Unequal unions? A comparative decomposition of income inequality in the European Union and United States

Abstract

This study applies improved household income data to measure and decompose trends in pan-European income inequality from 2006 to 2014. To contrast the relative significance of economic homogeneity versus the efficacy of welfare state and labour market institutions in shaping income distributions, we compare the structure of inequality in the 28 Member States of the European Union (EU-28) to that of the 50 United States. This comparison stands in contrast to the standard practice of evaluating the United States against individual EU Member States. Despite the greater relative heterogeneity of the EU-28 and our corrections for the underreporting of household income in the United States, post-fisc income inequality in the EU-28 remains lower than that of the United States from 2006 onward. Moreover, inequality appears to be rising in the United States, while it has remained stagnant since 2008 in the EU-28. In both unions, and particularly the United States, within-state income differences contribute more to union-wide inequality than between-state differences. In a counterfactual analysis, we find that if the EU-28 matched the between-state homogeneity of the United States, but maintained its relative within-country inequalities, pan-European inequality would fall by only 20 percent. Conversely, inequality in the United States would fall by 34 percent if it matched the within-country inequality of the EU-28. Our findings suggest that the strengthening of egalitarian institutions within the 28 Member States is more consequential than economic convergence in reducing pan-European income inequality. We highlight institutional challenges towards achieving a ‘more equal’ Europe and discuss implications for future EU policymaking.

Keywords

European Union federalism inequality social policy United States welfare state

Introduction

The rise of income inequality across advanced economies has become a central concern of political discussion and policy research. While most analyses of inequality focus on within-country income differences, the European Union (EU) has focused primarily on decreasing inequality between its 28 Member States. The Five Presidents’ Report, for example, devotes a full chapter towards the aim of ‘convergence between Member States’ (Juncker, 2015: 4), while the more recent European Pillar of Social Rights similarly offers a framework for achieving upwards convergence in working and living conditions across Member States (European Commission, 2017b). Despite these aims to reduce disparities within the broader EU, few evaluations of social indicators, such as poverty or income inequality, exist at the European level. It thus remains unclear the extent to which the current ‘European social model’ reduces pan-European income differences or how the structure of inequality within the EU compares to that of other federalist unions.

This study decomposes levels and trends of income inequality within the 28 Member States of the European Union (EU-28) and benchmarks its relative success against levels of inequality across the 50 United States. We give particular analytical focus to the role of economic convergence between states versus the strength of egalitarian institutions within states in shaping income differences within two unions.

This distinction of within- versus between-state equality helps to frame the vast set of differences between the US and EU-28. Compared to the EU-28, the 50 United States are relatively homogeneous with respect to their economic performance. Moreover, the states operate within a highly centralized federal government which, through its administering of social transfers, acts to reduce income disparities between richer and poorer states. Conversely, the EU is a collection of politically, culturally, linguistically and economically diverse Member States with limited centralization at the European level and few mechanisms for redistribution from richer to poorer states. At the same time, prior evidence makes clear that national welfare states within the EU tend to be more effective than the US at reducing within-country income disparities (Alber and Gilbert, 2010).

The diverse features of these two unions provoke several questions. In comparative research, the 50 United States are most often compared to individual EU Member States or clusters of Member States situated in Western and Northern Europe (Alesina et al., 2001; Brandolini and Smeeding, 2011; Corak, 2013; OECD, 2015). But when compared to today’s larger and more heterogeneous EU, does the US still feature comparatively high levels of income inequality? Moreover, does the structure of income inequality within the two unions suggest that economic convergence between states or the strengthening of egalitarian institutions within states is of greater relative importance? And to what extent does Europe’s lack of robust cross-state transfer mechanisms, a la that of the US, inhibit its aim to become a ‘more social’, or at least more equal, EU? These inquiries not only challenge prior comparisons of income inequality within the US and EU but also offer implications for policymaking decisions aimed at narrowing income differences within the two unions.

In addressing these questions, we add three primary contributions to the broader literature on social policy and income inequality. First, we utilize improved household income data from the EU and US to provide the most accurate portrait yet of differences in income inequality within the two unions. Specifically, we correct for the issue of income underreporting within US household survey data, apply a common definition of disposable household income to facilitate accurate comparisons across the two unions and correct for price differences across EU Member States and the 50 US. Despite the relative diversity of the EU Member States and our utilization of improved American income data, we still find that income inequality in the US is higher than that of the EU-28 after accounting for taxes and transfers.

Second, we clarify the relative importance of within- versus between-state differences in shaping levels of income inequality within the EU-28 and US. In the EU, we find that income differences between countries contribute to about 21 percent of pan-European income inequality (applying the Theil index) in 2014, though this relative contribution has decreased by 6 percentage points since 2006. Despite recent theorizations of state-level divergence of social and economic policies in the US (Bruch et al., 2016), we find that income differences within states contribute to more than 99 percent of the nation’s overall inequality levels.

Finally, we analyse the relative efficacy of economic convergence versus egalitarian institutions, such as more equitable labour markets and more robust tax and transfer systems, towards union-wide inequality levels. To do so, we construct a counterfactual in which the EU-28 matches the economic homogeneity of the US, but maintains its relative within-state inequalities. In this scenario, pan-European inequality would fall by 20 percent in 2014, far less than the relative change (34%) if the US matched the within-state inequality of the EU-28.

This collection of findings reveals that the strengthening of welfare state and labour market institutions is of greater relevance than economic convergence in reducing union-wide income inequality. Though the ‘European social model’ is, indeed, superior to the American model in narrowing income differences among individuals in countries as diverse as Sweden and Romania, the evidence nonetheless suggests that ‘social convergence’ between Member States must accompany economic convergence if further reductions in pan-European income inequality are to be achieved.

Theory and background

Union-wide inequality

Analyses of income inequality have most often focused on the national level to evaluate the sources, consequences and potential remedies of the widening gap between the ‘haves and the have nots’. Recently, however, the literature has turned its analytical focus to understanding global inequality, particularly in light of data advancements in many emerging economies and the scholarship of researchers such as Milanović (2016). Despite this shift in unit of analysis, few studies have adopted the EU as the focal point of analysis. We elaborate on three sets of reasons why a pan-European measure of income inequality serves as a useful complement to national and global measures of inequality.

First, a pan-European focus on inequality carries implications for established policy targets and social objectives. As detailed in The Five Presidents’ Report: Completing Europe’s Economic and Monetary Union, Juncker (2015) explicitly calls for upwards economic convergence between Member States, while the European Pillar of Social Rights calls, in effect, for greater ‘social convergence’ of the EU (European Commission, 2017b). Similarly, the United Nations’ Sustainable Development Goals (SDGs) explicitly identify inequality-reducing targets to be achieved by 2030, combined with ‘sustained, inclusive and sustainable economic growth’. A pan-European measure provides a useful tool to monitor the fulfilment of these policy targets, while an analysis of the structure of European inequality can inform future policy priorities.

Second, sociological research has offered some (albeit contested) evidence that individual perceptions of inequality and wellbeing are not limited to the boundaries of nation-states. Delhey and Kohler (2006), for example, suggest that Europeans tend to evaluate their position in society relative to other Europeans, not simply to other citizens of their country of residence. For this reason, many have criticized the national, rather than EU-wide, focus on trends in inequality (Beck et al., 2007; Heidenreich, 2016). Others, however, have disputed claims of a ‘Europeanization of reference groups’, suggesting instead that relative perceptions of wellbeing remain largely confined to national borders (Nolan and Whelan, 2011; Whelan and Maître, 2009).

Regardless of precise reference group, evidence suggests that rising levels of income inequality within the EU-28 may have consequences for intra-European migration, public trust and social cohesion. Standard migration models assume that income differentials induce migration from poorer to richer areas (Todaro, 1969), and recent trends in the EU serve as evidence to this claim: after the economic recession hit European economies in 2009, the level of migration from Southern and Eastern European countries towards wealthier countries increased (Fries-Tersch et al., 2016).¹ Moreover, increases in inequality between states in a common federation or union can negatively affect social cohesion and undermine support for the union’s institutions (Ferrera, 2005; Ferroni et al., 2008; Milanović, 2016). These issues appear to be particularly relevant in the EU as levels of trust in EU institutions have generally declined throughout the past decade (European Commission, 2017d)² and as frictions have intensified regarding financial obligations between some Member States (i.e. Greece and Germany, as well as public debate in the United Kingdom leading up to the ‘Brexit’ referendum).

Several previous studies have made progress in describing pan-European inequality trends (Beck-field, 2009; Benczúr et al., 2017; Bornschier et al., 2004; Brandolini, 2007; Darvas, 2016; Dauderstädt and Keltek, 2014; Eurofound, 2017; Firebaugh, 1999; Heidenreich, 2016). Of these, Heidenreich (2016) provides perhaps the most comprehensive portrait to date, estimating EU-wide inequality with European Union Statistics on Income and Living Conditions (EU-SILC) data and decomposing the mean logarithmic deviation of disposable incomes into between-country and within-country inequality. However, no analysis has, to our knowledge, combined four elements necessary for a more thorough understanding of pan-European inequality: household income data across all Member States over time, pre/post evaluations of market and disposable income inequality, decomposition of the relative importance of economic convergence versus national welfare states in shaping EU-28 inequality and benchmarking against the US using more accurate American income data. As we detail, the combination of these four features leads to greater clarity of the structure of pan-European inequality and more precise policy recommendations for achieving a more equal Europe.

US and the union of European welfare states

To contrast the relative importance of within- versus between-state income differences in shaping union-wide inequality, we benchmark the levels and structure of income inequality in the EU-28 against that of the US. Traditionally, indicators of poverty or inequality in the US are compared to those of individual EU Member States (Brandolini and Smeeding, 2011; Corak, 2013; OECD, 2015). Some scholars have argued, however, that such comparisons may conceptually undervalue the size and heterogeneity of the US (Alber and Gilbert, 2010; Smeeding and Rainwater, 2001). Consider that New York City, on its own, has a larger population (8.5 million) than roughly half the EU Member States. If inequality in the US is compared to the larger, more heterogeneous EU – encompassing all 28 of its Member States – does the EU still stand out as the more egalitarian union? In decomposing the structure of inequality within the two unions, we focus on two particular distinctions of the US that should facilitate lower levels of inequality within the country: its relative homogeneity and its mechanisms of cross-state redistribution.

The relative homogeneity of the US compared to the EU-28 is straightforward: the economic, political, social and culture differences between Sweden and Romania are, by any imaginable measure, likely to be larger than the differences between New Hampshire and Mississippi. Consider, for example, that disposable median income in Sweden is more than 10 times larger than that of Romania, while the richest state in the US has a median income only 1.6 times larger than the poorest state (Eurostat, 2017; Guzman, 2017). Even after correcting for purchasing power between countries and states, the relative differences remain large. Given this, the US has an inherent advantage in reducing levels of inequality: the relative difference between states is likely to be much smaller than the differences in incomes between the EU-28. Upwards economic convergence between Member States has long been a priority for the EU; from an inequality perspective, however, it remains unclear the extent to which convergence between countries affects pan-European income inequality.

It is well established that the ‘liberal’ or ‘residualist’ welfare state within the US is less conducive towards narrowing income differences than the more robust tax and transfer systems of Western and Northern Europe (Alber and Gilbert, 2010; Alesina et al., 2001; Esping-Andersen, 1990). Nonetheless, the American welfare state’s mechanisms of cross-state redistribution provide the US an institutional advantage (relative to the EU-28) in generating reductions of within-state inequality even in its comparatively poor regions. Several of the nation’s largest social programmes – Old-Age, Survivors, and Disability Insurance (OASDI); the Earned Income Tax Credit (EITC); the Supplemental Nutrition Assistance Programme (SNAP, generally referred to as ‘food stamps’) and Medicare – are predominantly federally funded. Though many states offer supplements to the EITC and each manages its own social assistance programme (Temporary Assistance for Needy Families), spending on state-administered social programmes remains small and is steadily declining relative to federal spending on social programmes (Falk, 2014). The result is that federal tax revenues, to which states with higher shares of high-income households contribute more, are redistributed in the form of transfer benefits to states with greater shares of low-income or vulnerable households. This system of ‘social federalism’ is largely absent in the EU, where social policy is legislated and administered at the national level.³ As a result, welfare states across the EU-28 are more diverse in size and scope: poorer Member States tend to have less financial capacity administer a robust welfare state, whereas richer Member States can perhaps more effectively narrow income differences through generous social policies. While several proposals have been put forth to create a European-wide unemployment scheme (European Commission, 2017c) or to redistribute funds towards social protection schemes in poorer countries (Jones, 1993), social policy nonetheless remains a national competence. Unclear, however, is the extent to which this decentralization prevents the EU-28 from achieving lower levels of inequality. It may be the case that the ‘European social model’ (Alber and Gilbert, 2010), or the tendency for its Member States to prioritize egalitarian labour markets and welfare states, is sufficient to produce lower levels of pan-European inequality despite the union’s lack of cross-state redistribution. Conversely, the differences between Member States – both in economic performance and redistributive capacity – may be too large to overcome the homogeneity and institutional advantages of the US.

Economic convergence and egalitarian institutions

With these differences of homogeneity and welfare state institutions in mind, we highlight two narratives that guide our measurement, decomposition and comparison of inequality in the EU-28 and US. The economic convergence narrative is rooted in the EU’s desire to ‘deepen its Economic and Monetary Union’ and achieve greater macroeconomic homogeneity among its diverse Member States (and its newer, less wealthy Member States, in particular; European Commission, 2017a; Juncker, 2015). If economic convergence were to be of greater consequence towards the reduction in income inequality, we might find in our analyses that between-state inequalities are of greater relevance than within-state differences in explaining union-wide inequality. In other words, the relative homogeneity of the 50 US is more significant than the equalizing effect of national tax and transfer systems. If this were the case, we might expect the data to point towards three findings: that income inequality in the US will be lower than pan-European inequality; that between-country inequality will contribute more to pan-European inequality than within-country income differences and that if the EU-28 matched the between-state homogeneity of the US, it would see greater relative reductions in inequality than if the US matched the within-state inequality of the EU-28.

Alternatively, the efficacy of national welfare states in narrowing within-country income differences may prove to be more consequential for union-wide inequality than cross-state homogeneity. We refer to this as the egalitarian institutions narrative: the role of national labour market and welfare state institutions in reducing within-country inequality may be more consequential than between-state homogeneity. If this were the case, we might expect to find that the EU-28, despite its relative diversity, has lower levels of inequality than the US; that within-country inequality will contribute more to pan-European inequality than between-state diversity and that if the US matched the within-state inequality of the EU-28, it would see greater relative reductions in inequality than if the EU-28 matched the between-state homogeneity of the US. We now turn towards testing these two narratives.

Data and methods

Harmonized household income data

We use household income data from EU-SILC to obtain pan-European income inequality measures and data from the US Current Population Survey (CPS ASEC) for the US. The two surveys allow for comprehensive and comparable definitions of pre-fisc (‘factor’) and post-fisc (or ‘disposable’) income for the 28 EU Member States and 50 United States. Our time span runs from 2006 to 2014, the first and latest years in which comparable household income data for each of the EU-28 are available.⁴

The CPS ASEC suffers from severe underreporting of means-tested transfers; thus, incomes are underestimated in many households in the bottom half of the income distribution. Prior research shows, for example, that between 40 and 60 percent of transfers from the Temporary Assistance to Needy Families (TANF), SNAP and Supplemental Security Income (SSI) programmes are not captured in the survey data (Meyer and Mittag, 2015; Parolin, 2017). To address this, we apply benefit imputations from the Transfer Income Model, version 3 (TRIM3) model developed by the Urban Institute. The TRIM3 augmentations match administrative records on social assistance caseloads across the 50 states to impute more accurate information on TANF, SNAP and SSI benefit receipt back into the survey data. Underreporting of means-tested transfers has not been identified as a similar issue in the EU-SILC data. This is largely due to the reliance in many countries on register data, which provide more precise estimates of household income.⁵ Thus, the available evidence suggests that use of the TRIM3 imputations provides a more accurate comparison of the two unions rather than ‘overcorrecting’ for the US.

We follow the framework of the Luxembourg Income Study (LIS) in harmonizing our income concepts between the EU-SILC and CPS ASEC. Our pre-fisc income variable includes gains from employment (including self-employment) and gains from rents, interest and dividends. Our post-fisc variable additionally takes into account all taxes and transfer income. We do not include imputed rents into our analyses. In Appendix 1, we provide an overview of the income components used to assess household income in the EU-SILC and CPS ASEC. We apply the modified Organisation for Economic Co-operation and Development (OECD) equivalence scale to adjust the income concepts for the household size. In robustness checks available in the online appendix, we also test our findings using the square root equivalence scale, but the findings are not substantively altered.

To ensure that the income concepts are comparable across all countries, we follow standard practice in correcting the income variables for national price-level indices (Milanović, 2011). We download price-level indices normalized to EU-28 = 1 from Eurostat (prc_ppp_ind), selecting them for different levels of aggregation (gross domestic product (GDP) or household final consumption expenditure (HFCE)). We present the results for the income distributions corrected for the GDP-based price-level indices, but provide the results for the HFCE-corrected income distribution in the online appendix. Price indices also vary across the 50 US. To account for this, we apply the US Bureau of Labour Statistics’ regional price parities (RPPs) to incomes in each state (Cover, 2016). The RPPs indicate, for example, that US$1 in the state of Alabama tends to carry more purchasing power than US$1 in Alaska.

We combine data for the 28 EU Member States into one sample for each year to produce pan-European estimates of income inequality.⁶ Our primary analyses observe the total population of each of the samples; in 2014, for example, this leads to a total sample size of 556,953 individuals for EU-28. In Appendix 2, however, we also present all inequality results for the working-age population (aged 18–65 years). The primary differences come when observing pre-fisc income inequality, as the estimates using the total sample includes many households headed by pensioners, which are less likely to have any factor income. In each wave, individual weights account for the different probabilities of selection and survey non-response at the individual level. Moreover, the weights are constructed in a way that it makes the merged sample of all EU countries representative of the EU population when it is applied to EU-wide statistics. We compute standard errors using the household identifier as primary sampling unit.⁷ We also use the standard population weights in the CPS ASEC for our US estimates.

Measuring inequality

Our primary measure of income inequality is the Gini coefficient, the most utilized index in inequality research. Other inequality measures, such as the generalized entropy family of indices, the Atkinson indices and some quantile ratios, were also computed and are available on request. After we present levels and trends of the Gini coefficient for the EU-28 and US, we turn towards decomposing the relative contribution of within-state and between-state inequalities. To do this, we exploit the property of decomposition in non-overlapping groups of individuals typical of many inequality indices. For example, the overall inequality measured by the Theil index (GE1) can be additively decomposed as the sum of the between-country and the within-country inequality

G E_{1} (Y) = G E_{1}^{B} (Y) + G E_{1}^{W} (Y)

(1)

with

G E_{1}^{W} (Y) = \sum_{m = 1}^{M} v_{m} G E_{1} (Y^{(m)})

(2)

where v_m is the country’s m’s share of the total income and $G E_{1} (Y^{(m)})$ is the inequality within m. In turn, $G E_{1}^{B} (Y)$ is the between-country inequality, in which each individual is assigned the mean income of the country in which he or she is surveyed. We decompose the within- and between-contributions by country of current residence (in the EU) and state of current residence (in the US). As such, the between-country component of the decomposition should be interpreted as inequality between the mean incomes of citizens residing in the different countries or states. We present only the Theil index (GE₁) decompositions, but data for mean logarithmic deviation (GE₀) decompositions are provided in the online appendix.

In a counterfactual analysis, we evaluate how EU-28 inequality would change if the Member States matched the between-state income homogeneity of the 50 United States, but maintained their relative within-state income inequality. To do this, we substitute the between-state contribution ( $G E_{1}^{B} (Y)$ ) of the EU with that of the US to produce a counterfactual estimate of pan-European inequality. We also perform a similar analysis to evaluate how inequality in the US would change if it were match the within-country inequality of the EU Member States, substituting the within-state ( $G E_{1}^{W} (Y)$ ) of the US with that of the EU. The Stata code to replicate our analyses can be found in the online appendix.

Results

With our equivalized, purchasing power parity (PPP)-adjusted income distributions for the EU-28 and US, we now evaluate levels and trends in income inequality. We first present trends in post-fisc income inequality within the two unions and then evaluate the role of taxes and transfers in shaping these results. We then decompose the relative contribution of within-state inequality versus between-state inequality towards overall income disparities within the two unions. As a final step, we construct a counterfactual in which the EU-28 matches the between-state homogeneity of the US, and the US matches the within-state inequality of the EU-28, to further evaluate the relative significance of egalitarian institutions versus economic convergence in shaping inequality.

Figure 1 displays the Gini index for net disposable income of households in the EU-28 and US from 2006 to 2014.

Figure 1.

Income inequality (Gini) in the US and EU-28.

Despite the relative diversity among the 28 EU Member States, post-fisc income inequality is consistently lower than that of the US. In 2014, for example, the Gini of the EU-28 was 0.35, while that of the US was 0.38 (a statistically significant difference). Though levels of inequality nearly converged in 2010, the peak of the recession, the differences have otherwise remained steady since 2006. Within the EU-28, we do observe slight evidence of a decline in income inequality over time: from 2006 to 2014, the Gini index fell from 0.36 to 0.35 – a small, but statistically significant decline. Conversely, the Gini in the US in 2014 was not statistically different than its value in 2006.

How does our evaluation of pan-European inequality change if we evaluate pre-fisc income inequality? Figure 2 adds the Gini coefficient from market incomes in the EU-28 and US. The gap between the pre- and post-fisc inequality levels provides a rough estimate of the role of taxes and transfer systems in reducing income inequality within the two unions.

Figure 2.

Inequality (Gini) among US and EU-28 Member States before/after taxes and transfers.

As the dashed lines indicate, pre-fisc income inequality has remained slightly higher in the EU-28 relative to the US.⁸ Though the average EU Member States performs slightly better than US with respect to pre-fisc inequality (an unweighted mean around 0.5 in 2014), between-country differences in pre-tax wages add enough to overall inequality to move the EU-28 line above that of the US. A key difference between the US and EU-28, though, is the relative trends: while factor income inequality has basically remained steady in the EU since 2006, it has increased from 0.506 to 0.532 in the US, a 5 percent increase, from 2006 to 2014. By 2014, pre-fisc inequality in the US nearly climbed to the level of the EU. The higher levels of pre-fisc inequality despite lower levels of post-fisc inequality reflect the relative strength of European welfare states: in 2014, European tax and transfer systems reduced the level of income inequality by 36 percent (from 0.54 to 0.35), while the post-fisc inequality fell relative to pre-fisc inequality by 29 percent (from 0.53 to 0.38) in the US. As there is no unified European welfare state that coordinates tax and transfer payments across countries, the reduction in inequality should be interpreted as the aggregate effect of the national and subnational tax-and-transfer systems. These findings provide initial support for the role of egalitarian institutions: although the 28 Member States are more diverse and have higher level market income inequality, the strength of national welfare states nonetheless drives inequality to a lower level than that of the 50 United States.

We now turn towards decomposing the relative contributions of income differences within countries versus between countries in explaining the inequality trends presented above. As detailed before, an increase in the between-country component of EU-28 inequality would signal divergence between countries, contra to the aims of the EU’s policy targets. We anticipate between-state inequality in the US to be substantially lower due to greater relative homogeneity among the 50 states, but give particular focus to extent to which the largely centralized American tax and transfer system reduces between-state inequalities in the US (in contrast, perhaps, to the decentralized and more diverse European welfare states).

The decomposition results are presented in Figure 3 (EU-28) and Figure 4 (US).

Figure 3.

Theil decomposition of income inequality among EU-28 Member States (disposable household income).

Figure 4.

Theil decomposition of income inequality among US (disposable household income).

We focus first on the EU-28. As Figure 3 details, inequality between the EU Member States accounted for about 27.2 percent of overall inequality in 2006. By 2014, the relative between-country inequality contributed to an estimated 21.2 percent of overall pan-European income inequality, a 22 percent decrease from its 2006 level. The general decline of the relative between-state contribution signals positive news for the EU’s push for convergence: country of residence may be increasingly giving way to intra-country differences in shaping EU-28 inequality, although this decline has recently stopped.

Conversely, income inequality in the US (Figure 4) is almost entirely a within-state story. In each year of analysis, less than 1 percent of overall inequality in the US is due to differences in income distributions across states. Despite divergence across the 50 states in social and labour market policies (Bruch et al., 2016), we observe no significant change in between-state inequality from 2006 onwards. Simply put, income inequality is high in each of the 50 states (the Gini index reaching a low of 0.32 in Utah, still higher than that of most EU Member States), while average incomes, particularly after accounting for regional price differences, vary little between states.⁹

To what extent do taxes and transfers affect the relative contributions of within- and between-state inequality? To evaluate this, we assess trends in the absolute (rather than relative) contribution of between-state differences in equality. Similar to our prior analyses, we depict these trends before and after taxes and transfers to provide a sense of how welfare states shape levels of inequality. Figure 5 presents these results.

Figure 5.

Between-state contribution (absolute) to inequality before and after taxes and transfers.

The dashed lines depict pre-fisc levels of inequality, while the solid lines depict post-fisc inequality. Again, Figure 5 displays only the between-country (or between-state) contribution to overall inequality. Within the EU-28, we see declining levels of between-country inequality from 2006 to 2009, but stagnation from 2009 onwards. For pre-fisc inequality, the EU-28 shows steady decline from 2006 (0.061) to 2014 (0.047). In 2013 and 2014, however, we find that taxes and transfers actually increase the level of inequalities between EU Member States, albeit by a very small level. This finding suggests that differences in the relative redistributive strength of national welfare states across Europe slightly widen differences in average household incomes across EU Member States in recent years. In the US, we find that taxes and transfers play only a small role in decreasing between-state inequalities: in all years of analysis, between-state differences in income inequality remain small both before and after redistribution with little change over time. But rather than increasing between-state differences, as in the case in the EU-28, the American welfare state has a small but statistically significant reduction effect on between-state inequality. The small differences in pre- and post-fisc levels of between-state inequality demonstrate that tax and transfer systems have a strong reduction effect on within-state inequalities (as detailed in Figure 2), but make little difference in average incomes between states. Economic convergence thus appears necessary to narrow between-state inequalities, but what effect might such convergence have on pan-European income inequality?

To assess this, we now turn to our counterfactual assessment of EU-28 and US inequality. As detailed, we test the extent to which inequality in the EU-28 would fall if the 28 Member States matched the economic homogeneity of the 50 United States, but maintained their relative within-country inequalities. We then test the extent to which inequality in the US would fall if the states matched the within-country inequality of the US, but maintained their relative cross-state homogeneity. From the data presented in the previous figures, we can deduce that the US will see a greater relative decline in its level of inequality: as the counterfactual states will be identical (both will have the within-state inequality of the EU-28 and the between-state inequality of the US), the union with the higher level of observed inequality (the US) will naturally see a greater relative decline. Of particular interest, though, is a depiction of the extent to which income inequality in either polity would fall in the counterfactual scenario. Figure 6 presents the findings from results.

Figure 6.

Counterfactual union-wide inequality in 2014 if the US achieved the within-state inequality of the EU-28 and if the EU-28 achieved the between-state homogeneity of the US.

To what extent could economic convergence further narrow pan-European income inequality? If the EU-28 were to match the between-state convergence of the US, but while maintaining the same relative income distributions, EU-28 income inequality would fall by an estimated 20 percent (from 0.214 to 0.171 using the Theil index decomposition). Conversely, if the US were to match the within-state inequality of the EU, it would see a 34 percent reduction in overall inequality (from 0.258 to 0.171). These findings support the egalitarian institutions narrative: more so than economic convergence between states, the ability of welfare states to reduce income differences within states is more consequential towards reducing union-wide income inequalities.

Discussion and conclusion

Diverging from inequality analyses that compare the US to individual EU Member States, this article decomposes and compares trends of pan-European income inequality with that of the 50 United States. Our corrections for the underreporting of means-tested transfers in the US allow us to present the most up-to-date, accurate and comparable portrait of income inequality within the two unions. Applying these data, we first evaluate the performance of taxes and transfers in reducing income inequality across the EU-28 and US and then decompose the relative contribution of within-state versus between-state income differentials towards union-wide inequality. We now summarize our primary findings and elaborate on their implications for the role of social policy in mitigating union-wide income inequality in the future.

Our findings reveal that despite the greater relative heterogeneity of the EU-28, its level of income inequality has been consistently lower than that of the US from at least 2006 onwards. This is largely due to the strength of national welfare states across the EU: although pre-fisc income inequality is higher in the EU-28, redistributive systems bring it down to a level lower than that of the US. We also find that the strengthening of egalitarian institutions within states is of greater consequence towards the future reduction of pan-European inequality than economic convergence between states. If the EU-28 were to match the economic homogeneity of the 50 US, its inequality would fall by about 20 percent – a notable reduction, but still less than the US would gain (34 percent reduction) if it matched the within-state inequality of the EU-28.

This does not mean, of course, that greater economic integration is an unworthy aim: a one-fifth reduction in pan-European inequality would be a notable achievement, and there are other benefits beyond the reduction of income inequality that would likely result from a more economically homogeneous Europe. Moreover, economic growth in the EU-28’s comparatively poorer Member States may strengthen the capacity necessary for such states to enhance redistributive systems, or it may coincide with a reduction of pre-fisc income inequality. However, the evidence is clear that the greatest gains to be made in reducing EU-28 inequality are through the reduction of within-country income disparities. Such findings are in contrast, however, with observed policy developments and country-level inequality trends: within most EU Member States, income inequality has risen throughout the past decade, while the generosity of certain social protections, such as levels of minimum income protections, has stagnated or declined throughout most Member States in recent years (Cantillon and Vandenbroucke, 2014; OECD, 2017).

Of course, large diversity exists in the extent to which EU Member States currently reduce levels of within-country inequality. A country-by-country breakdown reveals that taxes and transfers in Finland reduced levels of inequality by nearly 50 percent in 2014 (a change in the Gini index from 0.495 to 0.251), while the Bulgarian welfare state led to only a 26 percent reduction (from 0.502 to 0.370). Differences in within-state inequality reductions among the 50 United States are far narrower: the states with the highest and lowest relative reductions in inequality after taxes and transfers were West Virginia (37% decrease in 2014) and Texas (25% decrease). As detailed, one institutional advantage of the American welfare state is its mechanisms for cross-state redistribution: federally administered transfers act to redistribute funds from richer to poorer states, likely contributing to the smaller variance of within-state inequality reductions.

It is perhaps unlikely, however, that the EU will soon adopt a union-wide tax and transfer scheme to equalize the redistributive capacity of states as diverse as Finland and Bulgaria. Though Gerhards et al. (2016) find that a majority of adults in Germany, Poland and Spain support the implementation of a pan-European welfare state, recent policy developments within the European Commission (2017b) point towards the establishment of non-binding principles and recommendations. Nonetheless, attempts to reduce income disparities within the EU-28 must consider how to strengthen welfare state and labour market institutions and of newer EU Member States in particular. Consider that the six Member States with the smallest relative decline of within-country inequality after taxes and transfers – Bulgaria, Latvia, Cyprus, Lithuania, Estonia and Romania – are each part of the enlargement of the EU from 2004 onwards.

Proposals for and critical discussions on mechanisms to enhance social solidarity between Member States have emerged in recent years. The European Pillar of Social Rights sets out core principles that would, if acted upon, lead towards the enhancement of ‘egalitarian institutions’ within Member States. The Pillar explicitly prioritizes the development of fairer labour market conditions and broader social protections, touching on issues ranging from wage and gender equality to minimum incomes and unemployment schemes. Though the focus of the Pillar is of course broader than a concern for income inequality, progress towards its aim of creating a ‘fairer economic and monetary union’ would surely contribute to a narrowing of income differences (European Commission, 2017b). A potential weakness, however, remains the apparent lack of the Pillar’s binding enforcement mechanisms with respect to most of its policy recommendations.¹⁰ Prior EU strategies aimed at reducing poverty and narrowing income disparities (i.e. Lisbon and Europe 2020) have put forth similar policy recommendations. The Pillar’s principle of ensuring a ‘right to adequate minimum income benefits’, for example, mirrors a 1992 recommendation calling for ‘the basic right of a person to sufficient resources and social assistance’ (European Economic Community, 1992). Nonethe-less, minimum income protections in most EU Member States sit far below anti-poverty levels (Cantillon et al., 2017). As others have pointed out, these non-binding policy recommendations act as weak instruments towards reducing levels of poverty or income inequality (Cantillon et al., 2017; Daly, 2006; Saraceno, 2009) and concern exists that this may remain the case with the Pillar of Social Rights.

Vandenbroucke (2017), meanwhile, offers a more concrete call for a ‘European Social Union’ to ‘support national welfare states on a systemic level in some of their key functions and guide the substantive development of national welfare states’ (p. 4). Specific policy instruments to achieve these ends include automatic stabilization mechanisms in the form of, say, a European unemployment benefit scheme (European Commission, 2017c). As Vandenbroucke writes, and as the evidence in this article suggests, social convergence is as important as economic convergence if the EU is to see declines in inequality moving forward. The European Pillar of Social Rights and calls for a ‘European Social Union’ both offer promising steps towards that aim, but with an important caveat: if policy strategies and ‘key principles’ are established without an accompanying enforcement mechanism, it remains unclear whether such efforts will offer the progress needed to strengthen national welfare states and labour market institutions.

In closing, we call attention to several limitations of our study. Though we focus exclusively on income inequality, future studies may expand the scope of analysis to measure pan-European inequalities across age group, family status, race and ethnicity, education or other subgroups. Moreover, as data on household wealth become more accessible, a pan-European focus of wealth inequality would be of great relevance in measuring the broader disparities within and between EU Member States. An opportunity also exists to understand how regional variation at the subnational level contributes to national and supranational inequalities. Given ongoing residential segregation of high-income earners into cities, urban–rural divides in income differences deserve greater consideration. Though our analysis finds low levels of between-state homogeneity in the US, for example, this may mask significant urban/rural inequities within states, which may have rising political and social consequences (Lichter and Ziliak, 2017).

Finally, future research should continue to investigate how newer EU Member States, in particular, can overcome fiscal or political limitations to enhance their systems of social protection. As our findings suggest, the strengthening of welfare state and labour market institutions within Member States must accompany economic convergence if the EU is to become a more equal union.

Supplemental Material

Parolin_supplementary – Supplemental material for Unequal unions? A comparative decomposition of income inequality in the European Union and United States

Supplemental material, Parolin_supplementary for Unequal unions? A comparative decomposition of income inequality in the European Union and United States by Stefano Filauro and Zachary Parolin in Journal of European Social Policy

Footnotes

Appendix 1

Appendix 2 Acknowledgements

Direct correspondence to Zachary Parolin (Zachary.Parolin@uantwerpen.be): University of Antwerp, 2 Sint-Jacobstraat, 2000 Antwerp, Belgium. For helpful comments and feedback, the authors are grateful to Branko Milanović, Stephen Jenkins, Michele Raitano, Alessia Fulvimari, Tim Goedemé, participants at the 2017 International Conference on Inequality in Bologna, Italy, participants of the JRC Research Seminar in March 2017, two anonymous reviewers and the editors at JESP. The authors accept responsibility for all conclusions drawn from the data. The views expressed in the text are the private views of the authors and may not, under any circumstances, be interpreted as stating an official position of the European Commission.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Parolin acknowledges a PhD Aspirant fellowship from the Research Foundation Flanders (FWO).

Notes

References

Alber

Gilbert

(2010) United in Diversity? Comparing Social Models in Europe and America. New York; Oxford: Oxford University Press.

Alesina

Glaeser

Sacerdote

(2001) ‘Why Doesn’t the US Have a European-Style Welfare System?’ NBER Working Paper Series no. w8524. Cambridge, MA: National Bureau of Economic Research.

Beck

Grande

Cronin

(2007) Cosmopolitan Europe. Cambridge: Polity Press.

Beckfield

(2009) ‘Remapping Inequality in Europe: The Net Effect of Regional Integration on Total Income Inequality in the European Union’, International Journal of Comparative Sociology 50: 5486–509.

Benczúr

Cseres-Gergeley

Harasztosi

(2017) ‘EU-Wide Income Inequality in the Era of the Great Recession’, JRC Working Papers in Economics and Finance no. 2017/14. Brussels: European Commission.

Bornschier

Herkenrath

Ziltener

(2004) ‘Political and Economic Logic of Western European Integration: A Study of Convergence Comparing Member and Non-Member States, 1980–98’, European Societies 6: 71–96.

Brandolini

(2007) ‘Measurement of Income Distribution in Supranational Entities: The Case of the European Union’, Discussione Working Paper no. 623. Rome: Bank of Italy.

Brandolini

Smeeding

(2011) ‘Income Inequality in Richer and OECD Countries’, in Salverda

Nolan

Smeeding

(eds) The Oxford Handbook of Economic Inequality, pp. 71–100. Oxford: Oxford University Press.

Bruch

S.K.

Meyers

M.K.

Gornick

J.C.

(2016) ‘Separate and Unequal: The Dimensions and Consequences of Safety Net Decentralization in the US 1994–2014’ IRP Discussion Paper Series no. 1432-16. Madison, WI: Institute for Research on Poverty, University of Wisconsin–Madison.

10.

Cantillon

Vandenbroucke

(2014) Reconciling Work and Poverty Reduction: How Successful Are European Welfare States? Oxford; New York: Oxford University Press.

11.

Cantillon

Marchal

Luigjes

(2017) ‘Decent Incomes for the Poor: Which Role for Europe?’ JCMS: Journal of Common Market Studies 55: 240–56.

12.

Corak

(2013) ‘Income Inequality, Equality of Opportunity, and Intergenerational Mobility’, IZA Discussion Paper Series no. 7520. Bonn: Institute for the Study of Labor.

13.

Cover

(2016) Purchasing Power: Using Wage Statistics with Regional Price Parities to Create a Standard for Comparing Wages across US Areas (Monthly Labor Review). Washington, DC: US Bureau of Labor Statistics.

14.

Daly

(2006) ‘EU Social Policy after Lisbon’, JCMS: Journal of Common Market Studies 44: 461–81.

15.

Darvas

(2016) ‘Some Are More Equal Than Others: New Estimates of Global and Regional Inequality’, Working Paper no. 8. Brussels: Bruegel.

16.

Dauderstädt

Keltek

(2014) ‘Crisis, Austerity and Cohesion: Europe’s Stagnating Inequality’, International Journal of Health Services 45: 25–31.

17.

Delhey

Kohler

(2006) ‘From Nationally Bounded to Pan-European Inequalities? On the Importance of Foreign Countries as Reference Groups’, European Sociological Review 22: 125–40.

18.

Esping-Andersen

(1990) The Three Worlds of Welfare Capitalism. Cambridge: Polity Press.

19.

Eurofound. (2017) Income Inequalities and Employment Patterns in Europe before and after the Great Recession. Brussels: Publications Office of the European Union.

20.

European Commission. (2017a) Commission Sets Out Roadmap for Deepening Europe’s Economic and Monetary Union. Brussels: European Commission.

21.

European Commission. (2017b) European Pillar of Social Rights. Brussels: European Commission.

22.

European Commission. (2017c) Feasibility and Added Value of a European Unemployment Benefits Scheme. Brussels: European Commission.

23.

European Commission. (2017d) Standard Eurobarometer, Autumn 2017: Public Opinion in the European Union. Brussels: European Commission.

24.

European Economic Community. (1992) ‘Council Recommendation of 24 June 1992 on Common Criteria Concerning Sufficient Resources and Social Assistance in Social Protection Systems’, Official Journal 92/441/EEC. Luxembourg: Official Journal of the European Communities.

25.

Eurostat. (2017) Mean and Median Income by Age and Sex – EU-SILC Survey. Luxembourg: Eurostat.

26.

Falk

(2014) ‘Low-Income Assistance Programs: Trends in Federal Spending’, Congressional Research Service (CRS) Report no. 7-5700. Washington, DC: Congressional Research Service.

27.

Ferrera

(2005) The Boundaries of Welfare: European Integration and the New Spatial Politics of Social Protection. Oxford: Oxford University Press.

28.

Ferroni

Mercedes

Payne

(2008) ‘Development Under Conditions of Inequality and Distrust: An Exploration of the Role of Social Capital and Social Cohesion in Latin America’, IFPRI Discussion Paper no. 00777. Washington, DC: International Food Policy Research Institute.

29.

Firebaugh

(1999) ‘Empirics of World Income Inequality’, American Journal of Sociology 104: 1597–630.

30.

Fries-Tersch

Tugran

Bradley

(2016) ‘Annual Report on Intra-EU Labour Mobility’, Final Report, Directorate-General for Employment, Social Affairs, and Inclusion, European Commission, Brussels.

31.

Gerhards

Lengfeld

Häuberer

(2016) ‘Do European Citizens Support the Idea of a European Welfare State? Evidence from a Comparative Survey Conducted in Three EU Member States’, International Sociology 31: 677–700.

32.

Guzman

G.G.

(2017) ‘Household Income: 2016’, American Community Survey Briefs ACSBR/16-02. Washington, DC: U.S. Census Bureau.

33.

Heidenreich

(2016) ‘The Europeanization of Income Inequality before and during the Eurozone Crisis: Inter-, Supra- and Transnational Perspectives’, in Heidenreich

(ed.) Exploring Inequality in Europe: Diverging Income and Employment Opportunities in the Crisis, pp. 22–45. Cheltenham: Edward Elgar Publishing.

34.

Jones

(1993) New Perspectives on the Welfare State in Europe. London: Routledge.

35.

Juncker

J-C.

(2015) Five Presidents’ Report: Completing European Economic and Monetary Union. Brussels: European Commission.

36.

Lichter

D.T.

Ziliak

J.P.

(2017) ‘The Rural-Urban Interface: New Patterns of Spatial Interdependence and Inequality in America’, ANNALS of the American Academy of Political and Social Science 672: 6–25.

37.

Meyer

B.D.

Mittag

(2015) ‘Using Linked Survey and Administrative Data to Better Measure Income: Implications for Poverty, Program Effectiveness and Holes in the Safety Net’, National Bureau of Economic Research Working Paper Series no. 21676. Cambridge, MA: National Bureau of Economic Research.

38.

Milanović

(2011) ‘Global Inequality: From Class to Location, from Proletarians to Migrants’, World Bank Policy Research Working Paper no. 5820. Washington, DC: World Bank.

39.

Milanović

(2016) Global Inequality: A New Approach for the Age of Globalization. Cambridge, MA: The Belknap Press of Harvard University Press.

40.

Nolan

Whelan

C.T.

(2011) Poverty and Deprivation in Europe. Oxford: Oxford University Press.

41.

OECD. (2015) In It Together: Why Less Inequality Benefits All. Paris: OECD.

42.

OECD. (2017) Understanding the Socio-Economic Divide in Europe. Paris: OECD.

43.

Parolin

(2017) ‘Applying Augmented Survey Data to Produce More Accurate, Precise, and Internationally Comparable Estimates of Poverty within the 50 United States’, LIS Working Paper Series no. 696. Luxembourg: LIS Cross-National Data Center.

44.

Saraceno

(2009) ‘Concepts and Practices of Social Citizenship in Europe: The Case of Poverty and Income Support for the Poor’, in Alber

Gilbert

(eds) United in Diversity? Comparing Social Models in Europe and America, pp. 151–75. Oxford: Oxford University Press.

45.

Smeeding

T.M.

Rainwater

(2001) ‘Poverty across States, Nations and Continents’, in Vlemincx

Smeeding

(eds) Child Well-Being, Child Poverty and Child Policy in Modern Nations, pp. 33–74. Bristol: Policy Press.

46.

Todaro

M.P.

(1969) ‘A Model of Labor Migration and Urban Unemployment in Less Developed Countries’, American Economic Review 59: 138–48.

47.

Vandenbroucke

(2017) ‘The Idea of a European Social Union: A Normative Introduction’, in Vandenbroucke

Barnard

De Baere G

(eds) A European Social Union after the Crisis, pp. 3–46. Cambridge: Cambridge University Press.

48.

Whelan

C.T.

Maître

(2009) ‘Europeanization of Inequality and European Reference Groups’, Journal of European Social Policy 19: 117–30.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.27 MB