First Greg Mankiw wrote a good piece in the New York Times about how there’s sometimes a hazy line between ordinary income and legitimate capital gains. Then Uwe Reinhardt wrote a puzzling (at least to me) followup in which he concluded that we might as well just give up and tax both at the same rate. I have some questions for Professor Reinhardt.
|
Professor Reinhardt goes on to instance the case of a person who buys a vacation home for $500,000 and sells it two years later for $1.5 million, suggesting that it would be unfair to let this person hang on to all of this gain, so it should therefore be taxed at the same rate as ordinary income. This brings me to the next questions:
|
|
If you’re determined to tax the homebuyer’s gain at some rate other than 0% or 100%, how do you decide what that rate should be? In particular, why, in your column, do you want to equate it to the tax rate paid by an unrelated doctor? Why is the doctor’s tax rate any more relevant than, say, the price of margarine? |
|
Can you please write down the optimization problem which has as its solution “tax the doctor’s income and the homeowner’s capital gain at the same rate”? |
The above is, I think, the most important of my questions. I’m familiar with the vast literature arguing for a capital gains rate of zero, and I’m familiar with some of the dissenting literature arguing for a rate somewhat higher than that. But no matter how you define “optimal”, it seems like it would require an extraordinary coincidence for the optimal rates on capital gains and ordinary income to be identical. I’d like to understand what led Professor Reinhardt to expect such a coincidence, and nothing will make that clearer than an explicit statement of the problem he was solving.
Moving on, Professor Reinhardt does raise the twin spectres of fairness and “social engineering” (suggesting that the latter is somehow involved in the failure to tax capital gains), so I’d like to ask just two more questions toward understanding what he means by those terms.
|
Jack and Jill each work a day, and each receive two apple pies in wages. They each pay one apple pie in taxes, leaving them with one pie each. Jack eats his pie. Jill trades her pie for three Oreos and eats the Oreos.
|
|
Jack and Jill each work a day, and each receive two dollars in wages. They each pay one dollar in taxes, leaving them with one dollar each. Jack spends his dollar. Jill trades her dollar for three future dollars (via the stock market, or by lending it at interest).
|
These last two questions go to the heart of many popular misconceptions about capital taxation, which I’ve tried to combat in related posts here, here, and here.
Professor Reinhardt has visited our comments section in the past. I’ll be very glad if he has the time, patience and inclination to drop by again. If he does, then in the interest of a mutually enlightening discussion, I’ll probably be a bit more vigilant than usual about making sure that posted comments stay both civil and on-topic.
Can you put a link up to a paper you think is easiest to read which discusses the dissenting literature you reference in paragraph 3. Thank you.
Bradley Calder: As far as I’m aware, the hottest dissenting literature right now is the work of Saez, but none of it is particularly easy reading.
One of the arguments for taxing capital at a positive rate comes from the new public finance literature, see for example Werning’s paper on nonlinear capital taxation, http://dl.dropbox.com/u/125966/implementation.pdf
Not an easy read as well…
Marek: Correct me if I’m wrong, but I don’t think you’ll find anything in that literature suggesting that the capital gains rate should be any simple function of the rate on ordinary income.
Steve: That’s correct. But it argues for capital gains rate greater than zero. I understood that the dissent was only about that.
Odd. I’m not an economist but I find the economist version of Q3 much clearer than the layman’s. It’s a darned fine question.
I’m not sure how question 2b follows. It seems to me that the logical question to ask would be “should the government subsidize capital losses at the same rate at which it taxes capital gains?”
I suppose the question you posed responds to the potential argument that we tax capital gains at 100%, and therefore you ask whether we should subsidize losses at 100% (i.e. guarantee against losses). But if we’re trying to decide if there is an optimal non-zero capital gains tax, could we also expand this comparison to a non-zero, non-100% tax?
I hope that makes sense.
Gervais isn’t what I’d call easy, but I think it’s a little easier than Saez. (That said, I always say that if your results depend on strong overlapping generations style preferences [we all agree that future generations should have zero weight], then you also need to agree that we should increase debt as much as possible and try to pollute a bit more.)
http://dilbert.com/strips/comic/2009-02-24
Marek:
Steve: That’s correct. But it argues for capital gains rate greater than zero. I understood that the dissent was only about that
Right; that was what I meant by the dissent. Thanks for this helpful link.
How about this? The gummint gets one bite at the apple. Everything anyone does with after=tax income is immune from further taxation.
Optimize what? The earnings of productive people? Don’t tax any more than barely enough to fund the army, the police and the courts.
Regards,
Bill Drissel
I stopped reading when Reinhardt suggested that we “desist from using the tax system for any kind of economic or social engineering.” The only way to do that is to abolish the tax system.
Answer to Question 3: The problem is how the taxing authority can minimize cheating.
Since “sweat equity” is already taxed as a capital gain, maybe it’s not even cheating — just legal tax avoidance. But taxpayers have an incentive to try to re-characterize wages as capital gains, as long as the tax rate for capital gains is lower.
I believe the reasons for not taxing capital gains are more compelling than the government’s interest in deterring tax avoidance.
you could argue that the lower tax rate for capital gains is compensating for the riskiness involved in capital investment, where as there is little risk in a doctor’s salary because it is constant.
I find that he suggest that homeowners not be entitiled to capital gains laughable. Homeowner/propert owners are becoming capital owners/managers. They are a company of one (or few) responsible by their own hand or via contract for the maintenance or improvement upon sed property. Assuming we dont want the country waining into a nation of slum neighborhoods there is every reason to reap the social benefit of upkeeping property. In fact that serves as the reasoning why many believe that not only does a capital gains rate encourage imrpovement but also, per a discussion with Dr. Inman during some time at Penn, many local and state property taxes should be reformed to tax land rather than structures. As homes or properties increase in value so does the land, thus continuing a revenue, but not discouraging development/improvement. Especially true in old cities with severely dilapidated neighborhoods where the combination of capital gains incentive and reformed property tax incentive could work to encourge revitalization… so returning to the bigger point of capital gains, it again, gives me great confusion as to why one would dismiss the value of an incentivized capital gains rate regarding property owners.
I read an argument for capital gains taxes years ago in TNR by Michael Kinsley when he was editor.
Imagine Jack and Jill each have an initial endowment of $1. Jack invests his $1 in college and receives $3 in future wages. Jill buys Apple stock and receives $3 in capital gains. Why tax investment in human capital at higher rates.
I guess one could resort to a “tax incidence” argument to respond. That is, the workers at Apple will wind up paying Jill’s tax, because she will invest overseas and not repatriate.
Any other thoughts?
Steve,
I’m not sure if Questions 4 and 5 were meant to relate to each other, but they don’t. In 4, Jill has simply done a trade for a form of lunch that she likes better, but she hasn’t “invested” her pies. Three Oreos are simply the current market value (to Jill) of one pie. Whereas in 5, Jill forgoes consumption in order to achieve a better outcome for herself in the future (delayed gratification) – which I thought was the point of investment.
If you want 4 and 5 to relate to each other, then 4 should really be:
“Jill chooses to do without lunch today and sells her pie to someone who is STARVING hungry in return for the promise of three pies in the future. Assuming she is still alive by then, of course, is it fair that she should pay tax on the additional two pies she will gain by doing without lunch today?”
Or, in money terms:
“Jill chooses to invest her dollar instead of spending it, thereby doing without lunch. As a result of this Jill should be able to afford a slap-up meal later on because she has chosen to do without lunch today. Is it fair that Jill should contribute to a meal for Joe too from the returns on her investment?”
If you look at it like this it is clear (to me, at any rate) that the effect of taxing capital gains is to encourage current spending at the expense of longer-term investment. Perhaps that is what Professor Reinhart means by “social engineering”, but it isn’t what he seems to want.
Oh, and I’m not an economist either. That’s probably obvious.
Frances Coppola:
I’m not sure if Questions 4 and 5 were meant to relate to each other, but they don’t. In 4, Jill has simply done a trade for a form of lunch that she likes better, but she hasn’t “invested” her pies. Three Oreos are simply the current market value (to Jill) of one pie. Whereas in 5, Jill forgoes consumption in order to achieve a better outcome for herself in the future (delayed gratification)
In other words, Jill has simply done a trade for a pattern of consumption she likes better. Three future dollars are simply the current market value of one current dollar.
I don’t think it is too far afield to suggest that much of the difficulty here is caused by the taxation of income as a proxy for consumption.
A desire not to have the tax system skew the consumption/investment decision underlies most of the rationales for a zero capital gains tax rate because: (i) taxing only consumption is the way to avoid skewing consumption investment decisions; and (ii) not taxing investment income is theoretically equivalent to taxing income only when it is consumed.
The problem is that differentiating between investment income and ordinary income is difficult, especially when trying to keep in mind the ultimate goal of not skewing investment/consumption choices. It is one level removed from the real issue.
These problems would disappear if tried to tax consumption rather than income. I grant they would be replaced by other problems about how to distinguish consumption from investment (e.g. when you buy a house to live in)– but at least those problems would be closely related to the issue we were trying to solve, not problems that are derrivative of trying to tax consumption with an income tax.
Steve, you should try to separate the questions you have as a libertarian (e.g. “If it’s unfair for this homebuyer to keep 100% of his capital gain, why is it fair for him to keep 65% of it? If capital gains are per se unfair, why don’t we tax them at 100%?”) from the ones you have as an economist. It just confuses the issue. Reinhardt clearly wasn’t saying that it is inherently unfair for homebuyers to keep 100% — his point was that, in a world where we must collect taxes and we presently tax neurosurgeons at 35%, taxing homes at any less than that amounts to an implicit subsidy and is thus both economically inefficient (unless homeownership has positive externalities, which he seems to doubt) and unfair.
The answer to question 3 is pretty clear (and follows from what I just said): assuming no externalities, all capital gains should be taxed at the same rate. Given that both return on homeownership and return on medical school are capital gains…
JM:
assuming no externalities, all capital gains should be taxed at the same rate. Given that both return on homeownership and return on medical school are capital gains…
Ah. Does this assume, then, that the doctor’s earnings are a pure return to capital, i.e. that running a medical practice is a form of leisure?
JM:
Steve, you should try to separate the questions you have as a libertarian
It was unclear to me how much of Reinhardt’s argument was based on efficiency and how much was based on fairness, so I raised questions on both accounts. What I want is to understand his position better. If part of the answer is “I did not mean to invoke fairness, so the fairness questions are off the table”, or, for that matter, “I did not mean to invoke efficiency, so the efficiency questions are off the table”, that’s entirely reasonable.
On the other hand, it seems to me that on the most straightforward reading, there are a lot of appeals to fairness in that column.
Excellent post. I wish I could outline these points as well when I am discussing these issues with friends who view the world like Professor Reinhardt does.
When I use something like the Jack and Jill analogy, I suggest that Jill’s pie goes to a middle-class family that cannot afford to buy a pie today, but agrees to give Jill three Oreos at a later time in return for the privilege of eating one pie today. (After all, we shouldn’t live in a world where only the privileged get to eat pies.) If Jill had to give half of the Oreos up as tax, then she would probably make the family give her more Oreos in exchange for the pie. That means it would be harder for the family to get a pie today. Is making it easier for middle-class families to eat pies without having to save for them the type of social engineering that Professor Reinhardt wants to end?
I similarly have a hard time assessing what is fair and what is not, but there are surely practical problems with taxing people at vastly different tax rates for similar economic activities. In Mankiw’s examples, Carl only incurs a 15% tax on the capital gains he realizes after fixing up his own house, while Earl incurs a 35% tax rate on labor income fixing up someone else’s house. For the wider economy these two activities surely look alike.
This now opens up a nice profit opportunity for enterprising lawyers to disguise Earl as a (co-)owner using creative enterprise structures and phantom loans. It might also induce Earl to stop reporting earnings and enter the black economy. And I think it is highly likely that Carl is slacking off on his day job, forgoing promotions and underachieving because his wage earnings are taxed at much higher rates.
I live in a country with a zero tax rate on capital gains and a marginal tax rate on labor income (including payroll and local taxes) around 70% for most people. I can assure you that giving people the *opportunity* to save 70% in taxes if they can disguise their earnings as capital income leads to very wasteful organization of work. In Belgium you can see every day how the entire economy is organized around shifting the tax base to the low rates, and I am sure Italy and Greece are pretty similar.
It cannot be difficult to build a model that spits out the same rates as optimal if you introduce an evasion technology that is cheap enough. I think you are underestimating people’s willingness to game the tax system if they perceive it as unfair.
rich,
That’s a good point, and there’s certainly an argument that if we can’t tax human capital at a zero rate, maybe we should tax other kinds of capital at least somewhat. Two thoughts though
(1) The argument implies we’re trading off between two inefficiencies: (A) we want capital to be taxed as little as possible, but (B) we also want all capital to be taxed at the same rate. It’s very unlikely the solution is going to be to do B as much as possible and ignore A.
(2) We subsidize education a lot. We do this even though we have some evidence that a lot of education is signaling, which is a negative externality on the margin. So it’s at least a little ambiguous whether education is taxed too much.
Jo Van Biesebroeck:
It cannot be difficult to build a model that spits out the same rates as optimal if you introduce an evasion technology that is cheap enough.
Good point.
On the other hand….We start with the observation that it’s easy to disguise income as capital gains, so we tax capital gains and income at the same rate. Then we observe that it’s easy to disguise capital gains as interest or dividends, so the next thing you know, we’re also taxing interest and dividends at that rate. I don’t entirely like where this is going!
Anyone who lives somewhere where the marginal tax rate on income is ~70% is someone who lives under a highly inefficient tax regime.
The solution to that problem isn’t to take a different tax rate which is efficient and make it highly inefficient; rather the solution is to reduce the initial inefficiency.
At the risk of straying off topic, it is highly plausible that reducing the marginal tax rate on income from 70% to something considerably lower might actually increase tax revenues collected, becoming in essence an extremely rare free lunch. This is especially likely to be so if people are saying things like, “everyone knows there is massive tax avoidance going on” by people in the above tax bracket….
Before the haters chime in on my last paragraph, I want to preemptively point out that not all such tax rate reductions lead to higher tax revenues, which is a point that is “seemingly” lost on some politicians.
The problem is that differentiating between investment income and ordinary income is difficult…These problems would disappear if tried to tax consumption rather than income. I grant they would be replaced by other problems about how to distinguish consumption from investment (e.g. when you buy a house to live in)– but at least those problems would be closely related to the issue we were trying to solve, not problems that are derivative of trying to tax consumption with an income tax.
Excellent summary.
We subsidize education a lot. We do this even though we have some evidence that a lot of education is signaling, which is a negative externality on the margin. So it’s at least a little ambiguous whether education is taxed too much.
You have overlooked the “need”-based financial aid system, which extracts those subsidies from higher income families by overcharging them for college education.
It cannot be difficult to build a model that spits out the same rates as optimal if you introduce an evasion technology that is cheap enough.
Len Burman had a similar thought at http://www.forbes.com/sites/leonardburman/2012/03/15/capital-gains-tax-rates-and-economic-growth-or-not/
“Low capital gains tax rates do accomplish one thing: they create lots of work for lawyers, accountants, and financial geniuses because there is a huge reward to making ordinary income (taxed at rates up to 35%) look like capital gains (top rate of 15%). The tax shelters that these geniuses invent are economically inefficient, and the geniuses themselves might do productive work were the tax shelter racket not so profitable.”
I have to admit, I’m not a fan of Mankiw. He’s all over the place. A few weeks ago on this blog he was praised for his wisdom when he wrote an article in the NY Times about 4 ways to a better tax system. I wrote at the time that I thought it was a fluff piece, hopelessly short on details. Rule 4 in the fluff was “Keep it Simple Stupid.”
So I was intrigued when I saw last weekend that Mankiw had written a piece about carried interest. But by the time I finished reading it, all I could think was the whole point of the piece was obfuscation. It’s a pretty straightforward question: Why should investment managers have the compensation for their labor taxed at a far lower rate than all other professionals? But somehow, now that he’s a leading economic advisor to Romney, Keep it Simple Stupid isn’t so simple anymore.
Mankiw didn’t always feel this way. He wrote about the taxation of carried interest on his blog in 2007:
“Deferred compensation, even risky compensation, is still compensation, and it should be taxed as such. Paul Krugman hit the nail on the head with this question: ‘why does Henry Kravis pay a lower tax rate on his management fees than I pay on my book royalties?’
The analogy is a good one. In both cases, a person (investment manager, author) is putting in effort today for a risky return at some point in the future. The tax treatment should be the same in the two cases.”
http://gregmankiw.blogspot.com/2007/07/taxation-of-carried-interest.html
Twofer: I agree with Mankiw circa 2007 that deferred compensation is still compensation and should be taxed as such. I also agree with Mankiw circa 2012 that there’s a somewhat hazy line between deferred compensation and capital gains. These views are not in direct conflict.
“Low capital gains tax rates do accomplish one thing: they create lots of work for lawyers, accountants, and financial geniuses because there is a huge reward to making ordinary income (taxed at rates up to 35%) look like capital gains (top rate of 15%). The tax shelters that these geniuses invent are economically inefficient, and the geniuses themselves might do productive work were the tax shelter racket not so profitable.”
The key here is not that capital gains tax rates are low; it is that they are lower than ordinary tax rates.
Actually, that is the key point about “carried interest” (the main topic of the original Mankiw article). Carried interest involves tax arbitrage between tax exempt private equity fund limited partners (the great majority of PE funding in carried interest deals comes from pension funds and tax exempt university endowment funds, etc). Tax exempts are “trading” their tax benefits with the managing partners of private equity funds by giving up capital gains and ordinary deductions they would get if a higher management “fee” were paid (neither of which have any meaning to them) in return for lower management fees. This is classic tax arbitrage. Economists like to focus on the justification of lower capital gain rates, but they most often miss the central point about carried interest, presumably because they don’t understand the tax rules involved.
If one wants to complain about preferential tax rates, I would recommend that we focus first on the complete exemption we give to these “not for profit” enterprises. These “civic minded” organizations are at the heart of the carried interest problem.
Steve,
I’m not an economist (philosopher by training) so caveat emptor on the following thoughts. But the one aspect of the article that made me think was the bit about capital formation and human capital. From an economic perspective I don’t see a meaningful difference between human and physical capital. Inventing and manufacturing a machine that does complicated brain surgery and just teaching a human to do it are not all that dissimilar, even though the tax treatment of the ‘returns’ from that investment would currently be very different.
Accept this equivalence between types of capital, for the sake of argument. Reinhardt makes the inference that if we tax human capital formation then we should tax all capital formation. I think the better inference to make is that we shouldn’t tax human capital formation either, and heavily tax non capital-forming activities, i.e. consumption.
In short, pointing out the equivalence of physical and human capital strikes me as the beginning of a simple and highly persuasive argument for consumption-based taxation.
Steve,
Does the various literature on optimal capital gains rates account for the incentive to structure income as gains when the rates are meaningfully different?
Regardless of what the merits of capital gains taxes are in isolation, it seems their level relative to other tax rates has implications.
If we index capital gains to inflation, eliminate the corporate income tax, and tax all individual income at the same rate, is this still a sub-optimal scheme?
Thanks,
Brian
Brian:
Does the various literature on optimal capital gains rates account for the incentive to structure income as gains when the rates are meaningfully different?
Mostly no.
If we index capital gains to inflation, eliminate the corporate income tax, and tax all individual income at the same rate, is this still a sub-optimal scheme?
According to the usual Chamley-Judd analysis, this is still highly suboptimal if “all individual income” includes interest and dividends.
Richard Williamson: Thanks for putting this so well.
Human capital may be different from physical capital in that investment in human capital might have decreasing returns to scale. Factories for building capital can be duplicated, but the gains from learning basic math, reading, and writing may be much larger than the gains from going to college. And a high school education might be cheaper than a college one too. I don’t know, but I think that the zero-taxation result in the Chamley-Judd analysis depends on constant returns to scale to investment and so only applies to human capital if there’s constant returns to scale to human capital.
Landsburg wrote: Can you please write down the optimization problem which has as its solution “tax the doctor’s income and the homeowner’s capital gain at the same rate”?
You know, Steve, sometimes it seems like you go out of your way to not understand someone’s (probably wrong) point…
Suppose we learned that Obama had secretly ordered the IRS to give full 2011 income tax refunds to everybody who had worked as a community organizer with Obama in Chicago back in the day. Sean Hannity (or even another popular radio guy) goes ballistic, saying the IRS shouldn’t be giving people special privileges based on their personal association with Obama.
Would you have a list of questions for Sean Hannity (or his staff economist I should say), one of which was:
“Can you please write down the optimization problem which has as its solution “tax the non-Obama-friend’s income and the Obama friend’s income at the same rate”?
Or would you ask, “You are saying Hannity that it’s not fair for the community organizers to keep 100% of their money. Well holy cow, I guess you are in favor of freezing wages then. Do you understand the consequences of your emotional tirade?” ?
To be clear, Steve, I’m not even disagreeing with the taxation of capital gains versus labor income point (though we *have* crossed swords on that, at a theoretical level, in the past), but I think some of your questions here are needlessly patronizing. It was obvious to me what the guy’s point was, and yet you seem to be unsure of what he was saying or why it might seem unfair for a capitalist to earn an income and not pay the same tax rate on it that a brain surgeon pays on his income.
“Can you please write down the optimization problem”
I am going to make the same point that JM and JvB made above in a more formal way. Not because I am good at formalizing, this is just how I have been thinking about it.
We want to collect a total amount of tax revenue, T. We want to collect it in a way that is economically least damaging. Ignoring other sources, we are going to tax capital gains, C, and labor income, L. If we tax C at rate q and L at rate r, the revenue collected is T=qC+rL. As long as q and r are different, people have an incentive to move efforts and nominal income between C and L. There are two problems with this: one is that this artificial incentive creates an economically damaging distortion. The other is that it makes it harder to acquire target T (assuming qC and rL are both significant inputs).
There is another problem. I agree that it would “require an extraordinary coincidence for the optimal rates on capital gains and ordinary income to be identical,” but this statement implicitly admits that we don’t know what the optimal rates are. It would also be an extraordinary coincidence if we happened to hit on the two optimal rates, or even one of them. If we assume the two rates will, for now, be arbitrarily different from the optimal ones, is knowing the two optimal rates will be different an argument for keeping the two arbitary ones different?
PS: There is one case where the two rates could be the same, namely if Bob Murphy is correct and the optimal rate for both is zero. I think the chance of this is better than extraordinary coincidence.
@Bob Murphy: Steve could ask that question, and any answer would contain a function that assigns a high value to friendship with Obama. That would make the claim to such a privilege explicit, identifiable, and controvertible.
If there is a legitimate economic case to be made it should be possible to make it explicit enough for an equation.
As for the rest of your comment, Steve is trying to remove wiggle-room. That’s the very opposite of patronizing, the essence of which is to refuse to take the other guy’s argument seriously.
@Henri Hein: I think you’re right and that this is pretty funny, in the standard Chamley sort of model the long run optimal tax rate on capital gains are zero but I think also with many utility functions the optimal tax rate on labor income is zero as well, the government lives off of the interest from everything it taxed initially. So the model which says “tax the doctor’s income and the homeowner’s capital gain at the same rate” _is_ the standard model, at least in the long run. I guess that’s not what the question meant, though.
Ken B:
@Bob Murphy: Steve could ask that question, and any answer would contain a function that assigns a high value to friendship with Obama. That would make the claim to such a privilege explicit, identifiable, and controvertible.
I think you got this backwards. Bob has me asking the question of an *opponent* of special privileges for Obama-friends.
@Steve, Yhprum P Trebor: I did get him backwards.
AMTbuff,
You have overlooked the “need”-based financial aid system, which extracts those subsidies from higher income families by overcharging them for college education.
I agree we subsidize it less for higher income people, but I’m not sure that cashes out to say that on net we tax education for high income people.
Steve Landsburg wrote: “Twofer: I agree with Mankiw circa 2007 that deferred compensation is still compensation and should be taxed as such. I also agree with Mankiw circa 2012 that there’s a somewhat hazy line between deferred compensation and capital gains. These views are not in direct conflict.”
Steve, I’m no tax expert so I turn to various writers to try to gain an understanding. Clarity never comes from Mankiw. In theory, I suppose, Mankiw in 2007 could be interpreted as not being in direct conflict with Mankiw in 2012, but as a practical matter, there most likely is a conflict.
Using the tax rates Mankiw laid out, 35% for income and 15% for capital gains, Mankiw 2007 indicates that carried interest should be taxed the same as income, which I would interpret to mean 35%. Mankiw 2012 indicates that carried interest looks a lot like capital gains, and although he never quite finds the gumption to come out and say it directly, his argument would indicate that carried interest should be taxed like capital gains at 15% (again, he never writes this directly and that is why I think it was really an obfuscation piece, supporting the low carried interest tax rate paid by Romney, the man to which he’s a lead economic advisor – In the end, Mankiw only finds the courage to write “some reform may be appropriate.”).
One way I see to resolve the apparent conflict between the theoretical Mankiw and the requirements demanded by the practical world is to have the same tax rate for income, carried interest, and capital gains (either 35%, 15%, or something else). In that case there is no practical world conflict between Mankiw 2007 and Mankiw 2012 as carried interest looks like both income and capital gains. The problem with this solution is that is very close to what Reinhart proposes (with subsidy incentives), and one to which you appear to object.
Rather than asking questions of Reinhart, why don’t you ask Mankiw specifically what tax rate he recommends for income, for carried interest, and for capital gains, and why he recommends those specific rates? After all, Mankiw himself writes “some reform may be appropriate.” It would be helpful to know exactly what he, and presumbably his client, thinks it would be.
Twofer,
Both you and Landsburg (and Mankiw) are confused about the nature of “carried interest”. As I indicated above, it is pure tax arbitrage and has absolutely nothing to do with whether what a managing partner earns should be properly classifed as capital gain or as ordinary income. A partnership earns a fixed amount of capital gain which the partners are generally free to allocate in any fashion they see fit. “Carried interest” does not alter the fixed amount of that gain–it refers to the practice by which limited partner investors voluntarily cede part of their CG allocation to the managing partner. The default rule is that the CG is allocated according to contributed capital, but partners can agree to deviate from this.
This should not be of any concern whatever to the Treasury *provided that* the partners involved are subject to the same tax rules and rates. As it so happens, the investors who are giving up part of their CG allocations to the managing partners are mostly tax exempt investors. They don’t care how their income is classified because they are not taxable in the first place. They also don’t care that the increased CG allocation to general partners (rather than ordinary compensation) comes with the loss of the deduction they would otherwise get, because those deductions are meaningless, too.
The whole issue as to whether CG rates are appropriate with respect to gains realized on the sale of corporate stock is a complete red herring to the carried interest issue. Mankiw’s analogy to “sweat equity” is not particularly appropriate in this context. The real issue is tax arbitrage among taxpayers who are not similarly taxed.
Jo Van Biesebroeck
Beat me to the answer to Q3. Reorganizing your income streams is, for society as a whole, a fairly unproductive activity.
Steve, you then responded:
“Then we observe that it’s easy to disguise capital gains as interest or dividends, so the next thing you know, we’re also taxing interest and dividends at that rate. I don’t entirely like where this is going!”
Reading earlier arguments against the taxation of interest, and trying to reason over what might be different about eating two scones today, or two in the future (or perhaps in this thread they’ve become pies) it would appear that you may be implicitly assuming that it is obvious one should be able to “save” without paying any cost of carry. Believe me, if you find a way I would love to know what it is, I think this is something we all want. But, I’m not sure it’s fair to take it as a given that we should be able to do so.
For your typical person who’s willing to defer his scone consumption, but feels he is getting a bad deal because while he could’ve purchased 2 today and eaten them, if he puts his money in a time deposit and then pays taxes he may only be able to afford 1.8 scones after inflation, I say to him, go ahead and buy two today and store them. Then you’ll be able to eat 2 in a year. If the response is that you can’t store them, then it would seem that it is perhaps not unreasonable that you would pay a premium to be able to move your scone consumption around in time.
As for the main thrust of the thread dealing with taxation, I think Jo Van Biesebroeck is correct, and I think it is heading right where you don’t want it to head (taxation of dividends and interest). Because, again, reorganizing income streams is possible. So the best answer is probably to have one rate — or to only tax consumption directly? If you do only have one rate for income of all types, then it would seem you definitely want to keep it low, or else you wind up with a huge incentive for people to not sell appreciated assets that they might be able to consume (and hence avoid paying taxes on) — like their home. This, again, is a sign that perhaps a consumption tax works better — though I have my doubts as to how easy it would be to really implement correctly.
”Ah. Does this assume, then, that the doctor’s earnings are a pure return to capital, i.e. that running a medical practice is a form of leisure?”
If the doctor didn’t have a degree, he’d still be working, and therefore still would not be engaging in leisure during work hours. The doctor’s return on capital consists of the difference between the doctor’s earnings with the degree and without the degree.
If the doctor’s job is, in addition, more “leisure-like” than the non-doctor job, the increased “leisure-ness” is additional earnings on the capital.
Question 3 (Economist’s Version):
Can you please write down the optimization problem which has as its solution “tax the doctor’s income and the homeowner’s capital gain at the same rate”?
This is easy. “Minimise tax”
I have an answer for Q3 Economist’s version: “Find the tax policy that minimizes the total resources spent trying to get ordinary income reclassified as capital gains (or vice versa) in the eyes of the taxing authority.”
What do I win?
@Vivian – lol. It’s ok if I don’t understand all the ins and outs of carried interest, I’m just a random screen name. It’s probably ok if Steve doesn’t understand. He’s an influential economist, but I don’t know that his area of expertise is tax treatments (Steve, apologies if I’ve sold you short on this front — no slight intended). But it is problematic if Mankiw doesn’t understand. He’s writing articles about it in the NY Times, and he’s a lead economic advisor to the Republican front runner.
The way I’ve been thinking about the issue is simply from a tax return perspective: The income Romney still collects from Bain Capital, as part of his retirement deal there, is eligible for the carried-interest loophole, which allows fund managers to have their compensation for investing other people’s money taxed as capital gains, not earned income.
Thanks for the insight into the overall arbitrage opportunities — Although I can’t say I find it anything but expected (at least at a high level). Ok, I’m off. Planes to catch and bills to pay…all to be taxed at regular income rates (I KNEW I should have taken that fund manager position..).
The problem I have with capital gains tax rates is when Jack by nature of the industry he’s in (money mangagement timber etc) earns two pies and pays 1/3 of a pie in taxes. While Jill does the same job in a different industry, but after she earns two pies she pays one in taxes.
Steve – thanks for this incredible dialog and debate. Just what we need to argue this through (hopefully to a positive conclusion.)
What would be wrong with a single tax rate for all income (i don’t want to use the term “flat”) – set at, say, 17% – no matter the source? Wouldn’t that eliminate all of the maneuvering by lawyers and companies and HNW individuals? You get a realized gain, or income, from any source, and it is taxed at a specific rate, regardless – no deductions, no loopholes, no exceptions. Would this allow for a focus on true economic activity instead of evasion and avoidance? How would this fit into your fairness paradigm?
SmallBusinessGuy: If by a single tax rate on all income, you are including income from interest and dividends, then you are still distorting the incentive to save. This is surely not efficient, and for what it’s worth, it also strikes me as unfair (per my questions 4 and 5 to Prof Reinhardt).
The whole premise is that those with similar ability to pay should pay the same. He says “horizontal equity states that taxpayers with similar ability to pay should contribute the same amount.”
His offers glaring example where this principle is violated. In his example, the principle would be restored by taxing capital gains the same as ordinary income. He does not demonstrate that this tax regime would generally ensure the same outcome.
He clearly states that economic efficiency may be sacrificed to this principle.
If we look at your questions in this light, perhaps we can see what answer might be forthcoming.
Q1 only applies if there is some underlying principle about how we should treat these things. An example from your list might be “All injuries inflicted on backs should be punished equally”. Then it would not matter if the injury was inflicted by a mis-judged pat or a deliberate blow, only that an injury occurred. No such principle has been given to connect the options, so the question is a bit meaningless.
Q2a. The principle states that both should pay the same if both are able to pay the same. Therefore the only thing that matters is equity of rates, not the actual rate.
2b. If a housebuyer sells at a loss, the loss should clearly be tax-deductable against other income since his ability to pay has reduced.
2c Moot
3. The principle that those with the same ability to pay pay the same. The doctor and the nerosurgeon both have the same ability at that time, so both pay the same. The price of margarine has no bearing, but the doctors salary does.
Q4. Neither has any ability to pay since they are paid in food which they consume, therefore neither should pay anything.
Q5a. The principle clearly says that ability to pay is the determining factor. When Sue realises her 3 future dollars, she becomes able to pay more, and therefore must do so.
5b. Possibly.
He has questions of his own: “By what definition of the term would can one call the glaringly differential tax treatment of the real estate investor and of the neurosurgeon horizontally equitable? Indeed, by what theory of justice could one defend it on ethical grounds?
I can think of some answers to these questions also.
Harold
No such principle has been given to connect the options
Q1. In fact, Reinhardt gives the principle to connect income tax rates with capital gains tax rates: sometimes it’s hard to tell the difference, so always treat them the same. This is exactly the same as saying that since it’s sometimes it’s hard to tell the difference between intentionally and unintentionally causing harm when hitting someone on the back, so when determining punishment intentions should never be considered.
Q2a. The principle states that both should pay the same if both are able to pay the same.
Ability to pay is your only criteria? Comparing a 30 year old who made $1M in 2011, with little savings, to a 70 year old with little income, but $1M dollars saved, have equal capability to pay the same amount of taxes for 2010. Should they pay the same taxes?
Q2b. If a housebuyer sells at a loss, the loss should clearly be tax-deductable against other income since his ability to pay has reduced.
Say I made $80K in 2010, but go laid off in Dec 31, 2010 and was unable to find a job in the entire 2011, can I write off $80K in 2011 as “lost” income?
Q2c. Moot
Why? How is wanting the captial gains tax rate to be exactly the same as the income tax rate different from wanting the price of a house in 2100 exactly the same as in 2012?
Q3. The price of margarine has no bearing, but the doctors salary does.
Why does margarine have no bearing, but salary does, when talking about tax rates on capital gains?
Q4. Neither has any ability to pay since they are paid in food which they consume, therefore neither should pay anything.
This is a weird statment to say the least. Are you saying that if we all the sudden went to a purely barter economy, no one would ever pay taxes again? How is being paid $10 worth of food different from being paid $10?
Q5a. The principle clearly says that ability to pay is the determining factor.
Ability to pay is definitively not what Rheinhardt meant; I’m quite certain he would disagree with the 30 year old and the 70 year old above paying the same amount of taxes despite each having the same ability to pay the same amount of taxes for 2010.
Ken Arromdee:
“If the doctor didn’t have a degree, he’d still be working, and therefore still would not be engaging in leisure during work hours. The doctor’s return on capital consists of the difference between the doctor’s earnings with the degree and without the degree.”
It seems to me if you want to include education as being capital, where do you stop? Would you consider the difference between the avg wage one normally eventually gets and lower wage one often accepts during the first few years of a complex career capital? After all, you could look at this as paying for your “real world” education.
Are all the arguments in support of one rate here based on the practical problems of tax shifting, or is there a good *principled* argument in favor of double taxation? Bill Drissell’s “one bite at the apple” seems like a pretty nice simple conceptual rule as well. Reinhardt mentions this only in the context of corporate taxes and CGs (dividends seems like a cleaner example), and my understanding is this indeed a main argument in support of a preferential rate for carried interest (the tax arbitrage stuff is interesting as well; perhaps this has provided one industry with a way to avoid the double taxation). But it seems to me a more basic argument is if Chen’s CG accrued to income she already paid individual income tax on. This is more like saying one of the pies Jack (or Jill) received already had a slice taken out of it. The vacation home example seems to skip this step as well, e.g. it’s purchased *by* the neurosurgeon with the *after-tax* money he earned through “many hours of hard, physical and intellectual work”. What “theory of justice” says he should be treated differently if instead he spent it on a more fully depreciating asset like a big yacht?
Also “no new capital formation was supported by this trade in a stock sold by the company years ago” is too facile — a more liquid market with buyers bidding up share prices makes capital raising more attractive.
I’d also quibble with the “may be” argument Reinhardt offers (in order to debunk it) on the issue of double taxation due to corporate taxes: stock prices will generally be influenced less by an actual increase in earning than *expectations* thereof, so rather than saying they were “already taxed, it would seem more relevant to argue the market discounts the price Chen receives to reflect the higher future taxes. More importantly, he seems to want to dismiss the whole issue as “fluctuations in optimism and pessimism about the future” and linking this to irrational “bubbles”. Some ‘on the ground’ investors might think such expectations for earnings are typically a bit more fundamental, and much “volatility” reflects the fact that equity markets attempt to do a very difficult job of valuing what very long (theoretically infinite) cashflow streams. Clearly risk premiums are an important component of this, and the proper tax treatment of that could be a very interesting topic, again if not so glibly dismissed.
Finally as long as we’re at it, is there a valid economic argument for why we have a corporate income tax at all, other than a way to fool individuals about their true tax burden?
Is it valid to use “tax on interest and dividends distorts the incentive to save” as an argument against tax on interest and dividends, but NOT mention that tax on wages distorts the incentive to work?
Silly questions …
Q1: If there is a good reason to do so and it can be done in a manner that costs less than it gains, these things should be distinguished. Otherwise you make assumptions – like that incorrectly cited quotes are plagarism, that a pat on the back isn’t assault, and that you’re dealing with someone that’s mature.
What is the good reason to distinguish capital gains from income? Prof Reinhardt outlines a few reasons not to, and most people object to the fact that certain highly lucrative professions essentially get their normal income taxed at a much lower rate. Why should we prefer this current situation rather than changing it?
You don’t make a positive argument, rather you negatively attack his.
Q2a: If you frame it like that, obviously no taxes are “fair” and everyone should keep what they earned no matter the means. Too bad we decided we needed to pay for some things as a country. How about you tax any monetary gain at the same base rate, then explicitly write subsidies for things you want to encourage (buying houses for example)? Would that not cause minimal distortion?
b: I thought this existed in current law? Capital losses?
c: This is a sarcastic dick question.
Q3: I don’t know. How did they come up with 35% as the rate for income? What is the equation that came up with 15%? Doesn’t seem that precise of a number…
Why is 15% closer to optimal than 35%? What do you base that on? Just because it’s lower? If Reinhardt would have suggested lowering income tax to 15%, would you agree with his argument?
They are relevant to each other because they are both monetary gains realized by individuals. Why don’t you explain why they’re not relevant?
Q4a: “Fairness” again. I do regret that Obama made this a mantra. It’s annoying. Sounds like 3 oreos = 1 pie, I don’t really see a problem here. Maybe I’ll understand with the next zinger of a question.
b: I don’t see why it would be an instance of social engineering to not tax people on trades. If you want to trade money for a car, cool. Drive that baby all over the strip. If you then sell the car and realize $100 dollars income, I want to tax it like any other way you make money. Just like I realize my investment of time in my paycheck every other week. Is that social engineering?
Q5: What is a “future dollar”? If we want this to be in analogy to Q4 (and I think you did, somehow), I assume it’s an oreo that is worth $1/3 now and appreciates to $1 in the future. And the answer is, just as before, that I wouldn’t tax your purchase of future dollars or oreos, but if you then sold the oreos for pies (or realized the future dollars) I would tax them.
But the real irony here is that you suggest he gain “future dollars” as either stocks or as interest …. which are TAXED AT DIFFERENT RATES BECAUSE CURRENT POLICY IS DUMB. Here’s a questions – how is that not social engineering?
Thank you for the reply, Steve. Even though you characterize the equality of tax rates as a disincentive, the math is still in favor of saving. Yes, the compounded return from the savings/trading/capital gains is at the same rate, but if the return on the savings of $1 is $3, then the saver/investor is better off by 3X regardless – if the initial rate is low enough, (such as my hypothetical 17%) then why wouldn’t the incentive to save still exist? (Net worth of $2.49 vs $0.83) – assuming that the net return after taxes is still greater than the inflation-adjusted net present worth of the initial $1 (less taxes.) It’s hard for me to believe that your 3 Oreos in question 4, if taxed at the 17% rate, would be less valuable to Jill than the single pie, taxed at the 17% rate. So long as it is either-or (and not both the tax on pie and cookies also.) In Question 5, so Long as Jill is not being taxed on the same $2 of income as the initial earnings, then the future earnings can be taxed on an equal percentage basis without bias.
If their is a differential tax rate, then there is an inherent bias towards more efficient earnings, rather than absolute value of earnings (e.g. a day’s work vs a stock market gain in 1 day without work). I’m not sure why that is social engineering? Sure, we want everyone to find efficient ways to earn, but why differentiate in the public policy between one form of income vs another? If, from a public policy perspective, income is income, and there are no marginally increasing tax rates, then everyone is incented to earn as much as possible.
“Then we observe that it’s easy to disguise capital gains as interest or dividends, so the next thing you know, we’re also taxing interest and dividends at that rate. I don’t entirely like where this is going!”
“SmallBusinessGuy: If by a single tax rate on all income, you are including income from interest and dividends, then you are still distorting the incentive to save”
The sky would indeed be falling if interest was taxed as income… except that it already is.
http://www.irs.gov/taxtopics/tc403.html
“Most interest that you either receive or is credited to your account and that can be withdrawn without penalty, is taxable income. Examples of taxable interest are interest on bank accounts, money market accounts, certificates of deposit, and deposited insurance dividends”
So, here’s a question: why are cap gains taxed less than interest?
Got around to reading your supporting info. Seems like your other arguments can be summed up as “double taxation hurts savers”. However, you don’t take into account that saving is earning. It is a capital gains tax, not a capital stay-the-same tax. If, first, you make some money, and then, second, the money makes money for you, it is not oh-so-awfully-incentive-distortingly-evil to tax your two incomes twice.
“Can you please write down the optimization problem which has as its solution “tax the doctor’s income and the homeowner’s capital gain at the same rate”?
This is both naive and patronizing. All important questions is not decided by equations. Professor Uwe Reinhardt has made some important points which are persuasive. If you have any answer , you should be able to spell it out in simple equations. If you cannot, you simply don’t have any real argument. It’s all fluff!
Mike H:
Is it valid to use “tax on interest and dividends distorts the incentive to save” as an argument against tax on interest and dividends, but NOT mention that tax on wages distorts the incentive to work?
Yes, this is essentially the content of the Chamley-Judd Theorem. A tax on interest/dividends distorts the incentive to work much as a tax on wages does, and then imposes a second distortion on top of that.
When I first learned of this theorem, I was sure it had to be wrong because it seemed inconsistent with what I thought were obvious symmetries. It took me a long time to digest these ideas, but I get it now.
@iceman: “is there a good *principled* argument in favor of double taxation?”
Which double taxation are we talking about here? The corporate-then-dividends sequence, or the income-then-savings sequence?
Does the proof of the Chamley-Judd theorem assume that everyone has the same preference for savings over immediate consumption?
I have a counterexample to Chamley-Judd on my blog : I assume people have different preferences, and find that the optimal tax rates for investment income and earned income can vary wildly all over the place, even to the point of having subsidies for one or the other.
You’ll find the link in the Pingback, or here : http://2ndthotz.dr-mikes-maths.com/2012/03/steve-landsburg-asks-for-an-example/
Nowhere at the link you provided in your post did I see anything specific to capital gains. What I read was an analysis of the taxation of capital income, with no distinction being made among interest, dividends, and anticipated capital gains. Moreover, in the case of the overlapping-generations model, I found that the case for zero taxation of capital income is does not generally hold. Also, I only saw analyses that assume time-separable preferences with an unvarying labor-leisure tradeoff, so I’m not totally convinced of the complete generality of any of the results. After all, if people tend to concentrate their leisure toward the latter periods of life, their preferences don’t appear to be time-invariant.
I have also not seen you draw any distinction between capital gains that are forecastable (such as those that arise from retained earnings) and those that are the result of unanticipated relative price changes. It seems to me that efficiency dictates that the latter be taxed quite heavily, provided that inflation-indexed gains and losses are treated symmetrically.
Mike H; My best recollection is that Chamley’s original paper assumes homogeneous preferences. But there is a vast followup literature in which this assumption is dropped and the conclusion survives.
I’ll look at your blog later today.
@joena lopez:
Maybe you shouldn’t be quite so convinced by anyone’s argument if it isn’t quantitative: good arguments are usually better when made quantitative and more explicit. This is especially the case with arguments for policies with quantifiable effects. These CAN be quantified. This is what, in technical jargon addressed to a technician, SL asked UR for. If there is a legitimate economic case to be made it should be possible to make it explicit enough for an equation.
Ken: You said “Ability to pay is definitively not what Rheinhardt meant;”
I don’t know what he meant, I only know what he said. I took “horizontal equity states that taxpayers with similar ability to pay should contribute the same amount.” at face value, and re-phrased it as “ability to pay” is the determining factor. It seems consistent to me.
Q1. Exactly. If we were to apply such a principle, then motivation would have no influence on judgement.
Q2a. I am not sure if the ability to pay is counted over more than one year, but if it is done annually, then presumably both should pay the same under this principle.
Q2b. You pay what taxes you are able. What has happened in the past does not affect it. When you earn you pay, when you don’t earn, you don’t pay.
Q2c. It is moot because there is no need for the Govt to keep house prices fixed.
Q3. The price of margarine has no bearing because the principle does not mention margarine. If we had a principle that said “everyone must pay taxes according to the price of margarine”, then it would be a factor. If we have a principle that says “everyone should pay the same if they have the same ability to pay”, then of course the doctors pay is a factor.
Q4. I got this one wrong. Jill has no more ability to pay than Jack, 3 Oreos being equal in value to 1 pie. Therefore Jill should pay no extra tax.
Q5. Jill will not pay any extra tax until the future dollars are realised. This is the same as having put a pie in the freezer. When she takes it out of the freezer, she will have made no gain, and so presumably will have nothing to pay. If when she takes it out of the freezer it has turned into two pies, then she will have to pay something extra. So if the three future dollars are worth more than the 1 current dollar, there will be something to pay.
One could argue that she has more ability to pay at the moment she takes it out of the freezer. But she had equally less ability to pay when she put it in.
I am not claiming that it is a good principle, just that it appears to be the principle that was used. It does answer some of Steve’s questions, although it raises some others.
@Harold
The ability to place a pie in a freezer for later consumption means that she has a greater ability to pay today. Her pie should be taxed at a greater rate because she is choosing to not consume it all today.
Capital gains = “ability to freeze for future consumption”
Changed the loaded term ‘unearned income’ for the loaded term ‘income earned by saving’ and see how the arguments sound.
I do not know how we decide on ability to pay. I assume it is based on income only, from any source. The choice to save a greater or lesser proportion is not a measure of ability to pay, just a preference to pay. To keep things fair, if Jill puts a pie in the freezer and takes it out later, she should only pay once. This could either be when she earned the pie, in which case we need to index link the savings. Or it could be when she defrosts it, in which case we need to off-set money earned and saved.
Why create different incentive structures for different kinds of earnings? What is the rationale behind not taxing the extra $2 of Jill’s earnings but taxing the first $2? So long as the same $2 is not being taxed twice (as in the case of 3 oreos for 1 pie, assuming that they are of equal value and not that 1 pie = 1 oreo). This notion of “unearned” income versus “earned” income is what creates disparity. people should find the most efficient way to obtain income. In some cases, it requires labor. In other cases, intellectual effort. Sometimes, perhaps, luck. But the most effective for a society is productive labor (as opposed to inheritance, for example.) So, why force a disincentive to work? Tax everything equally – all income should be treated just as “income.” Earned, unearned, won in gambling, whatever.
@Henri Hein: either or both please.
@iceman: Oh please, not the perennial “Ginger or Mary-Ann” thing!
Harold,
Q2b You pay what taxes you are able. What has happened in the past does not affect it. When you earn you pay, when you don’t earn, you don’t pay.
This conflicts with your “ability to pay” principle. Above (in Q2a) you say the 70 year old and the 30 year old should pay the same because they each have equal ability. Now you’re saying the 30 year old should pay more because he earned more (the ability of the 70 year old to pay came from savings, not earnings).
Additionally, if “ability to pay” is the only criteria, how is selling a house for $50K more than you bought it any different from simply having a house worth more than $50K more than you bought it. In other words, if you want to compel those who sold their house for more than they bought it, why can’t you compel a homeowner to sell or take a line of equity and pay taxes on the difference between what the house was bought for and the current market value?
When she takes it out of the freezer, she will have made no gain, and so presumably will have nothing to pay.
This statement violates your “ability to pay” criteria. If Jill could put her pie in the freezer, but Jack, because he was hungry, ate his immediately, then it is obvious that Jill as a greater “ability to pay” taxes. Since she froze her pie, she clearly didn’t need it at all and was simply saving it for a later day for whatever purpose.
One could argue that she has more ability to pay at the moment she takes it out of the freezer.
Actually, it’s from the moment she put it in the freezer.
I am not claiming that it is a good principle
Fair enough.
The reason to tax capital gains and ordinary income at the same rate is that taxpayers choose which to earn, and by favoring one over the other, you encourage a certain type of investment/activity over another. As an investor, I can invest in investments that create capital gains (non dividend paying stocks), or those that produce ordinary income (bonds). Why favor one over the other?
@Michael Moran – I believe Steve would say that the relevant distinction is between earned income (money you earn by working) and unearned income (money you earn by investing your money). In that sense, your bonds, your dividend-paying stocks, and your dividend-free stocks are all giving you unearned income, which he says should be taxed at 0% – unless you happen to work for the company you own shares in, so that the capital gain or dividend is purely the result of your hard work, then it’s earned income after all.
His argument is sound, except for two missing pieces :
* the question of empirical evidence – does taxing capital gains actually, in the real world, have the disincentive effect he points out, and
* the fact that the proposal is impossible to implement in practice, since it’s too easy to lie about the source of your income or other things.
So, Steve, based on your comments on my blog (thanks for visiting) is it correct, if you want to maximise utility, to say :
* Homogeneous preferences => homogeneous tax, with no tax on unearned income
* Inhomogeneous preferences & inhomogeneous taxes => no tax on unearned income
* Inhomogeneous preferences but homogeneous taxes => Chamley-Judd breaks down
Michael Moran,
This is why capital gains should be treated differently from income.
@iceman: Alright, I will take a stab at making an argument, though I cannot promise it will be a good one.
First, let me get corporate taxes out of the way: I cannot think of any argument supporting those that are not political.
In terms of taxing income and savings, it actually struck me as a bit backwards to ask for a principled argument in favor. I understand the main argument against taxing interest to be that it discourages savings. I agree with that argument, but it sounds utilitarian.
My preferred principle is that when Alice and Bob each makes a buck, those two dollars should be taxed the same, regardless how they made it.
One problem I have with the double-taxation argument when it comes to investment income is that it assumes wages is the source of all investments. I don’t have data for the economy as a whole, but anecdotally, only a small portion of investments come from wages. By far the biggest portion come from equity.
@Henri : “but anecdotally, only a small portion of investments come from wages. By far the biggest portion come from equity”
but where does this equity come from? Doesn’t a big chunk of it, ultimately, come from wages, either the wages of the equity’s owner or of their forefathers?
@Henri, does your preferred principle also cause you to conclude that, in response to question 4, Jill should be charged an additional tax on the Oreos she receives in exchange for her pie?
@Ken B: Let’s not overthink things. It’s always been Ginger hands down.
@ Mike H: I mean wages in the technical sense, that is, money paid in compensation for labor or services. I think most equity is built outside of wages. Most entrepeneurs own their equity through sweat, not wages, and most wealthy people are entrepeneurs.
@Jon Shea: I did not understand question 4. I don’t think bartering should be taxed at all. The IRS seems to agree: they only crack down on the most systematic instances of bartering.
@Henry
I understand your point. If you define wages like that, then the idea that capital gains should not be taxed breaks down. However, economically speaking, there’s not much difference between
* me working for a year in my own company, and (thanks to my effort) getting a $60k dividend,
* me working for a year in someone else’s company and (thanks to my effort) getting paid $60k in wages.
Either way, I work for a year and get $60k. Good economic policy would try to treat these situations identically. Tax law, of course, doesn’t.
@Henri Hein: By “principled” I meant an argument based on anything other than the practical problem of income-shifting (which is not about double taxation but avoiding full single taxation). Efficiency would be one principle (‘discourages saving’ does sound like ‘social engineering’, but ‘remains neutral between current and deferred consumption’ not so much). Or something that says why taxing income more than ‘once at its source’ (e.g. ‘two bites at the apple’) is more right or fair. However, your statement “regardless how they made it” seems to duck those issues.
Interesting point about sweat equity; I guess I’d consider that a capitalization of the future (hoped-for) higher ‘wages’ which would be taxed as/when they materialize. Disguising that return as a dividend would get us back to income-shifting. And if someone else purchases the business in the meantime and pays a premium for that equity, I presume SL would say they will discount the purchase price for the taxes on the future wages so the seller has already paid.
On corporate taxes, I’m honestly just looking for any valid economic argument whatsoever.
In case anyone’s still listening, the report at http://www.ifs.org.uk/events/625 has an interesting proposal : tax investment income in full – however, offer a full tax deduction for invested capital.
Because capital income is fully taxed, there’s no incentive to disguise labor income as investment income. However, because invested capital is tax deductible, there’s no distortion with respect to present vs future consumption.
I guess there’s still the problem that people will disguise consumption expenditure as investment expenditure.
Regarding Question 2, I think that’s a deliberate misinterpretation of what he says. The point isn’t that the capital gains tax rate is unfair, but that the difference in tax rates for human capital formation (the surgeon’s pay) and capital gains are different. Consequently it’s hard to take anything else you say seriously since you seem to be devoted to attacking paper tigers.
I have never understood the rationalle for treating inflation adjusted capital gains avereaged over the holding period any different from any other income. Mankiw’s examples (all were real one year gains) seem to support this equivalence.
for question 3
Given that the capital gain is basically a return on an investment ( if we exclude the offsetting factors that part of the nominal return is just inflation and the owner derived a non-monetary non-taxable benefit/income stream from the use of the property) and the doctor’s income is primarily a return on investment in his education there is a rough economic equivalence between the capital gain and the doctor’s income and they can reasonably be taxed at the same rate. That could probably be formulated as an optimization problem fairly easily but I will leave that as an exercise for the reader.
Two weeks and still no response from Professor Uwe? I’m disappointed. Someone call him, pretend to be a reporter, and ask him these questions.