This July, New York Times columnist Thomas Friedman praised the elder Bush for “balanced conservatism,” because he “agreed to a compromise with Democrats to raise several taxes.” A few days earlier Washington Post writer Steven Mufson had written: “It may be time to reconsider the history of the 1990 budget deal,” because “as long as any increase in taxes is equated to political suicide, managing America’s finances will prove difficult.”
Last month, former Treasury Secretary Nicholas Brady wrote an article reminding us of his own role in that deal. In Brady’s retelling, “the combination of reining in spending and increasing revenue produced budgetary sanity, economic growth and a sense of calmness in our country for more than a decade.”
Unfortunately, none of that is true. Revenues went down rather than up. Spending went up rather than down. The economy slowed.
Real economic growth slowed to 2.5% from the fourth quarter of 1990 through the first quarter of 1996, after growing 4% a year from 1983 to 1990. There was, to be sure, a brief recession from the moment OBRA-90 was announced (GDP dropped at a 3.6% pace in the fourth quarter of 1990), and it lasted through March 1991. That is not necessarily a coincidence.
A renowned 2007 study by former Obama adviser Christina Romer and her husband David found that “tax increases are highly contractionary. The effects are strongly significant … The large effect stems in considerable part from a powerful negative effect of tax increases on investment.”
![]()
As Table 1 shows, federal tax receipts were 18% to 18.4% of GDP before OBRA-90, but then dropped to 17.5%-17.8% in 1991–92 when the Bush tax increase took effect, and remained depressed in 1993 when the Clinton tax hikes took effect.
Individual income taxes before OBRA-90 amounted to 8.3% of GDP in 1989 and 8.1% in 1990, but they dropped to 7.7% of GDP by the time of the 1993 tax increases (e.g., on Social Security benefits and high incomes) and were still only 8% of GDP in 1995.
The official rosy estimate of OBRA-90 was that the numerous income and excise tax increases would raise $146.3 billion over five years, with $9 billion of that coming from the 10% luxury tax on boats, airplanes, cars, jewelry and furs. Such wishful estimates may be one reason for the lingering myth that the Obama‐like 1991 experiment of combining higher tax rates with shrinking deductions was actually effective in bringing in more receipts.
The luxury taxes of 1991 were wisely repealed in August 1993 because they destroyed far more revenue as a result of lost wages and profits in the boating and aircraft industries than could possibly be raised by these excise taxes.
The Joint Tax Committee had estimated that the luxury tax would raise $6 million from airplane sales alone in fiscal 1991, but the actual take was $53,000. The 10% tax on cars priced above $30,000 brought in a mere $88 million in 1991 while losing much more in income and payroll taxes as a result of depressed auto sales and profits.
Even higher taxes on tobacco, alcohol and gasoline had a trivial effect, with excise tax receipts rising from 0.6% of GDP in 1989–90 to 0.7% from 1991 to 1993.
What about the spending side of the 1990 deal? Federal spending rose from 21.2% of GDP in 1989 to 22.3% in 1991 and 22.1% in 1992. Yet the Post’s Mufson nevertheless claimed “the 1990 budget deal was a success: It helped shrink the deficit, then at 5% of gross domestic product, by $492 billion — $850 billion in today’s dollars — over just five years.”
The facts are the opposite. The budget deficit was 3.9% of GDP before the 1990 deal, and it rose to 4.5% in 1991 and 4.7% in 1992. The deficit did not even get back down to the pre‐deal level of 3.9% until 1993, even though defense and international spending fell steadily from 6.5% of GDP in 1985 to 4.9% in 1991 and 3.1% by 1999.
Overall federal spending fell from 22.1% of GDP in 1992 to 18.1% in 2001 ? mainly because of greatly reduced military spending, not the 1990 deal. After 1993, the relatively modest savings in nondefense spending had much more to do with what President Clinton and a Republican Congress accomplished than with what President Bush and a Democratic Congress did in late 1990.
In short, the 1990 deal reduced tax revenues substantially for several years, and did not cut spending at all. Tax receipts promptly fell and remained below pre‐OBRA levels for five years. Nondefense spending went up.
The spending spree of the past few years is much more worrisome than what we faced in 1989–90, before the ill‐conceived budget deal made things worse.
Table 2 summarizes the budgetary situation before after President Obama’s American Recovery and Reinvestment Act of 2009 (ARRA). In 2006–2007, tax receipts and spending were near the long‐term norm, about the same shares of GDP that they were in 1994–1995 when income and capital‐gains tax rates were higher.
![]()
The recession and financial crisis of 2008 reduced receipts and added new bailouts and “stimulus” spending, but not much.
In 2009, however, spending as a share of GDP was an astonishing 6 percentage points larger than the average of 1997–2007. That was supposed to be temporary, yet the ratio of spending to GDP was just one percentage point lower by 2011.
Meanwhile, the cyclical revenue shortfall of 2009–2011 was unduly aggravated by unproductive temporary tax cuts — the president’s refundable “making work pay” tax credit reduced revenue by 0.3% of GDP, and payroll taxes fell by $46 billion in fiscal 2011 thanks to the temporary cut in payroll taxes.
So, what can we learn from the budget deal of 1990? The first lesson is that many of the same sorts of “tax increases” now advocated by many congressional Democrats — phasing‐out deductions and raising tax rates at higher incomes — did lasting damage to both tax receipts and economic growth in 1991–95.
The second lesson is that deficits did not begin to come down until spending was reduced, which was due to the hard work of President Clinton and a Republican Congress, not the bad deal President Bush struck with Democrats in 1990.
The U.S. has never managed to raise more than 20% of GDP in taxes, even at the peak of World War II, when personal and corporate tax rates were confiscatory. The idle hope that higher tax rates on top incomes could raise much revenue without damaging the economy just encourages procrastination on the spending side, just as the phony revenue estimates did in 1990.
The fact that federal spending has been increased to 24%-25% of GDP has created a problem that cannot be fixed in any way except by reducing spending and raising GDP growth. Repeating the 1990 tax policy blunders would not help reduce spending, and it certainly would not help raise GDP, so it would not help.
