« Mea Culpa: Positionality Data | Main | Spent's Main Argument »

May 18, 2009


Will folks like Frank at least agree that severely taxing beauty aids is one of the clearest policy implication of our evidence on positional effects?

What about you, will you adopt that position now you've seen the evidence on what is positional? You seem very much motivated to find reasons not to like positional taxes, so you highlight the evidence that leads to unpalatable consequences, but since your original argument against positional taxes based on a lack of evidence about what is positional no longer holds, will you revise your position?

(Incidentally, I *think* I can see a better way to conduct experiments on what is positional, that don't ask people to make explicit the extent to which their preferences depend on what everyone else gets, but I'll wait until I have the details worked out before I post)

But what about personal beauty, which our evidence suggests is one of our most positional goods?

I find this quite puzzling. I like to see beauty in architecture, gardens, artwork, and of course people. If offered a deal where I could make everyone else more beautiful while leaving my level of beauty fixed, I would definitely take it (even for my own sake). Doesn't this hint that beauty is not very positional (and indeed that it has positive externalities)? Or are my preferences just very different to the norm?

And since government spending seems far more positional than income, shall we greatly reduce our unprecedented levels of such spending?

Why do you think government spending is so positional? What am I missing?

Taxing beauty aids seems to be contingent on whether they result in inflation or compression. That is, if people using beauty aids are all proportionately more attractive, they add little. If, however, there is actual convergence to a certain level of attractiveness, then they seem potentially useful, depending on the initial distribution, particularly if being below average is worse than being comparably above average. If it's worse to be behind then ahead, and there's a fairly normal distribution, convergence is going to generate positive net utility, so it's unclear why you'd want to discourage it.

Also, if reducing current consumption is perceived as a loss, personal attractiveness is much less positional.

Of course, the other, bigger problem would be that taxing beauty aids might actually *increase* consumption (or, y'know, generate a black market, if it's high enough to be a real deterrent). Currently, wearing lipstick just makes your lips look better. But if it made your lips look better AND showed that you can afford lipstick, lipstick is of even higher positional value. This may be more of an offset than a reversal, but it's still a potentially large problem. (This would probably also hit poor people in poor communities harder than anyone else - http://knowledge.wharton.upenn.edu/article.cfm?articleid=1963)

Though admittedly I agree it would be useful to reduce consumption of such products, taxes seem rather ill-suited to accomplish that.

Paul, I will consistently follow the evidence in recommending taxes and subsidies. Of course the weight I apply to this evidence may not be that high for a while, while it is new and tentative. And I will post a lot more here on how it influences my policy.

Toby, the evidence clearly shows wide variance in who finds what how positional. I'll post more on why govt spending is positional.

psycho, the usual story is that beauty aids make it easier to distinguish among beauty levels. Many of them are already prohibitively expensive.

Could you explain in one of your posts what you mean by positional, and the context of the debate?

Is there some widely-accepted theory that status-signaling desires should be discouraged? The debate about which pursuits are "status-seeking" rather than "private pleasure-seeking" seems really silly to me - why would I care whether you prefer to increase your status by working hard and buying a fancy car vs slacking off and driving a beater. If anything, I gain more from trading with status-seekers - most such output is far more easily traded than one's leisure time.

Any externality (your gain in status by definition lowers everyone else's, since it's all relative) seems like an error on the part of the consumer of status, not the seeker of such. What we should tax is envious feelings, and we should subsidize happiness at others' wealth and good fortune.

A test of how much someone believes in their model is whether they accept implications of that model that go against their general world view.

There are a huge number of very smart Christians and Muslims. Consequently, for Robin and Eliezer a test of how much they believe in the "disagreement is disrespect" and "you can't agree to disagree" theories is whether they believe that there is a significant chance that the Bible and/or Qur'an provide greater insight into the physical world than does the work of scientists.

I say we appoint frank as handicapper general.


In my experience, very smart Christians and Muslims are much more inclined to accept the scientific evidence for the age of the earth and for evolutionary history than their less intelligent co-religionists. They also do not (typically) view the Bible and Quran as scientific textbooks, but rather as expressing cultural and psychological truths and guidance for humanity.

James, believers reject rationality. The agreement theorem concerns for rational people. The clue to your paradox is simply that rational atheist indeed disrespect theists. Nothing wrong with that.

Matthew - True most do accept science, but enough don't so that the "disagreement is disrespect" theory says you should give some weight to the possibility that they are right.

Arthur, most believers don't believe they reject rationality. Also, shouldn't a rational person give some weight to the views that other smart people have even if these views "seem" crazy. Therefore by definition a RATIONAL atheist can't completely dismiss the religious views of theists.

Government spending (on things like national defense or space exploration) is positional between countries, which means that reductions based on positionality don't make sense unless we get other countries to agree to the reductions.

Similar, sports are most obviously positional between teams within a single league (where a win for one team is a loss for another team). For an individual fan, sports consumption (e.g. the amount of games that you watch) doesn't seem especially positional. Agreements between teams within a league could make some sense (and sports leagues do have agreements to promote the interests of the league as a whole), but it's not clear to me what role Robin is seeing for government-imposed taxes or subsidies.

Education is not an ordinary consumption good because education makes people more productive. A more educated workforce is a more productive workforce, so the country has an interest in subsidizing education (even if the person being educated sees it largely as a positional game). (Similar arguments apply to some of the other goods, like government spending on health care research or foreign aid, which produce benefits to people other than those who are providing the funding.) There are also concerns about fairness or equality of opportunity - much of the subsidy for college education is there to give poorer people a chance to compete.

Finally, I think that it's misleading to describe this in terms of "envy," since I don't see any evidence that this emotion plays a major role in positionality. Much of the benefit of being beautiful, such as being treated well by others and being a desirable romantic partner, arises from being relatively beautiful compared to other people rather than from your absolute level of beauty. That's all it takes for beauty to be a positional good, and it doesn't require anyone to feel envy.

James, believers demonstrably reject rationality. You certainly should give some weight to other people's idea but that by itself doesn't imply convergence or agreement.

James, widespread religious beliefs do in fact weigh in my mind in favor of those beliefs. My claim is no that one should never disagree with anyone; that is impossible when others disagree among themselves.

Unnamed, yes, we would want international treaties to reduce national spending. It is not obviously in my interest for you to engage in an activity that makes you more productive. People in different sports compete to be impressive and gain fans.

Isn't politics itself much more positional than not? Politicians who spend more or hold more power in general seem to have status advantages over those who do not. I'm not sure how this affects the use of politics to reduce positional consumption, but I think its something to think about.

I like looking at beautiful women, regardless of how many other beautiful women might exist. I don't think beauty aids are nearly as positional as, say, expensive sports cars. Even among those few of us who appreciate expensive sports cars mostly for their performance, racing said cars is often done under governance which drastically reduces expenditures on positional upgrades.

Your examples are problematic because there are externalities as well as signaling effects involved. Beauty has a positive externality as long as you aren't competing against the people, and obviously useful education does as well.

What we could start taxing however, and which Mr. Frank might be instinctively averse to doing, are indie films and indie bands, and books on the NYT bestseller list.

But what about personal beauty, which our evidence suggests is one of our most positional goods? Yes, exercise also improves health, but it is very hard to see any large compensating side effects justifying makeup, hairdressing, and nice clothes.

Suppose the interviewees interpreted the beauty question as

a. You score 8 on a scale of 1 to 10; others of your gender score 10
b. You score 6 on the same scale of 1 to 10; others of your gender score 4

In this case, there is a large compensating side effect: the appreciation of members of the other gender.

Robin, do you know if Frank counts competition between producers to attract consumers (e.g. athletes attracting fans) as positionality? The examples that I've heard (and the quote in yesterday's post) all focus on consumption - consumers competing to do well relative to other consumers. Competition between producers isn't a problem if it leads them to produce a better product (though it depends, I suppose, on whether consumers find the product to be better in an absolute sense or only in a positional sense).

I think taxing status signaling spending is one of the best policies a government can implement. Just like taxing anything else with major negative externalities, and subsidizing things with major positive externalities.

However, makeup is really cheap, and it's a fuzzy category (makeup vs hygiene), so taxation would be hard to implement. One of the most expensive categories - surgery - is indirectly taxed already - because normal surgery is subsidized cost of surgeon services on the market are higher than they would otherwise be. So it's not really terribly great idea to tax it.

The same applies to other kinds of small categories of status products - its too expensive to organize a proper tax for them.

Sports competition performance is already treated like an extrernality, not by taxes but by outright ban on many performance enhancements (positional good) with serious negative health consequences (non-positional bad).

It's not obvious that education has negative externalities, most people believe its externalities are very strongly positive.

Tomasz, we spent lots on hairdressing and beauty-enhancing clothes. The main positional costs of sports are time and effort, and these are not taxed at all.

Even if we can agree on some activities to tax, we still have to consider what happens when we make one or several status signals more expensive. Instead of decreasing overall status-seeking, individuals may simply shift to less efficient and informative status signals, perhaps increasing the total cost and quantity of signaling. I do not see how Frank can answer this short of a tax and subsidy schedule for all conceivable activities.

Dagon: if all I care about is having a better barbecue than you, and I have $5000 to spend while you have $2000, then I don't care if high-end barbecues are heavily taxed, because even if my $5000 only buys $2000 worth of barbecue, it will still be better than yours. So the Government can enrich itself while minimizing the harm it does its citizens. That's the theory; it's not to do with discouraging it, but with extracting money in a way that causes less pain.

The primary argument against spending on positional status goods seems to be that, due to the inherently zero-sum nature, it induces wasteful spending, displacing other things, correct?

However, as has been mentioned, some positional goods (education, beauty, &c.) potentially have positive externalities. If the magnitude of these externalities is great enough, wouldn't it actually make sense to, if anything, encourage positional spending on them, since this will lead to people voluntarily spending additional resources creating said positive externalities? A society of well-educated, beautiful people seems strictly better than a society of less-educated, less-beautiful people with the same relative status positions.

See my added to the post.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30