Monday, January 3, 2011

Mankiw and Krugman

Mankiw's piece in the Sunday NYT and Krugman's column from this morning's NYT provide an interesting contrast.

Mankiw does a nice job of making the case for the conservative cause, though I'm not sure if there are any powerful people in the Republican party who actually embody this cool and rational approach. In any event, Mankiw's piece is a plea to the President to understand economic conservatives as standing for the following principles:

1. Focus on long-run issues: Milton Friedman was a strong proponent of this approach. When pressed for his views on short-run macroeconomics, Friedman's framework looked like an IS-LM model, but he was firmly against the use of short-run stabilization policy. According to Friedman, given policy lags and our lack of knowledge about how policy works, we would be better off not intervening to smooth business cycle fluctuations.

2. Incentives are important: The tax system and social insurance have to be designed with incentives in mind. For example, high marginal income tax rates discourage work. Note that this applies to the rich as well as the poor, if not more so.

3. Forget about income redistribution: Essentially all changes in fiscal policy have implications for the distribution of income and wealth. What I think Mankiw means here is that fiscal policy should not be designed specifically to redistribute income. For example, some objections to the recent tax bill from Democrats were related solely to the fact that the bill would retain the top marginal tax rate where it has been since the Bush tax cuts, instead of increasing it.

4. But inequality is a serious problem. We need to improve our educational system: Mankiw is not specific here. It's not clear if what he has in mind is an expansion of public education, or incentives for private education.

5. The demonization of conservatives won't get anything done.

Now, Krugman's column from today is not about these conservative principles, of course.

1. Focus on the short run: Mankiw, who certainly appears to be very much a short-run Keynesian (he of the Sticky Price Manifesto) seems to think that we have done enough stimulus, and it is time to move on. However, Krugman is very concerned, most of all about what he thinks is unwarranted optimism.

2. Don't worry about incentives.

3. Income redistribution is the main goal: Krugman states:
Jobs, not G.D.P. numbers, are what matter to American families.
Krugman focuses almost exclusively on the unemployment rate as a welfare measure. What he seems to have in mind is that, in any recession the dispersion in income across the population increases, and this effect is even larger in this recession. In the aggregate, we of course care about GDP numbers, but Krugman cares a lot about the distribution of income, and he wants to change it through the tax system and government expenditures on goods and services.

4. Public education? I think if you asked Krugman, he would come out in support of more resources for public education. In line with his focus on the short run, he does not mention that here.

5. Demonization: Krugman is of course well-known for this. You know that eventually he'll get around to "this Republican is a jerk," or "that Republican is a dope." This time out, he's a little milder than usual, but comes up with:
Realistically, the best we can hope for from fiscal policy is that Washington doesn’t actively undermine the recovery. Beware, in particular, the Ides of March: by then, the federal government will probably have hit its debt limit and the G.O.P. will try to force President Obama into economically harmful spending cuts.
Before we see it, we know that nothing good can come from Republican economic policy.

What are we to make of this? To me, Mankiw is making some sense, but I don't see any long-run vision coming from either political party, and Krugman isn't helping.


  1. "For example, high marginal income tax rates discourage work."

    Of all the canards I hear from the high marginals, this has absolutely GOT to be the stupidest.

    Let's say I make $200K net. My marginal rate goes up, and now I'm only making $190K. According to your statement, I'm going to do what? Work less? But if I do that, I'll only be making $180K. What am I doing then? Punishing the person who raised my taxes by cutting my own pay further? That's rediculous.

    The fact of the matter is that I wanted (or needed) my $200 back after my marginal went up, I WOULD WORK EVEN HARDER to get it back.

    So will you people stop making yourself sound stupid. Higher marginal tax rates may not be what you want for yourself, but they will NOT make you work less. if anything, they will make you work MORE.

    Stop the crocadile tears. Pay your fair share.

  2. Anonymous above is largely right.

    Mankiw is famous for intentionally misleading; in his position he's known full well for a long long time about the income and substitution effects, the backward bending labor supply curve, the empirics that it bends backward at about the upper middle class level, the fact that the wage per hour skyrocketed over the last hundred years, but people didn't work more hours in response; they worked less; people 100 years ago with their paltry hourly wages didn't work paltry hours in response to this "incentive"; they worked more hours.

    Mankiw knows all this, but as usual he intentionally misleads for the right wing cause.

    Krugman does care about the long run and would love to see a great increase in high return long run investments, like education, infrastructure, alternative energy, etc. He has many posts saying this, but this was a short column focused only on the current cyclical slump.

  3. Anonymous,

    I guess you haven't taken Econ 101. For an individual who can choose hours of work, there are two effects - income and substitution effects. When your after-tax wage goes down, the substitution effect makes you want to work less, and the income effect makes you want to work more. Theory tells you the net effect could go either way. Empirical evidence, while mixed, seems to suggest that the net effect, while negative, is small. However, that's not the end of the story, as this is only about the intensive margin - the choices of an employed individual. There is also the extensive margin, where empirical evidence suggests the effects are large. Facing high marginal tax rates, individuals can choose not to work at all, they can choose to retire early, they can choose to work in another country. Ask Keith Richards what "Exile on Main Street" was about.

  4. Anonymous,

    I guess you haven't taken Econ 101.

    I'm afraid that our Richard H. Serlin hasn't either... Or, at least, he hasn't read the empirical literature very carefully...

  5. I've taken Econ 101 (and got BA in economics). i have few questions/points:
    (1) What are the cross-elasticities which effect the substitution effect? How do they change with different income level? I presume that someone whose hourly pay went down from $10 an hour to $8.80 will work more hours (if available) to maintain the necessities. Is it different for someone making $50 an hour? $300 an hour? Does substitution effect even matter for someone making %500k versus $490k a year? Or $5m versus $4.90m a year? Will a neurosurgeon perform less surgeries a year because he gets paid 10% less per surgery? Will professional violinist play less shows?
    (2) Are work hours really a choice? I cannot think of most people (even in high income brackets, like big firm lawyers or doctors) picking their hours. You do what you gotta do.
    (3) You say empirical evidence is inconclusive. Can you recommend reading (like papers/books) reviewing the empirical evidence and analyzing it? How does it tie with policy?
    (4) To me, substition effect sounds very much like argument that reducing taxes increases revenue - it might be true if a lot of specific conditions are met (right amount of people being at right elasticity points etc.) but have not actually been seen in real life.

    Just some thoughts.

    With Best Regards,

    Aleks, D.C.

  6. P.S. I am not the same anonymous who posted the original post (and to whom Professor Williamson responds to). I apologize for any confusion.

    Aleks D.C.

  7. Also, how do substition effect apply to salaried people?

    Aleks, D.C. (credit for this question goes to my wife, blame goes to me)

  8. Aleks,

    1. If you are interested in specific elasticities, I would start looking in the "Handbook of Labor Economics." Individual labor supply behavior has been studied extensively by labor economists. In general, you are thinking about the right things. How much choice do particular individuals actually have over hours of work? Maintaining subsistence levels of consumption has to be important. I might be a bad example, as I have a lot of choice, but I would not have written a textbook if the royalty income was taxed at, say, 80%.

    2. But I might have the choice of working part-time at one firm, vs. full-time at another firm. Some people hold more than one job. I know firefighters who do part-time construction work on the side in flexible-hours arrangements.

    3. As macroeconomists, we are interested in aggregate labor supply elasticities. While individual wage elasticities may be small (for example because, as you point out, many individuals don't have much choice about hours of work), aggregate elasticities can be large, due to choices at the extensive margin. This relates to choices (as I mentioned above) about whether to work or not (e.g. choosing work in the home over work in the market, or choosing the timing of retiremment). Richard Rogerson knows a lot about this. I would recommend reading some of his papers. I can't give you particular references, as he is in the process of moving to Princeton, and his web site appears to be down.

    4. Sure, that was actually at the core of what the "supply side" people were talking about in the early 1980s. When you cut tax rates, if you had a large enough aggregate labor supply substitution effect, tax revenue could actually go up. You won't find many economists who think the empirical elasticity is large enough that this can happen. Reagan and Bush (W) both cut marginal tax rates, and tax revenue went down in both instances.

  9. Laffer seems to be correct when it comes to households in the top income bracket, say top 5% and top 1%. Perhaps not so much because of a high labor supply elasticity, but because of their ability to hire tax lawyers, move to another state, and change the timing of thier income.


  10. KP,

    This would be part of the argument of the "broaden the base, reduce marginal tax rates," people. The argument is that you simplify the tax code, eliminate deductions, and make the system more fair, without reducing total tax revenue. The problem is that the rich people who can hire tax lawyers can also hire lobbyists to get themselves new loopholes.

  11. Ok, let's get the opinion of a top economist in this area who knows the litterature up, down, and sideways, MIT economist Jonathan Gruber, from his 2007 textbook, "Public Finance and Public Policy", 2nd edition:

    Changes in tax rates appear to have relatively modest effects on total gross income; the total amount of income actually generated through work or savings does not respond in a sizable way to taxation.

    – Page 734

  12. I'll add if some of these greedy bastards want to move to Honduras to get even more millions and billions, in many cases the country is better without them. I wish the Koch brothers would leave the country -- we would be immensely better off.

    But ask yourself, why do so many of the rich live in New York and California and so few in Mississippi? If you only think in partial equilibrium, which Stephen often expresses a distaste for, then they should all live in some low tax southern state. But in general equilibrium we see that so many live in high tax states like California and New York for what they get in return -- vastly better universities, amenities, infrastructure, culture, hospitals, police, and many other public goods.

  13. Incentives matter, Richard. That's the whole foundation of our profession.

  14. I don't think incentives are the WHOLE foundation of economics. To me a huge essence is efficiency – getting the most total NET utility for the inputs, and really trying to maximize total NET societal utils, and growth, which is really a story of investment, especially public good idea/understanding/ science investment.

    Incentive can be complicated. You increase wage per hour and there's a substitution incentive to work harder, but it's well known in economics there's also an income incentive to work less hard – but there's much more, there's the level of consumption you've grown accustomed to; there's how people's utility functions adjust to society, and the level of wealth in society, as that changes decade by decade.

    But if all it was was a simple story, like Mankiw would intentionally deceive us into believing, of the higher the after tax wage per hour, the more incentive to work more hours, so the more hours people will work, then we would see that people worked hardly any hours at all 100 years ago for their peanuts after tax wages, and they work vastly more hours today for their massively larger after tax wages. But in fact, the opposite is true; people worked more hours 100 years ago.

    In any case, a top economist in the area summed up the literature above, and it's that, "the total amount of income actually generated through work or savings does not respond in a sizable way to taxation."

  15. "But in fact, the opposite is true; people worked more hours 100 years ago."

    1. A lot of things have changed other than after tax real wages in the last 100 years.
    2. I think you mean that the average worker worked more hours 100 years ago. (i) We don't measure hours of work for everyone. This is true for manufacturing workers, but what about everyone else? (ii) You have not been paying attention to what I have been saying about the extensive margin. More people are actually working than was the case 100 years ago. Particularly post-WWII, labor force participation has increased, driven by the behavior of women - that's arguably driven mainly by market wages.

    Why you think Mankiw is deceptive is beyond me. I don't agree with everything he says, but the guy is quite straightforward, particularly as compared to your hero Krugman, the King of Deception.

  16. Well, I won't get into cases where Krugman may or may not have been misleading, but what I like about Krugman is that it's clear from his track record that either way, what he says is for good and selfless causes, and ones I agree with.

    With regard to extensive margin:

    1) Gruber's summation is overall, considering all margins.

    2) Again, it's not simple. Women entering the workforce in far greater numbers certainly was influenced greatly by i) Big cultural changes -- 100 years ago they couldn't even vote ii) Changes in technology that freed up a lot more time to work outside the home. Having to maintain a home with no washing machine, disposable diapers, baby formula, day care, take-out,...took a huge amount of time, leaving little left for an outside job iii) The tax increases that Democrats would really like are for very high income, where even with a big increase, the money made outside the home would still be far greater than the money saved from working as a full-time homemaker.

  17. 1. You should go back and read Adam Smith. The discussion of the invisible hand is a gem. To paraphrase, he says something like: Greed is actually a good thing. In fact, people who claim to be selfless and devoted to the community are often doing less for us than the greedy bankers, or whoever. I tell my classes that the philanthropy of Bill Gates, while admirable, has done much less for the world than the creation of Microsoft, which made computer technology accessible to billions of people. In Krugman's case, the problem is that he sets himself up as a selfless defender of the poor, but he's actually a self-centred guy advocating policies that will do harm to society.

    2. Yes, the home production technology is very important. See this paper:

  18. Richard:

    Which paper are you talking about? Gruber and Saez (2002 J Pub Econ) say that high income households (>100K income) respond strongly to changes in tax rates. Also, they don't consider the extensive margin.

  19. Here's the Gruber quote, from his 2007 textbook, "Public Finance and Public Policy", 2nd edition, page 734:

    Changes in tax rates appear to have relatively modest effects on total gross income; the total amount of income actually generated through work or savings does not respond in a sizable way to taxation.

    This is an OVERALL statement about the effects of taxation in sum.

    I'm well aware of Adam Smith's invisible hand; I'm obviously not advocating communism. I'm also well aware that Smith gave many caveats, even with the primitiveness of the field in the 18th century, about things which would make the invisible hand not work. He also strongly advocated taxation to pay for high return government investments like in education, infrastructure, and basic research, things he knew had free rider and other externality problems if left to the pure free market.

    When Bill Gates started Microsoft the top marginal tax rate was 70%. It didn't stop him. He didn't say, this venture will only make me $25 billion instead of 50, it's not worth the effort.

    A big reason why you need good empirical work is because incentives are complicated. It's obviously not just, the more after tax pay per hour, the more hours people work, or people would have hardly worked at all 100 years ago, instead of working more hours than today. Investment bankers wouldn't have worked to the limits of human capacity in the 1990s for hourly after tax pay a fraction of what they make now – which brings us to the obvious problem of, if more after tax pay always makes people work more hours in response to this "incentive", then sometime over the last couple thousand years mankind would have reached either the limit of human endurance, or extremely low marginal effect from increased or decreased after-tax pay per hour.

    Clearly, there are big psychological/positional/prestige things going on. If your marginal tax rate is 70% (like the top rate was, or higher, for most of the 20th century), and your pre-tax income is $1,000/hour, this still allows you to drive a Ferrari from working long hours, when your peer group is only driving Porsches. It allows you to have one of the biggest houses among your peer group, and attract the most desirable women, because relative to your competitors you make the most, or close to. These positional/context/prestige incentives are huge, and it's very little about the intrinsic utility difference between a Ferrari and a Porsche, or even a Corvette and a Porsche.

    Psychologically, people will certainly behave differently if they make $100/hour, but it comes from $125/hour in wages taxed at 20%, then if it comes from $1,000/hour in wages taxed at 90%. It's the same $100/hour take home, and in a typical model with a typical utility function that's all that matters, but not with real human beings. So it's a reason to be concerned about extreme rates, and it makes hidden taxes like the VAT more attractive (the lack of progressiveness of a VAT can be offset by using the money raised progressively, like with free universal pre-school, nutritious free school breakfasts and lunches, and free undergraduate college).

  20. But for the poor, especially the future poor, it's investment that's crucial, especially investment in the advancement of science and medicine, and this is extremely high free rider, externality, zero marginal cost, high transactions cost stuff, that, as a result, is usually provided far more efficiently by the government, disceminated freely and easily over the internet, through taxation. Even if the rich work a little less hard, so we have less mansions and yachts and $100,000 watches, the gain to the poor and society vastly outweighs this by a huge increase in scientific and medical research and infrastructure and education.

    As I wrote in a recent post ( ), "Today total government spending on basic scientific and medical research is approximately $34 billion per year. That includes all of the government spending on basic scientific and medical research on curing cancer, arthritis, backaches, headaches, obesity, robots building robots, solar power, everything." So for just the tiny increase in taxes for the rich from not extending the Bush cuts for the top, we could TRIPLE, TRIPLE government scientific and medical research. Tell me that any supposed decrease in the work hours of the rich would outweigh that benefit for the poor and society in general.

  21. Richard-

    Look over the J Pub Econ paper he wrote with Saez. From the abstract:

    We estimate that this overall elasticity [of taxable income] is primarily due to a very elastic response of taxable income for taxpayers who have incomes above $100 000 per year, who have an elasticity of 0.57, while for those with incomes below $100 000 per year the elasticity is less than one-third as large. Moreover, high income taxpayers who itemize are particularly responsive to taxation. Our estimates suggest that optimal tax structures may feature tightly targeted transfers to lower income taxpayers and a flat or even declining marginal rate structure for middle and high income taxpayers.

    To reiterate, the overall response is small, but the response from high income households is quite large.

    Also, you seem a bit exuberant about the returns to basic research. All of the things that you mentioned (e.g. improving the treatment of cancer) have high value to individuals, and therefore there is strong incentive for private individuals and firms to innovate along these lines in order to earn profits. Most of these magnificently high return projects (ex-ante) have likely been done or are being done.

  22. While Mankiw was Bush's chief economic advisor, Bush pushed through massive tax cuts. Now Mankiw says conservatives care about the long term. Do you really not see the hypocricy?