Did Tax Increases or Spending Cuts Preface the 1990s Boom?

Report Taxes

Did Tax Increases or Spending Cuts Preface the 1990s Boom?

June 12, 2013 2 min read Download Report
Salim Furth
Salim Furth
Former Research Fellow, Macroeconomics
Salim was a Research Fellow in Macroeconomics at The Heritage Foundation.

Following Senate Budget Committee testimony, Senator Sheldon Whitehouse (D–RI) asked me seven “questions for the record.” The entire exchange will be publicly available, but a question about the U.S. economic boom of the 1990s deserves more attention.

Senator Whitehouse asked, “Are you familiar with the U.S. experience in the 1990s, during which tax rate increases in 1993 were followed by 7 years of economic growth at 4% per year, with 23 million new jobs created? How do you explain this prosperity following major tax increases?” My full response follows.

The early 1990s are a great example of the success of structural reform and spending cuts. Fiscal consolidation from 1993 on featured 67 percent spending cuts and 33 percent tax increases. International Monetary Fund economists recently quantified a detailed narrative of the tax increases and spending cuts during that era.[1]

Again, cautioning against drawing too much from a single example, the early 1990s featured steady fiscal consolidation in the U.S. as well as welfare reform (a key structural reform) in 1996. The table below shows the fiscal consolidation undertaken each year and the ensuing real per-capita gross domestic product (GDP) growth.

The year 1990 had fiscal consolidation mainly on the tax side. A recession followed. Consolidation accelerated with an even tax/spend split over the next two years, and the economy recovered—but less rapidly than after most previous recessions.

The year 1993 continued the spending cuts but with few tax increases, and the economy boomed at 2.8 percent growth from 1993 to 1994.

The large spending cut and tax increase passed in August 1993 had its greatest effects during 1994.[2] In particular, the Omnibus Budget and Reconciliation Act of 1993 sought major savings from Medicare and federal employee benefits,[3] which are good examples of the structural reforms I recommended in my testimony.

GDP per capita grew only 1.3 percent from 1994 to 1995. That’s not bad, reflecting a private sector that rapidly picked up the slack as government’s growth slowed.

From 1995 to 1998, fiscal consolidation was heavily on the spending side, and growth accelerated to a smoking 3.6 percent, and the deficit turned to a surplus.

The fact that growth was strongest right after spending cuts preponderated and weakest when taxes increased most is an excellent exhibit of the case for preferring spending cuts. Using a regression to quantify the correlations,[4] I find that a 0.1 percent of GDP cut in spending is associated with 1.2 percentage point higher GDP growth. And a similar tax increase is associated with 1.4 percentage point lower GDP growth.[5] These coefficients have no applicability out of sample, but they tell us that in the 1990s, U.S. higher taxes and lower growth went together like fire and smoke.

Good economists do not draw conclusions based on a handful of data points. The argument against spending cuts leans heavily on blaming Europe’s failed recovery for austerity. Just as the U.S. experience in the 1990s does not prove that spending cuts are expansionary, the European experience in the 2010s cannot prove that spending cuts are contractionary.

—Salim Furth, PhD, is Senior Policy Analyst in Macroeconomics in the Center for Data Analysis at The Heritage Foundation.

[1]Pete Devries, Jaime Guajardo, Daniel Leigh, and Andrea Pescatori, “A New Action-Based Dataset of Fiscal Consolidation,” International Monetary Fund, June 2011, pp. 81–85.

[2]Ibid., p. 84.

[3]Congressional Budget Office, “The Economic and Budget Outlook: An Update,” September 1993, p. 29, http://cbo.gov/sites/default/files/cbofiles/ftpdocs/76xx/doc7670/09-1993-outlookentirerpt.pdf (accessed June 11, 2013).

[4]I am not claiming causation on the basis of these nine data points. I include a time trend.

[5]Both coefficients are statistically significant at the 95 percent level, but it is still only a correlation.


Salim Furth
Salim Furth

Former Research Fellow, Macroeconomics