Project Link update

It's past time for my annual update for Project Link, my attempt to piece together the fragments of Statistics Canada's published data into coherent time series.

Statistics Canada's Attention Deficit Disorder means that I can never assume that the list of series that were current in one year will still be current in the next. Last year, I had to track down the replacement series for manufacturing and retail sales; this year I had to do the same thing for monthly exports and imports data. The base year for the economic accounts has also been updated from 2007 to 2012. I'm not going to complain too much about that last thing, though: these revisions go with the territory.

I've given myself the mission of adding more series each time I update the site. In 2017, I spliced 1953-1975 data from the Labour Force Survey to the current series. In 2018, I added income data, including estimates for Gross Domestic Income. This year, I'm going back to the LFS. One thing that jumped out at me from the 2017 update was that female employment rates showed an increasing trend starting in 1953. I never like it when I see trends that start at the beginning of the sample, because people are too quick to assume that the first observation in the sample is also the start of the trend. So this year, I pushed the LFS back to 1946.

Before the Labour Force Survey was set up, the only source for unemployment data was the surveys that the government sent out to the various labour unions, asking them to report how many of its members were unemployed. (Yes, really.) It started as a quarterly survey, and became a monthly survey in 1953. I had already deseasonalised the quarterly data and made monthly interpolations for the headline estimates for employment and unemployment rates; this update extends that to the main LFS estimates (employment, unemployment rates, labour force, participation rates and employment rates) for men and women.

So this is the biggest addition to Project Link this year:

It turns out that the turning point in female employment rates really was sometime around 1953. (Why?) I was expecting to see a decline in female employment rates immediately following the war (Chapter 17: In Which Rosie The Riveter Is Sent Home To Give Birth To The Baby Boom), but it's not something that jumps out at you. Of course, the first wave of the survey was in November of 1945, so maybe there was a drop earlier in the year that the survey missed. On the other hand, post-war demobilisation wasn't immediate, so there weren't yet many men returning home to claim jobs that had been taken by women. If anyone knows more about women in the labour market in the war and post-war years, I'd be grateful for some references.

Comments

[Frances Woolley]

Stephen -

Great post! Re your bleg:

The 1950s saw an acceleration of a trend that was already in place, rather than a distinct turning point. Costa's 2000 Journal of Economic Perspectives describes how white women's female labour force participation rates in the US and Great Britain rose steadily from the late 19th century onwards: here. This article by Maryam Dilmaghani here has a good bibliography that summarizes the Canadian literature. I don't know if anyone used the digitized old Canadian census data to extend the LFS survey data backwards yet, but I suspect you'd find similar trends to the one in Costa's article.

[Stephen Gordon]

I almost wrote "If anyone (and by 'anyone', I mean 'Frances') knows..."

Thanks for the references!

[genauer]

Thanks from me too