Tuesday, December 2, 2008

The (roughly) top 20 languages by GDP

There are many reasons to pick a particular language to learn. If there is a de facto second language of importance in your country (Spanish in the States, French in Canada, etc.), it probably makes a lot of sense to choose that language, particularly if you don't see any globe trotting in your future. If your relatives or spouse speak another language, it's not a bad idea to learn that language. And many still choose a language because they like the way it sounds (français, anyone?).

My preferred method of choosing which language to learn is based on its economic utility. As I do tend to be a bit of the globe-trotting type, I've never really limited myself to any region or the like. Without such limitations, it makes a lot of sense to choose the languages you learn based on the percentage of world GDP represented by speakers of those languages.

And that was roughly how I chose the languages I've studied: Chinese, Japanese, Spanish, German, French, Portuguese, and Italian. I knew which countries had the largest economies and I put their languages on a checklist, so it's hardly a coincidence that the languages I speak coincide with those at the top of the top-20 list below.

RankLanguagePercentage of world GDP

This list was put together by Unicode.org and the data covered by the list runs from 1975-2002, although projections through 2010 for the most part retain the same order. One interesting thing to note about the projections is that the "other" group declines to only 10%, meaning the relative importance of the top 18 languages increases. It's also worth noting, if unsurprisingly, that internet use by language largely corresponds with this list, according to data collected on Wikipedia.

There are two things that this data leaves me wanting. First is the obvious update of the data to cover through 2008, as well as longer projections going forward. Second, I'd love to see a chart based on the actual number of speakers of a all languages, rather than on the number of native speakers. For example, estimates for the number of English speakers vary from around half a billion to a billion, depending on the skill level at which you count someone as a "speaker". This would result in counting multilingual people multiple times, which might make it trickier to slice and dice the data to get a GDP figure, but I'm sure some enterprising statistician somewhere could get something we could work with.

Given the large percentage of GDP controlled by English speakers, it seems to be quite a rational choice that, if you don't already speak English, you learn it. I imagine that English's role as a global lingua franca would push it even higher if that enterprising statistician I mentioned above came along to give us the data.

I came across one memorable instance of seeing English function as the global lingua franca while interning during college in the Japanese Diet (Japan's legislature) in the office of Yuriko Koike, whose background includes a degree from Egypt and having literally written the book on speaking Arabic in Japanese. During one visit with some Arab dignitaries of some sort, they lamented that neither side was learning the other's language, but rather using English as a medium for communication. They suggested increasing learning on both sides, but somehow I doubt much ever came of it.


  1. Very good idea. For someone who plans to change countries once in a while definitely a good way to prioritize.

  2. I like the thought of learning languages based on economic utility. I think it's pretty reflective of opportunities to actually use the language - either for business or for tourism.

    However, I think this data is too old. A lot has changed in the world over the past decade -- in particular, this data ends where 9/11 begins, so the Bush policies aren't reflected. Neither is the global economic climate.

    Also, I think there are better numbers than GDP. I don't know a lot of economic terminology, but I'm aware of PPI, for instance.

    And finally, as a personal preference, I also think it would be useful to reflect the number of countries where a language is spoken. In doing so, I think Spanish, Portuguese, Russian, and Arabic get better placement on the list, and Chinese and Japanese get adjusted downward a bit. Not to discredit the economies in China or Japan, but its that whole eggs-in-one-basket thing that I dislike.

  3. There are more than one Chinese language and they cannot understand each others speech (Cantonese, Shanghainese, etc)

  4. Two comments...

    First, the data appears to consider Sinitic languages separately from each other, as Cantonese is mentioned separately (although I suppose that might just be Hong Kong and Macau being tallied separately).

    Second, Mandarin is still spoken by just about everyone most of us would ever need to do any economic business with in China, so I'd wager that China's position in the data above would not change much if you discounted all Chinese people who can't speak Mandarin.

  5. I would hesitate to count Cantonese, Shanghaiese and Mandarin as separate languages. The tones are spoken differently but the characters assigned to the tones are almost identical. Besides, you did business with Mandarin, not Shanghainese or Cantonese. Shanghainese and Cantonese are becoming peripheral in the Chinese-speaking world.

  6. I wouldn't.  They're mutually unintelligible, even if the writing's mostly the same.  The difference is a lot like that between Spanish and Portuguese.

    But regardless of where you fall on that point, I'd agree that they are getting pushed aside by the dominant Mandarin.