Following the publication of two new articles in the Annual Review of Economics and PNAS (my summary here), genoeconomics has been getting some press. From the Boston Globe:

But for all their throat-clearing and the cold water they feel compelled to throw on their work as they introduce it to the general public, they are confident that it’ll eventually be possible to match up patterns in a person’s genome with patterns of financial behavior. Perhaps parents could be alerted if their kids have genes that incline them to impulsive spending or wildly risky investments, Benjamin said. Or perhaps policy makers could use genetic information about a particular population—say, cigarette-smokers or alcoholics—in order to craft policy more effective at encouraging some behaviors and discouraging others.

While genoeconomics might provide exciting results, some of the information that is likely to emerge is already available to be used. We know from twin studies that patterns of financial behaviour are heritable. If a parent knows they are financially reckless, this can provide a probability of that behaviour for their child. You should worry about your child smoking or being an alcoholic if you have those characteristics yourself. We can obtain further information from the behaviour of siblings or other relatives. We do not need to wait for genoeconomics to deliver results before we can make inferences about genetically influenced behaviour.

Following the Consilience Conference and some suggestions for additions to my reading list, I have been convinced to read some more work by Robert Boyd and Peter Richerson. I’ve started with The Origin and Evolution of Cultures.

One quote in the introduction caught my eye:

[A]cquiring adaptive information from others also opens a portal into people’s brains through which maladaptive ideas can enter—ideas whose content makes them more likely to spread, but do not increase the genetic fitness of their bearers. Such ideas can spread because they are not transmitted as genes are. For example, in the modern world, beliefs that increase the chance of becoming an educated professional can spread even if they limit reproductive success because educated professionals have high status and thus may likely be emulated. Professionals who are childless can succeed culturally as long as they have an important influence on the beliefs and goals of their students, employees, or subordinates. The spread of such maladaptive ideas is a predictable by-product of cultural transmission.

While I might characterise the maladaptation differently – it is not being a professional in itself that limits reproductive success – this approach contrasts with the economic explanation. Economists usually frame the fertility decision as a rational quality-quantity trade-off. In addition to the usual economic assumption of rationality concerning the pursuit of an objective, there is also an assumption that the objective itself (fitness in a biological sense) is rationally chosen. There is no maladaptation, and I suspect this is the cause of some of the difficulty the economic approach to the problem has faced.

Evidence from twin studies implies that economic and political traits have a significant heritable component. That is, some of the variation between people is attributable to genetic variation.

Despite this, there has been a failure to demonstrate that the heritability can be attributed to specific genes. Candidate gene studies, in which a single gene (or SNP) is examined for its potential influence on a trait, have long failed to identify effects beyond a fraction of one per cent. Further, many of the candidate gene results fail to be replicated in studies with new samples.

An alternative approach to genetic analysis is now starting to address this issue. Genomic-relatedness-matrix restricted maximum likelihood (GREML – the term used by the authors of the paper discussed below) is a technique that looks to examine how the variance in traits can be explained by all of the SNPs simultaneously. This approach has been used to examine height, intelligence, personality and several diseases, and has generally shown that half of the heritability estimated in twin studies can be attributed to the sampled SNPs.

A new paper released in PNAS seeks to apply this approach to economic and political phenotypes. The paper by Benjamin and colleagues shows that around half the heritability in economic and political behaviour observed in behavioural studies could be explained by the array of SNPs.

The authors used the results of recent surveys of subjects from the Swedish Twin Registry, who had their educational attainment, four economic preferences (risk, patience, fairness and trust) and five political preferences (immigration/crime, foreign policy, environmentalism, feminism and equality, and economic policy) measured. The GREML analysis found that for one economic preference, trust, the level of variance explained by the SNPs was statistically significant, with an estimate of narrow heritability of over 0.2. Two of the political preferences, economic policy and foreign policy, had narrow heritability that was statistically significant, with heritability estimates above 0.3 for each of these. The authors noted that as the estimates are noisy and GREML provides a lower bound, the results are consistent with low to moderate heritability for these traits.

Educational attainment was also found to have a statistically significant result, although the more precise measurement of educational attainment and the availability of this data across all subjects made that result more likely.

This result is corroboration of the evidence from twin studies and provides a basis for believing that molecular genetic data could be used to predict phenotypic traits. However, one interesting feature of the GREML method of analysis is that after conducting this analysis with one sample, the data obtained does not assist in predicting the traits for someone out of the sample. This technique shows the potential of molecular genetic data without directly realising those results.

As a comparison, the authors examined whether any individual SNPs might predict economic or political preferences, but found none that met the significance test standard of 5×10^-8. Such a high level of significance is required to reflect the huge number of SNPs that are being tested.

The authors also conducted the standard comparison between monozygotic (identical) and dizygotic (fraternal) twins, which resulted in heritability estimates consistent with the existing literature, although with a much larger sample than typically used. Looking through the supplementary materials, the major surprise to me was that the twin analysis suggests that patience has low heritability, with a very low correlation between twins and almost no difference between monozygotic and dizygotic twins (in fact, for males, dizygotic twins were more similar).

The authors draw a few conclusions from their work, many which reflect the argument in a Journal of Economic Perspectives article from late last year. The first and most obvious is that we should treat all candidate gene studies with caution. Hopefully some journals that insist on publishing low sample size candidate gene studies will pay attention to this. Where they are going to be conducted, you need very large samples, and significantly larger than are being used in most studies being published.

Meanwhile, they are still hopeful that there can be a contribution from genetic research, particularly if the biological pathways between the gene and trait can be determined. This might include using genes as instrumental variables or as control variables in non-genetic empirical work. The use as instrumental variables does require, however, some understanding of the pathways through which the gene acts as it may have multiple roles (that is, it is pleiotropic). They also suggest that the focus be turned to SNPs for which there are known large effects and the results have been replicated.

On element of analyses of political and economic preferences that makes me slightly uncomfortable is the loose nature of these preferences. For one, the manner in which they are elicited from subjects can vary substantially, as can the nature of the measurement. Take the 2005 paper by Alford and colleagues on political preferences, which canvassed 28 political preferences. Many of the views are likely to change over time and be highly correlated with each other. And why stop at 28?

As a result, it may be preferable to take a step back and ensure that data on higher level traits are collected. I generally consider that IQ and the big five personality traits (openness, conscientiousness, agreeableness, extraversion and stability) are a good starting point and are likely to capture much of the variation in political and economic preferences. For example, preferences such as patience are likely to be reflected in IQ, while openness captures much of the liberal-conservative spectrum of political leaning. Starting from a basis such as this may also give greater scope for working back to the biological pathways.

The Social Science Genetics Association Consortium is doing some work in harmonising phenotypes across large samples. Hopefully their work will lead in this direction.

Over at Cheap Talk, Jeff Ely has posted on a presentation by Balazs Szentes at The Biological Basis of Preferences and Behavior conference. Ely writes:

Balazs Szentes stole the show with a new theory of the peacock’s tail. …

Suppose female peacocks choose which type of male peacock to mate with: small or large tails. Once the females sort themselves across these two separate markets, the peacocks are matched and they mate.

The female peacocks are differentiated by health, and within a peacock couple health partially compensates for the disadvantageous tail. In the model this means that healthy females who mate with big-tailed peacocks will produce almost as many surviving offspring as they would if they mated with peacocks without the disadvantage of the tail. …

Consider what happens when a small-tailed peacock population is invaded by a mutation which gives some male peacocks large tails. Since female peacocks make up half the population of peacocks there is now an imbalance in the market for small-tailed peacocks. In particular the males are in excess demand and some females will have trouble finding a mate.

On the other hand the big-tailed male peacocks are there for the taking and its going to be the healthy female peacocks who will have the greatest incentive to switch to the market for big tail. The small cost they pay in terms of reduced quantity of offspring will be offset by their increased chance of mating. The big tails have successfully invaded.

Szentes’s theory illustrates the mixed feelings I expressed in my recent post about some of the presentations at the conference. The model underlying the theory is clever and interesting, but Szentes’s focus is more on the game theory than the evolutionary problem that the model is proposed to address. The model offers limited insight into the evolution of the peacock’s tail as the assumptions underpinning the model do not hold.

First, the model requires an assumption of monogamy, which peacocks are not. As for most males with ornaments, the peacock’s tail is used to attract multiple partners to make up for the handicap that the ornament imposes (as the more established theory suggests). The model also assumes that the males are indifferent as to who they mate with (despite being monogamous), with high quality females unable to attract male interest above that of females of low quality.

Without those assumptions, the findings derived from the model no longer hold. In some respects, the sexes in the model appear backward, as a lack of males willing to give sperm is not an issue in most species. The low investment by males makes females the scarce resource.

I expect that these issues are of less concern to Szentes (or, based on his post, Ely) than they are for me, as his interest lies more in the model than its particular application. Szentes appeared to be aware of these critiques when they were raised at the conference.

If the model is not useful in explaining the peacock’s tail, what situation might the model describe? Instead of talking of disadvantaged males with tails, could we talk of low quality males and use the model to explain the persistence of low quality males in a population? This is an interesting model looking for a use.

*The videos are many of the presentations are now up.

The application of evolutionary biology to politics and policy spans the political spectrum. From Peter Singer’s A Darwinian Left to Larry Arnhart’s Darwinian Conservatism to Michael Shermer’s libertarianism, there is something in evolutionary biology for everyone.

One of the best of of these applications is by Paul Rubin in Darwinian Politics: The Evolutionary Origin of Freedom. While the arguments lead to conclusions that reflect Rubin’s political leanings, the book reads as though the evidence shapes the result and Rubin gives the evidence fair consideration.

Rubin’s basic position is that the political institutions of Western nations, and particularly the United States, are the best match with evolved human preferences. Humans seek freedom from dominance, with Western society maximizing that freedom. Political freedom allows citizens to form a reverse dominance hierarchy, with public pressure, wealth and constitutional frameworks limiting the ability of Western governments to exercise power. Western institutions also provide a framework that limits negative consequences of our evolved psyche, as the move away from kin based groups reduces xenophobic behaviour.

Rubin does not suggest, however, that humans are perfectly matched to our current environment. Rather, he argues that the current institutional frameworks do a good job of working around them. For example, one of the main threads of Rubin’s argument is that humans have moved from an evolutionary history of living in consumption hierarchies, which are effectively zero sum, to a world of productive hierarchies, whereby participation in hierarchies can boost production and be good for all involved. Humans do not always distinguish between the two (nor academics as Rubin makes clear), and as a result, envy may result as someone acts as though they are still in their zero sum world.

Rubin sees the best political framework as one that can deal with this tendency to envy while not damaging the productivity of the hierarchy. Rubin addresses these concerns through a couple of threads. One is to note that in a free society, movement between hierarchies is possible and people are likely part of many hierarchies. They will not always be at the bottom. However, Rubin paints an overly rosy picture (he should paint a rosy one – it just needs some tempering), as some hierarchies are more important than others and mobility is not a complete solution. If you are at the bottom of an employment hierarchy, your choice is likely to be which hierarchy you wish to be at the bottom of. As shown in the famous Whitehall studies, being at the bottom of a productive hierarchy may have costs (not that I implying that Whitehall itself is a productive hierarchy).

A more interesting points is when we move from utility to fitness. Rubin writes:

If an individual is highly productive and creates much wealth, social as well as private benefits will be generated; a productive individual will not normally absorb the entire surplus he will create. Thus, utility or wealth maximisation would imply that all will benefit from such increased productivity and should encourage it. However, if the added productivity is used to engross additional females, or if tastes evolved in an environment where this occurred, then in fact others will become less fit, although wealthier. In this sense, fitness and utility maximisation conflict. This may explain why many utility functions seem to contain elements of envy, even though envy is counterproductive with respect to consumption of wealth maximisation.

Rubin’s primary solution to this, already implemented throughout the West, is monogamy. Each man can only monopolise one female regardless of wealth. While monogamy undoubtedly acts as a reproductive leveler, Rubin’s analysis attempts to finesse his case too much. Despite monogamy in Western societies, there is still a large proportion of men at lower socioeconomic status who fail to attract a mate. Higher status women simply choose not to mate with them and lower status women have other avenues of seeking financial support. For example, over 40 per cent of Australian men between the ages of 40 and 44 and with incomes below $20,000 per year remain unpaired. This contrasts with just over 10 per cent of men of that age group with incomes over $83,000 per year. Monogamy levels the reproductive playing field but it is not completely flat.

For each of these points, Rubin is essentially right in his argument that productive hierarchies are beneficial and that monogamous societies are more stable and level the reproductive playing field. However, there is still a bottom of the hierarchy and consequences to it, and there will always be some degree of pushback due to this.

To assist those at the bottom, Rubin notes the evolved altruistic preferences for assisting those in need. As a result, some wealth redistribution may be supported. But if the focus moves from supporting the poor to clipping the rich, the output of productive hierarchies may be threatened. Further, Rubin considers that social support will remain popular as long as it is not overexposed to free-riding, with humans having strongly refined senses to spot those who are not pulling their weight.

Rubin’s least libertarian finding, apart from his implied support of restrictions on polygamy, relates to restrictions on drugs and other “anti-social” activities. Rubin argues that if consumption of these goods and activities is a form of competition between young males to signal status, restrictions on their use will be required to prevent above optimal use. While Rubin considers that the need to maintain a society’s prime age men at fighting strength is weaker than in our evolutionary past, a case can still be made for this form of control. It was interesting that Rubin chose to use a signaling argument at this point as he does not address the role of signaling in most of his analysis, such as in his discussion of “altruistic” gifts of game in ancestral societies or donations to charity.

Overall, Rubin’s arguments are clear, transparent and generally persuasive. It was an omission by oversight from my economics and evolutionary biology reading list, as I last read it a number of years ago, but it has now been included (thanks to Eric Crampton for the nudge).

I have just attended The Biological Basis of Preferences and Behaviour conference at the Becker Friedman Institute at the University of Chicago. It was a good conference with some high quality presentations. I will post on some of them over the next few months once I digest the presentations and papers (or they exit embargo).

In the meantime, the conference has triggered some thoughts on how economics will contribute to the evolutionary sciences, and how biology will be integrated into economics.

Most work generated by economists on the evolution and biological basis of preferences uses beautiful (to some beholders) but complicated mathematical models. As a gross generalisation, most presentations at the conference consisted of multiple slides of equations and proofs. Many questions and clarifications were required as the presenter progressed through the slides. The presentation usually ended with an attempt to explain the intuition of the model in plain language.

After being walked through one of these models, the first question to ask is whether the finding of the model might be true (noting the problems of defining truth). Are the assumptions even ballpark realistic? Is there any evidence that the process described in the model occurs or has occurred?

The next question is about insight. Did this model, even if not “true”, provide useful insight into the process or phenomena being explored?

Finally, does the model communicate the idea in a way that will be understood by people who might care about or should know about the work?

My impression was that many of the conference presentations fell at the first two hurdles. I sense that in many instances this was because the presenters were more interested in the mathematical dynamics and game theory than the final application of the model. This is, of course, a broader issue through much of economics, where the incentives push researchers towards beautiful but complicated models. Generally, these models will not be subject to testing and as a result, do not face the threat of being discarded if they are not supported by the empirical evidence. Having robust mathematics that supports the story being told has greater weight than the truth of the model or the manner in which it might communicate an insight. There is little direct competition between models.

Even where the first two hurdles were cleared, I am uncertain how often the product of the research is likely to be consumed outside of a small circle, generally consisting of other members of the field. Mathematical models are an excellent way to communicate to some people (my impression is that many in the conference audience prefer this approach), but simple, verbal explanations are the way most communication occurs.

In contrast, at last week’s Consilience Conference there was barely an equation in sight. The presentations provided me with a mountain of ideas that I have communicated in various ways ever since. Those presentations from this week’s conference that were based on a detailed mathematical model, while providing some fodder for thought, are unlikely to feature in many conversations. I am not sure the models presented are true or insightful, and even where they are, my first thought is how to come up with a different way of communicating the point. My posts over the next couple of weeks will largely feature those presentations that were the exception to the rule.

Having dug myself into a hole through some broad, over-generalised criticism, this critique is not to say that there are no useful and important ideas coming out of this work. My economics and evolutionary biology reading list has a sample of what I consider to be some of the more important contributions. Rather, my critique reflects my instinct that the field will remain a world unto itself, without being a significant contributor to the congruence of biology and economics. Research into the biological basis of preferences and behaviour has massive potential to affect the broader field of economics. It would be disappointing if this research did not achieve that result.

*The videos are many of the presentations are now up.

A quote from Arthur Jensen (From Steve Hsu. A fuller version of the interview can be found here):

[T]he outstanding feature of any famous and accomplished person, especially a reputed genius, such as Feynman, is never their level of g (or their IQ), but some special talent and some other traits (e.g., zeal, persistence). Outstanding achievements(s) depend on these other qualities besides high intelligence. The special talents, such as mathematical musical, artistic, literary, or any other of the various “multiple intelligences” that have been mentioned by Howard Gardner and others are more salient in the achievements of geniuses than is their typically high level of g. Most very high-IQ people, of course, are not recognized as geniuses, because they haven’t any very outstanding creative achievements to their credit. However, there is a threshold property of IQ, or g, below which few if any individuals are even able to develop high-level complex talents or become known for socially significant intellectual or artistic achievements. This bare minimum threshold is probably somewhere between about +1.5 sigma and +2 sigma from the population mean on highly g-loaded tests.

But IQ is not meant to capture everything of interest:

So-called intelligence tests, or IQ, are not intended to assess these special abilities unrelated to IQ or any other traits involved in outstanding achievement. It would be undesirable for IQ tests to attempt to do so, as it would be undesirable for a clinical thermometer to measure not just temperature but some combination of temperature, blood count, metabolic rate, etc. A good IQ test attempts to estimate the g factor, which isn’t a mixture, but a distillate of the one factor (i.e., a unitary source of individual differences variance) that is common to all cognitive tests, however diverse.

The fundamental insight that utility in economics should be based on the concept of fitness from evolutionary biology lies at the heart of Gandolfi, Gandolfi and Barash’s Economics as an Evolutionary Science: From Utility to Fitness.

The first half of the book is fantastic, as the authors describe the economic way of thinking and Gary Becker’s seminal work on families, marriage and reproduction. For someone unfamiliar with Becker’s work, it is a good introduction to what Becker was setting out to achieve. The authors then take Becker’s framework and build on it to give it an evolutionary basis.

The authors do this by constructing a model in which people maximise their long-term fitness through balancing investments in quantity of offspring, quality of offspring and other capital investments. It is by maximising long-term intergenerational wealth, which can be converted into quantity and quality of children as required, that an agent can maximize fitness. This provides a basis for the declining fertility in modern economies, with the authors arguing that investment in quality of children is a long-term fitness maximising strategy. It is not the number of children in the next generation that matters but the ultimate number of children.

I like this idea and approach, but ultimately, I am not convinced that it is true. To take an extreme case, what level of investment should a billionaire make in a child? Would they maximize their long-term fitness by having one or two children and bequeathing their entire fortune to them, or should they establish a harem and seek to have hundreds of children, each of whom could still have millions of dollars of investment in them. It would seem to me that the latter strategy would maximize fitness, particularly when you consider uncertainty across generations. One should reap while you can, but the rich do not appear to be doing this. There is an over investment in quality due to humans being in an environment to which they are not yet adapted.

Ultimately this is an empirical question, but the correlation between numbers of children across generations in developed countries suggests that those with a predisposition to sacrifice some quality for quantity are following the higher fitness strategy. The lack of effect of parents on child outcomes, as found through twin research, also suggests that parental investment has marginal returns at best.

If the current population is not optimizing fitness, the usefulness of the link between utility and fitness changes. Rather than seeking to equate each, the predispositions shaped by evolution can be used to determine the utility function, with the two differing to the extent the environment is no longer one in which those predispositions are fitness maximizing. For longer term modeling, the evolutionary system needs to be considered dynamically, as people adapt to the new environment.

In addition to the basic framework, the book contains chapters on sexual conflict and cooperation. While each are important considerations for the model, unfortunately they are not directly incorporated into it. In particular, it would be interesting to apply the discussion of signaling to the model, as the model does not provide an explanation for consumption above that required for survival.

The chapter on the evolution of cooperation treads ground covered in more detail by many other books, and unlike the earlier chapters, appears to be included for completeness rather than novelty.

Regardless of my critique, this is an important book. The challenge is for other scholars to take the framework laid out in the book and to start to adopt it for a modern environment.

The Consilience Conference on evolution in biology, the social sciences and the humanities wrapped up on Saturday, and it was generally a high quality conference. It’s strength was that most of the presenters were doing work across multiple fields, usually with an evolutionary twist. Conferences such as these often involve people trying to frame existing work around the topic, even if it is a weak fit, but here the presenters’ work generally fitted the subject nicely.

Some stray thoughts from the conference and presentations are below.

Until this conference, I had not realized the scope of the field of literary Darwinism. While I am slightly skeptical as to how far the idea can go, it was good to see some high quality thought being put into it. This article from Science provides some interesting background on the field, including comments from the conference organiser Joseph Carroll and one of the more impressive conference presenters, Jonathan Gottschall.

Barb Oakley presented on the idea of pathological altruism. The basic idea is that too much altruism is not necessarily a good thing. In some ways, it is an obvious point as altruism is costly to the altruist by definition and can be exploited by cheaters. However, that point is often forgotten in descriptions of the virtue of altruism, and when taken to extremes, can impose significant costs on both oneself and others. Some of the left-right divide might be explained by debate about what is considered to be the optimal level of altruism before the costs and potential for exploitation become too large.

Peter Turchin explored the quantification of history. Turchin argued that there is a need to move on from the tendency to accumulate theories without ever rejecting them. It sounds familiar to an economist.

Henry Harpending presented some ideas around measures of kinship. He was surprised that his research to date showed that, within most populations, there appeared to be little benefit in determining the degree of kinship of someone who may be a beneficiary of your actions. The variation in kinship is simply too small to matter. However, there is some value to considering kinship in mixed populations. Many of the results contained in the presentation are summarized by Harpending over at West Hunter.

Massimo Pigliucci’s summaries of each day (days one, two and three) are worth reading, although Massimo played the role of the token skeptic about the integration of evolution into social sciences and the humanities, and his comments reflect this.

The first day of the Consilience Conference has strengthened my feeling that support for group selection is growing in the social sciences. While the slant of speakers such as Edward O. Wilson and Herb Gintis is no surprise, the degree of support among many conference participants that I have spoken to was. The general argument is that the evolution of “altruistic” and cooperative behaviour requires group selection.

Again, the question in my mind is at what point do the evolutionary biologist critics of the group selection approach enter into this evolution of cooperation debate that is happening outside of their field? Social scientists relying on group selection generally do not receive attention from evolutionary biologists as their respective silos do not meet.

On the speakers, Edward O. Wilson’s advocacy of group selection over inclusive fitness in his keynote address was not particularly convincing, and his broader argument did not require the kin selection critique that he offered. A core point to his support of group selection was that individual level selection is responsible for sin, while group selection is responsible for virtue.

Of the arguments for group selection, the sin-virtue dichotomy is one of the weaker ones, and economics provides one of the best responses. Robert Frank responded to Wilson’s point at the end of his own presentation (which largely covered material from The Darwin Economy). He noted the benefits of exchange and the core insight (dating back to Adam Smith) that cooperative behaviour can emerge from self interested individuals. Despite Frank’s advocacy of Darwin as the father of economics, he has not forgotten the importance of Smith.

Further, I am uncomfortable with the sin and virtue dichotomy. The group selection argument relies on “virtuous” behaviour within the group, but for those outside the group, tough luck.

Massimo Pigliucci has offered a thorough summary of the first day’s events (not that I agree with some of his analysis).