We have heard a lot about the squeezed middle over the past few years. For the most part, the statistics have focused on the fact that average income have been stagnating (or declining) in both the U.S. and Canada.
A close look at the numbers reveals that the number of middle income earners in Canada is shrinking. Thing is, I’m not exactly sure that’s as dire as it sounds.
Rather than looking at incomes, the NELP researchers looked at employment growth by occupation. Since some occupations (hedge fund manager say) pay very well, and others (fast food counter server) pay fairly poorly, the relative growth rates of each can give a good picture of relative gains and losses amongst income groups, and a measure of inequality.
Looking at data on U.S. employment by occupations, the authors came to the conclusion that since the U.S. recession, employment gains have been concentrated in lower-wage occupations. The NELP researchers used monthly data on earnings in 366 separate occupations, and divided those into high-middle-low categories depending on the average pay at the start of the recession. They then tracked those occupations over the course of the last four years.
Their results suggest that a lot of downward mobility is going on in the U.S.. What they found was that U.S. in the ‘middle’ occupational groups have indeed lost ground since the recession. During the recession, employment in middle wage occupations accounted for 60 percent of total U.S. job losses, but since the recovery have only accounted for 22 percent of the gains. The job categories that have been gaining are in lower wage occupations, which were 28 percent of recession job losses, but have accounted for 58 percent of recovery growth. Higher wage occupations have basically come out even, accounting for 19 percent of recession job losses, and 20 percent of recovery growth. It’s not a lovely picture. The U.S. has not added enough jobs (the unemployment rate is stuck above 8 percent, compared to 5 percent pre-recession), and apparently the jobs they are adding are not particularly well-paying.
But that’s the U.S.; what about Canada?
Canadian occupational data is not available monthly, but Statistics Canada does a comprehensive survey of employment and earnings by industry. Accordingly, I looked at weekly earnings (including overtime) in Canada’s twenty-one major industry groups in July 2008, the month in which Canada’s economic output peaked (employment actually peaked slightly later, in October of that year). Earnings ranged from a high of $1,545 a week in mining and oil and gas extraction, through to a low of $336 a week in accommodation and food services. I then grouped the industries into high-middle-low categories, and tracked how each group has fared over the last four years.
And really, Canadian employment has fared pretty well, particularly by international standards, over the past few years. For the four years ended in June 2012, total employment is up by 2.6 percent. That may not sound like much, but it compares favorably to a loss of 3 percent in the U.S..
The industries that have added the most workers are the ones that pay the most money. Jobs in Canada’s highest paying industries have increased by 4.5 percent since mid-2008, far more than the average. The energy boom has played a big part in this – jobs in mining and quarrying industries have risen 8.6 percent over the period – but there are other strong gainers. Employment in construction is up by 6.5 percent, in finance and insurance it is up 4.4 percent, and in public administration it is up 4.1 percent. Alone among the seven sectors, only the ‘management of companies’ category posted a decline .
The seven industries that pay the least also added workers over the four years. Employment in these industries rose by 2.1 percent in total, although there were huge variations by industry. The strongest growth in the subsector – and in fact, the strongest growth in all Canadian industries over the period – was in the health care and social services category (which encompasses private, not public services), which was up whopping 10 percent. The two runners-up are Arts and Entertainment (4.3 percent) and Accommodation and Food Services (3.1 percent). Retail trade, real estate and leasing and administrative support industries all lost workers over the time period.
That’s all of the good news. The bad news is that the industries that typically employ ‘middle income’ earners have taken a big hit.
Employment in the middle category of industries dipped 3.3 percent over the four years ended in June 2012. Predictably, the manufacturing industries fared the worse. Durable goods manufacturing (which is heavily dominated by auto and related industries) now employs 11.8 percent less people than four years ago, while non-durable manufacturing is down 10.4 percent. The U.S. recession, a high Canadian dollar, a push to improve productivity and pare back high-wage workers – all the reasons for the adjustment have been well-documented. In fact, of the seven so called ‘middle earning’ industries, only Educational Services (which refers to private sector educational institutions, not public) gained ground over the time period.
So what to make of the data? For sure, you can look at it and say that middle income earners have been squeezed, which they have. By my calculations, pre-recession middle-earning industries accounted for 31.2 percent of total employment, and now that total is just 29.4. Then again, the top earning industries have increased their share of employment from 25.1 to 25.6 percent, and the lowest paying ones from 42.4 to 42.2 percent.
So here is the thing: is it a bad thing that there are less people in middle earning industries if there are more people in higher earning industries? We are not talking about the U.S. where the labour force is clearly downward mobility; in Canada, there have been a lot of jobs created in industries that pay a lot of money. Some of those jobs require specific education and skills, but that is not a bad thing. And true, some of the workers who lost their jobs in the middle earning industries will not be able to find work in the industries that are growing the quickest.
There is a re-shuffling of the deck going on, for sure. It seems, however, that we need to take a very close look at the statistics before adjusting any policies to interfere with the results.
 Data is from Statistics Canada’s Survey of Employment, Payrolls and Hours. The more commonly used ‘Labour Force Survey’ also provides a breakdown by industry, but does not have accompanying wage data.
 The break down was as follows: High earning industries are Mining quarrying and oil extraction; utilities, professional scientific and technical services; management of companies and enterprises; public administration; construction; and finance and insurance. Middle earning industries are Durable goods manufacturing; information and cultural industries; forestry, logging and support; wholesale trade; non-durable manufacturing; transportation and warehousing; and educational services. Low-paying industries are real estate and rental and leasing; health care and social assistance; administrative and support; other services (excluding public administration); arts, entertainment and recreation; retail trade; and accommodation and food services.