Walks, Dinca-Panaitescu & Simone (2016): Income Inequality and Polarization in the City of Toronto and York Region

Summary

"Income Inequality and Polarization in the City of Toronto and York Region, Part I: Examining Levels and Trends from Spatial and Non-Spatial Perspectives" is Research Paper 238 (May 2016) of the Neighbourhood Change Research Partnership (NCRP) at the University of Toronto, produced jointly with the United Way of Toronto and York Region and funded by the Social Sciences and Humanities Research Council of Canada. The verified full author list, from the PDF's title and bibliographic pages, is Alan Walks (geography and planning, University of Toronto), Mihaela Dinca-Panaitescu (Manager of Research, United Way Toronto & York Region), and Dylan Simone (then a PhD student, University of Toronto). This entry reads the full 41-page report, fetched directly from neighbourhoodchange.ca.

The report distinguishes income inequality (overall dispersion of incomes) from income polarization (a shrinking middle, re-concentrating into rich and poor groups — the "disappearing middle class"), examined at two spatial scales: non-spatial (individuals and households, regardless of location) and socio-spatial (among neighbourhoods, using census tracts as proxies). It covers 1980–2012 for the City of Toronto and York Region, benchmarked against the Toronto CMA, other GTA regions (Halton, Peel, Durham), and central Montreal and Vancouver.

The Core Argument / Findings

Measures. Non-spatial inequality is measured with the Gini coefficient; polarization with the coefficient of polarization (Walks 2013), preferred over the Wolfson, ER, and WT indices for cross-place comparability. Both are calculated for individuals and households, using before-tax-after-transfer income from Statistics Canada long-form census microdata (1980–2005, a 20% sample), plus Environics Analytics/CRA tax-filer data for socio-spatial 2012 figures.

Non-spatial trends. Individual income inequality was flat through the 1980s, then rose steadily from 1990 in both places, tracking the Toronto CMA and comparator cities. Household income inequality was consistently higher in Toronto and lower in York Region than the CMA average, reflecting household composition differences. Polarization followed the same shape — declining in the 1980s, rising steadily from 1990. The middle-income household share ($50,000–$124,999, 2012 dollars) fell from 49% to 38% in Toronto (1980–2012) and from 57.6% to 43.5% in York Region. The authors report this shift was driven more by growth in the lower-income group than growth at the top — high-income households in both places largely stagnated after 1990, unlike the national pattern (per other Canadian literature the report reviews) where top-income growth dominates rising inequality.

Socio-spatial trends. Neighbourhood-level inequality and polarization rose faster than their non-spatial counterparts in both places — the authors' central methodological point, meaning income segregation is not merely a mechanical byproduct of widening overall inequality but reflects active geographic sorting. The middle-income share of census tracts fell from 56% to 30% in Toronto (1980–2012) and from 77% to 66% in York Region. Toronto's socio-spatial inequality (individual incomes) nearly doubled over the period, ending roughly double the other GTA regions' levels.

Named drivers. The report attributes rising non-spatial Canadian inequality generally (via cited literature, not new analysis here) to the fading redistributive strength of the tax-and-transfer system since the early 1990s, and cites earlier work linking metropolitan earnings inequality to deindustrialization, changing occupational structure, and rising immigration rates (Bolton and Breau 2012), tempered by rising female labour-force participation. For the faster rise in socio-spatial segregation specifically, the authors point to gentrification — high-income inner-city neighbourhoods expanding at the expense of middle-income ones in Toronto after 1990 — and propose, partly as interpretation rather than tested decomposition, a feedback loop of self-segregation among higher-income households and mobility barriers concentrating lower-income households in place.

Relation to the Georgist Case

This tie is deliberately thin and should be reported honestly: the report does not discuss land value, economic rent, or land value taxation as a category, and the word "rent" appears exactly once in the entire document, in a passing background remark about landlords under-maintaining low-income rental stock — not part of any rent-theoretic argument. The paper's relevance to the wiki is geographic, not causal-on-land: it establishes that income segregation across Toronto and York Region neighbourhoods is large, growing, and growing faster than aggregate income inequality — the spatial substrate on which locational land value and housing-cost differentials (the subject of Housing unaffordability is a land problem and Rising land values and housing costs drive poverty) necessarily operate. It should be cited as background on Toronto's neighbourhood-level inequality geography, never as evidence that land rent drives polarization — the report makes no such claim and was not designed to test one.

Nuances and Limits

• Descriptive, not causal. The report documents levels and trends; it does not model land price, zoning, or housing-policy effects on polarization.
• Data seams. Pre-2006 census data lacks after-tax income, so the report uses before-tax-after-transfer income throughout — the authors argue this likely understates the "true" rise in inequality. The 2012 socio-spatial figures blend a different source (Environics Analytics/CRA tax-filer estimates) than the 1980–2005 census data; bridged with inflation-adjusted cut-points (Appendix 2), but a real methodological seam.
• Gentrification and self-segregation as drivers are partly interpretive — the most plausible reading of the results plus cited literature, not a formally isolated causal estimate; a promised Part II was to examine drivers by demographic group. Checked (2026-07-18, T2 verification): the Neighbourhood Change Research Partnership's own complete publications list (neighbourhoodchange.ca, covering 2005–2021, the authoritative source for the partnership's own output) lists this 2016 report only as "Part I" and carries no follow-up titled "Part II" in any subsequent year. No report literally titled "Part II" of Research Paper 238 was located, but the same three co-authors delivered the promised demographic-driver analysis through a different, United Way-branded lineage: Rebalancing the Opportunity Equation (United Way Greater Toronto, May 2019) — its own foreword calls it "this third instalment in our Opportunity Equation research" — breaks down GTA (Peel, Toronto, York Region) income-inequality trends from 1980–2015 by age, immigration status, gender, and racialized status, finding young adults, immigrants, and racialized residents increasingly disadvantaged relative to seniors, the Canadian-born, and white residents respectively. Two other related outputs were also surveyed but are not the promised Part II: Dinca-Panaitescu et al. (2017), "The Opportunity Equation in the Greater Toronto Area" (a related but distinct earlier instalment in the same Opportunity Equation series), and the multi-city edited volume Grant, Walks & Ramos (eds., 2020), Changing Neighbourhoods: Social and Spatial Polarization in Canadian Cities (UBC Press), which may separately fold in demographic-group analysis for Toronto within a book chapter — not confirmed, and not checked this session.
• Institutional working paper, not peer review, though the polarization index used is grounded in Walks' separately published work (2013).
• Regional scope. Findings are specific to Toronto and York Region; Vancouver and Montreal appear only as benchmarks.

Bears On

Feeds spatial/housing context for The young are increasingly locked out of land wealth (the intensifying geography beneath the intergenerational land-wealth divide) and Canada (Toronto as Canada's largest, most land-value-driven metro). Backdrops August (2020): The Financialization of Canadian Multi-Family Rental Housing, whose financialized-landlord penetration is unevenly distributed across this kind of neighbourhood geography. Should not be wired as evidence for the land-versus-construction-cost decomposition in Housing unaffordability is a land problem, since it does not test that decomposition.

See Also

• The young are increasingly locked out of land wealth
• Canada
• August (2020): The Financialization of Canadian Multi-Family Rental Housing
• Housing unaffordability is a land problem, not a construction-cost problem
• Rising land values and housing costs drive poverty

Sources

1. Alan Walks, Mihaela Dinca-Panaitescu, and Dylan Simone, "Income Inequality and Polarization in the City of Toronto and York Region, Part I: Examining Levels and Trends from Spatial and Non-Spatial Perspectives," Neighbourhood Change Research Partnership Research Paper 238, University of Toronto, May 2016. PDF — fetched and read in full — used for the author list, methodology (Gini coefficient and coefficient of polarization), all non-spatial and socio-spatial inequality/polarization figures, the cited drivers (deindustrialization, immigration, tax-transfer erosion, gentrification, self-segregation), and the absence of any land-rent argument.
2. Mihaela Dinca-Panaitescu, Laura McDonough, Dylan Simone, Ben Johnson, Stephanie Procyk, Michelynn Lafleche, and Alan Walks, Rebalancing the Opportunity Equation, United Way Greater Toronto, in partnership with the Neighbourhood Change Research Partnership, May 2019. PDF — fetched and read (front matter, foreword, contents); used only to confirm this report — not titled "Part II" but sharing three of Part I's co-authors — delivered the demographic-driver breakdown (age, immigration status, gender, racialized status) that Part I's own text promised as a future Part II.