Atkinson and Economic Inequality: Understanding Contemporary Society

Discussion of Tony Atkinson’s influence cannot be confined to his fundamental theoretical research on income distribution analysis. Through many seminal contributions, he has greatly enhanced our understanding of the distribution of labour earnings, income, and wealth in modern societies.

Although this work has constantly been informed by rigorous theoretical mod- elling—a typical example being provided by Atkinson (2008b, especially Part II)—this line of research is mainly empirical. It is distinguished by scrupulous attention to the quality of data that serves as a model for social science inquiry. Data quality is an aspect of scholarship that is not currently fashionable in mainstream economics. Yet, throughout his work, Tony has painstakingly tried to dig into the details and nuances of the available statistical information, often developing original ways to employ it. His ability to match conceptual theorising with an understanding of statistical production has gained him the respect of professional statisticians, as testified by the ‘Atkinson Review’ or his appointment to the European Statistical Governance Advisory Board (see below). Tony’s effort has provided a salutary warning against unwary use of statistical material, the understanding of which is a precondition for avoiding misleading conclusions, and arriving at robust results that are of value (e.g. Atkinson and Brandolini 2001). A second distinctive feature of this line of work is the constant attention to the international dimension.

Tony has been one of the pioneers, with Piketty and Saez, of the burgeoning literature on long-run trends in the distribution of income summarised in terms of top income shares (the share of total income held by the richest 5% or the richest 1%, etc.). The innovation is to use previously unexploited data from income tax administrative records, suitably adjusted, to derive series back to the beginning of the twentieth century. This enhances understanding of trends and cross-national differences in the light of historical trends to a much greater extent than is typically possible in many analyses of economic inequality (income survey data for most OECD countries typically go back only about 40-50 years at most). (See, inter alia, the cross-national studies collected in the volumes edited by Atkinson and Piketty (2007, 2010), and the magisterial review and synthesis by Atkinson et al. (2011) and Alvaredo et al. (2013).) The research shows that the experience of English-speaking countries differs significantly from those of others. (On this also see Atkinson and Leigh (2013). For Tony’s research on top incomes in the UK, see, for example, Atkinson (2005a) and Atkinson and Voitchovsky (2011).) In current research, Tony is extending the top incomes analysis to former colonial countries.

The rise in earnings and income inequality from the mid- to late 1970s onwards that occurred in the USA and the UK is often assumed by many economists to be a universal phenomenon. Per contra, Atkinson (2008b) showed that the rich industrialised countries of the OECD exhibited great heterogeneity in the patterns and timing of changes in earnings and income inequality, echoing the lesson of the top incomes research. This built on Tony’s pioneering and comprehensive documentation of income distribution differences across OECD countries 15 years earlier (Atkinson et al. 1995). More recently, Tony was one of the first economists to examine the relationship between financial crises and the distribution of income, again taking a cross-national and long-run historical perspective (Atkinson and Morelli 2011).

Tony Atkinson’s research has been ahead of its time, anticipating a number of topics that have recently become of substantial contemporary interest, and his contributions remain key reference points. Examples include his work on the distribution of wealth which made innovative use of inheritance tax data for the UK (Atkinson and Harrison 1978), and on earnings mobility (Atkinson et al. 1992) which reviewed the analytic possibilities provided by longitudinal data on earnings and income which were then just becoming more widely available. Tony’s research on inequality and poverty in Eastern Europe seized on a topic that became of fundamental importance thereafter (Atkinson and Micklewright 1992).

An even better example of prescience is Tony’s work on intergenerational mobility in income (Atkinson et al. 1983), based on an innovative follow-up survey in the late 1970s of children of the respondents to a 1950 survey, producing some of the first ever estimates of the degree of inheritance of earnings and income to match estimates of social and occupational class mobility that were more commonly available. These were pioneering steps: 30 years later, and with more data available, estimates of intergenerational income mobility now abound for many countries (see Jantti and Jenkins 2015). This is also a topic to which Tony has added analytical insights alongside empirical innovation. For example, it is now ‘well-known’ that summarising a father’s income using data for a single year can lead to a downward bias in the estimate of the degree of intergenerational immobility in income. Although the result is commonly attributed to Solon (1992), it appeared a decade earlier in Atkinson (1981b).