Capital income is taxed more heavily than wage income

[This is a follow-up for a longer piece at Econlog.]

Yes, the “official” top tax rate on capital income is lower than on wage income, but the very first thing you learn in public finance is that appearances can be deceiving.  The economic incidence of a tax is often very different from  the legal incidence of a tax.

While reading a recent book review by Robert Solow I came across a statement that made me want to pull my hair out.

We are politically unable to preserve even an estate tax with real bite. If we could, that would be a reasonable place to start, not to mention a more steeply progressive income tax that did not favor income from capital as the current system does.

Solow is a brilliant economist, indeed a Nobel Prize winner.  But this is an EC101 error.  I see this sort of nonsense so often it makes me wonder whether I am hallucinating.  Is it possible that the standard theory of taxation was overturned after I left grad school?  For those who have not taken public finance, here’s the intuition in a simple example:

I earn $100 wage income, and face a tax of 50% on wage income and 20% on capital income.  I have two choices, spend the money today, or save it for 14 years, in which time (at 5% interest) my money will double.  I pay 20% tax on the capital income.  So let’s look at the two options:

A.  Spend the money today:  $50 in consumption, vs. $100 in a tax-free world.

B.  Save the money for 14 years:  $90 in consumption in 14 years, vs. $200 in a tax free world.

In case A I face a 50% tax rate.  In case B I face a 55% tax rate.  I’d love for someone to explain to me how this sort of tax system “favors” income from capital.

There are certain terms that when uttered almost immediately lower one’s IQ by 5 points: bubbles, beggar-thy-neighbor, inflation, deserve, endogenous money, inequality.  Add “income” to that list.

Sound economics deals with consumption.  Taking about “income” is about as silly as talking about how many “fruits” are produced in California each year, counting both a watermelon and a blueberry as “one fruit.”  Just shoot me.

PS.  Of course this is just the tip of the iceberg.  Corporate income is triple taxed. Interest income is not adjusted for inflation, distorting inequality data.  Etc., etc.

PPS.  Did I just mention inflation?  Even worse, I put all four “i-words” in one sentence.  Please deduct 20 points from my estimated IQ.



51 Responses to “Capital income is taxed more heavily than wage income”

  1. Gravatar of NV NV
    24. April 2014 at 06:23

    Very clever argument by Prof. Sumner. By setting the denominator to a fixed number from a hypothetical “tax free” world ($200 in case B) and only changing the numerator, Prof. Sumner gets a nice result. Prof. Solo clearly thinks that both the numerator and denominator are not fixed ($60 and $140 in case B) and the cumulative tax rate is only 43%.

  2. Gravatar of Kenneth Duda Kenneth Duda
    24. April 2014 at 06:23

    Scott, you are too generous. If you pay the 20% tax on capital income each year, rather than all at the end, then your total consumption in 14 years is only about $87.50.

    The other way in which you’re too kind is you omitted the impact of inflation. Take your scenario and add 2% inflation per year. After 14 years, you get your $50 in principal back, and you have $50 in nominal earnings — but thanks to inflation, it takes a little more than $66 in post-investment money to buy what $50 bought at the beginning. Being taxed on that $16 of investment returns, which is needed simply to tread water, has always bothered me. I’ve always felt like I ought to be able to adjust my cost basis for inflation when calculating capital gains.

    Another way to view this basic unfairness is that suppose I have an investment vehicle whose return is always equal to the inflation rate (e.g. TIPS). I never actually increase my consumption by investing in these — my money is effectively parked. This is fine until tax time. Then you realize that you are actually falling behind on spending power, and the higher inflation is, the faster you fall behind. If we have 100% inflation per year, then you’re losing 10% of your spending power each year ($1 plus $1 in “earnings” minus $0.2 in tax is $1.8). Here is a case where money is truly not neutral, in a way that even the hardest-nosed freshwater economist could recognize (I’d hope), because the more inflation there is, the more your investment returns must exceed inflation just to break even after inflation and tax.

    No wonder wealthy people don’t like inflation! They have no need for it (what do they care about relative price adjustment in the face of downward nominal rigidity), it obviously kills the returns on any fixed-nominal-percentage investment, and what’s more, it actually increases their effective tax rate.


    Kenneth Duda
    Menlo Park, CA

  3. Gravatar of Philippe Philippe
    24. April 2014 at 06:33


    I might be missing something, but that doesn’t seem to make sense.

    If you earn $100 in wage income, and face a tax of 50% on wage income, then don’t you pay that 50% regardless of whether you choose to spend your income on consumption today, or save it for the future?

  4. Gravatar of Jack Jack
    24. April 2014 at 06:44

    I’m pretty confused here… I must be misinterpreting something.

    Scenario A:
    Earn $100, consume immediately = $50 of taxes on $100 = 50% ETR

    Scenario B:
    Earn $100 (presumably be allowed to defer this until you spend it? Otherwise paying 20% could only lower the average) which becomes $200 in 14 years. Pay $50 of tax on the original earned income and $20 of tax on the $100 gain = $70 of taxes = 35% ETR

    I guess I’m confused about your calculations for:
    B. Save the money for 14 years: $90 in consumption in 14 years, vs. $200 in a tax free world.

    Can you break this one down?

  5. Gravatar of Philo Philo
    24. April 2014 at 06:45

    “I put all four “i-words” in one sentence.” “Income,” “inflation,” “inequality,” . . . . What’s the fourth? “Interest”? “Is”?

    “Please deduct 20 points from my estimated IQ.” In my book you’re still a genius!

  6. Gravatar of MP MP
    24. April 2014 at 06:52

    Kenneth-I believe Scott is rounding.

    Philippe/Jack-Scott includes the double taxation of capital income. That’s how he gets the $90 & $200 in scenario B.

    $50 (after tax of $100 wages) invested at 4% (assuming 20% taxes every year on realized capital income) for 14 years is about $90.

    In a no tax world you start with $100 and it doubles to $200.

  7. Gravatar of MP MP
    24. April 2014 at 06:55

    Actually he could be assuming the capital income comes at the end, too. That is just under $90 in that case.

  8. Gravatar of Steve Waldman Steve Waldman
    24. April 2014 at 06:55

    It’s entirely unsurprising (and uninteresting) that paying both wage + cap income taxes takes a bigger bite off an entirely tax-free hypothetical than paying only wage taxes. The only way this would not be true is if cap income taxes are zero. That’s your policy preference, great. But there’s no information in pointing out that cap income taxes take more of a bite than no cap income taxes at all.

  9. Gravatar of cthorm cthorm
    24. April 2014 at 07:53


    If it’s not surprising, than you aren’t the target audience. The headline capital gains rate is almost always touted as meaning that we treat investment income more favorably than wage income. Scott is making a point he has made many times before: at time T that investment income isn’t taxed as highly, but it was already taxed as wage income at T-1. This is true in all but the most extreme cases (e.g. boards of directors that forgo salaries in favor of equity stakes).

  10. Gravatar of Russ Abbott Russ Abbott
    24. April 2014 at 08:10

    I just wrote the following post for my Google+ Stream.

    What’s wrong with this picture?

    I find most of what Scott Sumner writes worth reading. But today he offered the following example to show that capital income is taxed more heavily than wage income.

    I earn $100 wage income, and face a tax of 50% on wage income and 20% on capital income. I have two choices, spend the money today, or save it for 14 years, in which time (at 5% interest) my money will double. I pay 20% tax on the capital income. So let’s look at the two options:

    A. Spend the money today: $50 in consumption, vs. $100 in a tax-free world.

    B. Save the money for 14 years: $90 in consumption in 14 years, vs. $200 in a tax free world.

    How does that prove his point? If my $100 today came from capital income, the money I would have after saving the after-tax portion for 14 years would be $160–less 20% tax or $128, which is more than $90.

    On the other hand, if instead of investing my after-tax income I just worked enough to earn the same amount, instead of $90 I would have only $75 after 14 years (since we are assuming an income of $50 over 14 years, which if taxed at 50% yields an after tax result of $25).

    Either way you are better off if your income comes from capital than from wages. So I don’t see how Scott’s example works at all. Perhaps a monetarist can explain.

  11. Gravatar of Steve Waldman Steve Waldman
    24. April 2014 at 08:32

    @cthorn We do, objectively, treat capital income more favorably than wage income. Not all capital income derives from wage income, unless you want to trace things back to some stylized yeoman past. If i inherit $2K, the at 5% I earn the same $100 that our hypothetical wage earner does. We both consume today. She consumes $50. I consume $80.

    Scott’s headline is objectively false. It should read “Income taxed as wages then as capital is taxed more heavily than income taxed as wages only.” But that would surprise or interest no one.

    You, or Scott, can argue that because much capital income is generated from income first earned as wages, taxing capital income amounts to “unfair double taxation”. We can argue whether sales taxes paid on after tax dollars is “unfair double taxation”, or dividend taxes paid on already taxed corporate profits are “unfair double taxation” too. Those are arguments we can have. But there is no need to dissemble to have those arguments.

  12. Gravatar of foosion foosion
    24. April 2014 at 08:44

    @Steve, it’s probably worth mentioning carried interest compensation taxed as capital gains rather than at ordinary wage rates.

  13. Gravatar of ssumner ssumner
    24. April 2014 at 09:18

    NV, ????????

    Ken, I agree with most of what you say, and indeed have discussed inflation in other posts. All I would add is that when demand is depressed (as since 2008) many rich would benefit from a bit more inflation, as the gains to stocks would greatly outweigh the loss to bonds.

    Philippe. That’s right

    Jack, Wage taxes must be paid when the money is earned. Otherwise there’d be no problem.

    Steve, It may not be surprising to you that standard public finance theory says the optimal rate of taxation on capital is zero, but it is certainly a big surprise to 99.99999% of people.

    Russ, You are comparing apples and oranges. Either assume the $100 faces a tax in both cases, or not. You can’t assume different scenarios for tax today for the current and future consumption scenarios. There’s nothing controversial about my example. It’s public finance 101.

    Steve, Sorry, but what you are saying is flat out wrong. It makes no difference where the original $100 comes from. It’s purely a question of different tax rates on present and future consumption. Nothing controversial in what I’m saying. It’s public finance 101.

    To use the term ‘income’ for both wage and capital income is a monstrous abuse of language–it’s what confuses so many people when considering tax theory. The present value of $90 consumed 14 years from now is lower than the PV of $50 consumed today.

    Normally when people talk about the issue of where did the capital originally come from, they go on to argue for a one time capital levy, followed by a pure progressive consumption tax. There are good arguments both ways on that issue.

    foosion, I agree 100% on closing those loopholes. But for God’s sake stop forcing me to file a 1040 and pay taxes on my capital income. Just levy a progressive payroll tax on me. I’d gladly pay more money in taxes.

  14. Gravatar of Doug M Doug M
    24. April 2014 at 09:47


    Despite some accusations of gaming the numbers, I like what your example as it applies to the small saver and the “average citizen.”

    But do the rules change for the 0.1%. I am in a position where I have hundreds of millions of dollars of my wealth tied to the value of my company. I pay myself $1 million, buy my net worth is growing at $10 million / year.

    The value of my shares is tied to the my companies earnings, which are subject to corporate income tax rates. But I have clever accountants who can deffer paying that tax for several years.

    My company lets me use their planes, and boats, buys me club memberships, holds corporate retreats in exotic locations, and sends me around the world to do a little work and a lot of play. I don’t pay tax for most of this consumption. Neither does the company.

    I don’t pay taxes on the accretion of my shares until I sell them. And when I do, I pay that tax at the long-term capital gains rate.

    And if I really want to play with the tax man, I don’t sell my shares. I borrow against them and deduct the interest expense from my taxable income for a net negative tax rate.

  15. Gravatar of dw dw
    24. April 2014 at 10:13

    since i am not an economist by trade maybe i am missing some thing. but combining wage and capital taxes seems to make it much harder to see what your dring at. since most of us have one or the other but not both (unless you are talking about 401ks etc. but then those get taxed at wage incomes dont they?).

    so if you wages of 100$ taxed at 50%, you would have about 50$ left after taxes.

    but if you had capital income of 100$ taxed at 20%, would you not have 80$ after taxes?

    and it seems that with capital income you have a lot of ways to reduce even they 20% tax rate right? while the only real reduction of taxes you can have for wages is retirement investments (401ks etc), and the best you can to reduce taxes as such is to reduce your income about 14-15 % right? while capital incomes taxes can be reduced in a multitude of ways , and capital only incomes tends to be much larger, so some one can get really inventive in ways to cut taxes right?

    but i guess the real question would be why do we favor either? wage earners do tend to spend all of their money or more of than those with capital incomes done they? and i guess the main reason to try to favor capital. is that is needed to form companies, which creates jobs in theory any way right? except isnt it consumers who actually creates job and profits, not the capital isnt it? since no company will ever create any profits or jobs without some one buying (consumers) what ever its selling right?

  16. Gravatar of mpowell mpowell
    24. April 2014 at 11:04

    To use the term ‘income’ for both wage and capital income is a monstrous abuse of language

    Okay, so how do you explain the title of your post? Waldman still has the best of this exchange. You’ve shown that taxes on wages + capital > taxes on wages. That’s not interesting at all. This argument only develops strength in a world with very high nominal returns or over extremely long time horizons. That’s when a 20% compounded tax can actually grow larger than 50%. But the argument is still fairly weak because it is quite easy to avoid realizing capital gains on a yearly basis by simply holding equities.

  17. Gravatar of Steve Waldman Steve Waldman
    24. April 2014 at 11:30



    It is not surprising to me that some theories suggest the optimal rate on capital is zero, but that’s not what you expressed in this post (and those theories are wrong). You said “captal income is taxed more heavily than wage income”. That is false. It is an assertion of fact that cannot be redeemed without abusing common language.

    Your second claim is more interesting. You argue on the basis of present value that taxation renders the present value of future consumption endowed by saving less than consumption that could be enjoyed today. But taxation has very little to do with that. To compare the present value of current consumption and of future consumption, we need a rate of return and a discount factor. If the rate of return is higher than our discount factor, we will find that the PV of future consumption is higher than that of present consumption. If our rate of return is lower than the discount factor, we will find the opposite. Capital effect the rate of return actually available for future consumption, so if we choose a discount factor a priori, we might find that under some circumstances your assertion is true: taxes cause future consumption to be less valuable than present consumption. But under some circumstances, the rates of return even after capital taxes is higher than the discount rate, and your argument is false, or the average rate of return is is lower than the discount rate even before taxes, so taxes aren’t the issue and your argument is false.

    To distinguish these circumstances we need to determine the discount rate we intend to use to compute present value. At a certain level, that is arbitrary. I might claim to require $120 next year to be as satisfied as I would be with $100 in consumption today, so my discount rate is 20% and saving is not worthwhile with or without taxes. Or, I might be flush today and worried about a very uncertain future, and so be satisfied if I can have $80 a year from now for deferred $100 in consumption, in which case my discount rate is -20%, and taxes I might pay against a 5% opportunity don’t much discourage me.

    Rather than rely upon subjective time preferences, the usual approach to this issue is to assume that people discount future income at the best rate they can achieve at the level of risk they are willing to bear. Even if I’d be minimally content with $80 next year, I won’t except less than $105 if I can easily earn $105 by putting my money in the bank. So we use current market rates of return as our discount rate.

    But, and crucially, this logic requires that we use after tax market rates of return as our discount rate. If bank interest rates are 5% but interest is taxable at 5%, then the opportunity I will be satisfied with is 4%, and that is the rate by which future income would conventionally be discounted. Of course, that 4% may be much more or much less than the discount rate of my time preference, but market rates, after tax market rates, determine the rate by which I will actually judge alternative consumption paths. I’ll eat today if that 4% is too little, I’ll save if it’s too much. In either case I’ll value $104 in the future at no more than $100 today, because I’d only need $100 today to turn that into $104.

    So, tautologically, you are mistaken. Under the scenario you describe, the PV of $86.58 14 years from now is precisely $50 today.

  18. Gravatar of Morgan Warstler Morgan Warstler
    24. April 2014 at 11:32


    In 1999, you earn $100 – you pay $50 in tax, you eat $50 or stick $50 in 14 years * .05, and your bank account has $100 (really 99). You pay 20% on the $50 gain or $10. You have $90 left to eat.

    So today in 1999 dollars you have $63 to consume.

    So today, thinking forward, you’d effectively get $13 more consumption if you wait 14 years to eat the $50 in your wallet.

    It’s not super compelling.

    You want to go after the wealthy, but since you don’t resist the impulse, you’ll take a short cut that gives everyone this not super compelling deal.

    I’m a fan of going after capital class, but it has to be done in a way that the top 1/3 LIKE.

  19. Gravatar of Jason Jason
    24. April 2014 at 11:42

    A useful definition of income is that which will allow you to consume without diminishing and/ or possibly increasing your assets.

    Focusing on consumption inequality alone can mask heavy borrowing by the poor and the middle class to maintain living standards.

    I understand the problems you mentioned above Scott, but income isn’t “meaningless.”

  20. Gravatar of o. nate o. nate
    24. April 2014 at 12:08

    Hi Steve, I was wondering if you’ve posted elsewhere perhaps at greater length about the flaws in the public financing theory justification for 0% taxes on capital. I think your arguments in this thread are spot on, btw.

  21. Gravatar of Floccina Floccina
    24. April 2014 at 12:42

    Someone will say: “My secretary pays a higher tax rate than me” and includes matching FICA in the secretary’s tax. That shows that they know about tax incidence. Yet, they will ignore the taxes on retained earnings on the capital gains and taxes on distributed earnings on dividends. They seem sincere so I think it shows how bias can lead us astray.

  22. Gravatar of Morgan Warstler Morgan Warstler
    24. April 2014 at 12:47

    As publishing goes, Steve mainly argues with me on twitter.

  23. Gravatar of Ed Dolan Ed Dolan
    24. April 2014 at 13:12

    Russ wrote:

    “On the other hand, if instead of investing my after-tax income I just worked enough to earn the same amount, instead of $90 I would have only $75 after 14 years (since we are assuming an income of $50 over 14 years, which if taxed at 50% yields an after tax result of $25).”

    This comment seems to have puzzled some people, maybe because of its phrasing, but I think Russ is on to something. Let me rephrase what I think is his point:

    “On the other hand, if instead of investing my after-tax income in bonds I invested it in human capital that had the same nominal rate of return, I would definitely be worse off than if I invested it in bonds.”

    That is the sense in which capital is taxed more than labor.

    Sure, Scott’s example works if you compare a stream of inflows that starts with a highly taxed labor inflow (I won’t use the forbidden word income) and then switches to lightly taxed financial inflow. Try starting with a $100 inflow you receive as interest on a previous investment and then investing it in human capital. You get the opposite result.

  24. Gravatar of Steve Waldman Steve Waldman
    24. April 2014 at 13:13

    o — the best i have is this:

  25. Gravatar of Floccina Floccina
    24. April 2014 at 13:13

    @Steve Waldman
    You wrote:

    “To compare the present value of current consumption and of future consumption, we need a rate of return and a discount factor. If the rate of return is higher than our discount factor, we will find that the PV of future consumption is higher than that of present consumption.”

    But if the two earn the same income then in your example, they would face the same discount rate and if the expected return where greater than the discount rate why would they not both invest the same amount and if it where lower than the discount rate why would they not both spend same amount?

  26. Gravatar of Steve Waldman Steve Waldman
    24. April 2014 at 13:23

    Floccina — People have different subjective rates of time preference. Some really want future consumption or the insurance that comes with savings, some really prefer consumption today.

    Those preferences don’t alter the achievable rate of return they face, which is identical for both. Sometimes available investment opportunities net of all intermediation and tax costs will return more than one person’s time preference, but less than another, so they’ll behave differently. And indeed, real investors do behave differently! They don’t all refrain from saving because investment income is taxed.

    It’s incoherent to discount future values at subjective time preferences, though. Even if I prefer future money to present money, the present value of a future dollar is never more than $1 as long as I’m allowed to just hold dollar bills. On the other side, if I really prefer money now, it’s incoherent to discount future money at more than the rate at which I can borrow (including any fees and taxes involved in the borrowing), since I can turn future dollars to present dollars by borrowing.

    Time preferences differ, but discount rates are bounded below by achievable investment returns and bounded above by borrowing rates. All after fees, and costs of intermediation, and yes, taxes.

  27. Gravatar of Russ Abbott Russ Abbott
    24. April 2014 at 13:57

    Scott, I still don’t get it. You said I should “either assume the $100 faces a tax in both cases, or not. You can’t assume different scenarios for tax today for the current and future consumption scenarios.”

    How am I assuming different scenarios? If my income is $100 today, I can either spend or save the after-tax portion. If I spend it today, I’ll have $80 to spend today in my income is from capital or $50 to spend today if my income is from wages. That favors capital. On the other hand, if I save the after tax portion for 14 years and the result doubles, I’ll have $90 if my current $100 is from wages and $144 ($80 + 80% of an additional $80) if my current $100 is from capital. Again this favors capital.

    Capital comes out ahead in both cases–and for the same reason. After taxes today I have more left over if my $100 is from capital than if it is from wages. Whether I spend it today or save it for 14 years, I’m going to be ahead if I started with more, which is what happens if my income is from capital. That seems like such a simple idea that I don’t see what’s wrong with it.

  28. Gravatar of Russ Abbott Russ Abbott
    24. April 2014 at 14:12

    P.S. And as Ed Dolan pointed out, if the additional income during the 14 years is from (an investment in a source in income that is counted as) wages, I’ll wind up with less than if the additional income during the 14 years is from capital.

  29. Gravatar of Russ Abbott Russ Abbott
    24. April 2014 at 14:14

    P.S. And as Ed Dolan pointed out, if the additional income during the 14 years is from (an investment in a source in income that is counted as) wages, I’ll wind up with less than if the additional income during the 14 years is from capital.

    (For some reason WordPress is not letting me post this. Perhaps this additional comment will confuse it enough to let it through.)

  30. Gravatar of BC BC
    24. April 2014 at 14:18

    Scott, this is one of my pet peeves too. People suffer from some sort of time illusion or mental accounting or something. They look at the investment “income” in one period and just can’t grasp (or don’t want to grasp) that it came from giving up consumption in an earlier period. If someone earned a wage of 100 apples and was taxed 20% to end up with 80 apples and traded those 80 apples for 88 oranges, then no one would say (I would hope) that those 88 oranges (or 8 oranges or 8 oranges less inflation) are also “income” that should be taxed again. Yet, when one earns $100 of wages at time 0 and is taxed 20% to end up with $80 and trades those $80 of immediate consumption for $88 of consumption in 1 year, then they insist that $8 or $8 less inflation is new “income”. A future consumption basket can be exchanged for a current consumption basket at the prevailing exchange rate (the interest rate) just like apples can be exchanged for oranges at some exchange rate (relative prices). Both exchanges have zero net present value, but (some) people just don’t get it when an exchange occurs across time.

    Somehow, because we use the same word, “dollar”, for consumption in two different periods, they think that any numerical difference represents income. That’s like saying that if I exchange 1 USD for 1.10 CAD to exchange consumption in the US for consumption in Canada, then I have earned 0.10 in nominal “income” or that amount less some purchasing power parity (PPP) model amount in real “income”. Neither PPP nor CPI nor GDP deflators are exchange rates; exchange rates are exchange rates and, by definition, both sides of the exchange are equal in value.

  31. Gravatar of Jeremy Goodridgde Jeremy Goodridgde
    24. April 2014 at 14:18


    I have to say I have read your analysis on this dozens of times in different posts, and you have yet to convince me. Labor income is just HUMAN capital income. I don’t see how to look at any other way. If you don’t believe in the concept of income, that applies just as much to labor income. In business, workers can be paid in stock, options, or wages. It’s easily switched around. So, if you tax wages you have to tax capital income. It’s the same.

  32. Gravatar of ssumner ssumner
    24. April 2014 at 14:20

    Steve, You are making this way too complicated. It doesn’t depend on complex PV calculations. Let’s start with a baseline tax system where wages are taxed at a flat rate 50%. All saving is put into 401k type vehicles and then taxed at the full 50% when the money is pulled out and consumed. You pay 50% on your original wage income (minus 401k contributions), and another 50% on all the capital income. So far as I know that “Unlimited 401k” type tax system is regarded as neutral vis-a-vis saving and consumption. And notice you pay a full 50% on all the capital income. But a tax system with no 401ks, and a 50% wage tax and a 20% capital tax actually taxes capital at a higher rate than my hypothetical neutral tax regime. That’s why Solow is wrong.

  33. Gravatar of BC BC
    24. April 2014 at 14:26

    Jeremy, what you say would be true if you were taxed when you were granted your human capital when you were born. If someone figured out the present value of your future earnings and taxed you at birth, wouldn’t you agree that subsequently taxing you again as you converted that already-taxed human capital into wages would constitute double taxation? Well, since you weren’t taxed at birth, the wage taxes are the first level of taxation. After the wage tax, taxing you again simply because you choose to spend your wages on future consumption rather than immediate consumption is indeed double taxation.

    Compare the following: (A) You receive $100 in wages and save it for 1 yr at 10% interest. (B) Your employer pays you $110, but you have to wait 1 year to receive your pay. Both (A) and (B) are identical; you end up with $110 in 1 year pre-tax. But, if taxes are levied on both wages and interest, then you will receive less after-tax in (A) than in (B). That’s because the interest is taxed twice.

  34. Gravatar of Rajat Rajat
    24. April 2014 at 14:38

    Is there a ZLB on IQ?…

    Well, I did study ‘public economics’ and didn’t learn this (I had to learn it on this blog), so maybe even most economists just aren’t well taught.

    The hardest thing for most people to accept is that the capital most people accumulate comes from their savings out of their labour (or their own business) income (profit). The punters see rich people driving around in expensive cars or living in expensive houses and assume that their money must have come from somewhere other than their exertions. In most cases, this is wrong – most people’s wealth derives either from their own labour, or at most the labour of their parents. Very few of the rich people around today are living primarily on the fruits of capital passed down from before WW2.

    So, Scott, you’re correct in your Econlog post in saying that most people see lack of thrift as a type of disability. This is even more so in highly utilitarian Australia where what carries the day is the situation people are in now, today, not their lifetime choices. So if someone has little income or wealth today, they should receive welfare, regardless of the decisions they may have made earlier in their lives to consume rather than save. It’s a corrosive attitude, penalising thrift, forethought and saving. Progressives now want to increase taxes on retirement savings rather than increase the age for social security.

  35. Gravatar of Name Name
    24. April 2014 at 15:30

    But where is the line between investment and consumption?

    Say you hire someone to invest your money for you and pay them a percentage of the gains. Should that person be taxed? Presumably yes, because that is the consumption of a service they provide.

    But if that’s the case, why is it different when someone is providing that service for themselves?

    And if someone invests a lot of their money successfully, and therefore retires much earlier than they would otherwise, then their investments have cost society their extra labor that is no longer done due to retirement.

    I’m not arguing for a particular position here… I’m just confused.

  36. Gravatar of Rajat Rajat
    24. April 2014 at 16:14

    Name, tax systems have long had to grapple with the difference between income and capital gains. But there are court cases that deal with this. As for your funds management example, how is that any different to what happens now, when we do have taxes on capital gains? The fact is, stuff you do for yourself is not taxed under any tax system, whether it is managing your investments, mowing your lawn or caring for your kids. Someone who is completely self-sufficient pays no tax.

  37. Gravatar of TravisV TravisV
    24. April 2014 at 16:22

    “Tokyo CPI Spikes To Highest Since 1992 (Well Above Abe’s 2% Target)”

  38. Gravatar of Morgan Warstler Morgan Warstler
    24. April 2014 at 16:29

    “Name, tax systems have long had to grapple with the difference between income and capital gains”.

    No they haven’t they have chosen to chase capital gains and tea them.

    Again there are ways to get at the capital class without taxing capital gains.

    HINT: the trick is to make them UNABLE to find a safe investment. Think of it like a different kind of inflation.

  39. Gravatar of ray ray
    24. April 2014 at 17:30

    Let me try to restate Scott’s argument. It is not a criticism of Scott but I just want to make sure that I understand Scott’s point.

    Suppose an individual earns $100. Also suppose that his tax rate is 50%.

    Case 1: Consumes after-tax earnings

    After paying $50 in taxes, he can spend the entire $50 on consumption. End of story

    Case 2: Deferred Consumption

    Same facts as above except that the individual forgoes current period consumption and invests it in a saving vehicle. The investment horizon is 14 years and the rate of return is 5%.

    Future value of $50 investment: $99
    Taxes on investment income. I assume there is a recovery of basis for tax purposes and the investment income is taxed at 20%: (99-50)*20%=$9.80
    Present value of taxes paid in year 14: $4.95 (14 years @ 5%)
    Amount available for consumption in year 14: $90.20.

    Points to note
    1. The person that consumes today pays $50 in taxes and enjoys consumption of $50
    2. The person that saves (i.e. forgoes current period consumption for future consumption), faces an additional layer of taxes that is not faced by a consumer.
    3. In Scott’s view, the distinction between labor income and capital income is misleading. How does labor income set aside for future consumption become capital income? Also, why is the government not taking into account the fact that the individual is making a sacrifice in forgoing current period consumption?
    4. By imposing a capital income tax, the tax code is essentially penalizing individuals that set aside current period consumption for future consumption. In this sense, I think capital income tax is problematic.

  40. Gravatar of ssumner ssumner
    24. April 2014 at 18:35

    BC, Nice simple elegant example.

    Rajat and ray, Good comments.

    Name. There are some problems with distinguishing between consumption and investment. I’d regard college tuition as investment and three martini business lunches as consumption. But these problems arise under any tax system. A good reason to keep rates as low as possible.

  41. Gravatar of J.V. Dubois J.V. Dubois
    24. April 2014 at 23:38

    This is another set of good posts and I generally agree with the gist of it. But as it is, sometimes devil is in details. So far the best piece that (at least for me) illuminated the debate is this comment by Mark Sadowski:

    Some conclusions:

    1) Despite of what Scott asserts, in reality wage tax and consumption tax are not in fact identical. For instance there are empirical studies showing different impact of these respective taxes on growth.

    2) While “capital” taxes are bad “property” taxes are good! Property taxes are estate, gift and wealth ones. Now I understand that some people may say that estate is actually capital and that therefore since taxation of capital is bad then estate tax is bad. This is not true.

    So again, devil is in detail. Just an example, compare these two claims:

    1) “I am for progressive consumption tax and abolishing capital income and payroll taxes. I am also for complementing consumption taxes wiht property taxes – especially effective ones such as land value tax.”

    2) “I am for abolishing estate taxes as it is just tax on capital which are bad. I am also for abolishing all other taxes including consumption taxes and increasing payroll tax as these two are basically *the same*”

  42. Gravatar of Ed Dolan Ed Dolan
    25. April 2014 at 06:00

    I don’t want just to quibble here, Scott, but there is something else that bothers me about your example

    You say: “Save the money [$50 after tax from labor earnings] for 14 years: $90 in consumption in 14 years [with a 20% capital tax], vs. $200 in a tax free world.”

    This seems to assume that the rate of return on the bonds (or whatever other financial instruments you invest in) does not depend on the tax rate. However, if we moved from a 20% tax on interest to a tax-free world, wouldn’t we expect that to raise the price of bonds and lower their yield? In that case, wouldn’t the $200 be less than $200?

    Of course, a change in labor tax might change wages, too. Shouldn’t we rework the whole example to take these effects into account? Or is there some reason they don’t matter?

  43. Gravatar of Nick Nick
    25. April 2014 at 07:39

    I generally agree with your arguments but I think you are oversimplifying here. The real issue of “unfairness” as it relates to capital taxation is not the case of the doctor who makes $100 and has to decide whether or not to invest it and pay income tax on the $100 and capital gains taxes on the investment. The problem is with an executive who gets paid $5 in income, and gets a $100 loan by his company to buy $100 worth of stock, or receives $100 worth of at the money stock options. I think when a lot of people argue for higher capital gains taxes, they are thinking about making sure that there is sufficient income taxes paid in circumstances like this.

    That’s why the Buffett example with his secretary is so effective. People understand that the secretary is getting paid for doing her job, and pays income taxes at, say, 25%. They further understand that Buffett’s job is to invest, and therefore his investment earnings are really earned income! Similarly, wealth accumulated by company founders was never taxed under the income tax, even though most people would say that they are equity owners not because they invested after-tax earnings, but rather, it’s sweat equity, and really just earned income.

    Which highlights that maybe the best tax is a simple consumption tax. If you buy a yacht, you pay a ton of taxes.

  44. Gravatar of Adam Adam
    25. April 2014 at 13:16

    We’ve had this discussion before. Your view describes a word where all invested capital was initially taxed as wages.

    That is not the real world, where invested capital can come from funds that were inherited (yes, potentially another layer of tax) or, importantly, earned in ways that were never taxed as wages.

    Sure, it’s theoretically true that Louis Winthorpe III’s hedge fund could have earned him more in inverse proportion to the wage taxes that poor Louis Winthorpe Sr. paid, but it’s not true in any practical sense. The younger Winthorpe is not so foolish as to invest only his own money, and, of course, knows full well how to make use of leverage. The $100 million he earned last year from investing Muffy & Buffy’s trust fund was structured as capital gains if he was doing it right, and was constrained by his ability to get others to give him money to invest, not his personal wealth. Where’s the wage tax biting him with respect to that $100 million?

    I guess we can assume that Muffy & Buffy’s benefactors paid wage taxes on the seed money that led their trusts, and thus assume that Winthorpe could have earned even more in the wage-tax-free word, but that’s getting pretty far afield.

    Which seems to me to be the disconnect. If you put a lot of faith in the assumption that all capital was once wages that were subject to tax, then the textbook analysis is right. But how far should we be going back and how much other complexity should we assume away when setting policy?

  45. Gravatar of Negation of Ideology Negation of Ideology
    25. April 2014 at 17:27

    Scott – Shouldn’t we be comparing equal consumption scenarios rather than equal wage scenarios?

    “A. Spend the money today: $50 in consumption, vs. $100 in a tax-free world.

    B. Save the money for 14 years: $90 in consumption in 14 years, vs. $200 in a tax free world.”

    Should be

    A. Spend the $100 today, and have the need to work for $100 14 years from now to consume at that point.

    B. Save the money, and not need to work to consume $100 14 years from now.

    Person A and B would both consume $200 in a tax free world. Person A would consume $100 in the 50%/20% labor/capital tax regime you describe. Person B would consume $90.

    I realize the percentages are the same, 50% for A and 55% for B, but it seems to me to be a clearer comparison, especially for an advocate of progressive consumption taxes.

  46. Gravatar of ssumner ssumner
    25. April 2014 at 17:40

    JV, I agree that a wage tax and VAT are not identical in practice.

    I favor taxes on land and homes, but not other types of property like office buildings and factories.

    Ed, I suppose the numbers would change, but not the principle that a pure consumption tax is neutral between current and future consumption. By analogy, a 5% sales tax on all goods will surely affect the relative price of goods, at least slightly, but it’s still the baseline assumption to say the tax applies equally to all goods. You wouldn’t say that a 5% sales tax applied to all goods actually “favored” one particular good.

    Nick, There are some tricky issues with the self-employed, I certainly agree on that point. Of course those issues also occur with our current system. So we have to rely on the tax authorities to make wise decisions (unfortunately.) But that only applies to a tiny segment of the population. For most people, the system could be radically simplified.

    There’s also the 401k approach, make Buffett pay a wage tax on all of his income, from whatever source, when he spends it.

    Adam, Even if the current distribution of wealth is unfair, the consumption tax is still optimal. The only implication is that we should have a one-time wealth tax first, to grab the ill-gotten gains, then a consumption tax afterwards.

    People who earn money managing hedge funds should have their earnings taxed as labor income. There are many areas where I think the rich are currently paying too little in the way of taxes. But don’t let the perfect be the enemy of the good.

    It would be politically impossible to enact my proposed consumption tax without closing those loopholes and making the rates highly progressive.

  47. Gravatar of dtoh dtoh
    27. April 2014 at 04:54

    Reposting here. A couple of things.

    1. Nominal rates are a poor measure for the real rate of taxation on capital because they fail to take into account capital losses. For certain types of investments (e.g. equity investments in start ups) where returns are often negative and are also highly asymmetric (many losers and a few winners), low nominal rates (i.e. 23.8%) can easily result in real tax rates (taxes collected/(capital gains – capital losses)) that are well over 100%. [In contrast, there is no such thing as negative wage income.]

    2. Land and buildings tax – Yes, but based on use not asset type so any asset whose original purchase price is over $50k and which is used for private consumption is subject to an asset tax.

    3. Thus – a) national progressive sales tax, b) an asset taxes at the state and local level, c) no capital, corporate, wage, or other taxes.

    4. You might need a flat wage tax during a transitional period.

    5. You’re wrong on taxing hedge fund profits. (Read David Brooks recent editorial). If you do a separate post on this topic, I’ll debate you there, but I think you can argue that any tax on wages above some base level wage is effectively a tax on capital (albeit human rather than physical capital). Why should someone get taxed higher because they invest in a restaurant rather than a PhD in economics?

    6. Eliminate charitable deductions. Whey subsidize Gates/Buffet et al for mis-allocating capital.

  48. Gravatar of ssumner ssumner
    27. April 2014 at 14:56

    dtoh, Human capital formation is heavily subsidized, physical capital formation is not.

  49. Gravatar of Kolobok Kolobok
    28. April 2014 at 04:29

    Prof. Sumner,
    You say: “While reading a recent book review by Robert Solow I came across a statement that made me want to pull my hair out.”

    I was wondering: how come neither you nor any of your commenters mention which book review by Solow. My impression: the deliberate omission not to mention the specific context could be due to either intellectual dishonesty or pure demagogy. Or it could be both?

  50. Gravatar of ssumner ssumner
    28. April 2014 at 04:37

    Kolokok, Neither. Two points:

    1. I answered your question in the comment section to the very next post.

    2. You are an idiot.

  51. Gravatar of JL JL
    29. April 2014 at 02:25

    You are right that the current tax rate does not tax capital more heavily than labor.

    But the point is, that the current economic system more heavily favors capital and the rich who have capital to invest versus the poor who only have their labor and not enough discretionary income to build a real capital base.

    And therefore capital is not taxed nearly enough to achieve a meritocratic democracy (and IMHO a true democracy implies a certain level of economic, as well as political, equality), which is what we all favor.

    Under scenario (B) the capitalist still has 80% more to spend ($90 versus $50), which affords him various privileges that the wage earner does not possess.

Leave a Reply