One troubling aspect of the political debate over the debt ceiling is the constant repetition of the statement that “the US has never defaulted on its debt”. On the grounds that those who fail to learn from history are due to repeat it, I would like to set the record straight. I welcome readers to share this with colleagues with or without attribution. Fellow risk professionals and the general public deserve to know the facts. During its 235 years as a sovereign entity the United States has defaulted on three separate occasions - two of which are reported by Rinehart & Rogoff (2009) - and has also intentionally liquidated debt via inflation.
In 1782, the Treasury failed to pay interest on Revolutionary War debt. Following ratification of the Constitution, Treasury Secretary Alexander Hamilton resumed regular debt service in 1790, but deferred some interest payments for ten years (Riley, 1978).
For several decades prior to 1933, holders of Treasury securities were contractually entitled to receive interest and principal payments in either dollars or gold. At the time, many contracts contained a “gold clause”, which enabled payees to receive proceeds in the form of gold. During the 1933 banking holiday declared by President Franklin Roosevelt immediately after his March 4 inauguration, the federal government refused requests for interest payments in gold, remitting only currency instead. Congress later ratified this action by formally invalidating gold clauses in a "Joint Resolution to Assure Uniform Value to the Coins and Currencies of the United States," passed on June 5, 1933. In 1934, President Roosevelt officially devalued the dollar by increasing the price of gold from $20.67 to $35.00. Although contemporary press accounts characterized the government’s actions as an abrogation (Wall Street Journal, 1933, May 4), Treasury securities issued in June and August of 1933 were oversubscribed and a February 1935 Supreme Court decision upheld the government’s actions. While these actions were generally portrayed today as an attempt to halt gold hoarding or end price deflation, they also appear to have had a fiscal motivation. In FY 1933, the ratio of interest expense to federal revenues reached 33.15%, the only time this ratio has exceeded 30% since the post-Civil War era. The Roosevelt administration needed more funds to implement New Deal programs and wanted the flexibility to issue new Treasury securities unimpeded by gold convertibility (Wall Street Journal, 1933, May 27). On a cautionary note, Moody’s Municipal and Government Bond Manuals from 1933 and 1934, show that all US Treasury bonds carried Aaa ratings both before and after this default (Mergent, 1933-1934).
The United States Debt-to-GDP ratio reached its modern peak of 112.7% in 1945, primarily due to war-time borrowing. Interest rates at the time were relatively low, while tax rates were relatively high. Thus interest expense accounted for only 7.79% FY 1945 federal revenues – compared to a proportion of 9.88% in FY 2010 – suggesting that this high level of debt could be serviced without great difficulty. The Debt-to-GDP ratio fell rapidly after the end of World War II, largely as a result of high inflation that followed the relaxation of wartime price controls and Federal Reserve purchases of Liberty Bonds. Thornton (1984) reports that the Federal Reserve tripled its holdings of government debt between 1943 and 1946 by agreement with the Treasury. Consumer prices rose 18.1% in calendar year 1946 and 8.8% in 1947 (Bureau of Labor Statistics, 2011). By FY 1948, the Debt-to-GDP ratio had dropped to 79.9%. The ratio continued to fall through the early 1970s as economic growth outpaced the accumulation of debt. The ratio stabilized in the mid-1970s and then – after being temporarily suppressed by another round of inflation – climbed through the 1980s and early 1990s. This period was also characterized by high interest rates, increasing the Treasury’s debt service costs.
Meanwhile, as reported in The Economist (2011, June 23), the US Treasury failed to redeem $122 million of Treasury bills on time after another debt ceiling debate in 1979. This episode was purely a technical default, arising from systems issues.
The ratio of interest expense to revenues achieved a recent peak of 18.43% in FY 1991. Tax increases and spending cuts in the early 1990s, followed by rapid economic growth in the late 1990s substantially improved US debt ratios. By FY 2002, interest expense had dropped below 10% of federal revenues, while publicly held debt fell to less than 35% of GDP. Persistent large deficits over the last nine years have substantially increased the nations’ Debt-to-GDP ratio, while interest payments as a proportion of revenues have remained relatively stable due to low interest rates.
In 1933, an interest to revenue ratio exceeding 30% preceded a change in the terms of Treasury securities widely regarded as a default. After World War II, the record high Debt-to-GDP ratio of 112.7% was reduced in large measure through price inflation. Budget projections from various sources suggest that these two ratios will return to their historic highs during the late 2020s and 2030s in the absence of major policy changes. If these projections are realized, history suggests that holders of Treasury instruments will be subject to substantial risk.
While inflation would undoubtedly be the politically preferred method of restoring debt ratios to sustainable levels, outright default is also possible. Since the Federal Reserve chair’s term is not aligned with that of the President or Congress, it is possible that a chair politically disconnected from elected leadership may not succumb to pressure to monetize the debt. Further, regardless of political alignment, a Federal Reserve chair may decide that the price inflationary consequences of monetizing debt are worse than the consequences of a forced restructuring of Treasury debt. Jeff Hummel (2009), the first economist to predict a Treasury default, makes some more elaborate arguments as to why the federal government might choose default over inflation.
Conclusion: The US Treasury has defaulted in the past and it has a material risk of doing so again. Absent substantive budget reforms in the current debate, it is hard to see any justification for leaving the US at AAA.
* Marc Joffe (email@example.com) is a consultant in the credit assessment field. He previously worked as a Senior Director at Moody’s Analytics. This article reflects his personal opinion of sovereign rating practice. Although previously employed by Moody’s Analytics, the author no longer works at Moody’s and, when he did work there, his area of professional responsibility was software development and data collection. He had no professional experience as a ratings analyst, and no knowledge of Moody’s ratings practices beyond what is in the public record.
Economist (2011, June 23). The mother of all tail risks. http://www.economist.com/node/18866851.
Hummel, J. R. (2009). Why default on U.S. Treasuries is likely. Library of Economics and Liberty. Retrieved from http://www.econlib.org/library/Columns/y2009/Hummeltbills.html.
Mergent Corporation (1933, 1934). Moody’s Manual of Municipal and Government Bonds. Available online by subscription from Mergent Corporation, Fort Mill, SC.
Riley, J. C. (1978). Foreign Credit and Fiscal Stability: Dutch Investment in the United States, 1781-1794. The Journal of American History (65), 654-678.
Reinhart, C. M. & Rogoff, K. S. (2009). This Time is Different: Eight Centuries of Financial Folly. Princeton: Princeton University Press. Pages 112-113.
Thornton, D. L. (1984). Monetizing the Debt. The Federal Reserve Bank of St. Louis Review, 66(12), 30-43. Retrieved from http://research.stlouisfed.org/publications/review/84/12/Monetizing_Dec1984.pdf.
Wall Street Journal (1933, May 4). Editorial: No Payment in Gold. Page 6.
Wall Street Journal (1933, May 27). Acts to Cancel Gold Clause: Administration Bill in Both Houses Seen as Move Toward Long Term Bond Issue. Page 1.