Nobody serious believes that tax cuts pay for themselves, as I noted last week. But most senior Republicans flunk this test of seriousness.
Mallaby goes on to point out various respectableeconomists (pdf) and their academic research which has pegged the "ROI" (as it were) of assorted tax cuts at between 17%-50%. Note, please, that — while this kind of dynamic scoring is sadly missing from debates about the merits of tax cuts; few Democrats, for example, will acknowledge the increased growth when arguing against tax cuts — none of those figures is 100%, indicating that said tax cuts would "pay for themselves". Sure, tax policy is an oar in the water, but the economy is an aircraft carrier. The response from the Right seems to come in two forms, either (1) "but tax revenues are up!" or (2) "why do you want higher taxes?"
Mallaby’s argument is damaged by the inconvenient fact that revenues have actually increased since Congress and the President cut taxes! Mallaby seems to argue that tax cuts may work in practice, but they don’t hold up as a theory.
This is an intuitively appealing argument: if tax cuts are followed by increased revenue, there's a strong urge to impute some sort of causal connection. The fact that tax cuts undisputably do stimulate the economy makes the urge to see a direct causal relationship that much more appealing.
The problem, however, is that current tax cuts are but one small stimulus in an economic ocean. Sure, tax policy is an oar in the water, but the economy is an aircraft carrier. The tax policy oar contributes, sure, but compared to cutting the marginal rates by a few percentage points, monetary policy and the business cycle are nuclear-powered twin turbines.
How do you disaggregate the effects of tax cuts from the effects of the business cycle, Federal Reserve monetary policy, the recovering foreign economies and other economic effects. If you want to make a serious argument that tax cuts "paid for themselves" or even made the substantial contribution to increased revenues, then you have to come up with some ceteris paribus model — some way of distinguishing the long-term effects of the tax cuts from other economic noise.
Meanwhile, the evidence is that the marginal income tax rates we had in the 1990s do not seriously inhibit economic growth or tax revenue. This is not an argument for higher taxes, though.... Arguing that tax cuts at current rates do not pay for themselves is far from arguing that 70% tax rates are good policy. Occupying the (2) slot is John McIntyre at RealClearPolitics, who argues that Mallaby apparently "pines for the pre-Voodoo Economics days of the 1970's when the highest marginal tax rate was 70% and the country had anemic growth, high unemployment and a Dow languishing below 1,000." This is akin to arguing that those who oppose raising the minimum wage on the grounds that it is ultimately inefficient and harmful must want to go back to the days when workers were paid $.25/hour. Arguing that tax cuts at current rates do not pay for themselves is far from arguing that 70% tax rates are good policy.
McIntyre goes on to argue that generally-lower taxes produce more growth than generally-higher taxes, but he does so in the broadest possible terms... [Supply-siders] seem incapable of acknowledging that there is a left side of the Laffer Curve. At some point, further tax cuts are counter-productive.
Growth produces wealth, which leads to higher tax revenues and a more prosperous nation. Less growth produces less wealth and in turn lowers tax revenues. High tax rates retard economic growth; low tax rates encourage more growth. It really isn't that complicated.
And this is a problem with Supply Siders: they'll happily trot out the Laffer Curve, but they seem incapable of acknowledging that there is a left side of the Laffer Curve. At some point, further tax cuts are counter-productive.
Sure, additional tax cuts from our current level might continue to stimulate additional growth if we had corresponding spending cuts, but we don't. Further tax cuts without corresponding spending cuts are not tax cuts at all; they are merely "temporal tax burden relocations". (we'll pay you tomorrow for spending today!) It's perfectly possible to do this...but not forever.
Moreover, McIntyre compares "nations with high tax rates to countries with low tax rates", but nobody is suggesting an Argentinian tax policy. I don't want to see top marginal rates of 39.6%, but it's pretty apparent that these are not threatening to plunge our country into Argentinian/European-style economic sclerosis. please, make better arguments. All this to say: there are many good arguments for tax cuts and changes to the tax system. And it's worth reminding critics of the dynamic scoring models that indicate a partial "return on investment". It is not, however, necessary to trot out the eternal supply-side argument that tax cuts "pay for themselves". At some point, they just don't.
Argue for spending and/or tax cuts. But, please, make better arguments.
NOTE: Inevitably, a reader will point out that tax revenues have increased since the tax cuts have been implemented, explainthatwhydon'tyou,huh? In order to forestall such simplisme, I will simply note that, as Hale Stewart points out, tax "revenues have always increased during an economic expansion", and then reiterate my suggestion that they show me how they've disaggregated the effects of tax cuts from the effects of monetary policy, the business cycle, etc.
UPDATE (Dale): This is an area where I've also written extensively. Please see here, here, here, and here...among other places.
For a good visual illustration of the points I've made, see Kevin Drum, whose graph demonstrates that "weak recoveries eventually tend to improve, whether there are tax cuts, tax increases, or no tax changes at all." Alternately, see Max Sawicky for hard numbers.
While I do believe that there is a left side of the Laffer Curve, the best argument I heard for the tax cuts of 2001 and 2003 was that Congress had never adequately indexed the progressive tax rates for inflation.
$50,000 a year was a much heftier income in 1990 - warranting are higher rate of tax - than it was in 2000. As a result of this subtle process of inflation, government’s take - the fraction of GDP consumed by Government - had been steadily going up during the Clinton years.
The tax cuts put the government back to something like a fifth of GDP.
Sadly, the other argument that persuaded me in 2001 - that deficits would force Congress to live within its means - gave the pandering simpletons in Washington too much credit.
My response is the standard Libertarian one: Why, in the name of all that is holy, do you want to MAXIMIZE government revenue? That is a game of African dictators - extract the most possible from the populace.
The utilitarian response is that you want to compare the marginal benefit of the government spending the tax funds to the marginal cost of the tax. (which equals tax revenue plus foregone growth due to the tax increase needed to collect said revenue) Obviously, at the top of the curve MC is infinite (since there is no amount of foregone growth sufficient to collect the revenue). The equilibrium point has to be somewhere to the left of peak. (Assumptions: The marginal benefit of spending decreases as quantity spent increases. The marginal cost can be any ’supply curve’ shape you like, within reason, as long as it goes to infinity at the laffer peak. Without supply-side effects, the marginal cost of additional taxes is horizontal at one - that is, any government project with a return greater than $1 in value per dollar in taxes spent should be done. Laffer puts the critical point somewhere above a return of $1 per dollar spent. I cannot speculate how much higher that is. If you are at the Laffer maximum, you have too much government, already, unless your marginal government program also has infinite return in value to match the infinite cost.
The benefit of the government program is increased C+I+(X-M)(roads) plus direct social utility due to spending(cops) plus any value gained through more equitable distribution(welfare) - the unmeasurable, but undeniable truth that taking $1000 from a sufficiently rich person and giving it to a sufficiently poor person very likely increases net social utility. The true cost is the taxes (think in real resources consumed by the spending, if you like) plus the cost implied by "Supply-Side" intuition that taking money away from rich people reduces the motivation to do all of the productive stuff it takes to get rich, thus reducing overall growth, and poor people don’t have taxable resources in sufficient quantity to fund modern government.
The question then is where the left side begins. I’ve always thought it probably existed somewhere around 20% — the portion of GDP government always seems to collect regardless of where the top marginal rate lands. The problem, then, is that a good chunk of Bush’s tax cuts went to lower the 15% bracket to 10% for the first $7K of income, a cut that clearly exists on the left side of the curve.
I think you are underestimating the impact of of tax policy. Just how big is the oar with a flat tax* rate of 100%? Sure, the economy wouldn’t exactly stop even in that extreme condition, but it would do a whole lot of crashing and burning. You are right that it is a complicated topic in that there are multiple inputs and you can’t run the simulation again. It would even be difficult to find different economies, separated by time or space that could be considered equivalent for the sake of comparative analysis. That does not mean that any factors do no have a major impact one way or another. About as close a comparison I can make, and I can only make it so well, is the 91 and 01 recessions. One, whose recovery (2 years post) closely followed by a tax hike and it took almost 3 more years to start to take off. The other whose start was followed almost immediately by a tax cut took less than 3 years if not fully recovered, at least on its way.
BTW, here at least is an argument for the capital gains tax returning net positive.
* - for simplicity’s sake, instead of marginal rate of 100% above some arbitrary number, sufficiently small to achieve the same point
There are a few observable facts here. When a reduction in the marginal tax rate is discussed it is in terms of how much it will cost, but when implemented the reduced marginal tax rate produces significantly more revenue, much more than the general change in the economy. And it hapens every time a high marginal rate is reduced.
What is a tax cut? Lowering some marginal rates actually causes more revenue to come in, so it increases tax receipts. The results contradict the term. This likely does not apply to rates that were low to begin with, such as the 15% marginal rate, but clearly applies to higher rates.
"Further tax cuts without corresponding spending cuts are not tax cuts at all; they are merely "temporal tax burden relocations". (we’ll pay you tomorrow for spending today!)" The marginal rate goes down, actual revenue received goes up. What part of actual results do you find so hard to comprehend?
Tax receipts and business investments have increased twice as fast as GDP growth since tax cuts.
This is hardly a surprise, since the tax cuts were passed between 01-03, during the recession. Similarly, my income shoots WAY up on the 1st and 15th of the money, while my expenses far exceed my income on the other days of the month. Again, the business cycle and monetary policy do much more to explain this than does tax policy. (though I acknowledge that tax policy DOES have an effect)
I think you are underestimating the impact of of tax policy. Just how big is the oar with a flat tax* rate of 100%?
Very big. And should a situation ever obtain, you can be certain that Mr Mallaby, Mr Mankiw, myself and others would be making the argument that it would have a substantial effect. That, however, is not the situation that currently obtains.
I think you may be confused about the ’91 and ’01 recessions, though. By almost every measure, the ’91 recession was shallower than the ’01 recession. Granted, that’s largely due to exogenous circumstances (geopolitical stuff, etc), but the current recovery was fairly slow to arrive. And, again, there are vastly different situations within each recovery, too. It’s hard to pin the credit/blame on "tax policy".
When a reduction in the marginal tax rate is discussed it is in terms of how much it will cost, but when implemented the reduced marginal tax rate produces significantly more revenue, much more than the general change in the economy. And it hapens every time a high marginal rate is reduced.
You know when tax revenues go up? Almost every d*mned year. This is like arguing that coin tosses cause touchdowns. (after all, many coin tosses are eventually followed by touchdowns!)
I was reacting to your blanket, "tax policy is an oar in the water" remark. You didn’t say, "recent changes in the tax policy is an oar in the water" or anything close to it. Just being pedantic. I don’t have time right now to find the numbers to back this up. As I remember, the ’91 (actually I think it started in ’90 and basically over by ’91) wasn’t that bad but it did help elect Clinton (because it hadn’t recovered enough for people to notice), it did ok with Clinton, then hit a rough patch, which helped the Repub revolution and didn’t really take off as we know it until about ’95. The ’01 recession was also rather shallow, but was considered a jobless recovery. The ’lack of jobs’ is hard to qualify since it is compared to historically low unemployment at the end of the bubble. The GDP numbers were pretty good ’03 and beyond and by the second inaugural, we were basically job loss neutral.
The only simplistic notion I saw was to say "tax revenues always increase in an expansion" To which I reply, how do you stimulate an expansion, with monetary policy AND tax cuts, and how do you smash an expansion? Through monetary policy AND tax increases. It also disingenuous to suggest that supply siders never mention that there is a left side of the laffer curve. Of course we do, I have debated where it is many times. The point of equilibrium goes up and down with confidence in the economy but its probably at a marginal rate of about one third of your income. So, we are probably still on the right side of the curve.
You’re right on the money, so to speak. The Laffer curve is a legitimate model — an identical one is used to explain the optimal amount of seigniorage in an economy — but it’s generally used as a marketing tool for tax cuts.
I would like to see some respectable research that establishes that we are in fact on the right side of the Laffer curve. I don’t think that’s the case, or we would otherwise see the tax cuts paying for themselves (an ROI of 100%).
There are other arguments for tax cuts, and I supported Bush’s tax cuts when they were made, but I anticipated that spending would slow and that hasn’t happened.
In the very long run, the tax cuts may indeed prove beneficial by reducing the size of government — we have a round of tax cuts about every twenty years and it seems to put a ceiling on government spending.
However, the "supply siders" (that’s not a real school of economic thought) over-do things by making exaggerated claims about the ability of tax cuts to pay for themselves.
People always overestimate the effect of tax rates on the behavior of individuals.
First, a lot of people work hard enough to earn $X, and won’t work any harder once they reach that point. For those people, cutting tax rates actually causes them to work less. There are many reasons for this, including becoming accumstomed to a certain lifestyle or having a certain level of minimum monthly expenses.
Second, the instances of tax avoidance (other than out-and-out fraud) amongst the high income and/or high net worth individuals comes at very unique times. The estate tax, for example, causes people to do all sorts of ridiculous things, solely to reduce their taxes. This, IMHO, has no impact on economic growth. Other instances generally involve people selling a business they spent 10, 20, or 30 years building themselves and are now selling implicating a huge capital gains tax, or are executives selling large blocks of stock from their options. I don’t think you see a lot of CFOs or CEOs at large companies slacking off because of the marginal tax rates. As Jon notes in relation to the economy, there are other things that motivate people that swamp the effect of marginal tax rates.
We could increase the marginal income tax rate to 60%, and the impact on economic growth would be negligible.
Jon this whole thing seems a part of the "We’re P.O’d at teh Republicans for Spending So much" theme here at QandO. So just like Democrats advancing ANY argument to advance Universal Health Care, you seem to be advancing ANY argument that says "Things aren’t good, rein in spending"... just a thought.
In response to the overall point: Yes, there’s no specific correlation between tax cuts and increased revenue. Just like there’s no specific correlation between "getting more sleep and increasing fluids" and getting over a cold. Yet it’s still a good idea, and getting more sleep and increasing fluids is probably the most effect you can have on an amazingly complex system of recovering from an illness.
Methinks the economist doth protest too mightily.
I’m now more convinced tax cuts are a good idea based on the Laffer Curve than before you started. Willful ignorance, perhaps.
I understand THAT McQ... it’s just now we’re tarting it up in theoretical discussions of tax policy and the Laffer curve, as Jon argues that THE ECONOMY AND TAXES WOULD HAVE GROWN ANY WAY... what we need to do is control spending, because WE’RE ON THE OTHER SIDE OF THE LAFFER CURVE! Mayhap we are, mayhap we’re not, but it’s IRRELEVANT to the point that spending is too high. As I posited, this is the same as and just as disingenuous as arguing it is inefficient for businesses to provide medical coverage, so voila let’s have Universal Health Care.
Jon this whole thing seems a part of the "We’re P.O’d at teh Republicans for Spending So much" theme here at QandO.
Does it? Well, duros to you for noticing that we’re PO’d at Republicans for spending so much. We both appear to be operating well.
However, my above appeal for better arguments stands on its own, regardless of the Republicans intransigence on spending. Poor arguments may sometimes make good demagoguery, but they still make, you know, poor arguments. As I pointed out in the post, there are still some good reasons to favor tax cuts. However, as I also pointed out, without corresponding spending cuts, the good rationales for tax cuts diminish rapidly.
Yes, there’s no specific correlation between tax cuts and increased revenue.
If that’s what you got out of this post, then you have completely failed to comprehend it. Worse, you seem to have gotten it exactly wrong.
The point of equilibrium goes up and down with confidence in the economy but its probably at a marginal rate of about one third of your income. So, we are probably still on the right side of the curve.
I’m going to go out on a limb and suggest that you have no idea at all where the inflection point on the Laffer Curve is. You’re just making up a number to the left of the current position out of nearly thin air.
If you don’t have growth because you exacerbated the recession by not cutting taxes then where are you now? And tax cuts need to be permanenet to stimulate spending rather than one-off savings.
That’s an aggregate demand argument, not a Laffer curve argument. And, yes, they must be permanent to permanently alter consumption patterns. However, to get out of a recession, you only need to stimulate AD temporarily.
I’m going to go out on a limb and suggest that you have no idea at all where the inflection point on the Laffer Curve is. You’re just making up a number to the left of the current position out of nearly thin air.
You are correct I don’t KNOW where it is but you are incorrect in assuming its not an educated guess. It is based on historical research. When New York state and California increased their income taxes back in the nineties, even though the economy was booming in the nation, they began to see a drop in business investment and in the rate of new hires and other leading indicators. the increases happened in both cases to put the effective marginal rate for the investor class ($250,000+/year) above 35%. My guess is based on that. Of course other factors are in play, what is the capital gains rate? are there good investment alternatives in other states? My belief, and I am currently gathering data, is that effective marginal rates much above 33-35% start to cause investments to shift elsewhere. Probably this fluctuates with market conditions. (notice, when I say effective rates, we currently have a national tax rate which goes up to 39% but that is not the rate most people actually pay, due to deductions/incentives etc. the effective rate is more like 33% right now.) source: http://economistsview.typepad.com/economistsview/2005/11/effective_margi.html source:http://econ.ucsd.edu/~rogordon/growth.218.doc
I am just beginning this research, and do not want to sound like I know the answers, but there are people who are studying the Laffer curve. There is a lot of disagreement on just what is the Effective marginal rate, so it could be a few points higher, but it wouldn’t be much. When we say the Nation has a top marginal rate of 39% you have to add a weighted average of all state income taxes so remember its higher than the stated amount.
My point on the tax cuts was not an argument based on the Laffer Curve but to point out that they might have been a good idea even if you don’t believe in Laffer.
According to the Skeptical Optimist guy, as long as your growth of GDP exceeds your growth of revenue as % of GDP, then you will be doing fine in eliminating budget deficits this way. He’s predicting April 2008 as when we will be "back in black" based on continuing growth etc.
Does the fact that revenues actually increased more after the early 90’s tax increase than they are now prove anything?
And, there’s no need to point out that there’s a left side of the Laffer curve. Simply take supply side wacko arguments to their logical conclusion and explain that a zero tax rate must=Infinite revenues.
When they get mad at you for throwing logic in tehir faces, laugh at them and vote for the Democrats.
This is quite a good exchange. I haven’t followed down the clicks; pardon my redundancy.
The best tax rate may depend on which tax you are talking about. Many economists think the correct tax rate on capital gains is zero. (Robert Lucas, "Supply Side Economics: An Analytical Review" first paragraph, for example.) A very recent paper by JR Reyling ("Dynamic Scoring") sets a "speed limit" on capital gains of 15% based on Laffer type analysis. Reyling’s "speed limit" on labor income is 34%. Of course pols, like car drivers, may treat posted limits more as minimums than maximums!
There are a couple of ways to look favorably at Laffer tax cuts. There is the possibility that taxes be so high that a judiciously chosen cut will immediately increase revenue. To judge by the academics the tax rate to be lowered would have to be like 70 or 80 percent or more.
The other way is to consider the "present value" of tax collections. Then if the increase in the economy and future tax collections was great enough to cover the cost of any immediate deficits, then you’d still be golden.
Academic work on the Laffer recognizes that there are two sides of the curve. All of the academics I’ve read think the tax rate should be below the level that maximizes government revenue.
Jude Wanniski ("The Way the World Works", page 98) thought the maximum was optimal but I think his argument is just plain wrong.
The Heritage Foundation’s Economy Watch has denied that tax cuts will pay for themselves except in unusual circumstances. -They advocate budget cuts along with tax cuts.
Note also that Bush only claimed that the deficits would be cut in *half* after 5 years (by 2008, I think). I don’t remember his mentioning the Laffer curve at all.
I’m going to go out on a limb and suggest that you have no idea at all where the inflection point on the Laffer Curve
Ha. I’m going to stay right here next to the trunk and point out that nobody has any idea where the inflection point is. Meteorologists, statistical ecologists, chaos theorists, and just about anyone who builds mathematical models of complex natural systems for purposes other than fleecing naive investors out of their life savings will be happy to explain why it is, literally (which is to say mathematically), not possible to be certain where the inflection point of the Laffer curve is located at any given time.
Then again, perhaps kyle is on the verge of turning the world of nonlinear dynamics upside down. For mere mortals however, the accuracy with which we can estimate the location of the (hypothetical) inflection point is coupled, as Jon points out, to the accuracy with which we can actually distinguish between the influence of the tax rate and the other variables influencing overall revenue. Which is something we just about can’t hardly do, really. Which means that we have practically no idea where along this strictly hypothetical (and quite dynamic — don’t forget that it changes in response to all those other variables) curve we are located at any given time. Which means that it’s somewhat interesting academically, but about as useful to policymakers as string theory is to an architect.
And, unfortunately, the rest and fluids argument doesn’t help, because to the extent that rest and fluids are considered effective (this is not actually uncontroversial) it is because they have been observed to be effective. Observed, y’all. Not hypothesized. Not "obvious to anyone with a lick of common sense," which seems to be the main argument for the practical usefulness of the Laffer curve. Observed.
Call yourselves dismal if you want, but don’t call yourselves scientists if you’re not going to do science.