Wealth and poverty gap: using tech to uncover the roots of urban inequality

Inequality in cities has not been eradicated by technological advances, and by some measures, our digital tools and their economic effects increase inequality. Technological advances and

tools may, however, further our understanding of inequality. Since 2011, I have been formally involved in research and teaching in urban planning at the University of Waterloo. My PhD

studies included original socio-spatial analytic research that highlighted the possibility that computational modelling could help us understand urban inequality. LESSONS FROM HAMILTON

Careful research into tax, land ownership and city clerk records by historian Michael B. Katz in the mid-1970s unearthed a pattern. His research into mid-19th century Hamilton, Ont.,

revealed that a relatively small segment of wealthy citizens occupied social and structural realities far different than the large (and largely poor) balance of citizens. Katz advanced our

understanding of urban poverty by noting the structural, rather than just personal, nature of inequality. Prior to Katz, poverty was often understood as a character or motivational flaw. For

the past two years, a sub-group of computer scientists, systems dynamics researchers and urban planning scholars have met annually at the Computational Social Sciences Society of the

Americas in Santa Fe, N.M., to examine the structural and dynamic roots of inequality. The conference sub-group aimed to reduce complexity while retaining the inequality dynamics we called

social primitives. They were not anthropological primitives but are artificial, digitally designed dynamics for clarifying inequality origins. STRUCTURAL INEQUITIES Urban poverty is not

simply an individual character flaw: inequality clearly includes structural dynamics, which means that wealth is not simply a matter of superior individual characteristics. Katz noted that

wealth may be as much a consequence of where you fit into the bigger structures of the city or the social class you happened to be born into. Structural dynamics means that something beyond

the individual played a role in generating and sustaining the inequality. Careful interrogation of the dynamics of income inequality present in Hamilton in the 1850s provided insight for

developing a greatly simplified digital model. Hamilton had a small, wealthy population that was relatively stable during the 1850s, a period of time where the much poorer balance of the

population experienced little economic gain. Most of the wealthy population noted in 1851 was still there a decade later. For the poor, records indicate that as many as two-thirds who where

there in 1851 had moved on a decade later. WEALTH AND STABILITY A notable driver in the historical record was the persistence of a wealthy segment of society — typically 10 per cent or fewer

residents. Their social, work and recreational circles gave them access to resources in the form of land ownership, ability to borrow money for investments and ownership of production.

Those advantages, in particular land ownership, did not seem to be available to the balance of the city who remained in a persistently fragile economic state. Our exploration of Katz’s

research and insights suggested that a key advantage of the wealthy was access to a range of social resources in the form of institutions that provided the money, trust and investment

opportunities. In our earlier published work on a model of primitive forms of cooperation, we developed our ideas about the wealth pattern in Hamilton by formulating a core question: What is

the minimal set of rules required to model the emergence of inequality in wealth distribution? We developed a model in the programming language R that comprised a two-dimensional landscape

whereby agents had to oversee areas of resources — in this case, sugar — randomly assigned to each plot. Agents were added to the landscape randomly and their survival was based on making

sure that their sugar levels maintained above a certain threshold. If two agents ended up beside each other and both had extra resources collected from the landscape, they could share those

extra resources when needed. We defined two agents or more working together to be proto-institutions. These proto-institutions represented the advantage of the wealthy in the Hamilton

context. The inequality of agents was measured and compared — those who were members of proto-institutions and those who were not. We began as simple as possible in order to try and observe

the early, nuclear emergence of inequality. When agents do not have membership in a proto-institution, they are completely dependant on what the landscape provides at each cycle. When agents

are part of an proto-institution, they have a means of banking extra sugar, benefiting from other member agents extra sugar and can survive cycles of low landscape sugar. The computational

model generates data over hundreds or thousands of cycles with a wide range of variable settings. Our early results suggest that the proto-institutions advance overall stability and wealth

but also increase agent inequality. Our focus on the simplest possible form of institutional emergence allows us to reduce the number of confounding variables while testing our hypothesis

about institutions and income inequality. Our model, while not a replication of 1850s Hamilton, was informed by the historical realities identified by Katz. CHANGE EXPERIMENTS The knowledge

gained from our agent-based model may provide clues about where to look for current inequality drivers in urban settings. An effective model could clarify how many and what kinds of

institutional types are required to narrow the inequality gap. We may, for example, hypothesize that informal developments in urban settings have fewer formal institutional structures. These

missing pieces of social infrastructure may contribute to persistent poverty. These _in silico_ experiments — conducted in artificial labs — support ongoing research to deepen understanding

about inequality and help identify which interventions are likely to be most beneficial. [ _Like what you’ve read? Want more?_ Sign up for The Conversation’s daily newsletter. ]