The Theories of Distribution and the Empirics of Inequality
Current empirical studies show an overwhelming evidence of an increasing inequality since the late 1970s in many advanced economies and notably in Anglo-Saxon countries: the labour share decreased and at the same time both wage dispersion and household inequality increased (for example, Checchi and Garcia-Penalosa, 2008: 604).
Confronted with this evidence, the reader may wonder what the explanatory power of the main theories of distribution that we have reviewed is.The recent, highly successful book by Thomas Piketty (2014), while referring to no mathematical (or econometric) model nor to any formalized theory of distribution, can be instructive in this respect. A big issue on which he insists is the excess of the “rate of return to capital” over the output growth rate for long periods of time, which is an automatic driver of an ever increasing capital share and an ever decreasing labour share (ibid.: 232-3). Even though this evidence is broadly consistent with the “Cambridge equation” referred to above, more should be said on the reasons why profits have been so high relative to growth and average wages could obtain such small advantages from technological progress. Another important issue concerns the modes of the increasing earnings inequality, which displayed in past decades a fanning out of the upper tail of the distribution and a relative fall at the middle (ibid.: 304-10; also Atkinson 2007; Autor et al. 2006). A marginal productivity theory of labour demand cannot explain these wage developments: a more nuanced view should include some other elements.
It is hardly possible, we suggest, to explain the main stylized facts of income distribution without integrating economic arguments with arguments relating to other spheres of social life. The distribution of wealth and power among individuals and social groups, the advantages and disadvantages that they can get from specific technological developments, the institutions that regulate labour and capital markets, and so on, cannot be without effect on income distribution and its changes through time. In this respect, the open-ended analyses of income distribution have some advantages over the selfcontained theories.
ArRIGO OpOCHER