May 08, 2003

Linguistic Externalities

Edward Hugh reads Stephen Roach, and asks a very good question about future prospects for economic growth around the world:

BONOBO LAND: In yesterday's post Stephen Roach was once more eulogising about the growth (and deflationary) potential of services outsourcing. He now has a two part story: manufacturing in China and services in India (and Ireland??? - what no-one has quantified yet in all this is the language-based externality attached to being a country with a large English-speaking community. By the way what about South Africa in this context?)...

Posted by DeLong at May 8, 2003 11:15 AM | TrackBack


My guess is that the language variable is so correlated with other stuff that quantifying a specific role for being an English-speaking country is impossible. Looking just at Western countries, there are certain clear distinctions between the Anglophone and mainland Europe economies, as The Economist likes to point out every so often -- somewhat smaller size of government, more pragmatic monetary and fiscal policies. I believe that the linguistic externality is important, but have no idea how someone would prove it.

Posted by: P O'Neill on May 8, 2003 11:44 AM

Although the theory is plausible, I seem to recall some counter examples as well. Japanese investment in Germany is quite heavy and the Jamaican economy is not well known for its ability to attract international capital, and the economic history of the Netherlands and Scandinavia compared to France and Italy does not lend one to think that English knowledge is quite as essential as all that.

I will bet that there is an externality related to language that is more readily quantifiable: the relative accessibility of the people of a nation to one language, or no more than some small number of standardised languages, under some form of effective linguistic management. Without this condition, it is nearly impossible to operate a translation industry that can introduce materials from the rest of the world to the non-mulitlingual masses. Also, without effective language standardisation, it is very hard for international interests to hire or train people to communicate in the local language.

Indonesia is an excellent example of this process. Bahasa is a semi-artificial language invented to serve as a national language for a state with roughly 700 languages. Bahasa is taught in universities all over the world, while Javanese - the next largest language - is only occasionally taught elsewhere, and languages like Acehnese are almost never taught abroad. Furthermore, Bahasa is a languge designed not to have the complex system of honourifics of Javanese and maintained by people whose job is to make sure that there is, as much as possible, a single, universally known way to talk about new ideas and new technologies. Eliminating barriers through universal use of Bahasa - now spoken by some 75% of Indonesians - has, I think, been a meaningful economic factor in Indonesia.

The same process is occuring in a number of other countries: Papua New Guinea, the DR Congo, Pakistan, Vanuatu. There are quite a few countries where some not dissimilar process has occured within reasonably recent history: France, Italy, Thailand. The degree to which this processes has advanced seems to be related to economic prospects, although it may be difficult to ascertain which is the cause and which the effect.

I think the economic effect of language barriers is real enough, but I suspect there are few economic gains to national anglicisation - which imposes huge costs for what I suspect are relatively small gains - while there are huge medium term gains to national linguistic standardisation.

Posted by: Scott Martens on May 8, 2003 01:53 PM

At the other end of the spectrum we have Japan. A country populated by a people as inept at the English language as it gets. Of course, it's everything else being equal : it may be that Japan scores just so well on other determinants of development that it's been able to make up for its "language barrier".

Yet, the fact that Japan is nowhere with respect to software probably has something to do with this. In any case, regarding products the production of which is not deeply rooted in language, it would seem to me that smart distribution argreements coupled with an elite who can carry the key information to domestic producers should be enough. In fact, Japanese producers are known to know their foreign demand exceptionally well (an order of magnitude better than we do theirs, and even, sometimes, ours.)

Posted by: Jean-Philippe Stijns on May 8, 2003 02:33 PM

Speaking English or having English as national lingua franca has, in itself, bugger-all to do with economic success, as the near or total failure of a number of English speaking ex-colonies from Liberia to Nigeria demonstrate.

English is the international lingua franca, and the most essential second language globally, but it's not going to replace people's mother tongues. Knowing enough English to fill in an invoice is not the same thing as being able to dream, write a poem, or woo a loved one, in English.

Posted by: StrontiumDog on May 9, 2003 11:24 AM

"Speaking English or having English as national lingua franca has, in itself, bugger-all to do with economic success"

Um, I think I was talking about the future, and about information exchange. I also didn't mention economic 'success'. I just implied there might be an in-built advantage, and that for eg India, or South Africa might be able to get leverage from this advantage in a way that China can't, perhaps thats why China is a manufacturing and not an information hub. There is also the little distinction between necessary and sufficient conditions.......

The curious thing is that people are really terribly sensitive on this topic, the interesting thing would be to look at why.

Posted by: Edward Hugh on May 10, 2003 04:30 AM
Post a comment