Shrinking welfare states? Comparing maternity leave benefits and child care programs in European Union and North American welfare states, 1985–2000

Abstract

This paper tests whether changes in program design, coverage, and government funding of maternity and parental leave and child care and early childhood education (ECE) programs can be observed over the past fifteen years in European and North American countries. Analysis of cross-national and cross- time data reveals, first, that reports of the welfare state’s demise do not hold true in these areas. In most of the countries surveyed, the number of places for children in child care and ECE expanded, as did the duration of maternity/parental leaves (although the benefit levels declined in some countries). Second, the data reveal some policy convergence by the late 1990s, although not as much as functionalist theories would expect. Statistical analysis of factors to account for the continued though decreasing divergence reveals that traditional theories of welfare state variation hold greater explanatory power in the late 1980s than in the late 1990s. Demand factors continue to play a role in maternity leave duration but while a combination of demand, political and spending factors help to account for variation across countries in the 1980s, by the 1990s none were relevant in explaining variation in three of our policy fields. These results suggest that the scope of substantive policy provision is even more dependent on domestic policy choices.

Keywords:

• Child care
• early childhood education
• Europe
• maternity/parental leave
• North America
• welfare states

Acknowledgments

An earlier version of this paper was presented at the 13th International Conference of Europeanists, 14–16 March 2002, Chicago, Illinois, and circulated at the Annual Meeting of the Research Committee on Poverty, Social Welfare, and Social Policy (RC19) of the International Sociological Association, Toronto, Ontario, 21–24 August 2003. The authors acknowledge the helpful comments of Steven Bernstein, Yolande Cohen, Janet Gornick, Grace Skogstad, and Joseph Wong.

Notes

Although many European countries have a long tradition of social policy provision in these areas – some countries have provided maternity leave benefits, for example, since the late nineteenth and early twentieth centuries – what is new in many of the countries surveyed is the expansiveness and generosity of these benefits.

Further time-series analysis would have to be conducted in order to determine the significance of the within-country changes over time.

The impact of this latter factor is difficult to test statistically because of the small numbers involved.

Another important factor could be the type of electoral system in place. Countries with a single member plurality (SMP) system are generally associated with lower levels of female representation in government and have heavier constituent workloads than proportional or multi-member systems. For this reason we would expect that countries with SMP would have lower levels of service provision than countries with proportional representation systems. However, only three countries (UK, USA, and Canada) use unmodified SMP and thus the number of cases is too small to analyze statistically.

And, indeed, regression analysis revealed that EU membership did not make a difference.

Results of correlations are available from the authors.

Deleting social spending from the model produces similar results for the models. In 1985 this has the most effect on dependent variables maternity leave and early childhood education, variables for which spending was a significant predictor. Removing spending produces an adjusted R2 of 0.260 for maternity leave and −0.058 for ECE. None of the other variables is significant in this amended model. For the later model, the adjusted R2 are slightly elevated for maternity leave and benefits, to 0.780 and 0.632 respectively, identical for child care and decreased for ECE.

For excellent summaries of debates and arguments about the impact of globalization on national politics, see Garrett (1998) and Schwartz (2001)

Ranking based on an average of these four indicators across sixty-two countries is as follows: Ireland (1); Sweden (2); Netherlands (3); Denmark (7); UK (10); Finland (11); Austria (16); France (12); Portugal (15); Canada (17); Spain (18); Germany (22); Norway (24); Italy (36); Greece (41); USA (50) (it does not rank Belgium or Luxembourg).