Updated tax data are in and the spinners will spin. But the numbers can speak for
themselves.
The Mid-Session Budget
Review has been released, showing that federal tax receipts for 2005 are coming in at a rate
that may total nearly $100 billion more than was estimated last February. Let me say at the
outset that, as someone who has had real concerns about the deficit, this strikes me as
unambiguously good news.
Nevertheless, it’s clearly not quite as good news as some have portrayed. Mark
Thoma called my attention to this story from the Washington Times:
This is especially welcome news to supply-side tax-cutters who argued all along that lower tax
rates spur stronger economic growth, which, in turn, creates more jobs that increases tax
revenues. That is happening now.
It’s embarrassing news for President Bush’s diehard Democratic critics, who predicted his tax
cuts would worsen the budget deficits and drive the government deeper into
debt.
The proposition that by cutting the tax rate we might actually increase tax revenue is a
theoretical possibility, although it’s the sort of surprising suggestion for which one would
want to see rather persuasive evidence before taking too seriously, and such evidence certainly
seems hard to find in the historical record.
The closest thing I’ve seen to a numerical analysis that might suggest evidence of its operation
in the current episode is that by Kevin Hassett, though, in fairness to Kevin, he really seemed to be arguing that the tax
cuts hadn’t reduced revenue that much rather than that they’d actually led to an increase in
revenue. Even so, he was only able to support that conclusion by comparing the current forecast
for 2006 with a forecast for that year first made in 1999. Angry
Bear has some compelling concerns about Kevin choosing the base year to be 1999 rather than
2001, as well as some of the other details about how the 1999 projection relates to events as
they actually unfolded.
In any case, this seems to me to be a case where the data speak pretty clearly without the
need for much reinterpretation. The graph at the right displays federal government receipts and
expenditures, each as a percentage of GDP, for the last 35 years. Tax revenues fell from 20.9%
of GDP in 2000 to what is now projected to be 17.4% of GDP for 2005. Federal expenditures rose
from 18.4% to 20.1% over the same period. With tax revenues way down and expenditures way up,
voila, a big surplus turned into a much bigger deficit. In terms of this simple arithmetic, 2/3
of the swing from surplus to deficit would be attributed to lower tax revenues and 1/3 to higher
spending.
It’s hard to see how one can argue that the tax cuts set in motion some economic changes that
would be starting to show up only now or even later. The logical mechanism for such a delay
would presumably come through investment spending. But as Angry Bear again
noted, national saving and investment as fractions of GDP are both down since 2000, so the data
so far would seem to work against such an effect. Rather than the first fruits of an incipient
big return on the tax cuts, it seems much more natural to me to attribute the recent upswing in
tax receipts to the natural
consequence of a cyclical recovery and some
revenue mechanisms that could prove to be temporary.
Nonetheless, the above graph also suggests that using 2000 as the base year for fiscal
comparisons may be a bit misleading, since tax revenues for that year were at their highest
level and expenditures were at their lowest level, as percentages of GDP, of the previous
quarter century. An alternative baseline is to look at what the average values for these two
magnitudes had been over 1970-2000. By this metric, despite the rise since 2000 in government
expenditures as a fraction of GDP, spending is still below the average historical value of 20.9%
observed during 1970-2000. On the other hand, tax revenues in 2003-2004 represented a lower
percentage of GDP than at any point of the preceding quarter century. With the mid-session
revisions, the gap between expenditures and receipts currently stands at 2.7% of GDP, almost
identical to the historical average gap of 2.6%, albeit with both magnitudes shifted down a
little less than 1% relative to the historical averages.
Of course, some might reason that the device I’ve used here to scale the data– dividing by
GDP– assumes away the whole issue, in that the argument could be that tax revenues would fall
as a fraction of GDP, but the level of GDP grow so quickly that total revenues increase.
However, by this reasoning, the other part of the program should be a decrease, not an increase,
in government spending as a fraction of GDP. More to the point, the graph at the left shows that
GDP growth for each year during 2001-2003 was well below the historical average. It is true
that real GDP growth for 2004 of 4.4% was 1.2% faster than the historical average. But as is
also apparent from the graph, any growth slower than this in the recovery phase from an economic
recession would have been somewhat unusual.
It seems to me that any way you want to slice these numbers, unusually low tax revenues have
to be the biggest part of any account of where these unusually large budget deficits came from.
There may be good reasons for wanting high government spending, and there may be good reasons
for wanting low taxes, but the reality is that we have to choose between the two. It would
certainly be nice if cutting tax rates caused tax revenues to go up. But wishing doesn’t make
it so.
I’m not much of a macro guy, so I don’t read too much about this stuff, but…
Has anybody been able to come up with a Laffer curve for the US economy with actual, reliable, meaningful numbers attached? One guy I know swears that the maximum is at about a 90% tax rate, so we *must* be to the left of the maximum, but empirically I wouldn’t know one way or another.
My background is simply an MBA and CFA studies, but I have a different take on the data. Looking at the tax revenue data based a PERCENT of GDP could be misleading – especially the “unchanged” data on Investment as a percent of GDP.
Isn’t there a possibility that the tax decrease indirectly INCREASED the GDP by freeing up investment capital for new corporate projects and venture capital? Companies were able to expand, adding wage growth to the economy. Boosting the denominator of the percent-of-GDP equation is a big part of the Supply-Side formula, from my understanding.
Certainly if you look at investment flows into the stock market, they have increased dramatically as the cost of realizing capital gains has declined by 40%.
I read somewhere that the deficit numbers don’t include the expenditures on the wars because of the way the appropriations were done.
Is this true???
Another very good discussion! All I might add was said very well by a CBPP discussion noted yesterday by Brad DeLong and echoed over at Angrybear.
“However, by this reasoning, the other part of the program should be a decrease, not an increase, in government spending as a fraction of GDP.”
Well, no, not if government spending increases. The argument is that reducing the marginal tax rate can increase the tax revenue. The hypothesis may or may not be true, but the size of government outlays is beside the point.
Regarding Barry’s question:
I’m pretty sure that the 90% number comes from Varian’s Intermediate (undergrad) Micro text. But if you want to figure it out for yourself, just do the following:
Note that revenue is R(t) = t*I(1-t), where t is the tax rate, 1-t is the take-home share, I is taxable income, which is a fctn of t if you believe in microeconomics (why this has come to be called “supply-side” rather than “utility-maximizing” is an interesting question for the intellectual historians). Of course, I am simplifying by assuming a flat tax rate. But I don’t have time to write another dissertation right now.
Take logs and differentiate:
d ln R(t)/dt = (1/t) + I'(1-t)/I(1-t)
Note that the peak of the laffer curve occurs where this derivative equals zero. So we set
I'(1-t)/I(1-t) = -1/t.
Now multiply both sides by 1-t, and you get that the l curve peaks at t* such that
-E = (1-t*)/t*,
where -E is the neg elasticity of taxable income (-E must be positive if you believe that there is an opportunity cost to revenue-generating activity). All serious arguments about the l curve’s peak come down to figuring out what the value of E is, since
t* = 1/[1-E].
Now, it’s only a slight caricature to say that the luskin-bush-novak-delay-kemp types seem to think that -E = infinity, so that every single tax cut raises infinity. Sigh. More respectable folks think -E may be as high as 3, in which case t*=0.25. For technical reasons, I think the evidence for that claim is totally dismissable. A more plausible long-run estimate is probably something in the neighborhood of 0.4, and perhaps as high as 0.6 or as low as 0.
In the 0.4 case, t*=1/[1.4]=5/7, or over 70%.
If I remember right, the 90% estimate is based on earlier estimates of the labor supply elasticity using hours worked, rather than taxable income, as the endogenous supply-side (!) variable.
In any case, t* is a hell of a lot higher than today’s tax rates.
ok, slight typos in that previous post. corrections at
http://cardcarryingmember.blogspot.com/2005/07/laffing-our-way-to-transversal-trouble.html
Thank you, James, for presenting an effective empirical case against the Supply Side Fantasy. You have provided a great public service.
But I wonder if can you tell me why liberal economists have found it so difficult to attack the spurious reasoning of Supply Side Economics on theoretical grounds? It seems a rather simple matter to point out that income tax cuts are CONTRACTIONARY in and of themselves in the same way that savings are a pure “leakage” from the economy. We isolate the leakage effect of savings because we acknowledge the “injection” effect of lending. In the same way, we can also isolate the contractionary effect of tax cuts if we acknowledge the compensating expansionary effect of spending borrowed money.
A government that is only permitted to spend money that it raises through taxes would be forced to contract the economy if it ever wanted to cut taxes. This is because at least some of the money that taxpayers would receive from a tax cut would be saved. Since not all money that is saved is lent out to borrowers, there is a net leakage of money out of the economy (government spending would be decreasing more than private spending would be increasing).
Tax increases, on the other hand, are clearly expansionary. This is because at least some of the money that is collected from tax payers would have been saved. When the government takes money that would have been saved and spends it, the result is a net increase in aggregate spending. In other words, it’s expansionary. The more savings are taxed, the more expansionary the government’s tax hike will be.
Even if an income tax cut were designed to give refunds only to people who would be certain to spend all of it (poor people), it would still not provide any net stimulus to the economy. When spending-cut-dollars match tax-cut-dollars,the money that refunded taxpayers would get to spend would have been spent by the federal government ANYWAY.
Nearly every introductory economics textbook in America today mentions tax CUTS as one of the federal government’s expansionary fiscal policy tools and fails to mention that tax INCREASES are far more effective than tax cuts in stimulating the economy when the money that is collected in taxes would have otherwise been saved.
Why is it that liberal economists are not beating Supply Siders over the head with this theoretical argument, James?
James Kroeger, related to the point you’re making is the observation that the advocates of tax cuts have at times interpreted the effects as coming from what I would call the supply side (raising saving) and at times from the demand side (raising consumer spending). As you point out, which actually happens would also depend on factors such as what you’re assuming happens to government spending.
As for which issues to emphasize, personally I have always found empirical evidence more persuasive than economic theory. On the theoretical side, there are certainly other issues one could bring up besides the ones you mention having to do with plausible magnitudes for various elasticities.
Forgive me for requesting further clarification, James, but I’m having difficulty seeing the theoretical “weakness” you are alluding to:
“…advocates of tax cuts have at times interpreted the effects as coming from what I would call the supply side (raising saving) and at times from the demand side (raising consumer spending). As you point out, which actually happens would also depend on…”
Well, it seems as though we could assume the extremes in order to determine the limits of a range of possibilities. If all of a tax cut is spent on consumption, then there would be no net increase in aggregate spending once the government has reduced its spending by an amount equal to the tax cut.
If, on the other hand, 100% of a tax cut were saved in commercial banks, then only 90% of the tax cut could possibly end up being spent on investment (and transaction costs) because of the Fed’s reserve requirement. If the elasticity of supply is so elastic that 100% of the 90% is borrowed by investing firms, there would still be a net decline in aggregate spending after the first iteration.
Any conceivable mix of the spending & saving options will always give you a net negative change in aggregate spending following a tax cut, with the single exception of simply breaking even when all tax refunds are spent on consumption.
On another point, if I understood correctly, you (implicitly) stated that the worth/accuracy of Supply Side predictions depends in part on “what you’re assuming happens to government spending.”
The very essence of Supply Side claims is that TAX CUTS will cause economic growth to occur. It is a claim that they cannot rationally justify by combining an expansionary initiative (like borrowing money) with the contractionary initiative that they dearly love (tax cuts) and then simply declaring that their beloved tax cuts are an expansionary policy whenever the expansionary component of their combo-policy is of sufficient dimension that it is able to overcome the contractionary component’s effects and finally achieve some actual growth.
I was taught that economists are supposed to be very exacting in their attempts to identify the specific effects of changes in specific variables. Why are we letting Supply Side apologists get away with such glaring theoretical obfuscation?
Ever since the US passed the income tax in 1916, we have cut tax rates (as opposed to taxes) at twenty year intervals, and each time the result has been an economic boom. This was true of the Mellon cuts in the 1920s, the Kennedy cuts in the 1960s, the Reagan cuts in the 1980s, and now the Bush cuts in the 2000s. The only exception is the 1940s when, after all, we were in an all-out war.
Citing the revenues as a percentage of GDP is not on-point, because the whole idea is to grow the economy.
Citing the “low revenue” years 2001-2003, is not on-point, because the tax rate cuts did not take effect until 2003.
That there were other periods of growth does not change the fact that, when rates were cut, the result was long periods of growth–four of the longest in our history.
The idea is not that tax revenues increase via the Laffer curve. That applies only at very high rates, which are restricted to “the rich.” The idea is to improve return on investment and thereby to generate more of it over time.
Four times in 80 years–that’s about as positive as you can get in macroeconomics.
PS. The last years of the 1990s also benefitted from what were, in effect, tax rate cuts: the capital gains rate cut itself, the lowering of inflation to almost zero (inflation is similar to taxes to an investor), and the growth of the internet, a tax-free zone.
“Why are we letting Supply Side apologists get away with such glaring theoretical obfuscation?”
The theory which postulates tax cuts are good for our country is preferred by most voter-consumers. It just has a better ring to it.
“The theory which postulates tax cuts are good for our country is preferred by most voter-consumers. It just has a better ring to it.”
It may have a better ring to it, but that is only because most voter-consumers have not been taught some important truths about about how the marketplace works.
All too often, voters and policy makers make the mistake of believing that everyone would be better off if they were ALL to imitate an action that clearly benefits a single individual. But as a rule, it is only *exceptional* behavior that is rewarded by markets.
If all of us were savers and none of us borrowers, what would be the benefit of saving? If all of us were to win the lottery, then none of us would actually win anything, would we?
The same important principle applies to tax policy. If everyone is given a tax cut, then none of us actually benefits in terms of purchasing power BECAUSE MARKETS WORK.
How can voters and policy makers make intelligent decisions re: tax policy if they are not constantly aware of this ultimate economic reality?
I submit that voter-consumers would feel differently about tax cuts if economists were to constantly and loudly preach the ultimate futility of Tax Cut Populism.
A couple of points.
The empirical derivation of the peak of the Laffer curve is interesting but ignores how people really behave. We don’t have one rate everyone pays even at the same level of income. There are all kinds of loopholes. As incomes and rates go up, individuals spend more time and effort exploiting the loopholes. Note John Edwards tactic of classifying income as dividends to avoid the Medicare tax.
Second, most people believe that they can spend their money much more effectively and efficiently than the government can. Economists have to assume away this belief in order for the simple macroeconomic models to work. You get a lot more economic growth from private expenditures than government expenditures in the real world. Note Japan’s experience over the last decade.
Scott G–
I think I noted in my comment that it’s a gross simplification to derive the peak of the curve wr2 “the” tax rate, so I certainly agree with your first point. On the other hand, the same basic analysis applies to any given person. We could certainly incorporate multiple MTRs facing a given person and then aggregate over the income distribution to sort out where the peak is for given tax *schedules*. That exercise would just be a more notationally dense version of what I provided above (and corrected at my blog).
On your second point, let me make a couple comments. First, the derivation I provided is very much a microeconomic, not a macroeconomic, one. Second, there is no basis to your claim that “Economists have to assume away” the belief that “most people” have “that they can spend their money much more effectively and efficiently than the government can”. This view is actually a bedrock one among the vast majority of economists, when it is applied to allocation decisions with respect to private goods.
I don’t know where you heard that economists assume away efficiency of private decisions (again, when these decisions concern private rather than public goods, and when there are no market failures involved). Wherever it was, I doubt you heard it from a professor of economics.
You claim that pvt expenditures increase growth more than public ones. This is not my field of expertise within economics, but that statement strikes me as relatively meaningless. You have to know what is being purchased with these expenditures, what the existing stock of public capital looks like, and so on.
For example, consider the productivity of building roads. When they don’t yet really exist, roads are probably one of the most economically wise investments available (hard to trade stuff with people when you can’t bring the stuff to them or from them). Roads are also unlikely to be efficiently provided without govt expenditures due to their nonrival and nonexcludable features (which vary with design, traffic level, and so on).
So the efficiency of public investment in roads will depend a great deal on the existing transportation network, but in many cases it will be quite possible (if not likely) that public expenditures on roads will make a lot more sense than private ones.
Calling Bush’s Bluff on Tax Cuts
Jim Hamilton, who knows a little bit about how to look into such matters, blew a great big hole in that claim. No wonder Bush’s credibility is falling so fast…
Reduce Taxes to Reduce the Deficit?
I have to agree with Mark Thoma that this argument is highly suspect (well Mark simply says suspect, not hihgly suspect). Mark points to this quote from a Bloomberg article, President George W. Bush urged Congress to make permanent the tax cuts enacted…