Jude Wanniski, a writer for the Wall Street Journal, coined the term “Laffer Curve” after a concept promoted by economist Art Laffer. Laffer himself says the idea goes back to the 14th century
The idea is that if one wants to maximize the government’s tax revenue, there is an optimal tax rate. (Ignore for the moment whether or not you think this makes good economics in the long run, or whether or not you think this is even moral.)
Laffer noted that if the tax rate is zero, then the government gets no revenue. But likewise, if the rate is set at 100%, the government also gets no tax revenue. Mainstreamers say that there is no incentive to produce income at 100% tax rate, and this is true. But even more importantly, there is no means: a 100% tax rate is pure capital destruction.
The “Laffer Maxima”, i.e. the tax rate which maximizes the tax take, is somewhere between 0% and 100%. The Wikipedia article shows a picture of a Laffer Maxima at 70%, and implies that although it’s somewhat controversial this may be the right number.
There are two points about the Laffer Curve that are important to consider.
First, what in the world makes any economist think that he can gin up some differential equations and compute the right value for this Maxima? In the first place, every market is composed of an integer number of people transacting an integer number of trades, and each of those trades consists of an integer number of goods. People do not behave like particles in an ideal gas—they have reason and volition. The very idea of modeling a large number of people with equations is preposterous. Never mind that degrees are awarded every year to economists who purportedly do just that.
Second, what makes anyone think that the Laffer Maxima is a constant?
Let’s do a thought experiment that is in the vein of the Austrian School of economics. Let’s consider the boom-bust cycle, or what Austrians note is really the credit cycle. The central bank first expands credit, which flows into wealth-creating as well as wealth-destroying activities (malinvestment). As the expansion ages, an even greater proportion of credit funds wealth-destroying activities. Sooner or later the boom turns to bust. Malinvestments are liquidated, people are laid off from their jobs, portfolios take big losses, tax revenues decline, etc.
One clue can be found right there, in my description of the bust: tax revenues decline.
OK, maybe the Laffer Curve remains static and the only thing that changes is the absolute tax dollars?
Let’s continue comparing the boom and the bust phases. In the boom phase what’s happening is that economic activity is being stimulated, i.e. beyond what it would naturally have been. This fuels demand for everything: commodities, labor, construction, fuel, professional services, etc. And all of the people hired in the boom are demanding everything too. It feeds on itself synergistically, for a while.
At this stage, the frictional cost of taxes may be masked by the lubricant and fuel of credit expansion. This is especially so when everyone feels richer and richer on paper. People spend freely and we saw this in spades in the most recent boom that ended in 2007.
Now let’s look at the bust phase. The net worth of most people is falling sharply. Many are laid off, their careers, and sometimes lives, shattered. A huge component of the marginal bid for everything is withdrawn. People struggle to make ends meet. Budgets are stretched to the max.
I submit for the consideration of the reader that in the bust phase, any change in the tax rate drives a big change at the margin of economic activity. The tax rate is more significant in the bust phase than it was in the boom phase. The Laffer Maxima is not a hard-wired, intrinsic value of 70 (or 42 for fans of Douglas Adams). Like everything else in the market, it moves around. It is subject to the forces of the markets.
I will close with an example. Consider the marginal restaurant. Let’s say it is generating $25,000 per month in gross revenues. Net of $24,700 in expenses, it is generating positive cash flow of $300 per month. Why would the owner even keep it open? Well, times may get better…
Now, let’s say the tax rate goes up a little, say 100 basis points. The restaurant, making little money, pays essentially no taxes anyway. So this does not cause a direct impact. But what about the patrons of the restaurant? If their blended tax rate was 25%, then an increase of 100 basis points (i.e., to 26%) is a tax increase of 4%. These people will have to reduce their budget by 4%.
One logical place to cut is eating out. Suppose that they reduce their spending in the restaurant by $1,000, in aggregate. Now our restaurant has $24,000 per month in gross revenues. But its fixed costs cannot be reduced. And even the labor can’t be reduced in this case. The only reduction will be food supplies. So let’s say food supplies are reduced 1/3 of $1,000, or $333. So now the restaurant has expenses of $24,367. Whereas it formerly made $300 profit per month, now it makes a loss of $367 per month.
The owner can’t continue this very long. And so he closes shop. He defaults on the loans on the fixtures and tenant improvements, lays off 8 people, leaves the electric and gas companies with fixed infrastructure which no longer produces revenue for them, etc.
The impact to the economy (and hence to the total taxes collected) is negative and disproportionate to the tax increase.
© Dec 29, 2011 Keith Weiner
What you’re saying (taxes shouldn’t be increased because of impact to marginal economic activity) sounds good in theory but totally fails in practice because of the unequal distribution of wealth (or spending power, if you wish) among the market participants.
When you have massive income inequality such as today’s America, marginal tax rate increases on the wealthy will have very little effect on their spending activity. But a responsible government can use the increased revenues to either pay down the debt or use it for public sector services – which will help the poor and the middle class.
Increasing taxes on the wealthy do not hurt growth. That is a myth which can be easily debunked with data.
Bush-era Tax cuts are the single most important culprit in blowing up the Federal deficit. This can be proved with data as well.
Real wages are falling and this especially hurts the low and middle income households. So anybody who thinks real wage inflation will not solve the debt burden of households is married to ideology, not pragmatism.
http://jaredbernsteinblog.com/the-best-of-cbpp-graphics/
A wise man once said “Greed is the sin of capitalism and envy is the vice of socialism”. What we really need is a balance between the two, because either extreme has shown to hurt nations and economies. America is a perfect example of extreme greed gone wild.
*dang it*, my facts and logic have to be wrong, because it just isn’t *FAIR*!
You facts are logic are ideologically correct, but fail utterly in real world.
I have shown data to debunk your assertion that raising taxes will impact marginal economic activity . It won’t and you can’t disprove through ideology. Prove it practically!
Instead of arguing my point on a practical basis, you are stuck in a Randian fantasy.
Jesse, in my paper I make a few points:
1) the Laffer Maxima moves
2) the value depends on where the economy is in the boom-bust cycle
3) my example was about a tax increase on the patrons of a marginal restaurant
Your response is twofold:
1) trying to refocus this debate onto “the rich”
2) accusing me of ideological blindness to facts
I will stop here to note the irony.
Keith,
Are you going to confine your analysis only to the “marginal restaurant” or are you planning to generalize it to the whole economy?
I thought you are alluding that a tax increase on the whole economy is going to have a negative impact, using the marginal restaurant example. If you are, then that is practically wrong based on the data I present in my first comment.
I am not refocusing the debate onto anything, I was trying to get to the bottomline of your argument on the whole economy.
Where do you stand on the question I ask in this comment?
Jesse: You have presented no such thing. A quick look showed several different things:
1) some material about the so-called “1%”
2) data that assumes ceteris paribus, when the very point of my piece was that this is not true–the rising side of the boom is quite different from the falling side of the bust
3) equates consumption and investment
It is one thing to show that billionaires do not eat less, or drive less, or even fly planes less when the tax rate rises. That’s probably true. And this would prove your case if the only thing that mattered was consumption. But it is quite another to show that they don’t invest less under higher taxes. Today’s investment is necessary for tomorrow’s economic activity, much less growth.
Keith – you haven’t answered my question. Are you generalizing your analysis to the whole economy? The whole economy is much more complex than a “marginal restaurant” and this should be rather obvious.
It is a dogmatic assumption to say that higher taxes always cause lesser investment. We have to quantify how high is too high. If higher taxes causes lesser investment, why did GDP and unemployment actually get better under Clinton era tax rates?
Jesse_Fan,
even Clinton insiders would admit that the strong economy and fiscal budgets of his Presidency were based on curtailment of gov’t spending; and moreso, the tech boom, which brought both massive productivity growth as well as investment spending that earned high returns. there was more to tax, and then it all blew up in 2000 when Bill was on his way out.
i’m not arguing for or against increasing tax rates as i don’t think leaving them where they are now or letting the Bush tax cuts expire will have a material impact on revenues. maybe some in the short-term but that won’t last long. at widely varying tax rates for the top bracket since 1950, we always go back to tax revenue of 18% of GDP, and the band is pretty tight.
Jesse_Fan:
you say the Bush tax cuts were the biggest contributor to our Federal deficit issues and that this can be proven. please do so — i do not think you can.
please don’t cite keynesians like jared bernstein in your proof.
Why are Keynesians considered always false? Is it because of your dogmatic inclination towards Austrian?
You see, Keynesians are sometimes correct, and you’d be better off paying a bit of attention to them. Yes, they are massively wrong on several fundamental issues, but doesn’t mean that they’re always wrong. It is dangerous to be dogmatic, it is wise to be pragmatic.
I’ll cite an example where a full-bred Keynesian is actually correct:
http://krugman.blogs.nytimes.com/2011/09/06/treasuries-tips-and-gold-wonkish/
We went to a war that had full Congressional approval and at the same time (which was a war fought for totally different reasons than originally stated), we cut taxes on the wealthy. If as a nation, we decide to spend more — we also have to agree to be taxed more. Because national spending rightfully comes through taxation. Note: I said rightfully.
If you’re of the opinion that taxation itself is immoral, then I have nothing to argue with you as you wouldn’t even need a government in the first place.
i simply asked you to prove your point, you did not and that is because you cannot, and you cannot because what you said is arithmetically incorrect.
i am not dogmatic. you are now changing your tune, you never said a thing about wars (or other spending), you said it was tax cuts most responsible for deficits, which is simply not accurate.
i am not of the opinion that taxation is immoral. it is not ideal but not immoral in my opinion. certain services are done best by gov’t, they are few and far between but they do exist, and they must be paid for.
krugman, ha. you said one thing i agree with, “i have nothing to argue with you …”
If I may, I have a little clarification about differential equations. “Solving” an ordinary differential equation (ODE) can mean two different things. You can have a “general” solution to the ODE, which tells you the type of solution e.g. the type of curve. To get an exact solution from the general solution, you have to supply “boundary conditions”, additional information.
For example: a guitar string can be modelled by a 2nd-order differential equation, and the general solution to that is a periodic sine wave i.e. it has a resonant frequency. To actually plot the vibration, you need to add boundary conditions such as a linear displacement (hit the string), the properties of the string, and so on.
I am not an economist and haven’t studied the theory behind the Laffer curve, but if it is the result of a differential equation, or more than one, the same principle tells me that the differential equation(s) alone will tell you the shape of the curve, but not where the maxima would be. That would require a good selection of boundary conditions, and they could vary. So, not only would I be sceptical of any suggestion that there is a constant Laffer maxima (your second point), I would also expect an economist to take great care in the selection of boundary conditions. Garbage In -> Garbage Out, and all that … ;-)
brian: thanks for your comment. I read you as saying that working with differential equations is part Art as well as Science, and necessarily includes some judgement and a view of how things are which is obtained outside the mathematics itself.
As I understand it, Laffer did not come up with the idea based on differential equations (nor did the 14th century Arab who Laffer credits with the idea), but as a thought experiment.
The best academic research on this issue would be: “The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks” by Christina D. Romer and David H. Romer http://www.econ.berkeley.edu/~cromer/RomerDraft307.pdf
The paper was published in June 2010, while Christine Romer was Chgairman of the Counsil of Economic Advisors to President Obama. Quoting from the report: “This paper investigates the causes and consequences of changes in the level of taxation in the postwar United States. Our results indicate that tax changes have very large effects on output. Our baseline specification implies that an exogenous tax increase of one percent of GDP lowers real GDP by almost three percent… and that [capital] investment falls sharply in response to exogenous tax increases.”
This was highly embarrassing to the Obama Administration, because it essentially stated that for every percentage increase of government as a percentage of GDP, there is approximately a three percent reduction in total GDP.
As Chairman of the Council of Economic Advisors, her job was to provide the President with objective analysis and advice on a broad range of economic matters. The CEA is generally considered the most influential economic team in the country because it directs the economic policy of the White House and the President. Obviously, Barack Obama had no interest in objective advice regarding government crowding out the private sector.
Chriss: Thanks for your comment. You raise a topic I didn’t touch, because my piece was already getting lengthy.
And of course, depending on what the government spends on, this could further subtract from GDP. Let’s say they hire armies of regulators and litigators to go after productive companies, the destruction of GDP could be even greater.
David Romer – Ugh
I picked up this little gem two days ago.
http://www.marketwatch.com/story/emotion-not-economics-marked-stimulus-debate-2011-12-28
Mr. Romer thinks WE are just too dumb to get it, we need massive Keynesian Stimulus, its for our own ignorant good.
Keith nice piece. Taxes are life blood of the Nanny State, Mr. Laffer loves so much. The State (historically) taxes to implosion and then the cycle starts again, it never has been able to control its appetite or find Mr. Laffer’s equilibrium.
The State is an alcoholic, never satisfied and always wanting more, the benevolence is to starve the beast of its drink.
Thanks Californio. In a future piece, I want to delve into a more radical idea: *EVERY* thing is in disequilibrium. There is no such thing as equilibrium! :)
When you draw a laffer curve, just like when you draw a supply and demand curve, you assume “ceteris paribus”: the curve can be viewed as a snapshot and a snapshot in the next second will give you a different picture: so there is no optimal tax rate that you can measure, since measuring it is impossible.
Hi Keith,
this may seem a bit critical, but please, I am not trying to demean your efforts, I am wanting to highlight a debilitating cultural problem in economics. What I would like to point out, is that the article contains a series of intellectual/philisophical arguments on the Laffer curve. This is the history of economics, intellectual/philisophical discussions that are not connected to reality and are devoid of any significant data. Any side of a debate can usually make an apparently relevant “thought experiment” to back up their case. This makes “thought experiments” an interesting intellectual entertainment and requires them to be immediately followed up by high quality data/analysis.
I come from a physics backgound, in that field, the currency is good quality data, analysis and argument. This has led to physics advancing enormously in the last 150 years. Compare it to macro-economics, which is still travelling in the same tired old circles.
So fancy arguments about the Laffer curve are only for the already converted or the intellectually challenged. To make real progress on the Laffer curve you need real studies with decent data. The paper mentioned by your contributor, Chriss Street, is a good starting point, as they try doing an analysis on real data. The next step is these sort of papers need a thorough review to determine if the techniques used and the results obtained are valid. I would also say that it is probably not the final answer. Take a lesson from physics, in that a single paper rarely settles an issue.
To re-iterate, to make real headway on the Laffer curve, we need real data and data from many different scenario’s. As mentioned by some of the contributors the peak position and shape of the curve will probably be very dependent on the situation.
Keith I would also point out that it is quite possible that some characteristics of large numbers of people can be modelled by equations. Maths is amazingly versatile. As economists are generally the dummies of the mathematical world, the fact that they have had little success in this field, has everything to do with their poor mathematical/analysis abilities.
Again Keith, this is not a personal attack. It is a comment on the culture of macro-economics and that it needs to lift its game.
Your restaurant example basically states that in the event of a tax increase during a bust at all remain constant except for the consumer who will look to reduce expenses by not patronizing the restaurant and thus causing the adverse ripple effect. The smart restaurant during this period will also look at off-sets… Renegotiating with suppliers, reducing overtime or overhead etc… This is where your argument is flawed you are simply attempting to prove a theory that is inherently bias