The average high school graduate spends two years studying a foreign language. (Digest of Education Statistics, Table 157)  What effect do these years of study have on Americans’ actual ability to speak foreign languages?

I started by looking at the Census, but it asks only about “languages spoken in the home.”  Gallup has a survey finding that one-in-four Americans can speak a foreign language, but it offers no further details that would allow us to measure degree of fluency or the effect of foreign language instruction.  After nosing around for better data, I turned to the General Social Survey.  As usual, I was not disappointed.

In 2000 and 2006, the GSS asked over 4000 respondents the following three questions*:

1. Can you speak a language other than English?  [Responses: Yes/No]  (OTHLANG)

2. How well do you speak that language?  [Responses: Very well/well/Not well/Poorly or Hardly at All]  (SPKLANG)

3. Is that a language you first learned as a child at home, in school, or is it one that you learned elsewhere?  [Responses: Childhood home/School/Elsewhere](GETLANG)

The results showed an even smaller effect of foreign language instruction on foreign language fluency than I expected.

25.7% of respondents speak a language other than English.  Within this sample, 41.5% claim to speak the other language “very well.”  Within this sub-sub-sample, just 7.0% say they learned to speak this foreign language in school.  If you multiply out these three percentages, you get 0.7%.  The marginal product of two years of pain and suffering per high school graduate: less than one student in a hundred acquires fluency.  (And that’s self-assessed fluency, which people almost surely exaggerate).

If you lower the bar from “very well” to “well” the picture remains grim: merely 2.5% of GSS respondents claimed to reach this level of foreign language competence in school.

Fans of foreign languages will probably just respond, “That’s why we have to pour more resources into foreign languages.”  I say it would be far better to give fans of foreign languages a free economics lesson.  Here goes:

1. Lots of stuff that sounds good isn’t worth doing.  “Learning a foreign language” sounds noble, but so does “climbing Mount Everest.”  The wise calmly weigh costs and benefits instead of being carried away by words.  Any honest scale will tell you that the costs of foreign language instruction dwarf the benefits.  Think about it: Even ignoring teachers’ salaries, we’re currently burning two years of class time per graduate.  The payoff?  Making less than one student in a hundred fluent.

2. Doubling an input normally less than doubles output.  The world usually has what economists call “diminishing returns“: you can improve outcomes by spending more money, but the more you spend, the less efficacious each dollar becomes.  The fact that two full years of instruction have almost zero effect implies that massive spending increases would be required to noticeably raise foreign language fluency.  Think about all the Canadian adults who don’t speak French after a decade of required study.

3. Foreign language fluency is more common in other countries for a reason.  People around the world strive to learn English.  Why?  Because English fluency frequently helps them get good jobs, meet interesting people, and enjoy culture.  Pretty obvious, right?  To understand why Americans don’t learn foreign languages, simply reverse this reasoning.  We don’t learn foreign languages because foreign languages rarely helps us get good jobs, meet interesting people, or enjoy culture.  Americans start in an unusually abundant and diverse economic, social, and cultural pool, so we have little reason to stray.  And if Americans do decide to sample other pools, we can literally travel the world without needing to learn a word of another language.

I may sound like a typical philistine economist.  So let me confess that I personally got a lot of value out of my two years of college German.  I’m an opera fan; knowing a bit of German enhances my experience.  But this hardly means that most Americans would benefit from learning a foreign language.  All romance aside, requiring Americans to learn foreign languages makes about as much sense as requiring them to hear operas.  What inspires the few, torments the many.  Elites who relish foreign languages and opera should show some tolerance for the rest of humanity instead of calling for government spending to correct a “problem” that’s only in our minds.

* To be precise, the GSS asked over 4000 respondents the first question; it only followed up with the other questions if the answer to the first question was “yes.”