Skip to content

From the archives

The Trust Spiral

Restoring faith in the media

Dear Prudence

A life of exuberance and eccentricity

Who’s Afraid of Alice Munro?

A long-awaited biography gives the facts, but not the mystery, behind this writer’s genius

Spending Power

Can compassion and efficiency be combined in the use of public funds?

John Richards

Early Intervention: How Canada’s Social Programs Can Work Better, Save Lives and Often Save Money

James Hughes


232 pages, softcover

ISBN: 9781459408777

In pre-industrial society, the family was the primary source of social support. It still is in much of the world. However, voters in wealthy industrial countries have been willing over the last century to mitigate the extremes of a market economy via the welfare state.

Not that this willingness is boundless. At a time of severe fiscal crisis in Sweden in the 1990s, for example, the government appointed Assar Lindbeck, a respected economist, to review the country’s social programs. He prepared a lengthy report, which he later summed up with the quip that the welfare state is a wonderful innovation of the 20th century—provided the workers are all Lutherans and the administrators are all Prussians.

The opening sentence of James Hughes’s new book, Early Intervention: How Canada’s Social Programs Can Work Better, Save Lives and Often Save Money, might be interpreted in a similar light: “Canada’s social safety net is showing its age.”

Lindbeck and Hughes agree on the need to undertake, from time to time, fundamental reviews of social policy. But, as I discuss later, the two tackle the problem in decidedly different ways.

Hannah Wilson

Hughes has spent his career as a social worker in the trenches, running a mission for the homeless in Montreal, and as a deputy minister of social services in New Brunswick. He has a career that permits him to write an overview worth reading. In nine chapters he surveys the history of social policy in Canada, strategies to manage child protection, early treatment for autism, bullying in schools, healthcare improvements (particularly in treatment of mental illness), design of income support programs targeting the poor and housing the homeless. The concluding chapter develops his overarching theme: intervene early.

Hughes is at his most convincing when discussing concrete examples. A near-universal phenomenon in “settler societies” is a high incidence of child neglect among indigenous families faced with the cultural difficulties of integrating into an industrial society. Hughes has praise for the New Zealand early childhood intervention protocols, in the case of Maori. They engage the entire family and have a better record than most child protection agencies in avoiding the trauma of child apprehension. Early intervention via the extended family not only yields better outcomes, Hughes reports. It saves money.

Another example Hughes discusses at length is the Lovaas theory of intensive behavioural intervention with young autistic children. Treatment may readily cost over $75,000 annually per child. Expensive, but nonetheless a good investment he concludes. The benefits are the higher quality of adult life for the autistic child and lower costs of treatment when the autistic child is an adult. There is much debate surrounding the benefits of Lovaas treatment, but, again, Hughes may be right.

Hughes may also be right that “housing first” is the appropriate strategy to pursue for the homeless. The homeless typically have many problems over and above a lack of income to obtain housing. Frequently they suffer mental illness and addiction, and have many minor criminal convictions and few acquaintances to provide emotional support. Before attempting to treat the multiple problems the homeless often present, the housing-first strategy is to provide housing, with no—or very few—questions asked. This, too, is an expensive undertaking, but may be cost effective in the sense that the outcomes are at least as good as most alternative interventions, and the high cost of providing the housing is offset by avoiding costs such as hospital emergency visits, criminal behaviour, and so on.

At some point in reading this book, the reader may feel a certain misgiving. The proposed innovations have a common feature: spend more now, in expectation of better outcomes and cost-of-service savings later. Hughes does not have Lindbeck’s affection for tough rules and fiscal discipline as essential to successful welfare states. Perhaps because Lindbeck and I are both economists, I share his perspective. If our safety net is aging and in need of renewal, surely there are some program areas where the optimum innovation is spend less now, in expectation of better outcomes and no cost-of-service increase at a later date.

The classic example of a Canadian social policy innovation where less turned out to be more is the decision reached by most provincial social service ministries in the mid 1990s to tighten dramatically the regulations governing access to social assistance among those deemed employable. The result was a near halving of the poverty rate over the next decade, as measured using Statistics Canada’s low income cut-off.

The percentage of Canadian households falling below the after-tax LICO thresholds is the most frequently cited index of poverty in Canada. The LICO thresholds adjust for family size and community size, and they are also adjusted for inflation. Because the real value of what can be bought at the thresholds has not been benchmarked since 1992, LICO is increasingly becoming an absolute poverty measure: the value of what can be purchased at threshold incomes remains constant from year to year. The LICO poverty rate rose in the early 1980s recession, fell in the subsequent boom, and peaked above 15 percent in the early 1990s recession. Since the mid 1990s, the LICO poverty rate has fallen nearly in half. It rose slightly in the post-2008 recession, but in the most recently reported year (2011) it is under nine percent.

Not surprisingly, reliance on social assistance is linked to the state of the economy. The share of Canadians receiving social assistance was about five percent in the late 1970s, rose to seven percent in the early 1980s recession and remained constant over the decade. In the wake of the early 1990s recession, reliance on social assistance peaked above ten percent of the population. At that point most provinces abandoned what had been the tradition in Canada of providing generous access to welfare benefits. Some provincial governments (for example, the Tories in Ontario) resorted to Sturm und Drang in making the change; others (for example, the NDP in British Columbia) relied on stealth in order not to alienate core supporters. At the time there was alarm among social policy advocates as to the implication on poverty rates. Overall, the employment rate rose and the poverty rate fell. The lowering of post-1995 LICO poverty rates is most evident among female-headed single-parent families. From 1995 to 2011, their LICO poverty rate declined from over 50 percent to below 25 percent. They substituted higher market earnings for lost social assistance.

At 23 percent in 2011 (the latest available statistic), the single-parent LICO poverty rate is still more than twice the national average, and the explanation for the decline is more complex than the above implies. In the late 1990s, the federal government introduced the National Child Benefit System, a modest negative income tax for low-income families with children. Most provinces lowered welfare benefits dollar for dollar with NCBS payments. Families received no benefit from the NCBS while on welfare, but the NCBS has allowed families to exit welfare at lower earning levels. Having exited, these families face lower clawback rates on incremental earnings than they faced when on welfare.

I agree with Hughes that, in seeking areas of social policy in need of reform, mental illness deserves a high priority. Mental health reform is hard to do well. First, evaluating success is open to more ambiguities than in most other health programming. It is relatively easy to measure under-five mortality rates. More difficult to measure but still relatively straightforward are reading and mathematics outcomes among children from disadvantaged families. It is far more difficult to evaluate programs intended to tackle, say, depressive disorders. Second, mental health covers a diverse range of services, from special needs programming in schools to homeless services and care for the old suffering dementia.

It is hard to disagree with Hughes that early intervention is usually preferable to late in almost all areas of health, education and social programming. However, early does not usually imply lower budgets. To undertake the mental health reforms that Hughes wants means probably spending considerably more. Even with no reforms, continuing with the status quo will probably entail an increase of two percentage points in gross domestic product in the cost of social programs over the next three decades. (One percent of GDP is about $20 billion.) The primary reason will be the high cost of providing health care to the old, and the shift in the age distribution as the boomer generation populates the over-65 age cohorts. This brings us to a brute fact about social spending in a democracy. It depends ultimately on the willingness of citizens to pay the required taxes.

Hughes goes easy on the Harper administration: “the Harper Conservatives … have been disinclined to repeat the slash-and-burn practices of the governing Liberals in the 1990s.” This is too kind to the Harper government, too dismissive of the Chrétien Liberals and ignores context. The reason Canadian governments, both in Ottawa and the provinces, underwent budget cuts in the 1990s was not a heartless desire to slash and burn; it was the fiscal irresponsibility of Canadian politicians over the previous two decades and widespread popular resistance to solving the national public debt problem via higher taxes. Not once since 1975 had the politicians, summing up Ottawa and the ten provincial budgets, balanced their accounts. Instead, they engaged in mutual recriminations.

Throughout the 1980s and early 1990s, the provinces blamed Ottawa for arbitrarily cutting the federal contribution to shared cost programs; Ottawa accused the provinces of violating the spirit of shared-cost social programs by overspending on them. In neither arena did politicians address the inefficiencies in their respective program portfolios. By the mid 1990s, Canada’s net debt exceeded 100 percent of GDP; we were the third most indebted country in the entire Organisation for Economic Co-operation and Development, behind Italy and Belgium. We had no choice but to lower debt. Given the large gap between taxing effort in Canada and the United States (about ten percentage points of GDP), it was inevitable that the emphasis would be on expenditure reductions as opposed to tax increases.

If we are to make sense of present-day taxation levels and spending priorities, we need to pay attention to where Canada sits relative to our OECD partners. Let’s focus on the period from 1992 to today. Has Canada been relatively generous in social spending and aggressive in taxing, or relatively conservative?

In answering these questions it is useful to rank OECD countries into quarters, grouped from the highest to the lowest government spenders (Figure 1) and the highest to lowest taxers (Figure 2).

Let’s look first at government spending. As shown in Figure 1, the top quartile is the minimum spending rate among the top quarter of OECD countries with the most generous benefits. In 1992 it was 56 percentage points as a share of GDP. At that time, Canada’s own rate of government spending was close to 54 percent, which meant we were close to joining this top quarter of high-spending countries. Our spending subsequently declined so that by the second half of the 1990s we were close to the typical, or median, OECD country. Then, after 2000, our spending continued to drift down—so much so that during the past five years we have come close to joining the quarter of countries with the lowest spending as a share of GDP. The maximum spending rate among the bottom quarter (the bottom quartile) has hovered around 40 percent of GDP over the last two decades, with the exception of the aftermath of the 2008 recession when spending in nearly all OECD countries, including Canada, rose. At the same time, the spending gap with the United States has tumbled considerably, so that we now spend only about two percentage points of GDP more than the Americans do. What can we conclude from all this? Based on what other OECD countries are doing, Canada has gone from being a generous to a conservative spender on social programs.

Now let’s look at taxation. As shown in Figure 2, as Ottawa and the provinces eliminated their respective deficits during the late 1990s, Canada’s taxing effort fell. So in the first half of the 2000s Ottawa (under Chrétien) and the provinces were able to lower their combined taxing effort by about four percentage points of GDP. This placed Canada’s taxing effort close to the median OECD country. There was no widespread public demand at the time for further tax cuts. By 2006, when the Conservatives entered office, the tax gap with the U.S. had been cut nearly in half relative to a decade earlier. Given their ideological preferences, Harper’s Conservatives, in their years in office, pushed our taxing effort down by an additional two percentage points. Canada is now close to joining the bottom quarter of the lowest taxing OECD countries—in other words Canada has become a “low tax” country.

In conclusion, when compared with our OECD peers, we are now relatively conservative in terms of government spending and have relatively low taxes. So there is validity to Hughes’s claim that there is room to spend more on social programs—as long as Canadians can be persuaded to pay the required taxes. For many economists (including me), the preferred tax option is a new nation-wide carbon tax that rapidly ramps up to become a major incentive to curtail activities that generate greenhouse gases.

If the next government agrees to some tax increases—which is far from certain—there remains the complex and controversial matter of determining priorities. My own personal top priority for a spending increase is early childhood education targeting children in low-income or otherwise marginalized families. In particular, that means targeting aboriginal children, whether First Nations living on or off reserves, Inuit or Métis. But I agree with Hughes that better mental health programming should be high on the list. Some of his other priorities—such as those related to housing—would likely receive considerable attention as well. All that we can say for certain, however, is that the competition over specific priorities will be fierce.

John Richards is a former member of the Saskatchewan legislature and a professor of public policy program at Simon Fraser University, in Vancouver.