On February 21, 2006, an editorial by Senate majority leader Bill Frist appeared in USA Today. It is currently posted on his web site and is titled "Tax Cuts Make Money". It begins as follows:
Many people in Washington have long known a dirty little secret about tax-cut measures: When done right, they actually result in more money for the government.
Ever since the Senate approved the last major tax relief bill, in 2003, revenues have increased every year. In 2004, they went up 5.5%. Last year, they rose 14.5%, the largest increase in nearly 25 years.
Frist is correct about the 5.5% and 14.5% increase and the fact that the 14.5% increase was the largest in nearly 25 years (since a 15.9% increase in 1981). However, this gain followed a loss of 12% from 2000 to 2003. This was the largest 3-year loss since a similar 12% loss in 1947, at the end of the Second World War. The comparison is even starker if one looks at individual income tax revenues. From 2003 to 2005, those revenues increased 16.8%, the largest increase since an increase of 21.2% from 1998 to 2000, just five years earlier. However, there was a loss of 21% in individual income tax revenues from 2000 to 2003, the largest 3-year loss since at least the Great Depression. These numbers can be seen in the following table:
TOTAL AND INDIVIDUAL INCOME TAX REVENUES
(billions of dollars)
Total 1-yr % 2-yr % 3-yr % Indiv 1-yr % 2-yr % 3-yr %
Year Revenues Change Change Change Income Change Change Change
-----------------------------------------------------------------------
1934 3.0 0.4
1935 3.6 22.1 0.5 25.5
1936 3.9 8.7 32.8 0.7 27.9 60.5
1937 5.4 37.3 49.3 82.3 1.1 62.0 107.2 160.0
1938 6.8 25.3 72.1 87.1 1.3 17.8 90.8 144.0
1939 6.3 -6.8 16.9 60.5 1.0 -20.0 -5.8 52.7
1940 6.5 4.0 -3.0 21.6 0.9 -13.3 -30.6 -18.3
1941 8.7 33.0 38.4 29.0 1.3 47.3 27.7 2.2
1942 14.6 68.0 123.5 132.5 3.3 148.3 265.8 217.1
1943 24.0 64.0 175.5 266.5 6.5 99.4 395.1 629.3
1944 43.7 82.3 198.9 402.1 19.7 202.9 503.9 1399.6
1945 45.2 3.2 88.2 208.6 18.4 -6.8 182.4 463.0
1946 39.3 -13.0 -10.2 63.7 16.1 -12.4 -18.3 147.5
1947 38.5 -2.0 -14.7 -12.0 17.9 11.4 -2.4 -9.0
1948 41.6 7.9 5.8 -8.0 19.3 7.7 20.0 5.1
1949 39.4 -5.2 2.3 0.3 15.6 -19.5 -13.3 -3.4
1950 39.4 0.1 -5.1 2.4 15.8 1.3 -18.4 -12.2
1951 51.6 30.9 31.0 24.2 21.6 37.2 39.0 11.9
1952 66.2 28.2 67.8 67.9 27.9 29.2 77.3 79.6
1953 69.6 5.2 34.9 76.5 29.8 6.7 37.9 89.2
1954 69.7 0.1 5.3 35.0 29.5 -0.9 5.8 36.7
1955 65.5 -6.1 -6.0 -1.1 28.7 -2.7 -3.6 2.9
1956 74.6 14.0 7.0 7.2 32.2 12.0 9.0 8.0
1957 80.0 7.2 22.2 14.8 35.6 10.7 23.9 20.6
1958 79.6 -0.4 6.8 21.7 34.7 -2.5 7.9 20.8
1959 79.2 -0.5 -0.9 6.3 36.7 5.7 3.1 14.1
1960 92.5 16.7 16.1 15.6 40.7 10.9 17.3 14.3
1961 94.4 2.0 19.1 18.5 41.3 1.5 12.6 19.0
1962 99.7 5.6 7.8 25.8 45.6 10.2 11.9 24.1
1963 106.6 6.9 12.9 15.2 47.6 4.4 15.1 16.9
1964 112.6 5.7 13.0 19.3 48.7 2.3 6.9 17.8
1965 116.8 3.7 9.6 17.2 48.8 0.2 2.5 7.1
1966 130.8 12.0 16.2 22.8 55.4 13.6 13.9 16.5
1967 148.8 13.7 27.4 32.2 61.5 11.0 26.1 26.3
1968 153.0 2.8 16.9 31.0 68.7 11.7 24.0 40.9
1969 186.9 22.2 25.6 42.8 87.2 27.0 41.8 57.4
1970 192.8 3.2 26.0 29.6 90.4 3.6 31.6 46.9
1971 187.1 -2.9 0.1 22.3 86.2 -4.6 -1.2 25.5
1972 207.3 10.8 7.5 10.9 94.7 9.9 4.8 8.6
1973 230.8 11.3 23.3 19.7 103.2 9.0 19.7 14.2
1974 263.2 14.0 27.0 40.7 119.0 15.2 25.6 37.9
1975 279.1 6.0 20.9 34.6 122.4 2.9 18.5 29.2
1976 298.1 6.8 13.2 29.1 131.6 7.5 10.6 27.5
1977 355.6 19.3 27.4 35.1 157.6 19.8 28.8 32.5
1978 399.6 12.4 34.1 43.2 181.0 14.8 37.5 47.9
1979 463.3 16.0 30.3 55.4 217.8 20.4 38.2 65.5
1980 517.1 11.6 29.4 45.4 244.1 12.0 34.9 54.8
1981 599.3 15.9 29.3 50.0 285.9 17.1 31.3 58.0
1982 617.8 3.1 19.5 33.3 297.7 4.1 22.0 36.7
1983 600.6 -2.8 0.2 16.1 288.9 -3.0 1.1 18.4
1984 666.5 11.0 7.9 11.2 298.4 3.3 0.2 4.4
1985 734.1 10.1 22.2 18.8 334.5 12.1 15.8 12.4
1986 769.2 4.8 15.4 28.1 349.0 4.3 16.9 20.8
1987 854.4 11.1 16.4 28.2 392.6 12.5 17.3 31.5
1988 909.3 6.4 18.2 23.9 401.2 2.2 15.0 19.9
1989 991.2 9.0 16.0 28.9 445.7 11.1 13.5 27.7
1990 1032.1 4.1 13.5 20.8 466.9 4.8 16.4 18.9
1991 1055.1 2.2 6.4 16.0 467.8 0.2 5.0 16.6
1992 1091.3 3.4 5.7 10.1 476.0 1.7 1.9 6.8
1993 1154.5 5.8 9.4 11.9 509.7 7.1 8.9 9.2
1994 1258.7 9.0 15.3 19.3 543.1 6.5 14.1 16.1
1995 1351.9 7.4 17.1 23.9 590.2 8.7 15.8 24.0
1996 1453.2 7.5 15.4 25.9 656.4 11.2 20.9 28.8
1997 1579.4 8.7 16.8 25.5 737.5 12.3 24.9 35.8
1998 1722.0 9.0 18.5 27.4 828.6 12.4 26.2 40.4
1999 1827.6 6.1 15.7 25.8 879.5 6.1 19.3 34.0
2000 2025.5 10.8 17.6 28.2 1004.5 14.2 21.2 36.2
2001 1991.4 -1.7 9.0 15.6 994.3 -1.0 13.1 20.0
2002 1853.4 -6.9 -8.5 1.4 858.3 -13.7 -14.5 -2.4
2003 1782.5 -3.8 -10.5 -12.0 793.7 -7.5 -20.2 -21.0
2004 1880.3 5.5 1.5 -5.6 809.0 1.9 -5.8 -18.6
2005 2153.9 14.5 20.8 16.2 927.2 14.6 16.8 8.0
Source: Budget of the United States Government, FY 2007: Historical Tables, Table 2.1
Frist continues:
Total government collections, in fact, increased more after President Bush's 2003 tax cuts than they did after President Clinton's 1994 tax hikes.
Frist is evidently referring to the fact that total revenues increased 20.8% from 2003 to 2005 but increased just 15.4% from 1994 to 1996. However, if one looks at individual income tax revenues, that gain was 16.8% from 2003 to 2005, less than the 20.9% gain from 1994 to 1996. Even looking at total revenues, the picture changes greatly if one widens their view beyond two years. As the above table and following graph show, the 2003 to 2005 gain was a recovery from a very deep drop whereas the 1994 to 1996 gain was part of an 8-year rise in revenues.
The actual numbers and sources can be seen at this link. In any case, Frist continues:
In 2000 and 2001, the end of the dot-com bubble, the 9/11 attacks and a series of corporate scandals sent the economy into a tailspin. During the downturn, high taxes limited economic growth and kept receipts down. Although Americans were making some of the largest per-household tax payments in our nation's history, revenues plummeted in 2002 and 2003. When the major tax-relief measures kicked in, they restored the economy to health and helped deliver quarter after quarter of strong growth.
As the above table shows, individual income taxes plummeted 20.2% in 2002 and 2003. How could this have happened if Americans were "making some of the largest per-household tax payments in our nation's history"? In any case, Frist fails to make any mention of the 2001 tax cut. Of course, admitting that this tax cut may have had anything to do with the drop in revenues would greatly confuse the point that Frist is trying to make.
Frist continues:
Republicans' decision to reduce taxes on capital gains and dividends provides a good case study in effective tax policy. When we enacted these measures in 2003, the Congressional Budget Office estimated that revenues would decline by $27 billion over the next two years. Instead, it turned out that the tax cut stimulated investment and increased revenues by $26 billion — a $53 billion difference.
As Frist gives no source for these numbers, it was difficult to figure out exactly which revenues he is referring to. As it turns out, he appears to be referring to a January 27, 2006 National Review Online article by Donald Luskin titled "The 2003 Tax Cut on Capital Gains Entirely Paid for Itself". The following table summarizes the numbers given by that article:
Projected and Actual Capital Gains Tax Liabilities
(billions of dollars)
Change from
Release Date 2004 2005 Total January 2003
------------ ---- ---- ----- ------------
January 2003 60* 65* 125*
January 2004 46* 52* 98* -27
January 2006 71 80 151 26
* projected
Source: CBO Budget and Economic Outlook, January 2003, 2004, 2006
The basic argument put forth by Luskin and Frist appears to as follows: Actual Capital Gains Tax Liabilities turned out to be greater in 2004 and 2005 than the Congressional Budget Office projected that they would be in January of 2003. Therefore, the 2003 capital gains tax cut increased revenues.
It's hard to know where to start with this. It assumes that the 2003 CBO projections accurately predicted every event of the next three years except for passage of the 2003 capital gains tax cut. In addition, it assumes that the CBO accurately evaluated the effect of each of these events upon capital gains. Among the events that the CBO would have had to accurately predict and evaluate was the 41 percent rise in the S&P 500 from January 2003 to January 2006. The possibility that the CBO simply made an inaccurate projection, regardless of the reason, never seems to enter the minds of Luskin or Frist.
Frist concludes his editorial:
If we really want to avoid burdening our children and grandchildren with debt — which does represent a major problem — we need to reform entitlement programs. Within the lifetimes of today's college students, the combined budgets of Social Security, Medicare and Medicaid will consume all federal revenues, leaving nothing for defense, education, housing or any other program.
Making sure that our children and grandchildren don't face the burdens of debt requires that we reform these entitlement programs and set them on a sustainable course for the future. A sensible low-tax policy that keeps the economy growing will play a major role in confronting our fiscal challenges.
Frist is correct that we need to reform entitlement programs. One problem with our current programs is that they promise beneficiaries longer and longer retirements without providing any realistic way to pay for them. We need to be skeptical of any painless, free-lunch proposals to pay for those longer retirements. Likewise, we need to be skeptical of free-lunch proposals to provide tax cuts that pay for themselves. I suspect that much of the deficient analysis in Frist's editorial came from other sources. I hope he becomes much more skeptical and checks such analysis more carefully in the future before putting his name to it.