More on the Myth of Big Government – Canada vs US

Erin’s recent post prompted me to read Ferris and Winer’s interesting piece on the size of government in Canada and the US. The underlying data for the article have been usefully posted by the authors at (You’ll have to find the spread sheet posted at that site under author’s papers, and then look at the sheet for functional structure.)

As Erin corectly notes, the big message here is that the huge cuts to government spending in Canada in the 1990s – combined with modest increases in the US under Clinton – markedly cut the government spending gap between Canada and the US to the point where it is now close to non existent. If one accepts the authors’s adjustments to the commonly cited government spending as a share of GDP data, which I have no reason to quibble with, the gap comes close to disappearing. I agree with Erin that this is a fact that Canadian progressives have to take seriously, and that it is obscured by our tendency to invoke mythically large Canadian distinctiveness vis a vis the US.

Two caveats, however.

First, non defence related government spending is, according to the authors estimates, still more than 5% of GDP higher in Canada than in the US (37.5% vs 32.7% in 2004.) That’s a pretty significant difference, though it is very, very sharply down from the huge 16% of GDP (50% vs 34%) peak gap in 1994.

Second, I’d be inclined to second guess the ‘fact’ that transfers to persons are now higher in the US than in Canada- 11.9% vs 10.1% in 2004. (Canadian transfers to persons have fallen from a peak of 13.5% in 1993.) That’s a bit hard to square with the fact that Canada still has a much more sigificant set of social assistance programs than the US, and that OAS/GIS and the C/QPP as well as EI are at least broadly comparable to US programs. US military pensions and allowances may play a role here. Also, it is important to take into account that recent expansions to transfer programs in Canada – notably child benefits – take the form of refundable tax credits which are functionally identical to spending but are treated in the public and I think national accounts as a debit from tax revenue rather than as an addition to government expenditure.


  • Interesting stuff, Erin and Andrew.

    I’m not sure how income transfers were dealt with in the study, but in both the national accounts and the federal budget, children’s benefits are now included as expenditures. They used to be a non-transparent deduction from tax revenue, but a few years ago the feds made the switch.

    I’m wondering whether the difference in transfers has to do with the US EITC.

  • Andrew, thanks for the post and for pointing us to Stan Winer’s paper and Excel spread sheet. The spread sheet is a great resource and I have written a note to Stan Winer to thank him for making it available.

    In 2001 the Caledon Institute did a very careful study of child-related benefits in the tax-transfer systems of Canada, the UK, the US and Australia. I did a small follow-up paper in 2003 which is posted on the Caledon web site (‘Child Benefits Levels in 2003 and Beyond: Australia, Canada, the UK and the US,’ The Caledon Institute of Social Policy, 2003). I was surprised to find that for many ‘working poor’ families child-related benefits in the US were higher on a purchasing power parity basis than those in Canada. I have not done a recent up-date, and I suspect that Canada is now a little higher than the US, but not much.

    The US Social Security system provides higher benefits to retired labour force participants than does the C/QPP, but without a quasi-universal floor for the poor or for those who do not have a good contributory record such as we have in the OAS/GIS. I believe that the gross total retirement benefits paid through the state in the US are likely higher than those in Canada, but whether or not an individual’s benefits would be higher would depend upon where the person is on the income scale, family structure and history.

    The US also has food stamps, which is an odious program but neverthless pays out large benefits. Similarly, there is a growing housing voucher program.

    In general, the pattern seems to be that Canada is more generous for those with little or no income who are not in the work force: the US is more generous for the so-called working poor. Of course, most of the working poor in the US have no health insurance, so they are in any case not better of at all than thier Canadian counterparts. (As well, I believe that wage supplements for the working-poor are an attempt to socialize labour costs and keep wages down, although I recognize that many progressive economists do not fully concur in this view.)

    Overall, I suspect that the tax-transfer system in the US is about the same size as that in Canada as a percent of GDP, but that Canada’s is more redistributive.

Leave a Reply

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