Corporate Tax Rates and Unemployment
by Mike Kimel
Corporate Tax Rates and Unemployment
Cross posted at the Presimetrics blog.
Update: Reader JzB has noted that a large drop in the top marginal rate that appears in graph two may not have occurred. At this point I think I made a transcription error in pulling the data. I will check the numbers when I get a chance and redo the post. Apologies.
I’ve been kind of swamped, low on sleep, and doing a few book related things in my few waking hours that don’t work or parenting (buy a copy of Presimetrics!!!!), so posting has been light. But I thought I’d do a quick and dirty post today about a hot topic – the effect that taxes on businesses have on unemployment. The usual argument is that the lower the taxes on businesses, the more money they keep and pump back into, well, doing business, and thus, the more people they end up hiring. But is it true?
Now, since the talk right now is about cutting payroll taxes in particular, ideally I’d use that data. However, in a quick perusal at the IRS’ site, all I found was the corporate marginal tax rate. However, the folks who suggest tax cuts as a way to boost hiring aren’t particular – most of them feel any tax cuts will lead to more hiring. So let’s check that, at least, shall we?
Figure 1 below shows the data used in this post; the top corporate marginal tax rate (obtained from the IRS) is on one axis and the unemployment rate for individuals sixteen years and over (from the Bureau of Labor Statistics) is on the other axis. The latter series begins in 1947, but I decided to start with 1948 just to be far enough off from WW2 to avoid that effect as much as possible.
Now, consider the correlation between the top corporate marginal tax rate and the unemployment rate. If it is true that lower taxes = lower unemployment, the correlation between the two series should be positive. A positive correlation means the series should move more or less in the same direction; as tax rates rise, unemployment rises, and as tax rates fall, unemployment falls.
If the correlation is, in fact, negative, that means that lower unemployment tends to happen when tax rates are higher. Correlation may not be causation, but it would be very hard to argue that cutting taxes on corporations leads to lower unemployment if we do not see a positive correlation between the two series.
Now, obviously, it may take time for tax rates to do whatever magic they might have. So Figure 2 looks at the correlation between the top corporate marginal tax rate and the unemployment rate in the same year, the unemployment rate the next year, etc., all the way through ten years out. Its really hard to see how the effect of tax rates should last beyond a couple of years, but I figured I’d be thorough and put up the figures. I’ll take a pass at interpreting them, but feel free to reach your own conclusions.
Additionally, because whatever effect tax rates might have on unemployment might change over time, each correlation is computed several times: once for the entire 1948 – 2009 sample, a second time for 1960 – 2009, a third time for the period since 1970, a fourth for the period from 1980 and a fifth time for the 1990 – 2009 period. (I didn’t look at just post-2000 because the top corporate rate has been frozen during that period.)
So what does this say? Here’s my interpretation. If you look at the entire data set, the correlation between tax rates and unemployment appears to negative. That is to say, lower unemployment rates tend to be associated with higher corporate tax rates, not lower ones. Fast forward to the 1970s, and you see some of the oft-stated effect; higher taxes do seem to be associated with lower unemployment, and that effect becomes even stronger when you focus on the 1980s and beyond. However, the whole thing falls apart when you move into the 1990s and beyond, when the correlation is clearly negative, at least for the early years. The effect in out years is positive, but once again, its hard to see how the effect would lag that long. If I had to guess, the 1970s and 1980s were an aberration but I don’t have time to develop that thought right now.
My second interpretation… if I wasn’t some random guy who enjoys looking at data in an attempt to understand the world, but rather someone determined to reach a certain pre-determined conclusion, I’d pick my time sample very, very carefully.
How do you see it?
I think the 70’s and 80’s are pre Greenspan, that is times before the Fed Chm went Reagan voodoo economics.
90’s-oughties are fed run bubble economy.
From a finance point of view tax policy affects cash decisions, somewhat.
I think demand, mostly bubble related demand worked employment far more than taxes.
In short Laffer is not a theory, but a cocktail party game run at very high end parties to help a few enjoy all the fruits of bubbles and not pay their way.
My perspective is different. Corporations will hire people in the locations that make the most economic sense for them. For many of them that means Asia … but not because of “cheap labor”. The labor cost are raising in Asia and will eventually match those in the US. However, a large proportion of Asian countries do not have corporate income taxes. For that reason overseas operations will remain popular by allowing corporations to retain their profits. The US looses out twice. The jobs go overseas and the corporate profits remain overseas waiting to be invested (likely overseas).
Ben you are arguing that the big cost difference will go away some day (though we don’t know when) so it doesn’t matter, while the smaller cost difference, which could go away at any time because of legislation, does matter. Doesn’t add up.
Also doesn’t match well with observed behavior. As Chinese wages have risen, China has lost low-end manufacturing to Vietnam, where wages are lower. That followed the shift of factory jobs from high-wage G8 countries to China, Eastern Europe and elsewhere. Some G8 countries have taxes on corporate profits, others do not.
There is more going on that just wages or just taxes, but the argument that jobs move because of taxes rather than because of wages doesn’t fit the facts.
When taxes are higher, the incentive is to spend the dollars in growing the business, which provides lower current income, but higher income in the future. When taxes are lower, the incentive is to maximize profits by cutting employees and distributing as much as possible to owners and top management.
There are a number of problems with doing analysis on corporate taxes and unemployment. One of the big ones is that tax changes are sometimes made in response to economic cycles. To the extent that those changes are aimed at leaning against the wind, tax changes and the jobless rate will have a more negative correlation. In an effort to establish causation, policy efforts equate to spurious correlation.
It would be interesting to see a cyclically adjusted jobless rate against tax rates. (I don’t know of any cyclically adjusted jobless rate series, but I wouldn’t be surprised if one had been created by some academic somewhere.) The reason for doing this sort of comparison is obvious, yes?
As I have confessed many times I am not a numbers guy, but it seems to kme that taxes could affect employment 3 ways: First, there may be a tendency to hire people and produce goods or services in a lower tax jurisdiction. This is the fovorite argument of the GOP in my state of Wisconsin which the GOP frequently describes as a tax hell. In fact while Wisconsin’s tax burden is often in the top 10, the biggest motivation for business to move out of Wisconsin is lower wages in non union states. Second, certain taxes–like the payroll taxes–could make it unprofitable to add another worker. This is not true of marginal tax rates because by definition those taxes are only paid on profits If I can make just $1 by hiring a new employee, then in a riskless world I will do so whether the government takes 20 cents of that dollar or 80 cents of that dollar will make no difference. However because of the payroll taxes, where I might have made a dollar by hiring a new worker, I now lose $10 so I will not hire. Finally, the astute reader will note my reference to a riskless world, but of course we do not live or hire in a riskless world. There is likely some profit above a dollar that I have to achieve before I will assume the risk that comes with expanding production by hiring another worker. If that is $100 then marginal tax rates can make a difference. At a 50% marginal rate, I would have to anticipate that extra worker will bring in $200 of profit before I will undertake the risk. At 25% I need only expect an extra $135 in profit.
i think there ia a flaw in this analysis. you are assuming a 2 factor model (taxes and unemployment) which is too simple to explain the reality. i suspect that to a great extent you are mistaking correlation with causality and that you are ignoring the fact that effects lag.
if during times of economic duress, unemployment tends to go up and governements tend to cut taxes to “stimulate” then you get a negative correlation that has nothing to do with causality. what you need to be considering are inflection points, not absolute levels.
look at the late 70’s on your corporate chart. corp taxes were cut, and a year later, unemployment began to decline. then they were raised, and a year later unemployment headed back up. the cut in the 80’s came during a decline in unemployment already in progress, whic, after a quick blip, continued for a decade.
by looking only at absolute levels instead of inflection and trend, you are essentially arguing that turing up the thermostat causes the room to be cold because it always happens when the room is coldest.
Read the part of the initial post in which cactus talks about lags. All the way through ten years. No room there for “ignoring the fact that effects lags”. All kinds of lags considered.
his “lags” are not lags from changes, they are just decadal/multidecadal which is a meaningless metric. they are mostly comparing flat tax rates with fluctuating unemployment interspersed by rate changes at random times in the decade/period which alter the curves. that methodology is completely wrong. it’s not even laid out by individual decade so you can see what it did.
lag is “start at the change, then see what happened”.
look at the 2 changes in the 70’s.
can you possibly argue that in the 12 months following you didn’t see a change in the trend of unemployment both times?
in the 80’s, the cut was telegraphed well in advance, so it is perhaps undertstandable that unemployment led the change (though it kept dropping afterwards). alternately, perhaps that was just noise, but look at the periods after the changes in the last 40 years and the correlation is actually quite strong.
it’s the changes you want to watch, not absoulte levels. ceteris paribus, an economy will adjust to the changes in 2-3 years. then the correlation goes away, so longer periods will now show rate of change differentials.
his “lags” show nothing at all.
“let me recommend to you that you give up the chest-thumping “I’m right and his numbers don’t mean anything” approach.”
Oh you mean……just like you do?
“We are likely to learn a lot more about how the data shape up, before this is over.”
Then how do you know he is wrong?
While we’re at it, I should mention that one of the issues that has been pretty well covered since Mr. Laffer first put ink to cocktail napkin is interplay between economic activity and taxes. One of thing that has become clear from looking into the data, and that is pretty obvious just from thinking about it, is that changes in tax rates don’t matter much when rates are fairly low. They just don’t have much impact on the private outcome. Changes when rates are high have a very strong impact on private outcomes when they are high. So el tigre’s assertion that Mike has erred by not focusing exclusively on change disagrees with what we know about the impact of taxes on private activity.
How could I know that specific points in el tigre’s complaint are wrong without knowing the outcome of Mike’s future analysis? Wow. Thanks for making this easy. In fact, Jimi, I think even you can handle this one. Let me hand you a couple of analogies, and you give it a try. See if you can manage.
Mike is making a pie. Striped Kitty comes into the kitchen and says, “You can’t make good pie. It’s November!”
Mike is playing chess. Stiped Kitty watches Mike take the other guy’s queen with Mike’s bishop and says “You’re gonna lose. You always lose if you take a queen with a bishop.”
(Hint, in case you are having trouble: I don’t know if Mike’s pie is gonna be any good, and I don’t know if Mike’s gonna win the chess gain, but I know that Striped Kitty is wrong. Not that hard when you think it through.)
your assertations, while vigorous, don’t actually answer my fundamental issue with the analysis:
look at the 2 major tax rate chages in the 70’s. now look at unemployment is the flowing 36 months. in both cases, you had change in trend followed by considerable follow through. in both cases, there was strong positive correlation.
now look at the change in the 80’s. trend did not change, bcause it was already headed down, but again, unemployment dropped as taxes did.
this is strong positive correlation. it’s obvious visually and statistically.
if you have sufficiently foulded up the “lag” calculation that this does not show, then clearly, the math is wrong.
look at those 3 big tax changes. that is our set to study.
can you possibly argue that the 36 month lag in the 70’s is not extremely positively correlated?
if so, please lay out your argument. i’d like to see how you get there.
the 80’s is trickier because the trend did not change, but the effect was still positively correlated from what we can see.
alternately, look at the unemployment chart. it was in an uptrend from 1948-early 80s’ (a high tax period) it’s biggest drop was during a breif period of low taxes.
then from the mid 80’s to 2007 it was in a dowtrend while taxes were at a lower absolute rate.
this recession broke that trend, but was not caused by taxes.
so again, how do you get this negative correlation?
his math is bad. pure and simple. he is not looking at this the right way and is mistaking action at extremes and a noisy system for negative correlation. the low unemployment in the 40’s and 50’s was driven by somehting other than tax policy.
do this math yourself.
break the system into high tax periods and low tax periods then see if unemplyment trends up or down in them.
“One of thing that has become clear from looking into the data, and that is pretty obvious just from thinking about it, is that changes in tax rates don’t matter much when rates are fairly low”
O.K. great, then we both agree Obama is wrong and there is no need to raise taxes in any income group. At least we agree on something!
Sure, other things matter. But this is the umpteenth post I’ve done on taxes, using all sorts of data and looking it in all sorts of ways, and this was the closest I’ve ever coming to spotting the elusive positive effect of low tax rates. That positive effect really hides itself well, doesn’t it, usually disguising itself as its own exact opposite or nothing at all.
Depending on time, that may get into the to do list.
1. See my response to STR above.
2. I usually do include changes to tax rates in my posts as well. I was in a hurry.
3. That said, what you are saying, whether you realize it or not, is: “Sure, whether taxes are high or low doesn’t affect hiring, but a cut in the tax rate, now that would surely show us what we want to see.” The first part of that is already a major concession that the orthodox view of the effect of taxes is wrong. And the logical conclusion to the second part would be an eventual tax cut that dropped the rate to zero, and then what?
NO you don’t….and that is terrible analogy!
There is a big difference between fast money and slow money. Taxes go to the government and are spent and therefore circulate quickly causing a lot of economical activity that drives employment upward. Untaxed money go to CEO and stockholders who save most of it. This money is not spent as readily so it is slow money which has less economical activity and therefore does not do as a good a job of raising employment as taxes. The assumption that the rich invest most of their money in creating new jobs is false especially when the economy is down.
that is not what i am saying at all.
what i am saying is that the rest of the economic system is too noisy for you to compare the 1940’s to the 1990’s, do a correlation to tax rate, and think you found anything.
there are 2 many other differences for you to do a 2 factor regression and have it be meaningful.
the best way to minimize other factors and therefore maximize the relevance of a 2 factor model is to use a short timeframe. the best way to do that is to look at the differences accross a change for a few tears in either direction.
in doing so, we get a very clear positive correlation.
this result ought to make you very suspicios that the negative correlation you are teasing out is non causal.
the low unemployment of the 40’s-60’s was driven by our post war boom. it had nothing to do with taxes. increased global competition has increased our “natural” unemployment rate.
however, note that the long term rise in unemployment from the 60’s to the early 80’s went into multi decadal decline around the time of the tax cuts.
this and the behavior around the 3 big chages in the last 30 years are not sufficient to prove that tax cuts also cut unemployment (though there is considerable anecdotal evidence) but it certainly makes it all but impossible to caim that there is a causality based negative relationship.
For not being a numbers guy, it seems to me you have it quite right.
I think the big problem with the “taxes are everything” crowd is that they seem to ignore demand. They assume that there is 1 job, and it either stays here or goes to a competing location. And they assume that jobs are created simply by having low enough taxes to attract an investor, and low enough taxes to give the investor the money to invest.
But no matter what the tax setup is, new jobs and business growth won’t occur without the demand for them.
And when there is demand, demand that isn’t being met, more jobs will be created. The market should adjust to meet it and create those jobs. Say you have a very high tax state that is very unattractive for business. The increased demand would lower unemployment in “preferred” locations, until there is no more labor available. Competition for this labor raises their labor cost until the relatively low cost labor in the high tax state is now attractive. Demand is the key factor.
Actually, the IRS link does work in Chrome. My bad.
the biggest drop in unemployment came in during a breif period of low tax in the 70’s. look at the inflections in the 70’s. this data is VERY clear. unemployment has a VERY strong positive correlation in both directions.
No. There was a transcription error in the data set, and the apparent low tax period never happened. Constant top tax rate from 1971 through ’79.
all you are really doing is taking the post war boom with its low unemplopyment and assuming that was caused by taxes, which it wasn’t.
Again, no. I said nothing about causation. In fact, the main point of the exercise is to indicate that the happy cause and effect relationships between low taxation and all that is good and beautiful, that regressives like to talk about, is just so much fiction.
Not sure what else you have looked at but I did a comparison to top marginal PERSONAL income tax rates, which will include many small business owners, and you can see a relationship with unemployment. Now part of what the problem is in doing this type of comparison is that at some point, as we have now, many people will start to fall off the unemployment rolls and thus unemployment can appear to go down when in fact it is higher than ever. This happened under Clinton and is now happening to somewhat under Obama. But if you look at the tax rates on individuals you will see that Carter had taxes up around 70% (top rate) and unemployment was high (pushing 10%). Reagan cut taxes down to about 50% and then down to 31% and unemployment went down. Clinton jacked it up to 40% and unemployement went up. Bush cut taxes and unemployment went down. Of course those are overall. You also see unemployment will start to fall after taxes have been raised for a long time because the small business owner has to raise prices and can eventually afford to hire again.
Corporations are a different animal.
Hey Ilsm, you have anything better than nonsense like “voodoo economics” and “cocktail party game”? Seriously. Any comment with substance?
For example, Volcker’s tight money policy, which Reagan agreed with, helped wring inflation out of the economy (at the price, in the short term, of a deep recession). Are you familiar with this history?
If everyone wants to go back to the golden age of America, maybe we should restore the high corporate tax rate that got us there, and get rid of the IRS all together. Remember why our forefathers left english rule? It was because of personal taxes being too high. Corporations should have to pay for the free market they exploit. if they don’t have a business plan that allows them to succeed at the rate needed for our country to thrive, some new entreprenuer with a better plan will replace them and replace the jobs. Free to succeed, free to fail. America!
Your facts are opposite of what really happened. Do more research.
Hi, I’m at a work computer and can’t access my usual accounts.
I recently did a regression analysis after hearing once again the usual line about lowering corporte taxes. I imediately thought of an article a few years ago in The Economist about Ireland dropping corp tax rates to 1/3 Us rates. Fast fwd to this year an article in NPR. This DID apparently attract some blue chip companies. However the unemployment for Ireland is still around 14.2%. I looked at historical rates compared to U.S. rates. They were comparable and moved in synch mostly.
chose 27 developed nations top marginal tax rates and unemployment levels and ran a simple linear regression. Negative slope on the line with a wide standard error and variance. Hypothesis test for presence of a slope came up negative. Good ness of fit (Rsquare) was .02. Obviously alot more things affect unemployment but that is stilly amount of the affect on unemployment from a change in tax rates. Coefficient on the linear model was -.4% at BEST.
My take. No significant relationship. BUT if you want to force it then it’s negative. Taxes down and unemployment up. The only logic I can come up with that is that in developed nations the tax revenue contributes to social investments that affect unemployment. otherwise it gets to be simply kept by the company.
Tiger – Tiger. Excelent point about the lagging factors. To compensate I’d need to make a multiple regression with lagged Tax periods. I’ll look at that if I can get the data.
Obviously much more affects unemployement than corporate taxes. That is a big reason why the Rsqr is so tiny and the coefficient so small also. But I wonder how different a conclusion one would come to if they built or worked with an already existing robust model of unemployment that includes Corporate Tax rates and took the partial derivitives. That assumes that corp. Tax Rates were significant enough to include in the model of unemployment.
I’m guessing that like much of liberatopian economics its an over hyped relationship.
Mike, The 70s and 80s were skewed by Big Oil. As the military industrial complex affected the data during the beginning of the 70s, Under regulated the oil companies and the newly formed OPEC bled the economy dry. Once a third factor comes into play that soaks up the currency in circulation, artificially, the other factors will be skewed.
Reagan restructured the tax code to move money offshore. Previous deductible items were negated causing a swing in corporate spending.