As fears of imminent recession have climbed in recent weeks, many are warning that the U.S. government does not have the budget capacity to mount an effective response to an economic downturn. With federal debt historically high relative to gross domestic product, the story goes, we cannot afford to increase federal spending or cut taxes to stimulate the economy.
That concern is misguided: Yes, we have a serious long-term debt problem, but no, that problem does not make anti-recessionary budget policy impossible or unwise.
Suppose the economy goes into recession sometime in the next year and that the Federal Reserve lowers the federal funds rate to essentially zero, as it did between December 2008 and November 2015. Now suppose that Congress and the president agree on a collection of spending increases and tax cuts twice as large as the 2009 American Recovery and Reinvestment Act. The Congressional Budget Office estimated that the recovery act increased federal spending and reduced taxes by a combined total of roughly $850 billion, so a stimulus that is twice as large would have a direct budgetary impact of about $1.7 trillion. (If an earlier Congress and president had agreed to follow the 2009 act with an additional stimulus bill of comparable size in 2011, we would have enjoyed a faster economic recovery with more job creation, and the country would be better off.)
Such fiscal stimulus would boost GDP significantly relative to what it would be otherwise: The recession would be less deep, less lengthy or both. As a result, fewer people would lose their jobs than if the stimulus did not occur, and people who did lose jobs would find new jobs more quickly. Fewer families would be evicted from their homes or be unable to pay their medical bills, and fewer graduates who enter the labor market would suffer a recession penalty in their future earnings.
Based on a large amount of economic research over the past decade, a reasonable estimate is that fiscal stimulus of $1.7 trillion under the conditions described - in particular, with the federal funds rate at zero - would increase GDP over the following few years by roughly one-and-a-half times as much, or about $2.5 trillion. Higher GDP means higher taxable incomes, so the federal government would recoup about $0.6 trillion of the gross budgetary cost of the stimulus, leaving a net budgetary cost of roughly $1.1 trillion.
That figure is very large by almost any standard, but it represents only about one year's worth of federal borrowing at our current pace. All else equal, adding that much debt now would simply bring the future one year closer: The ratio of federal debt to GDP that CBO projects under current law for, say, 2026 would arrive in 2025, and so on.
There is little reason to think that financial-market participants would find this fundamentally different from the path we are already on today. Nor would holding off debt by a year be worth the lost national output and additional human suffering of a deeper or longer recession.
To be sure, this simple math omits some relevant considerations. A recession would widen budget deficits automatically through lower tax collections and higher government spending on certain programs. Those automatic stabilizers are crucial anti-recessionary tools, but they would increase the amount of federal debt beyond CBO's current projections even in the absence of deliberate fiscal stimulus. Plus, the additional debt from fiscal stimulus would incur interest costs over time, so the incremental debt five or 10 years in the future would be somewhat larger than $1.1 trillion.
But reducing the severity of a recession would lead to higher output and employment not just during the recession but for many years to follow. A growing body of economic research suggests that short-term shocks such as the Great Recession do lasting damage to labor force participation. By lessening such damage, fiscal stimulus not only offers greater economic benefits, but also generates additional tax revenue that lowers the budgetary cost.
Moreover, interest rates on federal debt are at historically low levels and, based on readings in financial markets, will probably remain low for many years. Federal borrowing is thus less costly and creates less risk for the federal government than many of us predicted several years ago - and, according to economic analyses, does less harm tothe economy even in the long run. In addition, the funds used as stimulus do not disappear: They support government spending for public goods and services, orthey lower taxes so that households have more resources for their private activities.
In sum, we have plenty of capacity in the federal budget to undertake vigorous countercyclical tax and spending policies when the next recession arrives. Given the economic and social costs of recessions, we should undertake such policies.
- - -
Elmendorf is dean and Don K. Price professor of public policy at Harvard Kennedy School. He was director of the Congressional Budget Office from 2009 to 2015.