Go to Brad De Long's Home
Dole's Forty Percent Solution
Why "Dynamic" Effects Will Amplify Revenue Losses
from Tax Cuts
James Bradford De Long
James Bradford De Long is Associate Professor of Economics at the University
of California at Berkeley. He is a partisan Democrat. From 1993 to 1995
he worked for the U.S. Treasury as Deputy Assistant Secretary for Economic
Sometime in the near future the Republican standard-bearer, Robert Dole,
will announce his economic program: a large tax cut.
The last time big tax cuts were in the Republican Party's platform, little
attention was given to the impact of such tax cuts on federal finances,
the national debt, and the American economy. Politicians talked of how stronger
economic growth would lead to rising revenues. Politicians made gestures
in the direction of a small tribe of supply-siders, who claimed that cutting
tax rates would increase revenues. But the general feeling was that the
Democrats had had so much success with tax-and-spend that it was time to
try to counter it with spend-and-borrow, and that if the economy and the
budget got into serious trouble as a result of the tax cut they could always
fix it later.
Well. it''s later. The American economy and the federal budget did get into
serious trouble. Things are still not completely fixed. American productivity
and living standards are lower, and the country weaker, than had the1980s
seen the economic policies of, say, an Eisenhower.
Yet once again the Republican Party is set to run on a platform of large
federal tax cuts.
This time there will be no handwaving about cutting rates not reducing revenues.
This time tax cuts will be "paid for": offsetting spending cuts
will be proposed. Or, rather, some spending cuts--for it looks like former
Senator Dole will claim that faster growth will replace forty percent of
revenue lost, so only sixty cents' worth of spending need to be identified
for every dollar of tax cut.
But there is every reason to think that this claim is wrong, and that a
dollar's worth of tax cut, as assessed--"scored" is the word used--by
the Congressional Budget Office will widen the deficit by much more than
Why? Begin with the link between the federal deficit and national savings.
An extra dollar cut from taxes is an extra dollar added to the deficit,
an extra dollar borrowed from the public, and thus an extra dollar not saved
and invested by Americans in boosting the country's productive potential.
Cut taxes by $100 billion, leave them lower (and the deficit higher) for
a decade, and find after a decade that $1000 billion of America's productive
capital stock is missing--and that this loss of productive capital has reduced
GDP by some $100 billion a year, and that the missing GDP has caused a further
revenue shortfall of some $30 billion a year.
Continue with the link between high national debt and high interest rates.
If you had told someone in the late 1970s what the actual rate of inflation
has been in the 1990s and asked them to predict interest rates, they would
probably have forecast interest rates two or three percentage points below
today's values. Why are today's U.S. interest rates so high? A prime suspect
is the budget deficits of the 1980s that have left us with a high national
debt. If correct--and I think it is--a $100 billion tax cut sustained for
a decade would leave us with interest rates higher by perhaps a percentage
point, and with annual federal spending on debt interest higher (both because
of more debt and because of a higher interest rate on existing debt) by
some $100 billion.
Go a step further, with the link between high federal borrowing and the
collapse of America's export industries that was so evident in the 1980s:
a dollar borrowed from Japanese investors to finance the federal deficit
is a dollar that is not spent on America's exports. I still have not seen
a good analysis of the social cost of the destruction of midwestern export
industry in the 1980s as a byproduct of the overvalued dollar induced by
the federal deficit. My personal back-of-the-envelope guess--I won't even
call it an estimate--is that this is another $30 billion reduction in annual
GDP in this context, and another $10 billion in lost revenue.
Conclude with the positive supply-side effects of lower tax rates in boosting
economic activity at the margin and in reducing the amount of tax avoidance.
These effects are there. As Charles Schultze likes to say, there was never
anything wrong with supply-side economics that dividing by ten would not
cure. In assessing these effects the devil is in the details. But for a
tax cut aimed at America's broad middle class it does not seem unfair to
take Schultze at his word, and divide the supply-sider estimate by ten:
instead of a cut in tax rates not reducing revenues, suppose that the $100
billion tax cut generates an extra $10 billion in revenues from expanded
GDP--and throw in an extra $10 billion from reduced tax avoidance to be
But note that this is a two-edged sword. Spending cuts to offset tax reductions
have negative supply-side effects of their own: reductions in infrastructure
investment may be very costly in reducing growth, reductions in spending
that supports education lower Americans' skills and GDP, and reductions
in business investment may well be (I think probably are) much more expensive
in terms of lost GDP than standard rules of thumb allow. Do you gain more
in added GDP from lower taxes than you lose from the reduction in public
investment as spending is cut? I think not.
If you asked me for my best estimate of how much of the tax cut would be
offset by the revenues generated by faster economic growth, I would say
much less than zero: itt would shift the direction of the American economy
in such a way that after a decade the deficit would be higher not by $100
billion a year but by something like $240 billion because the higher deficit
leads growth to slow and interest payments to rise.
I know of no serious economist who thinks that the U.S. economy is in better
shape now than had we followed Eisenhower-style economic policies in the
1980s: everyone's assessment of the past fifteen years is that the early
1980s tax cuts set in motion a dynamic that reduced savings and investment
and slowed growth--and thus amplified, not offset, the revenue lost from
the early 1980s tax reductions. Thus when I hear Republican political operatives
say that one dollar of tax cuts will only reduce revenues by sixty cents
(and that such an assumption is an act of conservative prudence), I get
frightened. I fear that the old Republican political operative medicine
of rosy scenario and falsified forecast is back.
So I hope Robert Dole backs off, and does not do what his handlers are priming
him to do. Policies based on rosy scenarios are bad for the country, if
implemented. Yet campaign promises that are not implemented poison the presidency
that repudiates them.
Go to Brad De Long's Home
Associate Professor of Economics Brad De
Long, 601 Evans
University of California at Berkeley; Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax