The role of income inequality and social policies on income-related health inequalities in Europe

Several reviews [13‐15] address the influence of income inequality on population health. Even though the conclusions are mixed and only partly suggest a negative effect of income inequality on health, the authors acknowledged the relevance of income inequality for health inequalities [15]. Income inequality affects the average population health via two possible mechanisms. First, there is the neo-material perspective that suggests that the unequal distribution of income leads to underinvestment in human, physical, or cultural capital, as well as in the educational system and medical services [5, 16]. Privileged groups within a society are able to use privately managed services, and therefore, are no longer interested in maintaining public services. If the quality of public service provision is poor, households act rationally by opting out and investing in the private alternatives available, which further undermines the financing of public services [17]. Lower socio-economic status groups are more affected, since they are more likely to be dependent on public services and infrastructure, e.g., schools, transportation, and medical services. Higher income inequality would lead to a widening health gap between the people who have little income and who rely on public services, and those who do not.

The second mechanism is psychosocial: everyone in society is subject to social comparison; people look up and down the social ladder and evaluate their social status. Seeing that others are better off than oneself leads to increased stress levels, which eventually could damage mental and physical health [13]. As a consequence, all people of all income levels are affected by health inequalities—independent of absolute poverty. The observed income gradient in health often is interpreted as evidence of this mechanism [9, 18] since it describes how health worsens with every step down the social ladder.

Subramanian and Kawachi [19] studied the effect of income inequality on the subjective health of different population groups in 50 US states by applying a two-level binomial hierarchical mixed model. They analysed whether income inequality at the state level affects the health of different socio-economic groups differently. Amongst other socio-economic factors, they looked at equivalised household income. They did not find a varying effect of state income inequality on different socio-economic groups, such as e.g. the five income groups they looked at, but an equally negative effect of income inequality on the subjective health of all socio-economic groups. They concluded that there is no association between income inequality and socio-economic health inequalities. On the other hand, in a previous study [20], which examined the subjective health of three income groups (low, middle, and high) in 39 US states, they found a cross-level interaction effect for income inequality and individual income on health.

An analysis of the relationship between wealth and health in 16 countries using the SHARE data is presented in [21]. Across all countries, the authors found a positive and significant effect of wealth on health (net of income), but not for income on health (net of wealth). Health was measured using a Physical Health Index based on 41 items that asked for physical limitations and various illness symptoms. By applying Hierarchical Linear Models, they found a significant cross-level interaction between income inequality and wealth, which supports the thesis that income inequality affects the relationship between wealth and health. However, generalizations should be drawn carefully, since their results were influenced by the special case of the United States. Once the US was excluded from the analyses, the effect of income inequality on the relationship between wealth and health was no longer statistically significant.

Whereas income inequality represents the unequal distribution of earnings, social policies reflect the attempt to redistribute earnings through taxes and social security contributions. These policies aim at increasing individual resources not only through financial benefits but also via social services, e.g., public education, public health, and social security expenditures (social insurance and social assistance). Social policies alleviate the tight budget constraints that impact the lower income strata. Furthermore, people with lower income do not only benefit from targeted social assistance, but also from public investments in general, such as the availability of primary care facilities or public transportation, which are health beneficial [22, 23].

Two possible mechanisms help to determine the relationship between social policies and health inequalities: first, social policies affect health inequalities indirectly by reducing the harmful effects of income inequality on health inequalities; and second, the provision and availability of public services directly promotes health. Health inequalities are reduced, since the lower income group benefits especially from public services. While the first mechanism follows a psychosocial approach, the second follows a neo-material approach.

When applying the various interpretations and operationalisations of welfare state policies, the evidence shows that generous social policies are positively related to population health and reduce health inequalities (for reviews see [2, 3, 24]). Many studies have analysed the relation between the welfare state and population health, mostly by applying a regime approach as a social determinant [2, 3]. In a review article [3], of 73 empirical and comparative studies that discussed the role of politics on population health and health inequalities, the authors specified 31 studies that focused on welfare state generosity as a social determinant. More than half of the evaluated papers in this category found a positive association between welfare state generosity and population health or lower health inequalities. In another review [2], the articles are divided into the Regime approach, which covered welfare state regime typologies; the Institutional approach, which studied policy programs; and the Expenditure approach, which analysed the relationship between social or health spending and average health. Even though the authors could not find results for the Regime approach as consistent as did [3], they found a positive association between generous policies (the Institutional approach) and health in general. Five articles were subsumed under the Institutional approach, which analysed health inequalities. Of these five, only one study found a reduction in income-related health inequalities, while the four other studies did not find that the benefits and social policies had any positive impact on reducing health inequalities. The studies that followed the Expenditure approach found that both social and health spending were associated with smaller socio-economic health inequalities.

In an early review of mostly descriptive studies on socio-economic health inequalities [25], it was found that the Nordic countries—characterised by a Social-Democratic welfare state model that includes higher levels of social benefits and services—did not have lower income-related inequalities in self-rated health. Using data from the European Social Survey 2002 and 2004, another study [26] found that the smallest income-related inequalities in self-rated health occurred not in the Nordic countries, but rather in the Bismarckian countries (e.g., Germany, France, Austria, or Belgium).

The few review studies that exist show that there is little research regarding the impact of welfare institutions or social spending on health inequalities. Only around a third of the studies (9 of the 31) that followed the welfare state approach in the literature review of Muntaner et al. [3] discussed socio-economic inequalities in health. Notably, in the review of Bergqvist et al. [2], half of the reviewed articles (28 of the 54) studied socio-economic health inequalities. Nevertheless, in both articles the authors argued for future research with a focus on the relationship of social spending and the health of different socio-economic groups (see also [27]).

When looking at income inequality and its implications for health inequalities, Wilkinson [1] claimed that income inequality negatively affects everyone’s health. This claim seems to suggest that the average population health could worsen, but this reduction would not impact health inequality either positively or negatively. It overlooks the fact that people in the lower income strata are especially vulnerable because they have few buffers to protect them. Income inequality especially impacts the health of the lower income strata, which leads to increasing differences in the health of different socio-economic groups. Thus, I hypothesise that income inequality increases health inequalities, regardless of whether the mechanism is via disintegration or the disinvestment in public services.

Social policies are related to an increase in average population health [3]. In sum, improvements of the health of lower income groups outweigh the redistributive burden for higher income groups (e.g., via higher taxes) [28]. Here, generous social policies are indicated by the extent of social spending. Social spending serves as a proxy for the availability and quality of public welfare. The higher the social spending, the better are public welfare services and institutions [22]. The more that is redistributed from overall economic performance (either in the form of direct benefits or in investments in public spheres, such as education or health care), the greater the benefit for lower income groups. Therefore, I hypothesise that generous social policies reduce health inequalities.

Furthermore, I assume that income inequality and social policies have additive effects on health inequalities, but they do not influence each other’s impact on health inequalities.

When running the model using the health dummy ‘less than good health’ as the dependent variable, 23 out of 42 countries displayed significant income-related health inequalities (i.e., the effect on health of being in the lowest income quartile compared to the highest income quartile was significant, p < .05). Countries with non-significant findings were scattered across Europe; there was no cluster found according to specific regions such as, e.g., Scandinavia or Eastern Europe. The highest inequalities are found in Germany: the probability of having less than good health is around 26 percentage points higher for respondents in the lowest compared to the highest income quartile. The lowest significant effect is found in Greece: the probability of having poor health in the lowest income quartile is only seven percentage points higher. The two measures for health inequalities and the significance level of the effect of income on health are found in the Appendix B of the present study.

In accordance with previous research, I confirmed the health gradient in income for both the health dummy and the original variable of subjective health for most countries. Not only did the weakest income group assess their health worse than the highest income group, but also the groups in between fell into a similar pattern: the lowest compared to the highest income quartile was the worst off, but the second income quartile was still more disadvantaged than the third quartile is, when compared to the highest quartile.

Table 1 presents the results for the first indicator of health inequalities, which was based on the health dummy variable. There is some support for the hypothesis of a positive association (.34) of income inequality and health inequalities: a higher Gini index is related to higher health inequalities, although not at a conventionally significant level (Table 1, Model 1). When the other two macro determinants were introduced (Model 4, 5 and 7), the Gini index gained significance throughout all model specifications.

The relation between social protection expenditures and health inequalities could not be confirmed. As expected, SPE and health inequalities are negatively related (−.25), but the relation does not reach significance (Table 1, Model 2). When running the analysis in which both GDP p.c. (logged) and SPE were introduced as macro determinants (Model 6), the standardised regression coefficient of SPE on health inequalities was heavily reduced, which indicated an importance of GDP over SPE.

GDP p.c. (logged) had a negative effect on health inequalities meaning that economic performance reduced income-related health inequalities (Table 1, Model 3). The effect of economic performance on health inequalities was linked to income inequality (Model 5). While controlling for the Gini index, the standardised regression coefficient of GDP p.c. (logged) on health inequalities increased and was significant at a higher level. Still, income inequality was the dominating explanatory effect for health inequalities: models including the Gini index showed the highest adjusted R². Furthermore, the Gini index appeared as the highest standardised regression coefficient compared to the log of GDP p.c. and SPE.

When income-related health inequalities were analysed based on the index of dissimilarity as the dependent variable in the model, the results were similar (Table 2). Running models with each macro determinant separately, the direction and size of the coefficients were found to be very similar to those in the models discussed above. The Gini index was positively related to health inequalities (.39, p < .05), i.e., higher income inequality was linked to higher health inequalities (Table 2, Model 1). Fig. 2 illustrates the relation between health inequalities and income inequalities in 42 European countries.

Also, higher social protection expenditures were related to reduced health inequalities (−.25, n.s.) (Table 2, Model 2). The non-significant findings between SPE and health inequalities are not due to the small number of cases, namely 42 countries, as the correlation between SPE and average population health was significant (p < .001). When compared to the other indicator of health inequalities discussed above, GDP p.c. (logged) was not significantly correlated with health inequalities this time (Table 2, Model 3). Again, income inequality was the dominating explanatory effect for health inequalities.

When a sample is small, such as 42 countries in this case, single data points could be critical for estimating the regression and interpreting the effects of the results [54]. When operationalising health inequalities, using the dummy variable of poor health, the case of Germany stands out, since it has the highest income-related health inequalities, but ranges in the middle of the three macro determinants. Previously, lower or medium health inequalities were found in Germany [26, 55]. Regression diagnostics showed a noticeable overall influence (Cook’s Distance, DFFITS) of the data point of Germany due to large residuals. Estimating the analyses without the exceptional case of Germany gives different, i.e., more significant, results: the effect of social protection expenditures on health inequalities increases and becomes significant at the 5 % level. Similar modifications can be found for the effect of GDP (logged) on health inequalities. The Gini index is not significant.

Nevertheless, the analysis of 41 countries, excluding the outlying case of Germany, supports the conclusions of the previous analyses. Even though income inequality as a single determinant no longer plays a significant role in health inequalities, the pattern is again very similar to the analyses of all 42 countries. However, it is notable that GDP (logged) seems to be the most important determinant for the association to income-related health inequalities, contrary to the findings based on all 42 countries.

Implementing the index of dissimilarity as the dependent variable, regression diagnostics showed that Moldova might have an overall influence on the regressions, but a regression analyses that excluded the case of Moldova did not offer any different insights than running the analyses with all 42 countries.

With respect to future studies, the limitations of this study should be discussed. In 2008, the European Values Study covered the whole geographical area of Europe. Although the EVS represents a unique dataset that integrates various European societies, it may include field work that varies in quality across different countries.

For macro-comparative analyses, low numbers of units of analysis are typical [3]. In this case, the number of countries analysed (42) was an inevitable constraint that should be kept in mind when interpreting the results. Usually, to study people nested in countries, the typical approach is to use simultaneous multilevel analysis; instead, to gain more detailed information on single countries, I used a two-step approach—I extracted country-specific effects of household income on subjective health from the micro level at the first step, and subsequently introduced them as dependent variable at the macro level in the second step. This led to the finding of the outlying case of Germany: high health inequalities are combined with a medium level of Gini index, social expenditures, and GDP, as well as medium subjective health at the mean (see Appendix A and Appendix B). Future research could show if this is a specific finding and hence an artefact of the EVS data, or whether income-related health inequalities did indeed increase compared to findings based on earlier data.

Since little research has used a comparative approach to focus on inequalities in health [51], an agreement on the best indicator for socio-economic health inequalities does not yet exist. Subjective health includes both the physical and mental aspects of health. Even though it is often criticised because it is based on individual perceptions, subjective health is widely used in research on population health as well as health inequalities [57]. Since this present study is based on within-country income-related health inequalities, cross-national differences in response styles of self-assessed health [58] are negligible. The question whether socio-economic factors such as income influence respondents’ self-assessment of health, which would bias the estimation of health inequalities, is not solved yet. Jürges [59] finds that response behaviour varies according to socio-economic groups. On the other hand, Van Doorslaer and Gerdtham [60] conclude that income-related health inequalities are ‘unlikely to be biased by such reporting tendencies’ (p. 14).

A strong point of this present study is that it tests two different dependent health variables in country-specific models at the first step. When using the effect of income on health, both health variables have certain advantages and disadvantages as indicators of health inequalities. The interpretation of marginal effects at the mean is more straightforward when running regressions on the health dummy variable. However, after combining the categories, less information was obtained compared to using the original 5-point response scale; also, the way the categories were combined is perhaps controversial. Therefore, health inequalities were also calculated on the basis of a dummy variable of (very) poor health versus fair and (very) good health as recommended by Etilé and Milcent [61]. Probably due to the rather low share of respondents with (very) poor health, only a few countries displayed significant income-related health inequalities. Since it was questionable as to whether this health dummy was an appropriate indicator for health inequalities if it targeted such a small number of respondents, I decided against presenting those results.

Regarding the index of dissimilarity as an indicator of health inequalities, I discovered that using the original 5-point response scale as a metric rather than an ordinal variable led to approximately equivalent results at both the first and second step.

Studying income-related health inequalities across countries imposes the challenge to generate one variable for income across a variety of countries. In this case, the variable had to ensure that respondents’ income in Luxembourg was comparable to respondents’ income in Moldova—to name two extreme cases. Additionally, some countries had a high rate of missing values. Both factors were taken into account when computing the income variable but nevertheless could be interpreted as a limitation of this study. For future studies, education instead of income might be an interesting measure for socio-economic health inequalities. However, given that half of the EVS dataset consists of post-communist countries, where a good part of the adult population was educated during Communism and equal access to education was emphasised [62], educational health inequalities would need to be interpreted carefully, for they might not adequately describe socio-economic inequalities.

While the Gini index is a widely used and recognised indicator for income inequality, one single predominant measure for the impact of social policies in comparative health inequality research is missing. Dahl and van der Wel ([63], p. 60) even claimed that ‘a social expenditure approach is new in this field of research.’ Using social protection expenditures in the percentage of GDP as a quantitative measure for social policies should be understood as just a starting point for further analyses. The number of various countries in the EVS made it impossible to find one single data source for social protection expenditures. However, with Eurostat, I found a database encompassing 30 countries (see Appendix B). Furthermore, I took reasonable care in data investigation for the other countries and tried to double-check with other sources, e.g., national statistics. Although social protection expenditures already are a specification of the comprehensive understanding of social policies, it would be interesting for future research to look at the effects of schemes of social protection, i.e., minimum income protection, on health inequalities.