Someone at The Economist blog Free Exchange writes,

Everyone now agrees that “institutions matter”, but this is partly because “institutions” is a catchall word that can include almost anything anyone cares about, if one squints hard enough and tilts one’s head at just the right angle. Government, social structure, and economic practices get squashed into the word “institutions”, whereupon we can all agree that the things that matter, matter.

The uncomfortable part of this, however, is that some of the institutions that clearly matter are indisputably cultural. Corruption, which eats away like acid at most development projects, is a lot easier to fight if your citizenry sees bribery as a moral outrage, rather than a convenient way to evade inconvenient rules. Keeping your girl children out of school, and sending the boys to religious schools that eschew math and science in favour of memorising religious texts, dramatically disimproves the prospects of competing economically in the modern world. And people whose cultures distrust foreigners, or any strangers outside the extended family, are unlikely to enter into the complex web of trust in strangers that supports a modern economy—or accept the legitimacy of a democratic government elected by millions of countrymen they have never met.

I like to represent this as a pyramid, with the beliefs that people have at the bottom, their social and political institutions resting on top of their beliefs, and economic outcomes resting on institutions.

An alternative hypothesis is that people’s beliefs are subject to tipping points, and institutional changes can create differences in beliefs. Implicitly, I think I had something like that in mind in my argument against polygamy, because I thought that polygamy might unleash male envy in violent forms.On that issue, Bryan wrote,

why are men’s attitudes markedly less egalitarian than women’s?

This does not directly confront my argument. Suppose that in prehistoric times, men were polygamous. Men became highly competitive, and their instincts with respect to one another became envious and hostile. However, groups that adopted monogamy were able to dampen the male fear of not having a mate, and this resulted in societies with sufficiently high trust levels to develop institutions that supported economic progress. If you set a monogamous society and a polygamous society side by side, ultimately the monogamous society would achieve larger scale and greater economic development.

Now, what I am suggesting is that if we were to legalize polygamy, eventually the process could run in reverse. The ability of males to trust one another could decline, and men might revert to a set of beliefs that make it impossible for markets and other co-operative institutions to persist.

The post on the Economist blog started out by talking about Iraq.

There seem to be only two possible explanations for the current conflagration:

1. the country could have been made stable, but the Bush administration mucked up the occupation in its early phases, and now things have spiraled out of control
2. a dictatorial strongman like Saddam was necessary to hold the place together; once he was removed, violence was inevitable

In my essay, I implicitly argued for the second position. That is because my view is that beliefs tend to change slowly. Iraq was held back by Iraqis’ beliefs that trust only extends to a narrow tribe, tribes compete for power on the basis of guns and violence, and government employment is a means to steal from the public rather than serve the public.

The first position (that the United States messed up the occupatoin) would be based on the view that institutional changes could have tipped Iraqis’ beliefs. Had we established order quickly, for example, we could have caused Iraqis to expect that peaceful economic competition would supplant violent tribal competition. I can see this in theory, but I remain highly skeptical in practice. I think it’s a far-fetched hope to try to build a pyramid from the middle.