Inequality and Well-Being in Transition: Linking Experience and Perception to Policy Preferences

Abstract

This chapter reviews the dynamics of economic inequality in the region from the early 1990s until today, distinguishing between actual and perceived inequality dynamics. It then looks at whether inequality matters for the well-being of people, and for their policy preferences. The evidence reviewed shows that the early 1990s saw a sharp increase in inequality, although reliable data from that period is patchy. Over the past 2 decades, on the other hand, income inequality has been declining in the majority of transition economies. These dynamics in observed inequality are somewhat at odds with perceptions of inequality being relatively high, and the chapter reviews some of the possible drivers of the discrepancy, with a focus on inequality of opportunity and considerations of fairness.

1 Introduction

The 1990s inaugurated, in countries of Eastern Europe and the Former Soviet Union (FSU), a period of deep structural transformation across many dimensions of life and society—economic, political, social, and institutional. One of the characteristics typically associated with this period is an increasing level of economic inequality. Branko Milanovic, a leading scholar of economic inequality, associates the first 2 decades of transition with a “dramatic shift in the role of Eastern European/Former Soviet Union (FSU) countries from an ‘inequality reducing’ world middle class to an ‘inequality increasing’ downwardly mobile group” (Milanovic 2005, 44). While this chapter will present a more nuanced picture of inequality dynamics since the fall of the Berlin Wall, the 1990s have indeed heralded increased hardship, downward mobility, and even poverty for many (World Bank 2018), as well as considerable riches for some.

The increasing levels of economic inequality over the past 2 or 3 decades have been observed across a number of industrialized nations, notably in the United States (Atkinson 2015). These dynamics have sparked a renewed debate about the degree and implications of economic inequality both between and within countries. Against this background, this chapter has two main goals. First, it aims to review the dynamics of economic inequality in the region starting with, data allowing, the beginning of transition, and tracing it until today, distinguishing between actual inequality dynamics and inequality perceptions. Second, the chapter aims to provide a review of the literature that tries to answer the question whether inequality (and increases thereof) matters for the well-being of people in the region, and for their policy preferences and choices, namely for their preferences for redistribution, which are connected to their tolerance of economic inequality. The chapter will conclude by pointing to some existing knowledge gaps and fruitful avenues for future research in this area.

2 Economic Inequality in Transition Economies: Magnitude and Dynamics Over the Past 3 Decades

In this section we will review the evolution of economic inequality both across the transition economies and for the region overall vis-à-vis other regions of the world. Before proceeding with this discussion, however, it is important to define the terms that the chapter will be referencing, and to clarify some of the assumptions and measurement issues involved in quantifying economic inequality.

2.1 Measuring Economic Inequality: Concepts, Data, and Methods

While global inequality is today at the forefront of public discourse, it is not always clear what “global inequality” refers to. When economists discuss economic inequality, they typically refer either to the distribution of wealth (or the distribution of ownership of assets, as in the above example from Forbes), or to the distribution of income, as measured by current income received by individuals. In some cases, income inequality may refer interchangeably to inequality in incomes and inequality in consumption, as measured by household expenditures recorded in household surveys. The choice is typically driven by the types of data household surveys collect.¹ These differences matter, as inequality in the space of wealth tends to be greater than inequality in the space of incomes, which, in turn, tends to be greater than inequality in the space of consumption.² In this chapter, when discussing the evolution of economic inequality in transition economies, we will be referring to inequality measured with data from household surveys, either in the space of incomes, or that of consumption, depending on data availability.

Another important consideration, when discussing inequality at the regional, or global, level, is whether we are considering inequalities between countries or between individuals. Milanovic (2005) provides a useful typology of economic inequality, composed of three distinct concepts of inequality. Concept 1 inequality refers to unweighted international inequality, where comparisons are made between countries, represented by their income or GDP per capita, without taking account of their population sizes. Concept 2 inequality refers to population-weighted international inequality, where comparisons are the same as in Concept 1, except that now we acknowledge the fact that changes in per capita income in Russia, for instance, may have a greater impact on the region than changes in Armenia. In other words, Concept 2 inequality accounts for differences in population sizes, but still ignores inequality within each country—every individual from a given country is assigned that country’s per capita income. Finally, Concept 3 inequality refers to inequality among all citizens of a given region (or the world), ranking all these individuals, as captured in representative household surveys, and ignoring which country they come from, to arrive at a measure of regional (or global) inequality.

The discussion here refers to inequality in the space of outcomes (wealth, or incomes, or expenditures). In recent years, the literature has also emphasized inequality in the space of opportunities, and not just outcomes. For instance, a number of studies have relied on pioneering work by John Roemer (Roemer 2000) that aims to distinguish between inequalities due to circumstances individuals have no control over (es gender, ethnicity, the socio-economic status of parents) and inequality due to differences in effort. This literature points out that not all inequalities are equally objectionable, from a normative point of view, and inequalities due to circumstances may have greater claims to be remedied than inequalities due to differences in effort (World Bank 2005; Ferreira et al. 2008). We will return to this discussion later in the chapter.

Finally, the degree of inequality can be measured through different statistics. We will rely here on the most common measure: the Gini index, which measures how much the distribution of income departs from a situation when everyone has exactly the same income, that is, the Gini index would take on a value of zero. In the opposite extreme case, when all the income belongs to one person, and everyone else has zero, the Gini index would take on a value of 1 (or, using commonly used normalization—100). Other commonly used measures include various ratios, such as the 90/10 ratio, or the ration of the income of the richest 10 percent of the population to the poorest 10 percent of the population, or other measures such as the Theil Index, or the Atkinson Index.³

2.2 Dynamics of Economic Inequality in Transition Economies

The collapse of the socialist block has been associated with a significant economic contraction in the early 1990s. Output declined by some 40 percent in the Baltics, by more than 45 percent in Russia and by almost 65 percent in Ukraine (Svejnar 2002). While in some countries, particularly in Central and Eastern Europe, the economic contraction only lasted a few years, the recovery has been slow. By 1999, only Poland and Slovenia had reached the same level of GDP they had in 1989.

This has also translated into a rapid deterioration of household welfare and an increase in poverty in the region. According to Povcalnet data from the World Bank, the share of population of the Europe and Central Asia (ECA) region with incomes below the international $1.9/day poverty line increased from 2.9 percent in 1990 to 7.9 percent in 1999, while the share of population below the Upper Middle Income threshold of $5.5/day increased from roughly a quarter of the population to over 45 percent; during the same period, the share of the middle class population (defined here as having incomes in excess of $15/day) fell from 26 percent to 16 percent (Fig. 27.1).

Fig. 27.1

Evolution of poverty in the ECA region. (Source: Povcalnet, The World Bank)

The increase in poverty in transition economies was associated with a considerable increase in the degree of income inequality. This is the case both at the country and at the individual level. Consider first the Concept 1 inequality across countries in the region, according to the above nomenclature. In order to compute this measure of inequality across countries, we can take data on per capita GDP for each country, expressed, for comparability purposes, in purchasing power parity (PPP) 2011 international dollars.⁴ Only 18 transition economies in the World Development Indicators (WDI) database have GDP per capita estimates going all the way back to 1990, whereas for 2000 onward data is available for all 28 countries in the sample. For this reason, we present two series, the Gini index for the unbalanced sample for the 1990–2018 period, and a Gini index for a stable sample of 18 countries for which GDP estimates are available in every year (Fig. 27.2). It can be seen that the Concept 1 inequality Gini index increases from about 0.29 in 1990 to 0.40 in 1996, if we take the balanced sample, or 0.38 for the full sample in 1994 and then declines to about 0.29 by 2014, remaining roughly constant over the past 5 years.

Fig. 27.2

Concept 1 inequality across transition economies. (Notes: Author’s estimates based on data from the World Bank’s WDI database. GDP per capita expressed in PPP 2011 international USD)

At the regional level, the Gini index of inequality for the Europe and Central Asia (ECA) region (Concept 3 inequality in the above nomenclature) increased from about 25 in 1990 to 35 in 1995—a very large increase that resulted in ECA going from the region with the lowest level of inequality in 1990 to surpassing the level of inequality observed in industrialized nations, as well as in South Asia. It should be noted, that during the period 1990–1995 increasing inequality was a more general phenomenon, observed across all regions with the exception of Middle East and North Africa (MENA). Nevertheless, the magnitude of the increase in inequality in the ECA region stands out even against this secular trend.

As the transition economies stabilized around the mid-1990s, both poverty and inequality began to decline. Supported by strong economic growth throughout the 2000s, that was generally inclusive, as captured by the dynamics of indicators such as the shared prosperity premium, or the rate of income growth of the bottom 40 percent of the population in each country that is in excess of the average rate of income growth,⁵ the overall level of inequality in transition economies declined by 2015 to a level that is again below the level of inequality observed in other regions in the world, or among industrialized economies (Fig. 27.3). The decline in the within-country inequality overtime was also associated with within-region convergence, as evidenced by the decline in Concept 1 inequality among transition economies (Fig. 27.2).

Fig. 27.3

Inequality across regions. (Source: Ferreira et al. forthcoming)

How does this picture of inequality dynamics look if we zoom in from the regional to the country level? Ferreira et al. (forthcoming) examine inequality dynamics around the world during the 1990–2015 period. They find that for the Europe and Central Asia region, out of 5 countries for which comparable data is available for the 1990–2000 period, in 3 inequality went up, in 1 inequality went down, and in another county it remained stable. In the subsequent period (2000–2015) inequality increased in 3 out of 16 countries in the ECA region, and fell in 9 out of 16, remaining stable in 4 countries.

These trends of growing inequality within countries during the 1990–2000 period among transition economies appear to mirror inequality trends at the global level, with inequality increasing in more than half of the countries in the global sample in the first period, and falling in three-quarters of the global sample during the second period (2000–2015). Note that this is in stark contrast with inequality dynamics among industrialized nations, in which inequality at the national level appears to have risen throughout the entire 1990–2015 period (Table 27.1). If one takes a longer view, in the ECA region during 1990–2015 out of 9 countries inequality went up in 7 and fell in 2, whereas overall in the world, the Gini index increased in 32 countries by more than 1 point and fell in 23 countries.

Table 27.1 Dynamics of country-level inequality around the world, 1990–2015

If we abstract from the initial post-transition economic collapse of the early 1990s, and take the 20-year period of 1995–2015, inequality fell in 7 out of the 15 transition economies in the comparable sample, and rose in 4 as well as in Turkey, having remained within +/− 1 point in further 3 countries (Fig. 27.4). Among the FSU states, including in Ukraine, Moldova, Kazakhstan, and Russian Federation inequality fell after an initial increase during 1990–1995, and falling levels of inequality were also observed in Slovenia, Estonia, andPoland. At the same time, in a number of new EU member states such as Hungary, Latvia, Lithuania, and Romania, inequality rose during the same period. At the global level, inequality fell or remained stable in two-thirds of the countries and rose in one-third of the countries.

Fig. 27.4

Change in the Gini coefficient in transition economies, 1995–2015 (percentage points). (Source: Ferreira et al. forthcoming. Note: Inequality measures based on consumption data in Georgia, Kazakhstan, Moldova, Russian Federation, Turkey, and Ukraine, and on income data in all other countries)

What can we conclude from these broad trends in inequality since 1990s? While the necessary data for a full comparison between inequality at the beginning of transition and today is clearly incomplete, available data show that the degree of inequality in transition economies (in the Europe and Central Asia region, as defined by the World Bank), has increased both in the sense of Concept 1 inequality and Concept 3 inequality (from a regional Gini of 27.5 in 1990 to 31.5 in 2015), and within-country inequality also increased in most countries. However, this is largely because of the considerable increase in inequality in the immediate aftermath of the collapse of the Soviet Union. Between 1995 and 2015, inequality in transition economies as a region (both across countries and across individuals) has been on the decline, and this is also the case for country-level inequality dynamics—in two-thirds of the available ECA countries the Gini index fell over this period. This decline is not unique to transition economies, and mirrors inequality trends over the same period across most regions and developing countries. Industrialized nations are an exception in this regard, having registered a rising level of inequality, both as a group, and among most individual countries.

This brings us to the current situation in the region, depicted in Fig. 27.5 using the most recently available data from World Development Indicators (WDI). While the overall level of inequality in Transition Economies may be low compared to other regions in the world, there is also quite a bit of heterogeneity across Transition Economies, with the Gini index ranging from 25 in Slovenia to almost 40 in Serbia. It is difficult to discern from looking at Fig. 27.5 any clear regional patterns; while a number of New EU member states have relatively lower Gini indices, this is not the case in Romania, Bulgaria, or the Baltic states, all of which are in the top 10 countries with the highest Gini indices. Likewise, in the Balkans, Serbia and North Macedonia have high levels of inequality, whereas in Kosovo it is relatively low; and within the Commonwealth of Independent States, the Gini indices range from very low in Belarus to very high in Russia.

Fig. 27.5

Gini index of inequality across countries, circa 2015. (Source: Author’s estimates based on WDI data)

2.3 Perceptions of Inequality Changes

The picture of falling inequality over the 1995–2015 period stands in contrast to the growing degree of concern, in the media, but also in academic and policy discourse, with the level of inequality in general, and with its consequences for individual and societal well-being. Given that much of the research on inequality and top incomes originates primarily in the United States, one could be led to conclude that concerns with respect to inequality, are primarily a problem of (or at least heavily weighted toward) industrialized nations, where inequality has indeed been on the rise over the past 25 years. Do the citizens of transition economies share the perceptions of increasing inequality in the region? Are they concerned about the degree of inequality in the region or in their countries? We turn to these questions now.

We thus investigate perceptions of inequality over time. One data source that does allow for such long comparisons is the International Social Survey Program (ISSP), which had questions trying to elicit the respondents’ perceptions of inequality in their countries in the 1992, 1999, and 2009 survey rounds. Specifically, they were shown diagrams representing five different types of societies, with increasing levels of inequality, that roughly represent a span of the Gini index from 0.20 to 0.42 (Gimpelson and Treisman 2018). The respondents were asked to pick the diagram that best describes their country.

Bussolo et al. (2019) examine inequality perceptions over time in the ISSP sample, by constructing a “net equality perception” measure, which equals to the difference in the share of the population choosing the most equal society and the share of population choosing the most unequal society, such that positive values indicate a higher share of population reporting that they live in a very equal country vis-à-vis the share of population who think that they live in a very unequal country. They find that in transition economies there is a wide belief that societies are unequal, which persists throughout the entire period 1992–2009. For instance, in Bulgaria, where perceptions of inequality are highest, net inequality perception is minus 49 percentage points in 1992, becoming even more negative at minus 67 percentage points in 1999, before falling to minus 60 percentage points in 2009. In other countries in the ISSP sample (Czech Republic, Hungary, Poland, Russia, Slovak Republic, and Slovenia), the net equality perception is still negative, even if somewhat smaller in magnitude. Overtime, perceptions of inequality in transition economies appear to worsen between 1992 and 1999, subsequently improving between 1999 and 2009. In Western European countries, in the other hand, the net equality perception either hover around zero in countries like France, Germany and Spain), or tends to be positive (i.e. more people think they live in an equal society) in countries like Austria, Cyprus, Norway, and Sweden. For instance, in Sweden it increases from 24 to 31 between 1999 and 2009. Dynamically, inequality perceptions in Western European countries appear to worsen between 1999 and 2009, in contrast to the dynamics in Eastern Europe.

Another data source that allows us to look at inequality perceptions for a larger set of transition economies is the Life in Transition survey (LiTS), which has three waves of data collected in 2006, 2010, and 2016. In the LiTS, the respondents are asked whether they agree with the statement “the gap between the rich and the poor in this country should be reduced.” This is somewhat different from the examination of inequality perceptions in the ISSP, because the question compounds the positive assessment of the perceived degree of inequality in a country with a normative assessment with respect to the need for reducing the perceived level of inequality. Figure 27.6 plots, for each country and wave of the LiTS, the share of adults who either agree or strongly agree with the statement that the gap between the rich and the poor should be reduced, for the 2006 and 2010 survey rounds, and separately for the 2010 and 2016 rounds. Several conclusions can be drawn from this graph. First, in 2010, the share of adults who deem the gap between the rich and the poor in their country to be too large is substantial, according to the most recent data comprising of at least half of the population in all countries in the sample, and in 23 out of the 32 countries in the sample, more than three-quarters of adults would like to see a smaller gap between rich and poor. Second, the perception of inequality, as captured by this survey question, was even higher, on average, in 2006, having subsequently fallen in the 2010 survey round (left panel), and then having increased again slightly in 2016 (right panel). Third, the fall in the share or respondents who think that the gap between the rich and the poor should be reduced between 2006 and 2016 rounds of the survey is consistent with the observed decrease in the degree of economic inequality in the region. It is surprising, to some degree, to observe a lower preference for redistribution in 2010 in the aftermath of the financial crisis, although it is plausible that the effect of the financial crisis would have taken some time to translate into updated beliefs about inequality, and certainly the greater preference of redistribution in the latest 2016 round of the LiTS would be consistent with the very difficult and drawn out recovery and subdued growth in the aftermath of the crisis.

Fig. 27.6

Attitudes toward inequality across countries, 2006, 2010, and 2016. (Notes: The graphs show the share of adults in each country and each survey round who either agree or strongly agree with the statement “The gap between the rich and the poor should be reduced.” The line in each panel is the 45-degree line. Source: Author’s estimates based on LiTS data)

Inequality perceptions can also be compared to actual inequality statistics. Figure 27.7 plots the share of adults who either agree or strongly agree with the statement that the gap between the rich and the poor should be reduced against the Gini index of inequality, separately for the 2010 round of the LiTS against inequality in 2010 (left panel) and from the latest round of the LiTS against the latest available inequality statistics from 2015 (right panel). Inequality perceptions and realized inequality are positively correlated, both in 2010 and in 2016, but this correlation is not very strong. A given level of income inequality, as measured by the Gini coefficient, can be associated with very different average perceptions of inequality across countries, and likewise, similar perceptions of inequality can be observed in countries with rather different values of Gini indices. Bussolo et al. (2019) similarly confirm that there exists a weak correlation between perceptions of inequality based on the ISSP Social Inequality dataset and actual Gini indices of inequality for the same country; they also find a weak correlation between inequality and preferences for redistribution based on the agreements or disagreements with the statement “it is the responsibility of the government to reduce income differences between people with high incomes and those with low incomes.” At the same time, they find a strong correlation in the data between perceived levels of inequality and demands for redistribution.

Fig. 27.7

Inequality perceptions and actual levels of inequality in transition economies in 2010 and 2016. (Notes: Inequality perceptions from the LiTS 2010 and 2016 survey rounds. Gini coefficients from the WDI database. Inequality perceptions based on the share of adults in each country and each survey round who either agree or strongly agree with the statement “The gap between the rich and the poor should be reduced.”)

Thus, for the region overall, there appears to be a discrepancy between widespread perceptions of inequality being very high, and empirical evidence suggesting that (i) the level of inequality in transition economies is not very high, on average, when compared to other regions of the world, and (ii) the level of inequality has been declining over the past 20 years after the initial increase during 1990–1995. However, these discrepancies between actual inequality and perceptions thereof are not specific to transition economies only. Gimpelson and Treisman (2018) find, across a number of datasets and countries that respondents predict poorly (slightly better than by chance) both the levels of inequality in their countries, as well as the trends in inequality, or other distributional statistics such as the top one percent’s share of wealth, average salaries nationwide or for specific jobs, or the country’s current poverty rate. When presented with diagrams representing different societies that differ by their degree of inequality, only 29 percent worldwide choose the diagram that most closely resembles the post-tax-and-transfers Gini of their countries. In a number of European countries (Estonia, Slovakia, Croatia, Hungary, andUkraine) over 90 percent of respondents chose the wrong diagram as representative of their country (ibid.).

What may cause such misalignment between actual inequality and perceived inequality? Many factors could be at play. People may simply not know, and Gimpelson and Treisman (2018) present compelling evidence in favor of partial knowledge of inequality levels and dynamics. Other studies have similarly found perceptions of social mobility to differ from actual experience of social mobility (Alesina et al. 2018; Narayan et al. 2018). Alesina et al. (2018) find evidence that in the United States (in Europe) perceptions overestimate (underestimate) mobility vis-à-vis mobility measures that can be observed empirically from the data.

With actual inequality levels not readily observable, individuals can rely on a number of other macro and micro variables to infer the extent of inequality in their countries. Bussolo et al. (2018) find that macroeconomic variables such as the Gini coefficient, the unemployment rate, the poverty rate, and government expenditures on education together explain a quarter of the total variation in individual perceptions of inequality. In addition to these macro variables, individual circumstances such as employment status, also influence inequality perceptions. For instance, perceptions of inequality correlate with the level of education expenditures among the employed, but not among those who are unemployed; higher poverty rates correlate with inequality perceptions among the 24–34 years age group, but not among those who are 45 years of age or older; objective measures of inequality correlate with perceptions of inequality among those who experienced perceived downward mobility with respect to their fathers, but not among those who think that their job status is the same or better than that of their fathers (Bussolo et al. 2018).

Beyond partial knowledge, and especially when respondents are asked whether the degree of inequality is too high or needs to be reduced, their perceptions are also informed by some reference points that frame these comparative statements. In the case of Transition Economies, inequality increased following the collapse of the Soviet Union, and the perceived greater degree of equality pre-1989 may still loom large in respondent’s minds. Indeed, studies have found that some 20 years after the beginning of transition, the pre-transition level of well-being is still an important determinant of subjective well-being today (Senik 2009; Cojocaru 2014a). It may also be the case that individuals make inferences on national-level inequality, which is hard to observe (Senik 2004), from local-level inequality. Frank and Levine (2007) argue that “the within-reference group level of inequality for an individual is likely to correspond more closely to the degree of inequality in the city in which [the person] lives than to the degree of inequality in his home country” (Frank and Levine 2007, 13). Cojocaru (2016) finds, based on data from several transition economies, that relative status perceptions are more salient for well-being at the local (city) level, when compared to regional, or national levels.

Another important consideration is with respect to the evaluative space in inequality perceptions. While inequality statistics presented in this chapter, or inequality perceptions that are usually queried in surveys, refer to (usually implicitly, rather than explicitly) inequality measured in the space of disposable incomes, this need not be the phenomenon that respondents have in mind when they answer. Individuals could be basing their perceptions, on, for instance, market income inequalities in the space of earnings, not accounting (partially or fully) for the redistributive effects of various taxes and transfers. To get an insight into this we can draw on a number of recently completed studies in the region that follow the so-called Commitment to Equity methodology (Lustig and Higgins 2013), which allows us to measure the degree of inequality across a number of income concepts, including: (i) market income, or household income before any tax-benefit interventions; (ii) market income plus pensions, which includes contributory pensions and exclude pension contributions; (iii) disposable income, which starts with market income plus pensions and then subtracts direct taxes and social insurance contribution and adds direct cash transfers; (iv) consumable income, or disposable income minus indirect taxes plus indirect subsidies; and finally (v) final income, which adds to the consumable income in-kind transfers such as public education and healthcare expenditures. The fiscal system does appear to have a significant redistributive effect in Transition Economies for which estimates are available. Contributory pensions alone reduce the Gini index by 10 percentage points or more vis-à-vis the inequality in the space of market incomes (Fig. 27.8). Accounting for indirect taxes and transfers, and the imputed value of services such as publicly provided health and education reduces inequality even further (Cojocaru et al. 2019).

Fig. 27.8

Gini indices across income concepts. (Source: Cojocaruet al. 2019)

Yet, however large the differences between pre-fiscal and post-fiscal inequality may be, it’s not clear how much these differences drive the discrepancy between actual and perceived inequality. In particular, Gimpelson and Treisman (2018) test whether respondents are good at estimating both pre-tax inequality and post-tax inequality in their countries, and find that their perceptions align well with neither of the two.

Another possibility is that they may not be thinking of inequality in the space of outcomes such as income (whether disposable or otherwise) at all; rather, they could be thinking of inequality of opportunity in some broad sense. Bussolo et al. (2018) show that while inequality in transition economies has been relatively stable in recent years, there is, at the same time, evidence of increasing concentration of wealth, increasing labor market polarization characterized by a hollowing out of the jobs in the middle of the distribution, with intensive and routine tasks, and an increasing generational divide, with young cohorts losing ground. In particular, they find that younger age cohorts are facing higher income inequality at every point of the life cycle relative to older generations. Narayan et al. (2018) similarly find intergenerational mobility in Europe and Central Asia to be worsening for the age cohorts growing up following the collapse of the Soviet Union. In other words, irrespective of the dynamics of income inequality as measured by national Gini indices or other statistics describing the distribution of incomes, perceptions of inequality may be driven more by these considerations of increasingly uneven opportunities for success in the region.

3 Inequality and Welfare

How relevant are these changes in inequality in the region, both actual and perceived, for individual well-being? Inequality measures such as the Gini coefficient, are merely statistical representations of the (usually, disposable) income distribution within a given population. Are high values of these inequality statistics intrinsically nefarious when it comes to individual welfare? Differences in outcomes (such as incomes) may be driven by many factors, and, when they are related to factors such as differences in effort across individuals, it is not clear whether such differences should be viewed as being detrimental to individual, or social, welfare. Indeed, the philosophical literature on inequality, has argued that inequalities that matter for justice should be inequalities in the space of resources (Rawls 1971; Dworkin 1981), opportunity for welfare (Arneson 1989), access to advantage (Cohen 1989), opportunities for a good life (Arneson 2000), capabilities (Sen 1980), or opportunities (Roemer 2000). These normatively-informed concepts of inequality are oftentimes difficult to measure, however, or disentangle from overall inequalities in outcomes. With these caveats in mind, the discussion of the links between inequality and well-being can be structured in terms of linkages at the macro level, and at the individual level. We turn our attention first to macro linkages.

3.1 Macro Linkages Between Inequality and Welfare

At the macro level, to look at the effect of inequality on economic outcomes and objective well-being, one needs to look first and foremost to the literature on economic inequality and growth, as economic growth is widely recognized as the key engine behind poverty reduction and improvements in living standards. Existing empirical evidence suggests that, on balance, higher inequality has a retardant effect on economic growth, although there a lot of heterogeneity in the literature, with both negative and positive associations between inequality and growth, depending on definitions, countries included in the sample and methods used (Boushey and Price 2014). De Dominicis et al. (2006), in an early meta-synthesis of the literature on inequality and growth find that among the various empirical estimates in the 22 studies they review, the correlation between inequality and growth was negative in roughly 40 percent of the values, close to zero in another 40 percent, and positive in about 20 percent of the estimates. The relationship between inequality and growth tends to be more negative in low income countries. Similarly, inequality at the bottom of the income distribution tend to be more pernicious for growth (Voitchovsky 2005). There is also some evidence that inequality in wealth is negatively associated with growth (Deininger and Squire 1997; Birdsall and Londono 1998). A number of studies also find higher levels of inequality to be associated with shorter duration of growth (Berg and Ostry 2011; Ostry et al. 2014), although there are methodological questions with respect to these findings (Kraay 2015).

It was noted earlier that some of the discrepancy between observed inequality and perceived inequality may be due to the latter being driven by dynamics of inequality of opportunity, more so than of inequality of outcomes. Some of the most recent literature on the links between inequality and growth also confirms this. Marrero and Rodriguez (2010), using state-level data from the United States, find that while there is no statistical relationship between growth and inequality of outcomes, there is a negative relationship between inequality of opportunity⁶ and growth, and a positive relationship between inequality resulting from differences in effort and growth. Aiyar and Ebeke (2019) similarly find, in a cross-country setting, that the relationship between inequality and growth is more negative when intergenerational social mobility is lower, a situation that tends to be indicative of greater inequality of opportunity (Narayan et al. 2018).

In addition to having a retarding effect on growth, inequality has also been found to mediate the relationship between growth and poverty reduction. The ability of economic growth to reduce poverty, or the growth poverty elasticity, has been found to be higher when inequality is lower (Hanmer and Naschold 2000; Ravallion 2001). A recent World Bank study notes that the goal of eliminating extreme poverty by 2030 will not be reached with distributionally neutral growth, especially in a period of overall slowdown in economic growth. Reducing within-country inequality, especially in countries with large concentrations of the poor, is estimated to be essential in eliminating global poverty (World Bank 2016).

Inequality has also been found to be strongly correlated with socio-economic mobility, as shown in the now famous Great Gatsby curve, which shows a strong empirical association between higher levels of inequality and greater intergenerational immobility across high income countries (Corak 2017), but also across a much larger set of developing countries (Narayan et al. 2018). This means that high levels of inequality can lead to inequality traps: with children born to parents at the bottom of the income distribution being much more likely to remain at the bottom of the income distribution themselves as adults. While by international comparisons, the association between parent and children’s outcomes (intergenerational persistence) in transition economies is relatively low, it is worrisome that for the latest two cohorts in the data (children born between 1970 and 1990), and thus the generations that grew up and reached adulthood in the aftermath of the collapse of the Soviet Union, the degree of persistence has increased and is currently almost on par with levels of mobility recorded in lower middle income countries, and much below the levels of intergenerational mobility in high income countries (Fig. 27.9). In other words, the transition period has been associated with a deterioration of social mobility in transition economies, when in other parts of the world the recent trends have been in the opposite direction—that of generally improving mobility over time, with the exception of the 1970s cohort.

Fig. 27.9

Intergenerational mobility across 1940–1980 birth cohorts. (Source: Narayan et al. (2018) and author’s estimates based on the Global Database of Intergenerational Mobility (GDIM) database. Notes: Intergenerational persistence measures the degree of association between the education of parents and children, using the coefficient from regressions of children’s years of education on the education of their parents. Higher values of this regression coefficient indicate greater persistence across education, and hence lower relative mobility)

Fiscal policies play an important role in reducing inequality and promoting social mobility. For instance, transfers can aid low income families in ways that improves long-run outcomes of children. Likewise, taxation can influence the amount of resources that can be passed from one generation to the next (Narayan et al. 2018). The discussion in the previous section highlighted the fact that fiscal systems in Transition Economies can have an important redistributive effect—the degree of post-fiscal inequality can be much lower than pre-fiscal inequality. They also help reduce poverty—in countries like Belarus, Moldova, Russia, and Ukraine, the fiscal incidence analysis has shown the poverty rates based on the comparison of a given poverty line with a welfare aggregate based on disposable income to be considerably lower than if market income is used as a welfare aggregate (Cojocaru et al. 2019).

It is also widely understood that the mediating pathway from inequality of outcomes to socio-economic mobility is that of inequality of opportunity. Narayan et al. (2018) confirm empirically the importance of parental characteristics (other than education) for explaining income persistence across generations. (Brunori et al. 2013) argue that the reasons why higher inequality makes intergenerational mobility harder are likely related to the fact that “opportunities for economic advancement are more unequally distributed among children.” Data from the latest round of the Life in Transition Survey (LiTS III) indeed show a positive association between overall income inequality in transition economies, and the extent of inequality of opportunity in these countries (EBRD 2016). Moreover, the data reveal that the extent of inequality of opportunity in transition economies is higher, on average, than in Western European countries like Germany or Italy; the extent of inequality of opportunity is also generally higher in countries of the Former Soviet Union, than in the new EU Member states or countries in the Western Balkans, although this is not universally so, with relatively high levels of inequality of opportunity in EU countries such as Bulgaria, Estonia, Latvia, and Romania. Roughly one-third of the overall income inequality is found to be attributable to inequality of opportunity, defined, following Roemer (2000), in terms of circumstance types based on characteristics such as urban/rural birthplace, parental education, gender, ethnic minority/majority status and parents’ membership in the communist party. Among these circumstances, differences in parental background are the most prominent—this factor account for more than half of the overall inequality of opportunity in a third of the transition economies; gender is the second most important factor, accounting for between a quarter and half of overall inequality of opportunity in most countries (EBRD 2016).

These results should not lead one to conclude that inequality in transition economies is, for the most part, fair, on account of only one-third of it being attributed to unequal opportunities. The authors note that these estimates represent a lower bound because (i) the list of circumstances considered is not exhaustive, and if other circumstances were available in the data, the share of inequality of opportunity in overall inequality would likely be larger (in any case, no smaller), and (ii) estimates based on income underestimate true differences in opportunities because they exclude people who are out of the labor force (EBRD 2016). Indeed, the same data suggests that three-quarters of adults deem connections to be at least moderately important (more than half think them very important or essential) to get a good job in the government sector. More than two-thirds think connections are at least moderately important to get a good job in the private sector, and these perceptions of inequality of opportunity are associated with expectations of future socio-economic mobility (Cojocaru 2019).

3.2 Micro Linkages Between Inequality and Welfare

A number of experimental studies show that individuals have a preference for equity, in the sense of preferring equitable outcomes, engaging in cooperation and having strong other-regarding preferences (Thaler 1988; Camerer and Thaler 1995; Fehr and Gachter 2000; Fehr et al. 1997; Fehr and Schmidt 2006 and references therein). This aversion to inequality would imply that inequality would have a direct negative effect on individual’s utility. This is what Clark and D’Ambrosio (2015) call the normative view, whereby individuals make judgments on inequality within a given reference group irrespective of their relative position in the reference group. Following this normative view, a growing literature takes subjective well-being, which is increasingly commonly reported in individual and household surveys, as a proxy for individual’s utility to answer the question whether inequality has an impact—either positive or negative—on individual well-being. The benefit of this approach, when trying to examine the welfare effects of inequality, derives from the fact that it is difficult to answer this question by way of reliance on revealed choice analysis, as data on choices between environments with varying degrees of inequality are seldom available. On the other hand, it is relatively easy to ask the respondent directly about their attitudes about inequality, and about their individual well-being, and then use statistical techniques to probe the association between the two.

Several studies have investigated empirically the link between inequality and life satisfaction in transition economies. In one of the earliest papers, Sanfey and Teksoz (2007), relying on data from the World Values Survey from the first four survey rounds (covering the period 1981–2002) investigate the link between life satisfaction and inequality, conditional on other correlates of life satisfaction, separately for transition and non-transition countries, and find inequality (measured by the Gini index) to have a negative association with life satisfaction in transition economies, and a positive association in non-transition economies. The authors suggest that the negative correlation between inequality and well-being in Transition Economies is on account of the lingering dislike of inequality that was characteristic of the socialist systems.

Guriev and Zhuravskaya (2009), also based on data from the World Values Survey (waves 3 and 4), investigate the determinants of the “unhappiness gap”, or the lower level of happiness reported by respondents from Transition economies relative to respondents from non-transition economies. They find that accounting for income inequality measured by the country’s Gini coefficient reduces the gap between life satisfaction in transition and non-transition economies; in other words, inequality is one of the contributors, alongside income volatility, deterioration of public goods, and the depreciation of human capital, to lower levels of life satisfaction in transition economies, consistent with the findings of Sanfey and Teksoz (2007). In a subsequent study, Guriev and Melnikov (2018) find that this happiness gap, still present in the data in 2010, disappears by 2016, consistent with the predictions in Guriev and Zhuravskaya (2009), a result confirmed both by the Life in Transition Survey data and the Gallup World Poll data (see also Nikolova 2016). The follow-up study does not investigate the contribution of inequality to the closing of the happiness gap, but it should be noted that this gap closes during a period when inequality in transition economies is falling, in contrast to the inequality dynamics in industrialized countries, and some other regions of the world, although certainly this need not be the only, or even the main, difference between transition economies and other countries over the 2010–2016 time period.

Yet, not all studies find the relationship between inequality and individual well-being to be universally negative. Berg and Veenhoven (2010), in one of the largest cross-country studies of the relationship between inequality and life satisfaction across 119 nations find no statistically significant relationship between inequality and subjective well-being. A well-known study by Alesina et al. (2004) similarly find no relationship between inequality and reported well-being in the United States (across states), although they find a negative relationship in a sample of 12 European countries over the period 1975–1992, although Berg and Veenhoven (2010) claim that this negative relationship may be an artifact of the sample in Alesina et al. (2004). Clark and D’Ambrosio (2015) provide a comprehensive review of the literature on inequality and well-being, and find that among the 29 studies (either cross-country or single-country) that estimate the relationship between well-being and the Gini index of inequality (or some other statistical measure of inequality), 8 studies find no statistically significant association between inequality and well-being, 14 studies find a negative correlation, and 7 find a positive relationship.

Clark and D’Ambrosio (2015) conjecture that such a heterogeneity of estimates with respect to the relationship between inequality and well-being is due, in part, to methodological issues. In particular, not all studies that they examine in their review adequately control for relative income when estimating a conditional relationship between inequality and well-being, which compounds what they call the comparative view, or the perceived importance of one’s own position in the income distribution relative to others, and the normative view, which reflects one’s attitude toward inequality irrespective of one’s position in the income distribution relative to others. Other methodological caveats include the fact that correlations between well-being inequality indices assume that (i) the degree of inequality as captured by the Gini index, for instance, is observable to the respondent; and (ii) the inequality measure used in the regressions is estimated over a relevant comparison group (which is unobserved). If either of these (largely untestable) assumptions are violated, it is not clear why one would expect to observe either a positive or a negative correlation between inequality and well-being, and even when one is observed in the data, how it may be interpreted. In this regard, it is not clear that national-level inequality is either observable to individuals, or is the relevant reference point. For instance, Kuhn et al. (2011) find shocks to relative status based on winning a lottery to be salient within very local postcode areas in the Netherlands (comprising roughly 20 households), and even then, restricted in large part to a household’s nearest neighbors. Furthermore, since inequality indices such as the Gini index are constant within the groups for which they are estimated, a relationship between individual well-being and group inequality is, empirically, a relationship between mean satisfaction in a group, and that group’s inequality. If an individual’s utility function is concave in income, then there will be a negative relationship between mean well-being in a group and the group’s inequality, even if inequality has no effect on well-being at the level of the individual (Atkinson 1970).

Several studies probe these implicit assumptions. Senik (2004), using data from the Russian Longitudinal Monitoring Survey (RLMS) for the period 1994–2000, find no relationship between national-level inequality and life satisfaction. To address the issues of observability, the authors also compute inequality indices at the regional level, and the Primary Sampling Unit (PSU), although these too are not significantly correlated with individual well-being. Cojocaru (2014a), using data from the 2006 round of the Life in Transition survey for all transition economies with the exception of Turkmenistan, finds that while there is no significant relationship between PSU-level inequality and individual well-being, this is a methodological artifact of trying to make inference on individual behavior from group outcomes. When an alternative specification, based on the Fehr and Schmidt (1999) specification of inequality-averse preferences is estimated, a significant and negative relationship between individual well-being and the Yitzhaki index of relative deprivation (Yitzhaki 1979) is found, conditional on own income and reference group income, which is indicative of aversion to inequality both in the New EU Member States, and in the non-EU countries in the LiTS sample.

Another strand of literature, described by Clark and D’Ambrosio (2015) as the comparative view, considers that individuals rely on their position in some reference group relative to others for purposes of self-appraisal. There is now a large literature (comprehensively reviewed by Clark and D’Ambrosio 2015) that shows that relative comparisons matter for self-evaluation. We do not review this literature here in part because relative status may be important to individuals in ways that need not relate directly to economic inequality in the sense described in this chapter; the implications of relative status considerations for individual well-being are a related but separate issue analytically. There are, however, two key areas of overlap where relative status concerns and inequality aversion are directly related. First, a number of studies investigate the importance of relative status concerns by relying on Yitzhaki’s relative deprivation index (Yitzhaki 1979; Deaton 2001; Eibner and Evans 2005; D’Ambrosio and Frick 2007) and find a negative relationship between relative deprivation defined this way and individual well-being or health outcomes; Cojocaru (2014a) confirms this negative relationship in the case of Transition Economies as well. Hey and Lambert (1980) establish formally that if there are two distributions where one Lorenz dominates the other, such that the latter is more unequal, there will be more relative deprivation in the Yitzhaki sense at every level of income in the more unequal distribution. In other words, a negative well-being effect of relative deprivation can, although does not have to, be indicative of aversion to inequality and thus a negative relationship between the degree of inequality and individual well-being.

Second, relative status can serve as a mediator for the relationship between inequality and well-being. Delhey and Dragolov (2014), based on data from the European Quality of Life Survey, examine three mediating factors—generalized trust, status anxiety, and conflict—as mediating channels between inequality and well-being. They find that among Western European countries the main reasons why individuals dislike inequality is generalized distrust; neither status anxiety nor perceived conflicts among rich and poor or management and workers appear to mediate the effect of inequality on well-being. In transition economies, on the other hand, the key mediating pathway is status anxiety, with sufficient mediating power to reduce the contextual effect of inequality on well-being to statistical insignificance. Schneider (2019) similarly finds evidence of social status mediating the relationship between inequality and life satisfaction in a sample of mostly Western European countries—individuals in more equal societies report, on average, higher social status, and subjective social status fully explains the link between inequality and life satisfaction. Furthermore, Schneider (2019) also finds that inequality affects the strength of the link between status perceptions and well-being—in countries with higher levels of inequality social status is more important for life satisfaction.

The previous section has highlighted the important distinction between inequality of outcomes and inequality of opportunity, and that disparities driven by circumstance, effort and luck may warrant different normative judgments. The nature of the process that generates the distribution of income in a society will affect the link between inequality and well-being, in both the comparative and the normative views described above. For instance, Grosfeld and Senik (2010) find inequality to be positively associated with subjective evaluations of the economic situation in Poland in the early years of transition (1992–1996), the relationship between inequality and well-being turning negative in the second half of the transition period (1997–2005). They suggest that the evidence is consistent with the Hirschman tunnel effect (Hirschman and Rothschild 1973), whereby at the beginning of transition greater differentiation of incomes was perceived as a positive signal of greater opportunities, whereas over time, growing inequality amid unfulfilled reform expectations has led to disappointment and skepticism with respect to the legitimacy of the enrichment of reform winners, and economic inequality began to be perceived as being unfair. Senik (2009) uncovers systematic differences between “Old” Europe and “New” Europe, with inequality aversion being less prominent in transition economies (and in the United States) vis-à-vis Western European countries, which they interpret as being consistent with the evidence that reference group income is positively correlated with well-being in Transition Economies (in line with the Hirschman tunnel effect) and negatively correlated in western Europe. Like Grosfeld and Senik (2010), they interpret inequality in the 1990s in transition economies as a signal of opportunity, whereas in Western Europe inequality aversion is likely driven by perceptions of fairness.

Cojocaru (2014a) finds that inequality aversion in transition economies is not intrinsic, but rather driven by perceptions of fairness. In particular, inequality averse preferences are observed among those who think that need in society is due to injustice, but not among those who think that it is due to laziness. Similarly, those who think that effort and skills were key to success prior to 1989 whereas now success is driven by connections (and therefore would deem inequality to be unfair) are found to be averse to inequality, but not those who think that connections were key prior to 1989 but effort and skills are key to success now (and therefore would deem inequality to be fair) do not exhibit inequality averse preferences.

3.3 Inequality and Policy Preferences

If respondents hold inequality-averse preferences, or if status considerations play a role in determining one’s well-being is a way that is being amplified by economic inequality, then we would expect inequality to be associated with demands to reduce income disparities, by way of redistributive policies or otherwise. Alesina and Angeletos (2005) show that in theory, when individuals are averse to unfair social outcomes, in societies with a greater degree of unfairness generating the income distribution, aggressive redistribution will be desirable, because anticipation of high taxes makes it optimal to exert low effort, making a high share of the heterogeneity in the income distribution the result of luck, which makes redistribution optimal ex post. Note that inequality can affect preferences for redistribution even when individuals are entirely self-regarding. In the canonical Meltzer and Richard (1981) model where redistribution policy consists of a flat tax and an equal lump sum transfer, the degree of redistribution preferred by the pivotal median voter will be a function of the degree of inequality, as measured by the distance between median and average income (see Bussolo et al. 2018 for a discussion), however, there is little empirical evidence to support the Meltzer-Richard model (see Alesina and Giuliano 2009 for a detailed discussion). Inequality is also not the only, or even the key, determinant of redistributive preferences, and a review of this literature is beyond the scope of this chapter (see Clark and D’Ambrosio 2015 for a review). We focus, instead, on the links between inequality and preferences for redistribution, and within this area, we restrict our attention to the studies focusing on transition economies.

Alesina and Giuliano (2009) provide a useful taxonomy for the various channels through which inequality can affect preferences for redistribution First, inequality can enter in the utility function indirectly, such that individuals do not care about inequality per se, but rather about its effect on their consumption flow. In particular, inequality could lead to sub-optimal education levels, or to higher crime rates, such that the affluent would support redistribution for the externalities that reducing inequality would produce. Second, inequality could enter directly individuals’ utility functions; for instance, individuals could hold libertarian, or communist, or Rawlsian views with respect to “social justice,” which would lead them to support different levels of redistribution. Finally, individuals’ views about inequality can be intertwined with some sense of fairness, such that their demands for redistribution would be informed by the extent to which inequality may be perceived to be unfair or not.

It should be noted that the differences between actual inequality and perceived inequality in transition economies that were noted earlier will matter when it comes to preferences for redistribution. When the two diverge, it is perceptions of inequality that will determine one’s beliefs and policy preferences, irrespective of the actual degree of inequality (or social mobility) even though perceptions are clearly informed by reality at least to some degree. The United States is a well-known example of a country where preferences for redistribution tend to be relatively low, on account of perceived high social mobility, even though empirical evidence shows that social mobility in practice is quite low, and lower than in many European countries where citizens prefer, on average, a higher degree of income redistribution. (Gimpelson and Treisman 2018) show that both within and across countries, the relationship between actual inequality and demands for redistribution was tenuous at best, whereas perceptions of inequality closely track demand for government redistribution and reported class conflict. Bussolo et al. (2019), based on ISSP data, similarly find that preferences for redistribution are not correlated with the actual Gini index of inequality, but are correlated with perceptions of inequality, where the perceptions are based on the net equality perception concept that was described above.

With this caveat in mind, several of these pathways appear to be borne out in the empirical data from post-socialist countries. For instance, Bussolo et al. (2019) finds support for redistribution concerns being driven by inequality entering directly into the utility function. Those on the left of the political spectrum are more insensitive to their perceptions of inequality when forming their demand for redistribution—they have a strong demand for redistribution regardless of the inequality level they perceive, whereas those who are right-leaning vary their demand for redistribution as their perceptions of inequality change. Cojocaru (2014b), using data from the 2010 round of the LiTS, finds evidence that fairness considerations (measured through perceptions of whether success is determined by informal connections, or need in society is determined by injustice) are important determinants of redistributive preferences in EU countries, but less so in non-EU transition economies, and conjectures that one possible explanation could be that outside of the European Unioninequality of opportunity is more widespread, leading people to adapt to it, which in turn attenuates the link between perceptions of unfairness and preferences for redistribution. Cojocaru (2019) finds, on the other hand, that using the same data from the 2010 LITS, perceived inequality of opportunity⁷ is an important determinant of preferences for redistribution; among those who perceive a greater degree of inequality of opportunity, not having connection is associated with greater demands for redistribution, and, among those without connections, perceptions of greater inequality of opportunity also heighted preferences for redistribution.⁸

4 Concluding Remarks

The end of 2019 marks the 30th anniversary of the fall of the Berlin Wall. The social, economic, and political transformation that has taken place over the past 30 years in countries of Eastern Europe and the Former Soviet Union has been monumental. One of the commonly invoked markers of the post-socialist transition is the considerable increase in the income inequality in the countries of this region. The evidence presented in this chapter paints a more nuanced picture. Inequality did increase considerably in the first half of the 1990s, but has since experienced a steady decline, such that by 2015 inequality in the ECA region was much lower than in 1995, and also lower than in all other regions in the world, including industrialized countries. Over the past 15 years inequality has been declining in most countries in the region, for which the necessary survey data are available.

There is, at the same time, a stark difference between relatively low and declining levels of observed inequality in transition economies, and the widely held beliefs that the region has grown to be very unequal, and preferences for the level of inequality to be lower. Research has shown that this is due, in part, to the fact that individuals have very little knowledge of the actual level of inequality in their countries, as measured using data from household surveys. But this lack of knowledge is only a partial explanation. Recent literature also documents a growing degree of inequality of opportunity in transition economies, declining intergenerational mobility for the most recent birth cohorts, and a growing degree of polarization in the labor market. Younger cohorts are also being faced with higher levels of inequality through their life cycle compared to older cohorts. A large majority of adults believe that it is difficult to gain access to key opportunities in life, such as a good job or university education, without informal connections.

These perceptions matter. If opportunities are perceived to be unequally distributed, it can lead individuals at the bottom of the income distribution to under-invest in human capital and form lower aspirations for the future, thus perpetuating inequality traps across generations (see Narayan et al. 2018 and references therein for a detailed discussion). Bussolo et al. (2018) also document how the perceptions of the shrinking level of equity in the regions are putting fissures in the existing social contract in transition economies, through (i) a growing polarization in voting; and (ii) declining trust in institutions. Winkler (2019), using data from 25 European countries, including a number of transition economies, for the period 2002–2014, also finds that a 5-point increase in the Gini index of local inequality increases the likelihood of a voter supporting either a far-left or a far-right party by 4 percentage points.

There is also much that we still do not know about the relationship between inequality and welfare in this region. One key contribution to this literature would be to highlight to a much greater extent the heterogeneity across the countries in the region by undertaking comparative analysis. This chapter has reviewed, to the extent that the existing literature allows for it, the differences across transition economies, and in some cases between transition economies and western European countries, with respect to the extent of overall inequality, inequality of opportunity, and perceptions of inequality. There is also some evidence from the existing literature that different mechanisms may be mediating the link between inequality and well-being and between inequality and preferences for redistribution, in transition vs non-transition countries in Europe, or in New EU member states vs non-EU transition economies in the Balkans and the FSU. However, most of the studies reviewed in this chapter are based on cross-sectional cross-country datasets, even in cases such as Grosfeld and Senik (2010) that look at within-country dynamics overtime, and as such, our ability to provide a high degree of differentiation from data of this kind is necessarily limited, as is our ability to adequately address issues pertaining to individual heterogeneity.

One of the main conclusions of this chapter is that people in the region perceive inequality to be too high, and would like income disparities to be lower, or perhaps more accurately, they would like opportunities to be distributed more equally. This knowledge was made possible by the increasing number of studies relying on subjective well-being data and allow us to make inference on the relationship between inequality and individual well-being. However, to better understand the implications of increasing inequality of opportunity, and falling intergenerational mobility in transition economies, and in order to help policymakers, we would need to move beyond subjective well-being data and obtain more direct evidence on the effects of inequality (and inequality of opportunity) on key decisions and actions that individuals make; such as (i) their investments in human capital, (ii) their engagement in the labor market, (iii) their policy preferences beyond a general preference for reducing the gap between rich and poor, in particular related to the current policy debates on issues such as the best ways to protect vulnerable households that lose out from the changing labor markets or which taxes to deploy to reduce inequality.