Links this week:

1. Hodgson and Knudsen have set up a reading group for their book Darwin’s Conjecture: The Search for General Principles of Social and Economic Evolution. Chapter one has already kicked off.
2. Another from The Umlaut – Conspicuous Frugality.
3. Flip-flopping selection pressure in a modern population.
4. Following from my post on Douglas Kenrick and colleagues’ theory of Deep Rationality, below are a couple of short videos – one on How Mating and Self-Protection Motives Alter Loss Aversion, and the other on the upcoming book The Rational Animal: How Evolution Made Us Smarter Than We Think by Kenrick and Vlad Griskevicius, which also looks like it covers similar territory.

I have posted before about Gary Becker’s argument that the evolution of altruism can be explained by a version of his rotten kid theorem. In short, if an altruist cares about other people’s welfare in addition to their own and is willing to transfer their resources to others, an egoist’s action to harm the altruist may also harm the egoist as the amount that the altruist would be willing to transfer to the egoist will be reduced. As a result, the egoist will refrain from hurting the altruist, making the altruist better off than if they were an egoist.

As I raised in my post, Becker’s argument can run into corner solutions, whereby the scale of the gain to the egoist and damage to the altruist are such that the egoist is willing to harm the altruist. However, I only recently realised that the year after publication of Becker’s article, the Journal of Economic Literature published responses to Becker’s paper by Jack Hirshleifer and Gordon Tullock, along with a reply by Becker.

Hirshleifer’s critique focuses on the order in which the altruist and egoist’s actions occur. If the egoist has the last word, they will likely take advantage of the altruist at the end, meaning that the altruist should not be as altruistic to begin with. In the case of altruism between a parent and child, the child will normally have the last word due to differences in ages. Tools such as a legal system that allows the making of wills are required for the altruist to have the last word.

Hirshleifer then goes on to show that Becker’s analysis is powerful where the altruist can keep the last word, and an altruist may be selfishly better off than if they were planning an egoistic action. Their altruism restrains the behaviour of the egoist.

In reply, Becker called Hirshleifer’s comments perceptive, with the note that his scenario can only work among a small number of relatives or neighbours (so Becker had already dealt with part of my criticism).

After quibbling with the definition of altruism, Tullock’s criticism focuses on an altruist’s ability to know the preference ordering of the recipient of the altruism. Tullock suggests judgments of this type are almost impossible. The history of charitable administration and its attempts to prevent recipients from taking advantage of gifts suggests a knowledge problem.

Tullock argues that in this case, the egoist can abuse the situation. For example, they could stop working, which reduces their income considerably but their utility only slightly as they no longer have to work. If the altruist only looks at the slacker’s loss of income, the altruist may effectively overcompensate the egoist.

Tullock also talks of the potential for corner solutions based on different orderings of the size of the gain of the egoist, damage to the altruist and transfer from the altruist. He notes that Becker’s scenario can only occur where the damage inflicted on the donor is greater in size than the gift that would be given to the egoist in the absence of damage, which is greater than the gain to the egoist from damaging the altruist. This is only one of six possible orderings (although two of the others involving no or almost no harm to the altruist are the most common).

Becker’s reply to Tullock is sharp. “Although Tullock’s comment is much longer than Hirshleifer’s, it is less focused and less useful, and my response shall be brief.” His dismissal of Tullock’s argument is largely on the basis that Becker is not seeking to explain altruism to people 1,000 miles away, but rather to kin and close neighbours about whom the altruist will have more knowledge. Further, while this is a restrictive class, Becker (rightfully) considers it an important one.

Becker’s point on knowledge of kin and neighbours does not completely nullify Tullock’s, as parents have imperfect knowledge of their children. That same criticism could be applied to kin selection, which requires some degree of understanding of what actions will benefit kin. However, Tullock may be comfortable that the criticism would also apply to kin selection given that he prefers group selection based explanations of altruism and cooperation.

Even though I consider that I am across the literature at the boundary of economics and evolutionary biology, now and then an article pops up that I somehow missed. The latest article of this type is a 2009 article by Douglas Kenrick and colleagues, titled (as is this post) Deep Rationality: The Evolutionary Economics of Decision Making. I found it through Dan Ariely’s reading list for his Coursera course A Beginner’s Guide to Irrational Behaviour. Kenrick has also posted on the article over at his blog

I don’t feel overly guilty about not seeing this article earlier, as the authors have not referenced a lot of the literature in economics that I would consider relevant. Regardless, there is a lot to like about this article, particularly the way that it looks to incorporate an evolutionary approach into behavioural economics. I have often posted my criticism that much behavioural economics lacks a framework, without which it is just a list of biases and heuristics. It is good to see someone trying to offer that framework.

The authors’ basic argument is that people have evolved domain specific decision rules. Decisions depend on the current environment, plus the decision maker’s sex, mating strategy and stage in the life cycle. As a result, many decisions that are called inconsistent or irrational in behavioural economics are actually “deeply rational” to the domain in which the decision is being made.

In making their case, the authors start out with a brief kick at economics by noting that most economic theorists “have remained relatively agnostic about the roots of utility.” They do note the work of Gandolfi, Gandolfi and Barash, but otherwise do not mention the wealth of articles on the evolution of preferences by the likes of Arthur Robson, Larry Samuelson and others (my economics and evolutionary biology reading list gives a taste). Thus, when they suggest that we need to go deeper than Gandolfi, Gandolfi and Barash’s approach of equating utility to fitness, they miss some literature which does just that.

Regardless, the need to go beyond “fitness equals utility” by considering factors such as life history or differences in mating strategy is important. The authors suggest that we should consider human decision making as being geared to solve recurrent adaptive problems in different domains, whereby successful solutions in each are associated with increased fitness. The body of their article focuses on some examples of this approach.

In one section, they address attitudes to risk. Humans are normally risk averse, which Kenrick and colleagues suggest is consistent with empirical observations of loss aversion. Although this shorthand equating of risk aversion and loss aversion works some of the time, it sells these concepts short, along with the way that they are incorporated into Kahneman and Tversky’s prospect theory. Under prospect theory, people evaluate choices from a reference point, they show loss aversion (losses hurt more than gains) and they are risk averse when faced with two potential gains. However, in the domain of losses, they are actually risk seeking. When you combine these features with the human tendency to overweigh small probabilities, you obtain the fourfold pattern of risk attitudes. When an agent faces a moderate probability of a gain or a small probability of a loss, they will be risk averse. However, when faced with a low probability of a gain or a moderate probability of a loss, they will be risk seeking.

Kenrick and colleagues do make the important point that the attitudes to risk as predicted by prospect theory will vary with evolutionarily relevant factors. Men with mating motives will be more likely to take financial risks.  Women would not respond in the same way as women know that men give a lower value to the resources possessed by a mate. In the social domain, such as networks of friends, there tends to be loss aversion in both sexes, although this may reverse for men with mating motives.

This is a point of the article where a hat tip to the existing literature might have been most useful, as some economists have considered the evolutionary foundations of attitudes to risk. For example, Rubin and Paul examined the effect of mating motives on risk preferences in 1979. They developed a model where male fitness depended on attracting a mate, which was in turn a function of their resources (income). Rubin and Paul suggested that young men who do not have a mate are likely to be risk seeking in obtaining income as they have no mate to lose. Older men who already have a mate will tend to be risk averse, particularly given the huge level of income required to attract a second mate.

In another section, Kenrick and colleagues look at the economic approach to choosing a basket of goods within a budget constraint. They argue that the weighting of each good will depend upon the domain in which an agent is making a mate choice. For example, promotion of a colleague at work may influence status motives and accordingly, the worker’s preferences between more time in the office and leisure will shift.

They also make the interesting distinction between traits in a potential mate being necessities or luxuries. Consider a female who needs a male to have a minimal level of resources to make sure her offspring survive. Due to diminishing marginal utility (another economic concept) as the male’s resources increase, she may start to look at other traits if there are plenty of males with enough resources. The pattern of consumption will be that resources are a necessity, while other traits are luxuries. A similar pattern might emerge for male preferences, initially prioritising fertility related traits, but then considering other traits if there are plentiful fertile females. Thus, when the necessity traits are scarce, we might expect large sex differences in mate preferences as each sex focuses on obtaining their different necessities. As these traits become more plentiful, traits that are luxuries are sought. If there is overlap between the luxuries of one sex with the necessities of the other sex, we would see smaller differences between the sexes in the traits sought in mates.

One issue Kenrick and colleagues do not spend much time on is why evolution has shaped domain-specific and not general decision rules. This is addressed in the evolutionary psychology literature, but to sell their argument to economists, they need to sell them the constraint inherent in the modular approach. Most evolutionary analysis of economic preferences struggles to incorporate “irrationality” through constraints, often due to a view that evolution is the ultimate rationality machine (and most economists fixation, conscious or not, with rationality). Selling to economists the picture of constrained, path-dependent evolution that leads to modular decision making and “deep rationality” could improve the economic endeavour considerably.

Links this week:

1. From The Umlaut (worth adding to your feed) – Why Choosing to Make Less Money Is Easier Than Ever. The return of conspicuous leisure?
2. A new book Forecast: What Physics, Meteorology, and the Natural Sciences Can Teach Us About Economics may be worth a look (HT: Diane Coyle).
3. Should human genes be patented? I wonder how much of this debate will be surpassed by technological progress. Cheap genomes and the ability to upload to free or near free websites for analysis will make attempts to prevent unauthorised testing of a patented gene impossible.
4. I’ve updated my economics and evolutionary biology reading list, including adding a couple more articles on the evolution of preferences and genoeconomics.

One of the few areas where there is active research on the link between evolutionary biology and economics is the evolution of economic preferences (some papers in this area are in my economics and evolutionary biology reading list). Economic preferences are the way an actor will rank a set of choices based on characteristics such as the amount received, the probability of certain outcomes, and the timing with which the outcomes are received.

Preferences about the timing of an outcome are known as time preference (often measured as a discount rate). Someone who values goods now much more than goods received later has a high rate of time preference, while someone who gives goods in the future a relatively higher value has a low rate of time preference. We would call someone with a low rate of time preference patient.

So, what rate of time preference would have evolved under the forces of natural selection? Discounting the future makes sense in an evolutionary context as future outcomes might not be realised, population growth might make benefits received in the future worth less, and there is the chance of events such as death.

One of the seminal papers in the analysis of the evolution of time preference is Alan Rogers’s Evolution of Time Preference by Natural Selection (ungated version here). Time preference had been touched on before by R.A. Fisher, and in an earlier paper by Hansson and Stuart who examined intergenerational time preference. But Rogers’s paper was the first to look at this question on an intragenerational basis, which is the context in which economists usually consider it.

Rogers examined the optimal same-age transfer  that would be made from a mother to her daughter (e.g. from the 20-year old mother to the 20-year old daughter). In making such decisions, the mother would need to consider the remaining reproductive life of her daughter, that her daughter is only 50 per cent related to the mother, and the rate of population growth. As the transfer is same-age, the mother and daughter have the same remaining reproductive life. If there is no population growth (which was effectively the case for most of human history), only the degree of relatedness would matter and the discount factor is effectively one half per generation. Under these conditions, Rogers argued that the long-term real interest rate should be about 2 per cent per year.

As for the analysis by Hansson and Stuart, this rate appears low against measured discount rates, which are typically above 10 per cent per year. A simple static analysis of this nature is missing something. Some other papers grapple with this issue, and I will post about them soon.

Subsequently, Robson and Szentes argued that there are “serious problems with Rogers’ model” (ungated version here and extended version here), and that a particular rate of time preference should not flow from the analysis. They argued that unrealistic assumptions drive the results, including the assumption of identical offspring, which is not the case when offspring vary with age, and the assumption of a single same-age transfer where there are many possible same-age transfers (e.g. 20-year old mother to 20-year-old offspring, 30-year old mother to 30-year old offspring). Robson and Szentes also showed that the rate of time preference would depend upon the nature of the survival function faced at each age (i.e. the probability of death).

Regardless, Rogers’s paper is an important one, and despite a couple of earlier pieces of work on the evolution of time preference by other authors, Rogers’s paper is often seen as the seminal paper that kick started the evolutionary analysis of preferences. That is not a bad legacy to have.

Update: Alan Rogers responds in the comments.

John Brockman has wheeled out another good bunch of experts for the newest Edge question “What’s the question about your field that you dread being asked?”

One response by Richard Thaler is particularly interesting, who fears being asked “When will there be a single unified ‘behavioral’ theory of economic activity?” For those who know Thaler’s work in behavioural economics, his reason might be surprising:

If you want a single, unified theory of economic behavior we already have the best one available, the selfish, rational agent model. For simplicity and elegance this cannot be beat. Expected utility theory is a great example of such a theory. von Neumann was no dummy! And if you want to teach someone how to make good decisions under uncertainty, you should teach them to maximize expected utility.

Obviously, Thaler knows that this model is not perfect:

The problem comes if, instead of trying to advise them how to make decisions, you are trying to predict what they will actually do. Expected utility theory is not as good for this task.

However, Thaler is not convinced that alternatives such as prospect theory are up for the task, and he suggests that there will ultimately be a multitude of theories:

Just as psychology has no unified theory but rather a multitude of findings and theories, so behavioral economics will have a multitude of theories and variations on those theories. You need to know both physics and engineering to be able to build a structurally sound bridge, and as far as I know there is no general theory of structural engineering. But (most) bridges are still standing. As economics becomes more like engineering, it will become more useful, but it will not have a unified theory.

Thaler is being overly pessimistic – and I’m not sure that there are many theories of bridge building that can ignore the unifying framework of physics. He is right that the rational agent model is simple, elegant and powerful. The problem is that while behavioural economics can pick holes in the model on the basis of predicting how people make decisions, there has been limited attempt to generate a unified theory. Prospect theory is a useful tool for predicting behaviour, but the question that is rarely asked is why people act in that way.

I am optimistic about the role that evolutionary biology will play in filling this gap. Evolution is the ultimate rationality machine, and any actions that are not rational will be ruthlessly eliminated. This is what lies behind the power of the rational agent model. But evolution can only work with the material at hand, leading to a constrained rationality. Heuristics that use less energy and time can be favoured. Many adaptations are path dependent (Robert Frank’s Passions Within Reason gives one excellent account of how path dependence might have shaped human emotions). A changed environment can result in decisions that were once rational no longer being optimal.

Thaler points to the multitude of theories in psychology as an example, but psychology is now being reconstructed by evolutionary psychology, with many of the available theories unable to withstand the light of evolutionary theory. Economics, and more particularly behavioural economics, is slowly being examined using evolutionary theory and the unifying basis of human decision-making as an evolved trait. Those theories inconsistent with our evolved past will be discarded, and the commonality between those that remain will provide considerable unification across the field.

Links this week:

1. E.O. Wilson’s argument that great scientists don’t need math has already received plenty of responses. Wilson’s argument reminded me of one of Paul Krugman’s critiques of Stephen Jay Gould’s popular work, which were “literary confection” as they lacked math.
2. A free webinar with Geoffrey Miller and others on What Every Marketer Should Know About the Nonconscious.
3. Diane Coyle posts on a paper by Sergio Da Silva in which he argues that “economics fails to ground itself in the underlying knowledge provided by biology”.
4. While researching a paper, I came across this 20 year-old piece by Jared Diamond on the isolation and technological regress of the Tasmanian aboriginals.

Since the time of Darwin, the same evolutionary psychology debates have played out over and over. Here is Ronald A. Fisher in The Genetical Theory of Natural Selection (1930), addressing the type of argument that you can still hear today about the evolution of the human mind:

[I]t is often felt to be derogatory to human nature, and especially to such attributes as man most highly values — as if I had said that the human brain was not more important than the trunk of an elephant, or as if I had said that it ought not to be more important to us, if only we were as rational as we should be. These statements would be unnecessarily provocative: in addition they are scientifically void. And lest there should be any doubt upon a matter, which does not in the least concern science, I may add that, being a man myself, I have never had the least doubt as to the importance of the human race, of their mental and moral characteristics, and in particular of human intellect, honour, love, generosity and saintliness, wherever these precious qualities may be recognized. … [N]atural causation … introduces the strongest motive for striving to know, as accurately and distinctly as possible, in what ways natural causes have acted in their evolutionary upbuilding, and do now act in making them more or less abundant.

Less played out today are Fisher’s concerns about the direction of the selective forces acting upon our mental traits. Consider the interaction between the desire for economic acquisition and fertility, for which Fisher appears equivocal about which way the selective pressures might be acting. First he notes that those with economics ambition tend to curtail their fertility:

Parents in whom economic ambition is strong, will, in like circumstances, be more inclined to limit their families than those in whom it is weak. Consequently a progressive weakening of the economic ambition, or at least in the average intensity with which this motive is felt among the great body of citizens, is to be expected as a concomitant to the strengthening of the moral aversion towards family limitation.

But where there is potential for famine, economic ambition is required to survive.

The attitude of men and women towards their economic welfare cannot, however, be ordinarily reduced by this cause to indifference, for in countries in which the poorest class are frequently decimated by famine, it is apparent that a stage will be reached at which what is gained in the birth-rate is lost in the death-rate; and even where the extremes of distress are ordinarily avoided, some loss of civil liberty, and of the opportunities for reproduction, has been the common effect of indigence.

And even in developed countries, acquisitive impulses may be required.

Moreover the rational pursuit of economic advantage must, even in the most civilized countries, frequently place the individual in a position favourable to normal reproduction. It would, apparently, in most societies, be as disastrous to the biological prospects of the individual to lack entirely the acquisitive instincts as to lack the primary impulses of sex, notwithstanding that the abuse of either passion must meet with counter-selection. The moral attitude of civilized peoples towards money, as towards sex, must be therefore the product of much more complicated evolutionary forces than is his attitude towards infanticide or feticide, as might perhaps be inferred from the hypocrisy and fanaticism, the passions and the passionate inhibitions, found among long-civilized peoples on both subjects.

When we experience positive events, we feel happy. But happiness adjusts, with the effects of a positive event normally short-lived. Over the long-term, happiness tends to float around a stable mean. Happiness is also strongly related to our position relative to our peers. How happy we are with our income depends on everyone else’s income.

In line with the first law of behavioural genetics, it is worth looking for an evolutionary foundation to this pattern. How does happiness motivate us to do things in our evolutionary interest?

Evolution did not shape our happiness to simply increase or decrease in line with how events affect our fitness. We need a more nuanced explanation, which Luis Rayo and Gary Becker offered in two papers published in 2007 by asking how our ability to feel happiness would be affected if it is constrained. The long form of their argument is contained in the Journal of Political Economy, with a shorter version published in the American Economic Review Proceedings and Papers.

Rayo and Becker propose two potential constraints. First, there are limits to a person’s sensitivity to happiness. A person can only determine which of two alternative choices they should make if there is more than a certain size difference in happiness for the two choices. Second, there is a bound on the range of happiness that a person can experience (say, limits to nervous system signal strength).

To overcome limits to a person’s sensitivity, evolution could amplify the happiness response to make sure that we knew which of two choices made us happy. But if there is a limit as to how happy we can feel, this solution will not always work. Combining the two constraints, the strength of the happiness signal should be strong where it matters most – over the current relative decisions.

Rayo and Becker relate a couple of cases that are similar. If you move from the sunlight into a dark room, you initially can’t see anything, but your eyes adjust until you can distinguish between the relative shades. Another example comes from Arthur Robson, who likens happiness to a voltmeter. When you are about to measure an electric current, you must first set the voltmeter to the range in which you want to measure the current. If you set the range too high, the meter will barely move. If you set it too low, the reading will go instantly to the maximum value. The voltmeter must first be calibrated to the problem at hand.

To formalise this idea, Rayo and Becker developed several “happiness functions”. In one function, the agents first compare their income against their peers to determine their current social position. They then contrast their current social position against their relative social position in the last period. An advance in social position leads to happiness but it is only short-lived.

Under this function  a general increase in income across society does not increase happiness (consistent with the Easterlin paradox), and happiness will tend to revert to a mean. However, given recent arguments that the Easterlin paradox is an artefact of having happiness measured on a bounded scale, Rayo and Becker’s argument may need to more finely tuned.

This happiness function is also consistent with the positive correlation between income and happiness sometimes observed in cross-section data. People are subject to random shocks and those who have higher income are more likely to have received a recent positive shock.

One thing I didn’t enjoy about the Journal of Political Economy article is that it follows a tradition in much work on the evolution of economic preferences by using a metaphorical principal-agent approach to the analysis. The principal is nature, while the agent is the individual being shaped by evolution.  I’ve never been a fan of this approach, which is generally not adopted in evolutionary biology. It lessens the accessibility of what is often already hard to access work, and I am not convinced that any pay-off from the additional complication is worth it. I’ll post some longer thoughts on this soon.

Links this week:

1. There are plenty of good looking MOOCs popping up. John Hawks has announced his MOOC Human Evolution: Past and Future, starting early 2014. Dan Ariely’s behavioural economics course kicked off a couple of weeks ago, and this course on social and economic networks by Matthew Jackson looks worth a look. I seem to be adopting a speed learning strategy for MOOCs – wait until all the material is up and then binge.
2. How will the Flynn effect and the bright tax balance out? We’re looking OK until 2042.
3. Slavery and capitalism.