Introducing the Evolution of Inequality Project

The study of income and resource inequality has been the academic topic du jour this year, highlighted by Thomas Piketty's economic history opus [1] and a special issue of Science [2]. There was even a paper on the 1% vs. the 99% of academic publishing [3] which argues that in terms of citations, the rich tend to get richer. But how do these patterns emerge and evolve? Is it a merely a statistical artifact, or a reflection of how complex, hierarchical societies tend to evolve? For example, we might assume that extreme inequality is maladaptive. But upon simulating a range of artificial societies of different population sizes, initial degrees of stratification, and behavioral features, we might find that extreme inequality tends to occur under specific conditions.

This is the motivating factor for my interest in the topic. Unlike many of the more well-versed approaches to the topic, an alternative view of inequality is a cross between statistical distributions, nuanced views of human sociobiology, and the aftermath of social change. While most economists have taken a materialist view of inequality, I feel that evolutionary perspectives would be helpful in teasing out questions of inequality's origins. This might involve data as diverse as historical data, ethnographic data, evolutionary modeling, and behavioral/neurophysiological data. In the end, we will be able to provide a conceptual alternative to the usual discussion at the intersection of behavior, biology, and social change. An inclusive, multidisciplinary approach is a core component of this project.

My research organization (Orthogonal Research) is trying to initiate work on a project called "The Evolution of Value and Inequality". This project is an attempt to understand the emergence of these social inequalities as a set of evolutionary and biobehavioral phenomena. While the argument can be made that an evolutionary perspective might be helpful in understanding unequal allocations of resources, it is sorely lacking in the general discussion. The broadness of the initiative is necessary to make the connection between the pure inferential approach of evolutionary science coherent public policy outcomes. An initial grant application to the Washington Center for Equitable Growth (submitted January 2014) did not get funded, one reason being that the idea needed more conceptual and empirical fleshing out.

So upon doing some more conceptual refinement, I just finished submitting a second version of this proposal to the Institute for New Economic Thinking (INET). While I will not get into the technical details here, the basic idea is to construct adaptive computational models that mimic a social hierarchy (so-called hierarchical network models). Each node of these directed graphs are informed in their behavior by neuroimaging and other physiological sources of data on human behavior.

The project is focused around evolutionary models of social change (social and cultural change), the underlying assumptions of which can be verified by the collection of biobehavioral data (e.g. neuroimaging experiments). The empirical component is meant to test assumptions of individual and social behavior, and serves as an alternative to the rational expectations assumption that dominates much of conventional economics. However, it also refines many of the model-free findings that
characterize behavioral economics [4].

The evolutionary aspects of this work are also quite interesting. Because we are bridging the short-term (behavioral) and the longer-term (social evolution), there are at least three forms of adaptation: a social learning mechanism, a cultural evolutionary form of selection, and a neurophysiological imperative that satisfies various material, social, and existential needs of an individual. This gives us a fitness and selection criterion that is tangentially related to reproductive success. Subsequent evolutionary algorithms and simulations may bear out the evolutionary dynamics of value construction and social stratification.

Another contribution of this project is to link the statistical aspects of inequality with an evolutionary and demographic framework. The oft-referenced phrase the "1%" or the "0.01%" has its roots in an exponential (non-normal) statistical distribution called the power law. Power laws of various size tend to describe observed income distributions in many different types of society. As inequality increases sharply in a single society, or as different degrees of inequality are observed in different contexts, the power law and in conjunction with various stable states can be used as selection criteria.

Schematic of expected results. Both the C and L parameters refer to operations on intra- and bi-level hierarchical networks dynamics, respectively.

As proposed in the first part of this post, the resulting evolutionary algorithms and experimental inquiries provide us with a possibility space for given outcomes. Given a set of initial conditions, we can observe the tendencies of inequality of resource allocation. If there are common outcomes relative to a number of different initial conditions, this could tell us something cross-cultural and fundamental about the nature of inequality. Ultimately, the outcomes of this project could help to identify and predict opportunities to head off crises and as an architecture for achieving sustainable economic growth.

NOTES:
[1] Piketty, T.   Capital in the 21rst Century. Harvard University Press (2014).

[2] Citation for the special issue: Science, 344, May 23 (2014). One article with particular relevance to social evolution is: Pringle, H.    The Ancient Roots of the 1%. Science, 344, 822-825 (2014).

[3] Ioannidis JPA, Boyack KW, Klavans R.   Estimates of the Continuously Publishing Core in the Scientific Workforce. PLoS One, 9(7), e101698. doi:10.1371/journal.pone.0101698 (2014).

[4] Camerer, C.F. and Loewenstein, G.   Behavioral Economics: past, present, future. In "Advances in Behavioral Economics". C.F. Camerer, G. Loewenstein, and M. Rabin eds. Chapter 1. Russell Sage Foundation (2004).

Behavioral Economics Reading List. Russell Sage Foundation blog, March 23 (2012).