DP4003 Unilingual versus Bilingual Education System: A Political Economy Analysis
|Author(s):||Javier Ortega, Thomas Tangeraas|
|Publication Date:||August 2003|
|Keyword(s):||education, language policies, minorities, skills|
|Programme Areas:||Labour Economics|
|Link to this Page:||cepr.org/active/publications/discussion_papers/dp.php?dpno=4003|
We define an economy composed of two language groups. Value is created through bilateral trade between individuals who can speak the same language. The value of trade increases in each participant's level of education. We compare a bilingual education system, under which the individuals who take education become bilingual, with a unilingual system, under which the individuals attending school end up speaking the language of the politically dominating group only. Bilingualism is socially optimal when education levels are centralized. In the decentralized equilibrium, individuals (i) vote over education systems anticipating the future levels of education (ii) independently and simultaneously choose whether or not to take education. We show that in the unilingual system the returns to education for each member of the dominated group positively depend on the number of members of the same group attending school (a ?bandwagon? effect). Instead, under bilingualism, decisions to take education are negatively correlated across groups (a ?duplication? effect). For this reason, the equilibrium education levels may be higher under unilingualism, and there may be unanimity for unilingualism. We find that language conflict, whenever it arises, consists in a situation in which uniligualism is supported by the dominant group, while bilingualism is defended by the dominated group. We characterize also the conditions under which unanimity for bilingualism arises. The predictions of the model are shown to be compatible with the almost unanimous choice of a bilingual Finnish-Swedish education system in Finland (1919-22) and the choice of a unilingual French-language system in France (1789-94).