Social Policy in Ontario

Turns out income mobility in Canada isn’t as impressive as we thought

Posted on June 23, 2016 in Equality History

—

NationalPost.com – FP/FP Comment
June 22, 2016.   William Watson

Socialists will celebrate when, at the end of this season, legendary Dodgers’ baseball announcer Vin Scully retires after 65 years at the mic. The other day, as a Venezuelan player stepped to the plate, Scully offered this analysis of their preferred social system: “Socialism failing to work, as it always does, this time in Venezuela. You talk about giving everybody something free and all of a sudden, there’s no food to eat. And who do you think is the richest person in Venezuela? The daughter of Hugo Chavez. Hello! Anyway, zero-and-two….”

Though Forbes doesn’t have Ms. Chavez on its list of billionaires, there’s little doubt that, in many places, who your father is does determine how you do in life. Since 1999, when two StatCan researchers published one of the first studies correlating fathers’ and sons’ incomes, Canadians have taken satisfaction from the fact that, here, the correlation was relatively low, only about 0.2, whereas in the U.S. some studies put the correlation as high as 0.5. Now, new Statistics Canada research (by Wen-Hao Chen, Yuri Ostrovsky and Patrizio Piraino) puts us at 0.32, not quite U.S. levels, but closer than we thought.

Intergenerational income correlations are like golf: lower scores are better. If you’re running a completely feudal society and every son has the same status and income as his father, the correlation is a perfect 1.0: the higher his income, the higher yours is.

But if your and your father’s incomes are completely unrelated, the correlation is zero. In theory, the correlation could be negative. High-income fathers might systematically have low-income kids. That’s what China’s Cultural Revolution aimed to do: Successful people were sent to the country to live peasant lives and their “bourgeois” kids were stigmatized, which presumably made things harder for them.

Countries with lower father-son correlations aren’t necessarily more meritocratic. To judge that you’d need much more than this one number. But in this field, lower scores are seen as better and for two decades Canadians have been thinking we’re more like the low-correlation Nordic countries than the U.S. So, score one against passive-aggressive Canadian hubris.

The new StatCan study doesn’t find any errors in the 1999 one. It just uses more complete data. In theory, before calculating the correlations you’d like to know the lifetime income of fathers and sons. Because we haven’t been tracking these things forever, we’ve only a few years of income for both. Having more years now than the old study makes a difference: it seems how sons fare in early earning years isn’t always a good indication of how they end up.

What are the implications of Canada having higher intergenerational correlation of incomes than we thought?

Looking at the U.S. comparison, some people have been tempted to think: Bigger government (Canada), lower correlation of fathers’ and sons’ incomes. Ergo: big government good. That’s always been a dicey conclusion. In fact, the U.S. has a huge welfare state; it just prefers tax incentives to government cheques. Moreover, several non-Nordic European countries have big states, too, but also relatively high correlations. Now it turns out we don’t have as much income mobility as we thought, despite our big government.

Score one against passive-aggressive Canadian hubris.

The higher father-son coefficient also slightly undermines “The Great Gatsby Curve,” first drawn by the University of Ottawa’s Miles Corak, but named by Princeton’s Alan Krueger. It shows that the more inequality a country has to begin with the less income mobility it displays across generations (although Jay Gatsby, who rose from nothing, was mentored by two patrons sitting high atop the post-Great War U.S. income distribution: inequality actually helped him move up). Canada fit that curve. We had less inequality than the U.S. and, we thought, more income mobility. We still have less inequality but now it seems we don’t have quite so much income mobility.

The new study is also interesting for father-daughter results that show less of an influence of fathers’ income on daughters’ incomes than on sons’ (0.23 rather than 0.32). That’s a puzzle, possibly having to do with whether the data effectively capture women’s peak earning years.

The study also shows differences in mobility rates in different parts of the income distribution. At the very bottom, among the lowest earners, the correlation between fathers and sons is actually pretty low, which is what we want. Sons can’t move farther down the ladder than fathers already on the bottom rung. So the low correlation means they’re moving up. At the top end, however, the correlation is higher, though still less than 0.5. That kids of people at the top are less likely to fall isn’t that surprising.

Reading such studies always teaches me that what’s going on behind these simple numbers is very, very complicated. I hope policy-makers draw the same caution-inspiring conclusion.

< http://business.financialpost.com/fp-comment/william-watson-turns-out-income-mobility-in-canada-isnt-as-impressive-as-we-thought >

Tags: economy, ideology, participation, poverty, standard of living, youth

This entry was posted on Thursday, June 23rd, 2016 at 12:00 pm and is filed under Equality History. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.