A new perspective on the financial crisis

The blogsite Corpus Callosum has an interesting new plan for the financial crisis: see here.

Prestigious but wrong?

In his book Expert Political Judgment, Philip Tetlock examined the accuracy of forecasts made by political and economic experts. Among his many interesting findings, he found that experts with a track record of media appearances were more overconfident in their judgments than their office-bound colleagues; that is, the probabilities that they attached to possible future events of interest tended to be too extreme.

Now it is claimed, by Neal Young and John Ioannidis, that academic journals are affected by a similar bias. The pressure to publish, combined with the selectivity of top journals, means that the top journals are likely to publish dramatic research that will more often turn out to be wrong. An earlier study of 49 papers published in leading journals found that nearly a third of them had been refuted by other studies within just a few years, although it should be noted that they did not look to see whether the same was true for less prestigious journals.

A short account of the Young and Ioannidis paper can also be found at the Economist online.

Sex and violence in London: the Richmond Park red deer experience

It's easy to forget sometimes that London isn't just about cinemas, theatres, music, and other urban stuff. For people interested in wildlife there is plenty to observe. One interesting place to visit is Richmond Park, especially at this time of year because their large population of deer are in their mating season. During this period the strongest stags (male red deer) and bucks (male fallow deer) try to monopolise access to a large group of females. Occasionally, fighting takes place between the males, although this mainly seems to occur when the males are fairly equally matched. But for the most part, the younger, smaller, and less experienced males generally avoid overt conflict with their more dominant counterparts. Nonetheless, they still try to obtain access to the females, hanging around some distance from the group and waiting for a likely opportunity.

Having observed the red deer on various occasions, these opportunities seem to come about in various ways. For example, a dominant male may spot another male getting too close to the females and so move to chase him away. Whilst he is engaged in this activity, other males lurking nearby may try to move in and attempt to mate with a female. In the few observational trips I've made, with my colleague Tom Dickins (an evolutionary psychologist) I have never yet seen this result in a successful mating; either the dominant male returns or the younger stag does not find a receptive female. However, presumably matings do occasionally occur, otherwise such behaviours would not be worth the risk.

In the park environment, anything that disturbs a group of females, leading them to disperse somewhat, can lead younger/smaller stags to move in and try to round up a female. Such triggers could be people getting too close to the red deer, leading them to move, or even a group of fallow deer running towards the reds (as happened when we were observing during this past weekend). The following brief clips, taken on my mobile phone (hence the rather blurry quality), show (1) a dominant male trying to chase off challengers and round up his harem following such a disruptive event, and (2) the same male running back to join his harem

Although it cannot be seen in these clips, earlier in the day this male had sustained a wound just below his right eye and that side of his head was soaked with blood. Some students that were on this trip had observed this stag in combat with another, during which he sustained the wound although succeeded in driving the other stag away.

Following the disruption shown in the above clips, this stag and his harem moved to a more central area of open grassland, not far from another group of females with its own dominant stag. These two stags kept a wary eye on each other, occasionally moving towards each other but never getting involved in a fight. One moment of disruption occurred as a large number of fallow deer came running past, at which point a couple of nearby stags tried without success to round up a couple of females. The following clip shows some of this disruption, but doesn't capture the full scale of what was happening (many more animals were within view than could be captured in the brief camera shot):

After things settled down again, our dominant stag (nicknamed "Stallion" by the students) made a couple of unsuccessful attempts to mate with one of the hinds (female red deer), but eventually succeeding on a third attempt. After this had occurred, the stag seemed to become more confident, making early challenges to young males well before they had got close to the group. The clip below shows a generally quiet state of affairs among the group.

On this particular visit to the Park, we noticed that a couple of stags kept their harems in the ferns rather than in the open. These groups seemed to be smaller, but less susceptible to challenges from roaming stags. This led us to speculate that alpha stags have more than one strategy. The most dominant ones keep larger harems in the open grassland where the grazing is better, but where there is a greater risk of challenges. Other alphas, slightly less dominant, keep their smaller harems where the grazing is poorer but where disruptions are less likely to occur. However, whether or not this hypothesis is accurate will require further visits to assess.

Book Review: The language of God

The language of God: A scientist presents evidence for belief, by Francis Collins. London: Pocket Books/Simon & Schuster (2007).




Arguments over the existence of God have been raging since Richard Dawkins published The God delusion. We are used to hearing from theologians as to why Dawkins is wrong, though to my mind they often seem to repeat the same arguments that Dawkins himself takes apart in his book. What is far less common is to hear the case for God made by eminent scientists. That is why The language of God caught my attention.


Francis Collins is one of the world's leading scientists. In particular, he headed up the Human Genome Project, ensuring that the results were placed in the public domain, thus heading off the threat (as many saw it) of the genome being patented by Craig Venter's private enterprise initiative. For this, he and his team certainly deserve all our thanks. But Collins is also an atheist-turned-Christian. In this book he argues for both evolution and God, thus challenging both the biblical literalists (and other varieties of creationist) and those who would argue that God has effectively been abolished as the creator of life.

The language of God makes an impressive case for the evolutionary view of life, focusing particularly on refuting the claims made by the proponents of Intelligent Design (ID) and young earth creationism (the latter term distinguishes those who believe the earth is just a few thousand years old from those who believe that God set up the conditions for evolution to occur). For example, a key claim made by ID proponents is that the structures of organisms show "irreducible complexity". This means that, whilst it might be plausible for a single component to have evolved, there are actually multiple interlinked components and the probability that these could have coevolved is vanishingly small.

Darwin himself fretted that his readers would find it hard to accept that a structure as complex as the eye could have evolved. However, a study of different organisms shows how the eye could have developed in a series of discrete stages. Even the very simplest organisms derive some advantage from a pit containing a few light sensitive cells. In other organisms an additional advantage has been conferred because this pit has evolved into a cavity with just a "pinhole" to admit light. In yet other organisms, the addition of a jelly-like substance allows for a degree of focusing. And so on. It is also worth noting that the eye, had it been deliberately designed, would have to be considered something of a botched job, given that the light receptive rods and cones are placed beneath a layer of nerves and blood vessels (and the positioning of the optic nerve means that we have blind spots in our visual field).

Collins also notes that the "poster child" for ID, the bacterial flagellum can be accounted for by natural selection. This has multiple interacting parts which, the ID people say, could never have co-evolved. But protein sequencing of different bacteria shows that various components of the bacterial flagellum are related to an entirely different apparatus. So, just as natural selection predicts, this apparatus has been constructed by taking components that were used for one particular function and co-opting them for use in an entirely new function.

In short, the arguments put forward by the proponents of ID rest on invoking a "God of the gaps". The strategy is to identify a phenomenon that has not (allegedly) been explained by science and then invoke an intelligent designer (God) to account for the phenomenon. This is a flawed kind of argumentation; just because a phenomenon has not yet been explained by science does not mean that an explanation will not be forthcoming. And, as noted, science is increasingly showing how complex structures have evolved from simpler ones, just as Darwin predicted.

Francis Collins sees no conflict between his acceptance of evolution and his belief in God. He takes the position that God, in creating the universe, set up the conditions for evolution to occur. Collins notes that many among the religious faithful might find it hard to understand why God would have used such "an apparently random, potentially heartless, and inefficient process as Darwinian evolution" (p.204). But for Collins, God truly is a supernatural entity, someone who is outside of nature, outside of space and time, and can therefore know every detail of how evolution will unfold. From God's perspective, there is no randomness; everything is specified.

This is all well and good, but skeptics will ask "Why believe in God in the first place?" (noting that the book is subtitled A scientist presents evidence for belief). Collins' main argument is presented in the first chapter, titled From atheism to belief, and is referred to at various points throughout the book. In human behaviour Collins sees evidence for something he refers to as "The Moral law". I struggled to find a definition of this law; given that laws in science are precisely defined, this was something of a surprise. However, what Collins seems to be flagging up is that humans have a sense of right and wrong, and that we sometimes show selfless altruism - what C.S. Lewis (much cited in the book) called "agape" (pronounced ah-GAH-pay). The rather woolly specification of this Moral Law is rather worrying, though, given some rather strong claims that are later made, such as the suggestion that the religion you hold does not matter in bioethical debates because "The Moral Law speaks to all of us, whether or not we agree on its origins. Basic principles of ethics can be derived from the Moral Law, and are universal." (p.244)

Collins claims that we have no evolutionary or cultural explanation for why people would show selfless altruism, and again utilises an argument made by C.S. Lewis; this says that if God is outside the universe then he would not somehow reveal himself within the physical universe, but he would show himself within ourselves as an influence or a command to behave in a particular way. But as I read this chapter the only voice echoing within my own head was saying: "It's the God of the gaps, the God of the gaps!" In fact, just as the claims about irreducible complexity are being show to be wrong, so a resurgence of work in the psychology of moral judgment is showing just how selfless altruism could come about (see for example this review article by Krebs).

For example, evidence suggests that our early ancestors lived in groups of 100-150 people, where most people would have been related to each other to varying degrees. In such conditions, helping another person is likely to be equivalent to aiding the propagation of one's own genes. Thus, altruistic behaviour in modern life may occur partly because we are predisposed to regard those around us as our kin. Needless to say, though, people do not always behave in an altruistic fashion and, here again, psychologists are increasingly specifiying the conditions under which altruism is more or less likely to happen.

Regardless of how well advanced one regards the psychology of moral behaviour, the fact remains that Francis Collins' argument is invoking the God of the gaps, the very thing that he himself argues against later in the book when refuting the claims of the ID proponents. He also argues that the atheist position is a weak one because belief in a God of some kind is found in all cultures, thus it is atheists who need to justify their views more than theists. But, aside from the fact that not all cultures do believe in a deity as such (some preferring to believe in spirits of the earth, for instance), Collins argument here is essentially a version of the "Argument from Popular Opinion". This is always a weak argument, verging on a fallacy. Douglas Walton notes that this argument type can be strengthened where some of those asserting a particular view are "in a position to know", but that hardly seems to be the case here - in fact, some would regard it as blasphemous to claim to know God's intentions (though much of the time this is exactly what relgious people do).

Of course, the question as to why so many people hold supernatural beliefs, especially those involving belief in a God or Gods, is itself a question of interest to scientists. However, reading The language of God, I had the feeling that psychology is somewhat off the radar for Collins. In effect, he is arguing that the physical structure of organisms is the result of evolution, a process that was set up by God who had the foresight to know how things will work out. At the same time, he is arguing that "The Moral Law" cannot be explained by evolution and has somehow been implanted in us by God. But given that our thoughts and impulses, conscious and unconscious, arise through the operations of the brain and nervous system, and given that these are physical structures, then it is contradictory to invoke evolution for one but not the other.

In short, I found the arguments of the first chapter so weak that I almost did not make it to the rest of the book. In that sense, I cannot imagine that many atheist evolutionists will become theists as a result of reading this book. On the other hand, if this book is able to convince any young earth creationists / ID believers that the evolutionary viewpoint is correct, then that can only be a good thing. And it certainly is the case that Collins does an excellent job of showing why their views are untenable. But given how trenchant people's views tend to be on this topic, I wouldn't bet money as to how many creationists do turn their back on their views.

Science at the movies

A wonderful modern example of film noir is the movie Memento, in which the lead character - apparently following a head trauma - is unable to form new memories. The film follows the character Lenny (played by Guy Pearce) as he tries to track down the person responsible for killing his wife and injuring him.

As a psychologist, it is interesting to note that "Lenny" describes himself as suffering from a short-term memory problem. The DVD box also describes Lenny as having short-term memory loss, as does the Internet Movie Database, though Wikipedia does not (at the time of writing). In fact, as defined by psychologists Lenny does not have any problem with short term memory. His short-term memory is working just fine. Short term memory (or working memory as it is now called) is a memory store of limited capacity and duration. It can hold approximately seven items of information for about 30 seconds, without rehearsal. For example, when you first hear a telephone number you may just about be able to remember it for half-a-minute, if you're lucky. Lenny's problem is that he is unable to create new long-term memories, memories that last for more than about half-a-minute.

Real-life examples of such individuals are HM and Clive Wearing (various clips on YouTube, for example: here). In fiction, this anterograde amnesia makes for a great story, but in real life the condition is clearly far more debilitating to the individual.

Anyway, I've been sent off along this train of thought, about science in the movies, by watching the sci-fi/horror film The Fly once again (the David Cronenberg version that is, not the original). In The Fly a scientist studying bodily transportation through space accidentally merges his DNA with that of a fly. On this repeat viewing it suddenly dawned on me that the physical changes in the lead character (played by Jeff Goldblum) are preceded by psychological changes. Increasingly, the instincts of a fly begin to dominate those of the human. Eventually, the fly's instincts completely take over, although in physical terms there is more of a mixture of characteristics - a fly-like creature of human size.

In the middle part of the 20th century, a school of thought known as radical behaviourism argued against the concept of instincts and, indeed, any reference to "the mind". These behaviourists argued that scientists should only study things that could be directly observed, i.e. stimuli and responses to stimuli. In due course, though, this approach proved untenable. For example, Breland and Breland, students of the famous behaviourist B.F. Skinner, reported that many animals could not be trained using behaviourist principles because these organisms had instinctive behaviour patterns that were hard to overcome. For example, a racoon could not be trained to put two coins into a box; instead, it would only rub the two coins together. Breland and Breland argued that the behaviour of organisms could not be understood without reference to such instinctive behaviours, the evolutionary history of the organism, and its ecological niche.

Evolution also gets a mention in The Planet of the Apes. In this movie, astronaut Charlton Heston crash-lands his spacecraft about 2000 years into the future, though he does not realise - until the end of the film - that he is still on planet Earth. In this future time, the planet is dominated by highly evolved apes, whereas humans have become the slaves of the apes and have lost their capacity for language. In an interesting parallel with the modern arguments about evolution, it is a heresy to even suggest that apes may have evolved from humans. Although there is evidence to show that this is in fact the case, no-one is allowed to mention it. The sites containing the evidence are out-of-bounds and, at the end of the movie, are destroyed.

It is fun to consider the philosophical underpinnings and the level of realism of such movies. For instance, if human DNA and fly DNA were merged as in the movie, would it even be possible for a dramatically transformed living creature to be produced? What kind of behaviour might result? To what extent are we even now ruled by our instincts? Do we have the free will to completely overcome our instincts? In The Planet of The Apes, it is never explained why apes have become dominant or why humans have lost their capacity for language. What environmental circumstances might have caused this? Could such a change have come about in the time period since the fall of human civilisation (about a couple of thousand years)?

Minds over markets?

Whilst on my way home tonight I saw this advert for managed funds (at Stratford station in London, UK). Note that the last line of the small print does say that investments can fall as well as rise, but the large print has a fairly clear message.

I have no idea how well or badly this particular group has performed over a long period. However, in the market as a whole it is not possible to predict which managed funds will beat the market. Just by chance, some funds do beat the market index in any given year but most funds don't. And beating the market in one year does not predict success in the next year.

In his book "A mathematician plays the market", John Allen Paulos asks us to consider two stock pickers, Tom and Harry. Both are actually performing at chance level, but nonetheless Tom outperforms Harry over the course of the year. And this is exactly what we would expect from a chance process. You can see this for yourself through a simple coin tossing exercise. Toss a coin 101 times and keep a running tally of how many heads and how many tails there have been. At any given point in the sequence of coin tosses either Heads or Tails will be in "the lead" or there will have been even numbers of both. Most people seem to think that the lead will be continually changing, but actually such sequences typically produce very few changes of lead. It's not uncommon for one side of the coin, say Tails, to go into the lead on the very first coin toss and stay in the lead for the remainder of the trials. In fact, as far as a precise number of lead changes is concerned, the most likely number of such changes is zero. This counter-intuitive result was shown to be true by the mathematician William Feller in the 1950s. A popular account is given by John Haigh in "Taking chances: Winning with probability" (see also "A random walk down Wall Street", by Burton Malkiel). Of course, over increasing numbers of such sequences we'd expect the overall proportion of Heads and Tails to converge towards 50%, but in any given sequence the overall proportion is rarely exactly 50% and - as just mentioned - there are typically very few changes of lead.

So our stock pickers Tom and Harry are rather like our coins. In any given year, just by chance, one is most likely to have outperformed the other and is also very likely to have done so consistently across the year. People may well form the conclusion that Tom is a better stock picker, particularly as Tom will now place lots of advertisements flagging up his success.

Musings on the credit crunch: Built to last?

Why is it that some financial organisations are weathering the financial storm better than others? In the UK, the Spanish Santander banking group has just bought up the branches and deposits of the failing Bradford and Bingley building society. It seems that Santander is going from strength to strength, and the reason according to an article in the Independent newspaper is a combination of the strictly regulated nature of the Spanish banking system, the fact that Santander has avoided getting involved in the complicated and risky forms of loans that are now biting other organisations, and that Santander has been profiting from buying up a string of ailing organisations.

All this got me to thinking about business in general and the question of why it is that some companies are more successful than others. The business section of bookshops are full of tomes extolling the secret of business success. Of course, the skeptical mind cannot help wondering why, if there is a "menu" for creating long-lasting success, not everyone is managing to do this? I am reminded of what the racing driver Lewis Hamilton said recently, to the effect that he was determined to "cut out the errors" in order to win the F1 world championship. If it is so straightforward to not make errors, why were errors made in the first place?

It turns out, though, that most of the authors of business success books are little more than snake oil salesmen. Perhaps I'm being a little unfair here, as I'm not suggesting they are setting out to deceive, but nonetheless they appear to be deluding themselves and thus flogging nonsense to others. Phil Rosenzweig has analysed the research behind business success books in "The halo effect... and the eight other business delusions that deceive managers". One of the key flaws behind much of the research is that authors frequently analyse the performance of only successful companies. So if you find that most of these companies are led by a CEO who has a burning passion for the company mission, then the conclusion is that this is a behaviour that helps create company success. But wait a minute, what if the leaders of failing companies also have a burning passion for the company's mission? Well, who knows, because those companies weren't researched.

Another key flaw - and this is where the halo effect comes in - is that much of the research involves interviewing the leaders and employees of these successful companies. In these successful companies, both groups of people have good things to say about the way that the organisation goes about its business, about the enthusiasm and teamwork of everyone involved, and so on. But the problem here is that these comments are made in the full knowledge of the organisation's performance. That is, the story being told about the way the company operates is coloured by knowledge of its success. Rosenzweig describes experimental research in which people worked in groups on a problem solving exercise. Each group was then given feedback, except that the feedback wasn't genuine but randomly determined for each group. Group members then rated how well they had done on a range of issues. In a nutshell, groups who had randomly received positive performance feedback thought that they had worked together well as a team, whereas those randomly receiveing negative feedback did not.

It also turns out that the companies identied as "great" or "visionary" in books such as Built to Last turn out to have a long-term performance that is nothing special in relation to the market as a whole. In fact, the main factor in operation appears to be regression to the mean. For example, the book In Search of Excellence, published in 1982, identified 43 "excellent" American companies. Rosenzweig follwed up the performance of these companies post-1980 (when the original study ended). By 1984, during which time the S&P 500 index almost doubled, only 12 of the "excellent" companies outperformed the market, whilst the other 23 failed to keep up. Similar figures were observed for the period to 1989. (By the way, the now-troubled Fannie Mae was just one of 11 companies to be identied has having gone "From Good to Great" in the book of that name).

To take the title of another book, are some companies "built to last"? Rosenzweig notes that of the 500 S&P companies existing in 1957, just 74 were still in the S&P forty years later: "The other 426 were gone - nudged aside by other companies, or acquired, or bankrupt. And of the 74 survivors, guess how many outperformed the S&P 500 over that time period? Only 12 out of 74. The other 62 survived, yes, but didn't thrive." (Rosenzweig, p.102). Thus, the evidence indicates that the companies that last the longest are not actually the highest performers.

More rigorous research indicates that managerial performance does have some influence on company performance, but to such a modest level that it shouldn't be the stuff of headlines or bestsellers.

What the figures above indicate are that it is incredibly hard to predict the performance of companies. But when things go wrong, it seems everyone has a diagnosis, usually one of two polar opposites: either a company strayed too far from its core, or it failed to innovate. Much of the time, though, it's not clear whether these explanations really hold water at all. Indeed, pundits sometimes want to have it both ways, criticising a company for being unadventurous, but then when the company does something innovative and still suffers falling performance, the opposite criticism is introduced - they strayed too far from their traditional customer base.

I note that here in the UK, prior to the credit crunch, the opposition party often claimed that the successes of the economy were little to do with the government, but were the result of policies that were put in place when the previous government was in power. Now the opposition story is different - the current crisis is all the fault of the government who have failed to regulate the City! I'm no expert, but I can't help feeling the truth may lie somewhere in between. Whilst few people predicted the current problems, nonetheless there have been some mutterings about the potential problems caused by the complexity of derivatives. On the other hand, during times when things are going really well it would be a brave government who says they're not going to allow their organisations to engage in the same practices that seem to be working elsewhere (though see the comment about Spain and Santander earlier). I find it hard to imagine that most opposition parties would have been doing anything different if they had been in power.

I'll finish with this quote from the British investment banker David Freud, about his dealings in the City:

"The currency was not cash but chaos. Transactions invariably took place at the edge of feasibility conducted under a competitive background under great time pressure. I found few committees of experts considering all the available evidence in wise conclave. Much more typical were decisions taken on the fly, by whoever happened to be available, based on a fraction of the full information. "
(From "Freud in the City", 2006, pp.355-6)