The shocking truth about an IMF working paper: Labour market mobility in Canada

Travis Fast

Some mornings I wake up hoping that my capacity to predict the “robust” findings of liberal economists will have dissipated and that something counter intuitive and not part of the conventional liberal economic wisdom will appear. To that end I started reading through IMF reports this morning, hoping that I could not pre-scribe the predictable findings on central questions like the degree of labour mobility and its relationship to unemployment. To wit, I came across a working paper entitled Shocking Aspects of Canadian Labor Markets. Imagine my thrill. However, right from the summary I realized that the economists who wrote the paper fancied themselves as persons plain with whit and irony. The conclusion reads :

Labor markets within Canada seem to become more flexible as one moves to the
west. Migration plays a much more important role in labor market adjustment in the western
provinces than the Atlantic ones. Turning to the central provinces of Ontario and Québec, the
evidence suggests that Ontario has a significantly more flexible labor market than its
neighbor, consistent with microeconomic evidence on migration. Further analysis indicates
that migration appears to be the main process through which labor markets adjust over time,
with real wage differentials being a minor factor. Finally, the adjustment process appears
relatively similar across macroeconomic disturbances. In short, labor adjustment appears
very different east and west of the Ottawa river.

Shocking hey. The paper not only confirms liberal economists predeliction for blaming the victim but also the long held stereo-types in the Canadian body politic. And just how do they derive such a bold conclusion? Easy they construct their metric of labour mobility by dividing total net regional migration by total regional population. Eh voila, those bloody Eastern Canadians are a bunch of sedentary hillbillies who refuse to move to where the jobs are .

I have long been a skeptic of conventional economic wisdom; and even more so of conventional regional stereotypes so I decided to check the data. As far as the authors’ constructed metric of mobility goes it is correct. ‘Correct’ as in they got the numerator and the denominator right. But the metric is flawed. Why net migration? Surely the question concerning mobility is one of an individual’s willingness to leave the province or region . That is, what we want to know is how many persons as a percent of the total population are willing to leave their province of origin. How many persons who arrive in that region are of no consequence to the question of mobility because mobility is a question of the incidence of out migration. Net migration is therefore a lousy metric.

Let me hammer home this point with a little more force. Why should BC gain points on the mobility meter because there are more individuals entering the province than leaving? What do the truck-loads of senior citizens fleeing the Prairie winters have to do with labour mobility? ‘Nothing’ is the reasonable answer unless of course one is plumbing for a confirmation of the conventional wisdom.

The only reliable semi reliable statistic for mobility is out migration; flawed as it is. Why flawed you ask? for the reason just outlined above. Why should Saskatchewan gain points for labour mobility if the bulk of out migration is from the retirees moving to the relatively balmy lower mainland of British Columbia. Indeed, for these reasons we should be more skeptical of central and prairie province out migration figures as they do not distinguish between workers and retirees.

But let me run the the out-migration as a percent of total population for the regions as a metric of labour mobility. Below the regional statistics mirroring the IMF working paper on regional aggregations are presented.

What a difference a metric can make! By this metric labour mobility is higher in the East, lowest in Central Canada and middling in the West. All this of course makes very good sense to a Marxist or heterodox political economists. Labour is a special kind of commodity which is doubly free. Workers are free to contract with whichever employer they so chose but chose they must because they are ‘freed’ from the means of production at the same time. So within this frame one would be led to predict that even though the bonds of family, community and culture be strong there is simply no substitute for a job. That is, that individuals leave their place of origin at all is a testimony to the relative power of labour markets to “make all that is holy profane.” Thus, it is no surprise to Marxist economist that mobility should be greatest in the areas of relative economic stagnation and less so in those regions of relative economic stability. Indeed, it is but the history of immigration to Canada.

If we disaggregate the regions into their constituent parts the picture remains, by and large the same, although it is curious that central Canadian mobility rates, i.e., Ontario and Quebec, should be declining given that in the 2000s the hot labour markets are decidedly in the West. What then are the results of this brief study of mobility? The exact opposite of the conclusions drawn in the IMF working paper study. Labour mobility has tended to be highest in the provinces with stagnant economies and lowest in the provinces with stable employment. As the relative economic fortunes of the regions has shifted so too has the degree of mobility. Although this result is tempered by the most recent data. As labour mobility is highest in East despite the oil induced boom in GDP growth and lowest in the central Canadian provinces despite the decline of the manufacturing sector and aggregate unemployment rates. The east can be excused owing to the fact that an oil price led boom has not translated into a white hot labour market. That is, relatively high unemployment remains and that, as any Marxist political economist would predict, is a good determinate of outward-migration. That the Western provinces still place second despite white hot labour markets and central Canada still places third in the new millennium, is a plague upon all theoretical houses. So indeed there are, after all, some shocking aspects of Canadian labour markets to be explained.


One possible explanation for the low mobility in Ontario could be its history of relatively high levels of employment which means that when workers are laid off they tend to be able to both qualify for EI and eventually find new employment in the province.  In any case, we would need better data to see what internal provincial migration looks like.  It could simply be the case, given the size and diversity of both Quebec and Ontario’s economy, that workers do not relocate to other provinces but relocate to to other regions within their respective provinces.  That is, surely moving from the Sioux to Toronto ought to count as mobile labour.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s