IJE Advance Access originally published online on May 8, 2008
International Journal of Epidemiology 2008 37(5):1105-1108; doi:10.1093/ije/dyn078
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Commentary: Politics and public health—some conceptual considerations concerning welfare state characteristics and public health outcomes
Professor of Health Equity Studies and Assistant Director, Centre for Health Equity Studies, Stockholm University/Karolinska Institutet, SE-106 91 Stockholm, Sweden.
E-mail: olle.lundberg{at}chess.su.se
Accepted 2 April 2008
Espelt et al. have published a paper1 on differences between European welfare states and how these differences are linked to health inequalities among the older part of the population. Although many comparative studies of international variations in health inequalities have drawn conclusions about the pros and cons of different welfare state set-ups, the issue has not been properly studied. Partly, this could be due to the conceptual and methodological problems involved when one attempts to relate international variations in complex welfare state structures on the one hand with mortality, ill health or health inequalities on the other. And because of the complexity of the task, the analytical choices made when designing a study become even more crucial than in regular individual-level epidemiological studies. Examples of such choices include what kind of welfare state characteristics we believe to be of importance for public health outcomes; how data on these characteristics are handled and what kinds of public health outcomes are likely to be affected. I believe that the choices made by Espelt et al. need to be examined, since they have important consequences for our understanding of the links between welfare state characteristics and public health outcomes.
A fundamental question is of course what it is about welfare states that affect the health and longevity among their populations and that also vary systematically across different types of welfare state. Ultimately, I would argue, it is the resources available to people that will be of importance for the levels of and inequalities in health in a country.2 These resources are generated within the family, in the market and also through the welfare state. Welfare state institutions will thereby contribute to people's resources, either directly through transfers and services or indirectly through policies that affect people's possibilities to generate resources in the market. The degree to which welfare state institutions do so and the extent to which this in turn is linked to health and health inequalities is the key issue, therefore. Consequently, it is features such as the coverage and generosity of cash transfer programme like unemployment insurance, sickness insurance, family support or pensions that should be in focus if we are to find out what it is about welfare states that are important to people's health. Or, for that matter, the availability and quality of services provided.
And as welfare state research has demonstrated, there are large variations in the way social insurance programmes are organized across different welfare states, and these differences are also related to differences in outcomes on the individual level, as reflected most clearly in cross-national variations in poverty rates.3,4 Hence, there seem to be a good case for assuming a relation also with public health outcomes, although such links are certainly more complex than links between transfer programme generosity and poverty.
But rather than focussing on differences in the amount of resources that welfare states provide to their citizens, Espelt et al. looks at politics. While political power and political forces are of key interest for an analysis of how differences among welfare states are established, there is certainly no 1:1 relationship between the two. Instead of clarifying things, there is a clear danger that an analysis that looks at welfare states only in terms of what parties have been in government will introduce a serious misclassification problem to an already complicated issue. Basically, we then assume that social democrats or Christian democrats will set up institutions and programmes that generate the same kind and amount of resources to their citizens irrespective of the historical, political and economic context. The assumption is that it does not matter if the comparison is Sweden in the 1950s or France in the 1990s; nor does it matter what sort of parliamentary basis for the government exists—political parties with a certain label are basically assumed to pursue politics that generate the same resources for citizens. This appears very unlikely, to put it mildly.
A further problem is that Espelt et al. have clustered welfare states on the basis of the amount of time after 1950 that different political parties have been in the government. This has resulted in three groups of countries; a social democratic (Sweden, Denmark and Austria), a Christian democratic (The Netherlands, Germany, France and Italy) and a late democratic (Portugal and Spain). Indeed, there is a strong tradition in welfare state research where welfare states have been grouped into a smaller number of regimes or types according to some underlying basic principles. The main example here is Esping-Andersen's seminal work where he identifies three regimes, namely the Liberal, the Conservative-Corporatist and the social-democratic regime.5 This categorization is based on the relative importance of state, market and family for people's welfare, and in particular the degree to which people are dependent on the market to maintain their livelihood.
But this tradition has mainly been focussed on the principles behind welfare states rather than the potential consequences for their citizens, and in the latter case welfare state typologies have clear shortcomings as analytical tools. One is that they are usually one-dimensional—in the case of Esping Anderson focussing on general aspects of cash-transfer systems. While there are certainly general principles that differ across countries, it is not necessarily the case that pensions, sickness insurance, unemployment insurance and family policies are all organized according to the same principles within each cluster we identify, or even within the same country. Furthermore, the principles adopted in other areas of the welfare state, such as social care services or health care, are not necessarily the same as those adopted for transfer programmes, although more generous transfer systems tend to be mirrored by more generous services.6 Another problem is that important cross-national differences may be averaged out within each cluster of countries.
In the case of Espelt et al. the important assumption behind the clustering is that parties with the same label pursue the same kind of policies across time and space. Again, this is obviously not the case in real life—the policies implemented by social democrats in Sweden changed quite a lot from the early 1970s to the late 1990s, just to take one example. And more importantly, governments in Sweden and Austria have set up radically different welfare state institutions, despite the fact that they have included social democrats to a larger extent than in many other countries. This, in turn, can be related to differences in political and historical contexts that are larger and more complex than the similarities in terms of the number of years that social democratic parties have been in the government would suggest.7 Consequently, in most welfare state typologies Austria is an ideal type of the corporatist welfare state relying heavily on occupation based solutions, whereas Sweden is the ideal type of the Scandinavian or Nordic welfare state with an emphasis on universal policies. To lump these two distinctly different countries together is not especially helpful if we want to understand how different policy solutions are translated into public health outcomes.
So while the country clusters may be helpful for descriptive purposes, they are much less useful if we really want to open the black box and analyse what aspects of the welfare state that are of importance. Especially, if we are interested in linking welfare state characteristics to public health outcomes it will be much more fruitful to study aspects like coverage and generosity in specific programmes and how these co-vary with public health outcomes rather than to merely relate country-cluster averages to each other. In doing that we must go beyond simple labels and welfare state clusters and try to study how the design of specific programmes are related to important outcomes. In other words, I believe we need to look at what is done rather than what it is called. If politics matter it is the institutions, the programmes and the resources they provide to citizens that matter, not the label attached to the government parties. This is not to say that party politics are of no interest, but if researchers want to contribute to improved public health policies we should be able to tell what kind of policies that work or not, not what kind of parties’ people should vote into the government.
Another central issue is considering the expected effects on public health and health inequalities that should flow from different welfare state arrangements. Espelt et al. choose relative inequalities as their main outcome, without much motivation for why and how relative inequalities in health between social classes should be expected to vary in size across clusters of welfare states. While they do report such differences, the finding that southern European countries have larger relative inequalities than northern European countries does not seem to be in line with previous findings.8,9 The expected outcomes of welfare state characteristics should be considered carefully, since the conclusions regarding the performance in terms of public health in different welfare state regimes can depend strongly on the outcome measure chosen. This is illuminated in Table 1.
|
First, relative inequalities seem to be very similar across these countries, while absolute inequalities are not (although it should be remembered that data reflects the situation in the 1980s). But more importantly, perhaps, is that the consequences of inequalities in terms of lives lost among manual workers are profoundly different, and that the relative inequalities actually obscure this variation across countries. The prime example of this is Sweden and Ireland where relative inequalities are almost identical. Yet, the mortality level among manual workers is much lower in Sweden. As a tool for evaluating welfare policies measures of relative inequalities are simply not adequate,12 since Swedish manual workers apparently have better possibilities to avoid premature mortality than their Irish brothers. Whether this is due to welfare state policies or not is of course another question, but the main point here is that we cannot draw any valid conclusions regarding welfare state performance based on relative inequalities only. The reason for this is that relative inequalities are dependent on the situation for more privileged groups as well. The fact that Swedish non-manual employees have extremely low mortality must be considered a success and not a failure, even if that contributes to relative inequalities on par with that in Ireland.
While the absolute mortality levels among workers is not a measure of inequalities per se, cross-national variations in this outcome definitively reflect cross-national variations in the degree to which the less-privileged groups in a society have a possibility to achieve a decent living. That, in turn, is likely to be related to how well the welfare state is able to buffer and compensate for inequalities generated in the market. The ability of the welfare state to buffer and compensate for market generated inequities is a result of welfare state institutions and programmes and the principles and characteristics of these. In the Swedish case, universal rather than targeted and means-tested social policies have been a key feature, including as one central element, that social insurances and services are designed for the population at large, and not only for the poorest. That the middle and upper classes thereby also are included are often argued to be a central feature behind the high degree of generosity in universal systems, which in turn means that the poorer segments of society usually fare better with universal systems.3,5 But also middle classes benefit from universal social protection, and if universal policies contribute to better health and lower mortality among lower and middle classes alike, such policies will have a huge impact on public health in general but not necessarily on relative health inequalities.
While I have been raising some critical remarks, I believe that the paper by Espelt et al. is important since it raises the question of the importance of macro-level policies for public health outcomes. However, if we really want to know what kind of policies and programmes work in improving population health and reducing health inequalities, we will have to focus on the characteristics of the specific welfare state institutional mechanisms that link politics to public health.
 |
References |
|---|
 Top References |
|---|
1 Espelt A, Borrell C, Rodriguez-Sanz M. Inequalities in health by social class dimensions in European countries of different political traditions. In: Int J Epidemiol. (2008) 37:1095–105.
2 Fritzell J, Lundberg O. Health, inequalities, welfare and resources. In: Health Inequalities and Welfare Resources.—Fritzell J, Lundberg O, eds. (2007) Bristol: Policy Press. 1–18.
3 Korpi W, Palme J. The paradox of redistribution and strategies of equality: welfare state institutions, inequality and poverty in the Western countries. ASR (1998) 63:661–87.[CrossRef]
4 Ferrarini T. States and Labour Markets. Institutions, Causes and Consequences of Family Policy in Post-war Welfare States. (2006) Cheltenham: Edward Elgar.
5 Esping-Andersen G. The Three Worlds of Welfare Capitalism. (1990) Cambridge: Polity Press.
6 Anttonen A. Empowering social policy: the role of social care services in modern welfare states. In: Social Policy and Economic Development in the Nordic Countries.—Kangas O, Palme J, eds. Houndmills, Basingstoke: Palgrave Macmillan. 88–117.
7 Esping-Andersen G, Korpi W. Social policy and class politics in post-war capitalism: Scandinavia, Austria and Germany. In: Order and Conflict in Contemporary Capitalism. Studies in the Political Economy of Western European Nations.—Goldthorpe JH, ed. (1984) Oxford: Clarendon Press. 179–208.
8 Mackenbach J, Kunst A, Cavelaars A, Groenhof F, Guerts J, and the EU Working Group on Socioeconomic Inequalities in Health. Socio-Economic inequalities in morbidity and mortality in Western Europe. Lancet (1997) 349:1655–59.[CrossRef][Web of Science][Medline]
9 Dahl E, Fritzell J, Lahelma E, Martikainen P, Kunst A, Mackenbach JP. Welfare state regimes and health inequalities. In: Social Inequalities in Health. New Evidence and Policy Implications.—Siegrist J, Marmot M, eds. (2006) Oxford: Oxford University Press. 193–222.
10 Fritzell J, Lundberg O. Fighting inequalities in health and income: one important road to welfare and social development. In: Social Policy and Economic Development in the Nordic Countries.—Kangas O, Palme J, eds. Houndmills, Basingstoke: Palgrave Macmillan. 164–85.
11 Kunst A, Groenhof F, Mackenbach J, and the EU Working Group on Socioeconomic Inequalities in Health. Mortality by occupational class among men 30-64 years in 11 European countries. Soc Sci Med (1998) 46:1459–76.[CrossRef][Web of Science][Medline]
12 Vågerö D, Erikson R. Socioeconomic inequalities in morbidity and mortality in Western Europe. Lancet (1997) 350:516.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Int. J. Epidemiol.:
- Inequalities in health by social class dimensions in European countries of different political traditions
- Albert Espelt, Carme Borrell, Maica Rodríguez-Sanz, Carles Muntaner, M Isabel Pasarín, Joan Benach, Maartje Schaap, Anton E Kunst, and Vicente Navarro
Int. J. Epidemiol. 2008 37: 1095-1105.[Abstract] [FREE Full Text]
This article has been cited by other articles:
|
|
 |
|
  J. Beckfield and N. Krieger Epi + demos + cracy: Linking Political Systems and Priorities to the Magnitude of Health Inequities--Evidence, Gaps, and a Research Agenda Epidemiol. Rev., November 1, 2009; 31(1): 152 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A E Sanders, G D Slade, M T John, J G Steele, A L Suominen-Taipale, S Lahti, N M Nuttall, and P F. Allen A cross-national comparison of income gradients in oral health quality of life in four welfare states: application of the Korpi and Palme typology J Epidemiol Community Health, July 1, 2009; 63(7): 569 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Muntaner, C. Borrell, A. Espelt, M. RodrIguez-sanz, M I. PasarIn, J Benach, and V. Navarro Politics or policies vs politics and policies: a comment on Lundberg Int. J. Epidemiol., June 2, 2009; (2009) dyp220v1. [Full Text] [PDF]
|
|
|
|
|
|
A Espelt, C Borrell, M Rodriguez-Sanz, C Muntaner, M. Pasarin, J Benach, M Schaap, A. Kunst, and V Navarro Answer to the commentary: Politics and public health--some conceptual considerations concerning welfare state characteristics and public health outcomes Int. J. Epidemiol., January 8, 2009; (2009) dyn350v1. [Full Text] [PDF]
|
|
|
|
|
|
O Lundberg Politics and public health--some conceptual considerations concerning welfare state characteristics and public health outcomes Int. J. Epidemiol., January 6, 2009; (2009) dyn351v1. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||