Coping with Budget Uncertainty: Gleanings from a Hutchins Center Conference

We weren’t sure what answers we’d get when the Hutchins Center on Fiscal and Monetary Policy posed the question: What’s the smartest response from Congress and the president to the uncomfortable fact that today’s projections about future budget trends are surely going to be off?

And, to be frank, we ended the day with as many questions as answers, as you can tell if you read the papers, watch the video or skim the transcript. For the rest of you, here a few things I gleaned.

The economist’s approach. Alan Auerbach of University of California at Berkeley made the academic economist’s case that it’s not enough to find a way to restrain spending or raise taxes so the U.S. doesn’t continue to pile up debt under best-guess projections. Uncertainty about those projections means the government should do even more, essentially taking out insurance against unlucky breaks, or as he put it, “economic disaster.” That argument derives from economic theorists’ view of how individuals should behave when confronting uncertainty, he said. “Uncertainty means our policy choices will always turn out to be wrong in some sense. We’re going to have to make adjustments.” And it’s better to do more now to “lessen the consequences of uncertainty than simply to ignore it and let things happen.”

Not all economists are convinced. Peter Diamond, the MIT Nobel laureate challenged the argument that economic theory of individuals makes always applies to the government. Diamond, deviously citing earlier work by Auerbach and Kevin Hassett, argued that “in some circumstances it may turn out that more uncertainty about projections is a reason to delay legislative action and so to save less currently.” Brookings’ Henry Aaron argued that there is so little useful information in long-run projections that, most of the time, they can and should be ignored altogether. But as I listened closely, the academic economists really were challenging Auerbach less on his economic logic and more by invoking political reality. Indeed, Aaron’s concern is that long-run projections become ammunition in political fights rather than actually contributing to better decisions.

The political reality. Those who have played in Washington’s big leagues doubt politicians will ever take Auerbach’s advice. (And so does he: “I understand there’s a political problem whenever our theory says you should be putting resources aside, but yet, the resources are there and it’s tempting to spend them.) It’s hard enough to persuade politicians to do enough to narrow the gap between projected spending and projected revenues; it’s almost impossible to contemplate doing more in response to uncertainty. Rep. Jim Cooper (D, Tenn.) accepted the Auerbach prescription. “It’s almost a hallmark of civilized people to be able to delay gratification,” he said. “But that is very unpopular speech material. We seem incapable of doing that today.” No one challenged that view.

There was, however, a repeated distinction between Social Security and Medicare. There were frequent references to past decisions to make changes to Social Security far in advance (including the 1983 legislation to raise the age at which individuals qualify for full benefits and the general comfort with automatically adjusting benefits for inflation.) There was agreement that it’s easier to make credible long-run forecasts for Social Security as well as to fashion changes to that program. Health care is tougher. Credible long-run forecasts are much harder to make. It’s far from clear how to slow the growth of health care spending in the future without causing adverse effects on health. And legislating changes today that take effect later either aren’t credible or require delegating more authority to technocrats or boards of outsiders than is palatable to Congress.

Outside observers and inside-the-Beltway insiders agreed that it’s hard for Congress to deal with long-run deficits at times when the government is flush or running surpluses but also to legislate future spending cuts and tax increases when the economy is sick and in need of the opposite. When the economy is lousy, as it has been, White House veteran Gene Sperling said, “The sweet spot on fiscal policy is to have a single piece of legislation that both expands demand in the short run, but gives confidence on long-term fiscal policy.” That requires building automatic on-off switches into law.

Automatic pilot. Given the obvious inability of Congress to respond swiftly – or, sometimes, to respond at all – when the economy takes an unexpected turn, there is widespread interest in building automatic adjustment mechanisms into legislation though not much agreement on which sorts of mechanisms work best. NYU’s David Kamin noted that Congress failed to increase the amount of fiscal stimulus even when it became clear that the 2007-09 recession and subsequent recovery were worse than predicted, and suggested that a self-aware Congress could have anticipated that possibility and designed a mechanism in advance that would have automatically injected more fiscal stimulus. Sperling favored a provision that would automatically send more money to the states if the economy disappoints; Congress simply moves too slowly, he said. But Harvard’s Martin Feldstein emphasized the recent recession was unusual; most recessions are shorter so automatic mechanisms risk putting more fiscal stimulus after the recession has ended.

For Social Security, one often-discussed option is to automatically adjust the age at which Americans are eligible for full Security benefits for changes in life expectancy; the British are about to do that. Several suggested that policy is unfair because the rich tend to live longer than the poor. Feldstein said that’s a problem could be addressed by adjusting the age more for higher-wage workers and less for lower-wage workers, though that may be easier to describe to economists than explain to Americans.

The communications challenge. Those charged with supplying budget projections to politicians – Doug Elmendorf of CBO and Robert Chote of the U.K.’s new Office for Budget Responsibility – think long and hard about how best to convey uncertainty to politicians, who aren’t much interested in it. There are fan charts, probability distributions, confidence interval and ranges. But injecting a dose of reality, Bill Hoagland, who spent a quarter-century on the Senate staff, recalled one member of Congress telling him: “Young man, we don’t appropriate in ranges.”

Chote offered an intriguing approach to conveying the uncertainty around his agency’s estimates of the cost of particular proposals (known as “scoring” in the trade.) It qualitatively describes its confidence in the data, models and predictability of behavior underlying the estimates are on a six-grade scale (very high, high, medium-high, medium, medium-low or low.) Such judgments can, of course, be more politically treacherous than a best-we-can-do numerical estimate, but they do convey a lot of information.