The good news on welfare caseloads in Canada should be universally applauded. According to StatsCan, the numbers of non-handicapped recipients fell from 3.1 million people in 1994 to under 2 million in 2000, and the bill for social assistance for the able-bodied went down from $14.3 billion to $10.4 billion. Significant disagreement can be expected, however, about the cause of the decline and its meaning for poor people.
Numbers can deceive and these ones must be put in context. First, their time frame makes the improvement look better than it actually may be. The StatsCan data, which actually measures from 1992 to 2000, show caseloads peaking in 1994 as North America was emerging from a particularly nasty recession. Second, many jurisdictions reduced benefit levels and tightened rules for eligibility. Both actions have disincentive effects on those who might otherwise have applied for assistance.
An indictment based on the second factor has already been handed down by one Toronto-based group of advocates for the poor. “The sharp drop is not a result of significantly less people living in poverty or in need of assistance,” according to a spokesperson, “. . . [but] of governments making it significantly more difficult to be eligible for assistance.” She cited a rule in Ontario, where a single mother whose monthly income exceeds $931 a month can’t collect welfare.
She has a point, but falls into the intellectual trap of using taxable income as a measure of poverty. Most poor people have many other resources than reflected by that bottom line, from families, non-governmental organizations and charities and from unrecorded, or underground economy income. Real poverty in Canada – as measured by a more reliable indicator, whether one’s basic needs for shelter, food and clothing are met – fell from 35% of the population in 1951 to just under 5% in 1994.
Economic expansion since then is often cited as the biggest factor in the subsequent decline of welfare caseloads. Although StatsCan’s number crunchers carefully avoid any attempt to suggest cause and effect, they do say that the incidence of social welfare participation “generally followed the business cycle at the national level.” The importance of family assistance in the mix is also reflected by the fact that caseloads over the decade rose slightly for single individuals, while they fell quite sharply for couples, with or without children.
A comparison of entry and exit rates from social assistance indicate that tightened eligibility requirements may have shaken a lot of people out the system in British Columbia, Alberta and Ontario, which did the most to introduce American-style restrictions. As StatsCan rather coldly puts it, “the stock of SA participants likely changed.” Alberta, for instance, rerouted adolescents leaving home back into scholastic and training programs rather than open-ended assistance.
The importance of steering “human capital” into productive activity is made clearer by the results of welfare reform in the United States after 1996, when successful programs in states like Wisconsin were replicated nationally. Overall poverty, child poverty, black child poverty, the poverty of single mothers and child hunger have all declined substantially since then. Contrary to conventional wisdom, the decline in welfare dependence has been greatest among the most disadvantaged and least employable group, single mothers. The most dramatic change in Canada was also experienced by single mothers. Their welfare participation rate declined from a peak of one-half in 1995 to a third in 2000.
But the belief that a rising economy is solely responsible for declining welfare rolls is not borne out by the American experience. The Heritage Foundation, a large think tank, looked at caseloads over eight periods of economic growth prior to the 1990s. In none of them did participation rates drop, and during two of them, the late 1960s and the early 1970s, the welfare caseload grew substantially. Only during the expansion the 1990s did it contract. “How was the economic expansion of the 1990s different from the eight prior expansions?” the researchers asked. “The answer is welfare reform.”
That is supported by the work of Dr. June O’Neill, former Director of the Congressional Budget Office, who analyzed changes in caseloads and employment patterns from 1983 to 1999. Her conclusion was that in the period after the enactment of welfare reform, policy changes accounted for roughly three-quarters of the increase in employment and decrease in dependence, while economic conditions explained only about one-quarter.
The StatsCan data seem to point in that direction as well. In Québec, where the rules changed the least for a major province, welfare participation rates remained high throughout the period.
The new numbers demonstrate that welfare reform works. It’s good for the people affected and good for the taxpayers who support the poor. Saskatchewan’s Minister of Finance, Harry Van Mulligen, who oversaw changes in his province’s welfare system in 1997 and spoke at the Frontier Centre in 2002, was asked what was most valuable about the effort.
“I think to reward the decision to go to work,” he answered. “At the end of the day, our economy and way of life is based on all of us working and contributing. If you support that, you support the decisions of people to move into the mainstream of our society and to be included, not excluded.”
