Fiscal and Monetary Policy Interactions

May 2007

Moderator: Welcome to Research Insights, an occasional podcast from the Federal Reserve Bank of Atlanta. The following is a conversation between two economists after a recent conference on fiscal and monetary policy interactions.

Ellis Tallman is a vice president over the macro policy group at the Atlanta Fed, and he will be asking the questions. Eric Leeper is director of the Center for Applied Economics and Policy Research at Indiana University, and he's also a professor of economics. Eric will answer the questions, and his answers represent his views and not those of the Atlanta Fed. Ellis, I'll let you begin with the first question.

Ellis TallmanEllis Tallman: Eric, the federal debt-to-gross domestic product (GDP) ratio has fluctuated recently, falling in the late 1990s and rising since 2001. Is this a cause for concern?

Eric Leeper: Well, from one perspective the recent rise in debt-to-GDP ratio is not unprecedented. The debt now is fluctuating in the neighborhood of about 35 to 40 percent of GDP. And over the postwar period we have had episodes as high as 70 or 80 percent. I think, however, that what's going on now is unusual relative to history for two reasons. One is that in the past most of the large run-ups in debt have been due to war expenditures. And what has been happening recently is largely not due to war expenditures, and so it's not clear whether we can easily extrapolate from past experience to draw inferences about the effects of the current debt run-up.

The second thing that differs from the past is that the vast majority of new issuances of government debt have been bought by foreign debt holders. In an environment where the central bank's primary concern is with price stability, which it is in many inflation-targeting countries around the world, there will be incentives, if fiscal policy does not adjust to these changes in debt, for the central bank to try to revalue that debt by generating inflation. This is something that the Federal Reserve is going to have to be on the lookout for.

Tallman: The Congressional Budget Office has projected that due to the aging demographic of the United States, the cost of Social Security and Medicare will force large expenditure increases in these entitlements. How might these fiscal issues affect monetary policy?

Leeper: This is a very difficult question to address based on current research. And the reason for this is that these future liabilities are often referred to as "unfunded liabilities." In models where people have good foresight—form rational expectations and make optimal choices—we have no way to understand the phrase "unfunded liability." So, to make sense of that in a rational expectations environment, you have to either relax the unfunded part and assume that people believe these liabilities will be funded in some manner, or you have to relax the liability part, which means that people believe that these commitments may well be reneged on.

Having said that, in rational expectations models we ought to be seeing the effects of large transfer expenditures even 20–30 years ahead. We should start seeing those effects appear in the data today. And those effects have been very hard to detect in the data. Maybe they're there, and we don't understand them. Maybe rational expectations are not the best assumption to make. Maybe people don't have as much foresight as our theoretical models suggest, in which case these effects may slowly start to appear in data, and this is something that the Fed has to be looking for very diligently.

Moderator: Thanks, Ellis and Eric. Again, we've been listening to Atlanta Fed Vice President, Ellis Tallman, who is asking questions, and Eric Leeper, director of the Center for Applied Economics and Policy Research at Indiana University. This concludes our Research Insight's podcast on fiscal policy and monetary fiscal policy interactions. The papers from this conference are available on our Web site. Thanks for listening and please return for more podcasts. If you have comments, please send us e-mail at