« Measuring Universities: Once More Unto the Breach | Main | Capital-biased technical change vs low interest rates? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

"I wonder if there's anything equivalent in biology?"

The thing that strikes me immediately is the huge impacts of latitude on just about every human outcome - trust in other people, strength of family ties, etc. Latitude is correlated with population density and disease, hence the absolute number of creatures competing for resources (in a Malthusian world, however, everyone is just at the brink of survival). Don't have time to think this through more, must get back to marking.

Actually, the discussion about people in trades and occupations getting together and discussing ideas is something I came across in Alfred Marshall's Principles - the quote: "When an industry has thus chosen a locality for itself, it is likely to stay there long: so great are the advantages which people following the same skilled trade get from their neighbourhood to one another. The mysteries of the trade become no mysteries: but are as it were in the air, and children learn many of them unconsciously. Good work is rightly appreciated , inventions and improvements in machinery, in processes and the general organization of business have their merits promptly discussed..." Marshall, Alfred (1890/1969) Principles of Economics, 8th edition (London: MacMillan and Company)p.225. Surely the tech people in Silicon Valley from rival firms get together for beers too. By the way, farmers are indeed the same as universities. They both have occasional crop failures.

You might want to have a look at the effects of the Bayh-Dole act, which greatly enhanced a university's ability to patent the results of publicly funded research.

Also the increase in claims by nonpracticing entities has become a problem. A company defends a patent lawsuit by making counterclaims based on the plaintiff's alleged violations of the defendants patent portfolio - mutually assured destruction. This means that to defend itself, a company either must have a large patent portfolio, or must not be a manufacturer and thus not practice any patents.

This is an oligopoly with rents - not, I think, a result a (non bought and paid for) economist would choose.


"It might be differences in the nature of competition between firms. If Samsung uses a good idea, that hurts demand for Apple's products, and vice versa. There's a small number of competing firms. It's an oligopoly."

Samsung and Apple compete by making products that are distinguishable from each other. Their patentable technology is not limited to production methods, but rather the goods that are produced themselves. Productivity is increased through a value added process - micro controller A + power supply B + display screen C + instruction set D gives product E that does something. People are willing to pay more than the sum of parts for the utility provided by product E.

"If a farmer's immediate neighbors use a good idea, that increases their productivity, that will have a negligible effect on the total supply and price of of wheat. Or the total demand and price of inputs like land and fertiliser. There is a very large number of competing firms. It's perfect competition. Unless the good idea spreads to all farmers across the world. But that's unlikely to happen quickly, if at all, so the farmer ignores that risk when he is talking with his neighbors."

Unless the farmer is doing genetic engineering or controlled cross breeding between strains of wheat, one farmer's wheat is not very distinguishable from a another's. Productivity is increased through a time saving process.

"From my limited experience, farmers are the same as universities."

Farmers might be, but Monsanto isn't: http://www.monsanto.com/newsviews/Pages/saved-seed-farmer-lawsuits.aspx

Without wanting to put too fine a point on it, I think you discovered what I believe to be one the main missconceptions of a lot of "Evolution" style thinking. I feel that most economists look at evolution, and then say evolution=competition='survival of the fittest' and then conclude, "hey, that's how markets work". The problem I feel with this line of thinking is that evolution, to my understanding at least, is just a process of adaptation. And I see no inherently true reason why competition should in every case be the superior method of adaptation. In fact in many cases cooperation (and thus avoiding the kind of 'weeding out' that normally should happen) is actually a far superior strategy and humans have many instincts designed exactly to further this cooperation. In that sense, I belive that the equivalence evolution=competition is almost certainly wrong. Of course that leads us to the unsatisfactory answer that universities cooperate because, well... we actually aren't designed to think like the Prisoner dilemma's representative agent.
Then again, I'm neither a biologist nor a psychologist, so I might just be writing nonsense

Frances: I wonder if it's some sort of trade-off between competing for limited resources with diminishing returns to population vs cooperating in small groups?

Livio: lovely find! "It's all in Marshall"!

Peter N: that stuff worries me too.

"This is an oligopoly with rents - not, I think, a result a (non bought and paid for) economist would choose."

But then I'm bought and paid for by Carleton University!

Frank: but even if Apple and Samsung were oligopolists producing identical products, like wheat, neither would want the other to find a new technology that lowered production costs, increased supply, and reduced market price.

Frances: yep. There's the big exception.

Alex1: I think I have heard some people say that economics is just a sub-branch of biology! I'm not sure I would go that far, though I do notice similarities, especially with ecology. And non-humans sometimes cooperate too, sometimes across species (those cleaner fish, bees and plants).

"I feel that most economists look at evolution, and then say evolution=competition='survival of the fittest' and then conclude, "hey, that's how markets work"."

You could almost argue it was the other way around. Darwin read Adam Smith and Malthus and then thought "hey, that's how nature works!"

But yes, we mustn't push the parallels too far. Though saying how far is "too far" isn't easy.

The propensity of farmers to share tips is something I've explicitly looked at, and it seems to confirm some elementary economic theory and the way you're thinking.

Broadacre (grain/hay) croppers happily share tips, as do dairy farmers. Horticulture farmers, especially market gardeners, are very tight lipped however. When I asked a capsicum (bell pepper) farmer if he discussed tips at the pub he looked at me as if I had done a turd on the ground. "We talk at the pub, but NEVER about farming".

Makes sense, there's often only one or two distributors for a given vegetable or fruit in a limited region, and it cannot be stored for as long as grain or dairy derivatives. There's also much more quality variation because they are not traded as standardised commodities, so each farmers can get a different price. If your mate increases quality or yield it will have an affect on quality.

Someday I'll get funding to look at this properly.

Richard: excellent! My guess is that these sorts of questions maybe aren't just fun ways to test our theoretical hunches, but might really matter for long run growth. I wish you luck with the funding.

There is a lot of differentiated competition, competitive avoidance, and even cooperation in evolution as well, from colonies, schools, and flocks, to symbiosis. There are probably a lot of inter relations among farmers too.

Economic as biology? Some french universities teach economics as it is the science of valuing property rights....
Competition? Didn't somebody ( I have a hunch about Hal Varian but I'm not sure, can't find it) wrote a paper arguing that what it is not competition as such but the number of possible solutions that free entry provides that let the superior one prevail? Like a plane crashing in the desertt and survivors fanning in all direction and one luck guy find the oasis. The best decision win but not by design. Just as mutations produces variants and something survives.

As EO Wilson pointed out, competition and cooperation are affairs of scales - they take place at all levels, and differently at each level. What equates a farmer to Apple? Or Apple to a university (rather than, say, a faculty or a single professor)? Further, the balance at each level will reflect/affect what is the optimum at other levels. Your kidneys do not ordinarily complete with your spleen, but your body will prioritise your kidneys over your spleen when the survival of you as a single human is at stake (and so on up to the social level and down to the cell level). So what matters here is the institutional/social framework. Universities "competed" by displaying a superior ability to generate certain sorts of ideas, but the ability to display is linked to the formation of a social world through other sorts of sharing. Business compete at the ability to generate money for some other people, but cooperate in other areas and so on. In other words, if it's a puzzle to explain it at one level, the answer probably lies at some other level(s).

Actually Bacteria and other single celled organisms will readily share genes by horizontal gene transfer. It really complicates the whole tree of life


But then most multi-celled organisms have sexual reproduction which is a bit of a mystery itself.

The long-awaited Darwin post! Thanks Nick. Not much time to comment now, will respond at greater length later. Very quickly, yes, Darwin was remarkable in terms of his intellectual range. Modern evolutionary biologists can be roughly arranged along a continuum from "lean and mean" theorists who interpret the world through the lens of very simple and elegant mathematics, and who emphasize the power of natural selection, to "the world is infinitely complex and idiosyncratic" types who see selection as just one evolutionary force among many and who see evolutionary history as the complex and contingent outcome of the interplay of a huge range of exogenous events and endogenous processes. There's another continuum, at least partially independent of the first, from people who do lab experiments on the sorts of organisms on which one can easily do such experiments, to people who do comparative analyses of observational data. And *all* of these people quite legitimately can and do claim Darwin as their intellectual ancestor!

On companies: Apple seems to be secretive about everything, but many companies appear not to worry too much about things that they don't regard as their products. For example, everyone knows about the Toyota method; everyone knows about the Google gyms and buffets. Those companies don't believe that those ideas are what makes them money (rightly or wrongly), so they don't protect them. But there are companies in which HR stuff (like salaries) are very closely guarded secrets.

On biology: evolution is pretty ruthless, so presumably the benefits and losses from sharing get worked out over the millennia. Any social animal that became extinct or became less social, the reason might have been because they shared too much.

Actually, Phil, the evidence is that social animals which became extinct sometimes became extinct because of *too much success* in the short term, depleting their food supply and causing a crash later.

Evolutionary "success" is time-scale limited, as any boom-bust cycle in ecology will show you. The most "successful" in the 100-year timescale is not the most "successful" in the 1000-year timescale, etc. It is a common mistake to think that evolutionary "fitness" is exogenous; it is environmentally contingent, and furthermore, the creatures change their own environment.

Long-term evolutionary "success" is a very tricky matter relating to robustness and adaptability; on the whole, the bacteria are the clade with the greatest long-term "success", due to their high reproduction and high mutation rates, along with aggressive horizontal transfer.

Interesting comments. Since you started with links, I have one link that I got from comments under on of Scott Sumner's posts. It is a series of lectures about Behavioral Evolution from Stanford University prof. Robert Sapolsky: http://www.youtube.com/watch?v=NNnIGh9g6fA It is really good and surprisingly he does not assume that students have any particular in-depth knowledge about biology or evolution

There are some interesting finds:

1. Evolution is not survival of the fittest, it is about passing your genes onto the next generation. It is OK for you not to survive as long as you as an individual pass copies of your genes onto the next generation. The most direct way of how to pass your genes is to reproduce. One example of why it is not survival of the fittest but passing of the genes is that many species have traits that have adverse impact on individual survival, but that help with reproduction. For instance peacock feathers are not very good at camouflaging him against predators, but they are very important to attract a chicken.

2. Therefore there exists a special form of "Group Selection" - called "Kin Selection". That is helping even at your own expense to pass the genes of their own. You share 50% genes with your brother and 25% with your cousins, that is the basis for a sentence Sapolsky uses a famous quote by J.B.S Haldane "I would lay down my life for two brothers or eight cousins". So being completely altruistic is more probable the closer kin the individual on receiving side is to you.

3. But sometimes we observe seemingly altruistic behavior even among unrelated individuals. Basically most of the time we are observing just a form of reciprocal altruism. Many species - even bacteria - found out that cooperation is mutually advantageous and they utilize it in their evolution strategies. And this is the ground economist can feel at home - it is the game theory stuff. Prisoner's dilemma and cooperation strategies where individuals are rational and maximize payoff.

Reciprocal altruism is another kind of a battle - individuals try to find a new innovative ways to cheat while at the same time they develop abilities to spot cheating. Sometimes there may be a noise and cooperation is misunderstood as cheating. Be it as it is, we can observe a very complex Tit-for-Tat among animals with all the above (kin selection and all that) as part of the explanation. Sapolsky for instance mentioned that while studying behavior of some apes it was observed that when one A mother did something nasty to mother B, it could be child of B that could later that day enact revenge on a child of mother A.

So long story short - cooperation is observed among animals of any size when there is payoff from cooperation. So if Samsung and Apple dominate the market it makes little sense for them to cooperate. Two universities can cooperate if there is another 20 universities out in the country and hundreds in other countries that can be exploited.

Or maybe it was your brother working in that other university who decided to take a little risk giving out secrets to you so that you can have promotion and increased chance to attract suitable female and pass on 50% of his genes to another generation :D

A H, Red Queen Theory is the most popular proposed resolution of that "mystery".

A few further comments Nick:

1. The broad issue you raise is one Darwin himself struggled with: conveying that by "competition" he meant "competition in a broad sense", not just the narrow senses of one-on-one combat, or two organisms competing for the same essential resources (e.g., two adjacent plants competing for the same light rays), or predators trying to catch prey that don't want to be caught. As you say, there is always *some* process that affects who leaves more progeny in the next generation, and who leaves fewer. It's quite possible for one organism to interact with another in such a way as to raise its *absolute* fitness (i.e. its absolute number of offspring), but reduce its *relative* fitness (i.e. its number of offspring, relative to the numbers produced by other members of the population). Relative fitness is the only thing that matters in evolution.

2. Many cooperative relationships in evolution are either between members of different populations, or between related members of the same population (where "related" typically, but not necessarily, means genetic relatedness). The former sort of case does indeed have close economic analogues, as you've discussed in other posts. This sort of case often involves things like gains from trade. The trading partners in this sort of case aren't competing to produce the next generation of the same population, so it's not really surprising to find them cooperating, exchanging information, etc. (Although nor is it necessarily surprising to find them exploiting or free riding on one another if they can--apparently-mutualistic relationships in nature often turn out to be exploitative or parasitic, at least in some times or places). The latter sort of case (cooperation and information sharing among related individuals) also isn't surprising in principle, although the costs of cooperating need to be sufficiently large relative to the costs, with the degree of relatedness governing how big "sufficiently large" is.

3. Re: whether there are evolutionary analogies to rival vs. non-rival goods, excludable vs. non-excludable resources, and to sharing of information and ideas among apparent rivals, I think it depends on how much precision you want in your analogies. I'm not enough of an economist to be sure of suggesting super-precise analogies, but I will throw some biological examples out there and let your economist's brain judge how well they map onto economics. Just the first examples that pop into my head, *very* far from an exhaustive list:

-many bacteria obtain iron (a scarce, nutritionally-essential resource for which there are no substitutes) by producing siderophores: iron-binding agents which are released into the environment. The resulting siderophore-bound iron is easier for bacterial cells to take up. But siderophore production is costly, and siderophore-bound iron can be taken up by any bacterial cell in the immediate area. So there is a clear incentive to free ride. Angus Buckling and Stuart West at Oxford are two evolutionary biologists who've done a lot of neat modeling and experiments on the evolution of siderophore production and other "public goods" in bacteria. Here's Stu West's publication list: http://www.zoo.ox.ac.uk/group/west/pubs.html

-in group-living animals there's of course a lot of sharing of information and ideas, such as good places to find food and techniques for obtaining it. There are neat stories of cases like one ape in a group figuring out a clever new technique for cracking a difficult nut or something, and the other group members watching and copying. These cases are interesting to think about, because we shouldn't necessarily take for granted that group members with information or new ideas want to share them with the group, even if the group members are all related. Probably in some cases they want to share, and in others there's no practical way to keep the information private (save by leaving the group, a move that might well carry severe costs outweighing the benefits of keeping the information private).

-not sure if there's any biological analogue to perfect vs. oligopolistic competition. Just based on your comment, the only thing that comes to mind is scramble vs. contest competition in bean beetles. These are small beetles that lay their eggs on the surface of beans like lentils. The eggs hatch, and the larvae burrow into the bean, eat it, and grow into adults that emerge and mate. When beetle population density is low relative to bean availability, so that no bean gets more than a few eggs laid on it, larvae within the bean compete in "scramble" fashion. That is, they each just try to eat and grow as fast as they can, and don't take any notice of each other. In such cases, the bean provides enough food for all the larvae to develop into adults, they're just racing to develop as fast as possible. The faster you develop, the sooner you get to mate and lay eggs, the sooner those eggs can hatch, etc. But if beetle population density is high relative to bean availability, so that a single bean gets dozens of eggs laid on it, selection favors a "contest" approach to competition. If each larva just eats as fast as it can, none will obtain enough food to grow into an adult and all will starve. So your only chance is to first kill (and as long as you're at it, eat!) some of the other larvae, and so that's what they do. There's genetic variation for the propensity to engage in scramble vs. contest competition, and so by rearing beetles under appropriate conditions in the lab you can select for beetles that compete in "scramble" or "contest" fashion. But I'm guessing this isn't quite what you mean by perfect competition vs. oligopoly?

-re: larger populations having more mutations (="ideas"), and so evolving faster, yes, but only up to a point. Once mutations are occurring fast enough, mutation supply rate is no longer the rate-limiting step for adaptive evolution. Instead, it's how fast the best mutation can replace the others and go to fixation. High mutation rates actually slow that fixation process down, a phenomenon known as clonal interference. For instance, if Apple has a really good idea that makes their product both better and cheaper than the competing Samsung product, the Apple product would presumably come to dominate the market very fast. But if around the same time Samsung also had a new idea about how to improve their product--not as good as Apple's idea, but still an improvement on their old product--then Apple's new idea won't come to dominate the market nearly as fast, because it only has a small "fitness" edge over Samsung's new idea.

-I'm sure it's often difficult in economics to precisely define markets--who are the suppliers and their customers, are different goods partial or complete substitutes--and how to think about "disruptions" that change the nature of the market (e.g., a non-excludable good becomes excludable for some reason, or whatever). Much the same is true in evolution. Organisms constantly surprise you by finding ways to change the rules of the game. Plants compete for nitrogen they extract from the soil, and one could imagine all sorts of adaptations that might make them better at engaging in that competition. But unless you knew that some plants had joined up with bacteria to *extract nitrogen from the air*, would it even occur to you that that was an option? Or, if you were around a billion+ years ago, would it ever occur to you that one way to respond to all the competition (broadly defined) in the ocean would be to *evolve to live on land*? That's the beauty of evolution by natural selection really--just throw up random variants (meaning, random with respect to their fitness effects), and then keep *anything* that works, where "works" means "increases relative fitness". Engineers often take inspiration from nature for solutions to particular design problems. How to design reusable tape that works like a gecko's toepads, or whatever. But I'm most impressed when some species evolves to solve a problem in such a way that the problem isn't "solved" so much as made irrelevant or non-existent.

Jeremy, why does only *relative* fitness matter in evolution? Is it because the success or failure of a variant is generally measured in relative terms? It seems to me that looking at absolute success should be just as meaningful, especially if a mutation can improve the survival of one organism without decreasing the survival of another. In other words, why is evolution a zero-sum game? Doesn't the possibility of cooperation or expansion to completely new territories imply that evolution is at least sometimes a positive-sum game?

I find the contrast with economics interesting. Economists spend a lot of time thinking about Pareto improvements (when they exist, how to obtain them). But if only relative improvements matter, Pareto improvements are ruled out from the start.


yes, the success or failure of variants is measured in relative terms. That's the appropriate measure. If a mutation occurs that increases survival of the bearer above that of the nonbearers, then all else being equal the offspring of that mutant will eventually comprise 100% of the population. This is true asymptotically even in infinite populations, which isn't much help in any case since real populations are finite. Evolution is always a zero-sum game in the sense that relative abundances of the different types comprising the population necessarily add up to 100% of the population, never more, never less. This isn't of course to say that diversity can never be maintained (for instance, relative fitnesses might be negatively frequency-dependent rather than constant). It is just to say that relative, not absolute, fitness is what matters. The possibility of cooperation or expansion to new territories doesn't change this.

Not being an economist, I only have a vague sense of what Pareto improvements are and whether or not they have any evolutionary analogue, so I can't be of much help on that.

See the work of Alan Grafen (specifically, his "formal Darwinism project") for a rigorous formal statement of what "fitness" must mean in an evolutionary context, and the precise sense in which (and conditions under which) evolution by natural selection can be said to be "optimizing". Steven Frank's expositions of the Price equation are another relevant, and somewhat more accessible, resource. Any evolutionary biology textbook will of course present things in terms of relative fitness, but won't necessarily go into depth or explain in a formal way why that's the appropriate choice.

("Pareto Improvement" means "One or more *people* are better off and no *person* is worse off". So "Pareto Improvement" is a speciesist(?) concept.)

The nearest evolutionary analogue to Pareto Improvements can think of is if the *absolute* abundance of one species (or gene?) increases without the absolute abundance of any other species decreasing. (It would probably be more useful if we restricted the domain, so that when two species cooperate both become more abundant (ignoring the other species).)

Thanks Nick. Yes, your suggested analogy basically sounds sensible to me.

Apologies for using jargon in an inter-disciplinary discussion. Thanks Nick, for clarifying.

Nick's analogy is basically what I had in mind. In welfare economics there is a long tradition of distinguishing between issues of efficiency and distribution. When we talk about efficiency, we ask whether there are any opportunities to make someone better off without harming anyone else (Pareto Improvements). When we talk about distribution, we take the overall amount of "good stuff" as fixed, and then examine how it is allocated between individuals. In folksy language, we could compare analyzing "the size of the pie" with "how the pie is divided up."

Economics tends to focus on the former; ecology (apparently) tends to focus on the latter. Is the difference that ecology is more descriptive than prescriptive, and that from a descriptive point of view it doesn't matter if you normalize the population size? (I am not aware of anyone who is particularly interested in maximizing biomass-per-square-meter for its own sake - though I would be interested to know if such research exists). Or is the issue that there really aren’t that many opportunities for mutual gain, so a zero-sum game is the good approximation? If so, then there is a real contrast in how economists and ecologists view the world.

Nick, perhaps you are right that it is just specisism. And I hope that this is at least tangentially related to your original post.

Brad: almost anything is at least tangentially related to this particular original post! ;-)

Yep, you are easily on-topic.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad