More on Rising Healthcare Spending

In a previous post, I quoted economic historian Robert Fogel on the income elasticity of healthcare. Fogel's claim of an elasticity substantially greater than one brought this email from MIT's Daron Acemoglu:

Dear Greg:

We noticed your blog on health care and I thought it might be useful to bring my research with my colleague Amy Finkelstein and our PhD student Matt Notowidigdo to your attention.

In this paper, Amy, Matt and I looked at the relationship between income and health care spending. Unlike the results you reference, our findings suggest that rising income cannot explain much of the rising share of GDP devoted to health spending. (In other words, we do not find evidence of an elasticity of health spending with respect to income that is greater than one). We think that the "assumed" relationship that health-care share of GDP should rise automatically as incomes rise is on much shakier grounds than most people realize.Of course, people with different priors will interpret the evidence differently, but we think in this case the evidence is interesting and informative. The paper is here.

Of course, one may ask, if not income, what is responsible for the dramatic rise in the health-care share of GDP. Amy has a very interesting paper on this, which you may have seen, estimating that the spread of health insurance may have played quite a large role in explaining the rise in health spending. So our view has now evolved,as a result of the empirical evidence in these papers, to the tentative conclusion that much of the rise in the health-care share of GDP may be due to policies and regulations related to private and social insurance and the way that the health market is organized (that dreaded word "incentives"). But again I am sure many people will not agree with this conclusion.

In any case, some quick reactions from us, which may or may not be useful to you.

Daron

Thanks, Daron.

Beyond a large income elasticity and the effects of incentives Daron describes, there is a third logical possibility to explain a rising healthcare share of GDP: an expansion in the range of products available to the consumer due to exogenous* technological change. As doctors figure out new and better ways to prolong and enhance life, we may rationally choose to buy these products. It might be tempting to view this effect as a large income elasticity (which is perhaps what Fogel is doing), for the technological change raises real incomes as well as healthcare spending. But the resulting parameter is not a true income elasticity, which measures how much more healthcare we buy if income rises while the range of products is held constant.

--------
*Of course, technological change is not completely exogenous. Surely, the incentives offered by such policies as the patent system and government research funding matter for medical advance. Here what I mean by "exogenous" is not driven primarily by the incentives determined by the health insurance system.

Card on Income and Substitution Effects

In Ec 10 we have been studying income and substitution effects. As a result, this passage from a recent interview in The Region with noted labor economist David Card caught my eye:

Region: As you may know, Ed Prescott has argued that different tax rates on labor in the United States and Europe explain why Europeans work fewer hours than Americans. Do you accept that explanation?

Card: I think that taxes could be part of the story. I would be surprised—given what I think is the credible range of estimates for the elasticity of labor supply—that tax differences are big enough to really explain the whole story.

It is conventional in one school of macroeconomics to assume that the elasticity of labor supply is quite high. And for some purposes that assumption may be correct. In thinking about responses to intertemporal or short-run shocks, for example, it is possible that the relevant elasticity is higher than labor economists have been able to estimate with conventional data and methods. A lot of work in the last 20 years has shown that the actual responsiveness of individuals to short-run fluctuations in wages may be bigger than the conventional estimates from the literature in the 1980s. Nevertheless, for the issue of taxes, we're really concerned about the long-run labor supply elasticity, which includes both the so-called substitution effect, representing the pure price effect of the higher wage, and the income effect. Those two go in opposite directions.

I believe that many labor economists in the United States—starting with H. Gregg Lewis, who was the intellectual father of modern labor economics—would agree with the view that in the long run the income effect dominates the substitution effect, so that over time, as societies become richer, people work a little bit less.

Region: They want more leisure.

Card: Yes, on average. That conclusion would be consistent with the long-run pattern of labor supply in the United States between 1890 and 1990. And in that case, one would normally assume that higher taxes [mean] lower wages and lead to a bit more work. That would have been my starting presumption, to tell you the truth: that the long-run labor supply elasticity is pretty small, and probably negative....

My own view would be that the plausible elasticity is not very big and that therefore the tax explanation won't go too far. But I don't want to get in a fight with Ed Prescott. After all, he's got a Nobel Prize and I don't [laughs].

I am not convinced that David is right. Here is the key question: When evaluating the effects of high taxes in Europe, compared to lower taxes in the United States, do we want to include both income and substitution effects, as David suggests, or just substitution effects?

The answer, I believe, depends on what happens to the tax revenue. If the tax revenue is wasted, then high tax rates are like low wages, and David is right. But suppose, more realistically, that the tax revenue is in effect rebated lump-sum to the taxpayers through a variety of entitlement programs (such as national health insurance). Then we are left with only a substitution effect. In this case, the effect of high taxes on the quantity of labor supplied is larger.

So I am more inclined to agree with Ed Prescott here. But I don't want to get in a fight with David Card. After all, he's got a John Bates Clark award and I don't.

Update: I emailed David to see if he wanted to comment, and he sent me this reply:

Thanks Greg. I am aware of Prescott's argument that the money collected in taxation goes back to the workers so the right analysis is one that ignores the income effect.

I don't think that is the right story myself: what fraction of the extra taxes that european workers pay do you think they view as yielding a rise in net income? My feeling is closer to 0 than 100%, but I can understand the alternative opinion. In that case, of course, the "cost" of bigger government is much smaller.

As my colleague Robert Barro pointed out to me, "Card is more likely to be correct the closer government expenditure is to being useless and vice versa for Prescott. Something of a role reversal."

Frank needs to read more widely

In today's NY Times, Robert Frank says there is little point to cutting marginal tax rates of high-income individuals:

Trickle-down theorists are quick to object that higher taxes would cause top earners to work less and take fewer risks, thereby stifling economic growth. In their familiar rhetorical flourish, they insist that a more progressive tax system would kill the geese that lay the golden eggs. On close examination, however, this claim is supported neither by economic theory nor by empirical evidence.

Apparently, Bob has not read this survey by Stiglitz and come to grips with this theoretical conclusion (from page 35 of the working paper):

Pareto efficient taxation requires that the marginal tax rate on the most able individual should be negative.

The reason for this conclusion is that a negative marginal tax rate on the most skilled worker induces him to work more, and if skilled and unskilled labor are complementary inputs, the wage for unskilled labor rises in general equilibrium.

Nor does it seem that Bob has read this empirical work by Gruber and Saez:

A central tax policy parameter that has recently received much attention, but about which there is substantial uncertainty, is the overall elasticity of taxable income. We provide new estimates of this elasticity...We estimate that this overall elasticity 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....We then derive optimal income tax structures using these elasticities. Our estimates suggest that the optimal system for most redistributional preferences consists of a large demogrant that is rapidly taxed away for low income taxpayers, with lower marginal rates at higher income levels.

Bob is perfectly free to believe whatever he likes and to advocate increasing the top marginal tax rate. But to suggest that there is neither theory nor evidence to support the beneficial effects of lower marginal tax rates on high-income taxpayers indicates a lack of appreciation of the academic literature in public finance.

Bob also makes this argument:

If lower real wages induce people to work shorter hours, then the opposite should be true when real wages increase. According to trickle-down theory, then, the cumulative effect of the last century’s sharp rise in real wages should have been a significant increase in hours worked. In fact, however, the workweek is much shorter now than in 1900.

This seems just wrong to me, if the goal is to analyze tax policy. When comparing work hours today versus a century ago, you have to consider both income and substitution effects of wages on labor supply, which are offsetting to a large degree. But, according to standard theory, the distortionary effect of taxes depends only on the substitution effect. The evidence cited suggests that income effects are larger than substitution effects, not that substitution effects are small.

New Taxes in Massachusetts

Today's Boston Globe reports:

Governor Deval Patrick said yesterday that he had come up with a way to pay for more police officers in Massachusetts: charge convicted criminals a fee. Unveiling his most detailed account yet of his plans for next year's state budget, Patrick said he would propose a "safety fee," which every person convicted of a crime would have to pay.

This fee can be viewed as a Pigovian tax on crime externalities. (It reminds me of the old quip, "A fine is a tax for doing something wrong. A tax is a fine for doing something right.")

Like many Pigovian taxes, this one is being criticized as having bad distribution properties:

Reacting to Patrick's announcement, advocates of prisoners' rights said the plan was unfair. Leslie Walker, executive director of Massachusetts Correctional Legal Services, which represents inmates, said about 85 percent of convicted criminals in Massachusetts earn less than $11,000 a year at the time of their convictions.

Another one of Patrick's proposals is less attractive from the standpoint of economic efficiency:

Patrick also pledged to support legislation that would allow cities and towns to hike the state's 5 percent meals tax to as much as 8 percent.

My guess is that restaurant meals have a large elasticity of demand (meals at home are a good substitute) and a large elasticity of supply (restaurant space can be converted to other business uses, and the labor can be redeployed to other industries and states). With a large elasticity of demand and a large elasticity of supply, a tax on this market would entail a particularly large deadweight loss.

A Guest Post from John Galt

David Leonhardt writes:

There is no question that the wealthy pay a higher overall tax rate than any other group. That is an American tradition. But there is also no question that their tax rates have fallen more than any other group’s over the last three decades. The only reason they are paying more taxes than in the past is that their pretax incomes have risen so rapidly — which hardly seems a great rationale for a further tax cut.

Really? What if the increase in their pretax income is in part attributable to the tax cuts? David seems to be treating pretax income as exogenous to tax policy, whereas there is good reason, both theoretical and empirical, to think that it responds to policy.

So I started wondering: How much of the increase of the the reported incomes of the superrich might be attributable to cuts in their marginal tax rates?  Let's do some very rough calculations to illustrate the possible magnitude of this phenomenon.  I will start my analysis before the first in the series of major tax reductions, which was the famous Kennedy tax cuts.

Over the past half century, the top marginal tax rate has fallen from 91 percent in the 1950s and early 1960s to 35 percent today.  Thus, the amount a person gets to keep at the margin has risen from 9 percent to 65 percent, that is, by a factor of 7.2.  If the elasticity of taxable income with respect to 1-t is one, as some studies find for high-income taxpayers, then the incomes of the rich would have risen by a factor of 7.2 as well.  If the elasticity is one-half, then their incomes would have risen by a factor of 2.7.  In either case, the change in pretax income attributable to the tax cuts is substantial.

By comparison, the incomes of the superrich (top 0.01 percent), as a share of total income, increased by a factor of about 5 over this period.  So, it seems that for plausible elasticities, a significant portion of that increase can potentially be explained by the cuts in the top marginal tax rate.

To be clear, I am not suggesting that we are now on the wrong side of the Laffer curve and that cutting taxes will increase revenue.  And I will be the first to admit that we don't really know the relevant elasticity for the upper tail of the income distribution.  But I am suggesting that it is a mistake to presume that changes in marginal tax rates have little effect on reported pretax incomes.

A Reply from Martin Feldstein

I am happy to lend this space to my Harvard colleague Martin Feldstein. -- Greg

 

Feasibility of the Romney Tax Plan – Reply to Comments

Martin Feldstein

This note is a reply to those who commented on my August 28 WSJ article (available here) about the Romney Tax Plan. The Romney income tax plan includes a 20% cut in all individual tax rates, eliminating the AMT, and eliminating the taxes on interest, dividends and capital gains for those with incomes under $100,000.  The resulting revenue loss is balanced in the plan by broadening the tax base for high-income taxpayers.

The Tax Policy Center (and others citing their report) claimed that the Romney plan is “mathematically impossible” and that the plan would inevitably lead to a large middle class tax increase or a rise in the budget deficit.

I found that that conclusion is not correct. It is possible to cut taxes as Gov. Romney indicates and to finance it with base broadening for taxpayers with AGI over $100,000. Governor Romney has not specified the base broadening that he would propose. My calculations presented here and in the WSJ are not estimates of the Romney plan but an indication that such a plan is feasible.

For the WSJ article I analyzed the most recent published IRS data (for 2009).  The cost of the Romney proposed tax cuts would be $219 billion in that year with no behavioral response (the “static estimate”) or $186 after a $33 billion reduction in cost caused by the behavioral response to lower marginal rates (with an elasticity of the tax base with respect to the net-of-tax share of 0.5.)   Those IRS data also implied that eliminating all deductions for taxpayers with AGI above $100,000 would increase the tax base by $636 billion.  I multiplied the $636 billion by a 30 percent marginal tax rate for the high-income taxpayers, implying $191 billion of extra revenue. That would be enough to finance the $186 billion revenue loss.  All taxpayers with AGI below $100,000 would have tax cuts and no tax increase. I concluded that even without further base broadening the plan is feasible and would not involve either a middle class tax increase or a rise in the budget deficit.

The critics of my WSJ piece raised 4 objections: (1) The 30 percent marginal tax rate is too high for these taxpayers because of the 20% Romney rate reduction. (2) The behavioral response (reducing the cost of rate reduction by $33 billion) is too large because the elasticity of the tax base would be lower than the 0.5 I assumed. (3) Applying the base broadening to those with incomes above $100,000 would create a “notch” with a jump in tax liabilities near that level. (4) The Tax Policy Center defined the middle class as all taxpayers with incomes under $200,000 while I used $100,000.

While I still believe the assumptions that I used in my analysis, I can modify them as suggested by the critics and still support my original conclusion by broadening the tax base in ways suggested but not developed in my WSJ piece. Eliminating a few of the “tax expenditure” exclusions and credits that are important for high-income taxpayers would raise more than enough revenue to compensate for assuming a smaller marginal tax rate, cutting the behavioral response effect in half, and phasing in the base broadening for individuals with incomes over $100,000 to avoid the notch.

More specifically, using a 25% marginal tax rate instead of 30% would reduce the revenue from eliminating deductions by 5% of $636 billion or $32 billion.  Cutting the behavioral response in half (i.e., using a taxable income elasticity of just 0.25) would raise the cost of the tax cut by $17 billion.  The cost of the “phase in” would depend on just how it was done but say another $15 billion of reduced revenue.  So instead of my conclusion that the revenue from eliminating deductions would exceed the cost of the tax cuts by $5 billion, these assumptions would imply a shortfall to be made up by other base broadening of $64 billion.

One part of that broadening could be eliminating the exclusion of employer payments for health insurance for those with AGI over $100,000. That  would increase income tax revenue by about $40 billion (out of the total revenue loss from the health insurance exclusion for all taxpayers of  $168 billion) plus an additional $10 billion of additional payroll tax revenue. (My estimate of this $40 billion is based on an imputation method developed by John Gruber based on data collected in the Medical Expenditure Panel Study.)

Eliminating the exclusion of municipal bond interest for taxpayers with AGI over $100,000 would increase tax revenue by an additional $15 billion.

Eliminating the child credit for those with incomes over $100,000 would increase revenue by an additional $10 billion.

So just those three changes to the list of base broadening measures would raise $75 billion or more than enough to exceed the $64 billion of potential shortfall with the very conservative assumptions noted above.

Additional tax revenue could be raised without reducing incentives to save or to invest efficiently by eliminating the exclusion for high-income taxpayers of such things as capital gains on home sales, the “cafeteria plan” benefits, and the capital gains at death.

One further point on the appropriate marginal tax rate (objection 1 above): although the top statutory rate is 35 percent, the effective top marginal tax rate is higher because of various phase-out provisions that affect high-income taxpayers (PEP, Pease, etc.) so my original assumption of a 30 percent marginal tax rate could be appropriate even with the Romney rate reductions.

The final objection is to my use of the $100,000 level to show that the middle class (i.e., those below $100,000 AGI) would experience no tax increases. The $100,000 level corresponds to 21 percent of all taxable returns and a significantly smaller fraction of all households.  I think it is very reasonable to say that people in that high-income group are not the “middle class.” The TPC focus on those with AGI over $200,000 limits that group to the top 4 million taxpayers who are three percent of all returns and five percent of all taxable returns.

So I think my conclusion stands: it is feasible to combine tax cuts and base broadening as Governor Romney suggests without raising the budget deficit or imposing any middle class tax increase. Critics might not like the Romney plan but they cannot call it “mathematically impossible.”

How distortionary are taxes?

An ec 10 student emails me a question about taxes:

Dear Professor Mankiw,

I'm a freshman taking Ec10. Prior to taking Ec10, I had no experience or interest in economics and was convinced that I wanted to be a doctor and to major in neurobiology. Now, Ec10 is easily my favorite and most intriguing course, and taking it, in combination with my reading Freakonomics, has definitely caused me to rethink my interests, goals, and life in general. Thank you for teaching this course.

However, my intrigue in Ec10 has led me to question some of the principles of economics. The first regards the idea that high income taxes distort work incentives. In theory, this makes sense, but I feel that in practice, the idea doesn't apply well, especially to the income tax. Do the distortion of work incentives and the deadweight loss really have enough of an impact to be seriously considered while forming tax policy? The reason I think that high income taxes don't significantly distort work incentives in practice is that most people probably aren't thinking about how much of their income is taxed unless it's April 15. Most people probably believe that working more means more money and don't consider how much they're taxed unless they have to physically write the check (e.g. with property and excise taxes).

[name withheld]

The key issue is the elasticity of labor supply. Some economists do believe, as you suggest, that this elasticity is small and, as a result, that taxes aren't very distortionary. Others believe that the elasticity is larger. Economist Ed Prescott has suggested that the main reason Europeans work less than Americans is the higher tax rates they face.

Think of it this way. Imagine that you are a painter deciding whether to accept another painting job this weekend. I am willing to pay you $500 for the work. It is possible that you would do the job for $500 but not for $300 (the amount you would keep after paying payroll taxes, federal income taxes, and state income taxes)? If you think that all workers make such decisions without regard to remuneration, then the assumption of inelastic labor supply is correct. In this case, taxes are not distortionary. But if some people respond to incentives and would do the job for $500 but not for $300, then that response is what makes taxes distortionary.

There are other margins to consider besides the tradeoff between work and leisure. Another is market production vs home production. To continue with our painting example, imagine that you have a choice between taking the job and staying at home to rake your leaves and fix your car. As a painter, your comparative advantage is likely painting, and it might make sense to hire a lawn service and a car mechanic to fill your other needs. But if the tax rate on your painting income is too high, you might pass on the extra job and spend your time on your tasks at home.

Finally, there is the issue of career choice. An article in today's NY Times mentioned a Harvard PhD in economics who left an academic job for a better-paying one in private equity. Based on the article, he seemed a bit wistful about leaving an academic job behind. At a higher tax rate on his new higher income, might he have stayed with the perks of the ivory tower? Perhaps. But, based on market prices, his talents are more productively applied in private equity, where he is filling the important role of allocating the economy's capital stock. If he gave up that job because of a higher tax rate, the loss to the overall economy would be measured by the deadweight loss.

On a related note: I am delighted to hear that you are reevaluating your career goals and considering a switch into economics, That is what the first year of college is all about. But I think we can agree that this important decision should ideally be based on comparative advantage and personal preferences, and not marginal tax rates.

Elasticity

We have been learning about elasticity in ec 10. One of the students emails me this clip from The Wire (a great TV show, by the way) on the concept. For those who have not seen the show, the main character in the clip, Stringer Bell, runs a drug operation but wants to go legit and has an apparent interest in economics. The person spying on Bell is Jimmy McNulty, a police officer who in a later episode finds a copy of Adam Smith's Wealth of Nations in Stringer's apartment.

Bob Frank replies

A few days ago, I expressed here my skeptical view of a recent column by Bob Frank. I offered Bob an opportunity to respond. Below I am reprinting, in its entirety, what he sends along. There is much that one could debate here, and I am sure the commenters will, but I will refrain. Since I picked this fight, and since I have ample opportunity in this forum to express my perspective, in fairness I will let Bob have the last word--at least for now.

First, my thanks to Greg for his gracious invitation to respond to his comments on my recent New York Times Economic Scene column. In that column, I argued against trickle-down theory’s claim that higher taxes on top earners would reduce economic growth. Here I’ll attempt to explain why Greg’s defense of trickle-down theory falls short.

Greg discounts the significance of the negative relationship I cite between wage growth and the average workweek over the last century. This relationship, he argues, is a consequence of the fact that the income effect of rising wages has offset the substitution effect (which is exactly how I described it). But the observed link is irrelevant, he explains, “because the distortionary effect of taxes depends only on the substitution effect. The evidence cited suggests that income effects are larger than substitution effects, not that substitution effects are small.”

Greg is right about what this particular piece of evidence shows. But he is mistaken in claiming that this evidence is irrelevant to my claim. Indeed, the argument I advanced in my column had nothing to do with whether taxes on the rich are distortionary. That’s an interesting question, and I’ll return to it in a moment. My only point in the column, however, was to question a very different claim—namely, that higher taxes on the rich would reduce work effort. That is precisely a claim about the total effect on work effort of lower after-tax wages. In other words, it’s a claim about the combined impact of the income and substitution effects. So the fact that the workweek declined over the last century in the face of substantial growth in real wages is directly supportive of my argument.

A necessary and sufficient condition for trickle-down theory’s argument to the contrary is that the elasticity of supply of labor with respect to real wages be significantly positive. The most comprehensive recent econometric study of labor supply elasticity in the United States will be published in the next issue of The Journal of Labor Economics. The authors, Fran Blau and Larry Kahn, estimate that the labor supply curve for men has been essentially vertical for many decades. The clear implication is that higher taxes on top earners, most of whom are men, will not significantly reduce work effort.

Greg also mentions research suggesting that higher taxes on the rich may reduce the amount of income they report to the IRS. Perhaps so, but that by itself would not imply any reduction in output. And with even the supply-sider Bruce Bartlett now conceding that tax cuts for top earners don’t boost total tax revenues, it’s important not to exaggerate the problem of unreported income. But irrespective of its magnitude, why isn’t the best solution to this problem a simpler and more strictly enforced tax code rather than tax rates that are too low to sustain minimally adequate public services?

Greg and I both have good health insurance, but millions of others have none. Absent the revenue necessary to fund universal health coverage, their ranks will keep growing as our current system of private, employer-provided insurance continues to unravel. President Bush’s recent proposal to make individually purchased health insurance tax deductible is unlikely to help. (As Stephen Colbert put it, “It’s so simple. Most people who can’t afford health insurance also are too poor to owe taxes. But if you give them a deduction from the taxes they don’t owe, they can use the money they’re not getting back from what they haven’t given to buy the health care they can’t afford.”) Without higher taxes on people like Greg and me, this problem will get worse.

Revenue shortfalls have also led to cuts in other important public services. I cannot imagine, for example, that Greg feels any more comfortable than I do about the Bush administration’s cuts in the Energy Department’s program for helping lock down loosely guarded nuclear materials in the former Soviet Union. But such programs cost money. Unless we can raise additional revenue to pay for this program, or unless we want to borrow even more heavily from abroad (loans that will eventually have to be repaid in full, with interest), it will remain underfunded. It is not satisfactory to assert that we can just reduce government waste. The president, who campaigned as an opponent of government waste, is the one who couldn’t find more wasteful or less politically protected programs to cut.

As evidence for his claim that I need to do additional reading, Greg cites a 1988 paper in which Joe Stiglitz argued that the socially optimal marginal tax rate on the most productive person might actually be negative. The reason, Stiglitz explained, is that inducing that person to work more could generate positive spillover effects for less productive workers. In the abstract, this is an interesting claim. (Is it any more than that? Stiglitz, for one, never thought to offer tax policy proposals on the basis of it.) But if we’re going to discuss externalities, then complementarities between skilled and unskilled labor are surely not the most important ones to consider.

For present purposes, by far the most important externalities are those stemming from the link between context and evaluation. As decades of behavioral evidence clearly demonstrates, virtually every evaluation is heavily shaped by local context. As Richard Layard put it, “In a poor society a man proves to his wife that he loves her by giving her a rose, but in a rich society he must give a dozen roses.” Because evaluation drives consumer choice, context is an important determinant of consumer demand. The upshot is that almost every consumer choice generates significant context externalities.
Consider, for example, a job applicant’s decision about how much to spend on an interview suit. His goal is to make a favorable impression. But his ability to do so depends far less on the absolute quality of his suit than on how it compares with those worn by other applicants. And when he spends more on a suit, he shifts the context within which other candidates will be evaluated.

Context externalities are pervasive. A good school, for instance, is one that compares favorably with other schools in the same local environment. The amount parents must spend to ensure that their children attend such a school is thus an increasing function of the amounts spent by other parents. The evaluations that guide an employer’s promotion decisions are similarly dependent on context. A worker’s odds of promotion depend less on his absolute performance than on how well he performs relative to his coworkers.

The dependence of evaluation on context lays waste to any presumption that individual decisions about how many hours to work or how much to spend on interview suits will be socially optimal. The general result predicted by theory is that if context shapes evaluation more heavily in some domains than others, too many resources will flow to the most context-sensitive domains and too few to the least context-sensitive domains. In my forthcoming book, Falling Behind, I summarize what available evidence says about the extent to which context differs across domains. For the discussion at hand, the relevant finding is that evaluations of leisure tend to be far less context-sensitive than evaluations of income. The implication is that individual valuations of leisure tend to understate social valuations. Thus people work longer hours in the hope of moving higher on the income ladder, only to discover that when others do likewise, their position remains unchanged.

It would be unfair to single out Greg for ignoring context externalities. After all, most of the standard economic models that serve as the basis for policy analysis make no mention of these externalities. At some point, the economics profession will look back in embarrassment about that fact. But even absent explicit mentions of context externalities, most practical policy analysts already seem to recognize that trickle-down theory’s portrait bears little relation to the behavior of people in the real world.

My point is not that people don’t care about money. On the contrary, when the pay in one occupation goes down relative to others, fewer people enter that occupation. Public school teachers, whose starting salaries were more than 20 percent higher than those of the average college graduate in the 1960s, now earn below-average starting salaries. So we are not surprised that fewer qualified people now enter teaching.

But trickle-down theory is about what happens when after-tax pay falls not just in some occupations but for top earners generally. In a largely meritocratic society like the United States, most top earners are extremely driven people. And as recent studies have shown, most of them will never spend more than a small fraction of their earnings. The trickle-down theorist’s insistence that they will begin slacking off in response to a small increase in their marginal tax rates strains credulity.

While serving as chairman of the Council of Economic Advisers, Greg actively supported the Bush tax cuts targeted at top earners by arguing that the cuts would spur them to work harder. Greg would have been astonished to observe such a response from his colleagues at Harvard. Does he have a behavioral model that leads him to expect different behavior from high achievers in other occupations? Or does he have one that explains why any such differences consistently fail to reveal themselves in the data? In the absence of a plausible behavioral model backed by persuasive empirical evidence to the contrary, I stand by my conclusion that trickle-down theory is supported neither by economic theory nor by empirical evidence.

The tax cuts that were sold by invoking this theory did little to promote the well-being of even the well-to-do Americans who were their ostensible beneficiaries. Money that could have been spent rounding up loose nuclear materials in the former Soviet union was spent instead on larger houses and more expensive cars. In light of what we know about the empirical magnitude of context externalities, the principal effect of such spending was simply to redefine what counts as adequate. As in the familiar stadium metaphor, all stand to get a better view, yet none sees better than if all had remained seated.

Greg titled his response to my column “Frank Needs To Read More Widely.” On that point, he is surely right. I don’t know Greg well enough to presume to know what he needs. But he would almost surely offer better policy advice if equipped with an economic model that better fits current scientific knowledge about human behavior.

Again, my thanks to Greg for inviting me to respond to his critique of my column.

Follow-up references

In my most recent Times column, I did not have the space to fully explain the body of work that follows up on the Prescott hypothesis that higher tax rates explain lower work effort and national incomes in Western Europe. For interested readers (that is, the more nerdy ones), here are a some relevant references together with brief excerpts (emphasis added):

1. Steven Davis and Magnus Henrekson

"Lastly, let us return to the recent studies by Prescott (2002, 2003), which consider the output, employment and welfare consequences of personal taxes in an equilibrium model with one production sector and a simple labor-leisure choice for the representative household. Our evidence supports the view that tax rate differences among rich countries are a major reason for large international differences in market work time. At the same time, however, our evidence strongly suggests that labor and consumption taxes operate with powerful effect on several margins: substitution between legal and underground activity, substitution between home and market production, the mix of market production activity, and the composition of market expenditures."

2.  Indraneel Chakraborty et al. 

"Americans work more than Europeans. Using micro-data from the United States and 17 European countries, we document that women are typically the largest contributors to the cross-country differences in work hours. We also show that there is a negative relation between taxes and annual hours worked, driven by men, and a positive relation between divorce rates and annual hours worked, driven by women. In a calibrated life-cycle model with heterogeneous agents, marriage and divorce, we find that the divorce and tax mechanisms together can explain 45% of the variation in labor supply between the United States and the European countries."

3. Alberto Alesina et al. 

"Our punch line is that Europeans today work much less than Americans because of the policies of the unions in the 1970s, 1980s, and part of the 1990s and because of labor market regulations. Marginal tax rates may have also played a role, especially for women's labor force participation, but our view is that in a hypothetical competitive labor market without unions and with limited regulation, these tax increases would not have affected hours worked as much. Certainly micro evidence on the elasticity of labor supply is inconsistent with a mainly tax-based explanation of this phenomenon, even though social multiplier effects may help in this respect."

4. Raj Chetty et. al.

"Based on our reading of the micro evidence, we recommend calibrating macro models to match Hicksian elasticities of 0.3 on the intensive and 0.25 on the extensive margin and Frisch elasticities of 0.5 on the intensive and 0.25 on the extensive margin. Hence, it would be reasonable to calibrate representative agent macro models to match a Frisch elasticity of aggregate hours of 0.75. These elasticities are consistent with the observed differences in aggregate hours across countries with different tax systems."

As I noted in my column, economists disagree about the how far the tax-based explanation goes. A reasonable reading of the literature is that lower labor effort and incomes in Europe are likely due to a combination of higher tax rates, stronger unions, and greater regulations.

Disincentives from Health Reform

Here is my column in tomorrow's NY Times about the marginal tax rates implicit in the health reform bill making its ways through Congress. Let me add a few additional observations on the topic.

1. Here are the CBO numbers on which the article is based. Unfortunately, the Times did not run the table of implicit marginal tax rates that I gave them based on the CBO numbers. But the example I used in the piece (an implicit tax rate of 23 percent) is representative. For lower income levels, the implicit marginal tax rate is even higher. Between $42,000 and $54,000, the implicit marginal tax rate from health reform is 34 percent.

2. When CBO estimates the budgetary cost of such bills, it holds GDP constant. If you think (as I do) that large increases in marginal tax rates tend to depress labor effort and thus GDP, then you should be wary of claims based on CBO scores that the health reform bill is deficit neutral. Lower GDP will mean lower tax revenue and thus a larger budget deficit.

3. How much do people respond to tax rates? Economists differ in their answer to this question. The latest thinking on this topic, by my Harvard colleague Raj Chetty, indicates that the elasticity of taxable income with respect to (1-tax rate) is about one half. So, for example, if a person starts with a marginal tax rate t of 0.3 and health reform raises it to 0.5, the percentage change in 1-t, using the midpoint method, is .2/.6, or 33 percent. With an elasticity of one half, his taxable income will fall by 17 percent. Thus, the economic impacts from these implicit tax hikes are sizable.

4. In my Times piece, I wrote, "None of this necessarily means that health reform is not worth doing. President Obama’s push for reform is premised on the belief that access to good health care should be a right of all Americans — a proposition better judged by political philosophers than economists. But we should not forget the cost of translating that noble aspiration into practical policy."

This passage may seem a bit passive-aggressive, as I appear to be criticizing the bill without really taking a stand. My aim, however, is to emphasize that economics alone cannot settle the debate.

Behind the healthcare debate is the classic tradeoff between equality and efficiency. Consider the following question, which is not about healthcare per se: Would you favor a substantial increase in marginal tax rates for millions of middle and upper income Americans to provide more resources for those toward the bottom of the economic ladder?

Your answer to this question cannot be determined by positive economics without adding in some normative judgments. But your answer should strongly influence your view of the health reform bill. The bill moves us closer to much of Western Europe by favoring equality and paying the price of reduced efficiency from much higher marginal tax rates.

That may be a policy choice Americans want to make. But before buying the merchandise being offered by Congress, I hope we all take a close look at the price tag.

CBO on Supply-side Economics

Lower tax rates induce people to work more. Greater work effort induces greater tax revenue, offsetting some of the initial losses in tax revenue from the tax cut. I believe this effect is one of the key tenets of supply-side economics. Most mainstream economists agree, to some degree.

The big question is about magnitudes: How big is this effect? In a new report, CBO reports this estimate:

CBO estimates that the change in tax revenues from the shift in labor supply would offset roughly 4 percent of the static revenue loss.

The report is not blunt, but its bottom line is simple: Supply-side effects are trivial.

I don't buy it. To understand why the CBO reached this conclusion, the key seems to be found in Table 2. According to this table, the earnings-weighted compensated labor supply elasticity is 0.14. With such a small elasticity, their model naturally yields small behavioral responses to changes in tax rates.

In my paper with Weinzierl, we reviewed some of the literature and concluded that a reasonable number for this parameter was 0.5. Kimball and Shapiro concluded it is even larger. If one plugs 0.14 into the formulas Weinzierl and I gave, one would find effects as small as those reported by CBO. So the details of the different models do not appear central; this part of the macroeconomic debate seems to boil down to a single parameter. Unfortunately, the academic literature on this topic is far from conclusive. But I am not sure it was prudent for CBO to settle on the low end of the plausible range of estimates.

Herb Stein is supposed to have once said, "There is nothing wrong with supply-side economics that division by ten wouldn't fix." If my reading of the CBO report is correct, then there is nothing wrong with it that multiplication by 3 to 6 wouldn't fix.

Update: An informed reader tells me this particular study is only part of CBO's analysis of tax policy. For example, it does not deal with the offsetting spending changes that must eventually accompany any tax change. As I explain in another post, one cannot fully analyze tax policy without thinking through how the government budget constraint is satisfied. My friend recommends this CBO piece to get a better sense of the full range of models used.

Geography and Mobility

A former student, M. Daniele Paserman, who is now a professor at Boston University, sends me the following email, which I thought was interesting enough to share (with permission, of course):

I bumped into your blog post on the Great Gatsby curve, and I was happy to see you raise the point about the arbitrariness of imposing geographic boundaries in measuring intergenerational mobility (why should one lump Connecticut and Mississippi together?)

Claudia Olivetti and I raise a similar point in our recent paper on the evolution of intergenerational mobility in the US between the end of the 19th and the beginning of the 20th Century. We measure a large increase in the intergenerational elasticity between the the cohort of children born in the 1850s and those born in the 1910s, but almost all of it can be explained by income divergence across regions. In fact, within the Northeast and the Midwest, the intergenerational elasticity was flat, or maybe even falling (it was rising in the South, though).

Spending and Tax Multipliers

A key issue facing the new Obama administration is to what extent the economic stimulus should take the form of spending increases versus tax reduction. One way to think about the issue is the size of the fiscal policy multipliers. The multipliers measure bang for the buck--the amount of short-run GDP expansion one gets from a dollar of spending hikes or tax cuts.

So what are these multipliers? In their new blog, Bob Hall and Susan Woodward look at spending increases from World War II and the Korean War and conclude that the government spending multiplier is about one: A dollar of government spending raises GDP by about a dollar. Similarly, the results in Valerie Ramey's research suggest a government spending multiplier of about 1.4. (Valerie does not present her results in multiplier form, but she emails me this translation: "The right column of figure 5A of my paper shows that for a log change of government spending of 1, log GDP rises by 0.28, implying an elasticity of 0.28. To back out the implied multiplier, we can use the fact that government spending averages around 20% of GDP. This implies a multiplier of 1.4.")

By contrast, recent research by Christina Romer and David Romer looks at tax changes and concludes that the tax multiplier is about three: A dollar of tax cuts raises GDP by about three dollars. The puzzle is that, taken together, these findings are inconsistent with the conventional Keynesian model. According to that model, taught even in my favorite textbook, spending multipliers necessarily exceed tax multipliers.

How can these empirical results be reconciled? One hypothesis is that that compared with spending increases, tax cuts produce a bigger boost in investment demand. This might work through changing relative prices in a direction favorable to capital investment--a mechanism absent in the textbook Keynesian model.

Suppose, for example, that tax cuts are not lump-sum but instead take the form of cuts in payroll taxes (as suggested by Bils and Klenow). This tax cut would reduce the cost of labor and, if labor and capital are complements, increase the demand for capital goods. Thus, the tax cut stimulates demand not only by increasing disposable income and consumption spending (the textbook Keynesian channel) but also by incentivizing more investment spending. A similar result might obtain if the tax cut included, say, an investment tax credit.

This hypothesized channel seems broadly consistent with the empirical findings of Blanchard and Perotti, Mountford and Uhlig, Alesina and Ardagna, and Alesina, Ardagna, Perotti, and Schiantarelli. The results of all these authors suggest you need to go beyond the standard Keynesian model to understand the short-run effects of fiscal policy.

My advice to Team Obama: Do not be intellectually bound by the textbook Keynesian model. Be prepared to recognize that the world is vastly more complicated than the one we describe in ec 10. In particular, empirical studies that do not impose the restrictions of Keynesian theory suggest that you might get more bang for the buck with tax cuts than spending hikes.

Cross-Price Elasticity of Demand II

If you are thinking about buying a smaller car or a hybrid in response to the high price of gasoline, here (from the Financial Times via Art Carden) is an alternative idea:

Farmers in the Indian state of Rajasthan are rediscovering the humble camel. As the cost of running gas-guzzling tractors soars, even-toed ungulates are making a comeback, raising hopes that a fall in the population of the desert state’s signature animal can be reversed.

“It’s excellent for the camel population if the price of oil continues to go up because demand for camels will also go up,” says Ilse Köhler-Rollefson of the League for Pastoral Peoples and Endogenous Livestock Development. “Two years ago, a camel cost little more than a goat, which is nothing. The price has since trebled.”

Cross-Price Elasticity of Demand V

The AP reports:

High gas prices drive farmer to switch to mules

High gas prices have driven a Warren County farmer and his sons to hitch a tractor rake to a pair of mules to gather hay from their fields. T.R. Raymond bought Dolly and Molly at the Dixon mule sale last year. Son Danny Raymond trained them and also modified the tractor rake so the mules could pull it.

T.R. Raymond says the mules are slower than a petroleum-powered tractor, but there are benefits.

"This fuel's so high, you can't afford it," he said. "We can feed these mules cheaper than we can buy fuel. That's the truth."

Thanks for Frank Stephensen for the pointer.

Here is the previous installment in this series.

Cross-Price Elasticity of Demand VII

The LA Times reports:

As Gas Prices Climb, So Do Scooter Sales

Scooters...are selling as fast as their little wheels can carry them from showroom floors. Sales have jumped 23.6 percent in the first quarter of 2008 compared with the same period in 2007.

Click here the the previous installment in this series.

Rap for Econ 101

Some years ago, I plugged some rap songs designed to be used in classrooms for introductory economics. The creators alert me that these songs can now be found here. They also note that the songs on elasticity and regulation are their personal favorites.

A Reading for the Pigou Club

A reader alerts me to an article in the journal Energy Policy:

Fuel taxes: An important instrument for climate policy

Abstract

This article shows that fuel taxes serve a very important role for the environment and that we risk a backlash of increased emissions if they are abolished. Fuel taxes have restrained growth in fuel demand and associated carbon emissions. Although fuel demand is large and growing, our analysis shows that it would have been much higher in the absence of domestic fuel taxes. People often assert that fuel demand is inelastic but there is strong research evidence showing the opposite. The price elasticity is in fact quite high but only in the long-run: in the short run it may be quite inelastic which has important implications for policy makers. Had Europe not followed a policy of high fuel taxation but had low US taxes, then fuel demand would have been twice as large. Hypothetical transport demand in the whole OECD area is calculated for various tax scenarios and the results show that fuel taxes are the single most powerful climate policy instrument implemented to date—yet this fact is not usually given due attention in the debate.

The Half-Full Glass of Economic Mobility

When people look at data on economic mobility, they see different things.  For example, it is well known that if your father had high income, you are more likely to have high income than if you father had low income.  According to this study (which I found thanks to a pointer by Paul Krugman), the elasticity of son's income with respect to father's income is about 0.5 in the United States.  How do you interpret this fact?

Some people might be tempted to see it as evidence against equality of opportunity.  After all, it shows that it matters where you started.  Rich parents can buy better schools, expensive tutors, fancy summer camps, and all sort of other great stuff for their kids.  How fair is that?

But what strikes me about that 0.5 number is not how large it is but how small it is.  As I understand it, that 0.5 estimate is roughly the correlation between father and son income.  That means that the fraction of variance of son's income explained by father's income--that is, R-squared--is only 0.25.  This last number is sometimes called the "heritability" of a characteristic.

By contrast, the heritability of IQ is usually estimated to be much larger than that.  At least some of the heritability of income must come not from inequality of opportunity but from the genetic transmission of talent.  Other aspects of talent, such as drive, energy, and spunk, might well have a genetic component as well, but they are harder to measure and thus we know less about them.  But the one that has been studied extensively, IQ, seems more heritable than income.

The bottom line: In light of the heritability of talent, it would be shocking if we did not find some significant heritability of income.  And that would be true even if equality of opportunity were perfect.

One further thought: The study cited above points out that economic mobility is greater in some European countries.  That fact does not surprise me, as those are nations with less inequality.  Moving up and down a short ladder is a lot easier than moving up and down a tall one.

---
For a related previous post and links to a couple relevant studies, click here.  See also this survey of the literature.