Tag Archives: Social Science

Hail The Radiation Model

By Robin Hanson · April 12, 2012 10:30 am · Comments (18)

We have a revolution in how to best predict transport and commuting rates:

The gravity law is the prevailing framework with which to predict population movement, cargo shipping volume, and inter-city phone calls, as well as bilateral trade flows between nations. Despite its widespread use, it relies on adjustable parameters that vary from region to region and suffers from known analytic inconsistencies. Here we introduce a … radiation model [that] predicts mobility patterns in good agreement with mobility and transport patterns observed in a wide range of phenomena. (more)

The gravity law assumes travel is proportional to the product of powers of the to and from populations, divided by some function of the distance between them:

The gravity law assumes that the number of individuals T_ij that move between locations i and j per unit time is proportional to some power of the population of the source (m_i) and destination (n_j) locations, and decays with the distance r_ij between them as

T_ij = m_i^a n_j^b / f(r_ij)

where a and b are adjustable exponents and the … function f(r_ij) is chosen to fit the empirical data.

The radiation model fits better by instead looking at how many people live closer than the destination location:

Step one, an individual seeks job offers from all counties, including his/her home county. The number of employment opportunities in each county is proportional to the resident population. … We capture the benefits of [each] potential employment opportunity with a single number, z, [independently and] randomly chosen. … Step two, the individual chooses the closest job to his/her home, whose benefits z are higher than the best offer available in his/her home county. … We denote with s_ij the total population in the circle of radius r_ij centred at i (excluding the source and destination population). … The radiation model is

T_ij = Ti m_i n_j / (m_i + s_ij)(m_i + n_j + s_ij)

… T_i … is the total number of commuters who start their journey from location i.

Amazingly, this better fitting radiation model only depends on distance indirectly, via population density. It suggests that while distance matters, it is almost never an overwhelming consideration. In the modern world, while political barriers are often insurmountable, distance is detail.

GD Star Rating
loading...

Tagged as: Social Science, World

Continuous Cooperation

By Robin Hanson · March 16, 2012 8:00 am · Comments (10)

In a prisoner’s dilemma, two sides have an incentive to defect, even though mutual defection is worse for both sides than mutual cooperation. It is well known that in theory and in reality people cooperate more when then expect to interact over more repetitions, and when they care more about the future.

It is hard to make people live longer, or care more about the future. It can be just as helpful, however, and often much easier, to make people interact more frequently. In the limit of continuous interaction, people should cooperate the most. My once co-author Ryan Oprea has a paper with Daniel Friedman in the latest AER, showing this:

We study [lab experiment] prisoners’ dilemmas played in continuous time with flow payoffs accumulated over 60 seconds. In most cases, the median rate of mutual cooperation is about 90%. Control sessions with repeated matchings over 8 subperiods achieve less than half as much cooperation, and cooperation rates approach zero in one-shot control sessions.

They introduce some new theory to explain details of this behavior:

Inspired by a strand of existing theoretical literature, we postulated a particular class of epsilon equilibria and derived formulas predicting how cooperation rates respond to adjustment lags and to payoff parameters. These predictions accounted well for the Continuous, Grid-8 and (trivially) One-Shot data. They also nicely explained a set of second-round data from Grid-n sessions, which varied the number of subperiods from 2 to 60. Thus the formulas correctly predict defection in one shot games, cooperation in continuous time and intermediate results on the path between the two. The underlying intuition is simple. When your opponent can react very quickly, defecting from mutual cooperation is likely to earn you the temptation paypoff only briefly and may cost you the cooperation payoff for the rest of the period.

So do online firms cooperate more when they can vary their prices more frequently? What rapidly-changeable actions would help nations to cooperate more?

GD Star Rating
loading...

Tagged as: Project, Social Science

What Is Econ Advice?

By Robin Hanson · February 6, 2012 2:35 pm · Comments (7)

Saturday I said:

I see two rather different ways to think about what economists do when we recommend policies:

Morals – We enter into a larger conversation about what actions are right. … While for our ancestors such discussion was often a prelude to concrete group action, today we more often argue about morals without such ways to coordinate in mind. …
Deals – … Conflicts … can be resolved at least in part via deals, both implicit and explicit. Finding and making such deals can be aided by neutral outside advisors, … Economists have a reputation for developing analytical tools useful for making suggestions about parts of deals. …

Academics who lean toward a sci/tech style tend to favor a deals view, while those who lean toward a humanities style tend to favor a morals view.

Bryan Caplan responded:

Fact #1: Robin has spent decades proposing unconventional policy deals. His track record is an abysmal failure. … Zero Hansonian deals have been embraced by any normal person. His proposals appeal almost exclusively to fans of economics, libertarianism, futurism, and science fiction. … Most human beings are far too conventional and stubborn to even consider Robin’s suggestions. And instead of trying to overcome this hurdle, Robin habitually raises the hurdle by criticizing conventional attitudes. …

Fact #2: People often have a very good reason to ignore deals: They have better ways to get what they want. Such as: persuasion, moralizing, trickery, and bullying.
Fact #3: The effectiveness of deal-making varies widely by person. …
Fact #4: The effectiveness of deal-making varies widely by situation.

I’m happy to admit that deal-making isn’t the only way we get what we want, and that its effectiveness varies by person and situation. Though I do think we make a lot of implicit deals. For example, asking “I’m in a hurry; would you mind if I cut in line” is offering supplication, and an owing of return favors, in trade for a favor.

I’m also happy to admit that much of what I do intellectually isn’t very directly targeted at promoting specific deals. Though neither is it very directly targeted at labeling specific acts as moral or immoral. To the extend that I do either of these things, I do them indirectly, which makes my acts harder to categorize. My work in prediction markets, however, has led to some voluntary mutually-beneficial changes in corporate practices, and I hope for lots more.

Imagine that economists were surveyed and had to choose how they’d best like to describe economic policy recommendations, as:

Morals – Arguing for the morality of actions,
Deals – Helping groups find and make deals, or
Showing Off – Academics do hard things in order to be certified by other academics as impressive, so that students, patrons, and readers can gain status by affiliating with them. Economic policy analysis is such a hard thing.

I’d bet that at least 25% would choose option #2, and even more among those whose style leans sci/tech. And #2 seems to me a better public face for economists to present to the world – economists will prosper more overall if they say this is what they are doing.

GD Star Rating
loading...

Tagged as: Academia, Social Science

Inequality Market Failure?

By Robin Hanson · February 5, 2012 5:20 pm · Comments (92)

In ordinary talk, you often hear arguments like:

There aren’t enough Jazz stations – government should subsidize them.
Too many kids today let their pants hanging low – that should be illegal.
Not enough kids want to be scientists – schools should push that earlier.

But to an economist, it is not enough to note that you do or don’t like something, to justify a policy to encourage or discourage it. We instead hold ourselves to a higher analysis standard – is there a net market failure sufficient to justify an intervention?

Except, alas, on (national at-a-time between-family) income and wealth inequality. There, most economists think it sufficient to just note that a policy influences inequality – they rarely feel a need to identity an associated market failure. For example, Christina Romer:

A successful argument for a government manufacturing policy has to go beyond the feeling that it’s better to produce “real things” than services. … The economic rationales for a policy aimed specifically at shoring up manufacturing largely fall into three categories. None are completely convincing:

MARKET FAILURES Government intervention can be justified on efficiency grounds if the free market won’t work well. … The market can malfunction if there are positive externalities across companies. … But large clustering effects have been hard to find. … A study of the semiconductor industry found that although learning by doing was substantial, most of the rewards went to companies doing the early investing. … We need a strong manufacturing base in case of war. … But it still doesn’t follow that all manufacturing deserves special treatment. …

JOBS A key argument for encouraging manufacturing is to create jobs and reduce unemployment. Unfortunately, those effects are probably small. … Today, we face a profound shortfall of demand. That truly is a terrible market failure, and it warrants government intervention. …

INCOME DISTRIBUTION A final argument for supporting manufacturing is distributional. Manufacturing jobs are seen as one of the few sources of well-paying jobs for less-educated workers. … But that is much less true today. … If increasing income equality is the goal, it might be wiser to put money into infrastructure than to subsidize manufacturing. …

Public policy needs to go beyond sentiment and history. It should be based on hard evidence of market failures, and reliable data on the proposals’ impact on jobs and income inequality. (more; HT Tyler)

Note that she even uses market failure to justify pushing jobs. But not for income equality – that is just obviously bad. I see the same thing over and over – only economists wary of equality-promoting policies talk market failures, and then they mainly ask what they are. For example Charles Lane:

Americans may never agree on an optimal distribution of income, either morally or practically. But they probably could agree that, to the extent possible, government should limit its interventions to bona fide cases of market failure. (more)

Well, no, Americans probably don’t agree on that. But you might hope economists would. Tyler Cowen has an article where he says there are market failures in the finance sector that increase inequality, and recommends fixing them. Which of course makes sense, but we’d want to fix those problems even if they reduced inequality.

The closest I could find to a market failure argument for reducing inequality was Ian Ayres and Aaron Edlin:

The progressive reformer and eminent jurist Louis D. Brandeis once said, “We may have democracy, or we may have wealth concentrated in the hands of a few, but we cannot have both.” (more)

But that is quite a stretch – there is no evidence that wealth concentration is threatening to stop our nation from being a democracy. And it is far from obvious that not being a democracy is a market failure.

As Obama has decided to make reducing inequality a central issue in his reelection campaign, we are going to hear a lot about it between now and November, including from economists. Could economists who support policies to reduce inequality please identify their market failure arguments?! Why lower our usual standards for this topic?

(I argued here that a poverty insurance market failure seems implausible.)

Added 2p: In case it is not clear, this post is directed to economists, in their role as economists. I’m not saying market failure is the only consideration anyone uses to decide policy, but I am suggesting that it is the main consideration that economists use in their role as holders of economic expertise. Economists don’t have much expert to say about whether we have too much or too little inequality, outside of their expert ability to discern and fix market failures.

GD Star Rating
loading...

Tagged as: Current Affairs, Inequality, Social Science

Deals Near, Morals Far?

By Robin Hanson · February 4, 2012 10:45 pm · Comments (8)

Unlike most social scientists, economists publish papers whose main conclusions are normative, not just positive. That is, we explicitly recommend policies, rather than just predict the consequences that follow from choices. Moral philosophers are the main other academics who explicitly focus on making recommendations. Compared to philosophers, economists are more standardized and structured, allowing more specific recommendations.

I see two rather different ways to think about what economists do when we recommend policies:

Morals – We enter into a larger conversation about what actions are right. As language let humans enforce social norms, we evolved strong moral feelings and a rich practice of arguing about forbidden and required actions. While for our ancestors such discussion was often a prelude to concrete group action, today we more often argue about morals without such ways to coordinate in mind. Economists mainly join this larger argument about what actions are right, and only incidentally induce specific people to take specific actions.
Deals – Groups of all sizes, from families to planets have conflicts that can be resolved at least in part via deals, both implicit and explicit. Finding and making such deals can be aided by neutral outside advisors, especially advisors who can suggest changes that might benefit all conflicting parties. Economists have a reputation for developing analytical tools useful for making suggestions about parts of deals. Economists mainly develop such suggestions for deal components, and only incidentally join larger conversations about morality.

My closest colleagues seem to mostly take a morals view, but many of my students like a deals view. I think I see a correlation whereby academics who lean toward a sci/tech style tend to favor a deals view, while those who lean toward a humanities style tend to favor a morals view. Sci/tech styles tend more toward math, precision, and local incremental contributions toward specific things and plans, while humanities styles tend more toward bigger pictures, wider-ranging applications, broader interpretations, and joining larger conversations.

In sum, how you think about economic recommendations may depend on whether your thinking leans near or far. It seems deals are near, while morals are far, and sci/tech folks lean near, while humanities folks lean far. Precise formal analysis is more near, while flexible more-metaphorical discussion is more far. Particular suggestions for particular conflicts of particular groups is more near, while general more accessible discussion about what choices tend to be good or bad is more far.

GD Star Rating
loading...

Tagged as: Academia, NearFar, Social Science

Regulatory Differences

By Robin Hanson · January 2, 2012 11:50 pm · Comments (56)

I recently discussed a puzzling regulatory difference: our applying work hour limits less to high than low status jobs. Many took me to be advocating fewer limits for low status jobs, and were eager to point out good reasons for work hour limits. But our not having a good reason for putting more work hour limits on high vs. low status jobs can equally well support adding more limits to high status jobs, rather than fewer limits on low status jobs.

John Cochrane similarly discussed a puzzling regulatory difference:

Ken Rogoff put in this little zinger

Medical care … fails to satisfy several of the basic requirements necessary for the price mechanism to produce economic efficiency, beginning with the difficulty that consumers have in assessing the quality of their treatment.

… really, Et tu Ken? It’s hard to know if the car mechanic is doing a good job. Get ready for the Federal takeover of the car industry. I can’t tell B grade exterior from A grade interior plywood, so we need a Federal takeover of home rehab.

Many readers probably take this as an argument for less medical regulation, and are eager to argue for or against that position. But pointing out that we have a similar difficulty assessing car mechanic and doctor quality can equally well argue for regulating car mechanics more, instead of regulating doctors less.

In general, people seem far more eager to collect respectable arguments for or against various specific regulations, than to consider the coherence of a pattern of regulations they endorse. They are satisfied to offer arguments for why janitors should have work hour limits, why musicians should not, why doctors should be highly regulated, and why car mechanics should not, all without much noticing or caring how much they treat similar cases differently.

This suggests that there is a lot of rationalization going on. That is, rather than choosing some principles and then consistently applying them, people instead pick various random policy positions and then search for justifications. There seems to be only weak pressures to even notice much less reduce how they and their arguments treat similar things differently.

GD Star Rating
loading...

Tagged as: Regulation, Social Science

The Puberty Puzzle

By Robin Hanson · October 25, 2011 9:10 pm · Comments (46)

Time magazine considers a big important puzzle:

By the 1980s, the onset of puberty, if not actual menstruation, had gone into free fall–a change so sudden and pronounced that something more than normal evolution must have been at work. In a landmark 1997 study of 17,000 [US] girls … more than 10% of white girls and an astonishing 37.8% of black girls were showing early breast development by age 8. … Later studies, one in 1998 and another in 2010, included Hispanics and produced similar results. On average, 2 out of every 10 white girls, 3 out of 10 Latinas and 4 out of 10 black girls are showing breast development by age 8. (more)

They consider some possible explanations:

Obesity, a well-established puberty accelerant, is high on the list of suspects. … Data from China and India similarly indicate that race by itself isn’t a factor but general prosperity is. Onset of puberty is on a downward march in those countries too. … But even in Europe, where the standard of living has been high for decades and diets haven’t changed much, something strange is going on. A study of girls conducted in Denmark in 2008 found that the average age of breast development there is 8.86 years, which … is a full year earlier than it was for Danes as recently as 1993. … Some investigators are focusing on environmental contaminants like PBBs and … bisphenol A … A number of studies have found that overweight boys may, if anything, suffer from delayed puberty.

Oddly they don’t even mention divorce and out-of-wedlock birth, factors that some theory suggests are crucial:

Father absence is indicative of the degree of polygyny (simultaneous and serial) in society. Polygyny of both kinds creates a shortage of women in reproductive age, and thus, early puberty will be advantageous. Available comparative data indicate that the degree of polygyny is associated with a decrease in the mean age of menarche across societies, as is the divorce rate a presumptive index of serial polygyny, in strictly monogamous societies. (more)

This theory has some empirical support:

As specified by evolutionary causal theories, younger sisters had earlier menarche than their older sisters in biologically disrupted families (n = 68) but not biologically intact families (n = 93). This effect was superseded, however, by a large moderating effect of paternal dysfunction. Younger sisters from disrupted families who were exposed to serious paternal dysfunction in early childhood attained menarche 11 months earlier than either their older sisters or other younger sisters from disrupted families who were not exposed to such dysfunction. (more)

I heard of this theory a while ago, but until now I hadn’t realized its radical implication: humans may have evolved adaptations to make major body/life features conditional on our social environment! If girl brains can order hormones to induce early puberty after seeing lots of nearby polygyny, how else might our bodies be contingent what our brains see about our social world? Do young brains see the level of violence, prosperity, or work complexity around, and adjust hormone-induced plans for body size, immune system strength, or brain resources? Could this adjustment explain recent trends in mortality, height, or intelligence? So many possibilities to consider!

Anthropologists often say that it is a mistake to look for “the” ancestral human environment or lifestyle, that what most defines humans is variety and adaptability. I’m going to take that view a lot more seriously from now on.

Added 4p: Why are people so much more willing to use strange chemicals to explain earlier puberty that other trens like increasing IQ, lifespan, and height? Is it because chemicals are bad, and therefore can only explain bad things?

GD Star Rating
loading...

Tagged as: Current, Forager, Kids, Mating, Social Science

Transferable Citizenship

By Robin Hanson · October 21, 2011 4:30 pm · Comments (21)

In 20th century USA, many feared that if a black moved into their neighborhood, “white flight” would greatly lower their property values. In many places it was illegal to sell your house to a black. But it was never illegal to sell your house to outsiders. At the national level today, however, we actually do adopt the extreme “no house selling” analogy: we don’t let folks transfer their citizenship to foreigners.

Imagine each US citizen could transfer her citizenship to a foreigner, as long as she found somewhere else to live, [added: hadn't recently given birth,] and wasn’t about to die, and the foreigner was at least as old, and wasn’t a terrorist, ex-con, etc. Benefits:

Citizenship could be collateral for loans for school, houses, etc.
We’ll prefer those who’d pay to come, over those who choose to leave.
Undesirable poor folk are especially likely to leave.
Retirees could be paid to retire more cheaply abroad.

Similar to those old rules against selling houses to blacks, we could also add more restrictions on to whom citizenship could be transfered. At least if we were willing to publicly own up to the racism, ethnism, etc. that such restrictions embodied.

Now folks like Julian Simon and Gary Becker have proposed selling citizenship before, without much success. But it seems to me better marketing to first focus on giving each person a direct benefit: more freedom to use an asset they already own, citizenship. First, get folks to see that selling citizenships makes as much sense as selling houses or club memberships. Then suggest that letting government add to the pool of citizenships for sale might raise revenue, and maybe help the economy as well.

This post was sparked by hearing a talk by Michael Clemens, who noted that we saw no effect on wages from either the Cuban boat lift that suddenly increased Miami’s population by 7%, nor the sudden elimination of migration restrictions within South Africa. Those sound like great pro-immigration arguments to an economist, but alas seem a bit too indirect for the public.

Added: Bryan suggests immigrants pay extra taxes, while Alex suggests giving visas to house buyers.

GD Star Rating
loading...

Tagged as: Politics, Social Science, World

Subscribe

Recent Posts

Recent Comments

Popular Tags

Less Wrong

Blogroll

Sponsoring Organizations