Linguistic skills and the intergenerational transmission of language

Language choices and skills have important economic consequences at both individual and aggregate levels. It has been documented that immigrants proficient in the host-country language enjoy enhanced employability and substantial wage premia. Moreover, these greater economic opportunities might be transmitted across generations. Studies in the United States and the United Kingdom found that parents’ English-speaking proficiency is positively related with children’s English proficiency. Thus, language-related economic inequalities persist across generations. This evidence raises two questions: through what mechanisms are language skills inherited, and whether similar results arise in bilingual societies.

In their BSE Working Paper No. 1053, “Linguistic Skills and the Intergenerational Transmission of Language,” Ramon Caminal, Lorenzo Cappellari and Antonio Di Paolo investigate the pattern of intergenerational transmission of language within the family in an asymmetric bilingual society: the Spanish region of Catalonia. Previous studies on transmission of language have focused on immigrants in societies with a predominant language. In the case of Catalonia, there are two main speech communities of similar sizes, Spanish and Catalan. Both languages are official and each of them enjoys the protection of a different layer of government. However, they do not hold symmetrical positions. Whereas all native Catalan speakers are bilingual, only a fraction of native Spanish speakers is fully proficient in Catalan.

Spanish is a very strong global language, with more than 400 million native speakers, while in the case of Catalan, less than six million people use it on a daily basis. Despite this, there is ample evidence for the value of bilingualism in Catalonia. It has been shown that knowledge of Catalan fosters employability, increases earnings, and increases the frequency of mixed marriages. In this environment, language transmission choices are far from trivial.

Theoretical framework

In a simple theoretical framework, the authors investigate the incentives and language choices in an asymmetric bilingual society. In this model, parents choose the language of usage with their children. The choice comes from optimally balancing a trade-off of preferences, predilection on native language, versus economic incentives, the benefits of using a second language to children’s future payoff. Therefore, the model predicts that an exogenous increase in parental second language skills induces a more intensive use of the second language with their children.

It is not straightforward to determine empirically the causality of the previous prediction. The researcher must isolate an increase in the skills of parents’ second language that is independent of the rest of factors influencing the choice of language spoken in the family. For example, unobservable characteristics, changes in the economic environment, in individual preferences, or mixed marriages could simultaneously affect the proficiency in the parents’ second language and the choice of language spoken. To isolate from these confounding effects, the authors exploit a natural experiment generated by a language-in-education reform that introduced Catalan-Spanish bilingualism to schools in Catalonia.

Empirical implementation

During the 1960s and early 1970s, there were large migration flows from the south and the interior of Spain to Catalonia. By the mid-1970s, 40% of the Catalan population was born outside the region. In that period, Franco’s regime (1939-1977) imposed Spanish as the only official language and the only one to be used in education.1 (In Franco’s regime, Catalan was excluded from most social activities but it was still intensively used in the private sphere and transmitted across generations.) As a result, for the first time in history, a very large fraction of the Catalan population was monolingual in Spanish.

With the advent of democracy, language policies drastically changed. A language-in-education reform, the Language Normalization Act (LNA), was approved in 1983, with the aim of making students fully bilingual. With its implementation, education slowly evolved from a system in which Catalan was excluded to one in which it was the main language of instruction in compulsory education. The implementation of the reform implied that, for cohorts already attending school in 1983, their exposure to bilingual education depended on the remaining number of years before completion of compulsory education. The authors take advantage of such reform to compare individuals with no exposure to bilingualism, those cohorts that already finished compulsory education in 1983, to those that were exposed to it. Moreover, as long as the language-in-education reform boosted oral skills in Catalan among native Spanish speakers (being native Catalan speakers already fluent in their native language), the authors exploit this heterogeneous effect of exposure to bilingualism at school as an exogenous source of variation that enables obtaining a causal interpretation of the results. They also control for birth-cohort effects, which implies the assumption that non-language trends across the cohorts are common between the two language communities.

The data used in the study is drawn from the Survey of Language Use of the Catalan Population produced by the Catalan Statistical Institute (IDESCAT). The survey contains unique information on sociolinguistic characteristics as native language, self-identification language, proficiency in both Catalan and Spanish, parental native language, and language normally spoken with the children. It also provides sociodemographic and linguistic characteristics of the respondent, his/her parents and eventually his/her partner. Using the data with the empirical strategy mentioned before, the authors find that an exogenous increase in parental second language (Catalan) leads to higher usage of Catalan with their children.


The main result from this paper is that, as parents improve their skills in the second language, the cost of switching to their second language falls. As a result, they are more likely to transmit their second language to their children. Better second-language skills improve the chances that the language is transmitted to the next generation. This is the first study to show that this mechanism also works for bilingual societies. The main policy lesson from the analysis is that any evaluation of language reforms that affect the skills of the current generation must take into account the positive inter-generational spillover effects.