Overcoming Bias

Overcoming Bias Has MOVED

Just go to OvercomingBias.com for the new posts.

Moving Hiatus

This evening at midnight Eastern Daylight Time, this blog will begin to move to a new hosting site.  For a smooth transition, please don't comment from then until 2am EST.

Tax Unhappiness?

Last week I described how Robert Frank wants to discourage "conspicuous consumption" via big houses, cars, and barbecues, because he thinks they cause large positional (i.e., relative status) harms on others.  After describing our best data on positionality, I asked if Frank would endorse its implication that we tax should beauty aids, education, and sports. 

Both Robert Frank and Geoffrey Miller eagerly point out when data shows folks being less happy doing what Frank and Miller dislike, but this doesn't obviously fit their big theories: it is not obvious why activities that produce positional side effects (Frank), or activities that are inefficient signals (Miller), should make us less happy.

There is a very different argument one could offer however, even if neither Frank nor Miller offer it directly.  One could argue that we should in general tax the activities that make people less happy.  To justify this we'd have to assume that humans tend for some reason to do too little of what makes them happy, but it doesn't require that we know exactly what that reason is.

Below is Table 6.1, from a 2008 paper on the "net affect" of 45 ways we spent our time.  (Net affect combines six emotions felt during the activity: happy, tired, stress, sad, interested, and pain.)  A simple policy of taxing low affect activities and subsidizing the others would have us subsidize not just parties, doing and spectating sports, exercise, playing with kids, walking dogs, and music, but also subsidize religion, eating out, and shopping.  We would tax not just work, commuting, home maintenance, and most housework, but also tax school and homework, writing by hand, buying medical services, doing personal medical care, and caring for adults. 

If you don't endorse these policies, you must have other policy criteria than just encouraging activities that directly make us happy.  That table:

Missing Liberaltarians

Officially, the libertarian world view is equally distant from standard liberal and conservative political views.  See for example, the World's Smallest Political Quiz, where liberals like social but not economic freedom, and conservatives like economic but not social freedom.  In practice, however, libertarians hang out more with conservatives than liberals. At least they do in the academic and think tank worlds I know.

I've been attending some "liberaltarian" dinners arranged by CATO's Brink Lindsey, where we discuss commonalities and differences of liberals and libertarians.  So I've been pondering why libertarians seem to connect more with conservatives than liberals.  Some say it is just a natural alliance of outsiders, but it seems more to me than that.  Some say it is because conservatives are more willing to adopt libertarian rhetoric in national politics, but that is just more data to explain.

Tyler Cowen's insight that ideology is mainly about who gets respect suggests an answer to me: libertarian heroes are more like conservative than liberal heroes.

• In the conservative view, we should most respect the pillars of local communities: dependable connected leaders who respect authority, do their job, help their neighbors, raise their kids, go to church, and go to war when needed. 

• In the liberal view, we should most respect passionate cosmopolitan subgroup activists: folks who identify strongly with an oft-disliked non-geographically-defined subgroup and who via sheer impressiveness of art or word gain their group a wider respect.

• In the libertarian view, we should most respect "self-made" men or women, able to achieve glory with minimal help from government, family, or community, if only such meddling outsiders would get out of the way. 

Libertarians support low taxes because individuals should be free to choose how their money is spent, rather than being forced to accept collective choices.  Conservatives support low taxes so that those who have worked hard for their money can show off the fruits of their labor and earn full respect for it.

Libertarians support gay marriage because individuals should be free to have whatever consenting relations they want.  Liberals support gay marriage because they want us all to officially respect gays as much as straights; gay activists have earned their group more respect.

It seems to me that libertarian self-made heroes are more similar to conservative community pillars than to liberal subgroup activists.  Self-made men are mostly not made in the bedroom; their glory shows more in their income than in their subgroup identity.

Perhaps some future Ayn Rand Two will describe new compelling libertarian hero characters, who are more like liberal activist heroes.  But until then I predict conservatives and libertarians will remain closer than official libertarian doctrine can explain.

Added 25May:  Andrew Gelman comments here, and Tyler says not to forget the villans, a great suggestion.  I should say this was an attempt to identify the signaling persona behind common ideologies, not the conscious rationalizations people give.  And that I'm not at all confident in this.  Also, it would have been better to say that libertarians support low taxes and gay marrriage because such policies allow self-made heroes to make more of themselves.

Reply to Moldbug

A Mencius Moldbug has written a confused and rambling 7400 word critique of futarchy. But since Mencius seems to have passion and potential, let me try to communicate.  Most readers may prefer to skip this post; it will get tedious.

Prediction markets are a fine idea, whereas decision markets are... well... retarded. ... Almost every conceivable application of a decision market ... does not produce accurate predictions.

So if PM good, DM bad, your complaints should focus on features that distinguish decision and prediction markets, right?

For a market to produce accurate predictions, there must be genuine experts in the market, and they must be substantially better-funded than the morons.

OK, except that morons may largely cancel each other, in which case you don't need as many non-morons.  But this issue applies equally to prediction and decision markets, doesn't it?

If no one has ever seen the Emperor of China's nose, can a prediction market predict its length? ... The worst case is that in which nobody has any way of actually calculating the prediction, but no one in the market is sure that this is the case. Your market signal will look exactly like that of an accurate prediction market, but predict nothing at all.

You can ask for a full probability distribution.  If speculators know they don't know anything, then they will give you a broad distribution that expresses a lot of uncertainty.  This is them telling you they don't know much.  In your nose example, they may just give you the distribution over nose sizes for elite Chinese.  And how is this issue different for prediction vs. decision markets?

Against Admirable Activities, Again

When my kids were young and played a new game, the pattern was clear: If they won, they liked that game and wanted to play it again.  If they lost, they didn't like that game.  Tyler Cowen once told me how this generalizes; the essential question of ideology is: who should be admired?  We tend to think it would be good for the world if policies and culture tilted a bit to more admire the activities that tend to make us look good. 

Such disagreements, however, shouldn't distract us from the fact that societies often agree quite a bit on what kinds of activities they admire.  For example, Frank and Miller's instinctive fear that the love of stuff obtained from distant soulless others corrupts one's soul is ancient.  Greek historian Herodotus ~430BC:

The Egyptians are divided into seven distinct classes—these are, the priests, the warriors, the cowherds, the swineherds, the tradesmen, the interpreters, and the boatmen. ... Whether the Greeks borrowed from the Egyptians their notions about trade, like so many others, I cannot say for certain. I have remarked that the Thracians, the Scyths, the Persians, the Lydians, and almost all other barbarians, hold the citizens who practice trades, and their children, in less repute than the rest, while they esteem as noble those who keep aloof from handicrafts, and especially honour such as are given wholly to war. These ideas prevail throughout the whole of Greece, particularly among the Lacedaemonians. Corinth is the place where mechanics are least despised.

I'm not exactly sure why traders have been so consistently disliked, though I suspect it has something to do with loyalty signaling.  But I am pretty sure that while societies consistently prefer to encourage more of the activities they admire, such choices often make them on net worse off.  I said two years ago:

Gottschall's Gotcha

Jonathan Gottschall reviewing Miller's Spent in Seed:

Miller may have made this final point a bit too well. I was not many pages into Spent before I found myself helplessly attuned to Miller’s own “narcissistic self-displays.” Miller reminds us frequently of his elite education, tells us that he owns several thousand books, lets on about his sophisticated taste in avant-garde art, makes offhand displays of his mastery of musical jargon (“timbral richness,” “isorhythmic motets,” “polyphony”), stresses his impeccable liberal credentials, and shows off his authentic verbal flair, his cosmopolitanism, and his soaring IQ (he argues —tendentiously —that elite university degrees function as covert IQ guarantees). So Spent functions not only as an attempt to popularize a vein of scientific research, but also as a means of selling the audience on the virtues of its creator: Geoffrey Miller—a smart guy, a bit of a Renaissance man.

There are two things to say about this. First, it is Geoffrey Miller, Renaissance man, who gives Spent so much of its winning personality, its narrative tang, and its consistent good humor. Second, Spent cued me in not only to its author’s self-marketing, but also to my own. For what is a book review if not—at least in part—a narcissistic self-display? What am I doing now, if not flaunting my penetration, my learning, my tough-minded yet charitable judgment, and—most narcissistically of all—my ability to take a decade of Miller’s life as a scholar, scientist, and close observer of American pop culture, and wrap it up neatly in a 1,200-word package—complete with an artful, preening flourish at the close?

Jonathan clearly "gets it."  Let me also admit: my blog posts are no doubt also designed, at least unconsciously, to signal my many features.

Spent = Gold + Schlock

On Sunday I reviewed Frank's Luxury Fever, which advocated taxing "conspicuous" relative to "inconspicuous" consumption.  I noted:

Frank offers no evidence whatsoever that the activities he dislikes and wants to tax in fact cause more inefficient status-seeking than the activities he likes and wants to subsidize.

But soon after I reported on data others have collected on this, and summarized their implications for policy.  This morning I gave the main argument from Geoffrey Miller's new book Spent: 

1. Signaling infuses most human activity.
2. Consumer capitalism marketers trick us into using unreliable signals.
3. We'd be better off to talk and customize more, and work and buy less.
4. Laws aren't the answer; let's make better social norms.
5. Let's also adjust a consumption tax to compensate for side effects.

On point 1, I completely agree; Miller's detail here is golden; spectacular really.  And points 4 and 5 make a lot of sense.  But points 2 and 3 are, alas, schlock.

Frank and Miller seem to dislike pretty much the same activities, but while Frank complains that we use them to "conspicuously" show off, Miller admits pretty much everything we do shows off.  Miller complains instead that showing off via purchases is "redundant" and "misleading" and not as "efficient," "rich," and "unique" as other ways to show off. 

Yet like Frank, Miller offers no evidence whatsoever that the signals he wants to discourage cause more inefficient harm than the signals he wants to encourage.  Specifically, Miller offers no evidence that:

Spent's Main Argument

Geoffrey Miller's The Mating Mind is probably the one book that has most influenced my thinking.  So I'll honor Miller by taking several posts to discuss his new book, Spent.  Yesterday I reviewed its predecessor, Robert Frank's Luxury Fever.  Today I give Spent's main argument, in Miller's own words:

1) Signaling infuses most human activity:

We are social primates who survive and reproduce largely through attracting practical support from kin, friends, and mates.  We get that support insofar as others view us as offering desirable traits that fit their needs.  Over the past few million years we have evolved many mental and moral capacities to display those desirable traits.  Over the past few thousand years, we have learned that these desirable traits can also be displayed through buying and displaying various goods and services in market economies. (p75)

2) "Consumer capitalism" marketers trick us into using unreliable signals:

The standard self-display strategy in most developed societies is to seek the highest-paying full-time employment permitted by one's intelligence and personality, and to use the resulting income to buy branded goods and services at full retail price. ... As a self-display strategy, it is very inefficient. ... Almost every other way of acquiring and displaying human artifacts or experiences sends richer signals about one's personal qualities. (p257) ...

Buying new, real, branded, premium products at full prices from chain-store retailers is the last refuge of the unimaginative consumer, and it should be your last option.  If offers low narrative value - no stories to tell about interesting people, places, and events associated with the product's design provenance, acquisition, or use. It reveals nothing about you except your spending capacity and your gullibility, conformism, and unconsciousness as a consumer.  It grows no physical, social, or cultural roots into your local environment.  It does not promote trust, reciprocity, or social capital.  It does not expand your circle of friends and acquaintances.  It does not lead you to learn more about the invention, manufacture, operation, or maintenance of the things around you. ... The alternatives listed above try to minimize retail spending not just to save money, but to maximize trait display power. (p270) ...

Against Makeup?

Though yesterday I complained:

[Robert] Frank offers no evidence whatsoever that the activities he dislikes and wants to tax in fact cause more inefficient status-seeking than the activities he likes and wants to subsidize.

A few hours later I reported that others have published evidence on which particular goods and activities induce more inefficient status efforts. So now the key question is:

Will folks like Frank consistently apply these results to recommend taxes or subsidies to reduce wasteful positional effort, wherever such waste may lie, or will they selectively cite only results favoring pre-existing political positions?

Some policy implications of this data on positional effects seem easy for folks like Frank to swallow.  While our evidence so far is tentative, it does seem to support subsidizing health, including air quality and device safety (though not necessarily medicine if it has little relation to health).  It also supports subsidizing insurance more generally; bads being less positional than goods gives a new reason to avoid inequality from bad outcomes, such as crime.  Our evidence also supports taxing work relative to leisure, though since we already have large taxes just like this, this evidence does not obviously support larger work taxes.

Many other policy implications, however, seem much harder for Frank to swallow.  Since sport effort seems especially positional, should we tax sports, instead of subsidizing them as we often do now?  Since education seems to be at least as positional as income, should we drastically reduce educational subsidies, or even tax it?  And since government spending seems far more positional than income, shall we greatly reduce our unprecedented levels of such spending? 

Perhaps Frank would suggest that other compensating side effects justify vast government spending as well as sport and education subsidies.  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.  Will folks like Frank at least agree that severely taxing beauty aids is one of the clearest policy implications of our evidence on positional effects?

Also, we observe a huge amount of variation in who sees what to be how positional.  This suggests that perhaps policy can influence this distribution of envy.  Shouldn't it be a top priority to find ways to influence folk's attitudes toward the gains of others, so that they don't feel as envious of such gains?

Added 19May: Actually the wording of these surveys has them capturing folk's expectations about all side effects created by these choices. So it is mostly side effects of which these folks were unaware that could compensate for these results.