Early retirement differences between Denmark and The Netherlands

Abstract

The goal of this study is to give a comprehensive empirical account of push and pull factors, situated at various societal levels, which may influence people's early retirement. Factors in two contrasting European countries have been analysed and compared: Denmark, where the rate of early exit/retirement is relatively low, and The Netherlands, where this rate is relatively high. Our analysis, looking into possible explanations for the difference in rates, includes factors that operate at the macro or institutional level of the labour market and the social security system, at the meso level of firms and companies, and at the micro level of individual workers. In addition, we not only look into the effects of structural, objective factors, which impacts are usually analysed from an incentive perspective, but also into the effects of cultural factors, i.e. of preferences, attitudes and values. Cultural factors related to early exit have not been the object of much international comparison thus far, although their possible significance is readily acknowledged. The differences in retirement behaviour in Denmark and The Netherlands are explained by the fact that Danish workers, at all three levels, are less experiencing structural and cultural factors pushing them out of the labour market, as well as experiencing factors pulling them into retirement, compared to workers in The Netherlands.

Introduction

An important problem of European welfare states is that, for some decades, many of them are confronted with a decreasing trend in the labour participation of older workers (e.g., Blöndal and Scarpetta, 1998, OECD, 2002). Partly, this is the result of explicit policies of the economically cumbersome late 1970s and 1980s when governments sought to promote early exit of older workers in the face of high unemployment levels. However, with the upswing of the economy in the 1990s, but most importantly as a reaction to the ageing of their populations and a decline in fertility rates, many European countries now try to stop and re-direct this trend. It is even the case that the Council of Europe has set specific targets, which at present are only met by Sweden, Denmark and the UK.¹

The fact that participation and retirement rates among older workers differ rather strongly between European countries suggests that macro push and pull factors, like differences in countries' labour market structures and opportunities related to social security systems, influence workers' exit decisions. This is what has actually been found in several large-scale, cross-national comparative studies among countries of the OECD and EU (Blöndal and Scarpetta, 1998, Gruber and Wise, 1999, Vlasblom and Nekkers, 2001, EU, 2003b, OECD, 2003b, Gruber and Wise, 2004).²

However, such comparative econometric studies leave unexplained quite a large part of the variation in countries' exit rates (OECD, 2003b, Esser, 2005). This is not surprising if one assumes that other factors play a significant role too, which may ‘push’ workers out of the labour market, or ‘pull’ them into retirement. Early exit may not merely be the result of older workers being fired in dire times, or of workers' calculations of possible gains from social benefits. They may also depend on employers' attitudes and policies towards older workers, on workers' health status, on their job satisfaction and work ethic, their social obligations, preferences for retirement age, etc., to name just a few. This suggests that, due to their confinement to macro institutional and structural factors only, most comparative econometric studies only offer a limited understanding of national variations in early exit among older workers.

In this article we aim at contributing to a broader understanding by comparing countries on a broad range of factors. In our analysis we will incorporate a series of push and pull factors, which may operate at the macro, institutional level regarding labour market and social security, at the meso level of firms and companies, and at the micro level of individual workers. In addition, we will not only look into the effects of structural, objective factors, which impacts are usually analysed from an incentive perspective, but also into the effects of cultural factors, i.e. of preferences, attitudes and values. Cultural factors related to early exit have not been the object of international comparison thus far, although their possible significance is readily acknowledged (EU, 2003a, Gruber and Wise, 2004).

Our broader approach comes at the cost of the number of countries we can compare. We focus on The Netherlands and Denmark, which suit our purpose well, because they vary quite remarkably in labour participation and early exit among older workers, but equal each other in an overall socio-economic sense. Both countries are small, Western-European, well-developed welfare states with mainly a social-democratic character, the differences in labour market structures are small since both countries have open, service-based economies, and they share a common path of having come out the recession of the 1980s and 1990s with a period of strong job growth, which resulted in low levels of unemployment (see also, Green-Pedersen, 2002, Van Oorschot and Abrahamson, 2003).

Our aim is not to assess, in any statistical sense, the relative significance of the various factors analysed in explaining differences in labour participation and early exit between both countries. It cannot be, since we only compare two cases. What the comparison will lead to, however, is a comprehensive picture of the typical constellations of factors within both countries, on the basis of which the differences in early exit patterns may be understood.

We will, firstly, describe and compare both countries' patterns in labour market participation and early exit among older workers. Secondly, we will compare both countries on a range of push and pull factors that are usually distinguished in the early retirement literature, and see how such factors can or cannot contribute to our understanding of differences in work and exit patterns.