Mitt Romney’s Taxes

Mitt Romney says his tax rate is “probably around 15%”. It’s not clear what he means by that (marginal rate? average rate? federal rate? federal-plus-state-plus-local rate?) but the New York Times is quick to point out that he’s a beneficiary of the “fact” that investment income is taxed at a much lower rate than wages and salaries, leaving him with a lower percentage tax burden than the working-stiffs he employs.

For at least the eighth time on this blog, I want to point out that this widely believed “fact” is not true.

To understand Mitt Romney’s tax burden, you have to compare him to his doppelganger Timm Romney, who lives on a planet with no taxes. In the year (say) 2000, Mitt and Timm both earned (say) a million dollars. Timm invested his million dollars, saw it double over the past decade or so, and cashed out his investment this year, leaving him with two million dollars. Mitt, by contrast, paid 35% tax in 2000, leaving him with $650,000. He invested it, saw it double, and cashed out last year, paying 15% tax on the $650,000 capital gain. That leaves him $1,202,500, which is about 60% of what Timm’s got. In other words, the tax system costs Mitt almost 40% of his income.

By contrast, people on our planet without investment income collect their wages, pay 35% in taxes, and spend what’s left. The tax system costs them 35%, while it costs Mitt almost 40%. In other words, people with investment income bear a higher tax burden, as a percentage of their income, than anyone else — and that’s before you even start accounting for the taxes on dividends, interest, corporate income and inheritance.

It’s true that there are some hedge fund managers out there who manage to game the system by disguising their wages as capital gains and thereby avoiding the wage tax altogether. That in no way undermines the main point.

Share/Save

108 Responses to “Mitt Romney’s Taxes”


  1. 1 1 HispanicPundit

    What about CEO’s getting paid in stocks instead of wages? Or stock options, etc.

  2. 2 2 Dmitry Kolyakov

    Professor Landsburg,

    Are you not forgetting to address some objections to this view of capital gains taxation?

    Just to sum up some of the previous discussion: the investment income (in the form of capital gains) is formed of time value of money, risk premium of the asset selected and premium for investment skill (also known as risk-free rate, beta and alpha). Which of these elements do in your view qualify for exemption from tax or some reduced tax rate?

    So far I am under the impression that you are in favor of giving an exemption on all these elements – am I right? I so far can see reasons to exempt only the risk-free element, however…

    As a former private equity professional Mr Romney is most likely receiving most of this income from his carried interest as a form of deferred compensation for successfully managing other people’s money and some of it for successfully managing his own money. Isn’t it in reality mostly a compensation for labor that is simply packaged in a tax-optimizing way?

    I am not sure why exactly the fact you mentioned is false – investment income as such is indeed taxed at a lower rate – I suppose that is what the tax code says. I guess you are effectively changing the definition of “taxed at a certain level”, so that still does not make the newspaper’s stated fact “not true”.

    The more profound view you are offering still is very useful and merits discussion. I guess you are somehow omitting the fact that most of our labor incomes (if none of us are serial paid organ donors, that is) are a return on past investment in education and acquiring new skills. Some of that investment is of course lost opportunity, but much is paid out of our taxed income. (There are sure some tax breaks for educational expenses in various jurisdictions, but certain other types of investment have their tax incentives too, so that does not change the big picture.)

    So let’s go back to your example: let’s assume a third member of their team, Mitimm who lives on Earth not far from the original Mitt and who also earned 1m in 2000. But being not as well-educated as Mitt, he decides (after paying his tax) to invest in his education with a hope of increasing his earning potential. So he pays the remaining 650K to a prominent professor who teaches him mysteries of economics. This investment pays off till his retirement in ten years when the present value of his surplus income before taxes reaches double the initial investment (assuming that education is just as good as any other investment). But as he gets taxed at 35% on his surplus labor income, in the end he is left with 650*2*(1-0.35)=845K. Compare that with Mitt’s 1,202,500.

  3. 3 3 ThomasBayes

    Steve: In your opinion, why do people have so much trouble accepting this view?

  4. 4 4 Matthew

    As an aside, it’s not clear to me that the performance fees earned from managing money should be viewed as wages rather investment income. The general partner is receiving a piece of the upside on an investment. Somebody paid the wage tax on the money that went into the pot. Additionally, the performance fees aren’t guaranteed.

  5. 5 5 Dmitry Kolyakov

    I guess the GP’s performance fees are viewed as compensation because they are not a result of his own investment and therefore are not his investment income. They directly compensate him for successfully performing his (or her) professional duties and taking a certain amount of risk – just like any other job compensation with a variable component.

    Should a real estate agent’s fee be treated not as a wage for taxation – he is getting a piece of the money his client is getting (including possible upside) after all? A lawyer’s contingency fee? What about performance bonuses – they are not guaranteed either…

  6. 6 6 Matthew

    I hear all that, but how can you change the way the profits are taxed from the perspective of the general partners without changing how they are taxed from the perspective of the limited partners?

  7. 7 7 Adam

    Taxing investment income is tricky because it essentially taxes inflation as well as real income, but once you separate the rates to account for taxing inflation, people with enough income to make it worthwhile will do their best to structure earned income as investment income (which is fairly easy under current law).

    Far better would be to remove inflation from the taxable gains and tax all income more equally.

  8. 8 8 Neil

    If all wealth consisted of accumulated earnings on which tax has been paid, this argument is correct. But much wealth, probably including Romney’s, does not consist of accumulated taxed earnings. Does anyone here think that Bill Gates’ $50 billion is accumulated taxed earnings?

  9. 9 9 Alan Wexelblat

    What I don’t understand is why Romney is so reluctant to point out that several of his opponents’ tax reform plans would in fact drop his rate to 0%. Why are they all het up about 15% when they want it to be zero?

  10. 10 10 Brian

    @Thomas – bad reasoning, I guess. Steve may have to go through it another 8 times before it sinks in for me. My simple brain says that Mitt made 1.65M in income (1M in wages, .65 in investment) and kept 73%. If someone else made 1.65M in wages, they would keep 65%. Same income, but Mitt pays less in taxes.

  11. 11 11 Ken B

    @Thomas bayes: Math is hard. Saying ’15% that’s so low’ and feeling righteous is easy.

    Speaking of the math is hard part, you read Godel’s original proof yet? :>

  12. 12 12 Greg

    If we repeat this process over N periods — that is, Mitt & Timm repeatedly reinvest their earnings — wouldn’t the initial income tax become increasingly inconsequential as N -> infinity?

  13. 13 13 Dmitry Kolyakov

    @Ken B: Math indeed can be quite hard, and i guess that what makes contribution of people like Prof. Landsburg who can put proper economics in plain language particularly valuable!

  14. 14 14 Jon Shea

    For about a year now I’ve had on my todo list: “Ask Landsburg about ‘carried interest’. It’s bullshit, right? I’m pretty sure it should count as income, not investment”. Thanks for at least hinting that my instinct was correct on that issue.

    More generally, how do we decide whether a given income source should count as labor or investment? For example, image I spend $10 to buy a lemonade stand. I spend $2 a day on lemons, and I make $1000 a day in profit. Of that $1000, what counts as labor and what counts as investment? (As I understand it, the IRS requires that you pay yourself a market salary for your labor role, and the rest you can treat as corporate profit. But that solution seems distastefully un-rigorous.)

    Finally, a little bit of Googling suggests that the majority of Romney’s wealth actually is from his days as a hedge fund manager, and would have been gamed as carried interest rather than income. So, it seems that there might be grounds for complaining about how much tax Romney has paid that even you might agree with.

  15. 15 15 Ken

    HispanicPundit,

    Stock options must be purchased. They aren’t “given” to anyone. A stock option means you have the option to purchase a stock at price $X for a given amount of time, regardless of what the current market value is at the time of purchase. For example, imagine Mitt is given 10,000 stock options, which can be optioned at $65/share. He can option these stocks between the dates 01/01/2012 and 12/31/2012, after which he no longer can option the stocks, ever.

    On March 7, 2012, the stock price is $50. Mitt obviously won’t option them because even though market price is $50 for the stocks, Mitt has to purchase them for $65/share.

    On August 23, 2012, the stock price is $66. Mitt might by option the stock options, then.

    Either way, no matter what the market price is for the stock options, Mitt pays $65/share, which is $650,000, which is what he has left after his income was taxed at 35%.

    Additionally, there are different ways to issue stock option, which determine whether any gains are taxed as income or not. By you assuming they will not seems like you’re not interested in the actual facts of how the gains are taxed.

    Not only do you seem to not understand the point of Steven’s post, but you seem to not understand how various incentive packages work and how they are taxed.

  16. 16 16 ConnGator

    Regarding Bill Gates’s $50 billion, most of it is in Microsoft stock. True, it has not been taxed yet but it also cannot be spent. Once Mr. Gates sells the stock, then it will be taxed at the capital gains rate and then he can spend it.

    Or, more likely, he will give away the appreciated stock and pay no taxes (and spend no money.)

  17. 17 17 Jeremy N

    I hesitate to say I disagree with your argument for fear that I have overlooked something.

    But it seems that you’re saying that Mitt’s tax burden includes investment returns he would have realized had the capital not been confiscated by the government.

    Timm’s income includes the original $1M and the $1M return on investment.

    Mitt’s income includes the original $1M and a $650K return on investment. He pays 35% of the 1M and 15% on the 650K. So he pays ~27% of his income but has 40% less than Timm. The difference is the return on the money that was taxed away.

    But that money never actually existed. (By the way you have to assume that Timm’s $1M investment yields the same return as Mitt’s $650K). It would be like saying that the wage tax is more than x% because the wage rate doesn’t include the consumer surplus I could accrue by the extra consumption of those tax dollars I send to the government.

    What’s the answer to that?

    Like most everything else, the debate boils down to a semantic one. That $350K investment return wasn’t paid in taxes to anyone, it never existed. Should it be considered part of the tax burden? Should we make up a new, confusing term for that kind of effective tax rate?

  18. 18 18 Will A

    Professor Landsburg:

    there are some hedge fund managers out there who manage to game the system by disguising their wages as capital gain

    I’m not sure what you mean by “game the system”. I think these hedge fund managers are taking advantage of tax laws they lobbied for. They aren’t disguising their wages in the way that Al Capone cooked his books. They aren’t avoiding the taxes they owe and then paying less as a plea bargain with the government. They are following the existing tax laws.

    Middle class home owners lobbied for tax breaks on home mortgage interest. Would you say that I’m gaming the system when I deduct my home mortgage interest?

    If your point is that anyone who is successful at lobbying congress for beneficial tax legislation or benefits from those lobbying efforts “games the system”, then why single out hedge fund managers?

  19. 19 19 Neil

    ConnGator,

    But that is the point. Without the corporate income and capital gains taxes that Steve claims are unnecessary double taxation, Gates’ wealth would go untaxed, at least by the Federal government since it does not levy a national sales tax.

  20. 20 20 Thomas Bayes

    @JeremyN: There are probably better examples, but here’s a way I think about this. Suppose you and I make a deal to split the payout if one of us wins a lottery. You win $1M and honor our deal by giving me $500K. (Ignore taxes for this.) You invest most of yours and I spend all of mine. Ten years later I have nothing left, but your investments have returned an additional $300K. What would you say if I said our deal compels you to give me $150K of your investment return? Would it seem like I was ‘taxing’ you twice on the same money?

    @KenB: No, I haven’t completed that yet. By the way, I like to tell people that there is nothing uncertain about the uncertainty principle. Is it also fair to say that there is nothing incomplete about the incompleteness theorems?

  21. 21 21 KS

    Dr. Landsburg–

    I have been reading your books and blog for years now, and the one issue where you fall repeatedly on your face is taxes.

    I have no formal training in economics, and even I know the law of diminishing returns — one of the basic laws of economics — favors the use of taxation as redistributive policy. Why? Because the law states the more you have of X, the less you value additional X. Conversely, the less you have of X, the more you value additional X. Hence, say person 1 has X1 money and person 2 has X2 money, where X1 > X2. If we redistribute y money in forms of taxes, such that person 1 now has (X1-y) and person 2 has (X2+y), we overall increase total utility.

    Not ONCE in your entire discussion(s) regarding “the man who can’t be taxed” did this even occur to you, which frightens me. Sure, the government cannot increase total number of resources by taking from one and giving to another. It can, however, increase total utility by doing so. The fact that an accomplished person such as yourself fails to even perceive this — due mostly to your preformed political beliefs — scares me as to the prospects of the economics “dismal science” profession in general.

    And to preempt your entirely obvious rebuttal, might I add that: (1) the greatest periods of job growth in our country in the past 100 years have increased with marginal tax rates that would make economists such as yourselves consider suicide, and (2) the effect of marginal tax rates on work incentives has been vastly, vastly magnified (yes, 80% is too high, and no, there is no difference between 33% and 39%).

    Sincerely,
    An academic in a real profession (medicine)

  22. 22 22 Will A

    Neil,

    Gates will probably never pay taxes on his wealth (at least while he is alive). However assuming we have an estate tax when Gates dies, the government will be able to build drones with part of his wealth eventually.

    I’m curious what your concern is. Is your concern that Bill Gates is getting away with something? Are you concerned the government won’t get to partake in his wealth?

  23. 23 23 Will A

    Correction:

    Gates will probably never pay taxes on the majority of his wealth

  24. 24 24 Neil

    Will A,

    Steve is claiming that capital and estate taxes and the like are unnecessary double taxes. I am pointing out that Gates’ vast wealth has not yet been taxed as income (except by the MSFT corporate tax), so without corporate, capital and estate taxes, Gates’ lifetime income would go untaxed. Yes, I want the government to partake of his lifetime income the same way it has partaken of mine.

  25. 25 25 CC

    KS- Steven Landsburg isn’t taking a position on more redistribution or less. He’s just saying that for whatever level of redistribution you want, there are more and less efficient ways to accomplish that. Specifically, he is claiming that taxing capital gains leads to less efficient outcomes *independent* of how much redistribution takes place.

  26. 26 26 Will A

    Neil,

    In the above argument he is not using the terms “double taxes/taxation” or “unnecessary”.

    In my opinion the double taxation argument is silly on both sides. The money I used to buy my Microsoft XBox has been taxed twice (income and sales tax). Microsoft has to pay corporate taxes on what I give then. Bill Gates then has to pay capital taxes. So in essence Bill Gates has experienced quadruple taxation (I don’t know how he survives).

    I personally would like a tax system that encourages a business owner to invest in his company (the tax on income is much higher than the tax on selling a business).

    I personally believe that the way a tax code is written can have a big impact on a society and I’m not convinced that taxing all income exactly the same for everyone is the best system.

  27. 27 27 Jeremy N

    Thomas Bayes,

    I believe your scenario doesn’t get to Dr. Landsburg’s point. Let me modify it to better reflect my interpretation of the situation.

    You and I make a deal to split the payout if one of us wins the lottery AND give the other one 25% of any money earned from investment of the reward. I win $1M and honor the deal by giving you $500K. I invest most of mine and you spend all of yours. Ten years later, you have nothing left, but my investments have returned an additional $300K. Of which, I owe you $75K as we agreed.

    Now Dr. Landsburg says I have only 45%* of what I would have had if we had made no arrangement at all. Therefore my tax burden is 55% ($875K), but I actually only paid you $575K. He’s saying the remaining $300K (which wasn’t ever earned because you spent the $500K instead of investing it), should be considered a tax burden, whereas I think that’s a shaky definition since I didn’t technically pay it to anyone and you didn’t collect it.

    *($500K+0.75*$300K) / ($500K+$300K + $500K+$300K)

  28. 28 28 Andrew

    Jeremy N –

    The “unpaid” and “uncollected” tax is simply the unseen. The next question would be to ask if there was a greater return by government on this unseen than what Mitt would have gotten.

    If the average male lives to be 80 and you are gunned down at 40, would you agree that you “lost” 40 years of your life?

  29. 29 29 Floccina

    One also should consider the incidence of the corporate tax.

  30. 30 30 TjD__

    Greetings,

    I think I found the flaw in your model. In order to double your money over 10 years you need every year 7% gains which is completely off, there is no investment that has done this over the last 10 years. If you take a more real percentage like 3% ( even that is hard nowadays ) your model falls apart.

    With 3% interest spaceMitt made 1.343.916 and earthMitt has made 873.545, so earthMitt paid in this case … 35%.

    I am disturbed by the round 35%, not sure if Excel is messing with me, feel free to point out any mistakes.

    T.

  31. 31 31 TjD__

    One more comment ( I would love to be able to edit my comments btw ).

    As I understand it, it is not only hedge fund managers but every owner of an LLC has that flexibility. Apparently there are 3.8 million of those out there..

    T.

  32. 32 32 Todd

    I think Thomas Bayes’ illustration of the lottery split agreement does a good job of demonstrating the reason that Steve and other economists consider taxes on capital gains to be a bad idea.

    In this scenario, if the saver were obligated to share his gains with the spender, then the spender gets the benefits of deferred consumption without bearing the cost of deferring his consumption. The incentive, then, is to save less. So it is with capital gains taxation.

  33. 33 33 Rational Animal

    Mitt, by contrast, paid 35% tax in 2000

    Citation needed.

  34. 34 34 KS

    CC– You are ignoring the fact that only rich people (basically) pay the capital gains tax. Therefore a higher capital gains tax is necessarily more re-distributive and better for society, even though in theory in may not be the most efficient if all else is equal.

    Everyone else– The effect of taxes on work incentives has been vastly overstated by (1) those who don’t want to pay higher taxes themselves and (2) those with a political agenda. As Warren Buffet has put it, in his 50 years of investing, he has NEVER known an investor not make an investment based on a certain taxation of the potential gain involved.

    Everyone always assumes a linear relationship between taxation and change in behavior. If I’m being taxed at 33%, and it gets bumped up to 39%, it is assumed I will work less hard. Oh, really? Can’t the effect be flat or fall minimally until a certain threshold, then dropoff dramatically? How many true relationships in the social sciences are linear?

  35. 35 35 Ken B

    @Thomas Bayes: My first thought was, no there isn’t. Then I thought, what if there happens to be a true but unprovable statement which is itself an ‘incompleteness’ result? So now I am uncertain.

  36. 36 36 Prometheus

    A considerable portion of Romney’s income comes from a retirement deal with Bain Capital that continues to pay him a small share of the firm’s profits. The wonky term for this cool stream of money is “carried interest” — the share of investor gains “carried” by the private equity or hedge fund manager.

    Capital gains are income from your own investments. Carried interest is income from other people’s investments. If you manage money for a mutual fund or a public company, you pay regular income taxes. Do it for a private fund, and you pay capital gains. That’s backward. That special treatment is what people object to.

  37. 37 37 Super-Fly

    Although the Mitts and hedge fund managers of the world are few, are you saying they don’t actually get taxed? (At least, not as much as in the examples)

    I think part (or all) of the reason that people are angry is that they only see Mitt’s taxes right now. He gets the benefits of govt even though he’s not currently paying anything. Would a consumption tax/sales tax would eliminate that problem because you pay it as you spend it?

  38. 38 38 Ben

    I’m looking forward to someone’s response to JonShea. It seems like the ideal answer to that question is that the corporate tax rate should be equal to the individual tax rate and the capital gains tax rate should still be 0%. In that case, it doesn’t matter how the revenue is divided up into individual and corporate earnings; it gets taxed the same way. But Steve prefers a corporate tax rate of 0% (in Ranking the Tax Plans), so I’m extra curious to read his response.

    Also, I thought this post captured the capital gains idea well:
    http://www.themoneyillusion.com/?p=7091

  39. 39 39 Andy B

    SL: “people with investment income bear a higher tax burden, as a percentage of their income, than anyone else”

    I assume “income” refers to nominal income. But what if we were to define income as that which purchases a certain amount of stuff at a specific time (i.e. translate money into real things).

    So in Steve’s example you can spend $1,000,000 on 2000 stuff or you can forgo that and instead invest your $1,000,000 and purchase the equivalent amount of stuff, adjusted for inflation, in 2012. Assuming average annual inflation of say 2.5%, then you would need $1,344,889 to do that.

    But your investment returns did better. You actually have $2,000,000 which leaves you $655,111 ahead which you can use to purchase extra stuff (in 2012 dollars). So the all in tax rate on your income in constant 2000 dollars stuff is ($350,000 + $111,533) / $1,487,112 = 31%. This is less than the 35% the person with no investment income pays on their earnings.

    So if we think of income as what you can actually purchase, then as long as my investment (i.e. forgone consumption) return exceeds inflation then I have excess investment earnings which are taxed at a lower i.e. 15% rate than other earnings.

  40. 40 40 CC

    KS, you wrote:

    “CC– You are ignoring the fact that only rich people (basically) pay the capital gains tax. Therefore a higher capital gains tax is necessarily more re-distributive and better for society”

    SL’s entire point is that he is not stating any opinion on how redistributive the tax system should be. You might argue that we should have a more graduated income tax or a less graduated one. He’s simply arguing that it’s inefficient to have a higher tax rate on people who save their income and spend it at a later time than you do on people who spend it right away.

    <>

    I work in “investing”, and after-tax payoffs play a role in everything I do. Incentives matter. I don’t know why people believe otherwise.

    <>

    No one is assuming it’s linear. But I also doubt it’s sudden in exactly the way you’re describing. In any case, you’re making the standard fallacy of saying “Well I don’t know anyone who works less b/c of taxes… there’s an income effect AND a substitution effect after all.” Greg Mankiw explained on his blog a while back why this reasoning is off. You generally want to look at just the substitution effect. He explained it better than I ever could.

  41. 41 41 CC

    Sigh… why can’t we edit posts? Anyway, here’s what I tried to post:

    KS, you wrote:

    “CC– You are ignoring the fact that only rich people (basically) pay the capital gains tax. Therefore a higher capital gains tax is necessarily more re-distributive and better for society”

    SL’s entire point is that he is not stating any opinion on how redistributive the tax system should be. You might argue that we should have a more graduated income tax or a less graduated one. He’s simply arguing that it’s inefficient to have a higher tax rate on people who save their income and spend it at a later time than you do on people who spend it right away.

    “Everyone else– The effect of taxes on work incentives has been vastly overstated by (1) those who don’t want to pay higher taxes themselves and (2) those with a political agenda. As Warren Buffet has put it, in his 50 years of investing, he has NEVER known an investor not make an investment based on a certain taxation of the potential gain involved. ”

    I work in “investing”, and after-tax payoffs play a role in everything I do. Incentives matter. I don’t know why people believe otherwise.

    “Everyone always assumes a linear relationship between taxation and change in behavior. If I’m being taxed at 33%, and it gets bumped up to 39%, it is assumed I will work less hard. Oh, really? Can’t the effect be flat or fall minimally until a certain threshold, then dropoff dramatically? How many true relationships in the social sciences are linear?”

    No one is assuming it’s linear. But I also doubt it’s sudden in exactly the way you’re describing. In any case, you’re making the standard fallacy of saying “Well I don’t know anyone who works less b/c of taxes… there’s an income effect AND a substitution effect after all.” Greg Mankiw explained on his blog a while back why this reasoning is off. You generally want to look at just the substitution effect. He explained it better than I ever could.

  42. 42 42 KS

    “SL’s entire point is that he is not stating any opinion on how redistributive the tax system should be.”

    Generally, those who argue for lower capital gains taxes, and those who favor a less progressive taxation system, tend to overlap strongly. I wasn’t commenting on this point per se as much as SL’s views on taxation in general, and his lack of understanding as to the benefits of a progressive taxation policy. I personally view capital gains as a form of selective taxation on the rich and super-rich, which is why I favor it.

    “I work in “investing”, and after-tax payoffs play a role in everything I do. Incentives matter. I don’t know why people believe otherwise.”

    The question isn’t, do incentives matter? The question is, is the decrease in work incentives induced by a change in tax policy offset by the benefits of redistribution? Given all the hyperbole about ANY tax increases you hear today, you would think a plan where the marginal rate on top earners was as high as 70% would induce economic catastrophe. Yet, it happened to coincide with the greatest period of economic expansion in US history (post-WW2 boom). I admit that doesn’t prove causation. It also doesn’t bode well for the theory that high tax kills job creators.

    Regardless of political views, hopefully we can agree that economics is as legitimate as a rigorous empiricist doctrine as scientology. And, coming from an actual science where evidence actually changes paradigms, it’s fun to point out every once in a while.

  43. 43 43 Dmitry Kolyakov

    KS: Could you please kindly remind what were the exact benefits of involuntary redistribution (such as through taxation) you were referring to?

  44. 44 44 ThomasBayes

    I know this is a bit off topic, but can anyone explain the difference in message that is conveyed by the data Paul Krugman shows here:

    http://krugman.blogs.nytimes.com/2012/01/18/check-out-their-low-low-taxes/

    and the CBO table here:

    http://www.cbo.gov/publications/collections/tax/2009/effective_rates.pdf

    Krugman’s graphic parses tax data to make it look like households that make high incomes pay 20% or less in federal tax. The CBO tables show that the Top 1% pays over 30% in total federal tax, the top 5% pays 29%, and the the top 10% pays 27.5%. Krugman’s table makes it look like you pay a lower rate if you make more. The CBO table shows the opposite. Is this just an issue of parsing the data, or is something fishing going on?

  45. 45 45 Floccina

    Inflation is also an issue.

  46. 46 46 KS

    “KS: Could you please kindly remind what were the exact benefits of involuntary redistribution (such as through taxation) you were referring to?”

    Sure. Or, you could refer to my initial post. Basically, the law of diminishing returns states that the more someone has of X, the smaller the increase in marginal utility for the next X they receive, and conversely. Thus, total utility will increase if you redistribute from those who have more to those who have less.

    Suppose we live in a country which only has 10 people. 1 of these 10 people owns 10 homes, while the other 9 are homeless. Even economists could probably figure out being homeless sucks. Now say the 9 people form a government and pass into law that the one person has to give up his 9 homes so that 1 go to each of them. Of course, that one person is going to be less happy by losing 9 homes, but he’ll still have one left. Each of the other 9 is going to experience an increase in happiness because they all go from being homeless to having one home. This “government” did not increase resources, it just shifted them around. But overall utility (or happiness) WILL increase, because the decrease in happiness felt by the one person is more than offset by the increase in happiness each person has going from zero to one homes.

    I am NOT advocating for a communist-type system where redistribution occurs until all wages are equal, because I do believe that with enough redistribution you significantly affect work incentives. But if the amount you redistribute is small in proportion to the initial amount the person has, then work incentives are minimally affected.

    In his entire discussion(s) on the man who can’t be taxed, not ONCE did this occur to SL. He continued to insist that the government cannot create resources through taxing, it can just shift them around. But that’s the point! You give to those who have less and total utility will increase.

  47. 47 47 nobody.really

    I work in “investing”, and after-tax payoffs play a role in everything I do. Incentives matter. I don’t know why people believe otherwise.

    EXACTLY! After-tax payoffs are what drives productive behavior – including the choice to invest your time in labor or in leisure.

    Republicans praise Reagan’s tax structure for unleashing a surge in productivity. So I’d be interested in seeing a tax structure designed to ensure that the top 10% receive the same (per capita, inflation-adjusted) after-tax payoff as they received under Reagan.

    Not the same tax RATE. The same after-tax payoff. That is, the top 10% would end up with the same number of dollars (adjusted for inflation and population growth) as they received in the 1980s. After all, if that number of after-tax dollars was sufficient to motivate productive behavior back then, why wouldn’t it be sufficient to motivate productive behavior now?

    (Ok, let me return to the original topic just long enough to ensure that my question bears no relationship to the original topic: I’d have no objection to restructuring the tax code to substitute some other taxes for taxes on capital, provided the new structure produced the same after-tax payoff for the top 10%.)

  48. 48 48 Ken B

    “After all, if that number of after-tax dollars was sufficient to motivate productive behavior back then, why wouldn’t it be sufficient to motivate productive behavior now? ”
    Yup, nobody.really should be paid now what he earned on his paper route in grade 4. After all if that number of after-tax dollars …

  49. 49 49 Jeremy N

    Thomas Bayes:

    The only difference I can find, and I think it accounts for the discrepancy, is that Krugman’s chart omits the effective corporate tax. I don’t know enough about taxes to know why it’s included.

  50. 50 50 Steve Landsburg

    KS:

    , the law of diminishing returns states that the more someone has of X, the smaller the increase in marginal utility for the next X they receive, and conversely.

    Actually, the law of diminishing returns describes what happens if you have a physical production process with multiple inputs, and you increase one of those inputs without increasing the others. The statement you are making a) has nothing whatsoever to do with the law of diminishing returns and b) relies on a whole lot of philosophical baggage (i.e. not just cardinal utility, but cardinal utility that can be compared across individuals) that has been pretty hard to swallow for most of the people who have actually thought hard about this. (I happen, incidentally, to be more receptive to that philosophical baggage than are many economists. But those of us who are receptive to it ought to recognize it for the baggage that it is, not present it as some sort of fundamental uncontroversial principle.)

    In his entire discussion(s) on the man who can’t be taxed, not ONCE did this occur to SL. He continued to insist that the government cannot create resources through taxing, it can just shift them around. But that’s the point!

    No, actually it’s got absolutely nothing to do with the point. The point there was that we had a guy who consumed (by assumption) zero resources; therefore it was impossible to transfer resource consumption away from him. When the whole point is that you can’t effect a redistribution, the desirability of that redistribution becomes quite irrelevant. You’re complaining that I failed to raise and address a question that was irrelevant in that context.

    Everyone always assumes a linear relationship between taxation and change in behavior.

    Please name one person who has ever assumed this.

    As others have observed, you are off on a rant about redistribution that is quite entirely irrelevant to the post at hand (and, if possible, even more irrelevant to the “man who can’t be taxed” post you keep referencing). I’m sure there will be many other threads on this blog where your comments will be valuable (at least if you take the time to state them more carefully, acknowledge your assumptions, and quote others accurately), but this is not one of them.

  51. 51 51 nobody.really

    “After all, if that number of after-tax dollars was sufficient to motivate productive behavior back then, why wouldn’t it be sufficient to motivate productive behavior now? ”
    Yup, nobody.really should be paid now what he earned on his paper route in grade 4. After all if that number of after-tax dollars …

    Ha!

    Of course, when I was in 4th grade I had a pretty impressive after-tax pay-off: I had housing, food, education, transportation, health care, in-home servants called Mom and Dad, etc. Given decades of real wage stagnation, I would not be surprised to learn that I do have a comparable after-tax payoff as I had in 4th grade – or less.

    I strongly doubt that the same is true of people in the top 10% of the income distribution. Why should we imagine that these people require vastly more after-tax payoff than comparable people in the 1980s?

  52. 52 52 Steve Landsburg

    nobody.really:

    The reason we usually believe that a reduction in compensation will lead to less work effort (by some, not by all!) is that if this were false, competitive pressures would already have reduced the compensation level.

  53. 53 53 Ken B

    @Nobody.really: Why the 80s, why not the 60s and the JFK tax cut? Because you aren’t making an argument you are trying to score a point — “oh burn!!!!!” — by thinking you are citing Reagan as a weapon against conservatives and republicans. But re-phrase your point in terms of JFK and the early 60s — or even earlier — and you will quickly see how illogical it is. I merely hightlighted one of the points of illogic.

  54. 54 54 nobody.really

    The reason we usually believe that a reduction in compensation will lead to less work effort (by some, not by all!) is that if this were false, competitive pressures would already have reduced the compensation level.

    Last I recall, price had two functions: production and distribution. Firms may feel the need to increase compensation to lure talent away from rival firms. Taxation may then reduce the amount of after-tax payoff without reducing production. Sure, at some point of taxation I expect that Albert Pujols would opt to chuck his $240 million contract to produce hits for the Los Angeles Angels and consume more leisure instead. But at what level?

    For whatever reason, people seemed willing to do the work required to enter the to 10% of the income scale in the 1980s for much less real compensation than people “require” today. Why is that?

    Why the 80s, why not the 60s and the JFK tax cut?

    Fine. Why not?

    Look, I’m not just trying to tweak Republicans. (That’s merely a side benefit.) My point is this: NOBODY in all of history has received the after-tax payoff of today’s rich people. To argue that reducing their after-tax payoff would eliminate their incentive to produce would seem to imply that NOBODY in all of history has had an incentive to produce. I find this argument implausible.

  55. 55 55 Super-Fly

    So, one more time. Can we say that people who are paid in stock or some other ‘non-wage’ are, in fact, paying less?

    Even though it’s probably only a handful of individuals, most people would regard that as unfair. I agree though that cap gains/dividends taxes are bad (well, inefficient at least). Is there another way to target those individuals?

  56. 56 56 Henri Hein

    I have some problems with this. As others have pointed out, there are ways other than wages to make your first million (in fact, I would think that everybody who manages to stash away their first million do so through additional earnings beyond their wages).

    Taxing wages only can lead to the same kind of sequence taxes. If I make $1000 and pay $500 in taxes and then spend $100 on a gardener who then pays $50 in taxes, would those $50 also be considered a double taxation?

    If the concern is a taxation of both investment income and wages, would it be considered a solution to only tax investment income and not tax wages at all? (The artifacts of the current tax code distinguishing between interest income vs. qualified dividends vs. unqualified dividends vs. capital gains confuses the discussion somewhat. Here I lump it together as investment income.)

  57. 57 57 Ken Arromdee

    “My simple brain says that Mitt made 1.65M in income (1M in wages, .65 in investment) and kept 73%. If someone else made 1.65M in wages, they would keep 65%. Same income, but Mitt pays less in taxes.”

    The .65M that Mitt made from investment depends on his ability to put money in the investment. Taxes not only reduce his investment income directly, they also reduce his investment income by virtue of the fact that he has less money to invest in the first place.

    The .65M more that the wage earner made is not affected in the same way.

    In the absence of taxes, the 1.65M wage earner would still be making 1.65M, but Mitt would be making 2M. You’re trying to treat the wage earner’s 1.65M and Mitt’s 1.65M as equivalent and they are not–one partly includes the effect of taxes and the other doesn’t.

  58. 58 58 Ken Arromdee

    “So, one more time. Can we say that people who are paid in stock or some other ‘non-wage’ are, in fact, paying less?”

    Assuming you mean “if someone is paid in stock, and only pays 15% taxes on the stock, are they paying less taxes”, then yes. Mitt is making the investments using after-tax income. Someone directly being paid in stock is getting the stock as before-tax income.

  59. 59 59 iceman

    @nobody.really: Somebody here mentioned that there are not just substitution but also income effects. Greater wealth makes increased leisure a more viable option. The more Pujols has stashed away the easier it also is for him one day to decide “you know I really don’t HAVE to place my head inches away from 100-MPH fastballs anymore.”

  60. 60 60 Neil

    The darndest thing is that it is quite possible to design a tax system that would not tax capital income, as Steve wants, but would ensure that all, or at least most, income is taxed, as everyone wants. The Hall-Rabushka flat tax which taxes earnings and all business income at the same rate does this.

  61. 61 61 Henri Hein

    “As Warren Buffet has put it, in his 50 years of investing, he has NEVER known an investor not make an investment based on a certain taxation of the potential gain involved”

    Buffett has said some amazing things, but if he really said this, that would take the cake. I have not read a single investment book that does not mention tips on how to optimize for taxation. Most investment articles will at least mention taxation optimization.

  62. 62 62 nobody.really

    Why do people in the to 10% of income demand a vastly larger after-tax payoff to be productive than did people in earlier eras?

    Somebody here mentioned that there are not just substitution but also income effects. Greater wealth makes increased leisure a more viable option. The more Pujols has stashed away the easier it also is for him one day to decide “you know I really don’t HAVE to place my head inches away from 100-MPH fastballs anymore.”

    A fine hypothesis! The supply curve for labor is backwards-bending: as we get richer, we consume more of pretty much everything – including leisure. So we have to offer people even MORE compensation to induce them to sacrifice their leisure and keep producing.

    Thus, the difference between prior eras and today is that the people in the top 10% are VASTLY RICHER now than in bygone days. Precisely BECAUSE they’re so rich, we need to offer them more compensation to keep them producing.

    This strikes me as plausible. But one obvious strategy for dealing with this dynamic is to tax people in the upper income brackets so that they don’t become so rich – and hence unproductive — in the first place. The lesson of the backwards-bending labor curve is that taxes can INCREASE the incentive to work.

    Admittedly, designing a society-wide tax to optimize the behavior of millions of people with differing utility functions is a challenge. But it’s a challenge we much confront no matter what tax code we design. I suspect a progressive consumption tax might have the most salutary – ok, the least pernicious – effect.

  63. 63 63 Ken B

    Nobody.really makes a good point (finally):
    Thus, the difference between prior eras and today is that the people in the top 10% are VASTLY RICHER now than in bygone days. Precisely BECAUSE they’re so rich, we need to offer them more compensation to keep them producing.

    I would add one important point though:

    Thus, the difference between prior eras and today is that the people in the bottom 90% are VASTLY RICHER now than in bygone days. Precisely BECAUSE they’re so rich, we need to offer them more compensation to keep them producing.

  64. 64 64 Will A

    nobody.really:

    Why do people in the to 10% of income demand a vastly larger after-tax payoff to be productive than did people in earlier eras?

    Maybe something to look at is whether a higher income tax than capital gains tax is what makes people in the top 10% productive.

    If this is the case, then comparing JFK vs. Reagan really wouldn’t mean much since the income tax was higher than the capital gains tax during both periods.

    If you want to look at the glorious 1950′s and 1960′s when the marginal tax rate was 90%, you also need to point to the 25% capital gains tax rate during that period.

  65. 65 65 Bob Murphy

    I don’t have time to read all of the above comments, so perhaps every point I make in this critique of Steve is already up there. Anyway, once again I find Steve fighting not against other pundits, but the English language.

  66. 66 66 HispanicPundit

    One more question, in all sincerity since I get this from progressives alot. How would you answer the comparison to say Alice Walton, who basically lives off of Wal-Mart stock and dividends inherited from her father.

    What is her tax rate vs that of a secretary? How would you answer that?

  67. 67 67 nobody.really

    Maybe something to look at is whether a higher income tax than capital gains tax is what makes people in the top 10% productive.

    If this is the case, then comparing JFK vs. Reagan really wouldn’t mean much since the income tax was higher than the capital gains tax during both periods.

    For what it’s worth, I wasn’t comparing results during the JFK Administration and Reagan Administration. I was comparing earnings in the past – pretty much ANY past – to earnings today. I embraced KS’s notion that a major incentive for productive behavior is after-tax payoff. I argued that you could tax the top 10% at a very steep rate and still provide them with a higher after-tax payoff than most people have experienced at any time in history. I used this observation to challenge the idea that these taxes would have much effect on these people’s inclination to produce.

    Thus, the difference between prior eras and today is that the people in the top 10% are VASTLY RICHER now than in bygone days. Precisely BECAUSE they’re so rich, we need to offer them more compensation to keep them producing.

    I would add one important point though:

    Thus, the difference between prior eras and today is that the people in the bottom 90% are VASTLY RICHER now than in bygone days. Precisely BECAUSE they’re so rich, we need to offer them more compensation to keep them producing.

    Perhaps the term VASTLY RICHER is insufficiently precise. The Tax Foundation publishes IRS stats on aggregate adjusted gross income (AGI) by deciles. The definition of AGI changed with the Tax Reform Act of 1996 so comparisons between 1980 and today are complicated. But the table’s data reveal that from 1980–1986 aggregate AGI for the top 10% grew 44% faster than for the rest of the nation. And from 1987-2009, aggregate AGI for the top 10% grew 50% faster than for the rest of the nation.

    Thus in 1980 the top 10% received roughly 30% of all reported income; today they receive roughly 50% of the pie – and it’s a much bigger pie. And this merely reflects REPORTED income; Cayman Island accounts are icing on the cake. (Or perhaps reported income merely icing on the unreported cake? Unreported income is estimated to increase with AGIs, peaking among those with AGIs of $1 million.)

    The idea that taxes might leave these people with insufficient incentive to work not only boggles my mind, it would boggle their parents’ minds.

  68. 68 68 nobody.really

    Addendum: Financial Samurai reports AGI stats for 2010. If those stats are comparable to the stats at the Tax Foundation, it suggests that in 2010 the AGI for the top 10% grew 83% faster than for the rest of the nation.

    And sorry ‘bout the formatting above!

  69. 69 69 Steve Landsburg

    HispanicPundit: Alice Walton’s inheritance was diminished by the taxes her father paid as he earned his money. She is still very rich, but she is much less richer than she would have been in the absence of those taxes.

    If Alice Walton’s dad paid, say, 30% of his income in taxes, then Alice Walton’s investment income is 30% lower as a result. That’s exactly the sense in which that investment income was taxed in advance. Any further taxes that are levied on that income increase her tax burden above 30%.

  70. 70 70 Scott H.

    Tax consumption only. This way no one is penalized for working or saving. People that use the most resources would be taxed the most. The end.

  71. 71 71 Mark

    Krugman posted an attempt to refute arguments against capital gains taxes:

    http://krugman.blogs.nytimes.com/2012/01/19/the-dubious-case-for-privileging-capital-gains/

    “In short: the low tax rate on capital gains is bad economics, even ignoring who it benefits.”

    Krugman’s past as an objective economist with a lot to contribute seems like a hazy blur. Or maybe I just imagined it in the first place.

  72. 72 72 Henri Hein

    “If Alice Walton’s dad paid, say, 30% of his income in taxes, then Alice Walton’s investment income is 30% lower as a result”

    Except he would only pay 30% of that part of his income that was realized. In Sam Walton’s case, I would not expect that to be a signification portion.

  73. 73 73 Dmitry Kolyakov

    Professor Landsburg,
    I would be really sincerely interested in your opinon on the issues i mentioned in the second comment to your post (also discussed in comments to your “ranking the tax plans” post). Thanks in advance.

  74. 74 74 Schvitz

    This is a ridiculously specious argument, constructed under some notion that people with no disposable income [your implied assumption] pay a smaller percentage in taxes than those who invest. That’s because you’re counting opportunity cost as though it were a real cost. Sure. If Timm didn’t have to spend money on food and rent, he could also invest. And if Mitt didn’t have taxes to pay, he could invest that, too.

    Let’s assume, instead, that, over and above his income needed to pay for the necessities of life [however defined], Mitt has $1m in disposable income, on which he pays $350000. He can either spend the after tax money entirely on yachts, presidential campaigns, whatever; or, he can invest some or all. The choice is his, and it’s voluntary. If he invests all and doubles it, he’ll have the $1,202,500 net of tax, as you calculated. The fact that he may have decided to piss it away on toys or ego gratification is a choice, and not a factor that would lead one to conclude that to do so is good tax planning, because it results in a lower tax rate.

  75. 75 75 iceman

    @nobody.really: “tax people in the upper income brackets so that they don’t become so rich – and hence unproductive — in the first place.”

    Your concern is that people will become less productive after they’ve been very productive (which is how one *becomes* rich); your solution is to decrease the incentive to be very productive “in the first place”.

  76. 76 76 nobody.really

    @nobody.really: “tax people in the upper income brackets so that they don’t become so rich – and hence unproductive — in the first place.”

    Your concern is that people will become less productive after they’ve been very productive (which is how one *becomes* rich); your solution is to decrease the incentive to be very productive “in the first place”.

    That’s the lesson of the backwards-bending supply curve for labor.

    But the larger thesis is that people have, throughout history, found adequate motivation to produce while receiving vastly less compensation than do the people in the top 10%. Ergo I see little harm in taxing these people such that they receive merely the levels of after-tax payoff that were sufficient to motivate previous generations of productive people.

    (Actually, I can imagine various argument to the contrary, but thus far the only one that has been offered has been the backwards-bending labor supply curve – a fine argument, but one that tends to reinforce my policy proposal rather than undermine it.)

  77. 77 77 Henri Hein

    nobody.really: maybe we have not offered any counter arguments because we are baffled by your assumption that incentives work in a binary fashion: either people are presented with incentives or they are not. That is so clearly false that it seems like you are mostly playing around.

  78. 78 78 Henri Hein

    Here is an alternate illustration. Take two people, Mads and Sam. Mads works as an wage-earning employee, makes $30,000 a year and pays $10,500 in taxes (using Steve’s 35% rate).

    Sam works part-time as a wage-earner and makes $15,000 a year and also part-time building sweat-equity in a business. After 10 years, Sam retires from his wage-earner job and pulls $30,000 from his business by selling his equity in small chunks. He pays $4,500 in taxes (using Steve’s 15% rate).

    Why should Sam pay less taxes than Mads?

  79. 79 79 nobody.really

    [P]eople have, throughout history, found adequate motivation to produce while receiving vastly less compensation than do the people in the top 10%. Ergo I see little harm in taxing these people such that they receive merely the levels of after-tax payoff that were sufficient to motivate previous generations of productive people.

    (Actually, I can imagine various arguments to the contrary, but thus far the only one that has been offered has been the backwards-bending labor supply curve – a fine argument, but one that tends to reinforce my policy proposal rather than undermine it.)

    nobody.really: maybe we have not offered any counter arguments because we are baffled by your assumption that incentives work in a binary fashion: either people are presented with incentives or they are not. That is so clearly false that it seems like you are mostly playing around.

    I’m sorry if I’ve given that impression. In repeating my argument, I sometimes fall into an abbreviated version of it.

    So let me emphasize what I understand to be the continuous nature of the motivation theory: People reduce their productivity if compensated less, and increase it if compensated more. This implies that there are more productive things that highly-compensated people would do if only they were offered more compensation. So I’d be interested in hearing people’s suggestions.

    How much more productive would the top 10% have been in the 1980s if only they had received compensation comparable to what the top 10% get today? What specifically do we imagine they would have done to become more productive, if only given a sufficient after-tax payout?

    And what is it that today’s top 10% are refusing to do because they are not yet sufficiently compensated? Again, specifics would help.

    And can we explain this dynamic in terms other than the backwards-bending labor supply curve?

    If it matters, I offer my questions sincerely.

  80. 80 80 iceman

    @nobody.really: I think in the real world it’s a better characterization to say people are compensated *for* being productive. That doesn’t necessarily mean change their compensation (first) and ipso facto determine their level of productivity.

    But before you move on from dismissing the income effect, I don’t see how it reinforces your point. It’s only realized as people achieve it. So some people may be most productive while *and because* they are trying to achieve it. You seem to assume you can limit their ability to reach their goals with no change in behavior. (“OK let’s try that again, only this time…” And if they do reach them again do you take away the gains again?) Of course some people may work hard for other reasons, like the love of creating something. And maybe others get pissed off by policies that don’t seem to value what they have *created* (i.e. taking nothing away from others) and say screw it. In all sincerity as well, who of us can respond to your request for “specifics” of what other people might hypothetically do? How about this — I’ll give it a shot if you can tell me what the next great technological innovation will be.

    Note I’m not claiming to know if any labor supply curve literally is backward-bending, other than to the extent your tax proposal determines to make it so (by weakening the substitution effect)

  81. 81 81 nobody.really

    I’m not claiming to know if any labor supply curve literally is backward-bending….

    Me neither. But if you Google “Backward bending” you’ll find many fewer racy images than you might expect, and many more discussions of labor economics. The theory of the backward-bending labor supply curve seems to be a widely-regarded view among a lot of fully-clothed, not especially attractive people. Pity.

  82. 82 82 nobody.really

    Ok, here’s my thesis:

    The reason we usually believe that a reduction in compensation will lead to less work effort (by some, not by all!) is that if this were false, competitive pressures would already have reduced the compensation level.

    Theory 1: The fact that Albert Pujols received a contract for $240 million/yr for 10 years to play ball means that if he had not received such an offer, he would have declined to play ball. Or perhaps he would have been less “motivated” while playing ball.

    Theory 2: The fact that Albert Pujols received a contract for $240 million/yr for 10 years to play ball was driven by rival teams competing for Pujols’s services. But Pujols would still be playing baseball even if his after-tax payout were vastly smaller than it will be. Pujols would eventually switch to another pursuit if taxation became so onerous as to render baseball a second-best option for him – but Pujols would likely bear very heavy taxes before he reached that point.

    I subscribe to the second view.

    I suspect that today’s top 10% are, by and large, very productive people. I suspect that the top 10% of people in the 1980s were less productive than today’s top 10% are. But that’s not because today’s top 10% are more MOTIVATED or have greater INCENTIVES. Rather, I believe that today’s circumstances enable certain people to better exploit their talents. Think about it: Would we really say that this generation of people is better compensated because we’re more motivated and incentivized than any other group in history?

    The people in the top 10% are not (by and large) villains; they are heroes. Good for them; good for us all.

    But precisely because these people receive greater compensation than ever before, I suspect that they can be taxed more heavily than ever before. As KS observed, people are motivated by (among other things) after-tax payoff. We can tax people in the top 10% more heavily and still leave them with an equal or higher after-tax payoff than prior generations of productive people experienced.

    I don’t begrudge anyone disputing this conclusion – provided you acknowledge that you’re rejecting the idea that people are motivated by after-tax payoffs. I sense some people conclude that people are motivated by the relationship between after-tax payoffs and EXPECTED after-tax payoffs: they’d rather expect and earn $1 than expect to earn $10 and actually earn $5. That’s a cool theory. Others suggest that people are motivated by the proportion of their productivity that they can capture for themselves; they’d be more motivated to produce and keep $1 than to produce $10 and keep $5. That’s a cool theory, too. But these theories deviate from the classical model, so I’m somewhat more skeptical of them.

    That said, let me make a few concessions:

    We don’t currently have a means to tax all use of time equally. As I noted above, at some point of taxation people will find it more rewarding to spend their time in less taxed pursuits – working abroad, or working in the black market, or pursuing leisure. The question is, how large is this affect for people who are already in the top 10%? In other words, how much would you need to tax Pujols before he would leave to play in the Japanese or Latin American leagues?

    As iceman observed, the backwards-bending labor supply curve means that people will reduce their productivity when they find themselves sufficiently rich. Taxing today’s currently rich people may induce them to reduce their hours; that may be unavoidable. But we could design the tax code to optimize returns from – and productivity of — future generations of highly-compensated people.

    Finally, there may in fact be certain behaviors that people will not undertake without very high compensation. I’m thinking of behaviors that pretty much nobody in history has ever undertaken because the risk/reward ratio did not make it seem appealing, but that might become appealing if the reward part of the ratio could be increased.

    For example, you might consider “selling your body” for sex, or literally selling your body for organs if the price were right. (I suspect these offers tend to attract people who were not in the top 10%, however.)

    Or consider the case of Lehman Bros. You might say that the firm was enticed to take extraordinary risks simply because the opportunity for a profitable outcome was equally extraordinary. Taxing people might in fact deter such bold behavior. Whether this would result in a productive or unproductive change is a matter of dispute. But arguably the better way to deal with the Lehman Bros. situation is controlling externalities (the moral hazard issue), not clamping down on the ability to reap big returns on really big gambles.

    Finally, would taxing the top 10% distort behavior and lead to deadweight social loses? I expect so. But which form of taxation would be better?

  83. 83 83 PrometheeFeu

    @Greg

    “If we repeat this process over N periods — that is, Mitt & Timm repeatedly reinvest their earnings — wouldn’t the initial income tax become increasingly inconsequential as N -> infinity?”

    I thought so too at first, but in order to do that, you have to assume that the size of the investment is inconsequential on the size of the return. From the beginning, Mitt is investing less money than Timm because he was taxed at 35% in the first period. So in the best case scenario (where they both spend their capital gains instead of reinvesting them) the 35% still matters a lot.

  84. 84 84 Rougaroo

    Professor Landsburg -

    Given your example, “everything else being equal – in a truly rational world” (as economists are wont to say) why then wouldn’t Mitt and Timm go to work as wage-earners? And what possible motivation would the Unnamed Working Stiff have in investing?

    The answer, of course, is an implied but unstated inequality. While the Unnamed Working Stiff is out earning money, Mitt and Timm’s *money* is earning money. Since UWS earns money, pays 35% in taxes and spends the rest (your hypothesis), he never has an incremental dollar to put into the investment world. Mitt and Timm, meanwhile, presumably value their free time more than the 5% tax difference or they would return to the wage-slave treadmill.

  85. 85 85 Steve Landsburg

    Rougaroo: People trade current consumption for future consumption for the same reason they trade apples for oranges. Some prefer apples; some prefer oranges.

  86. 86 86 Draco

    Without naming names, people who spend hours and hours thinking up schemes for how to make other people more productive while taxing them more (i.e. getting more of a share for themselves or for their causes) scare me. To be quite honest, I find them repellent. I shudder when I think of the uses history has found for such personalities.

    I wonder how much of it is just pure envy, versus a sincere altruistic desire to ensure that those who have less are well cared for? If the latter, wouldn’t asking nicely be a more ethical course than scheming to use the power of government to achieve charitable ends?

  87. 87 87 iceman

    @nobody.really: “precisely because these people receive greater compensation than ever before, I suspect that they can be taxed more heavily than ever before…I don’t begrudge anyone disputing this conclusion – provided you acknowledge that you’re rejecting the idea that people are motivated by after-tax payoffs.”

    I think you make a fun argument but one that’s more clever than correct (so I dispute the conclusion without the acknowledgment). Your ‘motivation theory’ focuses on the overall level of compensation, but in my experience when economists talk about incentives (at least as it relates to taxes) it’s generally in terms of *marginal* rates, i.e. the amount people get to keep of their next $ earned. It seems this ‘substitution effect’ can apply at any income level, and if anything the highest earners may be the *least* sensitive to it because of the income effect i.e. your backward-bending curve. Assuming you’re only talking about raising their tax rate and not actually confiscating wealth already earned (which as I’ve suggested might have its own pernicious effect on behavior), how do you see this as motivating them to be more productive, e.g. if they’ve already stashed enough away and you’ve just reduced their marginal incentive to earn more?

    If you want to stick with your example of elite athletes, I think what we really tend to see is that they’re most motivated when they’re playing to *get* that big contract; I can think of many examples where production actually falls afterward (especially in baseball where the contracts are guaranteed).

    I have to add on the correlation (or lack thereof, as you and KS like to cite) between growth and tax regimes over time, clearly it’s just correlation, and context is key. (If only I had a $ for every time I’ve heard that argument I might really have to worry about the estate tax.) Obviously there are many (thousands of) other moving parts in an economy, but if anything the Kennedy cut from 90% to 70% showed that directional changes in marginal rates can have an impact regardless of the initial level. The economy likely would have done even better had he cut it to 50% etc. Does anyone seriously think we would have stronger growth today if we went back to those levels?

  88. 88 88 nobody.really

    [W]ouldn’t asking nicely be a more ethical course than scheming to use the power of government to achieve charitable ends?

    Couldn’t agree more. I think asking nicely for people to help maintain living conditions for the elderly would be a much more desirable approach than compelling people into Social Security. And I think that asking Hitler nicely to keep his troops within the post-WWI borders of Germany would be a more desirable approach, too. If only you and I had been part of Roosevelt’s kitchen cabinet, imagine how much nicer the world would have been!

  89. 89 89 nobody.really

    [P]eople who spend hours and hours thinking up schemes for how to make other people more productive while taxing them more (i.e. getting more of a share for themselves or for their causes) scare me. To be quite honest, I find them repellent. I shudder when I think of the uses history has found for such personalities.

    Are you referring to time and motion experts? Admittedly, the dad in Cheaper By the Dozen seemed a little eccentric…

    Or are you referring to Congressional Republicans? They wanted to repeal ObamaCare on the grounds of a Congressional Budget Office study estimating that the law would lead to 0.5% a loss of people in the labor market. That is, when ObamaCare is implemented, we can expect some people who are working simply to have health insurance will elect instead to take leisure. Congressional Republicans labeled this behavior “job-killing,” and sought to manipulate public policy in order to keep these people working.

    But as far as I can tell, pretty much any employer will contrive to find ways to make workers more productive and to capture as much of that productivity for herself as possible. She may install an employee cafeteria in order to keep employees from spending too much time traveling to and from lunch breaks – and then may set prices in the cafeteria in order to turn that into a profit center, too.

    I admit, I also feel anxiety about the fact that my seemingly voluntary actions and preferences can be manipulated by circumstances, and that people make a study of how to achieve this outcome. If you find these thoughts too disturbing, you probably won’t want to read about the Milgram experiments, or about behavior economics (see, for example, Nudge), or about much involving advertising and persuasive speaking – including political campaigns. It’s all rather challenging to ideas of autonomy.

    Nevertheless, what’s the alternative? Do we want people to establish polices in ignorance of, or with disregard to, the predictable manner in which we are likely to respond to those policies?

    Where government is concerned, perhaps so. I don’t object to government trying to alter my behavior by providing me with factual information. But a lot of social science research suggests that people don’t alter their behavior based on unadorned factual information. Consequently I’m exposed to public service advertisements featuring sinister tobacco executives sneering and cackling about the harm they are inflicting on the public. I see ads suggesting that joining the Marines will enable me to slay fire-breathing monsters in the middle of a stadium. I see ads saying that playing the lottery will make me happy. I don’t doubt that these messages are much more effective than a plain statement of objective facts. But should government be in the propaganda business – even to achieve outcomes I find desirable?

    For what it’s worth, to theory of Ramsey-Boiteux pricing suggests that it is often most efficient – that is, least distorting to pre-tax behavior — to tax people with the least elasticity of demand. E.g., people with the fewest options. E.g., poor people. So when we study how to design taxes to achieve an optimal mix of revenues and productivity, the affluent are not the people I’m most concerned about.

  90. 90 90 KS

    Dr. Landsburg– I appreciate your reply, and I apologize for my delay in replying.

    “Actually, the law of diminishing returns describes what happens if you have a physical production process with multiple inputs, and you increase one of those inputs without increasing the others. The statement you are making a) has nothing whatsoever to do with the law of diminishing returns and b) relies on a whole lot of philosophical baggage (i.e. not just cardinal utility, but cardinal utility that can be compared across individuals) that has been pretty hard to swallow for most of the people who have actually thought hard about this. (I happen, incidentally, to be more receptive to that philosophical baggage than are many economists. But those of us who are receptive to it ought to recognize it for the baggage that it is, not present it as some sort of fundamental uncontroversial principle.)”

    I combed through my old economic notes, and I learned it was the “law of diminishing marginal utility” or “law of diminishing marginal returns” (my professor’s terminology, not mine). Basically, this law says the more you have of X, the less you value additional X — this can be interpreted as the less you’re willing to pay for additional X, or the less happiness/utility you receive by being given additional X. (This is why a gallon of water is far cheaper than a gallon of diamonds, even though water is essential to life and diamonds are, well, diamonds). The first part of your reply is merely semantics. The latter part I find more interesting, can you articulate exactly what the philosophical baggage underlying this axiom is, or refer me to a source that does so?

    “No, actually it’s got absolutely nothing to do with the point. The point there was that we had a guy who consumed (by assumption) zero resources; therefore it was impossible to transfer resource consumption away from him. When the whole point is that you can’t effect a redistribution, the desirability of that redistribution becomes quite irrelevant.”

    I’ll grant you this, with a caveat: say he kept his money in a bank. Therefore he spends nothing, so obviously you can’t reduce his spending. His bank does invest his money, and my point about desirability of redistribution is the same — maybe the government can spend that same amount of money in a way that is better than the way the bank invests is (and I define better as produces an overall increase in total social utility).

    “Please name one person who has ever assumed this.”

    Maybe no-one has stated it like that, but it is an implicit assumption when someone argues an increase in taxes will decrease the work incentives of the “job creators”. What if I’m taxed at 35% and you bump it up to 40%. Maybe I’ll work less hard, but maybe I’ll work even harder to achieve the same after-tax income. And certainly my function isn’t the same as yours, or anyone else’s.

    As an economist, you would agree that any analysis of the effectiveness of an intervention must focus on both the costs AND the benefits. Sure, there are costs to taxation. However, there are also benefits. Most, if not all, of the discussion regarding taxes on your blog solely concerns the costs (distortion of incentives, reduction of efficiency, etc…). Whether that’s because you just want to focus on this side and aren’t intending to examine the overall effectiveness of taxation, or whether it’s because you see taxes through a political lens that tends to ignore the benefits, is irrelevant. Completeness is lost.

  91. 91 91 Richard Williams

    Your comparison of Timm and Mitt is apples and oranges. At the end of the day, a person who makes $1,650,000 and pays $447,500 in taxes pays an effective rate of 35% — substantially less than the effective rate of 35% against a non-investor.

  92. 92 92 Richard Williams

    should read 27% for effective rate of investor taxpayer

  93. 93 93 Mike H

    Steve,

    I’d appreciate a post on or a link to something showing in reasonable detail why future consumption “should” be taxed the same as current consumption, and therefore why all “income” should be taxed equally. I’m led to believe that this is the main foundation of your argument that there should be no tax on investment income, but I haven’t seen it justified anywhere.

  94. 94 94 Ted

    Steve’s mirror world shows that Mitt pays 40% of what would have been his income in Timm’s tax free world, but it also provides a handy tool for exploring other revenue structures. Suppose the taxmen of Mitt’s world distributed the tax burden somewhat differently – instead of 35% of salary and and 15% of investment income, they might (crudely) try to maintain revenue by making the charge 50% of salary and 0 % of investment income. That would wind up costing Mitt 50% of his total income.

    Hmmm? Maybe it’s not investment tax that’s hurting his bottom line. Suppose we reverse the tax rates to 0% of salary and 50% ofinvestment income. In that case he only pays 25% of his Timm world income. That makes it look like it’s tax on salaries that really hurts the saver.

    Am I missing something?

  95. 95 95 KS

    @nobody.really–

    “People reduce their productivity if compensated less, and increase it if compensated more. This implies that there are more productive things that highly-compensated people would do if only they were offered more compensation.”

    This, to me, best summarizes the theoretical economist’s view on taxation, so I appreciate your brevity. However — and this will be my last post on this subject in this blog — I just want to counter with the following. Whenever taxation is discussed, these three should always be kept in mind.

    In my mind, the effect of taxation on human incentives is: (1) heterogeneous, (2) non-linear, and (3) not derivable by theoretical considerations.

    (1) Heterogenous: Taxation policy that affects one person’s behavior will have a very different, potentially-opposite effect on another person. One person might be dissuaded by a higher rate and work less; another might be motivated by the same higher rate to work harder (to achieve the same after-tax income). Who knows?

    (2) Non-linear: By this I mean you can’t predict how a marginal change in taxes will produce a marginal change in behavior, even within one individual. An increase of 10% in the marginal tax rate will have a different effect on behavior if you’re going from 30 to 40% than if you’re going from 90 to 100%.

    (3) Not theoretically derivable: This is the biggest failure of modern economics, in my opinion. The idea that complex relationships in the social sciences can be understood by a priori principles. We simply don’t know how a change in tax rates will affect behavior, and stringing a sentence together that “People will become less productive as you tax more” is simply bad science.

    With these three in mind, it should be a lot easier to wrap your mind around the fact that tax rates don’t correlate well with economic growth historically, or that the inverse may even be true.

  96. 96 96 Eli Rabett

    “People reduce their productivity if compensated less, and increase it if compensated more. This implies that there are more productive things that highly-compensated people would do if only they were offered more compensation.”

    Nonsense, people work harder and take additional jobs if compensated less. You gotta eat and you gotta pay the mortgage and then there are the kids.

    The simple disproof of this economist’s wet dream is the move of married women into the workforce when real wages started to stagnate.

  97. 97 97 Steve Landsburg

    Eli Rabett:

    people work harder and take additional jobs if compensated less….The simple disproof is…the move of married women into the workforce when real wages started to stagnate.

    Even if this were empirically accurate (which it isn’t), you’re confusing a wealth effect with a substitution effect. Your story is essentially that women work more not when their own wages fall, but when their spouse’s wages fall. That’s a very different phenomenon.

  98. 98 98 nobody.really

    [T]he effect of taxation on human incentives is: (1) heterogeneous, (2) non-linear, and (3) not derivable by theoretical considerations.

    And yet, we still need to find a way to pay for government services. Thus even if I accept that our knowledge of taxation is imperfect – and I do – how does this humility influence the design of a tax structure? We still need to proceed on the basis of the best information available, right?

    I guess you could say that this kind of humility influences the discussion of public finance by promoting some economists to oppose trading an income tax for a consumption tax. They argue that for any system of taxes, the devil is in the details, and we’re better off with the devil we know – and have gained a century’s worth of experience with – than with the devil we don’t. It’s a fair argument, I guess. Not much fun for blogging, though.

    @nobody.really: People reduce their productivity if compensated less, and increase it if compensated more. This implies that there are more productive things that highly-compensated people would do if only they were offered more compensation.

    [P]eople work harder and take additional jobs if compensated less. You gotta eat and you gotta pay the mortgage and then there are the kids.

    Yes, this describes the (previously-discussed) backward-bending nature of the labor supply curve. This dynamic represents a big fat exception to the principle that people work more when paid more. I share Landsburg’s view that this dynamic arises from a combination of an income effect and a substitution effect. I don’t share his implication that this renders the claim inaccurate.

    (I also don’t know how much wage stagnation influenced the trend of married women to join the paid workforce. I suspect other factors – the previous growth of women in college, the growth of electrical appliances in reducing the demand for domestic labor, the women’s movement, the pill – also contributed to increasing the number of married women working outside of the home.)

  99. 99 99 KS

    “And yet, we still need to find a way to pay for government services.”

    Agreed, and if the revenue has to come from somewhere, it’s a good idea to look at higher sources of revenue (ie, the rich). It’s also a good idea to not assume taxing the rich will hurt jobs/kill economic growth/etc…, especially if historical evidence suggests higher taxes on the rich have coincided with acceptable (if not superior) economic growth.

    America faces an issue of long-run fiscal sustainability, clearly. A proportion of this will require spending restraints, another proportion will require additional revenue. Yet one of our two-party electorate has committed itself to never accepting an additional tax increase on any upper income person or corporation ever — and defended this with a (1) willful ignorance of historical evidence, and (2) naive and narrow-minded theoretical considerations.

  100. 100 100 Greg MN

    Thomas Baynes asked “why do people have so much trouble accepting this [Steven's] view?” My answer is…

    In any given tax period, investment income is taxed at a lower rate than wages and salaries. In other words, a person with taxable income of one million dollars in 2011 will have a lesser tax burden if that income is investment income as opposed to wages.

    Steven’s view rests upon looking beyond one given tax cycle and assuming at least two things: first, the person had previously paid the higher income tax rate on all money subsequently invested and second, absent the taxes, the person would have invested all the tax savings and the investment results wouldn’t change. Considering neither of these assumptions is absolutely true, people can correctly reject Steven’s view. Over Steven’s eight attempts to present his view, many commenters have described situations where these assumptions are false, but Steven simply dismisses them and continues to present his view as though it is irrefutably correct. In this latest effort, Steven acknowledges people can “avoid the wage tax altogether”.

    So, my question is… How can Steven have the audacity to claim avoiding the wage tax altogether “in no way undermines the main point” when having previously paid the wage tax is central to his position?

  101. 101 101 Steve Landsburg

    Greg MN:

    So, my question is… How can Steven have the audacity to claim avoiding the wage tax altogether “in no way undermines the main point” when having previously paid the wage tax is central to his position?

    The fact taht some people have previously avoided the wage tax in no way undermines the main point that most people haven’t.

  102. 102 102 Greg MN

    “The fact taht some people have previously avoided the wage tax in no way undermines the main point that most people haven’t.”

    How would one determine the percentage of invested money which has avoided the wage tax? Due to the variety of ways to avoid the wage tax and the fungibility of money, I suspect it is impossible to do so with a sufficient level of accuracy.

    Sorry Thomas Bayes. I realize I misspelled your surname in my original post.

  103. 103 103 Eli Rabett

    Prof. Landsberg, When you say
    —————-
    Even if this were empirically accurate (which it isn’t), you’re confusing a wealth effect with a substitution effect. Your story is essentially that women work more not when their own wages fall, but when their spouse’s wages fall. That’s a very different phenomenon.
    —————

    Try when their FAMILY’s wages fall. Family income is what real people care about because that is what determines their standard of living.

    And as to empirically correct, wanna bet?

  104. 104 104 Steve Landsburg

    Eli Rabbett:

    Try when their FAMILY’s wages fall. Family income is what real people care about because that is what determines their standard of living.

    Precisely. In other words, this is not a substitution effect; it’s a wealth effect.

    As for the empirical facts, start here: http://www.nber.org/digest/nov05/w11230.html . The wage elasticity of female labor supply has fallen over the past 30 years from maybe .9 to something more like .4. It has never been negative.

  105. 105 105 Max Briggs

    I think that this is a very interesting topic, and I am happy to find some rational conversation about it somewhere. I have a somewhat related question about the capital gains tax.

    Many politicians say that the capital gains rate should be lowered because it “encourages investment”, which helps the economy. This makes sense to me for loaning money, buying bonds, or supplying seed money/venture capital, because these investments give money directly to companies to expand. However, when stocks are traded, the transaction is between the owner and the buyer. The company does not see any of that money, and cannot use it expand.

    So my questions are:

    If the rational for reducing the capital gains tax is that it spurs economic growth, then why are all forms of investment treated the same. If that was the intent, I would assume that capital gains on venture capital, loans, and bonds would be taxed at a lower rate than capital gains on stocks.

    I understand that this question is off topic, since the debate appears to be that the capital gains rate is unfair to investors in the first place, but I still think it is an interesting question.

  106. 106 106 Steve Landsburg

    Max Briggs: If the tax system discourages me from buying your stocks or bonds, then it indirectly discourages you from acquiring those stocks in the first place, and so has the same effect as a tax on direct equity investment or loans to businesses.

  107. 107 107 Maxwell Briggs

    Regarding your example above. I think it is interesting to extend the hypothetical a little bit. Say you have Timm Romney and Mitt Romney, from your example above, both investors, one taxed, one untaxed. Then you have another doplanger pair, say Newt Gingrich and Gewt Ninnrich. Gewt is a “laborer” who lives in a tax free world. He makes 1 million dollars in the first year, the same as Timm, then in the subsequent years he earns labor income, exactly equal to what Timm earns in capital gains. Newt Gingrich in the taxed world earns the same income, but is taxed at a 35% rate.

    If I did my math correctly, using your definition of tax burden (1-(income in taxed world)/(income in untaxed world)), Newt Gingrich still has a lower tax rate than Mitt. 35%, versus 40%, just as in your original example. Newt pays nearly $700,000 in taxes, while Mitt pays ~$440000.

    So, comparing effective tax rate: (1-(tax collected/income earned), Newt pays 35% of his income to taxes which is much more than the 22.3% payed by Mitt.

    So I guess the argument is essentially, which is the more relavent statistic, effective tax rate or total tax burden?

  108. 108 108 KS

    @GregMN- “Over Steven’s eight attempts to present his view, many commenters have described situations where these assumptions are false, but Steven simply dismisses them and continues to present his view as though it is irrefutably correct. In this latest effort, Steven acknowledges people can “avoid the wage tax altogether”.”

    You’re wasting your time. SL is clearly a very intelligent man, but he’s in a very dumb field, where theory trumps evidence, and narrow-minded linear modeling trumps multifactorial combinatorics. After all, SL gave an entire speech and wrote a WSJ editorial on how the estate tax is a bad idea because: (1) it affects rich people, (2) rich people consume more resources than they should, (3) it creates an incentive for rich people to consume even more resources than they should.

    The folly here is astounding, and simply too long to list point-by-point. Perhaps a metaphor would work best. If SL were a bodybuilding personal trainer, he would argue (on theoretical grounds) that the higher the weight of each rep, the less incentive there is to perform the rep since it’s harder, therefore, all bodybuilders should do 300 sets of 10 reps of 0.5 lbs each.

    Furthermore, the sheer idea that one can consider an intervention (like an estate tax) by only examining the costs and not the benefits at all would not pass in any other science (if you can call modern economics a science). It’s like a clinician arguing that taking aspirin is bad for you, since consuming a pill increases your weight, and increased weight is linked to heart disease. This ignores the fact that SL’s theoretical considerations of the costs are narrow-minded and linear.

  1. 1 You don’t think about taxation the right way: capital gains edition « Shewing the fly
  2. 2 Some Links
  3. 3 Steve Landsburg 2, English Language 0
  4. 4 Follow-Up on Labor versus Investment Taxes
  5. 5 Income and Consumption: It’s OK to Be Different
Comments are currently closed.