous instances of countries defaulting on their domestic currency debts, such as Russia, which did so in 1998. "Why would a government refuse to pay its domestic public debt in full when it can simply inflate the problem away?" they ask. "Sometimes, the government may view repudiation as the lesser evil." Default would certainly seem preferable to the social and economic unrest that hyperinflation has visited on Zimbabwe today or on Latin America in the 1980s.

Of course, it would be ridiculous to put the United States in the same company as Russia, much less Zimbabwe. Nonetheless, even if Washington never defaults, it can still suffer if questions about its ability to repay its debts affect its creditworthiness and thus its cost of borrowing.

Consider Japan, which spent the 1990s running up enormous budget deficits in an effort to kick-start economic growth. It failed, and gross government debt rose from 71 percent of GDP in 1990 to 115 percent in 1998, when Moody's Investors Service stripped Japan of its triple-A credit rating. Moody's figured that Japan's probability of default, while still low, had nonetheless risen -- even though that debt was owed almost entirely in yen. Japan's rating continued to slide until 2002, at which point its gross debt had reached 154 percent of GDP. Having its credit rating slashed was a stunning comedown for the world's second-largest economy and largest creditor nation. Nonetheless, Japan's borrowing costs weren't much affected, in part because almost all its debt was held by Japanese investors, who were not much inclined to sell it. The U.S. government might not have the same good fortune: More than half of its publicly held debt is owned by foreigners. (Of course, many of those foreigners are central banks with no attractive alternatives to the dollar.)

One key lesson of Japan's experience: It's not just the level of debt that affects a country's creditworthiness but the performance of the underlying economy. The size of the economy determines the resources the government can call on to repay its debts.

Economic growth was critical to the decline in America's debt-to-GDP ratio from its two previous peaks. The cost of World War II drove U.S. debt to an unnerving 109 percent of GDP in 1946, but the peace dividend and the postwar economic boom steadily brought the ratio down. Between 1981 and 1993, that ratio rose again, from 26 percent of GDP to 49 percent. It subsequently declined, thanks to the peace dividend from the end of the Cold War, the Clinton-era tax increases and a long economic boom.

It could, and probably will, happen again. But it's going to look a lot worse before it looks better. In another paper, Reinhart and Rogoff found that banking crises worldwide are typically followed by skyrocketing debt: It went up by 86 percent on average, after inflation, over three years, largely because the recessions that usually follow financial meltdowns devastate tax revenues and spur governments to spend aggressively.

So what's the moral of the story? The Obama administration should not focus on debt reduction now, which could actually undermine the prospects for a recovery in the real economy. With households and businesses trying to spend less and save more, the federal government must spend more and save less -- that is, borrow more -- in order to prevent a self-feeding downward spiral in economic activity. Once the recession is over, getting our debt burdens down will hinge on Obama's and Congress's willingness to confront the looming cost of Social Security and Medicare benefits for the aging U.S. population.

The chances of default remain pretty remote. But remote does not mean impossible. The best way to keep those chances remote is for policymakers to vow to get the deficit down once the recession is over -- and mean it.

gregip@economist.com


Greg Ip is U.S. economics editor of the Economist.