Deserving It

Bryan Caplan says he was the only libertarian at our recent Liberty Fund willing to defend the free market on grounds of desert. That's not true! It's too bad that Dave Schmidtz missed the first part of the desert discussion, since he has a whole (brilliant) section on desert in his new book Elements of Justice. And I, like Dave, am willing to defend the free market partially on grounds of desert. I think Bryan's confusing accepting Olsaretti's particular terms of justification with accepting some desert argument or other for the market. Olsaretti thinks, absurdly to my mind, that justifying the market requires justifying the differences in the distribution of income. When she asks whether desert can justify the free market she is asking whether income differentials across the entire national economy are deserved. I think this questions is like, as Elaine Sternberg put it, asking whether the market is ten feet tall.

I think desert can justify differences in distribution when people are working in a common cooperative enterprise. If you and I do a job that neither of us could have done alone, but I simply worked harder than you, then I deserve more than you. That is a bedrock principle of moral common sense. And people who add more value inside a firm deserve bigger raises and more frequent promotion than people who do less. Etc.

And there is a sense in which entrepreneurs or firms competing in the same sector can deserve to do better of worse. If I offer you a better product at a lower price than my competitor (if I offer terms of cooperation that includes a bigger surplus together with a bigger share of that surplus for you), then I deserve to be your partner in cooperation. Businesses that are more deserving of their customers, by offering better terms of cooperation, will tend to do better than those that offer slightly worse terms. If Bryan and I are are competing for customers, and I succeed by offering better terms, I think there is a pretty obvious sense in which my larger income will be a reflection of my desert. However, if Bryan is trying to sell hot dogs, and I am trying to sell little Buddha statues, and we both outcompete our rivals, but I end up making more than Bryan because of Angelina Jolie is seen in a movie fondly stroking a Buddha exactly like the kind I sell, then I think it is just as obvious that our differential will not be a reflection of my desert. Of course, entrepreneurs who correctly predict an increase in demand may in some sense deserve to do well. But it pretty quickly becomes incoherent to justify the size of income gaps across even a very small, not very complex economy in terms of desert.

That said, I'm open to Bryan's idea that there could be a significant positive correlation between some kind of deservingness or other and position in the income distribution, and that the correlation is stronger under free market institutions than under any alternative. To show that free-market distributions are more desert-sensitive than the alternatives would be neat. But demonstrating the desert-sensitivity of an entire distribution under certain institutions is rather different from justifying particular income gaps in terms of desert (e.g., Gates's deserving the differential between him and Buffet, or Bryan deserving the differential between him and me.) Actually, I think a comparative study of desert-sensitivity is a pretty awesome idea, though I wonder if even this is coherent in the end. Now, I don't know how you'd actually measure that, especially on Olsaretti's terms (she tries not to make a silly Big Bang argument—the Big Bang happened so nobody deserves anything—but seemed to me to make one anyway), but I think would be worth thinking through.

Let it not be said again that I am not a defender of desert! Here is my TCS essay on luck egalitarianism, and follow-up blog posts.

Liberty, Desert, and the Market

I had the good fortune this weekend to attend a Liberty Fund Socratic Seminar centered on Serena Olsaretti's Liberty, Desert, and the Market, which is a first-rate example of scholastic analytic political philosophy that attempts to arrive at strong normative conclusions about the justice of political institutions by analyzing extremely abstract concepts in a psychological and institutional vaccum. Truly, if you can take five or six contestable normative concepts, give all of them your favored gloss, line then up in just the right way, then you can get anywhere. My very favorite move is assuming your conclusion (this is always useful), using it to place “desiderata” on the analysis of a concept in question, and then pretending progress when you find that an account of the concept that satisfies your stipulated desiderata supports your conclusion. It's almost like not begging the question. Anyway, none of this is a problem with Olsaretti's book, which truly is a good example of its type. The problem is the kind of analytic political philosophy that Olsaretti, and lots of others, are taught to perform.

My biggest overall gripe is about the foundational assumptions of this kind of work—the framework stuff there is no argument for. Olsaretti wants to contest defenders of the market taking “the moral high ground.” She then basically identifies taking the moral high ground with justifying inequalities in the distribution of income and wealth. (She concedes that there are efficiency justifications of the free market. But that is either amoral, or the moral low ground. The fact that she thinks questions of distribution, as opposed to something else, define the high ground says a lot about her moral convictions, I suppose.) A lot of Rawlsian ideas pop up here and there. But they come out of context. Rawls at least says what the framework of redistribution in his theory is: an imaginary nation-state closed to the movement of people and capital across borders. He is sometimes clear about what is being distributed: the surplus from cooperation. Even with isolationist assumptions, it's not obvious that the state is coextensive with the relevant network of cooperation within which surpluses arise. These problems dog Olsaretti like crazy, since she is even less interested than Rawls in thinking them through. (She actually does not seem that interested in the idea of cooperation for mutual advantage, which is the most important idea there is.)

I kept finding myself wondering things like:

Suppose there are two unrelated networks of cooperation: A and B. Each has three members. The members of A cooperate and create a surplus of 99 units. The members of B cooperate and create a surplus of 75 units. Each member of each group agrees that “equal shares” is in this case the fair principle of distribution. So everyone in A gets 33, and everyone in B gets 25. Olsaretti then might ask “What justifies the difference in income between a member of A and member of B.” The answer is, nothing does. But that's not because it is unjust that I get 33 and you get 25. It is because we are not in this case in the circumstances of justice with respect to one another. What I get needs to be justified relative to what others in my cooperative network get.
Suppose your network is on Mars and mine is in Minnesota. In that case, Olsaretti doesn't raise the question of justification. Now suppose your network is in Canada. Still, apparently no need for a justification (even if my network is ten feet away from yours in Manitoba; we could shake hands across that imaginary line, but that imaginary line is important!) But Olsaretti seems to think that if your network is New Mexico, inside the same national boundaries as Minnesota, then, mysteriously, asking to justify the difference in holdings between members of two seperate cooperative networks is not like asking whether an F# is salty.

Now, I suppose that a market is a complex web of overlapping cooperative networks. You and I may in some circumstances be part of the same network, sometimes not, whether or not we have the same kind of passport. Are the differences in our incomes deserved? Fair? Just? Is chocolate quiet?

We could start like this. Assume a single, discrete cooperative network. Allow benefits to be distributed dynamically according to a constantly shifting set of agreements regarding local changes in distribution. Stop the clock. Now, assume a morally wise central planner who can rearrange benefits simply by blinking, and ask: Does the global distribution of benefits correspond with any plausible normative principle that a morally wise central planner might use to distribute benefits? The correct answer will be probably be “No.” Cool. Then we could proceed to ask questions like: are real political units ever discrete cooperative networks? Is a central planner an institution? Is there an institution that is a reliably proxy for both the moral wisdom and practical efficacy of the central planner? Is there a relevant normative principle for the central planner to employ? If not, is that because morally wise central planners do not make a lot of category mistakes? Etc. . .

Class, Education, and Meaning Manufacture

I was just talking to Brink about Annette Lareau's book Unequal Childhoods about the differences between the rearing and education of middle class and working class kids. This got me thinking, naturally, about the transformation of labor markets. People raising their kids to be cheap labor are having and will continue to have problems. Clearly middle class kids who develop human capital relevant to an information economy will do better. But a lot of the information economy will be automated or outsourced eventually as physical and human capital improves in China, India, etc. As Ed Leamer draws out in his entertaining and smartifying review of The World is Flat the less mundane and codifiable a job is, the less competition there is going to be for it (by both man and machine.) Here's his list from more mundane and codifiable to less:

·  Type this page.
·  Edit this page.
·  Write an article for an Economics journal.
·  Write a good joke.

One way to read this is that the really indispensible folks are those in the business of manufacturing meaning. My job is a meaning-making job in a pretty literal sense. I am here to help people understand what all this thinking other people have done means. (In the process, you end up having and conveying some new, meaningful thoughts.) But meaning manufacture is broader than that. It's basically ideas and aesthetics. If you can write a moving novel or ideas, a kickass rock opera, or a gutbusting monologue you are not about to be replaced by a robot or a Chinaman. Of course, only a small number of people will ever be able to do that sort of thing. But that small number gets bigger all the time as the number of people with the money to buy meaning gets bigger.
In any case, I don't know about you, but I don't want my future kids to be middle managers any more than I want them to be auto workers. Or even doctors or lawyers, unless they really wanna be. And I just don't like the hyperprogrammed minivancentric middle-upper class childrearing style. I'm not sure that's the best for building the human capital for future meaning makers. That's one reason I plan to home school my children. I fear their potential will be squandered in even the best schools. I'd like my kids to be educated not just to show up on time, effectively navigate bureaucracy, and compete well in social word/power games, but to creative and entrepreneurial with the ability to make the kind of meaning that other people need. Of course, I think kids should grow up to be whatever they want. But we all know the kind of education you get is going to have a pretty big effect on what you want. And it seem to me that responsible parenting is partly about trying to get your kids to want things that will enable them to have the best possible life. Can you get ahead of the curve by educating your kids to be poet entrepreneurs? Are today's middle class culture tomorrow's equivalent of today's working class values. Will the “cultural creatives” constitute a new and distinct class? Should you try to get you kids in?

Egalitarianism, the Entry

I had forgotten, until just now, that I wrote the original Wikipedia entry for “egalitarianism” way back in 2001. How's that for Internet philosophy geek cred?! It's pretty interesting to see that though there have been about 500 edits, the conceptual framing of my original mostly remains, along with a good bit of the original boring prose. Yay path dependency!
My first try could have been more rousing. But I give myself credit for not simply writing “First!”

Brainstorm on Positional Domination

This is not an argument of any kind. I’m not trying to make a point. This is thinking out loud. And you are going to help me.

Anne and Betty each prefer to positionally dominate the other—they both like coming in first better than coming in second. However, each has a different hedonic payoff from positional domination and subordination. How do we think this through?

Anne

Betty

A

1st;  1000h

2nd; 800h

B

2nd; 900h

1st; 900h

Which world state, A or B, does a benevolent planner choose?

In Pareto terms, the planner is indifferent. Both Anne and Betty prefer to come in first, but both can’t.  The Benthamite planner is also indifferent: same sum of hedons.

What about willingess to pay? Well, pay for what? Positional domination or hedons? When Layard says that people undermine their own welfare by seeking status despite the fact that it doesn't make them "happy," it sort of sounds like he’s saying that people sometime value status more than they value hedons, but are wrong to do so. Let’s get rid of the normative judgment and think about what it could mean to value positional domination independent of hedons.

Suppose that whoever pays most to positionally dominate positionally dominates. You can have an auction. Anne’s highest bid is $2000 and Betty’s is $2700. So does that mean that Betty gets a bigger hedonic payoff from dominating. No, by the stipulation of the matrix, she doesn’t. (And by stipulation of the matrix, the distribution of hedons is not the dimension of positional competition.) So does this mean that Betty is willing to pay more for a hedon? Maybe, maybe not. Why think Betty is bidding on hedons? It could be that Anne and Betty value the marginal hedon at the exact same rate. In which case, Betty is just bidding for positional domination, which she values for its own sake, not for the hedons that fall out of domination.  

Or maybe you can think of the choices between A and B as choices between packages of hedons and positional domination, which are independently valuable, but causally connected. This is value pluralism. There are lots of independent values: hedons, positional domination, etc. The value-to-money and money-to-value exchange rates may not be the same for each value. (Or in each direction; the endowment effect for a hedon and a dollar may be different. Misers may trade hedons for dollars, on the assumption that the dollars will pay off in even greater hedons, but, when the time comes, they are unwilling to give up dollars for hedons, so the dollars just accumulate.) And the money worth of some values might decline on the margin faster than others.

Suppose that there is a point of hedonic saturation (I believe this is true.) At the point of saturation, an extra hedon will have no money value, since there is in some sense nowhere to put another hedon. (Hedonically saturated states dry up pretty quickly though.) Suppose that positional domination doesn’t saturate, and remains ever valuable. It is never enough to be mayor, or governor, or president, or ruler of earth; there is always value in dominating on another positional dimension, or dominating a dimension of broader scope. So, one could be hedonically saturated, and unwilling to pay for another hedon, but not be positionally saturated, and perfectly willing to pay to become Generalissimo of the solar system.  

Suppose the willingness-to-pay planner chooses world state B on the strength of the higher money value of positional domination to Betty. Is such a planner really benevolent? If coming in second is a positional externality imposed on Anne, what is the value of the externality.

Are you confused yet?

I am.

I have also caused myself to wonder whether it might be possible to take money away from people under one description, and give it back to them under another, resulting in a net gain in hedons Could be! The issue wouldn't be the transfer per se, but the description under which the transfer takes place. (This has nothing obvious to do with the above.) Does support for the state depend on a kind of gratitude stemming from an (illusory for most people) sense that the value of public goods consumed is greater than taxes (direct and indirect) paid? It's like thinking somebody's your best friend since that gave you $4 bucks after they took $5 out of your wallet but gave you the $4 so warmly.

</brainstorm>

[Cross-posted to Happiness & Public Policy, which is back online.]

Funny Typo of the Day

From Bart Schulz's outstanding SEP entry on Henry Sidgwick:

Sidgwick's versatile and many-sided intellect—not to mention his keen wit—are typically better displayed in his essays and letters than in his best-known academic books. He was in fact much loved for his gentle humor (or “Sidgwickedness”) and sympathetic conversation, and his philosophical students prized him for his condor.

Prestige, Status, and Culture

A number of works I've seen on the importance of social comparison extrapolate in a fairly simple way from the existence of non-human dominance hierarchies to human status. (Frank, for instance, motivates his view of status by citing the general logic of competition for mates.)  Many go further and identify human status largely with position in the income distribution. Both moves are mistaken.

It's fair enough to point out that humans are primates, and so we should see some continuity. But humans also have language, higher cognitive abilities, and complex, cumulative cultural transmission and evolution. Humans, like other primates, eat. But human eating takes place in a rich cultural context, and the expression of human eating behavior is thickly culturally mediated. Like chimps, we like meat. Unlike chimps, we worry about eating it with the "correct" hand, or utensil, etc. Some of us won't eat certain foods because of culturally transmitted dietary taboos. One should expect that human social comparison and positional competition, even if it is universal, will also be thickly mediated by culture, and will be expressed in different ways in different cultures.

Furthermore, human status need not be a homologous to dominance. Here is anthropologist Joe Henrich in his paper with Franscisco Gil-White, "The Evolution of Prestige: freely conferred status as a mechanism for enhancing the benefits of cultural transmission." (Evolution and Human Behavior, 22, 1-32):

Although nonhuman status is still poorly understood, a single process appears at least strongly predominant: agonism (aggression, intimidation, violence, etc. — that is, force or force threat). The resulting social assymetries are referred to as "dominance hierarchies" in the ethological and behavioral ecology literatures. The privileges that accrue to dominant individuals are (1) in males, preferential reproductive access to females, food, and spaces, as well as disproportionate amount of grooming from others; (2) in females, preferential access to food and spaces, and disproportionate grooming. Despite some controversy, the evidence suggests that dominance correlates with fitness. (Cowlishaw & Dunbar, 1991; Ellis, 1995). The stability of dominance is often reinforced through "reminders": submissive behaviors (e.g., grooming, submissive displays, yielding space, etc.) from subordinate to superior, whether or not induced through intimidation by the latter.

In humans, in contrast, status and its perquisites often come from nonagonistic sources—in particular, from excellence in valued domains of activity, even without any credible claim to superior force. For example, paraplegic physicist Stephen Hawking—widely regarded as Einstein's heir, and current occupant of Newton's chair at Cambridge University—certainly enjoys high status throughout the world. Those who, like Hawking, achieve status by excelling in valued domains are often said to have "prestige."

Henrich's argument is that prestige is a feature of the human cultural capacity, which is adaptive because it saves the cost of having to learn everything yourself. I'm going to just continue to quote as at length because this really makes sense to me:

Once some cultural transmission capacities exist, natural selection favors improved learning efficiencies, such as abilities to identify and preferentially copy models who are likely to possess better-than-average information. Moreover, selection will favor behaviors in the learner that lead to better learning environments, e.g., gaining greater frequency and intimacy of interaction with the model, plus his/her cooperation. Copiers thus evolve to provide all sorts of benefits (i.e., "deference") to targeted models in order to induce preferred models to grant greater access and cooperation. Such preferred models may be said to have prestige with respect to their "clients" (copiers).

The above implies that the most skilled/knowledgeable models will, on-average, end up with the biggest and most lavish clienteles, so the size and lavishness of a given model's clientele (the prestige) provides a convenient and reliable proxy for that person's information quality. Thus, selection favors clients who initially pick their models on the basis of the current deference distribution, refining their assessments of relative model worth as information becomes available through both social and individual learning. This strategy confers a potentially dramatic adaptive savings in the start-up costs of rank-biased social learning. Finally, because high-quality information ("expertise," "performative skills," "wisdom," "knowledge") brings fitness-enhancing deferential clients, models have an extra incentive to outexcel each other. 

Because status-as-prestige isn't agonistic, there are clearly market-like gains from positional competition in a domain. There may be a fixed number of clients, and so competition for clientele may be zero-sum. But clients adopt models and defer to them because that makes them better off, not because of a threat. And insofar as reputation tends to tracks information quality, deference will be deserved.  

(The adaptive advantages of cultural transmission necessarily include  the danger of the success of maladaptive ideas. Boyd and Richerson explain why this must be the case in detail in Not By Gene's Alone. On Tuesday, I heard a good talk by Bob Subrick on how witch doctors in Botswana have successfully undermined WHO education efforts to contain HIV/AIDS by promoting widely believed false folk theories about the cause and transmission of the disease. This clearly involves a maladaptive allocation of prestige.)

So, the fact that we have a cultural capacity at all makes space for non-agonistic comparative advantage-based prestige/status. It would seem to me to follow that market systems, by promoting the refinement of the division of labor, promotes the multiplication of dimensions of excellence and therefore prestige. It is possible in market systems to gain the benefits of prestige and clientele from becoming a highly desired graphic designer, marketing consultant, or musician. 

Additionally, the fact that we have a cultural capacity is going to imply that, unlike other primates, status competition is going to be highly mediated by culture. It is possible to gain status among Mormons by being a good Mormon. There may be heated competition, even bitterness, over who is the best Mormon in the ward ("She's not really that good a mother!" "I paid my tithing plus five percent!" ), but such positional competition may serve on the whole to reinforce generally socially constructive norms. Indeed, it seems likely that one of the ways we judge the quality of a culture is by the way it mediates and channels potentially harmful universal human dispositions, such as status-seeking and tribalism. Mormon culture mediates status-seeking in generally beneficial ways. Redneck and ghetto culture doesn't.

The degree to which our place in the distribution of income/wealth is going to correlate with our status depends on culture. Some cultures and sub-cultures are more materialistic than others. Some are pointedly anti-materialistic. It's worth pointing out that comparative excellence style prestige is pretty clearly going to correlate with greater earnings in general. The higher the prestige in a domain, the fiercer the bidding from clients for access. That doesn't have to mean higher incomes, but on the whole it will. In this kind of case, higher relative position in the income/wealth distribution will be tracking, more or less loosely, excellence/prestige on some other dimension. Our high relative position on that dimension may make us both happy and high on the income/wealth distribution. But that doesn't imply we care much at all about the income/wealth distribution. (The world's best guitarist may have a lot of guitars, but he's getting more out of his status as the best guitarist, not as a guy who has a hell of a lot guitars.)

What is going to count as a positional externality is going to depend on what kind of position people care about. That's a matter of the kind of culture they're embedded in. It strikes me that policy types ought to take a step back and be willing to think about whether cultures and sub-cultures are in general peaceful, healthy, stable, and mutually beneficial. My colleague Jude Blanchette today gave me Alesina and Fuchs-Schundeln's paper on the effects of communism on people's preferences. They find that the longer people spent in communism, the more they prefer a heavily intervening state. What are we to make of that? Give them what they want because that's what they prefer? Or ask whether a system that leads people to want that is a good one to have? Similarly with cultures and subcultures. If a culture mediates status-seeking in such a way that certain acts of aspiration, success, and upward mobility might plausibly be understood as a kind of negative externality, should we plunge in and attempt to "rectify" the externality and push for efficiency relative to those culturally laden preferences? Or should we stop and ask if there is anything we can do to ensure that a culture that engenders these preferences does not long survive? 

[Cross-posted to Happiness & Public Policy.]

If You Would Like to Fund an Interesting Study, Call Me!

Thinking about the “fallacy of asymmetric idealization,” it occurred to me that it would be interesting to give personality, IQ, and heuristics and biases tests to high ranking U.S. bureaucrats and, better, Congresspersons. Would we discover that these are people who are especially good at decisions?

Well, no. We wouldn't. But it would be nice to know that, wouldn't it? And, anyway, this isn't the important thing. The important thing is the institutional structure—the incentives. What would be really interesting is a set of experiments that compared extremely intelligent people playing games that mimicked the incentives within bureacracies and in politics, versus below average people playing games that mimicked the incentives of well-designed market mechanisms. I think I know how this would turn out, too. But it would be useful to have experimental confirmation that no matter how smart and well-informed the bureacrats are, the quality of their decisions in the aggregate when acting in reponse to the incentives of political institutions is not in general likely to equal those of much simpler folk acting within good market institutions.