Friday, May 14, 1999
Drawing the poverty lineAndrew Coyne
National Post
For many years, the debate on poverty in Canada -- what is it, how much is there, and what should be done about it -- has been trapped between two poles. The right defines poverty as absolute privation; finds, on investigation, that few people are actually starving in Canada; and so concludes that nothing much needs to be done. The left defines poverty as relative inequality; proves, in repeated surveys, that some people are poorer than some others; and so concludes that nothing much has been done.
This is not helpful. Yet such is the suspicion between the two camps that any attempt to come to a shared definition of poverty has tended to fall to the ground. Left-wingers view any deviation from the traditional, relativistic measures, such as Statistics Canada's Low Income Cutoff (LICO), as so much statistical trickery, aimed at defining poverty out of existence. Right-wingers, for their part, persist in the delusion that their own "market-basket" measures, such as the Fraser Institute's Basic Needs Line, are the sole objective standard of poverty, permanent and universal.
It doesn't have to be this way. We do need to define poverty. But the issue is not how high or low to set the line, nor even whether it involves relative or absolute criteria. It is whether it is useful: that is, whether it allows us to measure our progress, against past performance and against other countries. For without a useful measure of progress, we cannot tell whether our chosen methods to combat poverty are working.
The problems with LICO are well known. More than relativistic, it is wholly arbitrary, defined as the level of income at which a family, if it were to spend as much as the average family spends on food, clothing and shelter, would have consumed at least 55% of its income. Why 55%? Because it is 20 percentage points more than 35%, the proportion of its income the average family spends on these necessities.
The problem with this, as with all attempts to measure poverty in relative terms, is that no matter how rich everyone in society became, the number of those defined as "poor" could conceivably remain unchanged. By such means are we told that Canada has higher numbers in poverty than some Third World states, or that there are no fewer Canadians living below the poverty line now then there were a generation ago -- assertions that are manifestly contrary to any common-sense understanding of the term.
But basic-needs measures do not, by themselves, resolve this problem. That is, what goes into the market basket of goods and services defining the decent minimum any family should be expected to live on will itself involve subjective, if not relative, valuations. We know what level of nutrition is necessary to keep the body alive, but we are far from agreed that that should define the lowest standard of living anyone should endure, least of all in a society as rich as Canada's.
Nevertheless, even if relative considerations creep in, it is progress to measure poverty in terms of needs, that is by reference to the actual conditions of people's lives, rather than as simple straight-line comparisons with average incomes. If the Fraser Institute's definition is too stark, then we might set ourselves to debating what other expenditures should go into the definition. As it happens, some efforts are being made in that direction, notably in a remarkable new study for the Canadian Council on Social Development.
The work of researchers David Ross and Paul Roberts, the study begins from a simple premise: "What if producing healthy children was the main objective of anti-poverty efforts in Canada?" Why children? Why focus on child poverty rather than poverty? Because for adults, poverty is as often as not a temporary condition. In poor families, however, especially single-parent families, it can persist for many years, in which time a child's life prospects may be permanently altered.
Drawing on recent findings from the National Longitudinal Survey of Children and Youth and the National Population Health Survey, the study shows that across a wide range of development measures -- health, behaviour, math and reading ability, participation in sports and other activities -- children in poor families display consistently worse outcomes than others.
How poor? Again, the data are remarkably consistent. In chart after chart, measuring one factor after another, outcomes improve sharply as family incomes rise toward the $30,000 to $40,000 range (for a family of four), whereon they tend to level off. That is, $30,000 or so seems to be the level at which a sort of rough equality of life chances is achieved -- the minimum necessary, the authors suggest, to escape "poverty of opportunity."
This is significant. Another sterile debate of longstanding has pitted "equality of opportunity" versus "equality of outcome." But what this new research indicates is that, to some degree, equality of outcome is equality of opportunity. How about we start the debate there?
Copyright © Southam Inc.