Deficits and Debt in the Lens of History
By Barry Eichengreen
This paper is part of an initiative from the Peterson Foundation to help illuminate and understand key fiscal and economic questions facing America. See more papers in the Expert Views: Lessons from History for America Today series.
U.S. debt and deficit problems look increasingly intractable. Federal government debt in real (inflation-adjusted) terms has increased for 24 consecutive years.1 Expressed in constant 2025 dollars, it is more than three times its value nearly a quarter century ago. Prior to passage of the One Big Beautiful Bill Act (OBBBA), the Congressional Budget Office (CBO) estimated that debt held by the public would rise from its current 100 percent of gross domestic product (GDP) to 117 percent of GDP by the end of 2034, on the assumption that nothing was done legislatively to address the problem.2 Subsequent legislation hasn’t helped. The CBO now estimates that the OBBBA, by widening cumulative budget deficits over the decade by a further $3 trillion, will raise 117 percent to 124 percent.3
A famous adage in economics, attributable to Herbert Stein, is that if something can’t go on forever it will stop. The premise is indisputable; the question is when and how. Slowing economic growth, rising interest rates, and congressional polarization render one pessimistic about an early and orderly resolution. The disorderly alternatives, which grow increasingly likely with delay, are not reassuring.
The United States dealt successfully with debt and deficit problems on at least two occasions in the 20th century: in the decades following World War II and in the 1990s. In the first episode, debt held by the public fell from more than 100 percent of GDP, comparable to current levels, to less than 25 percent in the mid-1970s. In the second instance, debt in the hands of the public fell from 48 percent of GDP in early 1994 to just 31 percent in the first half of 2001. Who says it can’t be done?
The current impasse has predictably redirected the attention of scholars and policy analysts toward these two episodes of successful fiscal consolidation. Unfortunately, the findings are not reassuring. Studies suggest that what worked to narrow deficits and reduce debts then will not work now. A different approach is required.
After World War II
No single factor can account for the sustained decline of debt-to-GDP ratios in the United States after World War II. That the decline had multiple sources is not surprising. The fall in the debt ratio from above 100 percent to below 25 percent in less than 30 years was stunningly large. It could only result from a confluence of favorable forces.
Those favorable forces start with fast growth of the U.S. economy. GDP growth averaged 4.2 percent in the 1950s and 4.5 percent in the 1960s — the two decades of fastest peacetime growth in the 20th century. The U.S. had pioneered modern mass production (symbolized by the assembly line but more systematically reflecting the reorganization of factory design to accommodate the availability of electric power) in the first half of the 20th century and applied these methods to additional sectors and industries after World War II. Investment was depressed in the 1930s and concentrated in war-related industries in the first half of the 1940s, so there was a singular opportunity to apply these new production methods to the consumer goods sector. In addition, there was a backlog of new technologies to be exploited, including those developed in conjunction with the war (e.g., radar, aircraft, electronics, early computers). Growth in lagging regions, notably the South, accelerated not least owing to the diffusion of air conditioning.4 Total factor productivity growth ran at nearly 2 percent per annum between 1958 and 1973, supported not only by the aforementioned backlog of technology but also by the build out of the surface road infrastructure.5 U.S. firms, which were global technology and productivity leaders, benefited from recovery and growth in other countries and the concomitant expansion of global export markets.
If real growth rates were high, real interest rates were low. Until 1951, the Federal Reserve, under the terms of its wartime agreement with the Treasury, capped interest rates on Treasury bills at 0.375 percent and on 30-year bonds at 2.5 percent, purchasing the requisite quantity of bonds at the corresponding prices. It being impossible to peg interest rates and target inflation rates at the same time, the Fed effectively lost control of the latter. Inflation fluctuated wildly, averaging more than 7 percent between 1947 and 1951, and making for a sharply negative ex post real interest rate over the period.6 In 1952 the residual maturity of marketable Treasury debt was still six years, meaning that bonds bearing low interest rates issued in earlier years remained in the market for the balance of the decade.
After the Treasury-Fed Accord in 1951, interest rates on newly issued bonds were allowed to fluctuate with market conditions and expected inflation. But realized inflation consistently exceeded expected inflation, again making for low and sometimes negative ex post real interest rates.7 Domestically, demand for Treasury securities was supplemented by ceilings on interest rates on alternative assets such as bank deposits (Regulation Q ceilings on interest rates on savings accounts only being phased out in the 1980s). Internationally, the demand for Treasury securities was supported by the efforts of European and other central banks to rebuild their foreign exchange reserves, which they held substantially in this form.
Economists (e.g., Blanchard 2019) regularly point to the real-interest-rate-real-growth-rate differential (r-g) as a determinant of debt dynamics. A negative r-g differential will allow the debt-to-GDP ratio to fall “of its own volition,” as it were. An r-g differential of -5 over the 1947–73 period (-1 percent short-term real interest rate and 4 percent real GDP growth rate) was decidedly favorable for debt reduction.
Then there is the fact that the federal government ran an average primary budget surplus (budget balance net of interest payments) of 0.9 percent of GDP between 1947 and 1973.8 Surpluses were consistently maintained over the period. In the 1950s and 1960s the primary balance was in deficit just three times: in 1953 (owing to Korean War and related national security expenditures), 1959 (due to the impact of the 1957–8 recession on tax collections) and 1968 (due to Vietnam War and Great Society spending). The Committee for Economic Development, with intellectual leadership from the same Herbert Stein mentioned previously, championed the idea that the budget should be balanced over the business cycle and that fiscal policy should rely on automatic countercyclical stabilizers. In 1960, Social Security accounted for less than 13 percent of total federal spending, and Medicare and Medicaid did not yet exist. Federal revenues as a share of GDP, in contrast, were slightly higher than today. The notion of cutting taxes to limit the size of government (“starve the beast” theories) did not develop until the late 1970s and early 1980s.
When my coauthors and I decompose the reduction in the U.S. debt-to-GDP ratio over the period, we find that almost exactly two-thirds of the total is accounted for by the interest-rate-growth-rate differential, while one-third is accounted for by cumulative primary surpluses.9 Acalin and Ball (forthcoming) conduct a related exercise. They conclude that 20 percent of the total fall in the debt-to-GDP ratio is accounted for by primary surpluses, 72 percent by growth rates and interest rates, and the remaining 7 percent by the interaction of these two sets of factors.10
The 1990s
Debt reduction between 1994 and 2001 was more limited in magnitude and duration but significant nonetheless. Again, factors supporting this reduction were multiple: they included the peace dividend bequeathed by the end of the Cold War; strong economic growth as productivity accelerated with diffusion of the digital technologies associated with the “new economy”; declining interest rates on the outstanding debt; and the Omnibus Budget Reconciliation Act of 1993, which cut Medicare spending and gave a modest boost to tax revenues.
Of the 3.6 percent of GDP fall in federal spending between 1990 and 2000, more than 61 percent came from reductions in defense spending. Cuts began already in 1990, shortly after the fall of the Berlin Wall, during the administration of George H.W. Bush. Expenditure on defense fell from 5.1 percent of GDP in 1990 to 2.9 percent in 2000. Active-duty military personnel levels fell by more than 30 percent from their 1989 peak, as the Army went from 18 to 10 active divisions and the Navy reduced its number of ships by half. There were base closings and reductions in procurement spending. These changes received bipartisan support. Thus, the budget surpluses of the 1990s cannot be understood without reference to the peace dividend.11
In addition, the Omnibus Budget Reconciliation Act of 1993 altered the political economy of fiscal policy. It reflected the strategy of a new Democratic president, Bill Clinton, informed by the arguments of his economic advisor Robert Rubin, that balancing the budget, by lowering interest rates, would jumpstart economic growth. This altered the traditional party dynamic where Republicans argued for cutting taxes while Democrats insisted on raising social spending. The 1993 budget, passed along strict party lines (Democrats voting in favor and all Republicans voting against), included cuts in Medicare and other mandatory and discretionary spending. It reduced payments to Medicare providers and lifted the cap on wages subject to the Medicare payroll tax. At the same time, it increased taxes on corporations and high-income individuals, raised the share of Social Security benefits subject to tax, and boosted taxes on certain fuels.12 The CBO estimated that the Act would reduce the budget deficit by $433 billion over five years (where $433 billion was approximately 6 percent of 1993 GDP). 56 percent of the savings would come from increased tax revenue, 18 percent from cuts in entitlement spending, 16 percent from cuts in discretionary spending, and 11 percent from lower debt-service costs.13
In terms of those lower debt-service costs, interest rates on the debt declined over the period, from 7.5 percent in 1990–1 to a low of 5.9 percent in 1994 and then an average of 6.4 percent over the remaining 1995–2001 period. In part this can be understood as a corollary of other measures making for balanced budgets — the argument made by Secretary Rubin. Meanwhile, inflation expectations as measured by the Michigan Survey of Consumers came down from the 3.1–4.8 percent range in 1988–90 to 2.7–3.9 percent in 1991, largely as a result of the intervening recession, and then settled at less than 3.5 percent in 1992, as the inflationary momentum of the preceding period dissipated.
The role of economic growth is controversial. Between 1995 and 2000, labor productivity growth in the U.S. nonfarm sector accelerated to about 2.5 percent a year, up from the average of 1.5 percent from the early 1970s to 1995, a trend widely attributed to the productivity-enhancing effects of new digital technologies.14 Average annual GDP growth accelerated to 4 percent, almost twice the rate of the preceding five-year period. It can be objected that the comparison is not entirely fair: the first quinquennium was punctuated by a recession, the second not. Henderson (2015) observes that growth was not significantly higher than that achieved by the United States in earlier periods when deficits were a problem. Indeed, when one removes recession years, the average annual GDP growth rate between 1970 and 1990 averaged 4.5 percent. And deficits returned — some would say with a vengeance — in the first half of the 2000s, this despite that labor productivity growth accelerated still further.
As the combined result of these measures, the primary balance swung from small deficit in 1994, to small surplus in 1995 and then to increasingly large positive values in 1998–2001, averaging +3.7 percent over the latter period. Of course, this was less than a permanent cure of the deficit problem, which returned after the turn of the century. These measures did not durably stem the growth of entitlements, bend the healthcare cost curve, stop defense spending from ratcheting up again after 9/11, or prevent renewed recourse to revenue-reducing tax cuts. But they did inaugurate a brief and, for the United States, exceptional period of primary surpluses and declining debt ratios.
The relative importance of primary surpluses and the real-interest-rate-real-growth-rate differential was the inverse of that in the earlier post-World War II episode. Primary surpluses were the single most important contributor to debt reduction between 1994 and 2001.15 The interest-rate-growth-rate differential mattered less because the real interest rate on government debt was now positive, at 4.5 percent. (Net interest payments as a share of debt held by the public averaged 7.0 percent over the period, while inflation ran at 2.5 percent.) Since growth averaged 3.8 percent, the real-interest-rate-real-growth-rate differential so measured did not contribute positively to debt reduction, though the modest fall in interest rates and acceleration in growth relative to the immediately preceding period prevented this element from enlarging the debt still further.16
Economic Measures
The clear message of this analysis is that what worked earlier to reduce high debts and prevent chronic deficits will not work now. Following World War II, debt reduction was facilitated by a favorable interest-rate-growth-rate differential. Interest rates were held down by the policies of a Federal Reserve System that had temporarily sacrificed its independence as part of the war effort, by tight financial regulation that limited the return on alternative assets, and owing to the appetite for U.S. Treasury securities of foreign central banks and governments anxious to rebuild their foreign reserves. It helped that inflation expectations persistently lagged behind realized inflation, plausibly as a legacy of price stability in the 1920s, deflation in the 1930s, and price controls in the 1940s.
None of these facilitating conditions is present today. The Fed is jealous of its independence — though aware of political threats to that status — and committed to pursuit of its inflation target. Possessing recent experience with inflation, consumers and investors are alert to its possible resumption. This suggests that any attempt to inflate away the debt will be ephemeral; interest rates will adjust. The United States has already moved into a higher interest rate environment, not just relative to the immediate post-World War II period but also relative to recent years. In terms of scope for financial repression, the deregulatory toothpaste can’t be put back in the tube. Bank depositors and shareholders will resist caps on deposit rates; they have the power to do so insofar as alternative savings vehicles have proliferated since the 1950s and 1960s. The U.S. dollar remains the leading reserve currency, but its share of global foreign exchange reserves has declined over the last quarter century. Central banks such as the People’s Bank of China have trimmed their holdings of U.S. Treasury securities, and questions are being asked about the dollar’s safe-haven status.
Growth rates are difficult to forecast. The International Monetary Fund sees U.S. growth as slowing from the 2.5–3 percent range achieved in 2022–4 to 2.1 percent or less in 2025–9. Some independent analysts are more optimistic: they see Generative Artificial Intelligence as giving a significant boost to productivity and GDP growth.17 Nevertheless, the CBO continues to project real GDP as growing by 1.8 percent per annum over the 2027–34 period. Leaving aside the most optimistic independent forecasts, few anticipate a growth spurt comparable to either 1946–73 or 1995–2000.
Combining a 2 percent GDP growth rate with a 4 percent 10-year Treasury yield and a 2 percent inflation rate (round numbers all), it is hard to see the interest-rate-growth-rate differential contributing materially to a reduction in the debt ratio.
In the second half of the 1990s, debt reduction was facilitated by primary budget surpluses, both their direct effect and their second-round impact on debt-servicing costs. Those primary surpluses in turn reflected a combination of spending reductions and tax increases, with an emphasis on the former. Expenditure cuts were possible because a benign geopolitical environment enabled deep reductions in defense spending. The opposite is true now: security threats from Russia and Iran and potential conflict in the South China sea create pressure for defense-spending increases. Cuts to Medicare, Medicaid and Social Security are politically fraught. These three categories of spending, together with defense and net interest payments, constitute 73 percent of the fiscal year 2024 budget, underscoring the fact that there is not much discretionary spending left to cut. The experience of the Department of Government Efficiency and the failure of the House Freedom Caucus and other fiscal hawks to append additional spending cuts to the OBBBA leave one pessimistic about the prospects.
By process of elimination, this leaves one route to fiscal consolidation and debt stabilization, namely raising additional revenue. The fact is that the United States is a low tax-revenue economy. In 2023, total tax revenue in the United States equaled 25.2 percent of GDP compared to the OECD average of 33.9 percent. Leaving aside Ireland, where GDP is artificially inflated by profit shifting, and the middle-income members of the OECD (Costa Rica, Turkey, Colombia, Chile and Mexico), the United States has the lowest ratio of tax revenue to GDP of any OECD country. Raising federal revenues sufficiently to bring the total tax take up to the OECD average would more than suffice to eliminate the overall budget deficit, much less the primary deficit.18
True believers in supply-side economics will insist that any and all increases in tax rates would be self-defeating; they would harm growth and lower the denominator of the debt-to-GDP ratio. Economists such as Alberto Alesina have argued that fiscal consolidation is sustainable only when it takes the form of spending reductions rather than tax increases.19 But their evidence is heavily based on the experience of countries, mainly in Europe, where public spending is higher than in the United States and taxes are excessive. It makes sense that confidence will be enhanced and investment and growth will be stimulated in such countries when fiscal consolidation is achieved through spending cuts, not when already excessive taxes are raised further. It also follows that U.S. experience is apt to be different.
Political Measures
But what is desirable economically may not be feasible politically. Sustained deficit reduction is possible only when different political factions can compromise and stick to an agreed course of action. Serkan Arslanalp and I (2023) considered cases since 1973 when countries were able to stabilize and substantially reduce high debt ratios. The most robust determinant of success (along with fast GDP growth) was a low level of political polarization. From this standpoint, the United States is not favorably situated. Elite political polarization has risen since the 1970s, as measured by the average distance between political positions across parties.20 The number of centrist lawmakers (moderate Republicans and Blue Dog Democrats) has declined over time. Survey data document analogous findings for the voting public: the share of ideological moderates has declined over time, and voters are increasingly sorting according to ideology.21 Comparisons with other OECD countries suggest that U.S. polarization is highest, whether measured at the elite or mass perception level, and that the United States is the OECD economy where polarization has been rising most strongly over time.22
Elite polarization means that members of Congress are willing to support either tax increases or spending reductions but not both. Mass polarization means that voters are likely to support candidates embracing one or the other strategy but rarely a combination of the two. Neither elite nor mass polarization is conducive to fiscal compromise or the maintenance of an existing fiscal consolidation strategy when the party in power changes.
What can be done about this is disputed, both because the causes of political polarization are disputed and also because, even when there is agreement on the causes, they are not easily remedied. Some blame high and rising levels of income inequality, which hollows out the middle class of moderate voters. But to the extent that the increase in inequality results from a combination of skill-biased technical change and imports of manufactures from emerging markets, it is not easily reversed.23 Others blame gerrymandering that creates safe districts for one party or the other, making the primary the key election and incentivizing candidates to adopt relatively extreme policy positions that appeal only to members of their party.24 Gerrymandering could in principle be restrained by the courts, but the federal judiciary has done less to limit it in recent decades, reflecting shifts in judicial doctrine, specifically evolving interpretations of the Constitution’s constraints on redistricting.
Still others point to the proliferation of new media, which allows voters to self-select into ideologically familiar and self-confirmatory audiences, resulting in incompatible understandings of policy issues. Again, this does not seem like a trend that is easily reversed. A final hypothesis blames legislators’ reliance on campaign contributions from groups and individuals with extreme preferences. Legal constraints on campaign contributions have, if anything, grown weaker in recent decades, as court rulings struck down aggregate limits on how much individuals can contribute across all federal candidates, and allowed the creation of Super PACs which can raise and spend unlimited funds independently of campaigns.
In principle, one can imagine treatments for each of these problems. More effective education and training might begin to level the pre-tax distribution of income, while more progressive taxation could level the post-tax distribution. Reintroduction of the Fairness Doctrine would require media networks to present differing viewpoints. District courts and a Supreme Court that interpreted constitutional issues differently might take a harder line against gerrymandering and unlimited campaign contributions.
Putting it this way suggests that there is no easy fix, either to political polarization or to the fiscal gridlock that flows from it. At the same time, this conclusion underscores the importance of getting started now on the relevant reforms.
References
Abramowitz, Alan and Kyle Saunders (2008), “Is Polarization a Myth?” Journal of Politics 70, pp.542–545.
Acalin, Julien and Laurence Ball (forthcoming), “Did the US Really Grow Out of its World War II Debt?” American Economic Journal: Macroeconomics (forthcoming).
Alesina, Alberto (2000), “The Political Economy of the Budget Surplus in the United States,” Journal of Economic Perspectives 14, pp.3–19.
Alesina, Alberto, Carlo Favero and Francesco Giavazzi (2019), Austerity: When it Works and When it Doesn’t, Princeton: Princeton University Press.
Alesina, Alberto, Roberto Perotti and Jose Tavares (1998), “The Political Economy of Budget Deficits,” Brookings Papers on Economic Activity 1, pp.197–266.
Arslanalp, Serkan and Barry Eichengreen (2023), “Living with High Public Debt,” in Structural Shifts in the Global Economy, Kansas City, MO: Federal Reserve Bank of Kansas City, p.303–360.
Barber, Michael and Nolan McCarty (2013), “Causes and Consequences of Polarization,” in “Negotiating Agreement in Politics,” in Jane Mansbridge and Cathie Martin eds, Task Force Report, American Political Science Association (December).
Bernanke, Ben (2006), “Productivity,” Speech to Leadership South Carolina, Greenville, South Carolina (31 August).
Blanchard, Olivier (2019), “Public Debt and Low Interest Rates,” American Economic Review 109, pp.1197–1229.
Boxell, Levi, Matthew Gentzkow and Jesse Shapiro (2024), “Cross-Country Trends in Affective Polarization,” Review of Economics and Statistics 106, pp.557–565.
Congressional Budget Office (1994), “An Economic Analysis of the Revenue Provisions of OBRA-93,” CBO Papers (January).
Eichengreen, Barry, Asmaa El-Ganainy, Rui Esteves and Kris James Mitchener (2021), In Defense of Public Debt, New York: Oxford University Press.
Eichengreen, Barry and Peter Garber (1991), “Before the Accord: U.S. Monetary-Financial Policy, 1945-51,” in R. Glenn Hubbard ed., Financial Markets and Financial Crises, Chicago: University of Chicago Press, pp.175–206.
Fernald, John and Shanthi Ramnath (2004), “The Acceleration in U.S. Total Factor Productivity after 1995: The Role of Information Technology, Federal Reserve Bank of Chicago Economic Perspectives 18, pp.52–67.
Field, Alexander (2007), “The Origins of U.S. Total Factor Productivity Growth in the Golden Age,” Cliometrica 1, pp.63–90.
Fiorina, Morris, with Samuel Abrams and Jeremy Pope (2006), Culture War? The Myth of a Polarized America, New York: Pearson Longman, 2nd edition.
Henderson, David (2015), “US Federal Budget Restraint in the 1990s: A Success Story,” Mercatus Center, George Mason University (June).
Klein, Alexander, Miguel Leon-Ledesma and Nicholas Crafts (2025), “A Long-Run Perspective on Unconditional Convergence in Manufacturing: Evidence from U.S. Industrialization,” CEPR Discussion Paper no.20488 (July).
McCarty, Keith, Nolan Poole and Howard Rosenthal (2006), “Does Gerrymandering Cause Polarization?” unpublished manuscript, Princeton University, UC San Diego and New York University (February).
McCarty, Keith, Nolan Poole and Howard Rosenthal (2016), Polarized America: The Dance of Ideology and Unequal Riches, Cambridge, MA: MIT Press 2nd edition.
Mitchener, Kris James and Ian McLean (1999), “U.S. Regional Growth and Convergence, 1880-1980,” Journal of Economic History 59, pp.1016–1042.
Rehm, Philipp and Timothy Reilly (2010), “United We Stand: Constituency Homogeneity and Comparative Party Polarization,” Electoral Studies 29, pp.40–53.
About the Author
Barry Eichengreen is George C. Pardee and Helen N. Pardee Chair and Distinguished Professor of Economics and Professor of Political Science at the University of California, Berkeley. He is a Research Associate of the National Bureau of Economic Research and Research Fellow of the Centre for Economic Policy Research. In 1997–98 he was Senior Policy Advisor at the International Monetary Fund.
Professor Eichengreen is a fellow of the American Academy of Arts and Sciences (class of 1997). He is a distinguished fellow of the American Economic Association (class of 2022), a corresponding fellow of the British Academy (class of 2022), and a Life Fellow of the Cliometric Society (class of 2013). He has held Guggenheim and Fulbright Fellowships and been a fellow of the Center for Advanced Study in the Behavioral Sciences (Palo Alto) and the Institute for Advanced Study (Berlin). For 15 years from 2004 he served as convener of the Bellagio Group of academics and officials. He is a regular monthly columnist for Project Syndicate.
Professor Eichengreen has been awarded the Economic History Association's Jonathan R.T. Hughes Prize for Excellence in Teaching and the University of California at Berkeley Social Science Division's Distinguished Teaching Award. He is the recipient of a doctor honoris causa from the American University in Paris, and was the 2010 recipient of the Schumpeter Prize from the International Schumpeter Society and the 2022 recipient of the Nessim Habif Prize for Contributions to Science and Industry. He was named one of Foreign Policy Magazine's 100 Leading Global Thinkers in 2011. He is a past president of the Economic History Association (2010–11).
His most recent book is In Defense of Public Debt with Asmaa El-Ganainy, Rui Esteves and Kris Mitchener (Oxford University Press 2021).
Expert Views
Lessons from History for America Today
We asked respected experts to examine America's present challenges through an historical lens and identify lessons learned for navigating our nation onto a stronger, more prosperous path.