Something Not Unlike Research

A worse than useless survey

post by Paul Kelleher

Robin Hanson points us to a forthcoming paper in Economics Letters by Olof Johansson-Stenman that purports to "provide survey-based support for the consequentialist assumption associated with welfare economics." Consequentialism is the normative view that an act or policy is right if and only if it maximizes the good (i.e. the net balance of good over bad). As such, it treats an act's rightness or wrongness as solely a function of outcomes. Modern welfare economics is consequentialist in that it aims to be a science of efficiency focused on welfare-maximization. But then again, plenty of welfare economists insist that what they study is but one sphere of moral investigation, and they would not agree that an interest in efficiency entails a single-minded interest in it.

Johansson-Stenman sent his survey to 2,450 Swedes over the age of 18 in the Spring of 2004. Forty-five percent responded (n=985). Here is the main question asked, together with the respondents' answers (click the image to enlarge it):

According to Johansson-Stenman, "the result is quite consistent with the consequentialist ethics underlying conventional economic welfare theory, since almost two-thirds chose this alternative." But this is an illegitimate interpretation of his results. As noted, consquentialism is the view that rightness and wrongness depends only and exclusively on the outcome. The survey, however, only asks people what they think the badness of an outcome "depends primarily on." So for all we know, the 67% who gave what the author characterizes as "the" consequentialist answer may actually agree with those welfare economists who are not consequentialists at all because they believe that other moral considerations (e.g. individual rights and liberties) matter too.   

The survey also makes the fatal mistake of asking its main question in terms of "badness" rather than "rightness" or "wrongness." To see why, consider this passage from a paper (pdf) by Frances Kamm, philosophy's leading non-consequentialist:

I shall distinguish between goodness, fairness and justice. To make one distinction clearer, consider the following case: A doctor must decide whether to stop a big pain in person A or a small pain in person B. She thinks, correctly, that she will do more good if she helps A. But she also remembers that yesterday B suffered a much bigger pain than A will suffer and no one helped B (A suffered nothing in the past). So she thinks it would be unfair to let B suffer again, even though she will do less good if she helps him. If it is overall right to do this, this means she does the morally better thing in helping him and the state of affairs in which B is helped rather than A is morally better than one in which A is helped. But this is not because it produces more good.

Here Kamm claims that sometimes considerations of fairness can conflict with considerations of goodness and badness. But note that someone using Kamm's conceptual scheme might well have given the survey answer that Johansson-Stenman associates with consequentialism. For the survey asks respondents to indicate which considerations they associate with "badness," and the simple (but not simplistic) Kammian answer is that welfare-related considerations are associated with badness, while fairness or equity considerations must be called in alongside welfare considerations to determine which action is "overall right" or "morally better." So it is quite possible that many survey respondents took themselves to be answering a question about only one part of morality, viz. badness, while also holding beliefs about other parts not asked about.

I conclude, therefore, that this survey is useless in helping to answer the question of whether most people are consequentialists. Indeed, since its publication has already led one influential writer (Hanson) to claim that "Most are consequentialist," I fear that it's actually worse than useless.  

Is Rawls Relevant in a Warming World?

post by Paul Kelleher

The short version: Tim Mulgan thinks John Rawls's A Theory of Justice has limited value for thinking about moral issues associated with climate change. I disagree. I describe the role that discounting plays in cost-benefit analyses of climate change, and argue that Rawls's discussion of discounting is useful and of lasting importance. 

Philosopher Tim Mulgan thinks that John Rawls may go the way of "paisley shirts and psychedelic music":

[O]ptimistic assumptions allow political philosophers to ignore the future. Consider Rawls himself.  His magnum opus (A Theory of Justice) devotes just ten pages to justice between generations; and even then his focus is tellingly limited.  As we accumulate wealth, we must decide how much to consume for ourselves and how much to invest for the future.  For Rawls, the only ethical question is: How much better-off should we leave our descendants?

Rawls wrote in 1971. Like paisley shirts and psychedelic music, his complacent approach has not aged well.  We no longer take it for granted that we will leave our descendants better-off, or even that we can.  And future people are no longer an afterthought.  They have moved centre stage, both in popular debate and in moral philosophy.

Mulgan may well be correct that the majority of A Theory of Justice does not speak to the issue of inter-generational justice. Nevertheless, I believe that what Rawls does have to say is of lasting importance. Consider the seemingly academic issue of discounting. In cost-benefit analyses of climate-related plans and projects--and in cost-benefit analysis generally--it is common for economists to discount the value (or disvalue) of goods (or bads) that occur in the future. Depending on the discount rate used, we--here and now--can have stronger or weaker duties to prevent health shocks that will occur in the far future. If we discount the goodness of future health and the badness of future ill-health, then it seems to follow that we have commensurately less pressing reason to prevent those shocks from occurring. After all, no one thinks that it's morally required for current generations to take on great burdens to prevent one hangnail in 200 years. Since the nature of our obligation clearly turns on the size of the future benefit or burden we're talking about, discounting has the potential to turn a future moral tragedy into something not that important.

The economist and philosopher John Broome illustrates the importance of getting the discount rate right:

Stern's conclusion differs from Nordhaus's principally because, on ethical grounds, Stern uses a lower "discount rate." Economists generally value future goods less than present ones: they discount future goods. Furthermore, the more distant the future in which goods become available, the more the goods are discounted. The discount rate measures how fast the value of goods diminishes with time. Nordhaus discounts at roughly 6 percent a year; Stern discounts at 1.4 percent. The effect is that Stern gives a present value of $247 billion for having, say, a trillion dollars' worth of goods a century from now. Nordhaus values having those same goods in 2108 at just $2.5 billion today. Thus, Stern attaches nearly 100 times as much value as Nordhaus does to having any given level of costs and benefits 100 years from now.

In a forthcoming short paper on discounting, I describe two different kinds of discounting and the main reasons why one might support them. Here they are in brief:

Kind 1/Reason 1: When individuals save money in bank accounts, they are trading present consumption for future consumption. But many individuals would not do this unless they were given an incentive in the form of an interest rate. Therefore, since individuals need to be in effect bribed to put off present consumption, individuals have a preference for consuming now rather than later. Some economists see this as revealing individuals' ethical opinion that future well-being is morally less important than present well-being.

Kind 1/Reason 2: There is a second reason that some people offer for discounting the moral importance of future benefits. The argument goes like this: unless we discount the value of future benefits (and the disvalue of future bads), current generations will be forced to make unreasonable sacrifices. After all, since what we do here and now will have effects that radiate out in the far future, a large sacrifice now will certainly create benefits in the future that outweigh the sacrifice we have to make. The only way to release ourselves from these surely tyrannical obligations to future generations is to count their well-being for less. Thus, we need to discount to protect ourselves from overdemanding duties.

Kind 2/Reason 1: Any use of current resources has opportunity costs---that is, we could have used them for other purposes, and these other uses may have yielded even greater benefits. Economists in general would like us to put scarce resources to their most efficient possible use. One possible use is a financial investment: we could invest the resources now and let them grow, at prevailing interest rates, to a certain size in the future. If it is impossible to achieve that same sized gain by investing current resources in environmental projects, then funding the environmental project "leaves resources on the table," so to speak. That is, we could give future people a larger benefit if we just invested the money and gave them the pin number to the bank account than if we funded the environmental project. To choose between projects, economists tell us to discount the future environmental benefits at prevailing interest rates. This discounted value is the amount we'd have to invest now to deliver future benefits of the same value as the environmental benefits. If it would cost more to deliver those benefits via an environmental project, then (the economists argue) we should go with the bank account instead.---The perceptive reader will notice that this second reason for "discounting" doesn't support discounting the moral importance of future well-being. Rather, it holds the moral importance of future benefits fixed and simply tells us how to deliver those benefits most efficiently. That is why I (following Dominic Rosser [pdf]) call it "discounting" and why I consider it a different kind of discounting from the first kind discussed above. 

As regards the first reason for Kind 1 discounting, Rawls would surely argue that when individuals are making their own private investment decisions, they should not be taken--nor do they take themselves--to be expressing a moral stance on whether the well-being of future people matters less from a moral point of view. It is Rawls's view that moral deliberation must be explicit and sensitive to specifically moral reasons. At least that's what I take the his argument from the Original Position to entail.

The short discussion of inter-generational justice in A Theory of Justice actually provides an explicit and decisive refutation of the 2nd reason for Kind 1 discounting. Rawls notes that duties to promote future well-being would be tyrannical only if we had a basic duty to maximize well-being. But that duty is questionable, and is only supported by the consequentialist moral framework he is rejecting in the book. The consequentialist framework is, however, the framework that many welfare economists use when constructing cost-benefit analyses, and this is why many of them are led to embrace discounting. If Rawls is correct that we do not have a simple duty to maximize well-being, then Kind 1 discounting is not needed to release ourselves from that duty.

Finally as regards the reason for Kind2 "discounting," Rawls seems willing to grant that it may have something to be said for it. He writes, "[T]he rejection of pure time preference [i.e. Kind 1 discounting]...does not rule out using an interest rate (in either a socialist or a private-property economy) to ration limited funds for investment" (TJ, p. 294 [Original 1971 edition]). In my forthcoming paper I offer some moral reasons for being troubled by Kind2 discounting, but like Rawls I am willing to admit that it may be useful in some cases. 

If Broome is right that the issue of discounting is crucial for properly understanding our duties to future generations, then A Theory of Justice--and especially the ten pages on inter-generational justice--may contain more that is of lasting importance than Mulgan's discussion would suggest.

Mankiw vs. Nozick on Just Deserts

post by Paul Kelleher

The short version: Greg Mankiw believes people should get what they deserve, and that what people deserve is based on the economic value of what they contribute to social production. Mankiw attributes this view to libertarian philosopher Robert Nozick. But Nozick explicitly rejects the view, noting a problem with it that Mankiw never addresses.

Students of economics learn that, under a certain set of assumptions, a competitive market economy leads to an efficient allocation of resources. However, since economic efficiency may not the only thing we care about, many economists are careful to note that there may be equity-based reasons to condemn otherwise efficient arrangements. Harvard economist Greg Mankiw thinks this is a mistake. In "Spreading the Wealth Around: Reflections Inspired by Joe the Plumber", Mankiw argues that when competitive markets promote efficiency, they do so by rewarding factors of production (including labor) in proportion to their productive contribution. Mankiw thus extols "the inherent equity of the competitive market": the market ensures that factors of production are paid what they deserve.

Mankiw claims that this "Just Deserts Theory" can be found in the work of the libertarian philosopher Robert Nozick (among others). But in fact, Nozick believed the area of economics that Mankiw draws on--Marginal Productivity Theory (MPT)--cannot reliably be used to assess individuals' productive contributions. Nozick did think that MPT could explain why individuals get what they get in market distribution, but this was not because Nozick thought the market does or should reward people in accordance with their contribution. Rather, Nozick's view is that people are entitled only to what others are willing to give them (either as gifts or in economic exchange). MPT helps us understand who will be willing to give what to whom, but that is it. So Nozick would say Mankiw is wrong on two counts: (1) people are not in fact entitled to their "Just Deserts"; (2) even if people were entitled to their just deserts, MPT cannot be used to identify what it is people deserve in Mankiw's sense.

Let me explain Nozick's argument for this second point. Nozick agrees with Mankiw that (at least under a set of "idealized circumstances") MPT can be used to predict the wage that rational firms will be willing to pay to a worker for performing a specific job. To illustrate, consider the following graphic*:

For economic reasons that Mankiw and Nozick agree upon, a rational firm will pay all laborers a wage equal to the value of the output generated by adding the last laborer, depicted here as laborer #10. To use some economic jargon, firms will pay a wage equal to the marginal value product of labor. Where Nozick and Mankiw disagree is on how this economic jargon should be interpreted. According to Mankiw, when wages are set in this way, it ensures that "each person's income reflects the value of what he contributed to society's production of goods and services...[E]ach person receives his just deserts" (emphasis added). Nozick claims that this is not an accurate characterization. Here is why: all we know from the graphic is that after adding the 10th laborer, the total product went up by 6 bushels of corn; but we don't know exactly how that outcome arose. For example, it might have been achieved by the 10th laborer simply producing 6 extra bushels all by himself. Or, those 6 extra bushels might have resulted from the 10th laborer producing 16 bushels while some other laborers simultaneously reduced their collective output by 10 (perhaps because they reduced their effort when the 10th man was added). In this case, the marginal value product of labor will still be defined by MTP as 6 bushels, and all the workers will still be paid a wage equal to the value of 6 bushels. But it will nonetheless be false to say (as Mankiw does) that "each person's income reflects the value of what he contributed." While it is true that adding the 10th laborer resulted in a net increase of 6 bushels, we cannot deduce any worker's actual contribution from this fact. 

Nozick summarizes this point in a footnote to p. 187 of Anarchy, State, and Utopia:

Thus marginal productivity theory is not best thought of as a theory of actual produced product...but rather as a theory of the difference (subjunctively defined) made by the presence of a factor.

Paul Samuelson makes the same point here:

In our example, output increases from ten bushels to eighteen bushels when one adds the second man not because the second man only adds eight, but rather because his presence on the field makes the situation such that the total output of both men is eighteen...[W]hat we wish to note is that by adding the second man, output was increased by eight. Thus, the marginal product of the second man is eight. But his actual contribution may be very different than this.

If Mankiw is right that each person's income should reflect the value of what s/he contributed, it does not seem that marginal productivity theory can tells us what we need to know. Importantly, this worry arises even if we grant the many unrealistic assumptions MPT seems to require for its validity. Of course there may be even more problems with using MPT as part of a theory of just income distribution. Amartya Sen discusses several of these in a very nice 1985 paper called "The Moral Standing of the Market" (which Mankiw doesn't cite).

 

*This figure comes from p. 28 of David Schweickart's After Capitalism.

What kind of mobility matters?

post by Paul Kelleher

Mark Thoma had a column yesterday on income inequality and economic mobility. He argues that greater income inequality may be less problematic if there were also greater mobility. If one's chances for moving up and down the economic ladder were no better or worse than anyone else's (including those who are currently at the top), then a longer ladder may not be such a bad thing.

Thoma identifies two kinds of mobility: intra-generational and inter-generational. Intra-generational mobility concerns the prospects for mobility that individuals and/or families face during their lifetime. Inter-generational mobility concerns "how easy it is to move between social classes across generations." In other words, relative to the prospects your parents faced, are yours better or worse?

I worry (but am not sure) that Thoma is focusing on both the wrong problem and the wrong solution. Near the end of the article Thoma claims, "Increasing mobility can help, but this alone won’t overcome the inequality problem." What, exactly, is "the inequality problem"? Reading his article, I don't see Thoma complaining about income inequality as such. Rather, I see him worrying about two more specific things. First, he is concerned about the power that those at the top often have to bend public policy toward their interests. Second, he is concerned about stagnant or even reversing living standards for those at the bottom and the middle. These two issues are connected to income inequality, but they are connected only contingently. For it is possible to imagine a society where the living standards of everyone improve significantly and steadily, but where those at the top improve faster than others, and so inequality grows. It is also possible (although perhaps harder) to imagine policy changes that hem in the political clout that greater wealth buys, so that political influence becomes somewhat disconnected from economic status.

Let's set aside the issue of political power and focus on the connection between income inequality and stagnating standards of living for the non-rich. Here there may be a kind of mobility that is more important than the two kinds Thoma introduces. His notions are both relative notions: they assume that mobility is a zero-sum game, so that if one person's mobility improves, another's automatically declines: if my chances of getting to the top increase, then the chances that a person at the top will drop down must also increase. But Lane Kenworthy--whom Thoma also cites--notes that there are also absolute notions of mobility:

Absolute intragenerational mobility refers to changes in income compared to the income one started with. Suppose a person begins her working career with an income of $25,000. If a decade later her income is $30,000 (adjusting for inflation), she has experienced upward absolute intragenerational income mobility. [...]

For intergenerational mobility, the distinction between absolute and relative is analogous. If a person’s inflation-adjusted income is $30,000 and her parents’ was $25,000 at a comparable point in life, she has experienced upward absolute intergenerational income mobility. Because of economic growth, we expect that such upward mobility will be the norm.

It is at least conceptually possible to achieve absolute mobility--of both intra- or inter-generational varieties--while income inequality remains steady or even increases. And if we again bracket the issue of unequal political power, one wonders how many people would care about stagnant or declining relative mobility if absolute mobility were improving appreciably. That is, if living standards at the bottom and the middle were improving nicely rather than deteriorating, how much would those in the middle care if their chances of rising to the very top were not improving? How much should they care? 

Over the last 30 years, increasing income inequality has been accompanied by stagnating living standards for much of the population. And it is probably true that improving living standards for the non-rich would lead to less income inequality. But it is not clear that it would or even should lead to greater relative mobility. Significant improvements in relative mobility, especially for those at the middle, may--may--require arrangements that carry unpalatable efficiency costs as well. On the other hand, if improvements in absolute mobility is the more important issue, then we may be able to boost mobility while avoiding the costs associated with distributions that are flat enough to permit easy movement between quintiles of the income distribution.

Finally, since absolute mobility isn't a zero-sum game, it may be achievable without having to displace the privileged from their position of relative advantage. That may turn out to be a politically useful ancillary implication of choosing to care more about absolute mobility than relative mobility. 

Beware the share

post by Paul Kelleher

Kevin Drum writes:

As you can see, the top 10% receive 41% of total income and pay 55% of total taxes. In other words, the federal tax system as a whole is progressive, but it's not very progressive.

I think this is not a good way of determining whether a tax system is progressive. Consider the following hypothetical claims that are nevertheless consistent:

1. The top 10% receives 50% of total income but pays 100% of total taxes. (On Drum's definition, this system is very progressive.)

2. But the only tax that exists requires everyone in the top 10% to pay one penny only, while everyone else pays nothing.

I think this shows that a group's share of total national taxes is not a meaningful number on its own, and shouldn't be used to determine the degree to which a tax system is progressive.  

Tax Facts

post by Paul Kelleher

Prompted by Pres. Obama's proposal of a "Buffett Rule"--that billionaires should not pay less in taxes than their secretaries--Greg Mankiw today writes (his emphasis):

If you can remember only one fact, make it this one: The middle class (middle quintile) pays 14.1 percent of its income in federal taxes, while the rich (top tenth of one percent of the population) pay 30.4 percent.

That fact does seem to establish that the the U.S. has a progressive federal tax system. Of course, Americans pay taxes other than federal taxes, so this fact does not alone establish the degree of overall progressivity of the U.S. tax system.

Moreover, there are facts other than the one Professor Mankiw selects that are worth remembering. Here are some of them in chart form, taken from Jacob Hacker and Paul Pierson's book Winner Take All Politics (click to enlarge):

 

* * *

The evident trend is tremendous growth in after-tax income for the richest Americans coupled with a marked decline in their federal tax rates over the last 30 years. The first chart does not report on income growth among the top one-tenth of one percent, which is the group Mankiw compares to Americans in the middle-fifth of the income distribution. For some perspective on that question, albeit with what I believe is pre-tax data, here's a graph from the Century Foundation's blog:  

 

My takeaway (although I'm willing to revise it): while the richest Americans do pay a higher federal tax rate on their income than do most Americans, most Americans have not enjoyed anything close to the income gains that the richest have enjoyed over the last 30 years. Mankiw has a paper in which he argues that this divergence in income growth is the result of skill-biased technological change. Hacker and Pierson offer a great deal of evidence against that interpretation in their book referenced above.

Welfare Economics -- A Follow-up

post by Paul Kelleher

This is a somewhat long follow-up to my recent post on welfare economics. I clarify my remarks and reply to some objections from the economist Julian Jamison.  

The Medicare eligibility age and the life course path of health

post by Bill Gardner

Most of us begin life with excellent health but -- breaking news! -- it declines over time. As it does, we have increasing needs for health care. However, some people get sick faster as they age, and hence are likely to need more health care sooner. This needs to be considered in the debate about raising the age of eligibility for Medicare.

Here is a graph from Anne Case and Angus Deaton of data from the National Health Interview Survey. The horizontal axis is the age of the person being interviewed. The vertical axis is self-reported health status (SRHS), running (confusingly!) from Excellent at the bottom to Poor at the top. This means that a rising curve means declining health. Please keep this in mind in what follows.

Nevertheless, the pattern in the graph is clear. The top lines are the SRHS of women and men in the lowest income quartile, and the bottom lines are SRHS from the highest income quartile. If you are in the lowest quartile, you enter the work force with less health than those in the highest quartile. Moreover, as you age, your health declines more rapidly than someone in the highest quartile. The largest income-related difference in SRHS is at age 60.

So, people in the lowest income quartile need health care at younger age than do those in the highest income quartile. They are also likely to have to pay more for their health care as a percentage of their income. Finally, despite Medicaid, having less income means you are less likely to be insured (see the bar chart to the right, based on 2009 data from the U.S. Census). The point, then, is that a reduction in the age at which you qualify for Medicare will hurt those in the lowest income brackets much harder than those in the upper brackets, because they need health care at a younger age, they have to pay more of their income for health care, and they are less likely to have other health insurance. 

Important caveat: The Case-Deaton graph doesn't quite show the life course paths people take, because of a problem called age-period-cohort confounding. This means, among other things, that older people have less health than younger people both because they are older, and because they were young in a (possibly) less healthy historical era. For example, if early childhood nutrition was worse in 1940 compared to 1990, that may have adversely shaped the development of a person who is 70 and in poor health today.

Free to choose

by Paul Kelleher

Krugman opens today's column on the Wolf Blitzer-Ron Paul exchange thus:

Back in 1980, just as America was making its political turn to the right, Milton Friedman lent his voice to the change with the famous TV series “Free to Choose.” In episode after episode, the genial economist identified laissez-faire economics with personal choice and empowerment, an upbeat vision that would be echoed and amplified by Ronald Reagan.

But that was then. Today, “free to choose” has become “free to die.”

As so frequently happens in these contexts, the libertarian bogeyman du jour is contrasted with the presumably more enlightened of his/her libertarian brethren. Liberals really love to call upon Friedrich Hayek to prove that libertarians can support the social insurance state without renouncing everything that made them libertarians in the first place. Krugman does just that today, quoting Hayek's words of support for “a comprehensive system of social insurance” to protect citizens against “the common hazards of life." 

I have no problem with this move, as long as it serves as the beginning of a meaningful conversation about whether Hayek's statement is in fact true, and not just an argument from (libertarian) authority. In that spirit, I present you with a passage from the preface (pdf) of  Income from Independent Professional Practice (1945) by one Milton Friedman and Simon Smith Kuznets:

Why I Picked On Welfare Economics

As I understand it, economics is a science of efficiency. Now, efficiency is pretty meaningless if we don't first specify what we're interested in the efficient production of. In some contexts, efficiency will measured in terms of the net amount of some plain commodity, such as apples or loveseats or iPods. But it is also possible to adopt a higher-order perspective using some other metric that somehow incorporates other particular commodities. One branch of economics, welfare economics, purports to investigate the efficient production of human well-being. Since the consumptions of apples, loveseats, and iPods presumably all contribute to human well-being, welfare economics is a candidate framework for understanding the nature and requirements of efficiency from that higher-order perspective.

Yesterday via Twitter I claimed (in part) that "sometimes welfare economics is just silly." I caught some flack for this, and rightly so. Since welfare economics is a broad tent encompassing a diverse array of techniques and fundamental commitments, it is probably unwise to generalize about welfare economics as such (although I did say "sometimes"). Moreover, since each version of welfare economics presumably wears its commitments and assumptions on its sleeve, it is always open to the economist who practices it to say, "Hey, don't blame me. I'm just telling you what falls out of the algorithm given the axioms and the assumptions." Finally, since welfare economics, in all its versions, is an account of efficiency only---welfare economics as such entails nothing about public policy, unless the only consideration relevant to public policy is efficiency. Some economists seem to operate as if they believe this (I give an example below). But there is no reason why one should take this as an essential feature of welfare economics.

So perhaps what I should have said is that sometimes welfare economists are just silly. Let me give you an example.