Deficits, Interest Rates and the Fed

This article was published in the Washington Times, July 4, 2004.

As the Federal Reserve finally begins to raise interest rates, those who have been patiently waiting to find some connection between budget deficits and interest rates naturally seized this opportunity to suggest it is the budget deficit rather than the Fed that is really to blame.

Stanley Collender, a longtime professional budget worrier, told CBS “Market Watch” if voters could be persuaded shrinking the deficit might help curtail Fed tightening, “it might focus attention on the budget in a way that hasn’t happened in the last couple years.”

Conservative columnist Bruce Bartlett wrote that because the Fed will be raising interest rates, “this will begin to change the terms of debate on fiscal policy, increasing the likelihood of a budget deal next year that will involve tax increases.”

Never mind that the Fed also raised interest rates when the budget was in surplus, in 1999 and 1969, and the effect was the same. A stubborn fiscal fixation nonetheless causes some people to insist the budget deficit, not the Fed, is somehow responsible for raising the interest rate at which the Fed stands willing to buy Treasury bills with new money.

The Quixotic quest to uncover some link between budget deficits and interest rates was the central theme of a paper presented at the January American Economics Association meeting by former Treasury Secretary Robert Rubin, Brookings Institution scholar Peter Orszag and economist Allen Sinai. My own critique of that study, among others, has just been published as “Deficits, Interest Rates and Taxes: Myths and Realities” at

One novelty in the Rubin-Orszag-Sinai paper — later imitated by the International Monetary Fund (IMF) — was to change the subject. Mr. Rubin, Mr. Orszag and Mr. Sinai dismissed the annoyingly invisible impact of deficits on nominal interest rates by saying, “The overall level of nominal interest rates is affected by many factors.” While deficits may not raise nominal interest rates, they inferred, deficits do raise real interest rates. “For purposes of assessing the effects of future budget surpluses or deficits, it may be more insightful to examine the spread between long-term and short-term interest rates.”

Rather than claiming actual deficits raise actual interest rates, Mr. Rubin, Mr. Orszag, Mr. Sinai and the IMF instead claimed estimated future budget deficits (1) increase only real, inflation-adjusted interest rates, and (2) widen the spread between long-term and short-term rates.

These two variations on what the original authors called “the conventional view” were presented as if interchangeable. Yet a wider gap between long-term and short-term rates is not all the same as a wider gap between interest rates and inflation. On the contrary, real interest rates are typically lowest when the yield curve is steep (such as 1993 or today), and real rates are highest when the yield curve is flat (such as 1981).

Because these incompatible variations on the conventional view have essentially dropped all pretense that actual budget deficits raise nominal long-term interest rates, their alternative allegations about effects on yields curves and/or real interest rates raise astonishing anomalies. If nominal long-term interest rates are unaffected, the only way deficits could steepen the yield curve would be by reducing short-term interest rates. That would be a paradoxical conclusion indeed, particularly since steeper yield curves are a bullish leading indicator.

To add to such confusion, the only way budget deficits could increase real interest rates without also raising nominal interest rates would be if larger budget deficits caused inflation to fall. Budget deficits would then have to be viewed as deflationary and surpluses as inflationary, which seems a bit too ingenious.

In April, the IMF came out with a similarly befuddled analysis, citing some of the same authors. Sustained fiscal deficits “lower national savings,” said the IMF, “and eventually raise real interest rates in both the United States and abroad, thus crowding out global private investment.”

Like Messrs. Rubin, Orszag and Sinai, the IMF even revived the quaint “twin deficits” theory — blaming the U.S. trade deficit on the budget deficit — without offering a shred of evidence (because there is none).

All the alleged effects of budget deficits, which range from causing the dollar to rise to making it fall, are suspiciously limited to the United States. The IMF could not and did not claim Japan’s larger and more chronic budget deficits cause current account deficits or high interest rates in Japan.

Yet they claimed their unique indictment of U.S. deficits was solidly grounded in weighty economic research: “A number of empirical studies find a significant impact of fiscal deficits on real interest rates.” In reality, that “number of studies” refers to a table of three. And only one of those three — a brief working paper by Fed economist Thomas Laubach — said anything about real interest rates. Besides, Mr. Laubach only surmised a murky relationship between estimated (not actual) deficits and expected (not actual) interest rates.

Another paper cited by the IMF merely claimed estimated future deficits widen the spread between interest rates on 3-month and 10-year Treasuries, echoing Mr. Rubin, Mr. Orszag and Mr. Sinai. But that just shows recessions worsen the fiscal outlook and prompt the Fed to ease.

The IMF analysis is premised on the doctrinaire assertion that “fiscal deficits lower national savings.” But that key assumption is inconsistent with recent experience among countries that moved from deficits to surpluses. In the United States, national savings averaged 18.2 percent of GDP from 1981 to 1989, when budget deficits averaged 3.8 percent of GDP, but fell to 17½ percent of GDP from 1998 to 2001, when the budget was in surplus. Britain and Australia had similar experiences.

Such evidence is consistent with “Ricardian equivalence.” That means the burden of government spending is roughly equivalent whether financed by borrowing or taxes (aside from tax distortions and disincentives). Government purchases crowd out private purchases, by diverting real resources from market-determined uses to politically determined uses, but this is no less true when government spending is financed entirely by taxes rather than partly by debt. Government transfer payments discourage productive activity among those receive them and those who pay, but this is no less true when transfers from taxpayer Smith to subsidized Jones are financed by greater taxes on Smith.

A rigorous new econometric study of the impact of U.S. budgets from 1976 to 2002, by Roger Kormendi of the University of Michigan and Aris Protopapadakis of the University of Southern California, found “no evidence of any positive effects of either current or expected future budget deficits on either real interest rates or current account deficits”

Unfortunately, all the evidence in the world won’t persuade those who see political opportunity in preaching highly variable, unsupported theories to advance their pet cause — higher tax rates. When facts conflict with conventional theory in an election year, conventional theorists tend to treat unconventional facts (and those who dare mention them) as dispensable nuisances.

Alan Reynolds is a senior fellow with the Cato Institute and a nationally syndicated columnist.