FRASER RESEARCH BULLETIN November 2020
Economic Performance in Ontario CMAs: A National Comparative Perspective by Ben Eisen and Joel Emes
Summary
Several recent Fraser Institute stud- Income growth has been significantly ies have documented the weak econom- stronger in the CMAs immediately prox- ic performance in Ontario over the past imate to Toronto than in those further 20 years outside of the Greater Toronto away in London and Windsor. Area (GTA), and to a lesser extent Ot- tawa. A national analysis shows that be- tween 2005 and 2015, Windsor fell from This bulletin builds on that analysis having the tenth highest median house- by examining economic performance hold income in Canada to having the 25th data from the Census Metropolitan Ar- highest out of Canada’s 36 CMAs. This eas (CMAs) of southwestern Ontario 15-place decline was the largest fall of including London, Windsor, Brantford, any CMA in Canada. London fell from Kitchener-Waterloo-Cambridge, and having the 15th highest median household Guelph. income in Canada to having the 27th high- est, the second largest decline in Canada. While generally the southwestern region of the province has struggled economically in recent years, there was substantial variation between the CMAs.
fraserinstitute.org FRASER RESEARCH BULLETIN 1 Economic Performance in Ontario CMAs
Introduction well as that of the individual CMAs within the region. We show that the extent of this relative Several recent Fraser Institute studies have decline for several cities in the region is unique documented the uneven nature of Ontario’s amongst Canadian CMAs. economic performance over the past twenty years, particularly with respect to job creation. Whereas the Greater Toronto Area and to a Ontario’s economic performance— lesser extent Ottawa have generally seen com- overall, and manufacturing paratively strong job creation, nearly all other Although there have been brief periods of regions of the province have not. More specifi- stronger growth, Ontario’s overall economic cally, many Census Metropolitan Areas (CMAs) performance in the early years of the 21st cen- experienced significant job loss in the years tury has been weak. Ontario’s economy was hit leading up to and during the 2008–09 recession hard by the 2008–09 recession, and the subse- and then experienced weak recoveries in sub- quent recovery was tepid. The 2008–09 reces- sequent years (Lafleur and Eisen, 2017). sion saw Ontario’s real GDP per person fall by This short bulletin builds on that analysis by 5 percent, which took five years to get back to providing analysis of additional data surround- the 2007 level (Statistics Canada, 2020c, 2020d). ing economic performance in the larger cities Growth in Ontario’s real GDP per person after of southwestern Ontario. the 2008–09 recession has averaged 1.2 percent per year compared to 1.9 percent in the previ- We define southwestern Ontario as the region ous period between major recessions (1992– of approximately 2 million people living in the 2007). CMAs of London, Windsor, Brantford, Kitchen- er-Waterloo-Cambridge, and Guelph, as well Due to the severity of the recession and the as surrounding smaller towns and rural areas. weak economic recovery, Ontario was near the Due to data constraints, the measures of sub- bottom of the Canadian provincial pack with provincial economic indicators measure devel- respect to a wide range of economic metrics. As opments in the large cities (CMAs),1 but not in a result, we have in the past characterized the smaller urban areas (CAs) or rural areas outside years immediately before, during, and after the of the CMA and CA boundaries.2 2008–09 recession as a “lost decade” for the province’s economy (Eisen and Palacios, 2018). This bulletin builds on past Fraser Institute studies by including measurements of chang- Although the effects of the 2008–09 reces- es in household income over the past two de- sion and the fact of a weak recovery are broad- cades and also considers the relative change in ly known, what is less well understood is that the region’s prosperity in a national context, as warning signs for Ontario’s manufacturing- heavy economy were evident for several years prior to the recession’s arrival. As manufactur- 1 The combined population of the five southwest- ing jobs migrated to developing economies, ern Ontario CMAs is 1.8 million. many North American manufacturing jurisdic- 2 In 2019, 82 percent (11,940,946) of Ontario’s pop- tions, including Ontario, started to see manu- ulation resided in CMAs while 8 percent (1,176,057) facturing employment declines much earlier in resided in CAs and 10 percent (1,449,544) in other areas. Statistics Canada (2020e). the decade. fraserinstitute.org FRASER RESEARCH BULLETIN 2 Economic Performance in Ontario CMAs
Figure 1: Manufacturing Employment in Ontario, 2001–2019
Number of jobs
Source: Statistics Canada, 2020a.
Manufacturing employment in Ontario peaked sector of any of the province’s regions. More in 2004 at 1.1 million jobs (figure 1). By 2007, specifically, a 2016 report showed that manu- that number had declined to 939,000 jobs—a facturing comprised just over 15 percent of to- decrease of 15.1 percent over just three years. tal employment in the southwestern region When the recession struck, the pace of manu- compared to just over 10 percent in the prov- facturing job loss accelerated, falling to 763,500 ince taken as a whole (Eisen and Emes, 2016). in 2010. Since then, there has been no rebound As a result of widespread job losses in the man- to speak of with respect to manufacturing em- ufacturing sector, southwestern Ontario was ployment in Ontario. Meanwhile, Quebec, an- hit harder than any other region of the prov- other manufacturing jurisdiction hit hard by ince by the 2008–09 recession. Total employ- the 2008–09 recession, shows a mild rebound ment fell by 5 percent in the region from 2008 in the subsequent years. to 2009, compared to 2.7 percent in the prov- ince taken as a whole, after which the region Employment trends in Southwestern experienced a slow recovery. Ontario CMAs Although the region taken as a whole has expe- The fact that Ontario’s manufacturing sector rienced weak job creation rates and (as will be suffered such severe employment losses which discussed in the next section), weak household have not recovered since the 2008–09 reces- income growth, there is substantial variation in sion has created specific challenges for the terms of the job creation rates among the re- southwestern Ontario region, which has the gion’s various large cities. highest share of its workforce employed in that fraserinstitute.org FRASER RESEARCH BULLETIN 3 Economic Performance in Ontario CMAs
Data on sectoral jobs by economic region or two large CMAs (with a combined population CMA are not publicly available. However, an ex- of over 900,000) is responsible for the region’s amination of total employment in southwest- low overall rate of job creation in this period. ern Ontario’s CMAs reflect the severity of the It is beyond the scope of this bulletin to assess prolonged period of weak economic perfor- all the possible contributing factors to this out- mance. Figure 2 shows employment levels in come. An almost certainly important point of the southwestern CMAs using an index, set- difference between the stronger and weaker ting 2001 levels at 100. This method allows a performing CMAs, however, is their proximity straightforward comparison between CMAs of to Toronto (Moffat, 2015). The three stronger different sizes. Overall job creation in Ontario is performing CMAs are much closer to Toronto, shown in the dark dashed line. and are in fact at the fringes of that metropo- Figure 2 shows us that three of the region’s lis’ large commuter shed. All are serviced by CMAs—Kitchener-Cambridge-Waterloo, the regional GO Transit system. They are also Guelph, and Brantford—actually experienced proximate to Pearson airport, and so better more job growth between 2001 and 2019 than connected to the global economy than outly- the provincial average. By contrast, London and ing CMAs. Meanwhile, London and especially Windsor experienced much weaker job growth Windsor are much further away from the pro- over this period. While total employment in vincial capital (which experienced strong job Ontario increased by 25.9 percent from 2001 to creation over this period). Another potentially 2019, that figure was just 9.1 percent in London contributing factor is that the Labour Force and 9.4 percent in Windsor. These numbers Survey determines employment by CMA where show that the weak job creation rate in these employees actually live—not where they work.
Figure 2: Employment Index by CMA (2001 = 100)
Kitchener-Cambridge-Waterloo Brantford Guelph Ontario
London
Windsor
Sources: Statistics Canada, 2020a, 2020b. fraserinstitute.org FRASER RESEARCH BULLETIN 4 Economic Performance in Ontario CMAs
This means that some jobs attributed to CMAs the prosperity of these CMAs relative to other near Toronto are actually taking place in the regions of the country during this period. GTA itself. As such, employment figures in Figure 3 shows that in 2005, southwestern On- CMAs proximate to Toronto are also partly at- tario was still a prosperous region of the country. tributable to job growth that is actually taking Out of Canada’s 36 CMAs, southwestern Ontar- place within the GTA itself. io’s five large CMAs were all in the top half of the Future regional economic analysis should con- country for this metric. Guelph had the fourth sider whether for analytical purposes it is use- highest median household income in Canada. ful to consider the CMAs in close proximity to In 2005, Kitchener-Cambridge-Waterloo had Toronto as distinct category of CMAs from the the seventh highest median household income. large outlying cities of London and Windsor, Windsor had the 10th highest median income in which generally share a pattern of recent eco- the country. London had the 15th highest and nomic performance that diverges sharply from Brantford the 17th highest median income in that of the GTA and surrounding areas. Canada—still in the top half of the country.
Over the next ten years, however, partly due to The relative change in prosperity the manufacturing job losses described above, in southwestern Ontario CMAs in a median household income levels (in 2015 dol- national perspective lars) in the SWO CMAs stagnated. Taken as a whole, Canada’s median household income in- The data above show that southwestern Ontario’s creased by 13.0 percent ($10,370) over this pe- manufacturing sector and the region as a whole riod. By comparison, median household income suffered significant employment losses in the in Guelph (6.1 percent, $5,904), Kitchener-Cam- years prior to and during the 2008–09 recession bridge-Waterloo (4.4 percent, $3,997), Brant- with a tepid recovery coming in the immediate- ford (4.0 percent, $3,310), and London (2.0 per- ly following years. Further, the data have shown cent, $1,660) increased at a much slower rate that employment growth was much weaker in and median household income actually fell in London and Windsor than in the southwestern Windsor (-3.1 percent, -$2,815). Again, economic Ontario regions more proximate to the GTA. outcomes for this indicator were significantly To better understand the impact of these de- worse in London and Windsor than in the rest velopments at the household level, we turn of the region, though the difference is less stark. now to examine changes in median household As a result of these trends, by 2015 the South- income between 2005 and 2015. Because this western CMAs were much less prosperous rela- metric is measured using census data, 2015 is tive to the rest of the country in terms of me- the most recent available data. This is a sig- dian household income than they had been in nificant limitation in that it does not permit an 2005. More specifically, the damage in the two analysis of this metric during the period from large CMAs further from Toronto—London and 2015 to today, but it is nonetheless useful in il- Windsor—relative to the rest of the country lustrating the severity of the economic chal- was particularly severe. Windsor fell from hav- lenges in southwestern Ontario’s CMAs be- ing the tenth highest median household income tween 2005 and 2015 as well as the changes in in Canada in 2005 to the 25th highest in 2015. fraserinstitute.org FRASER RESEARCH BULLETIN 5 Economic Performance in Ontario CMAs
Figure 3: Median household income by CMA, 2005
T B L N T B B L B T E 2015 $ Sources: Statistics Canada, 2016 Census of Population.
fraserinstitute.org FRASER RESEARCH BULLETIN 6 Economic Performance in Ontario CMAs
Whereas in 2005 Windsor’s median household had the largest drops in the table of any CMAs income was 12.1 percent higher than the na- in the country therefore can’t be attributed en- tional average, by 2015 it had fallen to 4.0 per- tirely to this boom period in energy-intensive cent lower than the national average. regions.
The story in London, Ontario is similar to These data show that a remarkable change Windsor. In 2005, London’s median household in southwestern Ontario’s relative prosperity income was ranked 15th in the country—the top compared to the rest of Canada occurred be- half of the national league table. In 2015, Lon- tween 2005 and 2015. This was especially true don had fallen to 27th place. in the province’s two large CMAs furthest from Toronto, London and Windsor. Windsor went These data suggest that with respect to medi- from being in the top ten and London from the an household income, southwestern Ontario’s upper half of Canada’s CMAs in terms of median economy had worse relative outcomes within household income to being in the bottom quar- Canada than any other region in the country. ter over the course of a decade. Windsor’s fall from 10th to 25th, a drop of 15 plac- es, was the largest decline in the table in Can- ada. London’s drop of 12 places was the second Developments since 2015 largest fall. Some southwestern CMAs have experienced an The remaining southwestern CMAs closer to uptick in employment growth in recent years, Toronto also moved down the league table particularly since 2015 (figure 4). Again, this up- over the course of this period. Guelph fell from tick has been stronger in the CMAs proximate fourth to eighth place, while Kitchener-Cam- to Toronto, although Windsor has participated bridge-Waterloo fell from seventh to 13th. These in the trend to a lesser degree. movements down the Canadian league table It remains to be seen how improved job perfor- still left these CMAs in the top half of the coun- mance in some southwestern CMAs will affect th th try. Brantford fell from 17 to 26 place. median household incomes since 2015. For in- However, this movement from near the top stance, the housing boom in Toronto and sur- of the list to closer to the middle was driven rounding areas that has taken place in recent as much by rapid ascents to very high median years has increasingly spread further afield household income growth in energy producing from the GTA, including in London. Labour regions during this time. For instance, in 2015, Force Survey data suggests construction em- six of the seven highest CMAs for this indicator ployment is driving significant job growth and were in either Alberta, Saskatchewan, or New- may help increase median incomes in the com- foundland and Labrador. In 2005, Calgary and muter shed of the GTA as shown in the next Edmonton had been the only CMAs from these census. oil-intensive jurisdictions in the top six. We will learn more about Ontario’s recent per- Of course, the rapid ascent of oil-intensive ju- formance on this metric when the 2021 Cen- risdictions for this indicator affected the rela- sus results are available. Developments in the tive position of all other CMAs, not just those rest of the country may also cause Ontario in Ontario. The fact that Windsor and London CMAs generally to improve with respect to fraserinstitute.org FRASER RESEARCH BULLETIN 7 Economic Performance in Ontario CMAs
Figure 4: Median household income by CMA, 2015
T B N L B T B L B T E 2015 $ Sources: Statistics Canada, 2016 Census of Population.
fraserinstitute.org FRASER RESEARCH BULLETIN 8 Economic Performance in Ontario CMAs
median household income relative to other Such a dramatic change, over the course of a CMAs across Canada. Specifically, dismal eco- single decade, helps illustrate the scale of the nomic performance in provinces with large en- economic challenges facing the region and pro- ergy sectors since 2015 may cause the perfor- vides some further insight on the nature of the mance of Ontario’s CMAs in the 2020 census to economic pain, at the household level, experi- improve relative to other cities on this metric. enced by many people in the region, particular- This is particularly true given that economic ly in the large cities that are furthest away from troubles in provinces with large energy sectors Toronto. New data in the 2021 Census will pro- are likely to see negative ripple effects outside vide further insight on how median household that single sector. The lack of construction em- incomes have changed in absolute and relative ployment in Calgary, for instance, may further terms since 2015.3 contribute to weak performance with respect to job growth and median incomes reflected in References the 2021 Census. Eisen, Ben, and Joel Emes (2016). The Five Soli- tudes of Ontario: A Regional Analysis of Labour Conclusion Market Performance in Post-Recession Ontario. Southwestern Ontario is a densely populated Fraser Institute.
Statistics Canada (2016). Income Highlight Ta- Statistics Canada (2020c). Table 17-10-0005-01 bles, 2016 Census. Statistics Canada Catalogue Population estimates on July 1st, by age and no. 98-402-X2016006.
Statistics Canada (2020b). Table 14-10-0096- Statistics Canada (2020e). Table 17-10-0135-01 01 Labour force characteristics by cen- Population estimates, July 1, by census metro- sus metropolitan area, annual (x 1,000). politan area and census agglomeration, 2016
Ben Eisen is a Senior Fellow in Acknowledgments Fiscal and Provincial Prosperity Studies at the Fraser Institute. He The authors thank the anonymous reviewers of holds a BA from the University early drafts of this paper. Any errors or omissions of Toronto and an MPP from the are the sole responsibility of the authors. As the re- University of Toronto’s School of searchers worked independently, the views and Public Policy and Governance. conclusions expressed in this paper do not neces- sarily reflect those of the Board of Directors of the Fraser Institute, the staff, or supporters. Joel Emes is President of Abacus Economics and a Fraser Institute Copyright © 2020 by the Fraser Institute. All rights Senior Fellow who rejoined the reserved. Without written permission, only brief pas- Institute after a stint as a senior sages may be quoted in critical articles and reviews. advisor to British Columbia’s provincial government. Joel holds ISSN 2291-8620 a B.A. and an M.A. in economics from Simon Fraser University. Media queries: call 604.714.4582 or e-mail: communi- [email protected]
Support the Institute: call 1.800.665.3558, ext. 548, or e-mail: [email protected]
Visit our website: www.fraserinstitute.org fraserinstitute.org FRASER RESEARCH BULLETIN 10