What Inter-Provincial Migration Trends Can Tell Us About Good Governance

It turns out we move a great deal less than our American neighbors

Government policies have consequences. Among them is the possibility that they might so annoy the locals that people actually get up and head for the exit. Given how parting can be such sweet sorrow (and how it’s a pain to lose out on all that revenue from provincial income, property, and sales tax), legislatures generally prefer to keep their citizens on this side of the door.

Nevertheless, migration happens. And when enough people do it at the same time, they sometimes leave economic and social clues behind waiting to be discovered. This graph represents net migrations since 1971 into and out of the four largest provinces:

It may just be possible to make out some broad patterns here. Quebec has never had a net inbound migration year (although there’s been plenty of immigration to Quebec from outside of Canada). But nothing matches the mass exodus of anglophones due to concerns over language and separation in the 1970s.

Curiously it seems that Alberta and British Columbia received far more migrants than Ontario around that time - although the actual numbers tell us that they were more likely to have come from Saskatchewan and Ontario than Quebec. By contrast, most disillusioned Quebecers found their way to Ontario. Besides the 70s, Alberta also enjoyed inbound spikes in the mid-90s, mid-00s, and early 10s. And it looks like they’re in the middle of another boom cycle as we speak.

The real value of all this data however, is in using it to test causation hypotheses. In other words, can statistical analysis tell us what it was that caused the migrations? And are some or all of those causes the result of government policy choices? Here are some possibilities we’ll explore:

• Household income trends

• Government debt

• Crime rates

• Healthcare costs

• Housing costs

Right off the top I’ll come clean with you: there’ll be no smoking gun here. I could find no single historical measure that came close to explaining migration patterns. However I was able to confidently discard some theories. That’s a win I guess. And other numbers did hint to intriguing possibilities.

Inter-provincial variations in household income, crime rates (specifically murder rates), healthcare costs (including prescriptions, eye care, and dental care), and even housing affordability had no measurable impact on migration. This was true for both correlation coefficients and lag analysis (where we looked at migration changes in the years following an economic event).

Rising unemployment had, at best, a minimal impact on outbound migration. And even then, it was only noticeable for Alberta and Prince Edward Island.

Of all the metrics I explored, the only one that might have had a serious influence in migration was provincial government budget deficits.