Is there really a link between language and economic behavior?

In the September/October print issue, I wrote about the work of Yale economist Keith Chen, who has suggested there may be a link between how languages treat the future and how the people who speak them behave:

[I]n what linguists call strong future-time reference (FTR) languages like English, a speaker says, "It will rain tomorrow." In a weak-FTR language like German, one simply says "Morgen regnet es"– literally, "Tomorrow, it rains." Mandarin Chinese has similarly weak time construction. Strong-FTR speakers have to do a little more verbal work to make it clear they’re talking about the future. 

Weak-FTR languages include German, Mandarin, Japanese, and the Scandinavian languages, while English, Greek, Russian, and Spanish are strong-FTR. In a recent paper, Chen compared European families of similar education, income, and religion and found that speakers of weak-FTR languages on average save more for retirement, smoke less, and are less likely to be obese. "Every one of the countries that we think of as an outlier in terms of savings is also an outlier in terms of how they speak about the future," Chen says. 

Chen also gave a TED talk on the finding here. 

Via Marginal Revolution, I was interested to see that Chen’s conclusion was being questioned by linguists, leading Geoffrey Pullum of the Chronicle of Higher Education to hold up Chen’s work as an example of the spurious correlations proliferating in the era of "Big Data":

Chen’s data on languages comes from the World Atlas of Language Structures (WALS), and his evidence on prudence from the World Values Survey (WVS). Both are fully Web-accessible. Sean Roberts, who studies language evolution at the Max Planck Institute for Psycholinguistics in Nijmegen, decided to investigate the other linguistic factors treated in WALS to see how they related to prudence. He compared the goodness of fit for linear regressions on each of a long list of properties of languages (the independent variables), using as the dependent variable the answers that speakers gave to the WVS question “Did you save money last year?”

The results (see this blog post for an informal account) were jaw-dropping. He found that dozens of linguistic variables were better predictors of prudence than future marking: whether the language has uvular consonants; verbal agreement of particular types; relative clauses following nouns; double-accusative constructions; preposed interrogative phrases; and so on—a motley collection of factors that no one could plausibly connect to 401(k) contributions or junk-food consumption.

In a follow up post, Roberts adjusted his model based on suggestions from Chen, finding that "the link between future tense and the propensity to save is more robust than the previous post suggested" and that, in fact, "the results showed that there was only one other linguistic variable that improved the fit of the model more than future tense." The only factor that he found to be more significant under the new model is whether a language has a distinction between perfective and imperfective — which also relates to time. 

Pullum notes that "it is still quite problematic for Chen that the third-ranking linguistic factor is whether the language has a velar nasal consonant: Why should having an ng-sound, as in singer, predict the savings behavior of the speakers of a language almost as well as the allegedly relevant future tense and perfective aspect marking".

And of course, even if the future tense and savings are correlated, there may still be a confounding cultural, linguistic, or economic variable that hasn’t yet been accounted for. As Pullum concludes, "the results are still coming in."