Part I
Definitions and approaches
1
Not just slicing the pie
The need for a broader approach to economic inequality
Pasquale De Muro
Introduction
In 1958, in the first edition of The Affluent Society, John K. Galbraith wrote that âfew things are more evident in modern social history than the decline of interest in inequality as an economic issueâ (p. 72). The current attitude is quite different, and in fact Milanovic´ (2007), one of the contemporary leading scholars of income inequality, wrote that âworld inequality is a topic whose time has comeâ (p. 3).
At first, because of the effects of late globalization and the post-Fordist accumulation regime, then because of the social impact of the Great Recession, the issue of inequality has recently regained attention in the economic and political debate. The dispute is particularly lively in the USA,1 where even President Obama and New Yorkâs Mayor De Blasio have addressed the issue in some of their public speeches, while Europe â once very concerned about social justice â is now mostly focused on macroeconomic imbalances rather than on its polarizing living standards.
Although we must welcome and support this renewed interest â that reconnects contemporary âeconomicsâ to classical âpolitical economyâ â we should, however, also notice that unfortunately the debate is still mostly focused on income and/or wealth distribution, i.e. âhow the pie is slicedâ (Rycroft, 2013), which is a very important but rather narrow view of economic inequality. Even if in the last decades some scholars have introduced and developed more comprehensive and multidimensional approaches to economic inequality (Maasoumi, 1986, 1999; Sen, 1992; Tsui, 1995, 1999 ; Weymark, 2006; Decanq et al., 2009; Decancq and Lugo, 2012; Aaberge and Brandolini, 2015), most of the past and current debate as well as of theoretical and empirical studies produced by economists concern personal income distribution or factor (functional) income distribution. This is more evident in the studies of the evolution and characteristics of contemporary capitalism (Piketty, 2013) and globalization (Cornia, 2003; Maskin, 2014).
A common justification for the focus on income and wealth distribution is that it is considered the âmotherâ of most other forms of inequality or the âfundamental inequality of capitalismâ (Piketty, 2013). Namely, (1) income distribution is generally deemed a paramount sign and indicator of overall inequality, (2) other inequalities are considered a reflection of income distribution, i.e. income may be used as proxy of other variables, and (3) a better income distribution is regarded as a necessary and often sufficient condition for reducing many other inequalities. In those monetary approaches,2 income or wealth (re)distribution is largely identified with the social justice issue.
Following a pioneering but not well-known article of Sen (1997),3 this work aims at showing that, notwithstanding their importance and role, the exclusive focus on income and wealth has some relevant drawbacks, and it suggests that a broader approach is needed. Section 2 of the paper concerns some relevant semantic problems. In section 3, the limitations of the conventional monetary approach are presented, while section 4 illustrates the complexity of inequality, across multiple dimensions and scales. In the last section, there will be some general conclusions.
Unequal meanings
Before proceeding to our analysis, it is useful to make some preliminary clarifications on the terminology that will be used, in order to avoid possible misunderstandings. The first clarification concerns the term âmultidimensionalâ, that has been cited above and is widely used in the recent studies on poverty and inequality measurement. In the vast literature on inequality, there is of course already some attention about the many dimensions of inequality. However, the term âmultidimensionalâ sometimes is used with different, inconsistent meanings. For instance, Sawyer (1989) wrote that âthere are many dimensions to inequality, and this would include inequality between and within social classes, between regions and countriesâ (p. 17). Likewise, also âraceâ,4 gender and sexuality are often considered as dimensions of inequality (Ferguson, 2016), especially by non-economists. On the other side, Stewart (2002), while introducing the concept of âhorizontal inequalitiesâ, emphasized that they are ÂŤmultidimensionalâwith political, economic and social elementsÂť (p. 2).
What Sawyer and other scholars mean is that there are many levels,scales(regions, countries, âŚ) or groups, units (class, gender, âŚ) to income inequality. However in that conventional approach there is only one metric: the monetary metric based on income or resources.
In this work, the term âdimensionâ will be used with a different meaning: dimension refers to the metric (or space) used for the inequality assessment. Therefore, if we analyse income inequality across regions or across classes, we are using different levels â e.g. regions and classes â but the same dimension â income. If we assess education or health inequality across regions or across classes, those will be different dimensions of inequality, as the metrics are education or health â rather than income â measured across different levels or groups. If, for example, we use jointly education, health and income across one or more units, that will be a multidimensional approach. In this sense, multidimensional inequality refers to the fact that inequality is measured and assessed using different metrics, not just different scales. When different units are used, such as individuals and countries, in this work it will be called multiscale inequality. An example of the latter can be found in the debate about the effects of globalization: scholars compare the trend of income inequality between countries (using the average or median national income), within countries (usually between individuals or households), and between all individuals in the world. Regarding inequality between countries, Milanovic´ (2007) defines it as âConcept 1â of inequality, or unweighted international inequality, if each country is not weighted according to its population size, while he defines it as âConcept 2â of inequality, or weighted international inequality if we weight each country. Accordingly, he defines âConcept 3â of inequality the distribution of income across all individuals in the world. As we can see, Milanovicâs concepts of inequality are multiscale, but not multidimensional. The same limitation characterizes the debate about the effects of globalization on inequality at any scale.
Furthermore, we may assess multidimensional and multiscale inequality at the same time. For instance, in the âInequality-adjusted Human Development Indicatorâ (IHDI) included in the Human Development Report (HDR) of UNDP since 2010, we have three dimensions (income, education, health) and two scales: between countries (though UNDP do not present a measure of intercountry inequality) and between individuals within countries.5
Another common misunderstanding concerns the supposed equivalence between income (or wealth) inequality and âeconomicâ inequality. Some authors that study income or wealth distribution, instead of using the general term âinequalityâ, prefer to use the expression âeconomic inequalityâ, which seems more precise and informs the reader about the fact that the focus will be on income or wealth. Therefore, economists who are aware of the many dimensions of inequality, consider income and wealth distribution as âthe economic dimension of inequalityâ (Atkinson, 2015, p. 14). On the contrary, according to Sen the distinction between income inequality and economic inequality is important, because of âthe presence of causal influences on individual well-being and freedom that are economic in nature but that are not captured by the simple statistics of incomes and commodity holdingsâ (Sen 1997, p. 398). Accordingly, in the following section we will not use âeconomicâ and âincomeâ inequality as synonyms.
A narrow focus
There is no need to prove that the economic and political debate on inequality is almost entirely focused on income and wealth distribution. There is an extensive scientific and grey literature on inequality and in those writings, with very few exceptions, inequality means unequal income distribution. A leading example is the several excellent works on inequality by Atkinson, the âgodfatherâ (Chu, 2015) of these studies. Even in his latest book on the subject, after discussing âwhat do we mean by inequalityâ, Atkinson concludes that âincome is only one dimension, and differences in income should be interpreted in the light of differing circumstances and of the underlying opportunities. But it remains the case that achieved economic resources are a major source of injustice. That is my reason for concentrating here on the economic dimension of inequalityâ (Atkinson, 2015, p. 14), which for him it is measured by income and wealth. The problem is that for many economists âachieved economic resourcesâ are the major source of injustice.
There are a number of interrelated reasons why the narrow focus on income and wealth is inadequate to the analysis of economic inequality, especially in a capitalist world-economy. Some of these reasons have been presented by Sen in his works, others have Marxian roots.
First, the focus on income and wealth leads to neglect or to leave in the background non-income economic and social inequalities that cause other not less severe and intolerable social injustices, which interact and couple with income inequality. Although, according to the conventional wisdom âachieved economic resources are a major source of injusticeâ, gender or ethnic discriminations in the labor market, for example, may be even a greater source of injustice, which may not be alleviated or compensated by better income distribution. For instance, the current participation of women to labor market (i.e. the âactivity rateâ) in Italy is around 54 per cent compared to 74 per cent for men: assuming that in general this is not a choice by women, that dimension of inequality is not less severe or unjust than the existing gender income inequality in Italy, neither can be measured or evaluated just looking at income and wealth distribution by gender. As a matter of fact, the exclusion from labour market is not just a deprivation of income and livelihood but also a source of many non-material severe deprivations and troubles that have been well documented by labour scholars. Another illuminating example is âracialâ health inequality in the USA: âAlthough health outcomes have been tied to income and social economic status, there are black-white differences in life expectancy of at least 3 years at every level of incomeâ (Devi, 2012, p. 1043). Is âliving lessâ not a major source of injustice?
In our examples we cannot assume or conclude âas generally is done in the economic and political debate â that income inequality has a policy priority over employment or health inequality, neither we can assume that the reduction of the former may automatically lower the latter. Also, we cannot think that employment and health inequality are not constituents of economic inequality.
Some economists focus on income inequality because they think that non-income inequality is the field of other social scientists, but unfortunately even many sociologists, political scientists and legal scholars largel...