Lionel Cottier, University of Lausanne
The impact of culture on individuals’ decisions is a question which has interested economists and other social scientists for a long time. People choose their job, follow a certain lifestyle and make choices based on many factors. Some choices are well understood, as they follow individuals’ preferences and institutional rules. However others are more ambiguous as they are influenced by beliefs and social norms, which are difficult to identify. Social policies aiming to address economic problems, such as labor market measures and insurances, are only effective as long as they correspond to people’s beliefs and norms. It is thus crucial to understand how these less tangible factors affect individuals’ behaviour.
The literature has already started to answer this question and it is now clear that both economic and cultural factors affect economic decisions. What is less clear is when and how we should expect cultural factors to come into play, which is the question that my research aims to answer. In this paper, I study how culture affects labor force participation near retirement in Switzerland, using the country’s cultural and linguistic divide.
Four national languages split the country into regions that do not necessarily follow institutional or geographical borders. Two main language regions emerge: the Romance-speaking one covers the South and West of the country, containing individuals speaking French, Italian, and Romansh; and the German speaking-one, which covers the North and East. These language regions show very different cultural traits, ranging from food habits to preferences on labor market regulations. By studying labor force participation at the language border, I compare individuals who differ in their cultural background but face the same economic incentives and institutional background. This allows me to circumvent the problem of endogenous institutional rules, which makes the estimation of cultural effects so difficult in a standard setting.
I first look specifically at male labor force participation in the ten years before their retirement age in 2000. Results show that Swiss-Romance individuals exit the labor force significantly earlier than their Swiss-German counterparts. One possible explanation could be that in their last working years, individuals enjoy greater financial and emotional stability, which allows them to follow their preferences and retire earlier if they wish to do so. This would in turn suggests that cultural factors come into play when the labor market is sufficiently flexible.
To test this hypothesis, I then re-do the analysis for 1970. The idea is that 1970 was a period where roles were more clearly defined—the husband was the breadwinner, while the wife took care of the household—and the welfare state was less supportive. In other words, the labor market was less flexible, meaning that cultural factors should play a smaller role according to my hypothesis. It is indeed what the results show, namely that culture played no role in men’s labor force participation. Finally I study labor force participation for males between 25 and 34 years old. If it is true that cultural factors kick in when individuals enjoy more freedom, then we should observe cultural differences for young workers entering the labor force for the first time. Again, results show strong differences between Swiss-Romance individuals, who enter the labor force very late, and Swiss-Germans.
Put together, the results suggest that cultural factors come into play when the labor market is most flexible. Results show that the behaviour of young and old workers is particularly affected. Also, the social evolution that took place in the last decades seems to give more room for culturally shaped determinants to affect labor supply.