Greg Mankiw’s marginal tax rate

Without any taxes, accepting that editor’s assignment would have yielded my children an extra $10,000. With taxes, it yields only $1,000. In effect, once the entire tax system is taken into account, my family’s marginal tax rate is about 90 percent. Is it any wonder that I turn down most of the money-making opportunities I am offered?

By contrast, without the tax increases advocated by the Obama administration, the numbers would look quite different. I would face a lower income tax rate, a lower Medicare tax rate, and no deduction phaseout or estate tax. Taking that writing assignment would yield my kids about $2,000. I would have twice the incentive to keep working.

The full column is here.  Perhaps Greg could do more to (legally) evade the estate tax, but that's hardly an argument for raising such a tax.

Alternatively, Dan Ariely asks:

Let’s say that your 1040 came with a little extra stuff: maybe a container with an alcohol content, or perchance something of the chocolate persuasion. What if your tax forms arrived in a gift box with some financial documents on the side? What if the instructions for filling out the form told you to type in your personal information and take a bite of chocolate, type in your W-2 information and drink some of the alcohol, add your deductions and try some of the nuts etc? What if we could live in a world where you actually looked forward filling out these forms?


When I read Mankiw's piece about the marginal tax rate increas due to the expiraction of Bush tax cuts for accepting a $1000 check for writing an article,

I just wanted to send $30 dollars for him to stop crying.

Because that is what the marginal federal change is over existing law: $1,000*.03=$30

But, then I thought, no one would be foolish enough to be confused with this piece. Not economists, not NYT readers

Nah, everyone is smart enough to figure this out. If they are not, they need to go back and review their differential calculus--when you hold everything else constant, and only one thing changes, and you look only at the change in federal law, you wouldn't be fooled.

Would you.

I was going to make a joke about Mankiw's wife ultimately making him stop blogging but incredibly Bill in all earnestness beats me to it.

Mankiw even succumbs to the stupidity by saying "I have been thinking — narcissistically, to be sure — about how higher taxes would affect me."

That's not narcisism. That's the definition of rationalism.


"First, I have to acknowledge that the Democrats are right about one thing: I can afford to pay more in taxes. My income is not in the same league as superstar actors and hedge fund managers, but I have been very lucky nonetheless."

The point is that Mankiw and other professors are much the same as actors (perhaps not Hedge fund managers) where it counts. They think extra money is just more houses and boats. Mankiw and Matt Damon can only really buy bigger houses, take more vacations, eat fancier dinners, and the like. They cannot really invest in things that provide jobs and scalable value to the rest of us. They don't understand what additional money does for actual business people. These "individual contributors' "multiplier" (to use the cockamamie lingo), in other words, sucks. And they don't get it.

I understand why they don't get that part. It's because they are not objective. They think everyone is like a professor or actor. (or they, like mulp think a business of expanding scale is a bad thing). But what I don't understand is why DeLong and Krugman and their...followers...think algebra equates to "crying." Is it because they make statements like "everyone is smart enough to figure this your differential calculus."? I think they think everyone is as intellectual as they think they themselves to be.

Oh, and Greg. The government hardly redistributes wealth. It destroys it.

Mankiw doesn't need the money - he's already wealthy enough. If he took the work he would simply save the money rather than spending it.

Because any rational high-earner will refuse the work, the editor will have to find a lower-earner to do the work. Low-earners have a higher marginal propensity to consume - i.e. if you give a poor single mother $1,000 she will spend it on baby food and diapers. This editor might find a young economist to write the article; and that young economist will buy a wedding ring for his fiancee or a new iPhone or the down payment on a new car.

So by taxing high earners, aren't we effectively giving more work to middle-earners and encouraging people to spend more?

by taxing high earners, aren't we effectively giving more work to middle-earners and encouraging people to spend more?

And effectively decreasing the organizational efficiency of society as we make do with the 2nd, 3rd, or 4th best choices.

I'm puzzled. Mankiw's incentive to work does not solely depend on the dynastic wealth he will leave behind, does it? Is there any empirical work on estate tax incentive.

The problem with Mankiw's argument is that the periods of fastest real economic growth since the NIPA tables have been kept have occurred when Federal Marginal top tax rates have been sixty or seventy percent or up. Think 1933 to 1940, the World War 2 years, 1961 to the 1964 tax cuts and from the 1964 tax cuts to 1968. In each of those cases, tax rates were probably well over twice the rate Mankiw would approve, but growth rates were also quite a bit higher during those periods than during the periods that had the sort of tax policy Mankiw wants to see.

Assuming that taxes affect economic growth (and corporate profits), they usually do it in a way very different than Mankiw is implying.


Medicare is going down the tubes and there is simply no way that Mankiw will be collecting it.

A lot of high income earners with the ability to modulate how much they work think this way. Anyone who reads the piece thinking he is whining or needs the world's smallest violin playing for him is totally missing his two main points, at least as I see them

First, Mankiw has written before that essentially his marginal time is either spent with his kids, or doing something that generates money that will eventually be left to his kids. So, for him anyway, the tradeoff is time with kids vs money for kids.

Second, earning money trades a good or service for that money; a good or service that presumably others value. Regardless of how much you may or may not value his particular services, he pointed out that there are plenty of people whose services you do, or would, enjoy or need that at higher tax rates will increasingly be unavailable. The incidence of taxation does not always land where the politicians say it does.


It's interesting that you can't even read what Mankiw writes without assuming people who are too smart to be fooled by it are fooled by it and that he's trying to fool people when in fact he explains in great detail everything you think he's trying to fool people with.

Including the estate tax in the marginal tax rate!? That is not analysis, that's propaganda. (Though as Bill rightly pointed out, it's not exactly ineffective as propaganda.) As is the whole nonsense about 'money for his children'. Even if he is sincere about that being the only good he wants to purchase, it's hardly intellectually honest.

Imagine the havoc reasoning like that would wreak if it were more widespread. Just think what kind of marginal tax rate I might be facing when I spend my income increase on cigarettes in New York State, or on speeding tickets for that matter.

And quite apart from that I find it hard to believe that somebody who is making the kind of money that Mankiw is making is working (at the margin) for the utility benefits of the money he receives. He is mainly working for the status benefits that the work brings with it, which of course will at least in part derive from gross payment.

"Including the estate tax in the marginal tax rate!?"

If you are thinking about what you can leave to your kids, that's exactly what it is?

By the way where is our esteemed government at with our estate tax? It would be nice to know that they are on the ball while we do our estate planning to make us feel good about the money we send to D.C.

Mei, Responding to your comment: "It's interesting that you can't even read what Mankiw writes without assuming people who are too smart to be fooled by it are fooled by it and that he's trying to fool people when in fact he explains in great detail everything you think he's trying to fool people with."

I do not think people are as smart as you are.

And, that what he says is public and all laid out for others to see is not an answer for a claim of deception: pickpockets pick your pocket in public, and, if you look carefully, you will see them do it. But, no one "sees" them do it.

He's right and wrong. It is interesting to see how he computes the money he makes now through the various and sundry loops/taxes it will encounter...and wrong in that he maybe, just maybe, wanted to mention the current tax rates and how they 'totally incentivise' him to must be massively increasing the marginal tax rate....oh wait...


Mei point is that Greg Mankiw does not strike me as someone who is intentionally deceptive. Neither does DeLong. Krugman does, but that's another story.

Mankiw makes a simple point for which he has to write 1500 words and I appreciate being reminded of it. If you really want to consider a cost, you have to consider the whole cost. Of course, most people don't have the wherewithal to do that and get the kids to soccer practice (just as we don't have time to set up SEPs or S-corps, etc.). But if we aspire to rationality, we should try.

Since we're including all sorts of returns on investment etc, why not include a return on investment of the solvency of the federal government and of roads and airports and so forth. I think Mankiw will then find that overall, it aint as bad a deal as he portrays.

I'd like to see more comment on Tomasz's point about Mankiw's assumed "8% risk-free nominal rate of return." Is this assumption as ridiculous as Tomasz implies? If so, it would certainly seem to undermine Mankiw's argument.

I think his children can earn their own income. Hasn't he heard of Buffet and Stanley and Danko over how demotivating inherited wealth is? We are doing them a favor. Yes, he could just gift it to them while he is alive and avoid estate taxes, but usually people like him want to exert control over them and what he really objects to is that lessened control.

Yancey, We are getting a bit off topic to Mankiw's marginal $30 loss, but I am willing to confront your statement: "Either Medicare is funded, or it is not. Which is it? Either Mankiw and his progeny pay his medical bills personally and through taxes, or they don't."

I am willing to confront it because it also has that ring of unclear thinking that people use to guide others with labels.

So, here goes:

Before we go there, let's ask a simple question: Has IBM currently "funded" its future bond obligations? No. Why not. Well, IBM is going to continue as a business, earning income and using part of that to pay for its bond obligation. Do we say: Oh, my goodness, IBM is insolvent because it CURRENTLY does not have payments for its bond obligation. OMG. That is an unfunded liablity. Quick, take a picture of of the IBM stockholder, he looks like the person in the painting, the Scream. No he's not fooled by that claim of unfunded liability.

Nor should you. But, I know you are not.

So, medicare is a bit different--it has been actually running a surplus (unlike IBM's bond) but it is not any different than the IBM bond obligation that is paid out of future earnings. Is there a demographic bulge in population; yes; and it built a surplus which meant lower taxes, particularly for the higher income levels.

You should be worried, though, about increasing medical expenditures, particularly relative to other countries. Here's a link to today's Economist View Spot the country that stands out from all of all the other countries in the world, and now ask yourself whethere there isn't more that can be done to reduce medicare costs.

And, by the way, if you follow the link to the Uwe Reinhardt article from which this is from, you will read Reinhardt's very pointed criticism of Mankiw's analysis of tort law as being responsible for these costs. But, that is a different story.

As this comment is as well.

Stay focused on Mankiw's $30 deterrence from working hard. I said I would send him $30 to stop whining, now I might also send him Estate Planning for Dummies as well. But, then, he could go to the public library to check it the book, unless it is closed for lack of funding.

Tomasz, not to rain on your parade, but you just used a nuclear bomb on a strawman. Mankiw never claims the 8% is risk-free. Reading comprehension is underrated and underutilized, especially when you are just itching to throw in a coup de grâce zinger about pulling things out of body orifices.

Yancey, Given that the only change that Mankiw is talking about is the change for the Bush tax cuts for the top bracket, do you or do you not agree that the change is $1,000*.03=$30. Everything else doesn't change. So, if yesterday he was willing to write an article for $1000 and today the law changed for the top bracket, the only difference from one day to the next is $30.

Is Mankiw's main incentive for working (at the margin) to leave money to his kids? If not, then his argument is not highly relevant. That said, I do not like inheritance taxes. They should remain zero. Better his kids should get the money than the federal government.


For somebody claiming Mankiw's column as deceptive, you strike me as extraordinarily dishonest yourself. Mankiw makes it quite clear that he is examining the difference caused by a NUMBER of different tax increases proposed/discussed/enacted in the last two years.

One of them is the top-level marginal tax rate (which is a 4.6% increase, assuming Prof. Mankiw's family earns more than ~400k per annum,which is not unreasonable). Others are: the phaseout of deductions/credits, the increase in Medicare tax, the large increase in capital gains/dividend taxes, and the increase in the estate tax.

Matt, OK, what you are saying is that instead of $30, it should be $46 dollars. Sorry.

Mankiw could get even more alarmist numbers if he assumes his children will spend their inheritance smoking and drinking. He could probably get that tax rate up to 97% or so.

Tax the rich, please.

As a designated rich person at $300,000/yr, no action on the tax code for 2011 will cost me an additional $8,900/yr in federal taxes. But raising rate only for the top 2 brackets will cost me just $759/yr.

So tax the rich, please.

Okay, I agree with Mankiw. Leave the rich alone, let's address the deficit through his preferred proposal of a carbon tax instead. Hop to it, Republicans!

Isn't there another way to look at this?

Suppose Mankiw is intent on leaving his children $2000. Then the tax increase forces him to work twice as hard, to leave the desired inheritance.

Indeed, if Mankiw were taxed much more heavily over the course of the career, he wouldn't have the luxury of only working for his children's inheritance. He's accept more work because he needed it to live the lifestyle he wants.

Thus I think the sword has two edges. It's possible to tax so much, that individuals do not think extra work is with the reward. But it's also possible to tax so little, that individuals amass such wealth that they think adding to the pile is not worth the extra work.

I love the way he bases the analysis on "without any taxes. . . assignment would have yielded my children an extra $10,000", as if "no taxes" was ever, or ever will be, an option in a developed nation.

Here's some good news for Greg. Half the households in the U.S. didn't pay ANY federal tax last year. But assuming he did, he can give EACH of his kids $13,000 every year ($26,000 if he's married) and pay no estate tax on those amounts. Better still, those college expense gifts, as well as medical expense payments, are unlimited and aren't included in those exemptions.

He can invest in those stocks and hold them until he passes them on to his children, never paying a capital gains tax on them. His kids will inherit them with a stepped-up basis, so they won't have to pay capital gains taxes on them, either, for the period their wealthy Dad owned them.

Add in the proposed $3.5M estate tax exemption ($7M including the wife) and ole' Greg won't pay a penny of estate tax, either. . . unless he's making a lot more than he claims.

Based on the outrageous prices for college textbooks, however, he might just be.


Here again was your claim:

the upper middle class has benefited from medicare, because they do not have to pay out of their current assets their medical expenses in old age.

If the upper middle class is going to benefit in this way, it means someone else is/was/will be shouldering the burden of the medical care they consume that is covered by Medicare/Medicaid. I only asked you who that someone is, and got no decipherable answer. So, if the upper middle class really benefits from Medicare, as you explicitly claimed, who is shouldering the burden not carried already by the upper middle class in their various federal taxes (the net benefit)?


You asked:

Yancey, Given that the only change that Mankiw is talking about is the change for the Bush tax cuts for the top bracket, do you or do you not agree that the change is $1,000*.03=$30. Everything else doesn't change.

Did you read the same essay I did? Those aren't the "only change[s] that Mankiw is talking about". There was the change in the top bracket that arises from the expiration of the Bush tax cuts (what you are focused on, and, in fact, used the wrong % change to boot in at least 2 comments already); and, though he didn't explicitly outline it, he also addressed the additional effect of the expiration of the tax cuts on investment income by showing the effect on the total return on investing the income from the writing job; he also addressed the proposed/passed tax increases advocated by the Obama Administration since it took office.

Look who has a massive claim on those who create wealth -- the super rich who sit on tax free government bonds and the government workers. Bothof these groups not by acfident have a massive pull on the levers of government power. Those who pruduce goods and services -- i.e. wealth? Not so much.

Again, from Bill:

If you don't believe me, ask yourself this question: who benefits more from medicare: (a) a person with substantial assets which would be depleted by medical expenses, or (b) a person with no assets at all. Answer: (a) and his beneficiaries. Person (b) will not die in the street, but will pay for the protection of Mankiw's assets from medical expenses. But, the medicare stuff is just a ruse, because nothing changed.

Really, aren't you omitting a little detail in ignorance, or in an attempt to dissemble? It really goes back to the very first question I asked you in this thread. If upper middle class people benefit on net from Medicare, who is carrying the additional burden?

Yancey, I'll lend you $1000 at 3.5 percent if you promise to pay me $350 on the basis of how you compute it.

That's a good rate according to the way you use decimals and percentages. In fact, I'll even lend you $1,000 at 2% if you pay me $200.

Manto, Sorry you missed my attempt at irony. I copied Yancey's math error to show the error. The delta in the rates is 4.6 per Yancey if Bush's tax cuts expire on the upper bracket expire. .046*$1,000=$46

By the way, have you wondered why Mankiw let's the $1000 accumulate only to $10k in year 10 when he shows there is a 90% marginal rate , and not year 15 or 20 when he would likely retire or die. It would be totally confiscatory, and you would wonder more about the math. Hell, he might even have to pay the government for having worked. I know, someone will say we do have to pay the government to work and this is all slavery.

At the 90% marginal tax rate he describes, Mankiw must not value his time well versus other alternatives. Strange that an economist would think that way. Leisure must have low value.

I'm sorry, Mankiw's article can only be characterized as grossly misleading. The bulk of it compares the 30-year fate of his $1000 earned today under two scenarios: current tax law including expiration of the Bush tax cuts for marginal incomes about $250,000, or NO taxes.

That makes for a pretty easy comparison, even if we disregard all the other best/worst-case assumptions: that the company he invested in would really yield 8%, but that it would then really pay 35% corporate taxes. Mr Mankiw, what proportion of NY and Nasdaq listed companies pay a real 35% tax rate, every year? Not many, I'll wager -- but you, as a professor of economics who are describing such a hypothetical company in a public argument, are obliged to know! And so on, and so on!

Note also that the estate tax rates he assumes are based not on that $1000 he earned, or even on the growth of that $1000, but on all his other assets and work, which may or may not grow or not grow over the next three decades. (After all, he might die tomorrow!)

At the end, he then pivots and pulls from his magic tax hat a number suggesting that if the $30 increase in his income taxes does not take place, the cumulative effect over 30 years is that his kids might receive $2000 when he dies (on the margin, right Tyler?), rather than $1000. (Presumably, inflation is utterly unfactored or assumed away in all this, as well...)

I'd like to propose a little thought experiment for all those who find Mankiw's argument persuasive:

I'll give you $30, today, or promise to give your kids $1000 in 30 years. (Yes, an 11.75% guaranteed return sounds good...but 30 years is a long time and a lot could change. Your kids might run off and all join a cult that worships aliens and takes all its adherents' money to spend on increasing mercury pollution to make the Earth more hospitable for its future masters, after all...or the Chinese hegemons might decide to lock up all economists and seize their assets...or the dreaded H5N2 flu might cut the world's population by 60%, leading to an era of harmonious peace and ecological recovery. Who knows?)

One can imagine someone quite wealthy, like Mankiw, as accepting this good return as a wager: but is the difference between the two choices enough to stop him from working? Would he turn down a request to write an editorial because only one of the two options was on the table?

Really, Greg?

So poor Greg comes to whine about his taxes going up. And he's intellectually dishonest enough to say it's all for his children. Break out the violins.

First, as someone has already commented, taxing Greg more might not make him work less. If he really cared enough about his children where he thought he positively had to leave them 10 000, then he would simply take another job to pay for the taxes. 2000 pre-tax leaves over 1000 post-tax, which will grow into the 10 000 after the magical 8% per year. Apparently he can do more work, because he's a slacker at Harvard who already turns down jobs (he admits refusing jobs, I guess, because he wants us to think he's just a great guy who doesn't just work for the almighty buck; IMHO it just makes him out to be a slacker who has got tenure). We only have Greg's word that he would work less; given his bias, that's not worth the pixels they're printed on.

Second, isn't Greg a big fan of Ricardian equivalence? Why, is there no equivalence between the governement taxing him in order to reduce its debt, and his desire to leave something to his children? Either he is untaxed and he gets to leave something to this children, then it's just his children who will be taxed to pay for the deficit. It's a wash, not completely (there is, for instance immigration), but some if not largely; in any case, it's as much a wash as Ricardian equivalence is a wash for stimulus spending. (Sure you can make assumptions so that it's not a wash, e.g. assume that the private individual earns 8%/year while the government borrows for less. For this particular assumption, you might want to ask Christine Whitman, ex-NJ governor, how this strategy worked for the NJ pension fund. If you can't ask her, the answer is: not so well.)

Bill, your reading comprehension is terrible. You think Yancey tried to calculate 3.5% of 1000 and failed. You are wrong. Re-read his post.

Mankiw could take the $10,000 and contribute it to Oxfam. Then his children would not get anything at all, so it would be a total loss.

A number of the commenters have claimed that Mankiw is only comparing two scenarios- no taxes vs what will apply after the end of the year. This is a terrible misreading of his essay. Here is the key part:

HERE’S the bottom line: Without any taxes, accepting that editor’s assignment would have yielded my children an extra $10,000. With taxes, it yields only $1,000. In effect, once the entire tax system is taken into account, my family’s marginal tax rate is about 90 percent. Is it any wonder that I turn down most of the money-making opportunities I am offered?

By contrast, without the tax increases advocated by the Obama administration, the numbers would look quite different. I would face a lower income tax rate, a lower Medicare tax rate, and no deduction phaseout or estate tax. Taking that writing assignment would yield my kids about $2,000.

In other words, he is showing you what the effects of Obama-tax-policy vs no Obama-tax-policy, or the difference between $1000 thirty years from now and $2000 thirty years from now. He concludes with:

I would have twice the incentive to keep working.

A number of you have quibbled with his selection of an investment return of 8%, but this is a standard used in a lot of pension accounting, and whatever number he selected is, at best, a guess about the future. If he had selected, let's say, 5% as the base return, then his $1000 grows to $4,300 in the absence of taxes. The adjustments for investment income taxes then reduce that rate by a similar factor, so that his return is 2.5% on $523 compounded for 30 y, or $1100, which, after the take of the estate tax (Mankiw's estimate of the rate that will eventually prevail, I will note) leaves $647 at the end of thirty years at the death of Professor Mankiw. Or, in his way of thinking about it, a marginal total tax rate of 84%, which isn't all that different from the value that arises from his arbitrary selection of an 8% return. The same calculations in his hypothetical using a 2% base return on capital gives a total tax take of 77% when all is said and done.

All Mankiw is doing in this essay is showing the effects of taxation, changes in taxation, and how they relate to how he might make a decision whether or not to produce additional value today. As far as I could tell, only one commenter actually advanced a half-way cogent argument that Mankiw is ignoring the investment returns his children will receive from the taxes collected from their father due to the $1000 earned today, and from that $1000's future value- and this is a reasonable criticism, but would it effect the decision he makes about whether or not to accept the assignment today? I don't know the answer to this question. If every person in Mankiw's position made the decision to not do the additional work, then those returns on government investment never materialize, but if some do and some don't, then Mankiw's children benefit from those that do, while Mankiw can still spend the additional time with his children today.

I would just note that 4.6% was your number

Actually, it wasn't his number. His number was 4.6% plus the increase in medicare tax that appears to be coming down the pike, plus the deduction phase-outs, and so on. There's a lot of stuff hidden in the tax code, and you have to consider it all when computing marginal rates. You seem to be trying extra hard to be obtuse on this point. Then again, when you lead off accusing Mankiw of "crying" (despite his having acknowledged that he can easily afford to pay more taxes), it's not too surprising that you're approaching the discussion with a healthy dollop of ideological bias.

In a larger sense, it doesn't really matter whether the tax Mankiw pays seems large to you. All that really matters is whether it's large enough to affect his decision to take on additional work. Clearly taxes as they stand are enough to discourage Mankiw from taking certain categories of freelance work. If marginal rates were to come down, he would probably take on more such work. If they go up, he will probably take on less. It seems fatuous to suggest otherwise.

Perhaps you don't care too much about how much Mankiw works, if you don't value his product (he acknowledges this possibility as well), but this logic applies to many high-earning professions. I think the best example from the article is dentists (and by extension other doctors). They have a great deal of flexibility in how many hours they work, and tax considerations undoubtedly play a role in that decision.

I'm not saying you have to feel sorry for the rich, or sympathize with them about paying high taxes, or anything else of the sort. But to pretend that raising their marginal rates doesn't affect their decisions is just absurd. And to castigate someone for pointing this out and trying to explain how it works just shows that this was always more about ideology than fact with you.


Well, he did point out that at the end that the expiration of the Bush tax cuts and the additional increases that are part of the current administration's proposed/enacted policies is a difference of a factor of two in the money he leaves his children. Now, some may consider this an unimportant difference in the two scenarios, but one must be up front in claiming this.

To How Bitter:


Has anyone worked out a simple, transparent formula for a graduated expenditure tax? I am thinking of a formula that would define tax brackets in some "non-arbitrary" fashion so as to generated a target revenue? links welcome: luke.lea at thanks

Comments for this post are closed