Pandering to the public’s ignorance
Tax theory is full of cognitive illusions. Things are never as they seem. Here are some examples:
1. In the long run it doesn’t matter how the payroll tax is split between workers and companies. Right now the split is 50-50 in the US, but this has no bearing on the actual incidence of the tax. If the tax were set up where 100% was paid by workers, it would not “favor” business in the long run (although it would in the short run, due to sticky wages.)
2. AFAIK, most countries levy VAT on imports, but not on exports. US politicians and pundits will sometimes demagogue this issue, claiming that this tax regime “favors” exports. It doesn’t. As far as I know all public finance experts agree on this point. It’s trade neutral. (In all of these cases there are obviously possible real world complications with second order effects—my point is the demagogues prey on the public’s tendency to use a common sense approach to taxes, which is almost invariably wrong.)
3. The VAT applies to consumer goods but not capital goods. Again, this is to avoid double-taxing the part of consumer goods that represents the cost of capital. In no sense does this mean the VAT system “favors” capital goods over consumer goods.
4. A flat 10% VAT and a flat 10% wage tax are essentially identical in the long run (ignoring enforcement problems.) At first glance the payroll tax seems to tax “labor” while the VAT seems to tax “consumption.” But that’s an illusion. They are both wage taxes and they are both consumption taxes. They are essentially identical.
Here’s what annoyed me so much about the Solow quotation in the previous post. Yes, there are a number of articles that provide some (weak) justification for a tax on capital income. These involve everything from risk/insurance, to enforcement/evasion problems. So I’m not claiming there are no arguments at all for some sort of capital income tax. But the baseline you start from is a zero tax, not a capital income tax equal to the wage tax. When Solow talked about our tax system “favoring” capital income he was clearly pandering to the public’s ignorance of tax theory. If Warren Buffett says he pays a lower tax rate than his secretary, that’s going to look unfair to almost everyone, even to GOP voters.
I think I have a better way of explaining all this, which might convince some commenters who were confused by the previous post. Many people are aware of the 401k pension plan in America. You contribute money you have earned, and do not pay tax on that money until it is withdrawn at retirement. If you allowed unlimited 401k contributions, and then taxed the amount only when it was withdrawn for consumption purposes (by your or your heirs) then you would have essentially converted the income tax into a pure consumption tax. AFAIK, this claim is completely uncontroversial among tax experts.
Now let’s think about a world of unlimited 401ks, with no taxes on capital income, just a 50% wage tax. You earn $100,000, and save $20,000 in the current year. That means you pay $40,000 in tax today, and consume $40,000 today. The other $20,000 goes into a 401k, where it grows to $60,000 over the next two decades. Then you take the money out and spend it. How much tax do you pay at that point? The answer is $30,000, as the wage tax also applies to funds removed from the 401k. You can think of that tax as having two parts:
$10,000 of the tax is a deferred tax on the $20,000 that you failed to pay tax on when you put the money into the 401k.
$20,000 is the tax paid on the capital gain of $40,000 that you earned on your invested money.
Thus to the average person it looks like the 401k account is not “tax free” but rather “tax deferred.” A politician or pundit would have a difficult time demagoguing that issue. Suppose they claimed that 401k holders were paying zero taxes on their investment earnings. Outraged 401k holders would say; “Wait a minute, I am paying the full income tax on all my investment gains when I take the money out and spend it. And even the wage tax I initially avoided is only deferred, I still must pay the full wage tax in the future.”
Under this 401k approach it looks like both wage earners and investors are paying the exact same 50% tax rate. No one in their right mind could claim this tax system “favored” capital income over wage income.
And yet our current tax system taxes capital income at a higher rate (relative to wage income) than the system just described.
You’d get the identical result if you applied the 50% wage tax to the full $100,000 (i.e. $50,000 in tax), if you assume they spent $40,000 on consumption today, and then invested the other $10,000 for 20 years, consuming $30,000 at that time, with no capital income tax. But the 401k approach looks much “fairer,” as it “seems” to tax capital income at 50%.
Of course Solow is a brilliant economist who certainly understands basic tax theory. Whenever I see smart liberal economists argue that our tax system favors capital income I immediately suspect them of pandering to the ignorance of the public. Tax theory is counterintuitive. A neutral tax system will often appear to favor capital, and European exporters, and businesses, and also appear to punish workers and consumers and domestic firms competing with European exporters. But these are cognitive illusions, and it’s about time we stopped pandering to these illusions.
If people want to make sophisticated arguments for special capital income taxes where evasion is rife (say hedge funds or entrepreneurs) then do so. But don’t use that as an excuse to double tax the money saved by ordinary working Americans. At a minimum they should remove any contribution limits on 401ks. Recoup the revenue with higher payroll taxes on upper income Americans.
And don’t liberals claim that working people can’t afford to save very much? Then my proposal shouldn’t cost much revenue.
PS. Some commenters wrongly think all this is some weird idea I concocted. Not so. I’ve seen Yglesias and DeLong discuss progressive consumption taxes. They certainly understand the points I’m making, even if they may not agree with all my policy preferences.
Prove me wrong and there’s a Nobel Prize waiting for you in Stockholm.
Tags:
24. April 2014 at 17:37
VAT on oil imports would be trade neutral? In normal ranges? But what about a heavy VAT on oil imports?
Or, are you arguing big picture, the dollar would nudge up in value cancelling other imports and exports? Or are you adding capital flows to the picture?
Are VATs different from tariffs? A 80 percent tariff on imported oil would not cut oil imports?
As my Uncle Jerry always said, “Even if that is true, I still don’t believe it.”
24. April 2014 at 17:50
Progressive Consumption taxes + GI/CYB FTW!
24. April 2014 at 17:58
Scott, the 401k approach might seem fairer to some, but the objection I anticipate is that people will say, “So a high-income earner who can afford to save will get the benefit of compounding returns on untaxed [or pre-tax, if you prefer] income. This will just widen the gap between rich and poor!” Illogical I know, but the problem is that the majority of voters are not thrifty; they live from paycheck to paycheck and as a result simply believe that if you can afford to save, you must have more than you need and thus should pay more tax.
24. April 2014 at 18:27
Ben, VATs are very different fro mtariffs, as they also apply to domestic goods.
Rajat, You said Aussies think:
“if you can afford to save, you must have more than you need and thus should pay more tax.”
Of course this is true, but it’s also true that millions of Australians who “can afford to save” do not in fact save. The test of whether you can “afford to save” is not whether you are saving, but rather your wage income.
24. April 2014 at 19:02
Yes, horizontal equity gets confused with vertical equity.
I think what happens is that many people – particularly the spoiled Gen-Ys (!) – judge the thrifty by the saving standards of the spendthrift. So if you have accumulated a lot of capital, people assume that you must have earned enough to not only spend as much as they did, but save as well. Hence, you should be taxed more.
24. April 2014 at 20:52
Whenever I see people claim that the Fed should prevent “spending” from falling, on the basis of “sticky wages” I immediately suspect them of pandering to the public’s ignorance. Allegedly they are too stupid to set optimal wage rates and so they need to be forced away from a free market in money so that their ignorant price setting can find enough demand.
No socialist in money can unhypocritically accuse others of pandering to the public’s ignorance. For if the public truly understood our monetary system, they would seek to abolish it. Indeed, the whole state is predicated on pandering to the public’s ignorance.
24. April 2014 at 20:59
Sumner doesn’t even understand the time value of money or the tax incidence, yet he is teaching college economics.
Such a disgrace.
Then again, he is a right winger. Facts, mathematics, and economics don’t stop him from spewing his worldview.
24. April 2014 at 22:02
So, lots of stories mixed here. Deferred taxation schemes (I don’t know 401K rules, I’m working from traditional IRA) have the well-known property of being time neutral so long as ones tax rate is constant. But neutral in what sense? The tax on consuming now is equal to the same tax compounded forward at whatever your investments earn, at the time of a future withdrawal. This is the form of neutrality that you describe.
But note that we politically acknowledge that this is a very generous tax regime towards capital: we explicitly and strictly (and quite correctly) ration participation in these retirement programs! Holding tax rates constant, a deferred taxation regime (traditional IRA) is equivalent to a pay-now, pay-no-cap taxes later regime (Roth IRA). This latter is the regime that you desire, so it’s understandable that you like to analogize it to the “tax-deferred-but-still-highly-taxed” sounding regime.
The issue is whether earnings that are derived from saving or deployment of capital should be taxed. To point out that there exists a regime that sounds like it taxes capital income but doesn’t may be confusingly persuasive to some people, but it is simply a distraction from that question. IRA-like schemes, traditional or Roth, don’t tax capital income. (unless there is an adverse change in tax rates… which direction of change is “adverse” depends on which kind of IRA.)
You still haven’t addressed the question, just hidden your policy preference in sympathetic and confusing stories.
There is of course a common answer. That is the Chamley-Judd result, which suggests a optimal capital tax rate of zero. But that’s a stylized result of a deeply unrealistic model. To call for that to be the baseline is simply to try to privilege ones own views. You are welcome to argue that the Chamley-Judd result is sufficiently realistic and relevant that it should form our null hypothesis, but to do that persuasively, you actually have to look out into the world and not into elegant identities.
Out in the world, it is to say the very least contestable that Chamley-Judd reasoning holds. Low capital taxes as we conventionally define them (which includes, of course, low taxation of lending to fund others’ mere consumption) do not seem reliably to lead to increases in real wages proportionate to aggregate growth and marginal productivity as that model would predict. Certainly the US experience post 1980 seems not so consistent with CJ expectations. Of course, one historical path is an uncontrolled experiment and not strong evidence any which way. But it is stronger evidence than simply holding the truth of an elegant model as dogma, or creating very particular scenarios that cast a predetermined view in the most sympathetic possible light.
Solow, by the way, has been perfectly consistent about all this. He has long criticized the Ramsey model underlying the Chamley-Judd result. [ http://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.22.1.243 ] He is not pandering or being disingenuous. What you, rather conveniently, take to be a settled result in public finance, Solow knows and understands perfectly well. He just doesn’t agree that the result is settled or even approximately correct. Neither do I.
25. April 2014 at 00:32
“The tax on consuming now is equal to the same tax compounded forward at whatever your investments earn, at the time of a future withdrawal. This is the form of neutrality that you describe.
But note that we politically acknowledge that this is a very generous tax regime towards capital: we explicitly and strictly (and quite correctly) ration participation in these retirement programs! Holding tax rates constant, a deferred taxation regime (traditional IRA) is equivalent to a pay-now, pay-no-cap taxes later regime (Roth IRA). This latter is the regime that you desire, so it’s understandable that you like to analogize it to the “tax-deferred-but-still-highly-taxed” sounding regime.”
This is not always true because of inflation and corporate income tax on equity savings, but let’s eliminate those factors.
If we do so, it is certainly true, but it strikes me that it catches you in a fundamental contradiction. The fact that if tax rates are constant, income deferred through a traditional IRA, or taxed up front through a Roth IRA or taxed immediately on wages that are consumed is, from a consumption standpoint, taxed equally and neutrally clearly demonstrates that this is not so in the existing regime in which wages and “investment income” are *both* taxed. If you concede that in the IRA regimes “neutrality” exists, how can such a regime “favor capital”? And, can such “neutrality” also hold in a regime that lacks those characteristics?
Or, are these inconsistencies simply explained by a “political acknowledgement” rather than an “economic acknowledgement”?
25. April 2014 at 01:16
Vivian.
There are no inconsistencies. Any taxation of any sort of income implies an opportunity cost of future earnings that would have appeared had the income been neither taxed nor consumed. It is silly to then call taxation of wage income taxation of capital income.
If the wage income is taxed and consumed immediately, there has been no capital income to tax. If Roth-IRA-style, wage taxes are deducted immediately and the remainder invested, then capital earnings accrue and are never taxed. If traditional-IRA-style, wage taxes are not deducted immediately but are taken upon withdrawal, then your capital earnings are still never taxed. Properly accounted, your investment fund has a second claimant, a fisc which agrees to ride along as an equity partner rather than withdraw its claim immediately. You keep the full capital income on your share of the partnership, the fisc enjoys the capital income on its share.
These are very simple, very consistent arrangements. They all involve wage taxation with no capital taxation. The fact that one of the three variations on identical economic stories is a bit harder to explain or keep the accounts straight might be useful as a source of tendentious confusion, but does not represent any sort of inconsistency. In all three cases, wage income is taxed once, capital income not at all.
25. April 2014 at 01:42
Prof. Sumner,
I got somewhat stuck by the annoyed and almost angry undertone of your recent postings on Piketty’s book (but maybe this is only in the ears of a non-native speaker). Either you do not mention the name at all (“a recent book review by Robert Solow”) or you refer to the “pied piper from France”. But putting these matters of etiquette aside, I am missing an aspect of capital taxation that is central to the book and the other writings of Piketty on this topic: the question of inheritance taxation. In his 2013 Econometrica article (with E. Saez) it is shown, e.g., that (based on utilitarian or at least welfarist reasoning) there exists a strong case for an inheritance tax if people differ along two dimension (labor productivity and bequests). In the case where there is heterogeneity only along one dimension (say labor productivity or the rate of time preference as used in your examples) the usual zero-capital-taxation result by Chamley, Judd and others is retained. In any case, I think that the appropriateness and “fairness” of capital taxation looks different whether it is only considered as a life-cycle issue or whether one also takes the intergenerational perspective into account.
25. April 2014 at 03:40
Steve: I always thought it was much simpler than any particular growth model:
People who prefer to consume bananas shouldn’t pay more tax than people who prefer to consume apples, even if bananas are cheaper than apples.
Similarly, people who prefer to consume apples in the future shouldn’t pay more tax than people who prefer to consume apples in the present, even if future apples are cheaper than present apples. (Positive real interest rates mean that future apples are cheaper than present apples.)
(Which is just another way of stating Scott’s point.)
What am I getting wrong?
(The tricky bit is: people who prefer to consume more apples shouldn’t pay more tax than people who prefer to consume more leisure, so ideally we should tax wage per hour rather than wage times hours worked, so we stop subsidising leisure. But that has big practical problems, because we don’t observe the wage rate that people *could* earn, plus it gets us into thorny problems about how much of hours worked is “voluntary”.)
25. April 2014 at 05:35
Steve, You are confusing two very different issues. One is whether a zero tax on capital income is desirable. I certainly agree that that is debatable. The other is whether people who think that capital is taxed at a lower rate means that the tax “favors” capital. Here’s an analogy: What would you think of someone who claims that a tax of two cents per watermelon and one cent per blueberry means the tax system favors blueberries? That would obviously be absurd. Comparing capital income and wage income as if they are the same thing is the same sort of fallacy
I notice that you did not address the gist of my post. Those who claim VATs favor capital over consumption are making the same error as Solow. Do you want to claim VATs favor capital over consumption? BTW, As far as I know all the VAT countries do this, including left wing countries which would have no reason to “favor” capital. Does Solow think VATs should also apply to capital? I can’t imagine why. He’d be one of the few economists in the world who believes that. And yet by excluding capital goods a VAT becomes equivalent to a wage tax. Would Solow claim that VATs favor exports? After all, they are excluded from the VAT. If not, why not? And why wouldn’t that argument apply to capital?
Yes, politicians who enact 401k laws say they are “favoring” savings and investment. Why should we be surprised in a world where most people (including even most economists) do not understand the logic of tax theory. These are same politicians who split the social security tax 50-50 in order to be “fair”. Surely you are not arguing that split reflects the actual incidence of the tax? So I think it’s fair to say we can discount appeals to how a tax break is sold to the public by non-economists. In contrast, tax experts often will argue that 401ks are good precisely because they are neutral between consumption and saving. They do not favor either activity. I’ll take the views of tax experts over politicians.
I’m pretty sure Solow knows that the public doesn’t understand that 401ks are equivalent to a pure payroll tax. He panders to the public’s ignorance with his “favoring” capital comment. Even if the optimal tax rate on capital is not zero, it’s extremely unlikely that the optimal tax rate on capital would be as high as on labor income. So Solow’s claim isn’t even justifiable if you switch the argument from the narrow question of whether capital is favored relative to a baseline assumption of neutrality between consumption and saving, to the completely different argument of optimality under different real world assumptions.
Your claim that IRA-type schemes don’t tax capital income is rather meaningless to a tax theorist. As I showed, IRA-type schemes such as a traditional IRA or a 401k look like they tax capital income at the exact same rate as wage income. Now it’s true that the system is equivalent to a pure wage tax (allowing a few minor adjustments in the law, like no requirement to withdraw money at age 70), but that just shows how misleading appearances can be. Your problem is that you started by claiming I was torturing the english language by claiming that a regime that taxes only wages does not favor capital, and that one that taxes capital as well as labor income actually taxes capital income at a higher rate. Then I showed you an exactly equivalent system (401k) that looks to the average person like it does tax capital income at exactly the same rate as it taxes labor income. That proves that appearances can be deceiving, and that the public’s understanding of this issue is almost nil. And this is true even if EVERYTHING ELSE I’VE SAID ON THIS ISSUE IS FALSE. (Sorry for the caps, I don’t know how to do italics in comment section.)
Whether a consumption (or payroll) tax regime actually taxes capital is a philosophical (or perhaps language) question that is beyond the scope of public finance theory. All economists can do is talk about tax regimes that are neutral between current and future consumption. Indeed pretty much all of economic theory (and much of finance as well) is about the impact of various changes on the relative consumption of different goods, at different points in time.
Over at the other post someone provided a very simple and elegant way of making my point:
“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.”
Is the interest taxed once or twice? That’s a philosophical question above my pay grade. But the absurdity of claiming that a payroll tax “favors” capital is clearly exposed, regardless of the words you wish to attach to the process.
(BTW, option B is essentially a mini-401k.)
25. April 2014 at 05:35
It hurts me at a personal level to see these kinds of fights over taxation. Steve Waldman, in particular I am talking to you. All the taxation in the world cannot make up for what is presently a completely missing marketplace in knowledge use for lower to middle income levels. People need to be able to create this marketplace for themselves, so that further taxation will not simply eat us alive.
25. April 2014 at 06:10
Mark, I plan to read the book after the semester is over. Because I haven’t read the book, I have no comment on it at this point. I didn’t mention Piketty’s name in the Solow post precisely for that reason. I didn’t want people to say “how dare your criticize Piketty when you haven’t even read the book.” I wanted the focus of my post to be on Solow’s comment, not Piketty.
If the pied pier comment was rude I apologize to Piketty. Perhaps “Siren song” is closer to what I had in mind. I was actually trying to criticize his followers, not him. Everyone agrees the book is skillfully done and on a certain level seems very persuasive. (I don’t mean everyone is persuaded, just that Piketty is skilled at making his points.) While I have not read the book, I have read the reviews. And I find many of the reviews to be far too willing to accept dubious claims, such that r > g has important implications for future trends in inequality. That’s what the comment referred to. However even where I’ve disagreed with the reviews, they’ve also been skillfully done. Reviews from people like Cowen, DeLong, Solow, etc, will help me when I get around to reading the book
I’ve read enough reviews, and seen enough interviews, to be fairly certain I will strongly disagree with much of the book. But I won’t actually comment on the book until I’ve read it.
Regarding your other points, I’ve always acknowledged that there are models that can be constructed where you move away from the baseline assumption of no taxation of capital. But the other side often forgets that you start with that baseline model, not with the presumption that “neutrality” means wages interest, dividends and cap gains are all taxed at the same rate. Solow basically started at that false presumption, which is how he ended up with the claim that our tax regime favors capital.
For instance, it’s very possible that the optimal tax on bequests is neither 0% nor the current 45%, but something like 7% or 27%. We simply don’t know.
Nick, Yes, that’s exactly my view as well. Once you start thinking about public finance in terms of present and future consumption, most of the discussion of taxes in the blogosphere seems to miss the point.
25. April 2014 at 06:22
This taxation discussion clearly has a focus on first order taxation effect; that is, the effect on decision makers when they consider the tax effect on their immediate personal decision. There are second order effects that can be very important such as the shift (under the 401K system) of near-term governmental financial needs to remaining non-401K participants.
Second and even third or fourth order effects can be of long term significance to an economy. I would go so far as to suggest that the U.S.A. persistent drift to greater wealth inequality is substantially due to these lower order taxation effects.
25. April 2014 at 06:37
New blog post:
On Piketty. Who cares about rich kids? Entrepreneurs are supreme.
https://medium.com/p/ff7e5e1c4f7
25. April 2014 at 08:11
Suppose we have a subsistence economy of 100 agricultural workers with no tools, each earning $10 per year and paying a flat 10% income tax. That’s total income of $1000 producing $100 of tax revenue. After tax consumption is $900.
Now, one day, one of those workers has an idea (he calls it a ‘shovel’). He decides to ‘exploit’ his fellow man, and the taxman (who don’t know anything about taxing capital which doesn’t exist). He drops out of the workforce to build shovels and rent them to the workers for $1 per year.
It would be trivially easy to come up with numbers for the above showing a big win-win-win. Everyone would be richer; the workers with their increased production, the govt. with higher revenue, and the ‘capitalist’ with his untaxed income.
But is capital ‘really’ untaxed?
25. April 2014 at 09:14
@ Benjamin Cole:
The point about the neutrality of VAT between imports and exports relates to a uniform VAT. It really is just the same thing as the equivalence of a uniform tariff and a uniform export tax.
25. April 2014 at 09:30
SCOTT —
Do you think the Corp income tax should be abolished? Is that “capital income”? What about the income tax on business activities of LLCs, which is passed through to the owners personal income tax and taxed like wage income. Is that capital income or labor income?
Jeremy
25. April 2014 at 10:22
Scott, O/T: Just how powerful can expectations be? Suppose there was a nation composed of 100% devout Neo-Fisherites:
http://noahpinionblog.blogspot.com/2014/04/the-neo-fisherite-rebellion.html
whom all expected that raising the overnight interest rate would raise the inflation rate. Is it possible that raising interest rates would then actually raise inflation? In other words, is it *theoretically* possible that a steering wheel, like Nick Rowe discusses here:
http://noahpinionblog.blogspot.com/2014/04/the-neo-fisherite-rebellion.html?showComment=1398289160702#c3401104455099959999
might actually be connected to turn the car left when you turn the wheel clockwise?
25. April 2014 at 10:57
I have one minor question
what is the ideal tax regime, that doesnt punish consumption, income, or capital?
and how do we make it so that its ‘revenue’ neutral as many seem to want it to be.
and why would raising interest rates raise inflation, since it would seem that almost all would tend to not buy as much (since a huge portion of consumption, aka buyers, borrow to purchase?)
25. April 2014 at 11:44
The mistake about SS taxes is not confined to liberals. It
shows up persistently with “conservatives” I’ve sparred with over Hobby Lobby. I claim that health insurance premiums, like the “employer’s share” of SS, is just part of the employee’s wage and as such the employer has no legitimate “religious” (or philosophical/principle) objection to what is covered by the insurance.
25. April 2014 at 11:47
Do you think the Corp income tax should be abolished? Is that “capital income”?
Speaking as lawyer with some graduate economics training and decades of experience working in the tax field I can say that the corporate income tax fails in spades the key tests of a “good” tax:
[] Equity: ‘treating like parties in like matter’. It lands with greatly *different* weight on parties in very similar circumstances. Not only do corporations in like situations pay *very* different real tax burdens under the same nominal tax rate, but the entire tax is passed on to individuals of course — inanimate objects and legal fictions do not pay taxes — and on to different people in different proportions via the relative elasticities of the parties. In this case it lands on workers, in that on shareholders, in the other on consumers, etc., in all different ratios. Unknown to all absent a complex case study.
[] Transparency. ‘How the tax works is clear to all’. The corporate tax is totally opaque. Jeebus, don’t even look at the Code, look at the thousands of pages of regulations for corporate taxes. If the tax was transparent people like me would be out of business and doing something productive with our lives. For starters, regarding the most elemental basics, how many voters realize what’s above in the “equity” paragraph? E.g., how many realize the incidence of the tax is likely falling on *them* instead of rich corporate fat cats?
[] Simplicity. ‘Ease and clarity of administration’. I won’t relate any of the running multi-year horror stories of administration and litigation. Just note the massive incentives and rewards that exist for gaming the system that exist for both sides, the taxpayers/lobbyists/lawyers and the politicians/regulators/bureaucrats. There are huge amounts to be made buying/selling tax expenditures and various other tax breaks — see those thousands of pages of Regulations. And all go up with each point of the tax rate. Remember that the deadweight cost of a tax rises not with the increase of the tax rate but by the *square* of the increase in the rate. (All of this occurring behind the opacity of the tax, and compounding the inequitable effects.)
This is all apart from the double taxation issue, and I’m not going to go at all into the economic inefficiency/distortion issues.
Notwithstanding the above, the corporate tax exists and will continue to do so for two reasons:
(1) Corporate treasuries are where the money is. Just as Willy Sutton robbed banks because that’s where the money was, the #1-above-all rule of tax politics is **the gov’t has to be able to collect the tax** (All kinds of very idealistic-seeming tax ideas quickly crash and burn into ash simply because it is impractical for the govt to collect the tax. See from the Stamp Act to Georgist land taxes.)
All corporate revenue flows through a single choke point that is exceedingly easy for the govt to tap. And public corporations have to be very honest about their revenue, they can’t hide it. (At least compared to near all other kinds of taxpayers.) In this sense, the corporate tax does pass the ‘simplicity’ test with flying colors, it makes actual tax collection very simple. The #1 rule rules.
(2) The average person’s massive ignorance about the subject (along with near all such political subjects). The typical eligible voter believes either that corporate income tax lands on nobody, on corporations *instead* of people, or at most only on “rich shareholders” (somehow not on working-class people who are shareholders through pension plans, IRAs, 401(k)s). Of course on the motivated political left these beliefs are creeds.
Politicians follow and cater to the median voter in general, and to their motivated political bases in particular, so the corporate income tax has a long life expectancy ahead of it. The real-life practical question is whether it can be improved (as it was in TRA ’86, though it has been degrading ever since).
What about the income tax on business activities of LLCs, which is passed through to the owners personal income tax and taxed like wage income.
It may be worth noting that the ‘regular (C) corporation’, which is subject to the corporate level income tax, is now pretty much extinct, except for legacy firms that for some reason can’t convert to some kind of ‘pass through entity’ status and for publicly owned corporations and firms planning to become publicly owned.
In that sense, the situation with the corporate income tax has been improving.
25. April 2014 at 12:22
well there is that Apple example. of paying no corporate taxes at all. any where
25. April 2014 at 13:02
Unlike Ernest Hemingway, you sent me to the dictionary (actually wiktionary) to look up what AFAIK meant.
25. April 2014 at 17:34
“At a minimum they should remove any contribution limits on 401ks. Recoup the revenue with higher payroll taxes on upper income Americans.
And don’t liberals claim that working people can’t afford to save very much? Then my proposal shouldn’t cost much revenue.”
I love this idea. In fact, I think it wouldn’t cost any revenue in the long run, it would be a net gain to the Treasury.
But if we insist on making it revenue neutral in the short run, we should be thinking more in terms of sin taxes on consumer credit. Getting the poor and middle class to shift from wasting money on interest to saving is a better way to promote wealth equality than any soak the rich tax scheme.
25. April 2014 at 18:29
Scott, what is the economic incidence of capital income taxes? For example, as I understand it, the earned income tax credit (EITC) is actually shared by both employers and employees even though the statutory incidence makes it appear that only the employee receives the credit. Muni bonds are tax-neutral (“tax free” in the popular vernacular) but their interest rates are typically lower than rates on taxable bonds of comparable risk. So, it would appear that the tax on taxable bond interest is at least partially paid by the borrower (in the form of higher interest) rather than the bondholder. Borrowers, of course, include many low income people, whether it be credit card debt, auto loans, mortgages, or student loans. Capital taxes could also diminish the value of human capital because, presumably, the taxes cause equity investors to demand higher rates of return before they will invest. (Equity investors end up owning a larger fraction of a firm relative to founders and employees than they would otherwise, for example.)
If capital income taxes punish both savers and borrowers, both equity issuers and equity buyers, then would it be most accurate to describe them as taxes on shifting the timing (and risk) of consumption rather than as penalties on savers? That would appear to make capital income taxes welfare destroying for *everyone* since both savers and borrowers benefit from shifting the timing and risk of their consumption to meet their individual preferences.
25. April 2014 at 22:12
Nick:
Why on earth shouldn’t people who consume later pay more than people who consume now, if they employ the unconsumed resources to generate income while they defer consumption?
Suppose that it is our intention to tax income. The idea of an income tax, its purpose, is to render the tax base broad. That’s why we choose an income tax, rather than say a consumption tax. We can raise a given fraction of total production without unduly burdening any particular kind of claimant. If we need 20% of GDP, everyone pays 20% of GDP out if their stream of current income. It is not a Pigouvian tax. Neither production nor consumption are “bad” — indeed both are activities at the core of economic life, both are things that we wish to encourage in general. There are some times and places where we might wish for more savings and investment relative to consumption, and times and places where we might wish for less. But an income tax is not a tool for privileging investment over consumption, it is a tool for pooling resources from a very broad base.
(Note that progressivity of taxation is a violation of the basic principle that favors an income tax, as it intentionally shifts the burden disproportionately to certain claimants. Whether that is a good idea is a separate conversation. I think it’s actually a very complicated conversation. Sometimes it seems to me that favoring capital income — sorry, Scott, you are trying to favor capital income — is embedded in a kind of tit-for tat with progressivity on the other side.)
In so much of this conversation, people who consider zero capital taxes neutral are really saying that deferred consumption should be encouraged over present consumption. But lots of times, like now, we’d really want to encourage present over deferred consumption, because we have fallow resources. A neutral income tax, one which taxes all factors of production, privileges neither. If you wish to defer consumption, you can use your unconsumed resources as capital and generate income. You’ll pay a fraction of that income as tax. In every period, all factors of production will be taxed on a proportional basis, so the tax base does not fall in a concentrated fashion on any particular type of claimant.
People who defer consumption remain fully paid for their time at the market rate. There is no a priori reason to take the pre-tax rate of return as a more valid discount factor than than after-tax rate of return. We may all have our subjective time preference, but the discounted value of future money is defined by what we can get, which is always a function of technological and institutional factors. People who defer consumption are always fully paid for the time value of their money when they are paid at the going rate, just as people in the labor market are fully paid when they accept wages that are affected by technological and institutional considerations, including payroll taxes. The market rate is the market rate, on wages and capital, and including all the imperfections and necessities of the imperfect world in which it arises. As you point out, usually (though not always, cf now-ish!), the world is not so cruel to those who defer consumption. People usually enjoy a positive real return, even after taxes. It would be reasonable, I think, to index the basis for capital gains for inflation, so that we are only taxing positive real returns, because that is our intention — however we produce, everyone pays a share of the income they enjoy, so everybody pays the same share of income.
Ultimately the issue is simple: in each period of time, we wish to purchase a public goods that represent a significant fraction of GDP. Generating that revenue from some subset of the factors paid out of production concentrates the burden. Indeed, the not-at-all neutral policy of setting a rate of zero on income from capital places our ability to fund public goods at the perverse mercy of technological change.
Suppose we invent robots that make us fantastically wealthy, but which largely substitute rather than complement labor income. We’ve seen labor’s share of output fall in practice, in part probably due technological change, so this is not outlandish. Labor’s share falls to 20%.
But, the technological change is universal, so we still have to spend 3% of GDP for the public good of national defense to remain competitive. We need to build laser beacon sky highways and space elevators. All told, we (as we democratically constitute ourselves) still wish to spend the same roughly 20% of GDP on the central government as we do today.
Our ability to generate that wealth has not declined. In utility terms, it should be much easier for us to raise 20% of GDP when we are very rich than when we are poor and slovenly like now. Yet, by limiting taxation to labor income, we’d have to set the tax on wages at 100%, which should strike one as perhaps something other than neutral.
25. April 2014 at 23:10
Scott,
VAT taxes are well-designed consumption taxes. Neither Solow nor I would say they favor capital over labor. They favor neither. They tax all of factors of consumption equally though indirectly. I’ll not dispute you on that, because I’ll not dispute you when you are right.
But you are far from right with respect to income taxes, which are not and are not intended to be consumption taxes. The only sense in which the income tax regime you propose in “neutral” is when the domain of choice is whether taxes should be paid at the time of earning or the time of consumption. The tax regime you favor is neutral between those choices, holding rates constant.
But outside of individuals deciding (in the US) between a traditional and a Roth IRA , there is no meaningful sense in which your preferred income tax scheme is neutral. It favors capital income, full stop. Solow is not erring or pandering, he is speaking plainly. You can argue, perfectly reasonably if the facts bear out the argument, that capital taxes should be favored because capital formation is a binding constraint on growth and/or because of complementaries between capital and labor render capital-preferenec Pareto improving. But you have to adopt a very peculiar definition of “neutrality” to apply the term to a scheme that would tax inherited wealth not at all and labor income at a rate 50% more than the average tax burden.
We can go through a million 401K, IRA scenarios, or weird scenarios where an employer does your saving for you (and which ignore the taxes paid by the employer on the deferred payment). None of that is helpful. If we wish to make intellectual progress, we should characterize your position as plainly as possible. The tax policy that you advocate is, or is economically equivalent to, one in which wages are taxed at an average rate of [ avg_tax_burden * (1 / labor_share) ] and capital income is taxed not at all. The one “neutralish” quality that regime has is that the absolute amount of tax an individual pays on [ wage income + derived capital earnings ] is independent of the timing of consumption.
But, your scheme is not even neutral between, say, two individuals who defer consumption for the same period but with different investment returns. If we both earn $200 today, we both are taxed 50%, then I realize a 5% return and you realize a 20% return, I will have been taxed 48% of total income and you will have been taxed 42%. (We can debate progressivity, but outright antiprogressivity seems like a not-so-desirable quality.)
Your scheme is not neutral between individuals who adopt the same consumption schedule but begin with different endowments. Suppose that I begin with $1000 and you begin with nothing, and the interest rate is 5%. We both earn $100 in wage income. My total income is $150, your total income is $100, and we both pay $50 in taxes. My overall tax rate is 33%, yours is 50%.
Your scheme is not neutral between people with identical consumption schedules but different wages. Suppose that we each wish to consume $50 in each period indefinitely. I earn $100, you earn $200 each period. The interest rate is 5% and the tax rate 50%. In period 1, I pay $50 and you pay $100, 50% of each of our income. Neutral, yay! But in period 2, I pay $50, you pay $100, but your total income is $205. So now, I pay the same 50%, but you pay only 49% of your larger income. Your overall tax rate decreases over every period, asymptotically to 0% even as your income grows without bound. My tax rate remains a 50% forever. Doesn’t sound so neutral.
The tax policy you prefer is neutral only if one presumes a universal rate of return on capital and then define neutrality to mean “tax burden insensitive to consumption schedule”. That’s not what most people mean by taxation neutral between the factors of production. Solow is absolutely right to claim the current system favors capital.
You are right is to point out that the tax policy you prefer would be similar to a consumption tax rather than an income tax. But we have not chosen to adopt a consumption tax. We have chosen to adopt an income tax, because it spreads the tax burden much more broadly. You are very welcome to prefer and advocate a consumption tax! But it strikes me as a bit cheap to hide that advocacy behind a redefinition of the income tax to a tax that resembles a consumption tax, especially if you claim with technocratic authority your redefinition represents a neutral baseline from which we should all begin.
26. April 2014 at 04:16
(it was a bit cheap of me to use the phrase “it strikes me as a bit cheap”. sorry for that!)
26. April 2014 at 05:14
Tom, I’ll look at that later.
Jeremy, Yes, abolished. When people actually run businesses, their share of the profit gets taxed as ordinary income.
dw. The ideal does not exist.
BC, That’s right, and that’s why it makes sense to talk in terms of a higher tax on future consumption. The language is more precise.
Steve, Lots of points, I’m afraid I disagree with all of them. Let’s start at the end, where it seems to me you accept my point.
“VAT taxes are well-designed consumption taxes. Neither Solow nor I would say they favor capital over labor. They favor neither. They tax all of factors of consumption equally though indirectly. I’ll not dispute you on that, because I’ll not dispute you when you are right.”
But that’s exactly my point about income taxes. A VAT is like an income tax that taxes capital at a zero rate. If Solow believes a VAT does not favor capital over labor, then he logically must also agree that our current income tax does not favor capital over labor, because it hits capital harder than an income tax with a zero rate on capital income, which is identical to a VAT. That seems conclusive to me, but clearly you disagree:
“But you are far from right with respect to income taxes, which are not and are not intended to be consumption taxes. The only sense in which the income tax regime you propose in “neutral” is when the domain of choice is whether taxes should be paid at the time of earning or the time of consumption. The tax regime you favor is neutral between those choices, holding rates constant.”
An income tax regime with unlimited IRAs is essentially identical to a wage tax or a VAT. So how is it not neutral? Perhaps I misunderstood you. But I strongly object to the “intention” argument. What policymakers “intended” may be relevant to a strict constructionist on the Supreme Court, but surely not to questions of economic theory. But it’s even worse, as the fact that almost all countries tax capital at lower rates, and many highly progressive European countries have no capital gains taxes, or no inheritance taxes, shows that policymakers clearly “intended” the tax top look more like an VAT than would a pure income tax. So if I’m wrong about intentions not mattering, and if you are right, it still doesn’t prove Solow’s “favor’s capital” argument as there was never any “intention” among policymakers for a pure income tax!
My basic argument is that a pure income tax is a silly “baseline” for tax incidence theory. We all agree that who writes the check is different from who pays the tax. The cigarette companies who pay a big tax to the Federal government pass much of that on in higher prices. We all agree on that. So how do economists respond to that problem, how do they figure out the true tax incidence? They do a before an after effect on consumption, just as finance people studying CAPM do a before and after test on consumption. Consumption measures living standards, and hence you look at the size of the hit on consumption as showing the burden of a tax.
Because you accept my VAT argument about being neutral vis a vis consumption and capital, I’ll also assume you accept that standard argument that a VAT does not favor exports. AFAIK all economists accept this argument, but I’d guess that 90% of the public (when they hear about it, think VAT “favors” exports. I see this example as a near exact analogy of my argument. Not just a close analogy, but near exact. In both cases it looks to the average person like something is being favored (exports or capital income.) In both cases it looks that way because the LEGAL incidence of the tax on the seemingly favored item is zero. In the export case economists think the public is suffering from a cognitive illusion, and that the fact that VAT does not apply to exports does not mean VAT favors exports. So why does my argument for the income tax seem so far fetched?
Your argument about the two people with different rates of return tries to calculate tax burden by dividing by income. But that simply assumes I am wrong. My whole point is that you can’t determine tax burden by dividing by income. Ideally we’d like to divide by some measure of “resources” or “ability to pay.” But income doesn’t measure those variables, as I showed clearly with my example of two brothers with identical lifetime wage incomes and different preferences regarding saving. Your mistake would be like dividing VAT tax payments on imports by total import value, and comparing to tax payments on exports (i.e zero) divided by total export value. That will make the burden look lower on exports, but trade theory tells us that the zero tax on exports does not favor exports. It does not distort trade.
I’ll continue on the next comment.
26. April 2014 at 06:09
Scott,
two questions about consumption taxes:
1. Given that it looks that there isn’t much consumption inequality, such a tax system would see “the rich” pay a smaller share of the total tax revenue. No?
2. Assuming that consumption is very inelastic, wouldn’t a tax payer reduce his/her savings in order to pay the consumption tax and wouldn’t that amount to taxing savings?
26. April 2014 at 06:14
Steve, You said:
“Solow is not erring or pandering, he is speaking plainly. You can argue, perfectly reasonably if the facts bear out the argument, that capital taxes should be favored because capital formation is a binding constraint on growth and/or because of complementaries between capital and labor render capital-preference Pareto improving. But you have to adopt a very peculiar definition of “neutrality” to apply the term to a scheme that would tax inherited wealth not at all and labor income at a rate 50% more than the average tax burden.”
I’m not basing my argument on which regime is optimal, but rather on the fact that in this area, as in dozens of other areas of economics, “plain speaking” hides the truth. Indeed I’d almost say the english language fails us here. For instance, I toggle back and forth between claiming that wage taxes don’t tax capital at all, and the claim that they tax capital income at the same rate as labor income. Why am I so confusing? Because the first claim is about the legal incidence of the tax, and the second claim is about the economic incidence of the tax. To review, the unlimited IRA hypothetical is equivalent to a pure wage tax. And yet the legal incidence of the taxation of capital income is 0% under a pure wage tax, and the legal incidence of the taxation of capital income under unlimited IRA income taxes is exactly the same as the tax rate on labor income! That’s a huge problem for your argument, and I don’t see you grappling with it effectively. You seem to be relying on common sense arguments that look towards the legal incidence of taxes. So I reject your claim that a consumption tax regime would exclude heirs, for the same reason that unlimited IRAs would not exclude heirs.
If you don’t like that argument, here’s another. A wage tax seems to exclude heirs, where as a VAT tax hits heirs. But the tax incidence of a wage tax and a VAT are identical (apart from second order implement issues), so it’s simply not possible that appearances can be correct in this case! Indeed if the consumption tax were implement via an income tax with unlimited IRAs (a very plausible possibility, favored by many economists) then these heirs would pay taxes at the exact same rate as wage earners. And yet you claim that a completely identical wage tax would let them off Scott-free (pun intended.)
Again and again I’m showing that appearances are deceiving. The export exclusion under VAT, the wage tax vs. the unlimited IRA income tax, the wage tax vs, the VAT. And yet I see you continuing to insist that a income tax with lower rates on capital favors capital because it looks that way “plainly speaking.”
You said:
“The tax policy you prefer is neutral only if one presumes a universal rate of return on capital and then define neutrality to mean “tax burden insensitive to consumption schedule”. That’s not what most people mean by taxation neutral between the factors of production.”
“Most people” don’t understand the nuances of tax theory, and suffer from cognitive illusions. If we poll the public on economic issues we find total absurdities. Most people I talk to believe imposing higher taxes and/or regulations on corporations will cause them to raise prices, but cutting taxes and regulations won’t cause them to cut prices! (They’ll “pocket” the extra profit.) As if the profit function wasn’t symmetrical. So let’s not talk about what “most people” think. Let’s talk about what’s really going on when we demystify the subject. What do the public finance economists think?
You said:
“We have chosen to adopt an income tax, because it spreads the tax burden much more broadly.”
Choosing to impose a higher tax on future consumption than current consumption in no way spreads the tax more broadly.
Nor do I assume a universal return to capital, as you claim. That assumption plays no role in my argument. (Although I concede it might play some role in the separate argument about the optimality of pure consumption taxes.)
You said:
“Why on earth shouldn’t people who consume later pay more than people who consume now, if they employ the unconsumed resources to generate income while they defer consumption?”
Because the cost of the unconsumed resources is already incorporated into the price the the future consumption goods they buy. So they are being taxed on those resources.
You said:
“But an income tax is not a tool for privileging investment over consumption, it is a tool for pooling resources from a very broad base.”
Then why have VATs? Wouldn’t the “base” be broader if a lower rate applied to each stage of the production process, raising equivalent revenue? Economists believe (quite rightly) that base broadening that is distortionary is unwise.
You said:
“Note that progressivity of taxation is a violation of the basic principle that favors an income tax, as it intentionally shifts the burden disproportionately to certain claimants. Whether that is a good idea is a separate conversation. I think it’s actually a very complicated conversation. Sometimes it seems to me that favoring capital income “” sorry, Scott, you are trying to favor capital income “” is embedded in a kind of tit-for tat with progressivity on the other side.”
This is a common mistake. The debate over progressivity is logical orthogonal to the debate over income taxes. Just as there is a NGDP target path that is equivalent to any given inflation target (contrary to popular belief) there is a consumption tax that is equally progressive to any income tax regime. BTW, I DO NOT favor any reduction in the progressivity of the US tax regime, take as a whole, and would support an increase in progressivity if combine dwith tax reform. Unfortunately it is impossible to measure the progressivity of a large tax regime. Indeed it’s impossible to even get a good estimate. The entire issue is shrouded in uncertainty. But in terms of LEGAL incidence, I’m fine with really high consumption MTRs on the rich. There’s no way my plan would be adopted without something like that.
You said;
People usually enjoy a positive real return, even after taxes. It would be reasonable, I think, to index the basis for capital gains for inflation, so that we are only taxing positive real returns, because that is our intention “” however we produce, everyone pays a share of the income they enjoy, so everybody pays the same share of income.”
It’s pretty generally agreed that it would be technologically impossible to do this, except at enormous cost. Think tax forms are annoying now? Imagine calculating the CPI every single time you take money out of your bank account by writing a check. So perhaps you would favor a second best compromise, such as lower tax rates on capital income? But Solow would argue that system “favors” capital—indeed he (essentially) does make that argument (unless he was figuring in the inflation distortion when he made his comment, which seems unlikely to me. Would he have said the same thing in 1979 when capital was getting hammered by inflation? Probably.) When I sell my 2 family house in a few years I’ll be hit with a high capital gains tax, much of it inflation. But the progressives will look at my income and say I’m one of the “one percent” that year. Because income truly shows “class,” or so they claim.
And finally, your robot scenario could be accommodated via a consumption tax that was implemented via an income tax with unlimited IRA contributions. I presume consumption would remain a substantial share of GDP in your scenario–you do say we’d be fabulously wealthy.
Totally unrelated, I thought you piece a wile back in defense of single moms was great–I have a related piece but haven’t had time to finish it yet.
26. April 2014 at 06:21
Pietro, The rich would appear to be paying a lower percentage of taxes (legal incidence would be lower), but they’d probably end up paying a higher percentage in terms of economic incidence.
2. No, it encourages more saving, relative to an income tax (which double taxes saving)
26. April 2014 at 06:22
Pietro, In other words, the rich would consume less resources, leaving more for the rest of us. Fewer yachts, more average cars.
26. April 2014 at 06:27
Scott wrote:
“An income tax regime with unlimited IRAs is essentially identical to a wage tax or a VAT.”
It would be different for a trust fund baby who never worked a day in his life, wouldn’t it? He would pay less tax on consumption (funded by spending down his inherited wealth) than he would under a VAT. Give me an estate tax and I’ll buy in to your plan.
Also, it seems unrealistic to think that we will ever have a simple tax code that isn’t riddled with loopholes and exceptions for various special interests, such as the one you mention (ways to characterize what should be wage income as capital income).
I’ll assume your argument is correct: a simple progressive tax on wage income is superior to a “simple” system of taxes on wage income, capital income, corporate income and payroll. Does it therefore follow that a progressive tax on wage income that is complex and riddled with loopholes and exceptions is superior to our current system?
26. April 2014 at 07:34
I’m scratching my head at the arguments here. Isn’t it true that under our current tax system, if you take two otherwise identical workers and vary their savings preferences, that the saver will experienced reduced lifetime consumption? If that’s true (and I’m pretty sure it’s true) then isn’t Scott correct?
26. April 2014 at 08:43
O/T: Mark Sadowski, I’m almost surprised to see you didn’t respond here:
http://longandvariable.wordpress.com/2014/04/23/if-i-was-devising-a-panics-and-bubbles-course/comment-page-1/#comment-799
26. April 2014 at 10:34
Steve, try to do point by point with Scott, this is fun.
But mainly Mr Waldmann,
I understand and agree with progressives desire to soak elites, but:
The ONLY true mongoose that can kill the Cobra (elites and their heirs), are ENTREPRENEURS.
Governments are TERRIBLE at killing Cobras, and GREAT at killing mongooses.
ALL DAY LONG we are treated to a parade of les enfants terribles (forgive me for speaking a dead language) rich kids who have sidled up to Big Government:
http://www.nytimes.com/2014/04/20/fashion/white-house-hosts-next-generation-young-and-rich.html?_r=0
—-
The whole thing is disgusting. It is a big fucking lie (sorry Scott). And YOU, Steve are not a liar.
Liberals who talk about ‘taxing wealth” DO NOT CARE AT ALL about whether wealth is passed down to poodle children THEY PREFER IT!!!!
That’s why there is no careful scalpel approach to simply FAVORING newcos and SMBs over old companies.
That’s why there is NO effort to end regualtions that hurt the mongoose.
Because ONLY mongooses can kill rich kids, government can’t and doesn’t want to!
We need to call them all liars. They want to grow government, and don’t care a single iota about the inheritance.
26. April 2014 at 10:41
Steve, right away in this thread you got your dander up about WHAT IF you inherited $2000.
It “offends” you the difference between what you earn and what you are given.
There is no offense. There is no difference between what you earn and what YOU decide to give to your kid.
When you eat a bite of steak, your kid can’t eat it.
When you forgo it, they get to eat it.
BOTH ARE MORALLY YOUR CONSUMPTION. You can’t eat for another, but you can enjoy them eating something by your hand.
Have you ever fed a pet? Rich children are poodles.
So AGAIN, don’t get confused, the fastest way to kill off a gang of poodles, is to let a n army of hungry entrepreneurs lose in their midst.
SO if you let government get in way of entrepreneurs, you are PRO POODLE.
Don’t be pro-poodle Steve, they are shitty dogs, that basically only pee and shake.
Stop protecting the weak soft underbelly of rich kids Steve.
Let us feast.
26. April 2014 at 11:04
Tom Brown,
I saw the link at Economist’s View on Thursday and chose not to read it (there are only 24 hours in a day). I read it now, along with some of the background links and the comments.
The back and forth between Tony Yates and Sakir Devrim Yilmaz (the “Panics and bubbles” course designer) in comments is fascinating. And I have to say I generally come down on the side of Yates.
Yilmaz’s suggested reading list includes a lot of material which I would classify as highly optional, in particular the section on Post Keynesian, Circuitist and ABCT economics.
http://www.post-crasheconomics.com/wp-content/uploads/2013/10/Bubbles-2013-14-Course-Outline.pdf
There is already so much that students need to know and it’s not clear to me that Yilmaz himself has even mastered what I would consider to be core knowledge. It’s more than a little disturbing that Yilmaz thinks Randy Wray, Scott Fullwiler, Riccardo Bellofiore and Claudio Borio and are more important reading than Diamond-Dybvig, Bernanke-Gertler and Robert Solow.
It’s like a diet of pudding and no meat:
https://www.youtube.com/watch?v=n5diMImYIIA
I guess I’m in danger of becoming cranky in my middle age.
http://rtsponderings.files.wordpress.com/2010/09/grandpa-simpson.jpg?w=584
P.S. The University of Manchester Post Crash Economics Society (PCES) sounds vaguely crackpottish. I would tell those damned kids to shutup and go do their homework.
P.P.S. It’s also interesting that Yilmaz thinks Stephen Williamson is “the most famous DSGE modeler” (highly disturbing if true).
26. April 2014 at 11:26
Mark, thanks for your thoughts. That’s funny. I was *sort of* kidding about you not responding: I think it’s amazing that you have the kind of presence that you do around the internet. I’m glad you found it fascinating.
On another subject, did you see Noah’s latest on modernity? I’d never heard of “neoreactionaries” … but from what I can tell they sound like they might be Fantasy novel reading, democracy hating, “traditionalists” (i.e. white supremacists). What do you think?
https://radishmag.wordpress.com/2013/05/24/heroes-of-the-dark-enlightenment/
26. April 2014 at 11:29
Scott, you said, “BC, That’s right, and that’s why it makes sense to talk in terms of a higher tax on future consumption. The language is more precise.”
I guess my question was whether tax is really higher for *future* consumption or if it’s higher for any *shifting* of consumption, whether one accelerates or delays consumption. Suppose with no taxes on interest, interest rates would have been 10%. I expect to receive a $110 cash flow in one year, but I want to consume $100 now. You receive a $100 cash flow now, but want to consume $110 in one year. With no interest taxes, I could borrow $100 from you and pay you back $110 in 1 year. We both are better off. If interest is taxed, however, you might require a pre-tax interest rate of more than 10%. As a result, I borrow some amount less than $100 from you and pay you back $110 in 1 year. You receive more than 10% interest pre-tax, but less than 10% after tax. So, after taxes, you consume less than $110 in 1 year. We *both* are penalized for shifting our consumption, me for consuming *earlier* and you for consuming later. If I had preferred consuming $110 in year and you preferred consuming $100 immediately, our total consumption would have been higher. The tax penalizes those whose cash flows don’t naturally (without borrowing and lending) match their desired consumption.
That seems like a strange principle on which to tax because a mismatch between timing of desired consumption and cash flows is not necessarily an indicator of wealth — poor people may want to borrow, for example — nor does it seem to be an indicator of value of government services consumed.
26. April 2014 at 12:32
I generally agree with Scott here, with nothing theoretical to add. I do, however, want to make an observation about politics (which is at least part of what the OP was about, right?). 20 years ago, with two economists (Phil Gramm and Dick Armey) in Republican Congressional leadership, Scott’s view (there should be no tax on the deferral of consumption — if that’s a fair characterization) had considerable support among Republicans. During the George W. Bush administration, at the vehement urging of Grover Nordquist, a high-limit save-for-anything IRA proposal was sent to Congress twice. Both times it was DOA, but, 401(k) limits were increased, and that increase was supported in the Republican party explicitly on the same basis put forth here — that income should only be taxed once and that returns to the deferral of income should not be taxed. Now — witness Dave Camp’s “tax reform” proposal — Republicans have totally abandoned these policy concerns.
In that regard, I would say, >of course< Keynesian liberals (Democrat economists) want to tax savings — Keynes thought saving was at best morally problematic. (According to Skidelsky: "I think [Keynes] recognized intellectually, in order for capital accumulation to go on, you have to have a high savings rate. But then, he didn't like saving. He didn't like postponement and always looking to the future as a moral and psychological quality. He admitted it was necessary but he hoped that it would give way as quickly as one possibly could to enjoyment, and getting enjoyment from the present rather than always thinking about the future.") And they are always trying to boost demand by getting people to spend now. That Republicans have given up on our side of this fight is tragic.
26. April 2014 at 13:37
Tom Brown,
Michael Anissimov said the following in the post Noah Smith linked to:
“â– Speaking of economics, why are we stuck in a perpetual recession? Do we really think that printing more money will be the way out of this one? How far can we go into debt before there is a fiscal collapse?
â– Our nation is deeply in debt, both household debt and government debt. In 1970, household debt was barely a thing. Today, US households are about $14 trillion in debt, roughly the US GDP for an entire year. Don’t even get me started about government debt.”
and:
“â– Structural employment is destroying millions of jobs. They are being sent overseas, never to return. The labor force participation rate is only 62.8%, a 35-year low. Only 43.6% of Americans have full-time jobs. Young people are earning less money than ever-about $10,000 a year less than early Boomers.”
That sounds like the usual knee-jerk inflation-fearing debt-phobic supply side nonsense that one typically finds in the right-wing blogosphere.
A scan of the map in the link you provided reveals some “Neo-Reactionary” economists are Hans Hermann-Hoppe,
Randall Parker and Tino Sanandaji. They all sound like adherents of some form vulgar Austrianism to me.
The map also seems to show the three main flavors of Neoreactionaryism are traditionalism, merchantilism and nationalism, which implies to me that Neoreactionaryism is just a 21st century spin on fascism.
And in general, after looking at the link you provided, I can’t help thinking Neoreactionaries view themselves this way:
http://2.bp.blogspot.com/-vXjLInkLCm0/UGEkE_gfMuI/AAAAAAAAC0E/Rqj7WbANBR8/s1600/tony-stark.jpg
When the rest of of the world just sees them this way:
http://2.bp.blogspot.com/-ojNRfH0PeWs/UGEkNvzz-tI/AAAAAAAAC0M/tIYUFjITgek/s1600/super-computer-nerd.jpeg
26. April 2014 at 14:04
Mark, Ha!… Yes, I’ve reused those pics from Noah’s bestiary too, and I agree they are appropriate here… they are my favorites in one of my favorite Noah posts ever.
26. April 2014 at 17:40
Not to digress, but 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%.
27. April 2014 at 14:53
Michael Byrnes, I’ve never much cared for that sort of argument. First we need to figure out the best system, and then think about what compromises are needed, and what second best policies might be better. But we haven’t yet even reached the first stage.
If other countries have tax systems that are not riddled with loopholes, why can’t we?
BC, Sorry, I didn’t follow that. Maybe someone else can answer.
Michael Barry, I didn’t know the GOP had given up on IRAs. That’s too bad.
dtoh, Good point.
28. April 2014 at 05:35
Scott,
First, thanks for being so even tempered. I’m dyspeptic at the moment, I’ll try to write up why later, but I appreciate your not rising to that bait.
(Let me put that more clearly. I have been something of an asshole. I still am, but I am sorry about that. Particularly since you are an unusually nice guy!)
We don’t agree about this, but your rebuttal is good. To respond, I’ll be slippery and, um, “clarify” (take back) something that I said. I erred when I wrote, “[VATs] tax all of factors of consumption equally though indirectly.” VATs don’t tax all factors of, um, PRODUCTION-not-consumption “equally”. Within each period, they tax the factors of production and the income of the purchaser according to some complicated pattern of incidence. Whether that’s “fair” is hard to say.
What is obviously true is that taxing each intermediate good separately would be stupid, it would discourage some perhaps technically-superior modes of production in favor of minimizing intermediate steps. So that’s important to this argument, I think. It is not true that labor and capital are taxed “equally” by a VAT. It is that they are not taxed in a way that would be particularly arbitrary and stupid. However, the situation that would make per-factor taxation stupid for a VAT, the bias towards less-roundabout modes of production, is not an issue with an income tax. An income tax “distorts”, no doubt, any way you do it. Any tax does. But the form of that distortion is very different than the absurdity that results from a per-stage-of-production excise tax, and may not be more harmful than what results from a VAT.
In an (non progressive, simple percentage) income tax all factors are taxed at the same rate. That is indisputable by construction. If we have F(K, L), then pre-tax income to providers of capital is rK, pre-tax income to labor is wL, capital taxes are t*rK, labor taxes are t*wL, total taxation is proportionate to the compensation paid each factor. I think this math is indisputable. We come down to a question of language. When we talk about taxes being “fair” or “equal” or “proportionate”, over what time horizon are we talking? If we are talking on a per-period basis, an income tax wins. And this is not an unreasonable way to think about things, regardless of whether one considers it the best way. It’s quite common, I think, even among economists very familiar with tax theory, because it’s straightforward and more elaborate approaches yield conclusions that remain tentative. The income tax vs consumption tax debate goes back at least to JS Mill and it remains unresolved. It’s pretty clear, for example, that neither Robert Solow nor Joseph Stiglitz consider a 0% appropriate, as a capital tax rate or as a baseline for a capital tax, which implies that they would not favor a consumption tax. Stiglitz provided much the modern basis by which the argument for a consumption tax is now made, and in the early 1980s he wrote favorably about reducing the capital gains tax. Perhaps they are both now deranged by politics, but perhaps they understand the models perfectly well but have updated their priors about the applicability or relevant parameters of those models. Perhaps then they have retreated to fairness in a per-period sense, not based on a strong assertion of optimality but as a reasonable, intuitive heuristic under constrained uncertainty. That strikes me as very much like a NGDPLT target, for example. Although plenty of celebrated models proclaim the optimality of inflation targeting, you (me too!) have updated our priors from experience and reject those models’ applicability. No elegant, widely subscribed model declares the optimality of NGDPLT, but given what we know it seems approximately right and a kind of parsimony recommends it over some alternative weighting of per-cap RGDP growth, population growth, and inflation. It’s a good-enough best guess upon which human economists and policy makers can easily coordinate.
You’d argue that we should not think on a per-period basis, but over a longer-term equilibrium. Your reasoning, I think, depends on the following suppositions: 1) All income is originally earned as wages (or else taxed at the wage rate); 2) All capital is eventually deployed as deferred consumption of labor income; 3) No scarce goods are consumed in the holding of capital that are not redeemed in full with returns at the end of the holding period; and 4) Absent taxes, the world would be in an efficient, optimal equilibrium such that the goal of tax policy should be to minimize any distortions; 5) Our only goal in tax policy is to minimize distortions from that no-tax equilibrium. If all of that were true, your case for favoring a consumption tax could be strong, albeit contingent on details of the model that you choose. (You’ve made me do some homework Professor S!) However, none of those conditions are actually true, and even if they were, the policy recommendation would be model dependent. [The best paper I’ve found on this is Atkinson and Sandmo (1980), the same Atkinson whose work with Stiglitz (1976) is the frequently cited as the modern foundation of consumption tax preference, under assumptions of separability of leisure and consumption. Atkinson and Sandmo show that model is in fact is very sensitive to those assumptions.]
Re (1) and (2), in the real world, lots of income is never taxed as wages. There is the obvious, resolvable-but-would-we-actually case of inheritances. There is also entrepreneurial income. Most of Bill Gates’ wealth was never paid as wages. (3) is a deeper issue we’ll have a hard time agreeing on, but see my piece here, and consider the stylized empirical fact that wealthier individuals fail to decumulate assets as any version of the permanent income hypothesis would suggest, and behave as if money enters directly into the the utility function in some form or another, and that form might be relative and therefore rival; (4) is trivially false, but much economic analysis proceeds on the basis that it were true; and (5) neglects the obvious importance of distributional goals, avoiding taxing people into starvation, etc. (If there were not these other concerns, we’d just choose a lump-sum tax and be done with it, right?)
When these conditions fail to hold, the equivalence of a wage tax and a consumption tax breaks down and we have no reason to presume any obvious sort of optimality to either a consumption tax or an income tax. Yes, we can still tax consumption. But the models that tell us to do so become suspect, since they rely upon capital taxation reducing labor provision on the part of would-be savers. Putting that aside, Atkinson and Sandmo show the zero capital tax is sensitive to the value of several elasticities and cross elasticities. That doesn’t redeem the income tax, exactly. Even if we know the optimal tax rate were nonzero, Atkinson and Sandmo point out that it’s quite unlikely it would happen to be the same rate as the wage tax. The best they can do is make qualitative conclusions, like (intuitively) that a high elasticity of labor supply to the after tax wage relative to a low price elasticity of future consumption (and therefore saving) prefers more capital taxation, while the opposite would prefer less. Other, more surprising cross-elasticities seem to matter too. Even in models where capital taxes do seem inferior, the scale seems small, especially considering that risky portfolios should be unaffected, since deductible losses offset taxable gains and desired risk levels can be restored by leverage. Recently, some guys called Piketty and Saez (before one of them left economics to become a supermodel) offered a direct challenge to the traditional Atkinson-Stiglitz and Chamley-Judd foundations of a consumption tax preference.
It oughtn’t be surprising that real life is too complicated for “simple” (umm…) models to describe, and that reality does not conform to the assumptions of simple models. Sometimes there are results that seem very robust: that survive lots of plausible model specifications, that accord with observed experience, that accord with the intuitions both of professionals and the polity. Sometimes, more usually, there are not such results. We are faced with an epistemological problem: How should we react when “model uncertainty” is very large?
The way we should not react, I’d argue, is to pick a tentative conclusion from uncertain and uncertainly applicable models and hew to them as settled results, appealing to professional expertise for authority. We simply do not know, to any meaningful degree, what tax system would be “optimal”, even within the straitened and abstract (and quite various) criteria economists use to define optimality. We don’t know enough to declare a “baseline”, nor do we have sufficient consensus, within the polity or within the economics profession, to declare someone wrong or insincere for expressing a view. It is undoubtedly true that some expressed views are influenced by politics. Mine certainly are! But that is true on all sides of a question. And politics themselves are endogenous. Stiglitz was favorably disposed to a capital tax preference before he was against them. I would have been too, around Y2K. A correlation between changes in ones politics and economics may result from a common third factor: actual events in the world that render older conclusions suspect. A rational Bayesian will take this into account, even as the formal literature takes time to catch up. (Yes, this new literature will be affected by political currents. As was the older literature.)
None of this fuzziness means that one cannot have a view and argue for the view! One way or another, we have to make individual and collective decisions under uncertainty, and we should do the very best that we can at it! And it’s great to persuade, or to try to persuade. Hopefully (not certainly), on average, the ability of ideas (however counterintuitive) to be explained and expressed persuasively is correlated to some degree with the likelihood that those ideas are useful. We should (and obviously I should practice much more than I preach) persuade and be open to persuasion respectfully. But we really have no basis to claim authority.
A bit more substantatively, I think we should heed Chris Dillow’s wisdom and focus more on mechanisms than models. It’s a fool’s errand, I think, to try to find a model good enough, and to measure parameters well enough, to determine one true “optimal tax policy”. It’s also foolish to imagine that any model we might characterize would be stable. History happens. Circumstance-variant policy rules might be more robust than fixed parameters. I don’t think your proposed NGDPLT rule, for example, represents optimal or eternally correct policy, but I think it represents much better policy than we have now and because it is a rule it will probably endure a while if implemented. (We might prefer different implementation strategies, but let’s leave that aside.) Still, even solid policy rules are unlikely to prove infinitely durable.
I don’t think there should be a tax preference for capital income right now, with “tax preference” defined on a per-period relative factor tax basis. I am drawn to that conclusion not because the models people use to support the zero-tax condition are worse than most economic models, but because like most economic models they are fabrications of questionable applicability worth taking seriously not a priori, but because we find that they usefully describe and summarize events in the world. (Look, ma! I’m Milton Friedman.) We don’t have reliable empirical techniques to judge our experiment with capital preference, we don’t have counterfactuals, but we do have a society in which wealth gains are clearly cumulating via relatively concentrated capital holdings and in which (counter to the assumptions required to apply Chamley-Judd-style models) financial capital availability does not seem to be a binding constraint. Although I know you don’t, I view accelerating wealth and income dispersion (as well as social segregation related but distinct from 1%-ish top inequality) to be toxic to aggregate welfare, via a variety of conventional and less conventional (to economists) channels. Capital taxation is a straightforward mechanism to mitigate wealth cumulation in ways that mere consumption taxation, under any plausible degree of progressivity, could not. (At the tippy top of the distribution, tiny in fractiles but large in dollars, incomes are many multiples of consumption.)
There is no reason you should lend much credence to these views if they don’t persuade you. I have no irrefutable evidence. Please do argue! But in terms of math and internal consistency, Piketty and Saez (plus others no doubt to come) can rival Atkinson and Stiglitz and Chamley and Judd. You can always write a model. Ultimately, as economists we have to be arguing about phenomena in the world and whether they resonate with our models. The world always comes first, even though we can generate consensus more easily around a math problem than about whether, say, in some achievable counterfactual, the Federal Reserve could have maintained a steady NGDP path. New Keynesian models whose instrument is an interest rate don’t dissuade you from saying yes, they could have, even though those models enjoyed “new synthesis” support from the profession prior to the crisis. You are not a crank for defying that consensus and positing an alternative that the mathematicians of the profession haven’t caught up with. You are correct and courageous, even when I think you are wrong.
But models from the 1970s and 1980s that don’t seem consistent with current developments won’t and shouldn’t dissuade Stiglitz, or Solow, or Piketty, or even me from suggesting that capital taxation might be a very direct mechanism to address what we perceive to be an acute problem not foreseen in the 1980s. We are not “pandering to ignorance” or contradicting reliable truths. We may be mistaken, but we are expressing well-informed if unconventional economic views on the basis of evidence at least as reliable as a few clever thought experiments.
28. April 2014 at 06:09
Steve, Because of my heavy grading load right now, just a quick comment for now. I’ll have an extended reply later.
You are actually a very polite and extremely thoughtful commenter, so don’t have any concerns on that account. Meanwhile, see if you like my new post better.
28. April 2014 at 07:43
Scott,
Some specific replies:
(1) Re: heirs, you are reading other things into that sentence than I intended. If we think of all income as originally labor income and we think of heirs in continuous terms, than we’ve no dispute that either a consumption tax or a wage tax would hit heirs as much as it would hit sires had they lived forever.
But if we are thinking in terms of fairness and discontinuous people, a bequest becomes and endowment, untaxed time value becomes an increment to the unearned income in that endowment. Capital taxation is itself a form of bequest taxation (and indeed Piketty and Saez analyze them as substitutes). If, from a fairness or meritocratic perspective you consider bequest endowments problematic, capital taxation ameliorates the problem and its absence exacerbates it.
(2) You said: “Choosing to impose a higher tax on future consumption than current consumption in no way spreads the tax more broadly.”
Yes, it does, if what you mean is taxing only current consumption. The timing of flows matter.
Consider a population of pure deferrer/capital-holders and pure wage-earner consumers, say there are half and half. A tax TY must be collected in every period. Labor share is constant at 0.5. Under an income tax, wage earners consume 0.5(1-T)Y in each period. Under a consumption tax, the most they can consume without borrowing is (0.5-T)Y. The difference is 0.5 TY per period.
Perhaps our consumers can borrow to make up the difference of 0.5TY, on the theory that at some point in the future, deferrers will start to consume their large stock of wealth, resulting in a reduction of the tax rate which will eventually lead to reductions in their taxes from which they can pay, in the limit, 0.5TY per period until their debt is eventually exhausted. If capital markets are perfect, there are no debt limits, and deferrers can’t defer over an infinite horizon, they must eventually consume, that would work.
But none of those assumptions are realistic. No, there are not pure deferrers, that’s not realistic either. But caches of capital do cumulate indefinitely in the world. (The very wealthy do not decumulate per the permanent income hypothesis, foundations are perpetual, etc.) People are likely to face debt limits, even if deferred tax assets should render them solvent over an infinite horizon. Plus, people face credit spreads, creating a deadweight loss for the population on whom the tax burden is initially concentrated. (They could not both consume their full 0.5(1-T)Y per period and remain solvent, if we imagine some epsilon sliver of bankers who merely consume the credit spread.)
Flows matter. An income tax puts less burden on the financial system than a consumption tax under inequality such that some people entities consume small fractions of their wealth and defer for long periods of time while others seek to consume the full present value of their wealth.
(3) You said: “Because the cost of the unconsumed resources is already incorporated into the price the the future consumption goods they buy. So they are being taxed on those resources.”
This is a normative view. There is no a priori reason why r is a more appropriate discount rate than (1-T)r.
“Then why have VATs? Wouldn’t the ‘base’ be broader if a lower rate applied to each stage of the production process, raising equivalent revenue? Economists believe (quite rightly) that base broadening that is distortionary is unwise.”
In the VAT case it would be foolish, because it would impose a costly bias into the technical process of production. It’s the distortionary that’s the problem, not the base broadening. You will claim that the extra cost of period n consumption is distortionary, but it is balanced to a degree by a reduction in period 0 discouragement of labor, and we get into the models and elasticities where details matter.
(4) Re progressivity: I never claimed that progressivity is not possible for a consumption tax! On the contrary, I was musing about whether it is desirable for an income tax. I am less sure of that than you’d probably think. But we agree, it’s perfectly possible to make a consumption tax progressive.
I don’t think the issues are quite orthogonal, though. Again, we run into your assumption that all saving is deferred consumption. But that’s counterfactual. A substantial fraction of savings cumulates indefinitely, rendering some levels of progressivity possible under an income tax regime impossible under a consumption tax regime.
“It’s pretty generally agreed that it would be technologically impossible to do this, except at enormous cost. Think tax forms are annoying now? Imagine calculating the CPI every single time you take money out of your bank account by writing a check.”
The Treasury department considered it feasible in the mid-1980s. See here, pp. 178-200, ht Bankman and Griffith. It would be much more feasible now. Really. The arithmetic isn’t hard, and bank Forms 1099 are generated electronically.
28. April 2014 at 07:47
Scott, I’ve been afraid to read the new post, I’m so backlogged on the old ones. Please don’t feel obliged to spend a lot of time. Your are generous in your evaluations, but felt very rude, am writing at length for absolution…
28. April 2014 at 08:00
Last one… I’ll not quote the full conversation. But.
Under my original comment, you had me to rights. If a VAT taxes factors equally or “fairly”, then a VAT plus extra tax on capital could not.
But, I’m a squirmy eel and I took that one back. The new party line (I’m being self-dismissive, but I do think this is right) is that the incidence of a VAT is neither fair nor unfair, just a matter of incidence due to elasticities different elasticities, between buyers and final and intermediate producers. We can’t evaluate the “fairness” of that, except to the extent that it represents a kind of market outcome.
But a VAT is nevertheless wisely designed, because explicitly taxing factors pro-rata would have terrible consequences.
With an income tax, taxing all factors does not have terrible consequences — even if there is a cost to the non-neutrality over time, it’s likely to be small. Taxing all factors proportional is presumptively fair, even though, sure, incidences are complicated and we are never sure who really ends up paying what. I won’t say superficial neutrality is uniquely “fair”, but it’s plausibly so and has the virtue of being performatively so (that is, it’s easy for people to think of it as fair).
29. April 2014 at 05:26
Steve, You said:
“In the VAT case it would be foolish, because it would impose a costly bias into the technical process of production. It’s the distortionary that’s the problem, not the base broadening. You will claim that the extra cost of period n consumption is distortionary, but it is balanced to a degree by a reduction in period 0 discouragement of labor, and we get into the models and elasticities where details matter.”
This is precisely my problem with explicit taxes on capital (that implicitly double tax capital). They distort the production process, causing firms to use too much labor and too little capital.
You said;
“I don’t think the issues are quite orthogonal, though. Again, we run into your assumption that all saving is deferred consumption. But that’s counterfactual. A substantial fraction of savings cumulates indefinitely, rendering some levels of progressivity possible under an income tax regime impossible under a consumption tax regime.”
I don’t think I agree with this. You can only put a burden on a taxpayer by cutting into their consumption. A infinitely-lived family of misers is not a problem for inequality. Check out the Steve Landsburg post on Scrooge. If they never consume, they never lower the living standards of the rest of the population. The only burden of a tax is less consumption by SOMEONE. Full stop.
You said;
“If, from a fairness or meritocratic perspective you consider bequest endowments problematic, capital taxation ameliorates the problem and its absence exacerbates it.”
I’m a utilitarian who doesn’t care about fairness and meritocracy, so I think bequests are good. They mean many people consume fortunes, not just one person. Declining marginal utility.
Again, the relevant choice is not bequests vs taxes, it’s bequests vs. consumption by the original rich guy. He can current evade bequest taxes by consuming his fortune. Of course I do favor progressivity, I just see no reason to hit the consumption of the children of the rich harder than the consumption of the rich person himself. If this were justified, then our social norms should be that leaving part of your fortune to your children is less ethical than spending it all on wine, women, and song. Is that really our social norm?
In Europe, billionaires are legally required to leave more than one half of their fortune to their children. You and I are probably equally outraged by that. But consider how far the social norms in egalitarian Europe are from the view that bequests are an evil. For American progressives, it would be like stumbling on a land where not only is one allowed to be a pedophile, but it’s legally mandated! Of course this is hyperbole, but I find it interesting that you don’t hear these European laws discussed very often.
You said;
“The Treasury department considered it feasible in the mid-1980s.”
Are they including every single check? Why do mutual fund companies have so much trouble getting the current (infinitely simpler) accounting correct? I don’t even consider our current tax system to be “feasible.” I have a Phd in economics and I have no idea if I am doing my taxes correctly. For me, figuring out taxes every April is a nightmare. (I have an unusually complicated situation, admittedly) So I’m personally very hostile to proposals to make the system more complicated. Especially when there is absolutely no economic rationale for indexing capital income for inflation, AFAIK. Our models say income from capital should be taxed at zero, or at a lower rate than wage income. Are there any models that say capital and wage income taxes should be identical? I guess it’s theoretically possible, but about as likely as me winning the lottery. Only in those models would pure indexation be optimal.
I agree with your last comment about appearance of fairness if the comparison is income taxes and VAT. Not sure I agree if the comparison is income taxes with no IRAs, and Income taxes with unlimited IRA ability (a disguised consumption tax.) My whole argument (as you may have noticed) is about cognitive illusions. Indeed “illusions” are pretty much all I ever blog about. “Never reason from a price change.” “Money has been tight since 2008.” “Tight money caused the Great Recession.”
Indeed the name of my blog.
I’ll address your longer comment in another comment.
29. April 2014 at 06:36
Steve, I’ll just make some general comments on your longer piece. Let’s start with the change of views of Stiglitz and many other progressives. Here’s where I object:
1. I don’t begrudge them the right to look more closely at second order effects and reach the conclusion that a consumption tax is no longer optimal. There are issues like risk and insurance, or enforcement problems, which can be added to models. That’s OK. But I don’t see the implicit jump to the conclusion that any tax that is not identical between wage and capital income (explicitly) actually “favors capital.” It seems clear to me that if your first pass on the problem is to go with a pure consumption tax. And later you tweak the model to add realism, you’d plausibly end up with a consumption tax plus a modest tax on capital (although of course there are also plausible tweaks that push you to subsidize capital.)
Even putting aside complications like non-indexation for inflation in capital taxes, but full inflation indexation in labor taxes, you have in the US explicit top rates of around 43% on wage income and 25% to 30 % on capital income. Whenever someone claims that system “favors” capital, surely the average listener is assuming that a neutral regime would apply the same rate to wage and capital income. But the Stiglitz who wrote the early consumption tax papers surely knows that’s a cognitive illusion. Even if the second order effects push the optimal capital taxation rate above zero, I see no reason to assume they push it up to 43%. So it still is not clear that our system “favors capital” in that sense.
This is a confusing issue because our debate is sort of toggling back and forth between whether it’s a cognitive illusion that capital income is not taxed (I claim the IRA thought experiment shows it is) and the separate question of what sort of tax regime is optimal. So when we discuss “benchmarks” it’s not quite clear what we mean. I’ve discussed optimality, now let me switch to cognitive illusions. I’d say the willingness of many progressives (not just Stiglitz) to embrace progressive consumption taxes at some point in their career (he’s never been a conservative, nor have DeLong or Yglesias) shows that they do understand my cognitive illusion argument at some level, even if they might describe it differently. Imagine them trying to sell the earlier argument to a room full of progressives who don’t understand the counterintuitive nature of economics in general, and tax theory in particular.
That’s essentially what I’m try to do here. I want progressives to first understand (and I mean deeply understand and internalize) Stiglitz’s earlier argument. If they later reject that policy proposal for the pragmatic reasons you discuss, so be it. But I don’t even see most progressives as having reached that stage. And that’s why I still think Solow is pandering. His “favor” comment simply confirms the prejudices of the vast majority of people (actually both liberals and conservatives) who don’t understand Stiglitz’s 1980s ideas about consumption taxes.
So because I agree that there are alternative models where some capital taxation is optimal, I won’t address all of your long list of arguments on that point. But some cautionary notes:
I think there is a danger that progressives will “latch onto” findings that they like, and ignore other findings. Here’s a quote (withembeded quotes–confusing) from an earlier post discussing Saez and Diamond in the JEP:
“Paul Krugman recently trumpeted the S-D conclusion that the optimal tax rate on high incomes is over 70%: (me)
Using parameters based on the literature, D&S suggest that the optimal tax rate on the highest earners is in the vicinity of 70%. (Paul)
That presumably comes from this statement by S-D: (me)
As an illustration using the different elasticity estimates of Gruber and Saez (2002) for high-income earners mentioned above, the optimal top tax rate using the current taxable income base (and ignoring tax externalities) would be Ï„ * = 1/(1 + 1.5 × 0.57) = 54 percent, while the optimal top tax rate using a broader income base with no deductions would be Ï„ * = 1/(1 + 1.5 × 0.17) = 80 percent. Taking as fixed state and payroll tax rates, such rates correspond to top federal income tax rates equal to 48 and 76 percent, respectively.” (Saez and Diamond)
End quote. So Krugman claims that the top rate is 70%, whereas the top federal rate in the model is actually 48 percent under our current lophole ridden tax system. That’s not far above the actual top rate of 43.4% (another number most progressives get wrong.)
So how do S-D justify the 76% rate as relevant?
“In the current tax system with many tax avoidance opportunities at the higher end, as discussed above, the elasticity e is likely to be higher for top earners than for middle incomes, possibly leading to decreasing marginal tax rates at the top (Gruber and Saez, 2002). However, the natural policy response should be to close tax avoidance opportunities, in which case the assumption of constant elasticities might be a reasonable benchmark.”
That’s right; “assume a can opener”. Just the thing people criticize me for doing. Assume away all public choice problems, and imagine a pure system. But then why not a consumption tax? Why is that not optimal? They do concede the theoretical arguments, but then add this:
“The existence of tax differentials between labor and capital also creates pressure to extend the most favorable tax treatment to a wider set of incomes. For example, in the United States, compensation of private equity and hedge fund managers in the form of a share of profits generated on behalf of clients is considered realized capital gains, although it is conceptually labor income.”
Why not tax hedge fund income as labor income? Presumably they would argue that it is politically difficult. So instead of recommending low taxes on capital, as theory suggests, they point to the danger of loopholes creating adverse side effects. But when it comes to “progressive” proposals like high MTRs, those practical problem are just swept away with the brush of a hand. I find that exasperating.
To conclude, I’m underwhelmed by the arguments of Saez, et al. They are essentially like a lawyer arguing a case in court. They will point to evidence in their favor, and ignore the rest.
Regarding your claim that consumption taxes are unlikely to be sufficiently progressive, tax “burden” comes from lower consumption. If (as you claim) consumption for the superrich is a trivial fraction of income, then it’s almost impossible to transfer resources from them to the rest of us. Money is a veil. If you want the poor to consume more, the rich must consume less. Unless the proposal is to have our poor consume the “seeds” that were going to produce next year’s crops. And I don’t see you making that proposal. What this really means is that a consumption tax would be much less regressive than it “seems,” i.e. more “cognitive illusions.” Indeed I believe that a consumption tax that is equally progressive to our current system would seem far more regressive. You could put a 99% tax on Buffett’s consumption, and it would be less than 1% of his income. But what does that fact really tell us about Buffett?
I’m not convinced that lots of wealth is never consumed. We haven’t yet reached the end of time. In the meantime, I’m happy if the rich don’t consume their wealth, and instead reinvest to boost real wages.
And I’m not convinced some income doesn’t come from deferred wages. Again, in a income tax with unlimited IRAs, that problem doesn’t appear to occur, and yet in the long run it’s identical to a consumption tax. (it’s different at the transition stage, but a one time lump sum wealth tax could address that “problem.”)
I don’t claim that our only goal is to reduce distortions. I think (utilitarian) redistribution is a worthy goal.
I’m clearly a pragmatist in macroeconomics, who puts little weight in “models” (such as DSGE). It’s not so much that I’m trying to rely on models, as I’m trying to get clear thinking–looking beyond cognitive illusions. I see the “model people” as those who tweak the simple consumption models to come up with justifications for income taxes. That’s fine, but that needs more modelling than what I do. I do insist however, that clear thinking requires a focus on consumption as the “ends” of all economic activity. If you start mixing up means and ends in a muddled way, you are likely to end up in a muddle. Obviously the good progressive modelers are able to do it right, but I see much of the debate in the blogosphere pretending like these cogntive illusion problems doesn’t exist. As an analogy, it would be like debating the optimal VAT tax on exports without realizing that the idea that VATs currently “favor” exports is a myth. Once you understand that the simple argument is a myth, then feel free to go on and develop more sophisticated arguments. Bring in all the second order effects.
And finally, does the increasing inequality in recent decades (nationally, not globally) change anything? Maybe, but maybe what it changes is that we should remove European laws forcing billionaires to give their fortune to their children, rather than charity. That we should loosen intellectual property laws. That we should weaken zoning, even though it will make income and wealth more unequal, because it will make consumption more equal.
29. April 2014 at 07:38
Scott,
Thanks for this worthwhile post and the very insightful comments. Let me offer a few thoughts.
1. If consumption is equal or relatively equal, why do we care at all from an equality of aggregate utility point of view if some people are much more wealthy than others.
2. The only things we should care about are: a) have we got the balance between consumption and investment right, and b) are the wealthy (those controlling assets) deploying/investing them well.
3. The current system has some problems (inheritors of wealth are not always the greatest investors) but I have a hard a time seeing any system of taxation on wealth which does not result in a much worse deployment of assets.
4. I like charity (giving money to people who need it), but most charitable giving is simply an exercise in producing goods at a price way above what anyone would actually pay for them and then subsidizing their consumption. It thrives because of the tax codes and the fact that many inheritors of wealth are unable or unwilling to do the hard work of actually deploying their assets in a way that optimally benefits society.
5. You’re wrong about hedge funds.
Thanks again for the post. (I’m busy deploying assets and don’t get as much time to read your blog as I used to. I’m hoping my current efforts will get me into the .0001% so I can spend my time reading your blog and getting my name prominently carved into the wall of some concert hall or above the entrance of some college building.)
30. April 2014 at 05:42
Scott,
Thanks for the long and thoughtful comments. We should probably mostly leave it there and move on. I guess I think a large part of our disagreement is just about the weight we put on models. I think an Atkinson-Stiglitz-ish view of capital as merely a purchase with labor of deferred consumption is interesting and important, but not a remotely sufficient description of what capital is and how it behaves. There are lots of other ways of thinking about capital, lots of other models, lots of effects of capital unrelated to the incentives of a worker choosing between consumption today or consumption tomorrow. So while you characterize as reasonable looking “more closely at second order effects and reach[ing] the conclusion that a consumption tax is no longer optimal”, I (and I think most other “progressives”) are looking at entirely different models, and sometimes summarizing observed dynamics in ways that wouldn’t qualify as a “model” under academic conventions. I do have some alternative models, under which high levels of capital taxation would plainly be optimal. But Piketty’s book, for example, is mostly model free: It simply characterizes the dynamics of capital as matters of history and accounting and what the accounts are likely to look like should history continue to unfold according to certain trends. No highly regarded economics journal would publish that sort of analysis. Yet, from a policy standpoint, there’s little reason to consider it less reliable than the fanciest DSGE model, or Atkinson-Stiglitz, for that matter.
So I still don’t think “progressive” economists are guilty of what you are accusing them of. You suggest that, as members of the economics profession, you and they share a baseline starting point, and you differ only by degrees in which you think frictions on second-order concerns modify that baseline. I think progressive economists like Stiglitz certainly know and understand the baseline that you are coming from, but they reject it as the default or best perspective from which to view the issue. It is not their baseline (as it is certainly not mine, even having done some homework to try to understand where you are coming from). This is not a dispute over second-order modifications, but over first-order frameworks. How should we understand capital, real and financial? Is what is conventionally modeled as “consumption” the only rival good in the economy, exclusion from potential consumption the only externality, such that a consumption tax as “taxing what you are taking out of the economy” is a good way to go?
Those are questions over which, I think, you and I would disagree a great deal. Don’t get me wrong, I don’t want to fade into a friendly we-just-have-different-values relativism: I think we have similar values, and I’d like to do a better job of persuading you of the inadequacy of so straitened a view of the meaning of capital. I think it would be possible to do with recourse to a concept no more exotic or progressive than indirect utility. With diminishing marginal utility, rival consumption, capital which among other things represents an option to consume, and systemic risk in the level of production and wage income, then the distribution of capital may correspond to a distribution of risk to consumption disagreeable on utilitarian grounds. That is true even if we use very conventional forms of utility, and holds much more true if we add realistic notions of habit formation and reference-group comparison to utility functions. On a macro-financial rather than micro-utility basis, there are models of capital formation quite orthogonal to productive use of real capital derived from deferred consumption of labor income, under which capital formation can represent a distorting tax on labor. These are all ideas, if I were better, I should be able to persuade you to take seriously, not based on fuzzy differences in “values” or “politics”. I’ll keep trying.
Thanks for putting so much time and thought into your responses, and sorry again for my being a bit of a dick at the start (probably through the finish) of our conversation.
1. May 2014 at 05:19
dtoh,
1. We don’t care.
3. I oppose wealth taxes.
Please don’t give to colleges. They are already far too rich. Give to a worthy cause.
You may be right about hedge funds, my actual position is “if they are evading taxes, we should do X”
If they are not, then I have no argument with you. I’d have to study the issue.
And thanks for the support.
Steve, Thanks for the very thoughtful comments. I’m too far away from grad school to debate this issue as effectively as I’d like. My general view is that while a pure consumption tax may be a bad baseline, for all the reasons you cite, a pure income tax is a far worse baseline, for all the reasons I focus on. And I think Solow was assuming a pure income tax as being the appropriate “neutral” baseline. I still think that’s based on a cognitive illusions, and will keep trying to convince progressives of that fact (even as I try to convince conservatives of the need for high MTRs on high end consumption.)
1. May 2014 at 10:03
Steve,
1. Can you model higher utility with realistic assumptions?
2. Can you sell it?
3. Can you execute it?
Ultimately, you need to achieve one or more of the following: a) better allocation between work and play, b) better allocation between current and deferred consumption, c) better distribution of consumption, or d) more optimal investment decisions.
Don’t really see how any income/capital tax model can get you there except on paper.
3. May 2014 at 22:07
Scott,
Two points.
1. When people complain about capital income being taxed at an unfairly lower rate than wage income, I think they are assuming that in both cases all the income is spent. Your infinite IRA example (and much else that you wrote) suggests that you believe that all income of any sort that is spent should be taxed at the same rate, i.e., a (progressive?) consumption tax instead of an income tax. Is that right? (I agree with you and Robert Frank on this.)
2. How would you treat purchases of jewelry, art, real estate, a major league sport franchise, and similar items? They are not consumed. So presumably they shouldn’t be taxed as consumption. They often appreciate. So they seem to act like capital investments. Yet owning them brings pleasure to the owner, similar to other purely consumption purchases. Not taxing them seems unfair to those who pay taxes on other items of consumption. If items like these are not taxed, I suspect it could be arranged to live a wonderful life and pay very little tax–because most of what one buys is (correctly) considered an investment. I think most people would feel outraged by this.
3. May 2014 at 22:50
P.S.
In my previous post instead of “a wonderful life” I should have written something like “a life of great luxury.”
Presumably rent would be considered consumption and taxed. (Right?) But what about living in a house that one owns? Would you have someone pay tax on its rental value? Many of the items in my point (2) might be treated this way.
3. May 2014 at 23:00
P.P.S.
If you take my previous suggestion to its logical limit, does that imply that anything that one owns that could be rented out should be taxed at that rental value? That would suggest that bank deposits should be taxed at the current interest rate because they could be loaned. Similarly for stock. It should be taxed at least at the rate charged to short sellers. Stock would probably have a very high rental value when there are controversial items that shareholders vote on. One could sells one’s vote (which would be taxed). If one didn’t one should be taxed on the income one could have received had one sold it.
More generally almost any asset could be loaned/rented. Should holding any asset be taxed at its rental value? That would, in effect, amount to a wealth tax.
3. May 2014 at 23:12
P.P.P.S. Taxes on unused assets might also be justified on the grounds that they are not contributing to the economy. Taxing them encourages putting them to productive use.
4. May 2014 at 11:08
PPPPS. When I suggested taxing something at its rental value I should have said tax something as if the owner receives income at the rental value. It’s the implied rental value income that’s taxed. The tax is not the rental value itself.
7. May 2014 at 12:45
Russ,
Very late to respond, but I have frequently suggested a national progressive sales tax and state/local taxes on non-financial assets being used for private consumption. This would cover homes, boats, diamonds, paintings, etc. and effectively act as a consumption tax on these assets.
7. May 2014 at 20:00
Hi dtoh,
Glad to see we’re on the same page. I was hoping Scott would reply, but so far he hasn’t. Would you mind telling me who you are. I’m pretty sure you’re not the Dance Theater of Harlem.
7. May 2014 at 21:44
Russ,
Scott knows who I am. I’ve been telling him this for the last two years. He’s pretty much on the same page.
7. May 2014 at 21:51
Russ,
Former banker (developed most of the early derivative products at GS). Now I’m mostly an entrepreneur but also sit on and chair some corporate boards. Also dilettante economist wannabe.
7. May 2014 at 23:20
dtoh,
OK. Thanks.
9. May 2014 at 05:24
Telling the truth in area like like is a sure route to electoral loss.
9. May 2014 at 05:35
One thing that I think can be added:
The goal is taxation is to move some consumption from individuals to public purposes. A consumption tax is seems best to do that. As Scott has said before it is nearly impossible to transfer consumption from the Gates and Buffets so you need to consider who consumes less when Government spends or distributes money.
9. May 2014 at 05:45
Sorry about the multiple post but…
BTW As a person who earned in the lower 20% until I was 29, it seems unfair to me that the cap on the IRA, Simple IRA and 401k is on annual contributions rather than on total value if the account.
Also BTW It seems quite perverse to me that the caps are different for the 3 types of accounts (IRA, Simple IRA and 401k).
16. May 2014 at 05:36
Mark Sadowski:
I don’t like spending time with discussions on economics blogs, but after all the non-sense I have come across in the last one month, I believe this has to be written. However, I will not waste my time to discuss further with charlatans who dare to call me a conspiracy theorist on their blogs; that is why I did not respond to some of those posts any more. A more fruitful discussion took place on Simon Wren-Lewis’ blog, I put the link below.
Some people just don’t get it. Why should I teach for instance the Diamond model in a module which is called “Alternative Theories of Economic Crisis?” What is alternative about the Diamond model? Or Bernanke-Gertler model? Or Solow? Really? Solow, or RBC models without money in a bubbles module? Is this a joke?
I find it fascinating that these answers to my post still find support. I said it very clearly before, and I will say it once again: You can teach anything you want in your own bubbles module. If you think that the students have a lot to learn from all those models you cited above, go on, teach them. It is up to me what I teach in my own module, what I include and what I don’t. Especially when the module is called “Alternative Theories”. It looks like some people just don’t understand what the word “alternative” means. I have never claimed that my bubbles module must be the only bubbles module taught in the university, it’s those like you and your friends who have that opinion for your mainstream modules. Did you even see how many mainstream papers there are in my reading list such as Brunnermeier, Cechetti, Bernanke, Goodhart? How many of my reading list are Central Bank publications? Did you see I said twice that this module was kept free of maths purposefully?
Here is what is funny: I have been accused of not knowing the core mainstream literature about the topic by people who have heard more than half of the papers in my reading list for the first time in their lives themselves and have no idea about what the papers say! In fact, it looks like it is those people who will have to read a whole library before they can discuss these topics, not me. Please read, and come up with arguments showing why they are wrong or “optional”. Otherwise, I will not take these posts seriously and write a response again. If your argument will be the stereotype response “oh, those papers are not proper models, they are not microfounded”, please read all the discussion here before you write because I don’t want to repeat myself.
http://mainlymacro.blogspot.co.uk/2014/04/when-economics-students-rebel.html?showComment=1398477861481#c6165295499742843574
And here is one of your very intelligent mainstream colleagues I respect a lot, because he actually understands what the mainstream is doing:
http://noahpinionblog.blogspot.co.uk/2012/03/why-bother-with-microfoundations.html
In my opinion, Cladio Borio or Andy Haldane both have a lot more to say about this topic than you or any of those people you cited above because unlike your fellow mainstream friends who have been falsified so many times but still have the face to preach us how to do macroeconomics, these people’s jobs depend on identifying/fighting bubbles. I am really fed up with posting papers for people to read because I know no one reads them, but I will try once again: http://www.bis.org/publ/work395.htm
“Essential features that require modelling
“The first feature is that the financial boom should not just precede the bust but cause it. The boom sows the seeds of the subsequent bust, as a result of the vulnerabilities that build up during this phase. This perspective is closer to the prewar prevailing view of business fluctuations, seen as the result of endogenous forces that perpetuate (irregular) cycles. It is harder to reconcile with today’s dominant view of business fluctuations, harking back to Frisch (1933), which sees them as the result of random exogenous shocks transmitted to the economy by propagation mechanisms inherent in the economic structure (Borio et al (2001)). And it is especially hard to reconcile with the approaches grafted on the real business- cycle tradition, in which in the absence of persistent shocks the economy rapidly returns to steady state”
“The second feature is the presence of debt and capital stock overhangs (disequilibrium excess stocks)”
How could this be done?
“One step would be to move away from model-consistent (“rational”) expectations” “Heterogeneous and fundamentally incomplete knowledge is a core characteristic of economic processes.”
“A third, arguably more fundamental, step would be to capture more deeply the monetary nature of our economies. As discussed in more detail in the next sub-section, models should deal with true monetary economies, not with real economies treated as monetary ones, as is sometimes the case”
“And in all probability, this will require us to move away from the heavy focus on equilibrium concepts and methods to analyse business fluctuations and to rediscover the merits of disequilibrium analysis, such as that stressed by Wicksell (1898)”
In summary: No equilibrium modelling, no exogenous shock modelling, no representative agent/firm/banks, no rational expectations, no complete knowledge, no non-monetary economies but in the true sense. So, Borio is implicitly saying there is no model in the mainstream that can explain business cycles. This is what I believe, and that is what I teach. I am well aware that after the crisis, there are many attempts that try to incorporate money and financial markets in mainstream models, I am sure I’ve read more of those papers than you have, but the methodology is still the same and wrong in my opinion. Therefore, I will not teach representative agent-equilibrium-exogenous shock models which I believe have nothing to say about this topic. If you want to do it yourself in your module, by all means, but you or any mainstream economist are not in a position to tell me what I can/cannot teach after the catastrophic failure of mainstream. The mainstream is not the only school which has important things to say, there are others, and they are as scientific as the mainstream if not more. (Once again, please refer to my post on Simon’s blog)
Further, if you had read the Post-Keynesian literature ONCE in your lifetime, you would know that what Bank of England (and BIS) says here in 2014
http://www.bankofengland.co.uk/publications/Pages/quarterlybulletin/2014/qb14q1prereleasearticlemoney.aspx
was said 26 years ago in 1988 here, http://ideas.repec.org/a/mes/postke/v13y1991i3p404-413.html
and endless times since then by some of those authors in my reading list (such as Bellafiore, Fullwiler or Wray who in your opinion are less credible than Solow(!)), while those papers you cited were claiming that banks are mere intermediaries, and money/debt does not matter. So they are not “optional reading” like you claim, they are ESSENTIAL reading, some of them are seminal papers on this topic. Those are the people I take seriously, not the supposed “core knowledge” that does not even understand the monetary nature of capitalist economies until the entire financial system almost collapses or Bank of England/BIS tells them so. So I recommend that before you attempt to judge my knowledge of “core theory”, you question your own knowledge of macroeconomics and methodology. In fact, you can try to write a paper to respond to Borio about financial cycle-macro modelling for instance and convince us that he’s wrong, I will be happy to read it.
I apologize for such a long post, I just saw the comments here.
Dr. Sakir Devrim Yilmaz