„4 The current issue and full text archive of this journal is available at *&£2k~ www.emeraldinsight.com/1746-5265.htm
BJM
5,2
156
Received July 2009 Revised October 2009, November 2009 Accepted February 2010
Baltic Journal of Management Vol. 5 No. 2, 2010 pp. 156-184
© Emerald Group Publishing Limited 1746-5265
DOI 10.1108/17465261011045106
The extended business case for childcare and leave arrangements in Western and Eastern Europe
Laura den Dulk
Department of Public Administration, Erasmus University Rotterdam, Rotterdam, The Netherlands, and
Pascale Peters, Erik Poutsma and Paul E.M. Ligthart
Institute for Management Research, Radboud University Nijmegen, Nijmegen, The Netherlands
Abstract
Purpose - The purpose of this paper is to propose an "extended conceptualization of the business case" including both organizational characteristics and institutional conditions to analyse employer involvement in extra statutory childcare and leave arrangements. Special attention is given to Central and Eastern European (CEE) countries.
Design/methodology/approach - The (multi-level) multinomial regression analyses included company-level data on human-resource practices of 2,865 firms nested in 19 countries, representing all European welfare state regimes.
Findings - The extended business case appeared fruitful in order to explain variations in employer involvement. Particularly, state support was found to be negatively related to employer involvement. In the liberal regime, employer involvement was high, but variations across organizations were significant. In CEE-countries, employer involvement was lowest, and did not vary by organizational business-case factors.
Research limitations/implications - The paper used data from a cross-sectional survey. To capture the long-term trends, dynamics and nuances in employer involvement within and across various institutional contexts, a longitudinal in depth study is needed.
Practical implications - While state support in many CEE countries is declining, the analyses showed that employers will not automatically step in by providing additional work-family arrangements. Social partners could use institutional pressure to stimulate a balance between state support and employer involvement.
Originality/value - The extended business-case perspective contributes to the theory on the institutional embeddedness of decision making of employers. Moreover, it adds to the knowledge on employer involvement in institutional contexts which have hardly been studied before.
Keywords Human resource management, Child care, Employers, Western Europe, Eastern Europe
Paper type Research paper
Introduction
With women entering the workforce in ever greater numbers over the past few decades, and with the associated rise in the number of dual-income households, many European governments have introduced statutory provisions to ease the combination of work and family life (Meyers and Gornick, 2003; den Dulk and van Doorne-Huiskes, 2007). In Scandinavian countries and post-communist countries, state provisions have a relatively long tradition, while in other countries these are more modest and more recently introduced. In case state provisions are limited, employers may become
involved and may supplement statutory provisions by offering additional formal and informal work-family arrangements, such as enhanced leave arrangements and extra childcare support (Appelbaum et al, 2005; den Dulk, 2001; Poelmans et al, 2003).
In Western Europe, under the influence of the neo liberal approach, the discussion on employer involvement, that is, the introduction of voluntary childcare support and leave arrangements that extend existing legislation, has been framed as a "business case". The emphasis on the business case for work-family support is in particular prevalent in Anglo-Saxon countries where state provisions are limited. Within the business case, the focus is on outcomes of work-family arrangements for organizations, and how organizational conditions affect costs and benefits of arrangements (den Dulk, 2001; Lewis, 1999). For instance, employers are more likely to introduce arrangements, such as childcare support, when this is expected to increase the productivity, recruitment and retention of valuable workers. The anticipated costs and benefits of work-family arrangements depend on organizational conditions, such as size, sector and workforce characteristics, resulting in variations of employer involvement.
Notably, however, most prior research on employer involvement has been done in Anglo-Saxon countries, such as the UK, the USA and Australia, countries characterized by a liberal welfare-state regime (den Dulk, 2005). The question can be raised, therefore, whether business-case arguments also hold in other institutional contexts, and whether we should not extend the business case by including institutional conditions besides organizational conditions. In order to answer this question, we need to extend our focus to employer involvement in other regimes. So far, cross-national research has been limited, and post-communist countries of Central and Eastern Europe (CEE) have seldom been included in existing research.
The present study aims to contribute to the existing literature in several ways. First, we extend research on work-family arrangements in organizations to other welfare-state regimes in Europe. More specifically, we focus on regimes in CEE. The inclusion of CEE countries in the comparative analysis is interesting since these economies have faced considerable transformations where established institutional arrangements eroded rather quickly. A relevant question is to what extent in CEE countries arrangements have moved to a situation where the business case will prevail in a similar way as in liberal market economies. Second, there has been very little large-scale empirical cross-national research on employer involvement in Europe. Third, our study focuses on employer data in order to capture their involvement over and above statutory requirements. Other studies are mainly based on employee surveys. These suggest, for instance, that employers in Germany, Austria and Southern European countries offer extra-statutory leave and childcare arrangements more frequently than employers in the USA or Canada (Organisation for Economic Co-operation and Development (OECD), 2001). However, information based on self-reported data from employees should be interpreted with caution. Employees are not always aware whether their workplace is offering additional facilities, or whether employers are merely following statutory requirements. Fourth, a four-country comparison of The Netherlands, Italy, the UK and Sweden, based on organizational level data, showed that employers are not likely to introduce a large number of work-family arrangements when advanced statutory provisions are present. In contrast, higher employer involvement was found in regime types with more minimal government provision (den Dulk, 2001). This four-country study indicates that the adoption of work-family arrangements by employers is a function of an interaction
between organizational conditions, and the level and nature of statutory provisions, and economic and labour market conditions. This paper extends this earlier study by including a larger number of European countries in order to examine the impact of different national-context determinants more thoroughly.
The present study focuses on the influence of the business case on the adoption of extra-statutory childcare and leave arrangements in organizations across different welfare state regimes, including CEE countries. In this paper, we extend the business case by including not only organizational characteristics, but also institutional conditions, like state involvement, societal debates on gender equality, and labour-market conditions. More specifically, we answer the question whether traditional business-case arguments apply similarly to employers outside the liberal, Anglo-Saxon countries. To do this, we elaborate the combined rational choice and institutional perspective as used by den Dulk (2001), arguing that employers' choice of extra statutory childcare and leave arrangements is also affected by institutional conditions.
Theory and hypotheses
The extended business case: a combined perspective
In the present study, we use a theoretical framework which is based on the rational-choice paradigm with a special focus on the institutional embeddedness of choices made by employers (Campbell, 2004; Jacoby, 2004).
Institutional theory. Since the 1990s, employer involvement has been object of study. In those years, much of the research which aimed at explaining the adoption of work-family arrangements by organizations started from a (neo)institutional perspective (Goodstein, 1994; Ingram and Simons, 1995; Osterman, 1995). Institutional theory emphasizes the institutional pressures that influence organizations to respond similarly to their environments. In all cases, these studies were Anglo-Saxon studies examining how employers respond to pressures in their environments. Institutional theory argues that organizations either adopt new practices because these are seen as the proper way to organize, or adopt new practices in response to powerful institutional pressures, such as demands of workers or government policies. These pressures are referred to as coercive, normative or mimetic pressures (DiMaggio and Powell, 1983; Tolbert and Zucker, 1996; Scott, 1995). Unlike the rational-choice perspective, the focus in the earlier studies on employer involvement was not on agency and choice, but employers were primarily viewed as passive actors responding to environmental changes (den Dulk, 2001).
Agency in the context of institutional pressures. The rational-choice perspective, in contrast, assumes that rational actors strive for the maximum realization of their goals (Coleman, 1990). Generally, actors will choose the alternative, or combination of alternatives, with the highest net benefit. Of course, costs and benefits of available options are influenced by constraints, i.e. the specific circumstances actors face. Traditionally, maximization of productivity and profitability has been considered the main goal of employers. However, the rational-choice perspective increasingly takes institutional pressures into account as being part of rational actors' choice frames. In the context of fiercer global competition, for example, present-day employers need to strive for other goals than short-term economic gain as well. Reputation, status in society, and harmonious employment relations, for instance, play an increasingly important role in the decision-making processes. Taking this pluralist perspective, work-family arrangements are viewed to contribute to organizations' goals, as these may contribute to productivity
and profitability of companies by improving employees' work-life balance (WLB), reducing absenteeism and turnover rates, improving employee identification and commitment, and subsequent outcomes. In fact, employees may even feel a "sense of entitlement" (Lewis and Smithson, 2006) to their employers' provision of work-family arrangements. In that particular case, it may be rational for organizations to develop these arrangements in order to gain status and a good reputation.
Organizational business case arguments. The rational-choice perspective can easily incorporate insights of the institutional theory regarding the impact of institutional conditions. Hence, in this study, it is assumed that employers weigh costs and benefits of the provision of work-family arrangements, given their particular organizational and institutional conditions.
With regard to organizational conditions, research that focuses on within country differences, for instance, shows that organizations who operate under similar institutional pressures, respond to these pressures differently. In many cases, large and public sector organizations are taking the lead with regard to the uptake of work-family arrangements (Appelbaum et al, 2005; den Dulk, 2001; Evans, 2001; Goodstein, 1994). Large organizations are obviously more sensitive to institutional pressures due to their visibility in the public arena (Goodstein, 1994). Moreover, they benefit from economies of scale, which reduces the costs of the introduction of work-family arrangements per employee in their organizations. Public sector organizations are also more often in the public eye and more likely to be judged according to government standards and norms. For private companies, in contrast, profit-related arguments are more important (den Dulk, 2001).
The proportion of women employees is also a relevant factor in employer involvement (Poelmans etal, 2003; Remery etal, 2003). Organizations with a substantial proportion of female workers may benefit more from the effect of work-family arrangements on productivity, absenteeism and turnover than male-dominated organizations (Lewis and Lewis, 1996; Goodstein, 1994), although the skill level of the female workforce may be an important mediator here (Ingram and Simons, 1995). Taking stock of the entitlement argument, a larger proportion of women employees can also result in more forceful requests enhancing pressures to conform; not responding to this demand may disrupt harmonious relations within the organization (Dex and Scheibl, 1999). A larger proportion of women in the organization can induce employee representatives to take up the issue. Union members within the company can also pressure employers to introduce work-family arrangements (Forth et al, 1997).
Moreover, the introduction of work-family arrangements as recruitment and retention tools will be particularly beneficial when organizations are experiencing a shortage or strong competition for valuable workers. Ackers (2003, pp. 227-8) concluded from his research that:
[...] the business case for harnessing human resources through work-family policies only works for certain employment groups, such as the more skilled and educated, during periods of full employment and labour scarcity.
Hence, based on the impact of organizational characteristics, employers should not be viewed as passive actors who always conform to external institutional pressures in similar ways. Instead, employers may fully conform to institutional pressures, or rather search for a compromise, make symbolic gestures, resist or even manipulate the institutional environment (Goodstein, 1994; Ingram and Simons, 1995; Oliver, 1991).
Childcare and leave arrangements
159
The need for an extended perspective. Owing to the lack of cross-national research, less attention has been paid to the impact of organizational conditions on employer involvement in interplay with national institutional pressures. The present study, therefore, aims to bring the focus on national institutional contexts back into the discussion, by focusing on both national institutional factors and organizational circumstances to explain employer involvement. Studying employer involvement, it should be noted that employers act in various institutional environments, such as country contexts or welfare-state regimes. Hence, whether there is a business case for the introduction of work-family arrangements not only depends on the organizations' circumstances or characteristics, but also on their institutional contexts.
Institutional-conditions-hypotheses. Focusing on the institutional and policy contexts affecting employer involvement, three relevant institutional conditions can be distinguished:
(1) the level of state support;
(2) normative pressures, including the societal debates on national gender equality; and
(3) labour-market conditions.
Based on these institutional business-case conditions, the following hypotheses regarding the influence of institutional conditions on employer involvement across countries can be derived:
HI a. In case the state takes full responsibility, there is less need for employers to offer additional work-family support. When state support is more moderate, on the other hand, enhanced childcare support and leave arrangements may offer a competitive advantage over other employers. Hence, we expect employer involvement to be lower in case of high state involvement.
HI b. Beliefs in a country about appropriate roles and behaviours of men and women are also likely to affect the business case for employer involvement. When women are seen as valuable workforce members, women's issues, such as the combination of work and care, are considered more important, and institutional pressure on employers to pay attention to work-family support will be larger (Lyness and Brumit Kropf, 2005). Summarizing, we expect employer involvement to be higher when gender equality is the norm.
HI c. The business case for employer involvement will be particularly strong in case of a tight labour market, in which retention and recruitment of personnel is difficult and highly competitive. Extended childcare support and leave arrangements may serve as a recruitment and retention tool. Hence, we expect employer involvement to be higher when the labour market is tight.
In the next section, we will discuss how European countries can be classified according to the institutional conditions included in this study. Special attention will be paid to cluster of CEE countries.
The business case across welfare state regimes in Europe
A classification. State involvement and the degree to which employers are expected to be involved in work-family support vary across countries in Europe. In Appendix, Table All,
we provide an overview of state support in the 19 countries in this study. To a certain extent, the degree of state support reflects welfare-regime types as explored by Esping-Andersen (1990, 1999). Based on his work and that of Blossfeld and Drobnic (2001), a distinction can be made between five different types of welfare-state regimes: the social-democratic welfare-state regime, the conservative regime, the Mediterranean regime, the post-communist regime and the liberal regime. Of course, in reality there are no pure regime types, differences within regime types do exist and some countries are difficult to classify (Hicks and Kenworthy, 2003). Nevertheless, the classification refers to the dominant character of the welfare package in a country, i.e. the role of the state, the family, civil society and the market. Therefore, the welfare-state typology can be viewed as a useful starting point for examining the way the institutional context shapes employer involvement.
By using the concept of de-familialization, Esping-Andersen (1999) refers to the degree to which welfare state or market provisions ease the burden of families' caring responsibilities. A de-familialized welfare-state regime is particularly characterized by an active public policy, including provisions such as childcare and services for elderly, which allows full participation of men and women in the labour market. In a "familialistic" welfare-state regime, in contrast, caring responsibilities are primarily seen as the responsibility of private households (Esping-Andersen, 1999). In the latter case, public policies may actually support care in the family, for instance by offering extensive leave arrangements (Leitner, 2003). When examining de-familialization through welfare-state policies, Esping-Andersen (1999) suggests that only the social-democratic regime is characterized by a high level of public work-family policies. Note, however, that CEE countries were not included in his analysis, and that Esping-Andersen used a limited conceptualization of de-familialization (Leitner, 2003; Szelewa and Polakowski, 2008).
The social-democratic regime. Finland, Sweden, Norway and Denmark are seen as examples of the social-democratic regime in which equality among its citizens and full employment for men and women is promoted. Dual earners are supported by an elaborate system of public work-family policies, which makes the combination of work and family life less difficult to manage. These four countries are characterized by universal services, such as a substantial public childcare system and a broad range of leave arrangements (Appendix 2, Table All). Moreover, gender equality is promoted by the introduction of paternity leave and specific daddy quotas within the parental leave system (OECD, 2007).
The liberal regime. Within the liberal regime, most closely resembled in Europe by the UK and Ireland, national regulations are limited and the development of work-family arrangements is left to market forces and framed as a business case. Because of the near absence of public provisions and the normative pressures put on employers (Lewis, 1999), there is a lot of scope for employers to develop work-family arrangements at the organizational level, in particular in a good economic climate with low unemployment (Appendix 1, Table AI). In this context, the employer involvement may offer a competitive advantage over other employers in terms of recruitment and retention. However, since organizational circumstances vary across industries and across individual employers, so do the costs en benefits of work-family arrangements. Therefore, diversity in employer involvement can be expected.
The Mediterranean and conservative regimes. Modest state support is also found in Mediterranean countries, such as Spain, Greece, Portugal and in countries with a more conservative regime, such as Germany, Austria and The Netherlands.
Childcare and leave arrangements
161
Both Mediterranean and conservative countries emphasize the role of the family and traditional gender roles. In contrast to the social-democratic regime, however, the societal debates in these regimes are less focused on gender equality. Because of this, the development of public work-family policies started relatively late and they are not as advanced as in the social-democratic and post-communist regimes. Moreover, unlike the liberal regime, there are less normative expectations regarding employer support for the combination of work and family life. Rather, the (extended) family is primarily held responsible. Labour-market participation of women tends to be low or reduced by the wide use of part-time employment in these countries (den Dulk et al, 2005).
The post-communist regime. In the post-communist regime of the CEE countries, traditionally, the state is an important actor regarding work-family provision. Under state socialism, women's labour market participation rate was high, and the common employment pattern was based on a family model of two full-time earners (Blossfeld and Drobnic, 2001). To safeguard fertility, and to enable the combination of work and care, the state developed extensive childcare services and leave arrangements. In fact, they were among the first countries to introduce work-family policies in Europe.
After the transition to the market economy, however, many CEE countries have witnessed a decline in childcare services and wage compensation for leave arrangements. In some countries, the decline has been more severe than in others (Pascall and Manning, 2000). For instance, Slovenia and Estonia are still characterized by high-state support for leave arrangements, while this is less the case in Bulgaria and Czech Republic where state support has diminished to a far greater extent (Appendix 2, Table All). Moreover, policy changes have been frequent in this area.
Hence, one cannot conclude that there is a general trend of re-familialization in CEE countries, rather country variations seem to prevail as well as policy shifts over time (Szelewa and Polakowski, 2008). With the diminishing of state support and the transition to a market economy the scope for employers to offer support is growing in many CEE countries. However, the transition to a market economy has often been accompanied by new experiences of job insecurity and unemployment, which is likely to limit expectations towards employer involvement. Moreover, the issue of gender equality at home was never much acknowledged or debated as in countries representing the social-democratic regime (Kocourkova, 2002; Pascall and Manning, 2000).
Ranking- and organizational-business-case-hypotheses. Taking the three institutional indicators included in this study into account (state support, gender equality and labour market conditions), it can be concluded that the business case in most CEE countries is likely to be weak, even though many CEE countries have witnessed a decline of state policies allowing scope for employer involvement. Based on an extended view of the business case, including both institutional and organizational conditions, in the following hypotheses employer involvement is ranked by welfare-state regime:
H2a. Owing to relatively high levels of statutory provisions, on the hand, and low expectations of employer involvement, on the other, we expect employer involvement in work-family arrangements to be the lowest in the post-communist regime of CEE countries.
H2b. Given its high level of state involvement, we expect organizations operating under the social-democratic regime to be next in the ranking of lowest employer
involvement, even though this latter regime is characterized by high national gender equality.
H2c. Countries operating under the conservative and Mediterranean regimes both place more emphasis on the role of the family, instead of the role of employer support. Mediterranean countries, in addition, combine lower normative pressure with relatively high-unemployment rates (with the exception of Portugal) (Appendix 1, Table AI), which is likely to lower employer involvement compared to conservative countries.
H2d. As countries in the liberal regime are characterized by modest state support, relatively strong expectations of employees towards employer support, and strong labour markets, we expect employer involvement in these countries to be the highest.
H3. Given the three institutional conditions typical for the organizations in the liberal regime, we will not only expect more employer involvement, but also more variation among employers depending on organizational conditions, such as size, sector, proportion of women, proportion of professionals and union membership. Therefore, we expect employer involvement in the liberal regime to show the largest variation along the lines of organizational characteristics.
Childcare and leave arrangements
163
Methodology
Data and design
The data source used for this research was the the Cranfield Network on Comparative Human Resource Management (CRANET) Survey on International Strategic Human Resource Management (HRM). This is a major survey of company HRM policies and practices in Europe and further afield. We used data from the 1999 to 2000 round of data collection. The choice of this time period is interesting, as it was the decade before several directives of the European Union (EU) came into effect (on equal treatment, on regulation of working time and on child care). This was a transition period when discussions on women and work may have influenced the early development of work-family arrangements in Europe.
The original 1999-2000 data set provides information on 7,038 organizations with more than 100 employees collected in various European and non-European countries (for details on the survey and the network, see Brewster et al. (1994)). For this study, we limited ourselves to 19 European countries (n = 6,267). The unit of response in each country was the "organization": this could mean the firm in its entirety, or a relatively self-contained unit within a company, such as a subsidiary.
The sampling frames used in each country were designed to produce stratified representative samples (by sector and size) and do so in the main for all the countries involved. However, due to slightly different sampling procedures in each country, the descriptive analysis cannot claim to provide a representative European overview. In the explanatory analysis, by including variables measuring organizational conditions, possible biases are controlled for.
In all but two cases the survey was conducted using a postal questionnaire. In two cases the survey was completed using face-to-face interviews with respondents. In most cases the response rates ranged from 12 to 20 per cent. The exceptions were Italy, where the response rate was just under 10 per cent (we therefore excluded Italy from the analysis),
and Sweden, where the response rate was in excess of 60 per cent. In each instance, the respondent was the person with responsibility for HRM in the business organization.
Translations of questions were slightly amended between national questionnaires to capture the nuances in meaning between languages. The questions were blind-translated twice by different translators (from English into another language, and then back again). Since the survey was not designed specifically to investigate work-family arrangements, it is unlikely that any completion of the survey would be biased towards firms with work-family arrangements. However, the emphasis on HRM strategies and practices in the questionnaire may have led to a disproportionate response from business organizations with relatively sophisticated and developed HRM systems and practices. On the basis that work-family arrangements tend to be associated with firms with well-developed HRM, this may have led to a slight bias in favour of the presence of these arrangements.
Operationalisation
The dependent variable. The focus in this study is on employer involvement denned as the provision of childcare and leave arrangements supplementary to statutory requirements. One of the sections in the CRANET-questionnaire was on compensation and benefits. The responsible HR-manager was asked whether the organization offers any schemes related to five arrangements: workplace childcare, childcare allowance, career break schemes, maternity leave (in excess of statutory requirements) or parental leave (in excess of statutory requirements). On the basis of these five single questions, a sum variable was constructed (values 1-5), indicating the number of childcare and leave arrangements offered by employers. Since the distribution of the dependent variable was not normal, we calculated a categorical variable with three values: (0) no provision of any extra statutory arrangements, which is used as the reference category; (1) the provision of one work-family arrangements and (2) the provision of multiple arrangements (more than 1).
The independent variables. Welfare regimes: for the purpose of our study, we selected a sub-sample of countries that represent the five European welfare-state regimes distinguished. The post-communist regime countries are the Czech Republic (n = 188), Bulgaria (n = 150), Estonia (n = 218) and Slovenia (n = 205). The liberal welfare regime is represented by the UK (n = 1,091) and Ireland (n = 446). Representatives of the conservative regime are France (n = 400), Germany (n = 503), The Netherlands (n = 234), Switzerland (n = 168), Austria (n = 230) and Belgium (n = 282). The social-democratic regime is represented by Sweden (n = 352), Denmark (n = 520), Norway (n = 391) and Finland (n = 290). The Mediterranean countries in this study are Spain (n = 294), Portugal (n = 169) and Greece (n = 136).
State support. The variable of state support at the time of the survey is derived from Appendix 2, Table All. Based on an extensive desk research, each country was rated according to public childcare provisions and according to the leave arrangements available to working parents on a four-point scale (high-state support (4) to low-state support (1)). High childcare support includes the right to a childcare place and high enrolment of both children younger than three years of age and older. Medium-high state support refers to substantial enrolment of children younger and older than 3, but no entitlement to childcare places (for instance, France and Belgium); medium-low implies almost no coverage of the young age group, but substantial enrolment among children older than 3. Low-state support refers to very limited public childcare.
Regarding leaves, length of leave, payment and leave for fathers were taken into account. High-state support was taken to refer to long, generous compensated leaves, including paternity leave and/or a specific daddy quota for fathers. Medium-high-state support for leave arrangements was taken to indicate long leaves and leave for fathers, but more minimal financial compensation. Medium-low state support refers to shorter leave periods, more unpaid leave and/or the absence of specific leave for fathers. Low-state support implies both the absence of paternity and parental leave, as is the case in Switzerland. Both scores were summated. The scores ranged from low (2) to high (8) state support (Appendix 1, Table AI).
National gender equality. To measure national gender equality we used the United Nations' gender-related development index (GDI). The GDI is based on the degree of gender equality regarding life expectancy, education and income (United Nations Development Programme, 2003). Among the countries in our study, Norway and Sweden have the highest ratings, while the CEE countries have the lowest (Appendix 1, Table AI).
Labour-market conditions. To cover the labour-market situation in general, we used the unemployment rate of a particular country in the year of data collection (Appendix 1, Table AI).
Organizational characteristics. As our theoretical framework suggests, workforce characteristics indicate the business case for employer involvement. For this reason, we included sector[l], organizational size, the percentage of women in the organization and the professional level of the workforce (percentage of professionals). The tightness of the labour market regarding professional workers can also be expected to affect organizations' willingness to offer work-family arrangements. Labour-market shortages were measured by means of a dichotomous variable representing whether it is difficult to recruit or retain employees in the professional categories (yes = 1; no = 0)[2]. Finally, the HR managers were asked about the proportion of the total number of employees in the organization who were trade union members. Their answers were taken as an indication of the degree of unionization of the employees in their firms. For the analyses, we used the relative degree of unionization of the firm, by dividing the firm's degree of unionization by the mean score of other companies in its particular sector and country.
Method
In order to test our hypotheses, we employ both descriptive and explanatory analyses. As our analysis includes company-level data of firms (n = 2,865) nested in 19 countries, the nested structure of the data is taken into account by using a multi-level random-intercept regression model (Rabe-Hesketh and Skrondal, 2005). We then performed multinomial logistic-regression analyses, using employer involvement as the dependent variable. The model takes on the following general form:
ya = B0 + Cj + Bi X ly+, ..., +Pp Xpy + síj
whereby are indicators of the use of one arrangements (first model) or indicators of the use of multiple arrangements (second model) for the z'th company in country;:
Childcare and leave arrangements
165
Bo
overall intercept.
= the country specific deviation from the overall intercept.
Bi to (3P = fixed parameters.
Sy = residual for the zth company in country j.
By conducting a multi-level analysis using countries as a random factor, the fixed parameters effects are controlled for the specific country effects. The analyses are done using STATA's (vlO.l) program GLLAMM (Rabe-Hesketh et al, 2004).
Results
Descriptive analyses
Ranking hypotheses. Table I shows the average number of work-family arrangements offered by employers across welfare-state regimes. On average, employer involvement in childcare and leave arrangements is highest in the liberal and the conservative regimes, and lowest in the post-communist regime, followed by the social-democratic regime. The descriptive analyses, therefore, supports our set of H2a-H2d, which predict employer involvement to vary across welfare state regimes. The ranking of employer involvement by welfare-state regime shown by the descriptive analyses is expected to result from the interplay of the three institutional conditions distinguished in this study: state support; gender equally and labour market conditions. In the next subsection, we analyze which of these three weighs most heavily in employers' decisions with regard to the provision of work-family arrangements.
Explanatory analyses
Institutional conditions hypotheses. In order to test the influence of each single institutional indicator, we employ a multinomial logistic-regression analysis, in which both institutional factors and organizational characteristics are entered in the equation. Table II presents the outcomes of three multi-level models explaining employer involvement with respect to the provision of one, or the provision multiple childcare and leave arrangements in excess of statutory requirements.
Model 2 presents the odd ratios explaining the relationship between state support, gender equality, and labour market conditions, on the one hand, and employer involvement on the other. In line with HI a, the odds of employer involvement are lower when more state support is provided. This holds true both with regard to one, and multiple work-family arrangements. In contrast to HI b and HI c, however, the odds of employer involvement are neither significantly related to the general unemployment rate in a country, nor to the GDI. Hence, the only significant institutional condition affecting employer involvement is the level of state support.
In Model 3, also organizational conditions are entered in the equation. Again, only state support is negatively related to employer involvement, both in case of one, and multiple arrangements, which supports our HI a.
Moreover, organizations are more likely to provide one work-family arrangement in case the organization employs a higher percentage of professionals and a higher percentage of women workers. Also when organizations face problems to recruit professionals and when organizations are larger, they are more likely to provide one arrangement. In case of multiple arrangements, a higher percentage of union members, a higher proportion of professionals in the organization, and a larger size increase the odds of employer involvement. In comparison with the public sector, organizations in the industry are less likely to offer multiple arrangements.
r/S
Ol
lo
oo co
S o
CM
CM CM
O
oo
55
oo o
o
Cvl
O CM
CO
lo cd
oo cm"
TO
a
o
T3
m 8
r/T <^
t/) CD		«7?
>		
CD		
	lo	Mean, : CRA
^ CD	p II	
mber ldcar	II CD M	Notes: Source:
8 < O	n s-H	
Childcare and leave arrangements
167
Table I.
Number of child care and leave arrangements
BJM
5,2
168
Table II.
Odd-ratios for the determinants of childcare and leave arrangements within welfare regimes
JO c
1
3
CO 1>- CD LO Ol CD t>- CI O
t>- ^ CI N O OO
CI CO ,—„ —. —. —. —. —. —. O O ,—, i—i —.
mn^iMOHcorocoHH ^NOjcoaiOioorocoH
i—i CO i—i i—i CO cm O O i—i O i—i
o o" o" o o" o" o" o o" o o
wooMincoGorococo^iM
NOim^HOOHCDincO
w co Nin o o
i o
H CD —. —. tr^ '—-'—-'—-'—-'—- ^ O- -—- <31 LO ,—, KHKirOO^IMIMmO^H HOiHCO^OOOKiNOlCli
rowrowrowooHOOH o" o" o" o" o" o" o" o" o" o" o" o"
Noooo^oimooNoooo
o o o o o o
I LO CD ^ C
S3«
cm oo cm
cm"
CD ,—, ,—,
^ m m
CO O CO
CD CO S
cm LO
I LO CO CD CD CO CD
q'1
CJ £j
! S-S
1 h a
cfi o oj 3 co u "T3   u O
m Oh Dh
" £ 01 ■' ■'
		
		
		
	O	
men	■fess	onal
: wo:	: pro	:essi
o	o	75
		
		
Tl    ^ m ^ f
73 mT7, % U   U ^
P   2 ^ S-h   S-h O
H pq O cl, Ph
So n i ft
o h
■a £?s
^ o £
T3
13
V
V
V
8S
■a £
.2 go
o
Z
Organizational business case hypothesis. The results presented in Table II refer to the impact of organizational conditions on the odds of employer involvement in all 19 European countries, regardless of them belonging to a particular welfare-state regime. In order to test H3 regarding variation in employer involvement within welfare state regimes by organizational characteristics, we need to do additional analyses. Table III shows the odds ratios of the determinants of employer involvement within each regime separately.
In line with H3, we find the number of organizational conditions affecting the odds of employer involvement to be highest in the liberal welfare-state regime. The odds of offering one arrangement vary according to sector, proportion of professionals and size of the organization. With respect to multiple arrangements, a similar picture emerges, with the exception that also union membership plays a role.
Hence, we find organizational characteristics to be most discriminating in the liberal welfare state regime than in other regimes. However, some organizational factors do also increase the odds of employer involvement in other regimes. In the conservative regime, for example, the proportion of women in the organization stands out as an important factor, besides the proportion of professionals, organizational size. With regard to multiple arrangements, employers are more likely to offer work-family arrangements in particular sectors. In the socio-democratic regime, both the percentage of professionals and size affect the odds of offering one arrangements. The provision of multiple arrangements is also affected by the shortage of higher educated workers and operating in business service sector. In the Mediterranean regime even fewer factors play a role: only sector, shortage of higher educated and size affect the odds of multiple arrangements.
Finally, no single organizational-level business-case variable affects the odds of employer involvement in the post-communist regime. Taking a closer look at the countries representing the post-communist regime, in case of adopting one arrangement, there are no significant differences across countries. In case of multiple arrangements, however, we find that the odds of employer involvement are lower in Estonia, Slovenia and the Czech Republic, taking Bulgaria as the country of reference. This variation across countries within the post-communist regime is in line with our HI a, suggesting that state involvement and employer involvement are negatively related. While state support is diminished within the post-communist regime in general, in contrast with Bulgaria, Slovenia and Estonia are still characterized by relatively high-state support for childcare and leave arrangements. Hence, the benefits for employers of employer involvement in these latter countries are lower than in Bulgaria.
Childcare and leave arrangements
169
Discussion
In this paper, we presented the results of a large-scale, cross-national empirical study on the involvement of employers in offering extra statutory work-family arrangements in the European workplace. We particularly focused on the extent to which employers in CEE countries offer voluntary childcare and leave arrangements in comparison with other employers in Europe.
In our theoretical framework, we fruitfully extended the conceptualization of the business case by focussing on how employer involvement is interrelated with the institutional contexts in which organizations in Western and Eastern Europe operate, and to what extent organizational business-case conditions interact with particular welfare
BJM
5,2
170
Table HI.
Odd-ratios for the determinants of childcare and leave arrangements within welfare regimes
Labels
(1) Liberal UK
Ireland Germany Belgium France
The Netherlands
Switzerland
Austria
Sweden
Denmark
Norway
Finland
Spain
Portugal
Greece
Bulgar
Czech
Estonia
Slovenia
Public sector
(2) Conservative UK
Ireland Germany
One arrangement
Multiple arrangements
Reference category 1.171 (0.282)
0.8
Reference category
Sweden
Denmark
Norway
Finland
Spain
Portugal
Greece
(0.232)
Industry	0.0506***	(0.0395)	0.0130***	(0.0104)
Personal services	0.108**	(0.0876)	0.147*	(0.115)
Process industry	0.0622***	(0.0482)	0.0424***	(0.0325)
Trade crafts	0.0371***	(0.0296)	0.0209***	(0.0165)
Business services	0.0673***	(0.0534)	0.0313***	(0.0247)
Other sectors	0.0302***	(0.0256)	0.00541***	(0.00533)
Percentage of women employed	1.007	(0.00497)	1.009	(0.00541)
Percentage of professionals	1.015**	(0.00484)	1.025***	(0.00510)
Shortage professionals	1.404	(0.293)	1.498	(0.334)
Unionization	1.210	(0.120)	1.451***	(0.145)
Organizational size	1.207*	(0.116)	1.639***	(0.157)
n		720		
Pseudoi?2		0.191		
Belgium	87.27***	(43.29)	12.73***	(7.101)
France	4.232***	(1.429)	3.833***	(1.248)
The Netherlands	37.66***	(22.32)	93.87***	(51.75)
Switzerland	14.78***	(5.871)	5.239***	(2.290)
Austria	1.145	(0.464)	1.134	(0.412)
{continued)
Labels
Bulgar Czech Estonia Slovenia Public sector
(3) Soc-democr UK
Ireland Germany Belgium France
The Netherlands
Switzerland
Austria
Sweden
Denmark
Norway
Finland
Spain
Portugal
Greece
Bulgar
Czech
Estonia
Slovenia
Public sector
One arrangement
Multiple arrangements
Industry	1.068	(0.663)	0.252**	(0.129)
Personal services	3.107	(2.436)	2.051	(1.353)
Process industry	1.241	(0.775)	0.194**	(0.102)
Trade crafts	0.765	(0.500)	0.0919***	(0.0552)
Business services	0.752	(0.501)	0.480	(0.258)
Other sectors	0.634	(0.467)	0.577	(0.344)
Percentage of women employed	1.019**	(0.00587)	1.018**	(0.00623)
Percentage of professionals	1.014*	(0.00580)	1.013*	(0.00567)
Shortage professionals	1.129	(0.282)	0.802	(0.199)
Unionization	1.026	(0.115)	1.128	(0.119)
Organizational size	1.797***	(0.189)	1.929***	(0.201)
n		703		
Pseudoi?2		0.299		
Reference category 2.478** 4.487*** 1.789
(0.858)	7.633***	(2.506)
(1.593)	3.667***	(1.346)
(0.653)	1.269	(0.501)
Industry	0.901	(0.401)	0.936	(0.390)
Personal services	0.537	(0.198)	1.154	(0.392)
Process industry	0.896	(0.320)	1.134	(0.396)
Trade crafts	0.695	(0.305)	1.774	(0.671)
Business services	1.421	(0.552)	2.892**	(1.036)
Other sectors	0.918	(0.490)	2.192	(0.947)
Percentage of women employed	1.006	(0.00518)	1.007	(0.00475)
Percentage of professionals	1.014**	(0.00473)	1.013**	(0.00414)
Shortage professionals	0.858	(0.199)	1.883**	(0.389)
Unionization	0.618	(0.212)	0.771	(0.222)
Organizational size	1.410***	(0.140)	1.301**	(0.117)
Childcare and leave arrangements
171
{continued)
Table III.
BJM
5,2
172
Labels
One arrangement
Multiple arrangements
Pseudoi?2
(4) Meditar UK
Ireland Germany Belgium France
The Netherlands
Switzerland
Austria
Sweden
Denmark
Norway
Finland
Spain
Portugal
Greece
Bulgar
Czech
Estonia
Slovenia
Public sector
780 0.117
Reference category
2.259* (0.917) 1.094
6.373*** (3.520)        12.46 *:
Table III.
(5) Post-corn UK
Ireland Germany Belgium France
The Netherlands
Switzerland
Austria
Sweden
Denmark
Norway
Finland
Spain
(0.494) (6.248)
Industry	0.264	(0.264)	0.0468**	(0.0476)
Personal services	0.639	(0.673)	0.322	(0.309)
Process industry	0.473	(0.426)	0.132*	(0.112)
Trade crafts	0.171	(0.174)	0.0830**	(0.0778)
Business services	0.709	(0.679)	0.121 *	(0.113)
Other sectors	0.290	(0.284)	0.117*	(0.108)
Percentage of women employed	1.015	(0.00818)	1.010	(0.00866)
Percentage of professionals	1.000	(0.00912)	1.014	(0.00879)
Shortage professionals	1.436	(0.520)	0.421 *	(0.173)
Unionization	1.113	(0.187)	1.181	(0.185)
Organizational size	1.164	(0.194)	1.609**	(0.262)
n		277		
Pseudoi?2		0.164		
{continued)
Labels
One arrangement
Portugal Greece
Multiple arrangements
Bulgar	Reference category			
Czech	1.188	(0.624)	0.133***	(0.0735)
Estonia	0.880	(0.529)	0.218*	(0.148)
Slovenia	0.627	(0.350)	0.0627***	(0.0450)
Public sector				
Industry	7.086	(8.162)	1.052	(1.156)
Personal services	7.188	(8.324)	0.340	(0.474)
Process industry	6.657	(7.619)	1.363	(1.446)
Trade crafts	6.715	(7.775)	0.756	(0.866)
Business services	6.943	(7.959)	0.495	(0.549)
Other sectors	1.158	(1.727)	0.591	(0.757)
Percentage of women employed	0.998	(0.00713)	0.992	(0.0102)
Percentage of professionals	1.004	(0.00822)	1.010	(0.0113)
Shortage professionals	1.452	(0.484)	0.783	(0.387)
Unionization	1.115	(0.194)	1.150	(0.206)
Organizational size	0.987	(0.158)	1.528	(0.338)
n			385	
Pseudoi?2			0.094	
Childcare and leave arrangements
173
Notes: Significance at: p < 0.05,    p < 0.01 and      p < 0.001; standard errors in parentheses
Table III.
state regime types. To examine the extended business case, we used several institutional indicators:
• the degree of state support for working parents (public childcare and leave arrangements);
• national gender equality (GDI); and
• labour market conditions (unemployment rate).
To study variations in employer involvement within particular institutional contexts, the three institutional indicators were discussed in relation to the five welfare-states regimes as classified by Esping-Andersen (1990,1999) and Blossfeld and Drobnic (2001). By investigating employer involvement across and within these different welfare-state regimes, the importance of the institutional embeddedness of organizational decision making was explored. In this final section, the most important results of the study are summarized and discussed.
Employer involvement across welfare state regimes
First, our study showed the average number of work-family arrangements offered by employers to vary across institutional contexts. In line with our "ranking-hypotheses", average employer involvement in childcare and leave arrangements could be ranked by welfare-state regime. These regime types represent, among other factors, a particular mix of the three institutional conditions distinguished in this study: state involvement; gender equality and national labour market conditions. By ranking employer involvement by welfare state regime, we found employers in the liberal regime to
provide the highest number of work-family arrangements, whereas the lowest number of work-family arrangements was found in the post-communist regime.
In line with our "institutional-conditions-hypotheses", employer involvement could be attributed to institutional conditions. Focussing on the direct effects of each single institutional variable, not all three institutional conditions distinguished in this study appeared to be important. In fact, only the degree of state support was found to be negatively related to employer involvement. Hence, well-developed public provisions were shown to diminish the likelihood of employers taking on an active role in the development of work-family support. This is in line with prior studies on voluntary employer involvement (den Dulk, 2001; OECD, 2001). However, none of the previous cross-national studies included CEE countries in their research design. In addition, previous studies based on employee reports do not highlight the leading position of employers in the liberal welfare states in Europe (OECD, 2001), but this might be related to the fact that employees are not always aware whether their employer is offering additional facilities or merely following existing legislation.
Strikingly, we did not find a significant effect of the other two single institutional conditions. National unemployment rates did not affect employer involvement significantly. However, our findings did suggest labour-market conditions to be important, be it at the organizational level. Hence, costs and benefits of work-family arrangements seem to depend on employers' perception of labour-market shortages with regard to higher-educated staff, rather than on macro-institutional level factors.
We also did not find a significant effect of the degree of national gender equality in a country with regard to life expectancy, education and income (GDI). This suggests that the normative pressures as being reflected in actual gender-equality indicators did not impact employer involvement directly. An explanation for this finding might be that normative pressures, expressed by more awareness for gender equality, affect the nature and degree of state policies on work-family arrangements. Therefore, although the GDI-factor in our research did not have the expected effect on employer involvement, we cannot rule out that normative pressures might have created a climate in favour of public provisions, rather than encouraging employers at the organizational level to become involved in work-family issues. In addition, national gender equality might be more relevant for the supportiveness of the work-family culture in an organization than the adoption of actual organizational policies as is suggested by the cross-national study of Lyness and Brumit Kroft (2005). In most institutional contexts, normative pressures exercised by a high percentage of women on the organizational level were also shown to have little impact on employer involvement. Yet, in the conservative regime, the percentage of women in the organization did have an impact. Possibly, the growth of female labour-market participation in this regime type has pressured both governments and employers to respond to employees' growing demand for work-family arrangements. As most governments in the conservative regime have only recently started to introduce national policies, the limited state involvement may have given more scope to employers to attract and attain female workers by offering work-family support. Hence, at the organizational level, normative pressures from the organizations' workforces may have engendered the provision of work-family arrangements.
In this study, significant effects were found for other organizational business-case arguments, however, not across all regime types, and not within all regime types to the same extent. In line with our "organisational-business-case-hypothesis", organizational
conditions were particularly relevant in the liberal regime. Obviously, in an institutional context in which the role of the market is emphasized, work-family arrangements are offered by employers anticipating that these arrangements support the business case, i.e. that their benefits exceed the costs. Regarding other regimes which are characterized by a weaker "organizational-business-case doctrine", for example the social-democratic regime, the business case became particularly relevant when the labour market for professionals was tight.
Employer involvement in CEE countries
Our study provided evidence that the business case for employer involvement within CEE countries was rather weak in the years 1999-2000, when the data were collected. In fact, none of the single organizational conditions appeared to be significant in employer involvement, and average employer involvement was low. Our findings, which were based on a large-scale data set, therefore, confirmed the impressions of national experts in CEE countries as reported in the European review on childcare and leave policies (Plantenga and Remery, 2005).
However, our empirical analyses focussing on the post-communist regime only, revealed country differences. This finding suggests variations in employer involvement across CEE countries. In their analysis of childcare and leave policies in CEE countries, Szelewa and Polakowski (2008) already showed large variations in existing policy models. Indeed, in some CEE countries, like the Czech Republic and Slovenia, state policies have developed in the direction of familialistic welfare state regimes. In contrast, some Baltic states, like Estonia and Latvia, have remained much more supportive of female labour-market participation and concentrate their policies on high public childcare arrangements, rather than on leave policies (Szelewa and Polakowski, 2008).
It should be noted, of course, that the post-communist regime is not the only regime in which the organizational business case appeared to be weak, when the number of organizational factors affecting employer involvement were taken into account. This also held true for the Mediterranean regime. In contrast to the post-communist regime countries, however, state provisions in Mediterranean countries were much more modest. This suggests that additional factors, besides state support, play a role. Most probably, in the Mediterranean context, the weak business case relates to the role of the extended family. Owing to the normative pressures placed on the extended family to take up family responsibilities, expectations towards employer involvement might remain low.
Implications
Our study found a negative relationship between types of state involvement and employer involvement. The negative relationship might refer to a general mechanism, in which employers are less inclined to offer additional provisions when governments take the main responsibility for the provision of work-family policies. It can be questioned whether the mechanism also holds in the opposite direction. Does employer involvement increase when governments step back with regard to the provision of work-family policies? The results of the present study, suggest that one should be careful to draw such a conclusion. In fact, in case government policies become more modest, employer involvement might not automatically increase.
Childcare and leave arrangements
175
This study was based on data collected in 1999-2000, a period in which many important transitions took place. Since that time, the development of public provisions, the demographics of the workforce, and other labour-market conditions, such as the increasing need for highly educated personnel, have all been in flux.
In the liberal, the conservative and the Mediterranean regimes, in particular, we have witnessed a further development of public work-family polices. Existing studies show that developments in childcare and leave arrangements in these regimes reflect the growing importance of policies of the EU in this area (Bleijenbergh, 2006). Since the 1990s, the EU has introduced several policy measures to create a base line of work-family policies in European welfare states (Ackers and El-Sawad, 2006). Since the Social Chapter at Maastricht in 1992, subsequent directives on parental leave and working time have been introduced. In addition, the EU has tried to engender a broad societal debate on the equal division of paid and unpaid work among men and women, and to stimulate women's labour participation. Among other factors, these policies and pressures have contributed to the development of public provisions with regard to work-family reconciliation. For instance, in the UK, the Labour Party has introduced parental leave, paternity leave, and the possibility to negotiate flexible working hours with employers for working parents after signing the social charter. In the early 2000s, the Dutch Government has formalised the right to two days paternity leave, and introduced short and long-term care leave. Slovenia recently introduced 90 days paternity leave for fathers (den Dulk et al, 2005).
Also in Eastern Europe, societal transformations have taken place since 2000. Although there is agreement among researchers that the post-communist countries started from similar institutional backgrounds of centralized and universal social provisions as rights, after years of reform it has become clear that former socialist countries are moving in different directions and have started to combine elements of different regime types. While in many post-communist countries public work-family arrangements are under pressure because of economic crisis, an insecure labour market, and the move towards a more residual-liberal welfare state, there are also countries in which public, universal work-family policies are reinstated in light of declining fertility rates (Rostgaard, 2004).
It can be expected, that in those CEE-countries where state involvement is on the decline, and given the weak business case for employer involvement, employers will not automatically step into compensate the loss of state involvement by providing additional work-family arrangements at the organizational level. New directions for the provision of work-family arrangements, therefore, demand a close study using the perspective of the extended business case, as suggested in this study. Should CEE-country policies be focussed on the maintenance of their former levels of state support? Or should they follow the more liberal example, and leave work-life provisions to the market, hoping that employers will fill the gap which is left by governments? This latter strategy should be looked at carefully. In case governments do step back, and relevant business-case arguments are absent, employees in the CEE-countries might be left with little support for the combination of their professional and private lives. In case state support will decline further, and business case arguments will gain importance, for instance, due to globalisation, employers might become involved in the provision of work-life arrangements. Like in the liberal regime, however, employer involvement induced by business case arguments, is likely to engender new labour market inequalities, for instance along the lines of educational levels. Social partners, however, could use institutional
pressures to stimulate employers to develop childcare policies and leave arrangements. They themselves may exercise pressure by developing policies to preserve a balance between state support and employer involvement.
What factors and forces become dominant, however, remains an empirical question. Future studies, therefore, should investigate the impact of more recent labour market changes and tensions at all analytical levels, such as the impact of the equal-opportunity doctrine imposed by European policies, increasingly affecting new member states, on the one hand, and the pressures arising from increased globalisation, the spread of multinational companies (MNCs) starting subsidiaries in CEE-countries, and related to that, the impact of the growing liberal doctrine putting emphasis on the market and the economic and social business case, on the other.
Childcare and leave arrangements
177
Limitations
Our study had a number of limitations. Employer involvement likely results from a larger number of institutional and organizational factors than discussed in this study. These factors are also likely to interact among each other. In order to understand employer involvement, therefore, more in-depth studies into the interplay between institutional factors is needed.
Our cross-sectional design, however, could not disentangle the interrelationships and interactions between the various institutional dynamics. A more comprehensive view on the ongoing developments with regard to employer involvement calls for a longitudinal design that allows taking more institutional and organizational developments into account.
Moreover, we only used one single indicator of employer involvement, by counting the number of childcare and leave arrangements. Of course, as our study compared a large number of organizations in 19 European countries, it would be impossible to inquire after, let alone compare, the actual content and quality of the work-family arrangements offered. For instance, childcare arrangements offered by employers can vary from a workplace nursery on the one hand, to resource and referral services which assist employees with finding suitable day care facilities on the other. Additional leave arrangements may vary from paid leave to the possibility of a more flexible take-up of statutory leave. Hence, some employers may be offering more comprehensive arrangements than others. Consequently, variation in employer involvement is likely to be even greater than is shown by our data that focused merely on the number of childcare and leave arrangements.
In conclusion, although the use of the present framework of welfare regimes, as developed by Esping-Andersen and others, was a useful starting point for our research, the variations within regimes we found may raise questions about convergence or new divergence within Europe. The impact of EU policies as guidelines for national policies and for growing practices in organizations may also invite reconsideration of this framework, and certainly invites further research on convergence and divergence.
Notes
1. The CRANET survey used the European standard of NACE sector codes which is also used by EUROSTAT.
2. Additionally, MNCs might also be seen as having a significant impact on work-family arrangements, for MNCs may follow ethnocentric, polycentric, etc. strategies. We performed the analysis with the variable if the organization is part of a national local organization or is
a subsidiary of a multinational (available from the authors). Adding this control did not change the results.
References
Ackers, P. (2003), "The work-life balance from the perspective of economic policy actors", Social Policy and Society, Vol. 2 No. 3, pp. 221-9.
Ackers, P. and El-Sawad, A. (2006), "Family-friendly policies and work-life balance", in Redman, T. and Wilkinson, A. (Eds), Contemporary Human Resource Management, Pearson Education, Harlow, pp. 331-55.
Appelbaum, E., Bailey, T., Berg, P. and Kalleberg, A. (2005), "Organizations and the intersection of work and family: a comparative perspective", in Ackroyd, S., Batt, R., Thompson, P. and Tolbert, P.S. (Eds), The Oxford Handbook of Work and Organizations, Oxford University Press, Oxford, pp. 52-73.
Bleijenbergh, I. (2006), "European social policies on combining work and family life: the gap between legislation and practice", in Guichon, A., van den Anker, C.L. and Novikova, I. (Eds), Women's Rights and Entitlements, Palgrave Macmillan, Basingstoke, pp. 62-81.
Blossfeld, H.P. and Drobnic, S. (2001), Careers of Couples in Contemporary Societies: From Male Breadwinner to Dual Earner Families, Oxford University Press, Oxford.
Brewster, C, Hegewisch, A., Mayne, L. and Tregaskis, O. (1994), "Methodology of the Price Waterhouse Cranfield project", in Brewster, C. and Hegewisch, A. (Eds), Policy and Practice in European Human Resource Management, Routledge, London, pp. 230-45.
Campbell, J.L. (2004), Institutional Change and Globalization, Princeton University Press, Princeton, NJ.
Coleman, J.S. (1990), Foundations of Social Theory, Harvard University Press, Cambridge, MA.
den Dulk, L. (2001), Work-family Arrangements in Organisations: A Cross-national Study in The Netherlands, Italy, the United Kingdom and Sweden, Rozenberg, Amsterdam.
den Dulk, L. (2005), "Workplace work-family arrangements: a study and explanatory framework of differences between organizational provisions in different welfare states", in Poelmans, S.A.Y. (Ed.), Work and Family: An International Research Perspective, Lawrence Erlbaum, Mahwah, NJ, pp. 211-38.
den Dulk, L. and van Doorne-Huiskes, A. (2007), "Social policy in Europe: its impact on families and work", in Crompton, R., Lewis, S. and Lyonette, C. (Eds), Women, Men, Work and Family in Europe, Palgrave Macmillan, New York, NY.
den Dulk, L., Peper, P. and van Doorne-Huiskes, A. (2005), "Work and family life in Europe: employment patterns of working parents across welfare states", in Peper, B., van Doorne-Huiskes, A. and den Dulk, L. (Eds), Flexible Working and Organisational Change: The Integration of Work and Personal Life, Edward Elgar, Cheltenham, pp. 13-38.
Dex, S. and Scheibl, F. (1999), "Business performance and family-friendly policies", Journal of General Management, Vol. 24 No. 4, pp. 22-37.
DiMaggio, P.J. and Powell, W.W. (1983), "The iron cage revisited: institutional isomorphism and collective rationality in organizational fields", American Sociological Review, Vol. 48, pp. 147-60.
Esping-Andersen, G. (1990), The Three Worlds of Welfare Capitalism, Princeton University Press, Princeton, NJ.
Esping-Andersen, G. (1999), Social Foundations of Post-industrial Economies, Oxford University Press, Oxford.
Evans, J.M. (2001), "Firms' contribution to the reconciliation between work and family life", Occasional Papers No. 48, Organisation for Economic Co-operation and Development, Paris.
Forth, J., Lissenburgh, S., Callender, C. and Millward, N. (1997), "Family friendly working arrangements in Britain, 1996", Research Report No. 16, Department for Education and Employment, Policy Studies Institute, London.
Goodstein, J.D. (1994), "Institutional pressures and strategic responsiveness: employer involvement in work-family issues", Academy of Management Journal, Vol. 37 No. 2, pp. 350-82.
Hicks, A. and Kenworthy, L. (2003), "Varieties of welfare capitalism", Socio-Economic Review, Vol. 1, pp. 26-7.
Ingram, P. and Simons, T. (1995), "Institutional and resource dependence determinants of responsiveness to work-family issues", Academy of Management Journal, Vol. 38 No. 5, pp. 1466-82.
Jacoby, S.M. (2004), The Embedded Corporation: Corporate Governance and Employment Relations in Japan and the United States, Princeton University Press, Princeton, NJ.
Kocourkova, J. (2002), "Leave arrangements and childcare services in Central Europe: policies and practices before and after the transition", Community, Work and Family, Vol. 5 No. 3, pp. 301-18.
Leitner, S. (2003), "Varieties of familialism: the caring function of the family in comparative perspective", European Societies, Vol. 5 No. 4, pp. 353-75.
Lewis, S. (1999), "Work-family arrangements in Europe", in den Dulk, L., van Doorne-Huiskes, A. and Schippers, J. (Eds), Work-family Arrangements in Europe, Thela Thesis, Amsterdam.
Lewis, S. and Lewis, J. (1996), The Work family Challenge: Rethinking Employment, Sage, London.
Lewis, S. and Smithson, J. (2006), Final Report of the Transitions Project for the EU Framework 5 Funded Study "Gender, Parenthood and the Changing European Workplace", RIHSC, Manchester Metropolitan University, Manchester.
Lyness, K.S. and Brumit Kropf, M. (2005), "The relationships of national gender equality and organizational support with work-family balance: a study of European managers", Human Relations, Vol. 58 No. 1, pp. 33-60.
Meyers, M.K. and Gornick, J.C. (2003), "Public or private responsibility? Early childhood education and care, inequality, and the welfare state", Journal of Comparative Family Studies, Vol. 34 No. 3, pp. 379-412.
OECD (2001), Employment Outlook 2001, Organisation for Economic Co-operation and Development, Paris.
OECD (2007), Babies and Bosses - Reconciling Work and Family Life: A Synthesis of Findings for OECD Countries, Organisation for Economic Co-operation and Development, Paris.
Oliver, C. (1991), "Strategic responses to institutional processes", Academy of Management Review, Vol. 16, pp. 145-79.
Osterman, P. (1995), "Work-family programs and the employment relationship", Administrative Science Quarterly, Vol. 40, pp. 681-700.
Pascall, G. and Manning, N. (2000), "Gender and social policy: comparing welfare states in Central and Eastern Europe and the former Soviet Union", Journal of European Social Policy, Vol. 10 No. 3, pp. 240-66.
Plantenga, J. and Remery, C. (2005), "Reconciliation of work and private life: a comparative review of thirty European countries", EU Expert Group on Gender, Social Inclusion and Employment, Brussels, available at: http://europa.eu.int/comm/employment_social/ gender_equality/docs/2005/reconciliation_report_en.pdf
Childcare and leave arrangements
179
BJM Poelmans, S.A.Y., Chinchilla, N. and Cardona, P. (2003), "The adoption of family-friendly HRM
r r> policies: competing for scarce resources in the labor market", International Journal of
°'Z Manpower, Vol. 24 No. 2, pp. 128-47.
Rabe-Hesketh, S. and Skrondal, A. (2005), Multilevel and Longitudinal Modeling Using Stata, 1st ed., Stata Press, College Station, TX.
Rabe-Hesketh, S., Skrondal, A. and Pickles, A. (2004), "GLLAMM manual", Working Paper Series. 180 Working Paper No. 160, Division of Biostatistics, University of California, Berkeley, CA.
Remery, C, van Doorne-Huiskes, A. and Schippers, J. (2003), "Family-friendly policies in The Netherlands: the tripartite involvement", Personnel Review, Vol. 32 No. 4, pp. 456-73.
Rostgaard, T. (2004), Family Support Policy in Central and Eastern Europe - A Decade and a Half of Transition, UNESCO, Paris.
Scott, W.R. (1995), Institutions and Organizations, Sage, Thousand Oaks, CA.
Szelewa, D. and Polakowski, M.P. (2008), "Who cares? Changing patterns of childcare in Central and Eastern Europe"', Journal of European Social Policy, Vol. 18 No. 2, pp. 115-31.
Tolbert, P.S. and Zucker, L.G. (1996), "Institutionalization of institutional theory", in Clegg, S.R., Hardy, C. and Nord, W.R. (Eds), Handbook of Organization Studies, Sage, London, pp. 175-90.
United Nations Development Programme (2003), Human Development Report 2003: Millennium Development Goals: A Compact among Nations to End Human Poverty, Oxford University Press, Oxford.
Appendix 1
Activity rates
GDI
Table AI.
Activity rates of men and women, unemployment rates 2000, WLB state support (leave and childcare arrangements, scale 2-8) and GDI by country
Country	Men	Women	WLB state support	Unemployment rate	Rank	Value
UK	83	68	3	5	11	0.928
Ireland	80	56	3	4	16	0.923
Germany	79	63	4	7	15	0.924
Belgium	74	56	5	7	8	0.931
France	75	62	6	9	17	0.923
The Netherlands	84	66	4	3	7	0.934
Switzerland	89	74	2	3a	10	0.927
Austria	80	62	3	4	14	0.924
Sweden	80	75	8	6	3	0.940
Denmark	84	76	7	4	9	0.928
Norway	85	77	8	4a	1	0.941
Finland	77	72	8	10	10	0.928
Spain	79	52	4	11	20	0.912
Portugal	79	64	4	4	23	0.892
Greece	77	51	4	11	24	0.886
Bulgaria	66	56	4	16	51	0.794
Czech	79	64	3	9	32	0.857
Estonia	76	65	6	13	38	0.831
Slovenia	72	63	7	7	29	0.879
Note: See Appendix 2, Table All for the calculation of state support
Sources: aFigure from 2001; employment in Europe (2000) and United Nations Development Programme (2003)
"O a
a
10
Maternity leave
Paternity leave
Parental leave
Availability public childcare
Denmark     18 weeks, 100 per cent paid (up to DKR 3,115 p/w)
Finland       17.5 weeks, paid between
100 and 60 per cent of earnings
Sweden Seven weeks pregnancy leave and 60 days of parental leave, paid at 80 per cent of earnings
Norway Nine weeks (part of parental leave), paid between 100 and 80 per cent of earnings (max €590)
France        16 weeks (1st and 2nd child), 26 weeks (3rd child), 100 per cent paid (max €2,432 per month)
Two weeks, paid
90 per cent of earnings
(up to a maximum)
18 days (extension one to 12
days possible when taking
parental leave), payment
between 100 and 60 per cent
Ten days, 80 per cent paid
(up to maximum)
Four weeks fathers' quota out of parental leave, payment between 100 and 80 per cent
Two weeks (three in case of multiple births), three days 100 per cent paid, afterwards up to a maximum
32 weeks per child, per parent, 90 per cent paid (up to max. DKR 3,115 per week) 158 days, payment around 60 per cent (home care leave up to child is three years of age, flat rate payment)
480 days to be shared between parents, 60 days reserved for each parent, first 390 days 80 per cent of earnings after that flat rate
42 or 52 weeks parental leave
per child to be shared between
parents, payment between
100 and 80 per cent
Three years per child, flat rate
payment in case of two +
children
Full coverage
Since 1996, all children under school age are guaranteed a municipal childcare place
Right to childcare for children between one and 12 years of age
Full coverage is a political goals since the 1980s
Full coverage for children older than three; 20 per cent of children under three are enrolled in nursery facilities and 20 per cent are cared for by a registered child minder
{continued)
o P h-
a «:
a> p ' O
tea
O <T> O &
o a
O SU
a- ^
a to
a, g a a
& p
£L fi>* s»
00
&3
Crq cd
3 i—■ rv
cd rjj o
jZS & V
W cd cd
>
a
00
to
Maternity leave
Paternity leave
Parental leave
Availability public childcare
Belgium
Germany
15 weeks (17 multiple births),    Ten days, payment between paid between 82 and 75 per cent 100 and 82 per cent of earnings
14 weeks (18 multiple births), 100 per cent paid
The
Netherlands
None
16 weeks, 100 per cent paid None
Switzerland 16 weeks, 100 per cent paid * None
Austria       16 weeks (or 20 for medical None reasons), 100 per cent paid
UK 52 weeks, six weeks paid at None
90 per cent; 20 weeks flat rate; 26 weeks unpaid
Three months per parent, per child, flat rate payment
Three years (incl. maternity leave) per child, payment is means tested, income-related benefit
13 weeks per parent, unpaid None
Two years, flat rate payment None
30 per cent of 0 to three year olds covered by public provisions; almost all three to six year old in education-based care
In West Germany 3 per cent and in East Germany 37 per cent of young children enrolled in public childcare. Since 1996, federal law entitle children older than three years to public childcare but places are on a part-time basis only Increase of childcare facilities since the 1990s, 22 per cent of 0 to four years old are enrolled in public childcare Limited and large regional differences
Very limited for children under three years of age Very limited, since 1997 National Child Care Act that is mostly targeted at three to four years old
{continued)
Maternity leave
Paternity leave
Parental leave
Availability public childcare
Ireland Spain
Portugal
Greece
Czech Bulgaria
Estonia
Slovenia
26 weeks, 18 weeks paid at 70 per cent
16 weeks (18 in case three or more children), 100 per cent paid
17 weeks, 100 per cent paid
17 weeks, 100 per cent paid (max €42 per day)
None
Two days ( + two days if another town), 100 per cent paid
Five days, 100 per cent paid
Two days, 100 per cent paid
14 weeks per parent, unpaid     Very limited
28 weeks (37 multiple births), None 69 per cent paid (max €25 per day)
19.3 weeks, paid at 90 per cent None of earnings
18 weeks, 100 per cent paid      14 days
21 weeks (105 days), 100 per cent paid
None
Three years per child (incl. maternity leave), unpaid
Three months per parent, unpaid
3.5 month per parent, unpaid
156 weeks, flat rate payment (€121 p/m)
24 months, partly paid (minimum wage)
34 weeks, 100 per cent paid
260 days can be shared between parents, 100 per cent paid
Very limited for under three; full coverage for four to five year old
Target set at coverage of 20 per cent for under three; 90 per cent coverage for five year olds
Very limited for children younger than three; more extensive between three and school age
1 per cent of children between six months and three years of age are enrolled in public care Decline of provisions, enrolment beginning of 2000 is circa 10 per cent One to three year olds; 95 per cent four to six year olds
Very limited for children younger than three, guarantee of provisions for children older than three but insufficient supply
Increase of provisions,
40 per cent one to three year
olds; 65 per cent four to six year
olds
Sources: OECD Family database (2006, December); available at: www.oecd.org/els/social/family/database; Plantenga and Remery (2005) and OECD (2007)
>
a
00
w
&3
Crq cd
3 i—■ cv
cd rjj o
jZS & V
W cd cd
About the authors
Laura den Dulk is an Assistant Professor of Work, Organization and Management at the Department of Public Administration, Erasmus University Rotterdam, The Netherlands. Her main area of expertise is cross-national research regarding work-life policies in organizations in different welfare state regimes. Current research interests include the attitudes, opinions and behaviour of top-managers towards work/life policies and the social quality in European workplaces. She participated in various EC research projects: "Quality of life in a changing Europe" (Quality) and "Gender, parenthood and the changing european workplace: young adults negotiating the work-family boundary" (transitions). Laura den Dulk is the corresponding author and can be contacted at: dendulk@fsw.eur.nl
Pascale Peters is an Assistant Professor at the Department of Business Administration, Radboud University Nijmegen, Institute for Management Research. In 2000, she received her PhD at the University of Tilburg, Faculty of Social Sciences. Her current research interests include new ways to work and work-life balance issues, in particular flexible work arrangements like home-based telework. The main focus in her research is on managers' telework attitudes; the telework adoption process; telework management and telework outcomes. She has published in a large number of international and national journals and books. In 2007, she co-edited a book entitled Competing Claims in Work and Family Life which was published by Edward Elgar. She co-supervised a PhD-project on the adoption and use of flexible benefit plans in The Netherlands. Currently, she co-supervises a PhD-project on the influence of work-life arrangements and work-life culture on Dutch female doctors' career ambition and career motivation.
Erik Poutsma is an Associate Professor of Labour Relations in the Department of Business Administration of Nijmegen School of Management, Radboud University of Nijmegen. He studied sociology (MA) at Rijksuniversiteit Groningen and received his PhD in Business Administration at the University of Nijmegen. He has many years of experience of conducting research in the areas of technological change, entrepreneurship, HRM and employee participation. He is partner of the CRANET. His main research interest is international comparative research of the domains HRM and employee involvement.
Paul E.M. Ligthart is an Assistant Professor of Strategy in the Department of Business Administration of Nijmegen School of Management, Radboud University of Nijmegen. He studied (experimental) social and organizational psychology (MA) at Rijksuniversiteit Groningen and received his PhD in theoretical sociology at Interuniversity Center of Social Science Theory and Methodology also at the Rijksuniversiteit Groningen. His main research interest is in the strategy effects of configurations of HRM-policies and practices, and configurations of innovation activities in the manufacturing industry. He is a member of the international research networks of HRM (CRANET) and the European Manufacturing Survey.
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints