Frances Woolley is a professor of economics at Carleton University
Linguists have found that language shapes our perception of colour, our awareness of social status and hierarchy, even the way that we describe everyday objects such as bridges or forks. Now a new study by Keith Chen of Yale University argues that language has a profound impact on economic behaviour, influencing savings rates and wealth accumulation.
Professor Chen begins with the simple observation that some languages make a stronger distinction between the present and the future than others. For example, a German speaker expecting precipitation could say, " Morgen regnet es". Literally, "It rains tomorrow." English speakers, on the other hand, must distinguish clearly between "it is raining now" and "it will rain tomorrow."
There is a surprisingly large variation in the way that languages mark time. All of the Germanic languages, except for English, are "weak-FTR (future-time reference) languages", making no clear distinction between present and future actions, as are most East Asian languages, including Japanese, Mandarin and Cantonese. Some African languages fit into the weak-FTR category too. Southern European languages, on the other hand, are "strong-FTR languages", drawing a clear distinction between present and future, as are Middle Eastern and most South Asian languages.
The list of weak-FTR languages is enough to give away the conclusion of Professor Chen's paper: people who speak such languages save more. Saving requires us to make a choice: I will eat that marshmallow after lunch, not now. English, with the obligatory use of "will eat", forces a person to think about delaying pleasure. Professor Chen argues that this awareness of delay, this consciousness of deferral, makes speakers of strong-FTR languages like English more likely to eat the marshmallow, to consume today instead of tomorrow.
Language indeed correlates strongly with differences in savings rates across OECD countries. Even controlling for demographic differences, unemployment rates and religious differences, strong-FTR countries like Greece have significantly lower savings rates than weak-FTR countries like Norway. The problem with cross-country studies, however, is that there are always alternative explanations for their findings. For example, many of the weak-FTR OECD countries are in northern Europe. Perhaps it is centuries of setting aside food for long cold winters, rather than language, that is behind these countries' higher savings rates.
Yet Professor Chen has other evidence to back up the connection between language and savings. Using the World Values Survey and the European Survey of Health, Ageing and Retirement, he has obtained information on the savings decisions of hundreds of thousands people. To make sure that his results are not driven by particular government policies or economic circumstances, he compares weak-FTR and strong-FTR speakers in the same country. For example, in Switzerland, he compares German speakers with those who speak French and Italian. In Singapore, he compares people who speak Malay and Mandarin (weak-FTR languages) with people who speak English and Tamil (strong-FTR languages). Again, he finds lower savings rates among those who speak languages that distinguish clearly between the present and future.
Moreover, it is not just savings that are affected by language. Professor Chen finds similar patterns when he looks at smoking and obesity rates. People who speak strong-FTR languages are more likely to smoke, and more likely to be obese. An English speaker, announcing his intention to get in shape, says "I will go running this evening." The words "will go" mean "not now, later." A Mandarin speaker, however, would say "I... this evening ... go running". The verb form is the same for running now and running later; the words "this evening" make it obvious when the running will take place. It is not that Mandarin speakers do not have ways of expressing future intentions or past happenings. Rather, Mandarin does not force its speakers to distinguish between past and present, now and later -- and, as such, does not encourage sloth.
I am not entirely convinced that speaking English, or another strong-FTR language, makes people fat, or causes smoking. These behaviours are the product of so many complex factors. But Professor Chen's paper made me think, and will continue to do so.
Follow Economy Lab on twitter
Frances Woolley