Do Tax Cuts Pay for Themselves?

            In a USA Today article published on December 14, 2010 titled, “Why tax cut is a bad deal,” Mitt Romney states, “In many cases, lowering taxes can actually increase government revenues.  If new businesses, new investments and new hiring are spurred by the prospects of better after-tax returns, the taxes paid by these new or growing businesses and employees can more than make up for the lower rates of taxation.”  The philosophy that tax cuts can pay for themselves has become pervasive among Republican politicians over the past 30 years, and is the direct result of the original Laffer curve argument that led George H.W. Bush’s to state that Ronald Reagan was following a policy of “voodoo economics”.

            Economists have long accepted the view that a reduction in tax rates can stimulate the economy and lead to increased output and employment.  But the extreme view that such a stimulation will lead to an increase in the tax base large enough that total revenues will actually rise has never been supported by facts or reality.  Yet, when Laffer first made his proposals in the late 1970s politicians like Reagan could not help but endorse a proposition that would allow a reduction in tax rates while also claiming to be able to balance the budget, or even increase spending.  And virtually every Republican politician since Reagan has used the Laffer argument to propose constant reductions in taxes.
            Let me review again the original Laffer curve arguments.  Laffer argued that a decrease in tax rates will lead to an increase in the incentives for taxpayers to work and produce since they will keep a larger share of the income they earn. In addition, any production and income that has been unreported as part of the underground economy will also surface and become part of the tax base.  But Laffer argued that this increase in the tax base will more than offset the decrease in tax rates, so total revenue will actually rise rather than fall.  Laffer could therefore argue that it was possible to cut tax rates, raise spending, and still not increase the size of the budget deficit.  Hence the now popular Republican phrase that “Tax cuts pay for themselves”.

             A number of questions and objections have been raised by economists. For example, the curve is extremely difficult to test empirically. At tax rates below some “optimum” tax rate, increases would lead to increases in revenue, as the rate increases more than offset any fall in the tax base.  At tax rates above the optimum, rate increases would cause a fall in the base that would offset the rate increase, so revenues would fall.  Above the optimum the opposite is also true: a decrease in tax rates would lead to an increase in tax revenue.  But even Laffer did not know for sure where the optimum tax rate was, and therefore did not know if rate increases would lead to revenue increases or decreases. But recent Republican arguments imply that any tax rate is too high and a cut will lead to an increase in overall revenues.  So even when current rates are at a maximum of 35 percent, and rates on dividends and capital gains are at a 15 percent maximum, they continue to call for a reduction as a means of raising more revenue.

            Further, Laffer did not argue that the process was instantaneous, i.e. that tax cuts would lead to instant increases in tax revenues. It takes time to create new investment and production incentives and for new income and tax revenues to be raised. But how long does it take? If tax revenues fall for the first few years is the theory disproved?  If tax revenues after one year are equal to previous projections without a cut, has the theory been proved or disproved? What about two years or three years later? Many forces work on the tax system, and the incentives to produce more or less are some of the most difficult to measure.
     
             But while economists recognize the truth in the arguments that reducing high marginal tax rates will increase saving and investment, it is not obvious that the additional production and income generated by that saving and investment will yield enough new tax revenue to offset the revenue loss of the original cut in tax rates. For example, assume that the marginal rates are cut enough that tax revenues are decreased by $25 billion. If the tax system collects an average of 20 percent of income, total incomes must rise by $125 billion before the lost revenue will be recovered. In the extreme case, this would require that all of the original tax cut would be saved, invested, and then generate $5 in new income for each dollar invested. It is more likely that less than half would be saved and invested and that the new income generated would be less than $5 for each $1 invested. Thus, part, but not all, of the tax cut would be recovered during the first year.

             While most economists would accept the logic that says tax cuts for the rich will increase saving and investment incentives, few accept the Laffer position that those incentives will be adequate to cause total tax revenues to rise rather than fall in the short-term. But incentives could certainly be great enough to cause output and incomes to rise. The question is not whether the chain of benefits exits, but rather the magnitude of those benefits. 

            But aside from the logical shortcomings of the Laffer argument, has experience with such tax cuts led to a surge in revenues that resulted in the Laffer predicted budget outcome?  There have been two Laffer experiments with tax cuts for the rich that were argued to produce an increase rather than a fall in tax revenues.  The Reagan 1981 tax cuts and the 2001/2003 Bush tax cuts were both predicted to have a Laffer response and rising tax revenues.  We will examine the growth in output, tax receipts, expenditures, and the overall budget for a five year period following each tax cut to test the validity of such claims.

 Table 1

Impact of the 1981 Reagan Tax Cut

Year   GDP    % Change  Taxes   % Change   Spending % Change  Deficit  % Change

1981    3126.8      --            599.3        --              678.2           --             79.0          --

1982    3253.2      4.0         617.8        3.1            745.7          9.5            128.0        62.0

1983    3534.6      8.6         600.6       -2.8           808.4          8.4            207.8        62.3

 1984    3930.9      11.2       666.5       11.0         851.9          5.4            185.4       -10.8

1985    4217.5      7.3         734.1       10.1          946.4        11.1            212.3        14.5

1986    4460.1      5.7         769.2        4.8          990.4          4.6            221.2          4.2

81-86     --         42.6            --          28.3            --            46.0               --         180.0
Source:  Economic Report of the President, 2010

            Several things stand out in this table.  In the first two years after the 1981 tax cuts tax revenues increased slightly and then decreased by virtually identical amounts, so the 1983 revenues were the same as the 1981 revenues.  Revenue then increased substantially in both 1984 and 1985.  But 1981-82 was a recession period when GDP grew slowly and government spending increased significantly, both contributing to a rising budget deficit, which nearly tripled from 1981 to 1983.  Both GDP and tax revenue increased by 11 percent in 1984 and the deficit fell by almost 11 percent as well.  Overall, for the five year period following the tax cuts GDP increased by 42.6 percent, tax revenue rose by 28.3 percent, and expenditures rose by 46 percent.  Do these numbers prove or disprove the Laffer prediction? 

            Revenues declined before rising, but then rose by a substantial percentage.  But was the rise a result of the tax cuts or other economic factors?  Examination of the five year period prior to the tax cuts is useful.  From 1975 to 1980 GDP rose by 70.2 percent, tax revenues rose by 85.3 percent, and government spending rose by 77.8 percent.  All of these increases are substantially greater than those following the tax cuts in 1981.  In particular, tax revenues rose more than three times as fast during the late 1970s and they did after the Reagan tax cuts.  If that percentage had continued through the eighties tax revenues would have been over $958 billion by 1986, or 24.6 percent greater than they actually were.  The argument for supply-side tax cuts is certainly not strengthened by the Reagan experience.

Table 2

Impact of the 2001/2003 Bush Tax Cuts

 Year    GDP    % Change   Taxes   % Change    Spending % Change    Deficit % Change

2001    10286.2       --        1991.2         --             1862.8           --          +128.2         --
2002    10642.3       3.5      1853.2        -6.9           2010.9           7.9         157.8      223.4
2003    11142.1       4.7      1782.3        -3.8           2159.9           7.5         377.6      139.3
2004    11867.8       6.5      1880.1         5.5           2292.8           6.2         412.7          9.3
2005    12638.4       6.5      2153.6        14.5          2472.0           7.8         318.3       -22.9
2006    13398.9       6.0      2406.9        11.8          2655.1           7.4         248.2       -22.0
2007    14061.8       4.9      2568.0         6.7           2728.7           2.8         160.7       -35.4
2008    14369.1       2.2      2524.0       -1.7           2982.5           9.3         458.6       185.4
01-06        --         30.3       --                20.9          --                 42.5          - -           293.6
03-07        --         26.2       --                44.1          --                 26.3          --            -57.4
01-08        --          9.7      --                26.8          --                   60.1          --            457.7

Source:  Economic Report of the President, 2010

            Do the Bush tax cuts prove the Laffer thesis?  For the first two years following the initial 2001 cuts, revenues fell by over 10.5 percent before increasingly substantially in 2004 and 2005.  But expenditures rose significantly between 2001 and 2004, so the budget deficit increased from a surplus of $128.2 billion to a deficit of $412.7 billion.  But the rising deficit was clearly a result of both falling revenues and rising expenditures.  Deficits then declined for three years as revenues grew faster than spending, but the 2008 recession caused the deficit to balloon upward again to its highest level of the decade, as revenues fell and spending expanded.
            Again, it is helpful to look at the five year period prior to the Bush tax cuts to compare growth in GDP, revenues, and expenditures for the two periods.  From 1995 to 2000, GDP increased by 34.2 percent, tax revenues rose by 49.8 percent, and spending increased by 18.0 percent, and the budget deficit moved from a $164 billion deficit to a $236.2 billion surplus.  Again, part of the budget change can be attributed to rising revenues and part to slower growth in spending.  But if the 1995-2000 trends in tax revenues had been continued, total revenues in 2006 would have been $2983, or 24 percent higher than they actually were.

            What these figures prove is that supply-side tax cuts lead to an initial decline in tax revenues, but to an increase in the third and fourth years out from the cuts, but then a slowing to a more historical trend.  But the initial fall in revenues results in lower total revenues during the five year period than would have been the case without the cuts.  GDP also grows at a similar or slower rate following such tax cuts as it did prior to the cuts.  The deficit growth is a combination of falling revenues and rising expenditures during the initial years, but as spending growth slows so does the deficit.  It is clear that the deficit will not fall unless tax cuts are also accompanied by spending cuts, but that is what economic theory would have predicted long before the Laffer curve became popular Republican theology.  The evidence clearly does not support the theory that tax cuts pay for themselves.