Falling Poverty Among the Elderly – A Canadian Success Story

I was recently trolling through Statistics Canada (LICO – After Tax) data on CANSIM and discovered to my slight surprise that the incidence of poverty (or low income as StatsCan politely prefers to describe it) has fallen very significantly among the elderly (age 65 and over.) I knew the rate was relatively low and had been falling, but it has declined very significantly, from 11.3% in 1989 (the last cyclical peak), to 9.8% in 1996 (when the poverty rate for all Canadians last peaked), to just 5.4% in 2006 (3.4% for elderly men, and 7.0% for elderly women.) Just 219,000 elderly Canadians lived in poverty in 2006.

Those figures are so low that Canada even looks great on the seniors poverty front in international comparisons (where poverty is defined differently as having income of less than one half the national median, again adjusted for family size.) The latest Luxemburg Income Study comparative data show, somewhat stunningly, that the poverty rate among the elderly in Canada is now even lower than in Sweden (5.4% vs. 6.6%) and hugely below the United States (21%) , the UK (16%) and most large European countries. We might even be able to stake a credible claim to having the lowest rate of poverty among seniors in the world (though these international compariosns are fraught with methodological cautions. http://www.lisproject.org/keyfigures.htm

Meanwhile, compared to 1989 – the most recent cyclical peak – poverty has actually increased for the working age Canadian population age 18 to 65 (from 9.4% to 11.3%) and has fallen only slightly among children (from 11.7% to 11.3%.) This is despite the fact that unemployment is significantly below the 1989 level, and reflects the growth of more insecure work and stagnant or falling wages at the bottom.

The main reason for the low poverty rate among seniors is Old Age Security plus the Guaranteed Income Supplement which currently provide a guranteed minimum income of $13,731 to singles, and $18,549 to elderly couples. These amounts are not too distant from the LICO After Tax cut off lines – $17,568 and $21,381 for singles and couples respectively in large urban areas, and lower in smaller cities and rural areas.

StatsCan doesn’t provide the data on CANSIM but the poverty gap among the elderly – the amount needed to push them all above the poverty line – would likely come in at significantly less than $1 Billion per year, a more than affordable proposition for the federal government.

That said, OAS/GIS has been in place a long time and – while matching inflation – improvements to it would not seem to explain the halving of poverty rates for Canadian seniors since 1989. A large part of the reason is likely the fact that today’s elderly couples at the lower end of the income spectrum -compared to those of 1989 – are more likely to be collecting two CPP cheques, the delayed result of women entering the workforce in big numbers back in the 1970s and 1980s. More may also be collecting two employer pensions. Many lower income seniors likely accumulated some significant wealth from house price increases.

Another factor may be that more of today’s elderly are working – 8.6% in 2007 compared to a low of 5.9% in 2001 – though employment rates for older workers are higher for the more highly educated rather than those at greatest risk of poverty.

Pension experts like John Myles and Bob Baldwin have underlined the important point that the incomes of today’s elderly reflect the economic circumstances and policies faced by that age cohort going back many years. They are still likely still benefitting from the maturation of the CPP/QPP and from high levels of private pesnion coverage of workers in the 1970s and 1980s. In other words, the fact they are doing well may be more a residue of the past than a reflection of anything we are doing particularly well today.

It would be useful to try and figure out more clearly what is going on. My suspicion is that falling private pension coverage, stagnant wages for the bottom 80% of Canadians, and high debt levels among the least affluent mean that non affluent baby boomers entering retirement may soon start to bump up poverty rates among seniors.


  • Yep! Canada has done a great job on old age poverty. If you look at the history, it can be explained but it’s mostly historical accident (no one had a grand plan). But who cares?

    We spend about the same (very low) share of GDP on public pensions (around 5%) as the U.S. but our old age poverty rates rival those of Sweden while those of the U.S. are four times higher.

    The big question is whether we can replicate this story with other marginalized groups (e.g single moms). And if not, why not?

    In the 1960s, old age was a virtual synonym for poverty. I think it was Michael Wolfson who said that what Canada has done with old age poverty totally confounds the biblical claim that “the poor will always be with you.”

    John Myles

  • The age group most threatened by poverty is one which gets little attention. It’s not helped by the fact that StatsCan doesn’t (seem to) use a more refined breakdown of that 15 – 64 group.

    The greatest number of people in poverty are those between 50 and 64 years of age. A grossly disproportionate number of them are women.

  • Michael Mendelson

    A terrific contribution as always Andrew.

    It would be interesting to see a careful decomposition of the factors contributing to the income of the elderly, especially over time. I assume that a huge part of our public expenditure goes to tax breaks (and now another mega-billion due to income splitting) that are mainly for the more well-off, so the actual amount of public expenditure for those with modest and lower incomes might be quite small. Someone, other than me, should do this work, or maybe it has been done.

    On Chrystal’s comment: you are so right! I would add only mainly women PLUS people with severe disabilities. This is why we (at the Caledon Institute) have developed what we call a new architecture for income security focused on people of working age. One of the elements of this is a program for those with severe disabilities which would be structured like the OAS/GIS and at OAS/GIS amounts. We have called this a ‘Basic Income.’ Presumably for these people at least poverty would all but disappear as it has among the elderly. This is about 300,000 to 500,000 people most of whom are now on welfare or the equivalent of welfare.

  • Michael makes an important point about the role of tax expenditures in re-distributing income upwards. The CCPA work is showing how effective this has been.
    For the 65 plus group, the results seem to say that targeting works. If we were to introduce an income security programme for single parents under 25, perhaps we could eliminate the mistreatment of young women and their children.
    Before I join the backslapping about poverty reduction in what shall soon by my age group, I would like to see some figures about the 65 plus near poor.
    No LICO style statistic is precise enough to tell us what happens to those just above the cutoff, to differentiate them from those below the cutoff.

  • It would also be very interesting to have a disaggregated look at the 65+ poverty levels – as increasing numbers of racialized community members make up this cohort – many of course being newcomers to Canada and thus having diluted or no access to certain of the old age benefits and income supports !?
    Does anyone know if this sort of analysis has been/is being done ?

  • I did a decomposition of changes in senior incomes a few years ago though it’s now a bit dated (1980-1995). The big gains after 1980 came from the maturation of the C/QPP (not OAS/GIS).

    With respect to Duncan’s concern (are we getting seniors just above the poverty line), the answer is no. Between 1980 and 1995 about three-quarters
    of the change reflected in the LICO rate and all of the change in the LICO-IAT rate was the result
    of an increase in the percent of seniors whose incomes were at least 1.5 times the low income cut-offs.

    Note too that the poverty gap of seniors below the LICO is quite small compared to other low income populations.
    John Myles. 2000 “The maturation of Canada’s retirement income system: income levels, income inequality and low income among the elderly.” Canadian Journal on Aging. 19 (3): 287-316 (Also available in Statistics Canada’s analytical paper series (#147).

    We also did a recent paper tracking “income security” (income replacement) for the “average worker” with tax data that shows very positive results (Stats Can analytical studies paper number 306). We have another, more reader-friendly, version of that paper in the works if anyone is interested.

    In the late 1970s old age poverty in Canada was in fact higher than in the U.S. and all the predictions I made during the Great Pension Debate (“Canada’s seniors were among the poorest in the world and things would never get better within the existing system”) have turned out to be wrong.

    We do show, however, that variation in income replacement rates for the average worker is the product of uneven coverage in occupational pensions and RRSP benefits (Michael’s point about our tax expenditures for these programs). While the average replacement rate is almost 80 percent, about a quarter of those in the middle quintile at age 55 had income replacement rates at age 75 of less than 60 percent.


Leave a Reply

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