What is a foreign language worth?

In a recent podcast the authors of Freakonomics, Stephen Dubner and Steven Levitt, addressed the question “is learning a foreign language really worth it?” In sum, Dubner and Levitt found that bilingualism can make people better decision makers and helps with executive function in children and dementia in older people. However, they also indicated that an American who learns a foreign language nets a mere 2% earning bonus. This earning bonus would translate into an additional $600 for an individual who earns $30,000 a year.
But is the benefit of learning a foreign language really as minimal as Dubner and Levitt assert? An article in The Economist seems to maintain that Dubner and Levitt may have misconstrued the true value of learning a foreign language.  
The article first asserts that Dubner and Levitt’s $600 earning bonus is entirely misleading and fails to take into account the long term effect. Rather than the $30,000, The Economist author believes that the more appropriate annual salary would have been $45,000, since this is the national average for individuals who graduate from a university. Moreover, Dubner and Levitt’s calculations fail to take into account salary increases and interest earned on the language bonus. For example, if the additional money earned from the language bonus were placed in a retirement account, the individual would amass nearly an additional $67,000 over the course of a 40 year career.  
Second, Albert Saiz, an MIT economist, found that the 2% language bonus used by Dubner and Levitt doesn’t accurately reflect the potential bonus earned by learning different languages. For example, Saiz found that the language bonuses for different languages are as follows: 1.5% for Spanish, 2.3% for French and 3.8% for German. Using the factors mentioned in the previous paragraph, the long term effect would amount to an additional $77,000 for French speakers and $128,000 for German speakers.  
But why do different languages account for different returns? Simply put, it all comes down to supply and demand, and in the American context more so supply. For example, in the US the number of individuals fluent in both English and German is much smaller than those fluent in English and Spanish and thus the German speakers are able garner higher language bonuses.  
Not only does bilingualism affect the earning capacity of the individual, but it has also been found to have a substantial effect on a country’s GDP (gross domestic product). Recently, a study conducted by the Cardiff Business School found that a lack of foreign-language proficiency costs Britain’s GDP nearly £48 billion ($80 billion) or 3.5% each year. With this in mind and considering an optimistic estimate that half the world’s population might speak English by 2050, there is still a considerable amount of demand for multilingual individuals.  
 In light of the aforementioned, it is clear that learning a foreign language may be a bit more valuable than previously thought, especially in languages such as French and German.
Best regards
und viele Grüße aus Charlotte
Reinhard von Hennigs