• ◊ Refresh the page if you do not see the animation
• ◊ Click here for the full screen version:

Update 2009-12-08: The animation now shows the preliminary US estimates  (i.e. without averaging).  The most recent month’s employment numbers are published earlier in Canada than in the US.  The animation now displays them too.

_________________________

Have some North American cities been less affected than others by the economic crisis? As national job markets have been severely hit – between July 2008 and July 2009, Canada lost over 300,000 jobs, and the United States over 5.5 million jobs – it has become clear that local economies matter.

The interactive map shows the year over year employment gains and losses for the largest 20 Canadian and 100 US metropolitan areas. The data are three month moving averages of the seasonally adjusted employment levels for each city obtained from Statistics Canada and the US Bureau of Labor Statistics.

” The map traces how the contagion spread from metropolitan areas in Florida, California and the industrial Midwest. It reveals that some cities have fared less poorly than others while for some this recession was yet another blow after years of decline”, explains Dr Peter V Hall, professor at the Urban Studies Program at Simon Fraser University.

Press to play the animation, or drag the cursor along the time line to view a particular month. The map will be updated monthly. Adds Dr Hall: “Keep watching it – the recovery will have to start somewhere”

This animation was inspired by two very good animations at TIP Strategies and Slate, which show the US story at the city and county levels, respectively.

Canadian labour statistics were obtained from the Statistics Canada Labour Force Survey (Cat. 71-001-X). Seasonally adjusted three-month moving averages for the 20 largest Census Metropolitan Areas were obtained from Table 5-1 and the seasonally adjusted national employment levels were obtained from Table 1.

US statistics were adapted from the Bureau of Labor Statistics Current Employment Survey. Three month moving averages were calculated for the total non-farm payroll employment, nationally and for the top 100 Metropolitan Statistical Areas. Year over year differences were calculated from these. All the data are seasonally adjusted except for cities in AR, DC, ID, KS, LA, NV, RI, TX, and VA, for which bls.gov only reports unadjusted employment levels. Seasonal adjustment appears to alter overall employment levels by 1-2% and has only a small effect on the final result.

There is a methodological difference between these surveys. LFS statistics are obtained through a survey of 53,000 households across Canada, while the CES is an establishment survey of 150,000 US businesses and government agencies. The US equivalent of the LFS is the CPS. While CES and CPS employment estimates tend to be close to each other, discrepancies sometimes occur.