EP is a subject that fascinates me and I like to discuss it. Even if you haven’t studied this the term is fairly self-evident. You might not know that in the broadest sense of the term it applies to all organisms, not just humans.
Evolution driven by natural selection is a fairly easy idea to grasp. But in specific cases it can be quite difficult to understand how a particular trait confers a selection advantage and that’s the fun part.
One starts the whole EP process by assuming that individual creatures are selfish. They do what helps them to survive and reproduce without regard to any other creature, including their own species, even their “kin”. With this assumption then does ‘altruism’ ever exist? – even as somehow “enlightened self-interest”
An example is a particular type of small bird that flocks with others for all its activities. The bird has a natural predator, some kind of raptor. Naturally these birds are good at spotting the raptor as it is out hunting for them.
The birds eat together as a flock in open fields. They’re not camouflaged while eating so it’s easy for the raptor to spot them. And they’re busy eating so they’re not always looking up. But the birds have a “trait” (a messy term, but “gene” isn’t really right either). When a bird sees the raptor it makes a warning call and then prepares to fly away. The others in the flock hear the call and go on alert as well. If one bird panics and flies, the whole flock flies while most evading the raptor.
Now flocking is an obvious EP advantage, just safety in numbers. Alone the raptor eats you, in a flock it eats your comrades so the bigger the flock the safer an individual is. Also in bird flocking behavior (often modeling by the ALife crowd) birds try to get away from the boundary of the flock, more inside. But: a) some birds have to be on the outside so all can’t crowd “inside” and attempting to do so just re-arranging the flock, and, b) inside isn’t entirely the safest because it’s now harder to see the raptor. So there is some optimal place to be in the flock which each bird seeks. These individual movements sometimes coordinate and the entire flock moves as a whole, the so-called emergent behavior. What’s interesting is that the aggregate behavior can not be deduced from each bird’s “rules” but simulation (or real life) reveals the aggregate movement.
But it’s the warning call that is of interest. By making the warning call that bird calls attention to itself from the raptor and thus is more likely to become the target of the raptor. So all the birds in the flock benefit from the warning call except the one who makes it.
In Dawkins’ Selfish Gene he also introduces using of game theory to explain evolution their risk-reward matrix. And using the simple math of natural selection, in a group with different traits (mutations) you can predict future evolution.
So, let’s say in the aforementioned flock some bird has a mutation and doesn’t make the warning call. It benefits from the other birds making the call but then minimizes its own risk. Thus it’s more likely to survive. And when it has offspring they’re more likely to also have this trait (since the first mutated bird will mate with an unmutated bird the offspring are not guaranteed to carry the trait). But let’s assume the mutated bird has multiple offspring that do carry the trait. All those offspring have an advantage, compared to the unmutated group, and therefore are more likely to survive and more likely to have offspring and so forth.
Now simple logic would indicate the no-warning-call trait would then become “fixed” in the population. But that’s contradicted by reality, real-life birds do make the warning call.
So how can this be explained?
A sub-population where the gene is completely fixed (no bird makes the warning call) raptors will soon learn these flock is easier to hunt. So raptors will adapt (just behaviorally, not genetically) and soon eat that entire flock and bye-bye mutant trait of no-warning-call, it won’t spread to other flocks, or at least all flocks, or else the entire species goes extinct (which could happen).
So the game theory approach says the trait won’t fix. It may exist, but it will have to co-exist with the making-the-call trait. So was it really an EP advantage to begin with?
Dawkins has a good account, that has been modeled, of “doves” and “hawks” – birds that are the same species but have the ‘dove’ trait or the ‘hawk’ trait (just like eye color in humans, no eye color even completely is fixed). Depending on the risk-reward matrix you hypothesize and then model you get different outcomes. Basically hawks always attack any other bird, doves always retreat. When doves encounter each other, both run away, both waste some energy. When hawks encounter doves, hawks use some energy chasing doves, but make easy kills and thus gain energy (these are the values that go into the simulation, how much it costs to hunt, what value accrues from a kill). Of course when two hawks encounter each other they both fight, one probably dies, but the other may be wounded even it is the winner thus having a net negative outcome. So more numbers to fill in the matrix.
Now what happens when you simulate introducing a hawk into an all-dove population (an all-dove population can be shown to be stable over time, but either an outsider to a sub-group or a mutation can introduce a hawk to the peaceful dove group). Clearly for a while the hawk(s) has a big advantage and so breeds more hawks (of course the doves are breeding too at the same rate). So the fraction of hawks rises until it reaches the point where now there are many hawks (most of the doves having gotten killed) and fighting between the hawks starts killing them.
Depending on the numbers in the matrix if you plot the hawk/dove ratio over time it turns out it oscillates. This actually isn’t too surprising (there is dynamic equilibrium but no static equilibrium) as most man-made control systems (like autonomous driving vehicles or rockets or whatever) have some range where they are unstable and oscillate.
Also usually the drop in hawk fraction is precipitous (like the so-called tipping point) whereas the drop in doves is gradual. For a while it’s good to be a hawk, until, boom, it’s bad to be a hawk.
Interestingly I think this behavior also describe market bubbles and crashes. With the feedback system having a delay or there is no external regulation bubbles are inevitable. And bubbles rarely have soft landings. So this is an inherently unstable system (unregulated trading). Like the actual flock of birds having mostly “altruistic” members in real life, or the dove/hawk fraction being temporarily stable with mostly doves, any claim that markets are self-regulating (at least without crashes as the “regulating”) seems false. But like any of these other systems it all depends on the numbers.
I’m just starting to look at a book that claims to see physiological effects in market traders. Their testosterone and other hormones rise as they are succeeding. It’s well known these hormones reduce fear thus inducing recklessness. This is probably an EP thing, just like the bird’s warning-calls. If you and I were in a band of humans and a sabertooth tiger was attacking a “rational” approach would be to run, not fight the tigers. If all in the band did that the tiger would have easy prey. OTOH, if some, esp. males, fight the tiger, that might be bad news for them, but good for the rest of the tribe.
Thus is defending against the tiger altruistic? No, it’s just a trait that must exist in some fraction of the band for the band as a whole to survive. So what induces the individual who does fight the tiger to do – recklessness, brought on by those very small hormonal systems. In short, a certain amount of fearlessness/recklessness has evolved in human beings. So it’s simple, bond traders are merely the people who express those traits more stronger. But what worked to keep humans alive may be really had behavior for an artificial system. The risk-reward matrix for dealing with tigers is different than it is for the bond market. So the oscillations the system will have will be different.
I used to believe markets are rational. That had a very specific meaning when I was in bizschool and the math of that has now been mostly discredited. But I don’t even, now, believe the broad meaning is valid. I think markets are irrational and thus prone to boom/bust cycle. Any claim to self-regulation can only be valid when you include crashes (no matter what was done in 2008 to arrest the crash eventually a new equilibrium, almost certainly total depression would have been reached and you could label that as “self-regulating” but disaster is hardly an outcome one would want to leave up to “self-regulation”). I actually saw Greenspan, when he was in his full Ayn Rand mode, defend the idea that even full-out fraud should not be externally regulated in financial markets; IOW, all external regulation will fail, only internal self-regulation is possible. This is actually one interpretation of the “flaw” he so famously announced in those congressional hearings.
In all the talk about economic ideas few, if any, include evolutionary psychology ideas. Human beings evolved without artificial things like markets or airplanes or computers. We have unconscious behaviors/traits designed, via natural selection, to let us survive natural threats, but does those traits/behaviors fail in a system of unnatural threats. I think this is the flaw in both economics and financial theory – they do not include biology and esp. EP, and instead assume “rational” self-interest. Given the evidence that humans can be subject to evolutionarily fixed irrational behavior this simplistic idea is just a mythical behavior used in microeconomics to model markets and believed in, religiously, by libertarians.
Perhaps, even we don’t end up back in a dark age instead, future economics will model markets more realistically and then enlightened external regulation can be appropriately applied. That is assuming we have a rational political system, which is a stretch at the moment.