Optimal Tax Rates for Generating Economic Growth According to Barro-Sahasakul Tax Data
By Mike Kimel
Optimal Tax Rates for Generating Economic Growth According to Barro-Sahasakul Tax Data
This piece is a bit more wonky than what I normally post.
I recently re-read “Macroeconomic Effects from Government Purchases and Taxes” by Barro & Redlick. I was struck by how different the conclusions they make about taxes are from what you get if you simply make a bar chart of the top marginal rate at any given time versus the growth rate over the next year.
Now, obviously, Barro & Redlick take a completely different approach… but at the bottom of everything is the data set they use (see Table 1 of the above referenced paper and this explanation of the “Barro-Sahasakul” data set). To cut to the chase, they use estimates of the average marginal tax rates paid by taxpayers rather than the top marginal rate that I used in the bar chart referenced above. Their overall marginal rate is made up of not just federal tax rates, but also social security tax rates, and even estimates of the state tax rates. It should be noted the Barro is an average rate, and since the average includes non-filers (who pay zero), the Barro rate is often well below the top marginal rate. The top Barro rate is 41.8% which occurred in 1981 (compared to top marginal rates of 90%+ from 1951 and through 1963). The Barro rate is also not correlated with the top marginal income tax rate (correlation going back to 1929 is -30%).
A lot of work clearly went into producing this overall marginal rate (I’m going to call it the “Barro tax rate” for simplicity). But does it explain economic growth rates any better than the top marginal rate?
I ran a quick and dirty regression…
Growth in real GDP from t to t+1 = f(Barro Tax Rate, Barro Tax Rate Squared, Top Marginal Income Tax Rate, Top Marginal Income Tax Rate Squared)
Data ran back to 1929, the first year for which real GDP was computed by the BEA. Top marginal rates came from the IRS Statistics of Income Table 23. And the Barro Tax Rate came from Table 1 of the Barro & Redlick paper. Since Barro rates are computed only through 2005, that’s when the analysis stops.
Results were as follows:
(Note… the errors got big during leading up to WW2, but I don’t think that invalidates this quick and dirty look.)
Here’s what I get out of this:
1. There is definitely a quadratic relationship between tax rates in one period and real economic growth the next.
2. If you’re going to pick either the Barro rate or the top marginal income tax rate, go with the latter. Its clearly better at explaining economic growth rates.
3. There may be something to be said about using the Barro rate and the top marginal income tax rate together. They do explain different things.
If you compute the “optimal tax rate” – the rate that maximizes economic growth implied by the regression – you get a Barro rate of 25% and a top marginal income tax rate of 64%. The optimal Barro rate was last seen in 1966, when the top marginal rate was 70% and the bottom rate was 14%. I’m guessing from this, and from looking at the Barro rate series, that this would imply that if you want to maximize growth, the top rate should be raised to about 64% and the tax burden on folks at the lower end of the income scale should be lowered. I’m not sure Barro would be pleased with these results.
I may return to this, but my next post should be the next in the series on GDP growth and the S&P 500.
As always, if you want my spreadsheet, drop me a line with the name of this post. I’m at my first name (mike), my last name (kimel – with only one m) at gmail.com. BTW… this spreadsheet contains a lot more wonky goodness!
Thanks to Sandi Saunders for getting me started wading through this particular pile of weeds.
An optimum Barro rate of 25% when the top marginal rate was 70 seems to be consistent with your analysis of a top marginal tax rate of 64%, at least with in the margin of error of these analyses.
Both point to the need to increase the top tax rates. I would also add more and higher brackets at the top.
Which still leaves the question of whether maximizing growth of GDP is what we want to do. Personally I’d settle for a little less growth and a little cleaner air, or pay levels and housing for the least of us that did not shame our humanity.
In any case I’d be a little leery of basing an economic policy on one number which has unexamined inner dimensions.
That said, I’m all for Kimel’s project if it encourages a few Presidents to risk raising the tax rates a little. My own kentucky calculations show that a 3% increase in the net federal income tax on income over 100k will pay down the SS Trust Fund debt (that’s money “we” owe to the people who already paid for their SS benefits) over the time that money will be needed. Before the great recession that would have paid down enough of the total national debt to get us back into “sustainable” range.
In any case it’s hard to see how we are going to continue as a country worth living in if we don’t pay enough taxes to cover our debts and pay for the things we need.
Social Security, it should be pointed out, pays for itself and has nothing to do with the national debt… except of course by having lent money TO the United States of America.
Raising the tax about 3% on ALL income would start to roll back the “deficit” right smart. And I’d bet we’d see an economic recovery, which, all by itself would pay off even the great recession deficit in about four years.
I suspect you are on to something with quadratic models for the effect of marginal tax rates.
But, and I’m sure I’ve said this before, you need a basic growth model. Throw in the usual variables such as labor force growth, gross fixed investment and some measure of human capital (such as secondary school attainment or average years of schooling). And then perhaps throw in an additional fiscal variable (such as the fiscal balance).
Do the results hold up? Are they robust? (No one will be happier than me. )
If you don’t do it, sooner or later I will.
Does the author assume that increasing the top marginal rate will result in greater revenue from the wealthiest taxpayers? It would be wrong to draw this conclusion based only on the statistics provided, especially given some contrary evidence of the past.
Does he attribute some correlation between the top income tax rate and GDP to a changing behavior in the richest class to avoid paying taxes? Are they working less or distributing more of their business receipts to employees? Are they less likely in high tax environments to address emerging markets, leaving opportunities for lower income individuals to pursue the risk?
Since this theory should not be isolated to the United States, it’s hard to understand why other nations have not adopted a 64% top tax rate. Maybe Belgium, Sweden, Germany, etc. know about the correlation between high tax rates and their citizens’ affinity for Monaco.
“Does the author assume that increasing the top marginal rate will result in greater revenue from the wealthiest taxpayers? It would be wrong to draw this conclusion based only on the statistics provided, especially given some contrary evidence of the past.”
The author is only providing evidence of a growth effect not of negative evidence of some oddball Laffer Curve effect.
“Does he attribute some correlation between the top income tax rate and GDP to a changing behavior in the richest class to avoid paying taxes? Are they working less or distributing more of their business receipts to employees? Are they less likely in high tax environments to address emerging markets, leaving opportunities for lower income individuals to pursue the risk?”
Possibly. But then this increases the growth rate doesn’t it? So what does it matter how it occurs?
“Since this theory should not be isolated to the United States, it’s hard to understand why other nations have not adopted a 64% top tax rate. Maybe Belgium, Sweden, Germany, etc. know about the correlation between high tax rates and their citizens’ affinity for Monaco.”
Empirical evidence suggests that the labor supply curve is extremely inelastic at high compensation levels. That’s partly because in most cases compensation is a flat fee at those levels, not an hourly wage. It’s also related to the fact most people don’t choose where they live based solely on tax rates.
I am not suggesting that rates should suddenly be doubled, or the maximizing GDP growth should be the goal, but when rates are half of the optimum values for maximum GDP growth, to talk about further cuts in the tax rate is stupid…and most likely harmful to the economy.
The top marginal rate needs to be higher than 70%. It needs to be 95% at the top. Lack of economical growth is not the only problem that rich people cause. They also you their wealth to corrupt and control government. They then use government as an attack dog against the poor and middle class. This causes the people to disconnect from their government and from sociaty which increases the chance that their will be social disruptions.
“The author is only providing evidence of a growth effect not of negative evidence of some oddball Laffer Curve effect.”
The author’s premise should be backed by a rational argument, because correlation does not imply causation. If you don’t know the ‘why’, then it’s just an exercise in probability using a credulous notion. Will a higher tax rate on the top income earners increase revenue from them? If so, why does GDP increase? If not, why does GDP increase? Why are you trying to force Laffer’s theory on a small segment of the tax base, when the author should have evidence from past tax receipts (especially when it is used overwhelmingly on total, national economies/populations/revenues)?
“Possibly. But then this increases the growth rate doesn’t it? So what does it matter how it occurs?”
If you don’t know the how, then you’re guessing. Would you be satisfied with a 65% tax on income over 78 million dollars if everything else about tax rates remained exactly as they are now (for incomes under 78 million)? Going by the argument put forth so far, this new tax bracket should solve all of the nation’s revenue and growth issues, no matter where we place the income level (78 million, 4 billion, 10 billion…). Also, if just raising the top tax rate is the great solution, how come there were years of negative GDP growth under high rates?
“Empirical evidence suggests that the labor supply curve is extremely inelastic at high compensation levels. That’s partly because in most cases compensation is a flat fee at those levels, not an hourly wage. It’s also related to the fact most people don’t choose where they live based solely on tax rates.”
Remember, we’re not talking about most people. Just the ones in the top 1%-5% of income earners. The ones who would consider moving to a tax haven like professional athletes, actors, musicians. My concern about changes in high earners’ actions isn’t along the lines of their labor contribution. It’s more about how they move their money when rates are doubled.
One importatant question for you: What was the effective tax rate for someone in the top 1% of income in those high tax years, say 1936?
Its also somewhat consistent with the data that shows states with higher tax rates get faster economic growth that states that charge less.
GDP is an imperfect measures of the economy. It doesn’t take into account income distribution or externalities and it has a whole host of other things wrong with it. That said, it is the “the number” that people work with. Its difficult enough to convince people that the data shows X when you are using real GDP. Its impossible when you have to make up the target variable. Get tax rates to the optimal rate for real GDP growth and then you can move on to things that would make ti better.
Mark A Sadowski,
Its a slow process. I have a full time job, a ten month old baby, and an assortment of critters that depend on me for sustenance. Blogging is a hobby. It also helps keep me sane. So getting there from here is slow.
If you want to collaborate on this, I’d be most pleased. You know how to find me via e-mail.
You may recall a good starting point for is this story: http://www.angrybearblog.com/2010/12/simple-explanation-for-strange-paradox.html
The author assumes this story:
It certainly governs the author’s behavior, and the behavior of other people he knows in the same boat.
The fact that a certain theory has gained cachet (i.e., low taxes = faster growth) doesn’t make it true. Yes, a number of countries worldwide have reduced their tax rates.
But… have you checked the top marginal rates in the countries you mention? It may not be 64%, but its not 35% either. (http://www.oecd.org/dataoecd/46/18/2506453.xls)
You want low top marginal rates, you want to go to Poland, the Slovak Republic, or Mexico.
As to people moving to Monaco. Yes, it happens. And yet, the countries that tend to do well in the long haul for their citizens tend to be able to have high tax rates. (I think I’ll write a post about this, with data, soon.)
But that’s like saying there’s no point in having a police force because some people are going to murder other people anyway.
See upthread for a link to the author’s explanation, or go straight here:
“ Also, if just raising the top tax rate is the great solution, how come there were years of negative GDP growth under high rates? “
The economy fluctuates. How come the Great Depression followed a decade of tax cuts? (The top rate dropped from 73% in 1921 to 24% in 1929. Similarly, the tax rate was 39.6% in 2000 and 25% in December 2007.) I mean please, allow for some variation.
As to people moving to avoid taxes… people engage in tax avoidance no matter how low the tax rate. Why do you think high income folks will pay Pricewaterhouse Coopers a fortune rather than go to H&R Bloch. Are they morons, or do they know that PwC knows more tricks?
I have enough of a hard time getting folks to accept empirical results. I cannot afford to take on arguments that I cannot quantify.
Please take your own missive to heart as well. Scolding Mike for something he is not trying to do, and then offering the going Galt counter, indicates your premise should be backed by a rational argument.
The finest Barro rate of 25% when the top marginal rate was 70 looks reliable with your analysis of a top marginal tax rate of 64%, at least within the edge of error of these analyses. Cash loans payday
For those who would like to go beyond Mike’s issue and more into the “why” of taxtion there are these postings:
Taxation’s Rhetoric: Today and yesterday’s economic crap
Then there are the following 3 articles on “why” here:
“See upthread for a link to the author’s explanation, or go straight here:”
The reason provided in the above link points to an assumed increase in income producing investments by the wealthy during eras of high tax rates. Yet we know that increased capital in income producing vehicles is influenced most by the investment environment and perceived opportunities. Increasing tax rates influences capital preservation the most.
“I mean please, allow for some variation.”
If there was a concern about variation, you would have removed outliers in your study. By using all GDP growth rates, a rational should be given for extreme values and their inclusion.
“As to people moving to avoid taxes… people engage in tax avoidance no matter how low the tax rate.”
The interest in tax shelters increases during times of high tax rates. Ask any tax attorney around in the 60s and 70s and they’ll tell you about how billions were kepts from the IRS. These vehicles weren’t investements, just schemes for the wealthy to legally report losses. It Maybe that’s how we’ll get the economy moving again: revamp the tax shelter industry to give unemployed lawyers jobs. I guess that’ll increase GDP…
“As to people moving to avoid taxes… people engage in tax avoidance no matter how low the tax rate. Why do you think high income folks will pay Pricewaterhouse Coopers a fortune rather than go to H&R Bloch. Are they morons, or do they know that PwC knows more tricks?”
And you want to encourage that activity with your tax policy?
Again, all we need to do is create a new bracket for individuals who make more than 5 billion a year. Tax them at 64% or whatever is deemed necessary and leave everything else the same. Everyone will be happy. The new bracket won’t apply to many, if any individuals and you’ll get the ‘ideal’ increase to the reported top tax rate. According to your argument, GDP should skyrocket and all problems will be solved.
“You want low top marginal rates, you want to go to Poland, the Slovak Republic, or Mexico. “
And the GDP growth rates of Poland, Mexico, and the Slovak Republic compared to Italy, Spain, Japan (47.7%), Greece? Why is there a problem with PIIGS, given their tax rates?
i respectfully… very respectfully… disagree. I think you need to keep up the good work, and GDP is the measure you have to work with.
But if we just stop thinking with “GDP is an imperfect measure…” and don’t get serious about what we really want, we are going to have to live in a poisoned planet with the “wealth” going to the people who know how to grab it.
I agree with Jerry… cutting taxes now is stupid. Raising taxes now would probably be smart. But very soon we are going to have to decide what to do with those taxes… and it should NOT be “maximizing growth.”
Bravo Mike, I think it is a damn good analysis. I had a simlar thought back a while, why cannot we have it as an amendment. Tax policy and rates are bound to ranges X,Y,Z.
To simplify this crisis in whole, it’s a balance sheet recession and the and the consumer who makes 70% of GDP, is flat broke. Banks got righted by the taxpayer (socialism for the rich), the multinationals that sent the jobs overseas are plenty healthy sitting on cash, small buisness who employee vast majority are struggling, The appropriate action for the government would have allowed the taxpayers (200k in debt each and all of us) to pull from future earnings to either extinguish debts or reinvest as deemed appropriate by the individual. This also would have redistributed the future taxes accordingly. If a fmily making 40k got 5k it is a larger percentage than someone making 200k and gettig 5k, it starts to heal the inadequate previous tax policy immediately. The balance sheet in question is the consumers. Correct the balance sheet of the consumer and recession ends.
Whether a middle class family dealing with underwater mortgage and paying more to inflation of consumer goods, or a poor person on wellfare goes to WalMart and spends 100% of that money. That is what increases GDP, not rich people adding to their paper assets wealth.
When the consumer is broke you have to get them moving again. Brazil did it with giving the poor people an opporutnity to participate, same thing goes here. The rich feed off them……. And now Brazil’s wealthy admit it. Capitalism has a fundamental flaw toward monopolies and oligopolies and concentration of wealth, their has to be a redistribution mechanism, otherwise cohesion of society collapses under the weight of the unequal distribution of income. Sometimes obtained by the wealthy by straight fraud and political favors.
“Correlation is not causation,” but it’s a damn good place to start looking.
Near as I can tell your comments re Kimel’s work are a pretty good demonstration of the futility of trying to find “reasons” that will satisfy someone who has reasons of his own.
Also near as I can tell your argument amounts to a tax policy based on giving in to the blackmail of “the rich.” Me, I wouldn’t argue with these people. I’d just call their bluff.
IMO, if you are going to maximize one variable, the median income is generally better than GDP, in terms of socio-economic well being. Right now, I think that the important variable to maximize is employment. If we just look at GDP, everything is fine now, isn’t it?
Kevin: “The author’s premise should be backed by a rational argument, because correlation does not imply causation. If you don’t know the ‘why’, then it’s just an exercise in probability using a credulous notion.”
I am waiting to see what causal connection there is, if any. However, I think that demonstrating the correlation is quite valuable, as it debunks the notion that reducing taxes bolsters economic growth, in general.
Brett Beerer: “Capitalism has a fundamental flaw toward monopolies and oligopolies and concentration of wealth”
I don’t think it is just capitalism. Agriculture may well be to blame. But yes, we need to counteract the tendency to concentrate wealth and power.
if Kevin’s argument is true, and higher tax rates drive rich people out of the market, that could be the mechanism by which they increase productivity.
Kevin assumes that the rich are productive My guess is that at best, they WERE productive. Now they are just more dead weight sitting on top of the economy preventing enterprise.
Even the Progressive Caucus proposal does not propose top rates remotely close to what we had before Bradley-Gephardt, but this history does suggest that many academics, economists, and business experts should join Mike in giving it serious attention and argue about implications and theory. In terms relevant to policymakers, top rates approaching Reagan/Kemp-Roth or JFK-LBJ would imply a 50% rate for the current top bracket. For JFK-LBJ, in 1965 60% kicked-in at about $625K, and 70% started at about $1.5M. The rates also applied to most dividends after an exclusion for small investors.
Beyond GDP growth, I also am curious about what such top rates would mean for our ability to pay for our government, which is what taxes are for. The current top bracket affects the top 1% of earners who currently are receiving 22-24% (over $1.7T) of all AGI. This is over double the share they got during the thirty years ending in the early-80s, when they were paying higher effective tax rates. So this would seem to be serious money, even before considering the other tax brackets.
I think you’re better off with a model that has some implied labor and investment elasticities. In addition, there’s just too much multicollinearity between top and average marginal tax rates for me to get excited about the regression. And I think growth responds to tax rates (and the degree of government investment expenditure and its ROI) with a multiple lag structure.
509 New York Avenue
Norfolk, VA 23508
Actually, GDP growth now is anemic, isn’t it?
Ummm… Poland and the Slovak Republic would have a dummy variable for recently rid of the Soviet Empire. There is such a thing as the catch-up effect.
As to Mexico… you really are of the opinion that Mexico is growing like gangbusters? So how many years do you think it will be before people are flocking to live in Mexico because of their high rates of growth?
As to Italy, Greece, and Portugal – here I was under the impression that nobody paid taxes in those countries, and that that’s been true no matter how low or high the tax rates have been.
1. I’ve done this before, with only top marginal rates. Results are the same.
2. In a nod to the multicollinearity issue, I mentioned this in the post: “The Barro rate is also not correlated with the top marginal income tax rate (correlation going back to 1929 is -30%).” Other issues may be a concern (this particular version of the model contains a small amount of autocorrelation but in my opinion not even the ballpark of mattering).
3. I agree that other variables matter. Obviously, the adj R2 is very low. But blog posts have to be bite sized to some degree. I’ve had other posts where I’ve added a number of variables, but this one was intended more to comment on the Barro-Sahasakul data set and Barro and Redlick’s results than anything else. See, I think Barro is very skilled and he knows what he’s doing. (That, in my opinon distinguishes him from, say, Christy Romer and perhaps Glenn Hubbard, but puts him together with Greg Mankiw.) Thus, when he gets a Romer/Hubbard/Mankiw style result, its worth asking – what did he do to get that result. Now, sometimes its the data set (e.g., Romer and Romer). And my first thought, given that he did use the Romer and Romer data set later in the paper, was that the problem was with his data set.
That was not the issue (which surprised me). Its the model specification. I’m pointing out one extremely obvious thing that a super-skilled guy like Barro has to avoid discussing in order to get the sort of results that the Mercatus Institute will publish.
“As to Mexico… you really are of the opinion that Mexico is growing like gangbusters?”
–Its GDP is growing at a faster pace than Spain, Greece, Italy, etc. The accuracy of Mexico’s growth is no less accurate than saying the U.S. grew 126 percent from 1933 to 1942 (in the depths of the Great Depression until it entered WWII).
“As to Italy, Greece, and Portugal – here I was under the impression that nobody paid taxes in those countries, and that that’s been true no matter how low or high the tax rates have been.”
—I don’t know about Greece, but the Italians (northern and central Italians at least) and the Portuguese pay their income taxes (as well as other taxes).
“Ummm… Poland and the Slovak Republic would have a dummy variable for recently rid of the Soviet Empire. There is such a thing as the catch-up effect.”
— Twenty years ago… and if we’re giving caveats, then we should factor in the post Cold War peacetime dividends into the the 90s… the rebuilding of Europe and Japan in the late 40s and early 50s… and any post War recessions. However, I was going on the reported GDP statistics.
Again, my original post wanted to know if raising the top tax rates increased revenue from the richest classes or had the opposite effect (either your opinion or thru facts). In other words, why would a policy that removed capital from the economy have a positive growth effect? That’s all. I made no judgements about the productivity of the wealthy, the morals or ethics of tax policy, nor income inequality.
(A cheap shot about Monaco, yes… but that’s because of the famous Europeans who purchase a residence there to escape taxes in their home country. An unfair commentary maybe.)
If you’re right coberly, then we should never find out why there are correlations. Don’t worry about cause and effect and maybe getting the two mixed up.
Again, what will be the effect of increasing the top tax rate be to the burden of the wealthiest segments of the nation? Also, what was the effective tax rate for someone in the top 1% of income in the high tax years, like 1936 (top tax rate was 79%)?
This doesn’t seem to include all the other taxes. Explicit and implicit taxes are raised on average every year (sales taxes, property taxes, state income, fees, tariffs, luxury taxes, etc). This go up and seldom come down and should be factored into any analysis of the effects of taxation on growth. I think a per-capita look at GDP would be more accurate as well.
To really get at the connection macro-economic factors would need to be excluded. For example oil cost and researves are impactful on the GDP however, have no connection to taxes. Similarly wars, natural disasters, and other systemmatic factors influence our prosperity but lack a real connection to the effects of taxation. Consider a major war like WW1 & WW2 – production (GDP) drops, tax rates go up, but the base isn’t there to tax so they only adjust slightly. Post war tax rates go up to pay for the prior expenditure and GDP skyrockets as the nation returns to work. There is a correlation but not a causation. This trend follows any significant expenditure as the nation has to get back to work to pay for the event that has transpired.
Robust sustained growth benefits from a low enough average tax rate on the wealthy that they have incentive to run and/or fund a business with a high enough marginal tax rate that they avoid taxation with deductible or depreciable expenditures that grow the value of their business.
If the top rate or capital gains rate are too low the wealthy pull equity out of businesses as personal income and the growth effect of the marginal dollar pulled out is smaller than the growth effect of the marginal dollar left in and likely spent on wages, plant and equipment, R&D or marketing.
The trick to a low average tax rate with a high marginal tax rate is high brackets.
The wealthy are paying too much tax and the problem got worse with the Bush tax cuts. Marginal rates need to be raised so they take less income and put more money into growing a business where increasing value can be sheltered from tax until the business is sold for a capital gain or it gets taxed in the estate.