The legendary economist Robert Solow once joked that every discussion about the relative performances of European economies "ends up in a blaze of amateur sociology." The joke has an element of truth. Economists are fond of invoking "culture" to explain large-scale outcomes they don't understand. This often comes up in discussions about Japan, whose macroeconomy defies every standard textbook theory. Culture, I often hear, is at the root of all the mysteries.
But "culture" is just one of many large-scale fudge factors that economists are tempted to fall back on. There's also "technology," which macroeconomists often invoke to account for inexplicable productivity changes. Or "power," which some left-leaning economists use as a rationale for outcomes that benefit the rich.
Now "trust" is in fashion and has earned a place on the fudge-factor list.
Before I criticize the use of "trust" to explain economic outcomes, let me admit that I am absolutely guilty of it. In a recent article, I worried about decreasing trust in U.S. institutions. Though I think my worry is legitimate, it's also easy for me and other writers and economists to take the trust thing too far.There are good reasons to think that trust, in some sense of the word, really matters for economic outcomes. In many interactions in markets and elsewhere, various forms of trust can help achieve results that are better for everyone involved. If buyers and sellers trust each other not to cheat, they can do more deals. If insurers trust drivers not to be reckless, they can insure them more cheaply. Cooperation — between employees in a company, for instance — is also dependent on trust. The famous prisoner's dilemma describes a situation in which trust can lead to much better outcomes for everyone — as long as the game is repeated many times. In open-ended relationships, where the number of interactions is uncertain, trust helps people cooperate repeatedly instead of just betraying each other over and over.
It's tempting to take this realization and start using as it an excuse to invoke "trust" to explain lots of things in the world around us. And so there is a boomlet in citing trust to explain things like happiness and economic development. But we should resist this temptation, for at least three reasons.
First, trust is hard to measure. The typical indicator of social trust comes from surveys. For example, the World Value Survey asks people whether they believe that "most people can be trusted," or whether one "can't be too careful." But this doesn't necessarily tell us about the kind of trust that matters in economic models.
Let's suppose that in one country, people have very close, trusting relationships with business partners, but distrust strangers on the street. That country might have the kind of trust needed to facilitate economic exchange, but might still rank low on the World Value Survey's measure. Trust is also relative. There could be a country where everyone trusts each other a lot compared to other countries, but because the culture has such a high standard for saying you "trust" someone, people respond negatively to the World Value Survey's question. This kind of problem crops up a lot in economists' surveys of self-reported happiness in different countries, so it's reasonable to expect it to rear its head here as well.
A second, related problem with trust is that there are different kinds. Although the World Value Survey asks about attitudes toward "most people," other surveys ask about trust in institutions, or the government. But there may be yet more forms of trust that the surveys don't tease apart. Trust in businesses may be more important than trust in strangers or trust in family. Alternatively, how quickly people build up trust with new people might be more important than how much trust they have in the people in the town they grew up in. There is just no automatic correspondence between survey measures and the objects in economists' models.
The third problem is the most insidious of all. Because trust is a concept that most people deal with in their daily lives, economists are naturally tempted to think they know how it works. For example, we may assume that because we trust people at our church, religious homogeneity produces social trust. We may therefore jump from the theoretical and survey literature on trust to the conclusion that nations should encourage religious homogeneity in order to make themselves richer and happier. But this kind of intuitive leap, though tempting, isn't supported by hard evidence.
So although trust, in some form, is probably important in our economic lives, we don't yet have the tools to measure it, we don't know exactly how it's important, and we definitely don't know how to control or alter a society's level of trust. Until we understand trust a lot better, it would be a mistake to rely on it too much when trying to explain the world around us.
Noah Smith is a Bloomberg View columnist. For more columns from Bloomberg View, visit http://www.bloomberg.com/view.