The Exploding Canadian Income Gap

Statistics Canada today released an excellent study of  high incomes and inequality -

Thanks to Michael Wolfson, Brian Murphy and Paul Roberts for getting this powerful data out into the light of day.

No big surprises here – the top end grabs a disproportionate share of all income, and their share has been growing apace. The tax data used in this study demonstrate the point more clearly than StatsCan income surveys, which tend to miss the very high end of the spectrum.  I find the study particularly useful for giving us much more insight into the changing distribution of market (employment and investment) income at the individual level, though family level data are provided as well.
Among individuals, the top 1% starts at $181,000 in pre tax income (in 2004), though average income for this group is $429,000 because incomes at the very top are extremely high (the top 0.01% or 1 person in 10,000 made more than $2.8 million in 2004, and averaged $5.9 million.)

The threshold of the top 1% of individual tax filers in 2004 was 7.4 times median income of $25,000 in 2004, up from 6.0 times as much in 1992 and 5.7 times as much in 1982.  The threshold of the top 0.01% was 115 times as much as the median in 2004, compared to “just” 57 times as much in 1992 and 55 times as much in 1982.

The share of all taxfiler income going to the top 1% has soared from 8.6% in 1992 to 12.2% in 2004, and the share of the top 0.01% has doubled, from 0.9% to 1.7% (though the authors show that the income share of the very top pales in comparison to that in the US.)  This is mainly a story of exploding wage inequality, though capital gains also contribute play t 0ugh troy
Another way of getting at fast-rising market income inequality is to compare the rise in real incomes between 1992 and 2004.  Real incomes were essentially unchanged for the bottom 80%, but the average inflation-adjusted income of those in the top 1% jumped from $268,000 to $429,000, and the average incomes of  the top 0.01% soared, from $2.5 million to $5.9 million.

The data show that the top 1% are disproportionately older men living in big cities (one third live in Toronto.)

The study also shows that the income tax “burden” on top earners is surprisingly light – the top 5% of taxfilers in 2004 received 25% of income, and paid 36% of income taxes (the only progressive part of the tax system.)  Effective tax rates for the top 1% and 0.01% have fallen quite sharply since 1992 . For the top 0.01% – those averaging $5.9 million in 2004 – the effective tax rate fell from 42% to 31%.
In summary, an extremely useful documentation of the grotesquely unequal  distribution of  Canadian income, and the big shift of before and after tax income to the very top in recent years. Grist for the mill of progressives in the debates ahead.


  • This study, like Saez and Veall, is an excellent contribution, and good reminder of just how much of the inequality story is at the very very top of the distribution. Statistics by quintile, and even those rare occasions with deciles, mask this. I know many people who make the top 5% (above $89,000 total income) and while they are doing comparatively well, they are not “rich”.

    I do have a few methodological quibbles about the study. First, the choice of years (1982, 1992 and 2004), the first two of which are troughs. It would have been nicer if we had business cycle peaks to compare (i.e. 1979, 1989, 2000 and 2004 as a final year).

    Second, the family estimates are undermined by the failure to adjust for family size. This is a pretty straightforward thing to do with the T1FF, and would give a better sense of the inequality picture at the family level.

    Finally, the tax discussion is only about income taxes, which are the only consistently progressive tax. Even there, we see some regressivity within the top centile. But if all taxes (or even just consumption and payroll taxes) were added into the mix, the pattern would change. As is, it gives the impression that the system works, that high income people pay a disproportionate share of the taxes. There is evidence (that I will be presenting soon) that this is just not so.

  • Here are two interesting articles on related issues:

    1. A survey done in the US stating that an increasing number of people view society as being divided between camps of “haves” and “have-nots”.

    2. An article in the Canadian Journal of Economics (vol. 40, n. 3) from this summer, which basically says that the rising trend in inequality started earlier than we usually assume (early in the 1980s), and that the redistributive role of taxes has been weakening since the early 1990s (apologies if it has already been mentioned elsewhere in the blog).

    “The tale of the tails: Canadian Income Inequality in the 1980s and the 1990s”, by Marc Frenette, David Green, and Kevin Milligan.

Leave a Reply

Your email address will not be published. Required fields are marked *