The labour market performance of vulnerable groups: towards a better understanding of the main driving forces

In various international reports studying employment across countries and over time (see for example OECD or IMF reports), some groups are systematically highlighted as more vulnerable or less present in the labour market. This is the case of the youths, older people, women, people with a foreign origin and those with a low level of education. The aim of this thesis is to better understand the main driving forces explaining the labour market situation of those vulnerable groups, with a special attention devoted to people with a foreign origin (Chapter 2 and Chapter 3) and women (Chapter 4). Both institutional settings and personal characteristics are deeply analysed.Using data from European countries, Chapter 1 provides robust estimates on how product and labour market regulation could influence the unemployment rate. Controlling for country fixed effects, endogeneity, and a large set of covariates, results show that product market deregulation overall reduces the unemployment rate. This finding is robust across all specifications and in line with theoretical predictions. However, not all types of product market reforms have the same effect: deregulation of state controls and in particular involvement in business operations tend to push up the unemployment rate, while barriers to entrepreneurship and barriers to trade tend to reduce it. Regarding labour market deregulation, our estimations show an increase in the unemployment rate in the short run, while a positive impact (i.e. a reduction in the unemployment rate) occurs only in the long run. Analysis by sub-indicators shows that reducing protection against collective dismissals helps in reducing the unemployment rate. This Chapter also analyses how unemployment’s effects of deregulations could vary depending on the type of workers. Although men and women are equally affected by product and labour market deregulations, workers distinguished by age and educational attainment are affected differently. In terms of employment protection, young workers are almost twice as strongly affected as older workers. Regarding product market deregulation, highly educated individuals are less impacted than low- and middle-educated workers.One of the main studied vulnerable groups with respect to labour market performance is people with a foreign origin. This group faces various obstacles when trying to find a job and therefore present an important employment gap compared to natives. Thanks to a novel database constructed with Belgian data, Chapter 2 provides a comprehensive quantitative assessment of the employment performance of first- but also second-generation immigrants compared to that of natives. Results show that not only first-generation immigrants face a substantial employment penalty (up to -30 percentage points) vis-à-vis their native counterparts, but also that their descendants continue to face serious difficulties in accessing the labour market. For descendants of two non-EU-born immigrants the social elevator appears to be broken. Indeed, estimates suggest that their employment performance is no better than that of their parents (whose penalty averages 19 percentage points). Immigrant women are also particularly affected. While they are all found to face a double penalty because of their gender and origin, for women originating from outside the EU the penalty is generally even more severe. Among the key drivers of access to employment, the Chapter highlights: (1) education (especially for second-generation immigrants from non-EU countries), and (2) proficiency in the host country language, citizenship acquisition, and (to a lesser extent) duration of residence for first-generation immigrants.Beyond personal characteristics, institutional settings and integration policies could improve the labour market integration of people with a foreign origin. Chapter 3 aims to empirically analyse the role played by integration policies in improving the labour market integration of immigrants in EU countries, not only directly, through the employment rate but also indirectly by influencing the intensity and the composition of immigration flows. Relying on the Migration Integration Policy Indicator (MIPEX), we find that different types of integration policies come with opposite significant effects. Their impact is also found to vary with immigrants' origin (EU or non-EU) and education level, suggesting that there is no one-fits-all integration policy. Among our results, we note that anti-discrimination policies as well as political participation appear to be effective in fostering the labour market integration of non-EU immigrants, especially if they have a lower level of education. Conversely, less stringent rules for accessing host country citizenship seem to reduce the overall employment rate of non-EU immigrants. Furthermore, based on the estimation of a gravity model, we show that more elaborate integration policies act as a pull factor for high-educated immigrants, in particular those coming from another EU country. There is therefore an indirect positive effect of integration policies on immigrants’ employment rate in the host country, as high-educated (EU) immigrants tend to integrate more easily into the labour market.Another group facing specific obstacles in the labour market is women. Despite a remarkable convergence in the economic roles of men and women in society, gender gaps in terms of labour supply, earnings, and representation in top jobs persist. In the past, one of the main cited drivers of gender gaps was differences in education. Nevertheless, over the last years, the share of tertiary educated women has constantly increased, even leading nowadays to a larger proportion of high-educated women than men in Europe. One of the remaining explanatory factors is motherhood and the induced choices made by women. While previous literature showed that motherhood has a negative impact on women’s labour market outcomes, the heterogeneity of the effect according to origin has been ignored. Based on the Belgian Labour Force Surveys between 2004 and 2019, Chapter 4 aims to estimate the effect of parenthood on the labour market position of men and women according to their origin and education level by using a propensity score matching technique. Our results show that women of foreign origin are more likely to experience a decline in employment after motherhood than native women, and this is particularly marked for women from North Africa and the Near and Middle East. These differences are less clear-cut when analysing the propensity to work part-time, with all women, irrespective of their origin, being on average 14 percentage points more likely to work part-time when they become mothers than their non-motherly counterparts. We find that the effects on employment and part-time work increase with the number of children. Our results also show that motherhood has no significant effect on overtime but on the probability of working atypical hours. Differences by origin also appear when we consider the educational level of women. Low-educated native women tend to leave the labour market, while high-educated native women are more likely to reduce their working hours when they become mothers. Conversely, low-educated women of foreign origin tend to reduce both their level of employment and their working time. As the comparison is made with women of the same origin, our results add to other differences in labour market outcomes by origin measured by previous literature.