WORKING PAPER SERIES EUROPEAN WOMEN WHY DO(N T) THEY WORK? NO 454 / MARCH by Véronique Genre Ramón Gómez Salvador and Ana Lamo - PDF

Description
WORKING PAPER SERIES NO 454 / MARCH 2005 EUROPEAN WOMEN WHY DO(N T) THEY WORK? by Véronique Genre Ramón Gómez Salvador and Ana Lamo WORKING PAPER SERIES NO. 454 / MARCH 2005 EUROPEAN WOMEN WHY DO(N T)

Please download to get full document.

View again

of 20
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Celebrities

Publish on:

Views: 24 | Pages: 20

Extension: PDF | Download: 0

Share
Transcript
WORKING PAPER SERIES NO 454 / MARCH 2005 EUROPEAN WOMEN WHY DO(N T) THEY WORK? by Véronique Genre Ramón Gómez Salvador and Ana Lamo WORKING PAPER SERIES NO. 454 / MARCH 2005 EUROPEAN WOMEN WHY DO(N T) THEY WORK? 1 by Véronique Genre, Ramón Gómez Salvador and Ana Lamo 2 In 2005 all publications will feature a motif taken from the 50 banknote. This paper can be downloaded without charge from or from the Social Science Research Network electronic library at 1 The views expressed in this paper are those of the authors and do not necessarily reflect those of the European Central Bank (). We would like to thank Neale Kennedy, Barbara Petrongolo, Jarkko Turunen and Etienne Wasmer for useful comments. Any errors are of course the sole responsibility of the authors. 2 Corresponding author: European Central Bank, Directorate General Research, Kaiserstrasse 29, D Frankfurt am Main, Germany; European Central Bank, 2005 Address Kaiserstrasse Frankfurt am Main, Germany Postal address Postfach Frankfurt am Main, Germany Telephone Internet Fax Telex ecb d All rights reserved. Reproduction for educational and noncommercial purposes is permitted provided that the source is acknowledged. The views expressed in this paper do not necessarily reflect those of the European Central Bank. The statement of purpose for the Working Paper Series is available from the website, ISSN (print) ISSN (online) CONTENTS Abstract 4 Non-technical summary 5 1. Introduction 7 2. Empirical model 3. Results Prime-age females participation Participation of females aged Participation of females aged Discussion and conclusions References Tables and charts Annex Data sources and definition 33 European Central Bank working paper series 35 3 Abstract This paper provides an empirical study of the determinants of female participation decisions in the European Union. The analysis is performed by estimating participation equations for different age groups (i.e. young, prime-age and older females), using annual data for a panel of 12 EU-15 countries over the period Our findings show that the strictness of labour market institutions negatively affects the participation rate. Decisions linked to individual preferences with regards to education or fertility are also found relevant to participation of the youngest and prime-age females respectively. The inclusion of a proxy to capture cohort effects is crucial in order to explain the oldest females participation. Keywords: labour force participation, labour market institutions. JEL classification: J21 4 Non technical summary GDP per head in Europe is around 65% that of the United States and the main reason for this gap is the relatively low labour utilisation in European countries. In particular, Europeans work 13% less hours on average than Americans do. While the existence of the gap on hours worked per head is not disputed, there is some disagreement as how to explain it. Some argue that the gap reflects social norms according to which Europeans value leisure more than Americans do, while others put emphasis on the influence of the institutional framework, in particular the disincentive effect of taxes on labour. This paper indirectly contributes to this debate by analysing another aspect behind the relatively low labour utilisation in Europe, i.e. low participation rates. Concretely, it examines the determinants of women s participation in Europe. Using annual data for a panel of 12 EU-15 countries over the period , this paper looks into the role that economic, institutional and social factors have played in explaining female participation rates and their developments in the last two decades. The analysis is performed by estimating participation equations for 3 different age groups (namely, the young 15 to 24 years old, the prime-age 25 to 54 and the older 55 to 64 ). We find that the influence of general economic variables on participation is not even across age groups. In particular, the unemployment rate, which is meant to mainly capture labour market pressures arising from business cycle fluctuations, appears to have some relevance in explaining participation, albeit only for the youngest and prime-age women. As expected, high unemployment tends to have a discouragement effect on these women, who, when considering that they have little chance to find a job, refrain from participating to the labour force. For all three age groups considered, the strictness of labour market institutions negatively affects participation decisions. Indeed, the declining tightness of the institutional framework ruling labour markets in some countries over the last two decades appear to have significantly favoured women s participation, in particular of the older age group. We also find that measures aimed at reconciling motherhood with professional life, such as maternity leave, seem to favour participation to the labour force. Along with the changing institutional framework, a number of specific decisions influenced by preferences, institutions and social norms appear to have driven participation developments of the different age groups. For example, for the younger age group, the increasing use of flexible forms of work (such as part-time employment) is found to explain a large part of participation developments between the 1980s and the 1990s. Education enrolment, which can be seen as an alternative to finding a job or being unemployed, is also a major driving force behind the decline in young female participation. Turning to prime-age female, a fertility variable, which can be seen as the outcome of the decision to have children versus participation to the labour force, significantly curbs labour force participation. Finally, the participation of the oldest age group appears significantly influenced by the lifetime behaviour of older women. If they did not participate at an 5 earlier age, they are not likely to participate at an older age. Once again, this points to the influence of cultural habits and social norms. In sum, our results discard any doubt on the influence of institutions on women s participation in Europe. Institutional factors, together with preferences and economic factors have shaped female participation decisions. Identifying these factors is all the more important since it appears as a prerequisite to the successful design of efficient policy measures. These measures should help attract female labour supply into the labour force, raise female participation towards the targets set by the 2000 Lisbon European Council and help Europe to be more dynamic and catch-up with the United States in terms of hours worked per head, and thereby GDP per head. 6 Introduction GDP per capita in Europe was around 65% of US GPD per capita in One possible reason for this gap is that Europeans work less than Americans do. Indeed, the ratio of hours worked per person in the EU-15 relative to the hours worked per person in the US was around 77% in 2000, which represents a significant decrease from a ratio of around 100% in This decrease was coupled however with an increase in relative productivity that left relative GDP per capita broadly unchanged, i.e. the increase in productivity did not have a catch-up effect in relative GDP per capita because labour utilisation in Europe was clearly below that of the US. In other words, if Europeans were prepared to work as many hours per person as Americans, GDP per capita in the EU-15 would be comparable to that of the US. While the existence of the gap on hours worked per capita is not disputed, there is some disagreement as how to explain it. The debate on whether Europeans work less due to preferences or due institutional features of the labour market has recently attracted considerable growing interest. The paper that initiated this debate was Blanchard (2004); Blanchard decomposes the change in hours worked per person as the sum of the change in hours worked per worker, the change in the ratio of employment to the labour force and the participation rate. He shows that the decrease in hours worked per person in EU-15 relative to the US during the period is only due to a decrease in the number of hours worked per worker, with no substantial change in the relative growth (EU-15 vs. US) of employment and participation rates. Blanchard argues then that the fact hours per worker (in particular full time workers) is responsible for the decrease in worked hours per capita is an indication that Europeans work less because they prefer to do so, thus valuing leisure relatively more than Americans. Prescott (2004) and Ueberfelt (2004) on the other hand, find that some institutional features, in particular taxes, play a determining role in shaping the dynamics of hours worked per capita. However, the latter papers do not disentangle whether the effect of taxes is via hours per worker, via employment and/or via participation rates. This paper contributes indirectly to this debate by examining the determinants of just one of the components of worked hours per capita: the participation rate; and more precisely female participation rates in Europe. We think that increasing the female participation rate in Europe is a feasible channel for increasing hours worked per capita and in turn GDP per capita. Indeed, participation levels in Europe are low, implying that around 76 million people aged between 15 and 64 are currently not actively participating in the labour market, the so-called inactives, of which two-thirds are women. 1 1 Europe in this paper refers to the 15 members European Union prior to the May 2004 enlargement. Both terms Europe and EU-15 will be used indiscriminately. 7 Labour market participation in the EU-15 has increased from 65% in the mid-1970s to nearly 70% in 2000 (see Chart 1), led by the positive contribution of female participation, which grew by more than 12 percentage points to reach around 60% in A similar increase in overall participation was recorded in the US over the same period, but EU-15 participation rate in 2000 was no higher than that of the US three decades earlier (in the 1970s). In 2000, the gap between US and European participation still reached 12 percentage points for women and 5 percentage points for men. This paper seeks to identify the determinants of the evolution of female participation in Europe. The analysis is performed by estimating women participation equations for three different age groups (young (15-24 years old), prime-age (25-54) and older age groups (55-64)) using annual macro data for a panel of 12 of the EU-15 countries over the period The separate analysis for each one of the relevant age groups of potentially active women can be easily justified. For example, fertility decisions are key for prime-age women but irrelevant for the old ones, while other factors such as social protection are important determinants for old women. An innovative feature of our paper is its focus on institutions as explanatory factors for participation rates. The analysis of the effects of institutions on unemployment and employment has largely developed in recent years since the dataset compiled by Blanchard and Wolfers (2000) and Nickell and Nunziata (2001) have become available. Nickell (1997), Elmeskov et al. (1998), Belot and van Ours (2000, 2001) and Bertola et al. (2003) are some examples of studies that focus on the direct effect of institutions, while Blanchard and Wolfers (2000), Jimeno and Rodriguez- Palenzuela (2001) and Nunziata (2002) follow an approach based on the interactions between institutions and shocks. We exclusively focus on the direct impact of institutions. To our knowledge this paper is among the first ones to look at the impact that institutions could have on participation decisions, other examples are Blöndal and Scarpetta (1999) who analyse retirement decisions and Jaumotte (2003), which also looks at females participation. Our findings discard any doubt on the influence of institutions on women s participation in Europe. Institutional factors can shape female participation decisions and therefore the dynamics of worked hours per capita and, in turn, GDP per capita and economic growth in the EU. However the degree to which these factors shape or interact with preferences is outside the scope of this paper. The strictness of labour market institutions negatively affects the participation decisions in the three groups considered. Indeed, changes in rules or in features that increased the overall flexibility of labour market institutions in many European countries in recent years (e.g. declining union density, decreasing employment protection, tightening of eligibility criteria for unemployment benefits, etc.) have generally supported female participation. We also find that institutional features By comparison, male participation rates fell by more than five percentage points to around 78% over the same period. Countries include Belgium, Denmark, Finland, France, Germany, Ireland, Italy, Netherlands, Portugal, Spain, Sweden and the UK. 8 aimed at reconciling motherhood with professional life such as maternity leave schemes favour participation of prime-age women. More linked to preferences, but also to social norms and institutions, fertility and education decisions have been dominant forces in determining the evolution of participation rates during the sample period considered for prime-age and young females respectively, while the extensive use of part-time employment has been relevant for both age groups. Finally, cohort effects, which are also linked to institutions, preferences and social norms, appear to be crucial to understand the developments in participation for oldest females. The paper is organised as follows. Section 2 presents the empirical model and discusses how the overall institutional framework and other variables traditionally suggested by economic theory are expected to influence labour market participation of women. Section 3 reviews the stylised facts and presents the estimation results Section 4 discusses the main findings and gives some conclusions. 2. Empirical model The general version of the empirical participation equation estimated for each group is the following: Log (PR it / 1- PR it )= βx it + γz it + b t +c i + ε it. The dependent variable is the logit transformation of participation rates of each group, for country i and period t. The participation rate is a variable bounded between zero and one. This can be taken into account by specifying the regression of participation rates on the conditioning variables as E(PR it ) = F (βx it + γz it + b t +c i ), where F is a probability distribution function. If F is chosen to be a logistic function, this model will lead to a linear equation with the transformed variable 4 log[(pr it /1-PR it )] as the dependent variable. b t and c i are period and country effects respectively. The explanatory variables include a range of plausible determinants of each age group s participation rate. These determinants may be organised into two main groups. Vector Z it gathers variables traditionally suggested by standard economic theory and usually found in empirical studies. Vector X it is meant to bring together variables reflecting the institutional framework ruling European labour markets and general preferences. For all age groups, the vector of market institutions X it includes the following elements: union density, employment protection, unemployment benefit systems and labour taxes, all of the taken from Nickel and Nunziata (2001). These variables are time varying and are available until In 4 The estimations in this paper were also performed using the actual (non-transformed) participation rate as dependent variable, the results did not show any relevant changes with respect to the ones presented in the paper. 9 addition, this vector includes a number of institutional features that are thought to be specific to each age group and will be discussed in more details in Section 3. The relevance of the institutional framework as a determinant of female participation decisions can be discussed on the basis of a standard insider-outsider model, in which prime-age males could be seen as insiders because they show a more stable attachment to the labour market, while women of all ages tend to be less attached to the labour market and can be seen as outsiders. Union density, defined as the ratio of union members to total employees, is a labour market institutional feature expected to be an important determinant of participation. The involvement of unions in the wage bargaining will lead to a higher wage compression and larger employment differences between insiders and outsiders, which will impact on the expectations of outsiders to get a job. Therefore, we expect more unionised economies to display lower average participation rates among the less attached workers, hence, women. During the 1980s and the 1990s, union density has declined in many countries and may have favoured a convergence in participation rates among age and gender groups. 6 Employment protection is another institutional feature that should play an important role in determining participation, while the overall influence may go in any direction. A stringent employment protection framework should restrain employers willingness to hire and fire workers. As a result, employment, hence, participation rates of workers with numerous lapses of inactivity should be negatively affected. At the same time, employment protection is also meant to provide a positive incentive to participate in a labour market where risks of getting unemployed are smaller. Employment protection is measured as a range from 0 to 2 depending on the relative strictness of employment protection in each European country. In general, employment protection is high in all European countries, but the strictness of it has significantly declined over the last two decades. The potential effects of unemployment benefits on participation follow a similar mechanism to the one of employment protection. To the extent that a more generous unemployment benefit coverage strengthens union bargaining strategies, it would reduce the employment rates of the outsiders and indirectly, their participation rates. On the other hand, however, the existence of more generous unemployment benefit systems could represent a positive incentive to participate in the labour market from the viewpoint of workers. Unemployment benefits may be described by two variables: the replacement ratio and the benefit duration. Replacement ratios are measured as the ratio of total unemployment benefits for the first year to average net earnings. They have generally decreased in the last two decades, although this has varied across countries. Unemployment benefit duration is a measure of the amount of time the unemployed keep on receiving unemployment benefits We have considered, at an early stage, the degree of wage bargaining co-ordination, but it was found to be insignificant in the estimations and, in general, highly correlated with other institutions. It should be noted that this is a quite imperfect measure of unionisation. There are countries with low affiliation rates like Spain or France where union agreements apply to all workers at a national level. 10 Developments in unemployment benefit duration during the last two decades have varied significantly across European countries: some have extended it, some have reduced it. Finally, labour taxes are also expected to be a relevant determinant of women s participation. Any increase in taxes leading to lower net wages will increase participation in order to keep the income level constant. However, the substitution effect (cheaper leisure) will tend to lower participation. Our labour taxes variable measures the sum of income and consumption taxes from countries national accounts. It shows a high dispersion among EU-15 countries and an in
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks