The Effect of Individual Income Tax Rates on the Economy, Part 6: 1981 – 1993
by Mike Kimel
The Effect of Individual Income Tax Rates on the Economy, Part 6: 1981 – 1993
This post is the sixth in a series that looks at the relationship between real economic growth and the top individual marginal tax rate. The first looked at the period from 1901 to 1928, the second from 1929 to 1940, the third from 1940 to 1950, the fifth looked at 1950 – 1968, and the sixth from 1968 to 1988. Because the Reagan era is so pivotal in the American psyche, though it was covered in the last post, I intend to focus on it again. The last post included the lead in to Reagan’s term, this post contains the follow-up to his term. In this post I’ll look at the period from 1981 to 1993.
Before I begin, a quick recap… both the 1901 – 1928 period and the 1929 – 1940 failed to show the textbook relationship between taxes and growth. In fact, it seems that for both those periods, there was at least a bit of support for the notion that growth was faster in periods of rising tax rates than in periods when tax rates were coming down. It is worth noting that growth from 1933 to 1940 was generally quite a bit faster than at any other peacetime period since data has been available, both on average and for individual years. Not quite quite what people believe, but that’s what it is.
In the 1940 – 1950 period, we did observe slower economic growth following a tax hike and faster economic growth followed a tax reduction. However, that happened when the top marginal tax rate was boosted above 90%.
Interestingly enough, though the so-called “Kennedy Tax Cuts” are often used as one of the prime exhibits on the benefits of cutting taxes, a look at the 1950 – 1968 period yields no such conclusion. Growth rates were already rising before the tax cuts occurred in 1964 and 1965, reached a peak when the tax cuts took place, and started shrinking immediately afterwards. The other period that is always pointed to as evidence that tax cuts spur growth is the Reagan years, which showed up in the 1968 – 1988 post. It turns out that put into context, the Reagan years produced one year of rapid but not particularly extraordinary growth a few years after tax cuts began. That’s it.
Real GDP figures used in this post come from Bureau of Economic Analysis. Top individual marginal tax rate figures used in this post come from the IRS. As in previous posts, I’m using growth rate from one year to the next (e.g., the 1980 figure shows growth from 1980 to 1981) to avoid “what leads what” questions. If there is a causal relationship between the tax rate and the growth rate, the growth rate from 1980 to 1981 cannot be causing the 1980 tax rate. Let me stress this point again as I’ve been getting people e-mailing me to tell me I’ve got the growth rates shifted a year. That is correct, and is being done on purpose (and is shown on the graph labels). To avoid questions of causality, the growth rate in year X used in this post is the growth rate from year X to year X+1. And when I say “to avoid questions of causality” – you’d be amazed at how many people write me when I don’t do this and insist that sure, higher tax rates seem to be correlated with faster growth, but that’s because when growth is faster governments feel more willing to charge higher tax rates.
With the preliminaries out of the way, let’s get started. The first graph shows the tax rates from 1981 to 1993 along with the t to t+1 real GDP growth rates.
It goes without saying that what the graph does not, repeat, does not show is that lower tax rates have much to do with faster economic growth. In fact, some of the slowest sustained economic “growth” that occurred during the Reagan-Bush years coincided with the lowest tax rates: 28% and 31%. The one standout year occurred when tax rates were at 50%, and had been at 50% for a few years. And yet, somehow this period has entered the public consciousness as Exhibit A that Tax Cuts Work.
That said, I’d like to point out that unlike the folks who venerate Reagan today, Reagan himself did have a reason, an excellent reason, in fact, to try tax cuts… at least the first round of tax cuts. Looking back from the vantage point of 1980 and leaving out the WW2 years, real economic growth when tax rates were in the 90% + range was lower than it was when tax rates were in the 80% to 89.9% range, and that was slower than when tax rates were in the 70% to 79.9% range, and that in turn was slower than when tax rates were in the 60% to 69.9%. That is shown in the graph below.
Note that all the information contained in Figure 2 was available by the time Reagan took office. If the information in Figure 2 is all you have, it isn’t unreasonable to wonder whether further reductions in the tax rate will lead to faster economic growth. Of course, Reagan did have a bit more information available. He also had growth rates from the last time tax rates were in the 24% and 25% range (i.e., the start of the Great Depression) which were negative… and which might have tipped him off that the relationship between tax rates and growth is actually quadratic. But I guess its a bit much to expect any of Reagan’s advisors to consider anything like a quadratic relationship.
In any case, we can combine Figures 1 and 2 to put the Reagan – Bush rates into context:
If you’re wondering, during seven of the 12 Reagan-Bush years, growth rates were actually below the average rate observed when top marginal tax rates were above 90%… and the really slow growth during the Reagan – Bush era occurred disproportionately when tax rates were at the 28% and 31%. That is to say, when tax rates were at their lowest levels in the Reagan – Bush era, growth rates were also at their lowest. And as the graph also shows, every single year, repeat, every single year of the Reagan Bush had a lower average growth rate than when tax rates were in the 60% to 69.9% range.
So… what we don’t from the Reagan – Bush era is anything that supports the notion that lower tax rates correlate with faster economic growth. (Note… correlation does not imply causality, but lack of correlation certainly does imply lack of causality.) We do see that it could have been a rational experiment for Reagan to cut tax rates from 70% to 50%. It was not a rational experiment, based on what happened at 50%, to cut rates further, and the result was easily predictable.
And speaking of rational… the story the data tells is strongly at odds with what is commonly believed. And this isn’t ancient history. Most of us lived through this. It isn’t rational for us to believe things that aren’t true. But collectively, we do. And its leading to crummy growth rates. What a surprise.
Next post in the series… 1993 to the present.
As always, if you want my spreadsheets, drop me a line. I’m at my first name which is mike and a period and my last name which is kimel (note that I’m not from the wealthy branch of the family that can afford two “m”s – make sure you only put one “m” in there) at gmail period com.
Can you expand on what you see as the implications?
It appears that I should conclude that central planning works better than market planning, since higher taxes (taking money away from market allocators) and higher government spending (centralized allocation of capital) lead to higher economic growth. (To conclude this I would have to attribute the failure of the Soviet Union to other causes, but that’s conceivable.)
A further implication is that all of us who are now at the higher marginal tax rates should, in your prefered world, see those rates go way up. We should look forward to taking life easier and working less (not an unappealing tradeoff), since it wouldn’t make any difference to our net incomes, and on your evidence our collectively higher levels of sloth won’t hurt the economy at all.
It’s certainly counterintuitive, so I want to make sure that’s what you are saying.
Is it?
Oh – a further implication appears to be “incentives don’t matter.” That’s both counterintuitive and contrary to well-established fact.
Do you have any thoughts on how negative incentives could not lead to negative outcomes?
Thomas,
I believe that GDP is an outdated for what Mike is trying to prove. The analysis for the effect of government spending is complex, which leads to one asking how accurate is the analysis of the situation where tax rate increase, but government spending decreases, and the opposite where tax rates decrease and government spending increases?
“The way the GDP accounts for government spending is biased: It assumes that if the government spends $200,000 on a contractor to repave a road, it creates $200,000 of genuine economic value. If Exxon Mobil pays an engineer $200,000 per year, that only shows up in GDP if the engineer finds an extra $200,000 of oil to sell, or builds a new machine that sells for $200,000, something like that.”
http://www.nationalreview.com/corner/260747/impact-spending-cuts-economy-veronique-de-rugy
Darren,
I remember being troubled when I first encountered Keynes’ analysis and realized that it treated G as contributing to GDP without at the same time subtracting T (the taxes that help to pay for G). After all, those taxed would either have spent the money (C) or invested it (I). Similarly, money borrowed by the government would either have been spent or invested privately. (And, contra Keynes, all money is either spent or invested, except for cash carried in people’s pockets, which does not vary by enough to account for economic cycles.)
I am also deeply troubled by the fact that GDP shows World War II as a boom, instead of as one of the greatest destructions of wealth in human history. The definition of GDP leads to people making utterly daft (and morally objectionable) claims that the war pulled the country out of the recession, instead of what it actually did, which was to make matters even worse! It also treats the subsequent peace as a massive welfare collapse (instead of a great blessing). Clearly GDP is a deeply flawed measure of the national welfare.
GDP’s mismeasure of war is equivalent to the government investing in breaking windows, then calling that progress. Which, in itself, hints at the problem I have with treating government expenditure as a straight contributor to GDP.
Perhaps an entirely different measure is in order: the annual change in the nation’s net worth (private assets plus public assets minus private debt minus public debt) – although this would get us into another tangle, about how to value the assets (market values reflect expectations of future value). Such a measure would not jump up when the government (or any investor) spends money, as the increase in government assets would be exactly offset by the funding source; but if high tax/high government spending does create growth, it should lead to higher subsequent growth rates in the measure (and vice versa).
However, what I was trying to get at was something different. I favor small government and low taxes in part because there is a consistent motivational/behavioral (i.e., microeconomic) theory for how that should lead to greater national wealth; and, in part, because I believe that liberty in and of itself has value, and thus that our institutional designs should move us toward it rather than away from it. On the other hand, I support non-zero government (versus zero government) because, again, there are microeconomically consistent theories supporting some, versus no government. If there were a consistent microeconomic theory showing how central planning and high taxes would lead to greater national wealth, I’d like to know about it. If I’m wrong in my policy preference, I’d like to know it – or at least, I’d like to be able to see the tradeoff between liberty, as a human value in itself, and material welfare, as a different human value. So, I’m asking if there is one. Note that Keynesianism isn’t one: it almost defines macroeconomics as an emergent phenomenon without regard to the underlying micro, and makes no effort to offer a connection between the two.
Thomas,
I know this is a simplistic statement, but it sounds like you are more interested in what theory says that what history says.
Thomas Boyle,
This series of posts is simply a look at the effect of changes in tax rates on economic growth. I have been taught my whole life that cutting tax rates leads to faster economic growth. Example A are the Reagan tax cuts. In the last two posts I put up a graph. I’m just not seeing what I’ve been told my whole life is there.
This is not an argument for central planning. It is not an argument against central planning. This is merely a look at the data. If you need an explanation, this is the best one I have: http://www.angrybearblog.com/2010/12/simple-explanation-for-strange-paradox.html
Note that it also neither argues for or against central planning.
That said, I’ve been posting on the relationship between tax rates and growth for a long time. I have for the most part focused on using simple graphs rather than esoteric statistics, though I think on occasion I’ve thrown in impulse response rates and simulations. The data looks the way it does unless one chooses to torture it. Based on what the untortured data seems to say, I would in fact support raising the top marginal tax rate.
Darren,
This series of posts is not looking at government spending at all. It is looking at taxation. The two are unrelated, as Reagan demonstrated by running up the debt.
Thomas Boyle,
1. This series of posts is not based on Keynesian theory. It is merely a look at taxation and growth rates that follow. Period.
2. “I am also deeply troubled by the fact that GDP shows World War II as a boom, instead of as one of the greatest destructions of wealth in human history. “
I suggest you go back to the third post in this series which looks at the WW2 years. Especially when looked at in context (say, after looking at the second post) you might find that the data actually backs of up this particular thought of yours.
3. “I remember being troubled when I first encountered Keynes’ analysis and realized that it treated G as contributing to GDP without at the same time subtracting T (the taxes that help to pay for G). “
I’ve had plenty of posts looking at Reagan’s activities in the past. A big part of the GDP gain under Reagan is due to his willingness to drive up G and simultaneously drive down T. In other words, a big part of his GDP growth is due to debt. My co-author and I actually estimate some of this in Presimetrics, if I recall.
Jerry Critter,
Give Thomas the benefit of the doubt. I’m not sure why, but there’s a massive mythology which gets repeated over and over and over and over. Sometimes an actual look at the data itself changes people.
Mike,
If you GDP as a measuring tool, then government spending plays a major role in the analysis.
Mike,
“It is merely a look at taxation and growth rates that follow.”
How do you determine the benefit or damage that is done by a tax rate change when when benefit and damage is not held constant by the “G” in the GDP Calculation?
“A big part of the GDP gain under Reagan is due to his willingness to drive up G and simultaneously drive down T”
The House and Senate was controled by his opposition, and refused to allow a tax cut without the increases in Government Spending. Also, you’re admitting to the fact that government spending blurs real economic growth in the private sector if GDP is the measurement.
You’re assuming that the “G” in the GDP calculation drove actual economic growth in the private sector, because revenues increased every year after 1983, but the tax rate lowering was not as important….how do you come to this conclusion?
Mike,
“It is merely a look at taxation and growth rates that follow.”
How do you determine the benefit or damage that is done by a tax rate change when when benefit and damage is not held constant by the “G” in the GDP Calculation?
“A big part of the GDP gain under Reagan is due to his willingness to drive up G and simultaneously drive down T”
The House and Senate was controled by his opposition, and refused to allow a tax cut without the increases in Government Spending. Also, you’re admitting to the fact that government spending blurs real economic growth in the private sector if GDP is the measurement.
You’re assuming that the “G” in the GDP calculation drove actual economic growth in the private sector, because revenues increased every year after 1983, but the tax rate lowering was not as important….how do you come to this conclusion?
Mike,
If you use GDP as a measuring tool, then government spending plays a major role in the analysis.
Mike,
I see you’re correct. Your claim that tax rates don’t affect economic growth rates at all, says that central planning is no more or less effective than market allocation of resources, from the point of view of economic growth rates.
Of course, since the data doesn’t support raising the top marginal tax rate, I’d suggest that basic fairness would recommend lowering it. Taxes are forcibly extracted, and thus have to clear a moral hurdle. This data doesn’t suggest that they do (on these grounds alone, granted)
Mike,
Thanks for that. Now that I’ve read it, I’m going to have to think about it. Several quick reactions:
1. It’s a (possibly) good example of what I’m looking for. If correct, it implies that consumption taxes would be a better tax system than income taxes. For example, it would allow ordinary income earners to invest in businesses also – all invested savings would behave like 401(k)s). That, in turn, points toward why it would be important to understand the mechanisms behind the apparent anomaly: they may point toward better approaches than “soak skilled/hard workers,” which is the result that leftist prescriptions currently tend to wind up having.
2. It’s not clear that, in and of itself, it would be enough to explain the apparent anomaly
3. Business investments are not tax deductible, as such. The income tax is payable on the company’s earnings, which do include a depreciation allowance for past years’ investments. I’ll have to think about how the math unwraps and get back to you. I think it’s an intriguing topic for study, though.
Darren,
You weren’t commenting around here at the time so you probably weren’t reading the site, but I wrote a series of posts on Gov’t spending less the increase in the debt. We also have a chapter on that in Presimetrics.
I’ve also looked at “private GDP” – that is, GDP less G.
That said, everything in the economy is interconected. Whatever measure you prefer will be affected by gov’t spending. And I don’t think you’re going to find taxes affecting whatever your favorite measure of the economy might be in quite the way you expect.
Darren,
“How do you determine the benefit or damage that is done by a tax rate change when when benefit and damage is not held constant by the “G” in the GDP Calculation? “
I’m not. As I noted upthread, I did posts in the past on “private GDP” and on GDP less the increase in debt. But I don’t see the issue here – how strongly do you think gov’t spending is with taxation? And what would you need to see be seeing the expect from tax changes appearing in the graphs?
“You’re assuming that the “G” in the GDP calculation drove actual economic growth in the private sector, because revenues increased every year after 1983, but the tax rate lowering was not as important….how do you come to this conclusion?”
I suspect you’re reading a different post. I don’t see how you get anything about G driving GDP growth anywhere.
As to revenues increasing yadda yadda. For crying out loud: I covered that years ago too, but this canard comes up so often that I remember the precise link: http://www.angrybearblog.com/2007/01/in-my-opinion-thomas-sowell-is-hack.html
Darren the Dems in Congress weren’t the ones pushing the increases in G under Reagan, which were focused on defense spending if you recall. I don’t think that we should decrement GDP growth rates under Reagan to the extent that they reflect defense spending increases. Notwithstanding a preference to leave out WWII years. GDP is an imperfect metric that includes many things of questionable value to the economy.
Darren the Dems in Congress weren’t the ones pushing the increases in G under Reagan, which were focused on defense spending if you recall. I don’t think that we should decrement GDP growth rates under Reagan to the extent that they reflect defense spending increases. Notwithstanding a preference to leave out WWII years. GDP is an imperfect metric that includes many things of questionable value to the economy.
Darren,
In the past, I’ve tried GDP less G, and I’ve tried growth in GDP less growth in debt. (The latter is actually in a chapter of Presimetrics.) You’ll have to search for them.
That said, all measures affect each other. You’re not going to get the effect of G to go away by using, say, Income.
Besides that, why should that cause the effect that you expect taxes to have to go away?
The implications for this are that high tax rates clearly fall more on economic rent than they do on productive activity. Hence, much of the activity they do discourage is a positive for the economy as it diverts the income to productive uses.
The discussion of Reagan brings a thought about your anaysis of taxation and growth. I’m not a big fan of Reagan or the theory that cutting taxes creates growth or jobs. But what if you consider Social Security and Medicare taxes? If you consider those then Reagan was not a tax cutter, but a tax hiker. He just did the reverse Robingood thing: taking from the poor and giving it to the rich. But then, maybe the correlations with tax hikes and cuts and growth in the economy might seem more plausible [GASP! I can’t believe I even said it].
zenzane,
Correct. Nevertheless, for this particular series of posts I’m focusing on marginal rates, those being the rates that orthodox economic theory and Rush Limbaugh both tell us matter.
Note to Dan – something is odd about the comments. When I check on my phone I see more comments than on my computer. What’s up?