Trade and natural selection

Economic theory tells us that trade makes the parties involved better off. Through trade, a person can specialise in the activity in which they have a comparative advantage. A person is better off even if they are trading with someone who is better than them at all activities. This is because the less productive person will still have a comparative advantage in some activities. By specialising, an individual can use income from the activity in which they have a comparative advantage to buy other goods and services.

Whether trade is beneficial is not as clear from the perspective of the genotype. To examine this, Gilles Saint-Paul authored a paper in which he examined how trade may affect the evolution of humankind.

Saint-Paul’s model involves a haploid population that has two productive activities – fight and defence. Haploid means that each loci, the specific location of the gene, has only one gene. Each person in the population has two unlinked loci, with the gene at the first locus determining fighting productivity and the gene at locus two determining productivity in defence. At each loci there can be one of two alleles, the higher productivity H or the lower productivity L.

This situation leads to four possible genotypes: HH, HL, LH and LL. HH has maximum productivity in both activities. HL is productive at fighting but not defence and so on.

First, Saint-Paul considers the situation where there is no trade and each person must do everything for themselves. In this case, the unproductive L alleles are eliminated and eventually the population consists only of HH genotypes. The LL types are eliminated as they are unproductive at both activities. The HL and LH genotypes are also eliminated as, while they are productive in one activity, they are unproductive in the other and must still provide for themselves in that unproductive activity.

Under trade, not everyone needs to be perfect. Each person only needs one H allele and they can then specialise in that activity and trade for the other service in which they have lower productivity. HL genotypes can specialise in fight and trade their fight for defence provided by someone else. Similarly, LH genotypes can specialise. However, both HL and LH genotypes cannot both exist in the population in the equilibrium as when they mate, they will produce some LL genotype children. As a result, HL and LH genotypes will have lower fitness than the HH genotypes until the L allele is completely eliminated from one locus. Once that occurs, all children will have H in one locus, ensuring they are a productive in one activity that they can specialise in. The population will end up a mix of HH and HL or, HH and LH.

Having looked at these two scenarios, the natural question is what would happen if a trading and a non-trading society were part of a larger population. Would one grow faster than the other? In equilibrium, the answer is no – both societies in equilibrium have the same fitness as all production is done by those with the productive H allele at the relevant loci. If both populations started from a mixed group consisting of all four genotypes, the group which trades would reach equilibrium first, so it would grow faster during transition. However, once the populations reach equilibrium, both grow at the same rate, so neither population is eliminated.

If environmental shocks are thrown into the mix, with these shocks changing whether the H or L allele is more productive, then the trading population might grow faster. This is because it maintains some diversity in the form of the L allele in at least one loci. When the productive advantage shifts, the non-trading society has only the newly unproductive alleles for each task. The non-trading society would then have lower fitness, although not directly from not trading, but rather from their non-trading-induced lack of genetic diversity.

One observation on this model is that it differs from the typical comparative advantage story told in economics. Economics shows that trade can make someone better off even if they are absolutely less productive at all possible activities. The genetic story tells us that even when there is trade, only those who have maximum productivity in at least one activity will be present in the equilibrium population. Trade may make a completely unproductive person better off in the short-term, but over the long-term, their unproductive alleles will be eliminated – totally in the case of no trade and from at least one loci in the case of trade.

I like the concept behind this model, but struggle to apply it to any examples. In societies with a history of exchange, do we see more genetic diversity? Conversely, are people in societies with little history of exchange productive across all areas which were relevant to their fitness? I am not sure there is evidence for either. More likely (or so I believe), have humans been trading in one form or another for so long that I am wasting my time looking for modern examples and that we should simply take Saint-Paul’s concept as one factor behind the diversity that we see today.

Saint-Paul, G. (2007). On market forces and human evolution Journal of Theoretical Biology, 247 (3), 397-412 DOI: 10.1016/j.jtbi.2007.03.021

2 comments

  1. I suspect you are right that all humans have been trading so long that this model has no explanatory power. But I like the concept so here is my attempt at two specific examples (which hopefully do not simply restate the paper which I haven’t read).

    Firstly, think about the prices in the trading society. If each individual wants to consume equal amounts of fight and defence, and prices are equal, HH genotypes will be happy with no trade. In order to induce trade, HL or LH genotypes will need to exchange what they are good at a rate of less than 1 for 1, that is, the price of the activity they are good at will be relatively lower.

    This leads to the two potential specific examples.

    Firstly, a society with HL or LH genotypes will be characterised by ‘income’ inequality while a society with purely HH genotypes will be characterised by ‘income’ equality. The empirical question then becomes, can a history of trading explain differences in the distribution of income between countries?

    Secondly, the model could be used to explain why comparative advantage between countries is observed. Consider the implications of allowing trade between societies, but not genetic exchange (of course, this would require goods to be produced not just the services of trade and defence). Trade would equalise the price differences between HH/HL societies (higher relative price for defence) and HH/LH societies (higher relative price for fighting). Trade between societies would therefore increase the utility of the HL and LH genotypes by effectively allowing them to trade with each other, even though they cannot coexist in the same society.

    1. On your income equality example, I am not sure if it could arise to a meaningful extent. Any trading premium that did arise would be bid down towards zero (assuming no cost to trade) as HH types compete between themselves for the premium.

      I like the second example, although that insight does not require Saint-Paul’s model – we could just take the work of Ricardo and claim that the differing productivity has a genetic basis. Trade in that sense also has the potential to make LL types better off.

Comments welcome

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s