The Oxford Handbook of Canadian Politics

John C. Courtney (ed.), David E. Smith (ed.)

[Link]
Published: 08 April 2010 Online ISBN: 9780199892204 Print ISBN: 9780195335354

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

Search in this book

CHAPTER

23 Continuities and Change in the Design of Canada's Social

Architecture 
Jane Jenson

[Link] Pages 417–433

Published: 02 September 2010

Abstract
This article discusses Canada's social architecture. It considers intergovernmental relations as it
illustrates the shifting intersections of institutional arrangements in four key welfare-producing sectors:
the state, the family, the market, and the community. The article discusses the policy instruments and the
program designs: both serve as a yardstick against which to measure the content of social policy. Special
attention is accorded to the income security instruments and programs of social policy of Canada after
1945, the main goal of which was to increase the social security of all categories of the population so as to
redistribute income from rich to poor.

Keywords: social architecture, intergovernmental relations, institutional arrangements, policy

instruments, program designs, social policy, income security instruments, social security
Subject: Political Economy, Comparative Politics, Politics
Series: Oxford Handbooks
Collection: Oxford Handbooks Online

CANADA has always been classi ed in the family of liberal welfare regimes, whose de ning characteristic is “a
preference for market solutions to welfare problems” (Myles 1998, 342). The labor market is central, but so, too,
are other markets, including those for retirement income via private pensions and for private services such as
supplementary health care or child care. Despite an underlying preference for market solutions, the ways to
institute them—via a mix of responsibilities among the family and community as well as markets and the state
—have varied throughout time and across the several jurisdictions that share responsibility for social policy in
Canada.
A second characteristic of the history and future of Canadian social policy is that it is profoundly shaped by the
institutions of federalism (Banting 1987). Disputes over social policy have been a focus of intergovernmental
 politics for most of the country's history, and compromises settled in that area have shaped intergovernmental
relations more broadly. The involvement and powers of each level of government to shape social policy has also
widely varied over time. If from Confederation to World War II the provinces and their wards, the
municipalities, were all powerful in the domain of income security, from the 1940s to the '80s, the federal
government dominated (Banting 1987, 47). Spending by the federal government signi cantly shapes support
for the unemployed, for seniors, and for children, but the provinces and territories retain a good deal of leeway
p. 418 for choices about both income transfers and services in these same domains. For example, Employment
Insurance (EI) paid to unemployed workers who meet its eligibility requirements, is a federal program.

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

Provinces and territories are responsible for training and for any bene ts offered to the unemployed not
eligible or no longer eligible for EI.
This chapter must, therefore, consider intergovernmental relations because it seeks to tell a general story about
the shifting intersections of institutional arrangements in four key welfare-producing sectors: market, state,
family, and community.1 It will do so by focusing on policy instruments as well as program design, because it is
choices about each of them that actually determine the content of social policy. In Canada's liberal welfare
regime, the main goal of social policy after 1945 was to increase the economic security of all categories of the
population more than it was to redistribute income from rich to poor, and therefore this chapter concentrates
on income security instruments and programs (Banting 2006, 431).

Analyzing the Social Architecture

Social architecture is the term used to describe the blueprints for welfare regimes and “welfare regimes …
bundle together programmes and policies which transcend the ‘welfare state’ narrowly conceived” (Goodin et
al. 1999, 5). In his work, Gøsta Esping-Andersen (1999, 33, 35) describes three pillars of this architecture: “[T]he
welfare state is one among three sources of managing social risks, the other two being family and market,” and
their interaction produces a “welfare triad.” This vision of three institutional locales has been widely adopted.
Nevertheless, it has also been criticized for failing to consider a location where much welfare is generated—
that is, the “third” or nonpro t sector. For example, Adalbert Evers (1998, 162–163) describes: “the ‘mixed
economy’ of market, state, voluntary organisations and the family,” and conceptualization of a welfare diamond
(Evers et al. 1994). Although remaining within the welfare regime tradition, this chapter adopts a four-sector
rather than only a three-sector approach to social architecture. It looks, in other words, at the mix of
responsibilities among family, market, community, and state, seeing a welfare diamond rather than triad
(Jenson 2004a).
Each country makes its own choices about the shape of its welfare diamond and therefore the relationships
among sectors. Across all regime types, by far the major source of welfare for most of the population is market
income, earned by the individual or a family member (Esping-Andersen et al. 2002, 11). However, regimes are
also structured around the assumption that there are numerous nonmarketized bene ts and services provided
within the family, such as parental child care, housework, and care for elderly relatives. The community is a
third source of welfare in most regimes, albeit with greater variability across regime types than for the previous
two sectors. This sector, with its mix of volunteers and paid workers, provides a range of services and supports,
p. 419 such as child care, food banks, recreation, and leisure either for free or on a nonpro t basis. Finally, in all
welfare regimes some portion of welfare comes from the state, via public services (such as child care, health
care, or other services for which we are not required to pay full market prices) as well as by income transfers.
Although a liberal welfare regime has a preference for market solutions, not even a liberal regime will leave
everything to markets. What a liberal regime does do more than a social–democratic or conservative one,
however, is both support citizens in their market choices and support markets providing certain services that
in other regime types are provided as public services. For example, after World War II, Canada provided a basic
 public pension to all seniors, as did many countries at the time, and in the mid 1960s it added a contributory
pension to the public system. However, the federal government also makes tax expenditures (in other words,
forgoes taxes) to encourage the development of private retirement pensions. These tax advantages to
individuals are Registered Retirement Savings Plans (RRSP), which enable tax-sheltered savings for retirement.
By making such tax expenditures, the government is also fostering a large market for pension funds populated
by the banks and other nancial institutions that offer RRSPs to their customers.2
Child care provides another example. In the Canadian liberal welfare regime, public policies send parents into

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

the market to purchase child care, providing some tax relief for their expenditures, and also shaping the market
for child care via regulations identifying the kinds of services for which parents can claim a tax deduction or
obtain a subsidy. The federal government provides a tax deduction for any kind of child care service as long as
the provider supplies a receipt (meaning, the provider is not working “under the table” and not paying taxes),
leaving parents to make the choice about whether they wish to hire an untrained baby-sitter or use a licensed
child care provider. Quebec, however, made another choice about how to structure the child care market. By
providing child care places at a very low at rate, the government's policy since 1997 has included an effort to
undermine the market for untrained, unlicensed, and “under-the-table” baby-sitters. It sought to encourage
parents to rely on either early childhood centers or licensed family daycare providers, by making such forms so
affordable that baby-sitters could not compete on price (Jenson 2002, 324).
These examples suggest that, although all types of welfare regimes may display convergence around the ideas
that it is important to ensure adequate income so that retired persons may live in dignity or that families need
help in balancing work and family responsibilities, liberal regimes will favor markets. However, they also
illustrate that there are still signi cant differences among interventions to shape markets or family behavior
within the category of liberal welfare regime.
Historically, social policy has relied on three policy instruments to construct its social policy architecture
(Banting 1987, 7–12; Myles 1998, 350–352):
1. Public assistance, intended to address the needs of those without adequate income from any other source
and using means testing as the main tool
2. Social insurance, based on labor market performance and using contributions as the tool
3. Social citizenship, using universality as the tool
p. 420 Recently, a fourth instrument has gained popularity: tax-based income supplementation, which relies on
income testing and is delivered exclusively through the tax system (Myles and Pierson 1997). The next sections
of the chapter will tell the story of social policy in Canada, tracking the policy design and range of instruments
in its social architecture over time.

The Canadian Welfare Regime a er 1945

One of the rst provincewide public assistance programs of the twentieth century was for poor mothers of
young children who had been widowed or abandoned by their male partner. Mothers' allowances were
eventually instituted in all provinces, but full coverage took thirty-three years, beginning in Manitoba in 1916
and arriving in 1949 in Prince Edward Island and Newfoundland (Boychuk 1998, 28ff). Nowhere were rates very
generous, however, and therefore charity provided by churches and other agencies often supplemented the
allowances.
The crisis of the 1930s revealed a depth of need that could not be satis ed by limited public assistance
programs such as mothers' allowances and municipal relief for the poor. Intergovernmental relations came
 under particular stress from constitutional debates about who had responsibility for the unemployed. Only
when the provinces accepted a constitutional amendment that surrendered some of their jurisdiction, could the
federal government set up the Unemployment Insurance (UI) program in 1940 that was (and remains) under its
exclusive jurisdiction (Banting 1987, 48–49).
The lessons of the Depression were not lost on policy makers planning for postwar reconstruction. There was
growing recognition, at least outside of Quebec, of the advantages of pan-Canadian and cross-provincial
sharing of risks. The federal government made a wartime announcement of a “grand design” for a new social

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

security system (Boychuk 1998, 42). During the next three decades, that new system was built. Its elements
were universal at-rate family allowances (1944); contributory unemployment insurance (1940); a universal
at-rate old-age pension (Old Age Security [OAS]) and a needs-based one, the costs of which were shared with
the provinces (1952); universal hospital insurance (1957); and public assistance, with federal–provincial cost-
sharing of the Unemployment Assistance Act (for the unemployed not eligible for UI, 1956) and support for the
blind (1951) and the disabled (1954). Mothers' allowances remained an exclusively provincial responsibility.
If social policy instruments were mixed, nancing was equally mixed. Unemployment Insurance was funded
by contributions from employers and employees as well as the federal government. The OAS depended on an
old-age security fund created with earmarked taxes. For the rest, the federal government asserted its
p. 421 constitutional spending power to “make grants for any purpose” and used cost sharing as the way to shape
social policy. Most provinces accepted, more or less willingly, the dollars that the federal government offered
and the conditions it set on the use of those transfers for programs for the disabled, the blind, hospitals, and
even for the improvement of their administration via research and experimentation (Banting 1987, 53). Without
half of the costs paid by Ottawa, few provinces would have been able to create a modern system of social
security; their own revenues were simply too limited.
In the mid 1960s, this piecemeal approach to addressing social risks was replaced by three major reforms. In
1966, a range of programs were combined into the Canada Assistance Plan (CAP), with which the federal
government promised to match provincial social spending. The Canada Assistance Plan consolidated the
Unemployment Assistance Act together with legislation providing public assistance to the physically disabled,
made federal funding available for bene ts for lone parents and poor families, and provided for a range of
social services including daycare (Mahon 2000, 595–596). The result was a signi cant convergence in program
design and bene t levels in social assistance across all provinces (Banting 1987, 94–96).
The Canada Pension Plan (CPP) and Quebec Pension Plan (QPP) also date from 1966. Together, they constitute a
compulsory social insurance scheme that covers virtually the whole of the labor force. It requires both
employees and employers to contribute toward a wage-related retirement pension at age sixty- ve as well as
long-term disability and survivor's bene ts. An innovation in intergovernmental relations, it permitted Quebec
to “opt out” of the CPP and establish its own QPP.3 This insurance approach was complemented by the addition
of a guaranteed income supplement (GIS) at the same time, and the gradual reduction to age sixty- ve of the
universal OAS pension.
Finally, building on the Hospital Insurance and Diagnostic Act of 1957, which had given the federal government
authority to enter into agreements with the provinces to establish a comprehensive, universal plan covering
acute hospital care and laboratory and radiology diagnostic services, the Medical Care Act of 1966 extended
health insurance to cover doctors' services.
This consolidation of three decades of social policy action was the capstone of the system. For Rice and Prince
(2000, 80) these three actions signaled the transformation of “social welfare in Canada from a system of
allocating bene ts on the basis of relief and means testing, to one of allocating bene ts on the principles of
universal entitlement, public service, and social insurance.” For John Myles (1998, 351), one could say that “by
the end of the period of welfare state reform in the early 1970s, the programmatic design, if not spending levels,
 of the Canadian welfare state was remarkably similar to that of Sweden.” By this, Myles meant that design
choices for programs and instruments privileged social insurance or social citizenship principles.
In these years, with their ideas as well as their actions, decision makers in both the private and public sectors
were redesigning the welfare diamond. The market sector was organized around collective agreements
p. 422 negotiated between unions and employers that provided private health, pension, and other protections to
union members and their families at a time that union membership rates were on the rise. This “private
welfare state” was an especially important component of income security for the retired, and a supplement to

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

universal health bene ts. At the same time, the state sector was moving toward providing a series of new social
citizenship rights (family allowances, health care, OAS) as well as redesigned social assistance programs (CAP)
and contribution-based social insurance for employment, workers' compensation, and the CPP/QPP portion of
pensions.
Families, too, had a major role in providing well-being. Both state and market sectors assumed that male
breadwinners and female caregivers were the norm; responsibility for their elderly relatives and children
remained overwhelming private with very few services, either public or private (Mahon 2000). Communities
were also in the picture. Although the voluntary sector was actively involved, little attention was actually
devoted to its role or contribution. The exception to this silence was in Quebec, where the Catholic Church had
been responsible for delivering welfare to the francophone population (schools, hospitals, charities, and so on)
until the Quiet Revolution of the early 1960s. Quebec's relatively tardy development of public spending led, in
the 1960s, to a lively debate about public and private roles and responsibilities. Elsewhere, however, the fact that
the Children's Aid Society was delivering virtually all child protection services in some provinces, that churches
ran schools and hospitals in part with public funds, that the Victorian Order of Nurses provided publicly
nanced home care, or that CAP required that nonpro t community-based associations deliver much of
publicly subsidized child care did not attract a great deal of attention.
Throughout all of this, preference for market solutions remained. That social policy spending should be limited
was a value in itself, which helps to account for the differences in spending levels if not design between Canada
and Sweden. For example, when policy makers were shown by, among others the Royal Commission on the
Status of Women, that families were struggling to juggle work and family life as women's labor force
participation rates rose, three new instruments were adopted. One used the contributory unemployment
insurance system to provide paid maternity bene ts in 1971. The second, re ecting the preference for
supporting market choices, led the federal government to provide a child care expense deduction, administered
through the tax system, rather than to fund universal child care services directly, as Sweden was doing at the
same time. The third involved provinces using CAP funds to offer child care subsidies to low-income families
so that they would be able to establish and help maintain the labor force.
This mid 1960s capstone to the social architecture was not in place very long, however, before reformers were
rearranging the basic building blocks. Indeed, the GIS was a precursor of an instrument that would gain great
popularity in the next decades. The GIS is available to seniors with the lowest incomes. As an income-tested
bene t, it is regulated via the tax system. This new instrument soon spread to other policy areas.
p. 423
Redesigning the Mix: Changing the Instruments

Michael Prince (2003, 127) identi es three periods that distinguish the federal government's actions with
respect to the social architecture since 1984. Efforts to restrain costs of social programs characterized 1984–
1988. From 1988 to 1997, restructuring the social role of government was the objective, under both Conservative
and Liberal governments. This period brought a number of signi cant changes in program design and
instruments. A third phase, which Prince terms “repairing the social union,” opened in 1997. The federal

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

government announced that it had tamed its de cit and could think again about investing in the social
architecture and reasserting its place in the social union (the name developed at the time to describe
intergovernmental relations in key areas of social policy). This third period lasted until the election of the
government of Prime Minister Harper in 2006.
The Canada Assistance Plan was profoundly affected in the second phase. The basic design of CAP had
remained relatively unchanged from 1966 to 1990 (Boychuk 1998, 48). Then, in what was the rst volley in what
would become an upheaval in intergovernmental relations in the realm of income security, the Conservative
government's 1990 budget imposed a “cap on CAP.” Instead of the open-ended nancing in which the federal
government matched the spending choices of the provinces, the federal government imposed a limit of 5%
annually on any increase in transfers to the three richest provinces. “The ‘cap on CAP’ decapitated [CAP] and
put all the provinces on notice that they too could no longer count on the same level of federal largess.” (Battle
1998, 329).
Then, in 1995, CAP was ended unilaterally by the Liberal government, in the name of its politics of de cit and
debt control and restructuring the social role of the government. In its place, the Department of Finance's
budget announced Canada Health and Social Transfer (CHST). Using a carrot-and-stick approach, albeit more
stick than carrot, the federal government released the provinces from almost all conditions on spending of the
money transferred to them.4 The exchange was that they lost billions of dollars of transfers and the guaranteed
countercyclical funding that allowed them to respond to rising loads of social assistance claims during times of
economic downturn (Battle 1998, 330).
The result of this reform has been signi cant reductions in levels of disposable income available to social
assistance recipients. Looking at change over time and comparing 2005 social assistance incomes with the
peak year of each province and family type, the National Council of Welfare has identi ed major losses. Overall,
one third of all household types lost $3,000 or more in income. In Alberta, the income in real dollars of a single
person on welfare decreased by almost 50% between 1986 and 2005, whereas from 1992 to 2005 in Ontario, the
welfare income of a lone parent with one child decreased by almost $6,600, and a couple with two children lost
just more than $8,700 (National Council of Welfare 2006, 48).
p. 424 Abolition of CAP and subsequent reforms to the social assistance regime in all provinces means that social
assistance is again a program intended to do little more than serve as a last-resort safety net for recipients.
Given the cuts to their real incomes, which were below the poverty line at the best of times, recipients must rely
increasingly on family solidarity, where it is available, and also on community support in the form of food
banks and other charitable supports.
Income security for the unemployed was also restructured in 1988–1997 to restrict eligibility (although not the
requirement to contribute) (Battle 1998, 332). An initial reform in 1971 had expanded the system signi cantly
and moved it toward serving goals beyond providing insurance for temporary job loss. The expansion included,
among other things, the introduction of bene ts for workers not previously covered (e.g., shing bene ts) and
new kinds of bene ts (sickness and maternity). Taking into account regional differences in unemployment
rates, it also increased the possible duration of bene ts in high-unemployment regions and reduced the
minimum qualifying work periods. This shift appeared to mark a “reorientation away from the logic of social
insurance towards a logic of social cohesion—increasingly aimed at bridging various fault-lines including
 those based on region as well as relationship to the labour market, such as distinctions between primary and
secondary income earners” (Boychuk 2004, 14).
Throughout the 1980s, Canada's overall unemployment rate was consistently above the Organisation of
Economic Cooperation and Development (OECD) average and during the recession of the early 1990s it reached
10%. The Conservative government of the day proposed a number of adjustments (including tightening
eligibility), but its major changes involved eliminating any government contribution to the UI fund and trying
to reinforce the market sector by increasing its responsibility both for funding UI and for developing training

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

programs (Campbell 1992, 32–33). It was the Liberal government elected in 1993 that undertook the most
signi cant overhaul and redesign, however.
Its reforms, which also renamed the program Employment Insurance, signaled a shift away from the 1970s'
focus on social cohesion as well as signi cantly reduced coverage (Banting 2006, 426). The percentage of the
unemployed receiving regular bene t payments fell between 1989 and 1997 from 87% of the of cially counted
unemployed to 42%. In other words, well more than half the of cially unemployed received no insurance
payments (Battle 1998, 332). Later reforms did lengthen parental leaves bene ts to twelve months and provide
coverage to part-time workers. Nonetheless, complicated and restrictive eligibility rules have made conditions
more stringent for certain types of part-time workers, especially part-time workers in high-unemployment
areas and new entrants to the labor market working part-time (Rice 2002, 115).
The move to EI as well as the cap on CAP and the creation of the CHST all had major effects on the social
architecture. The federal government and several provinces were much in uenced by the ideology of
neoliberalism, which advocated a smaller role for the state and much greater reliance on markets, families, and
p. 425 communities. Throughout these years, the programs and instruments of social citizenship in particular
were criticized and lost their popular legitimacy. Thus, beyond the goal of husbanding spending overall, during
the periods both of cost restraint (1984–1988) and restructuring the social role of government (1988–1997)
there was pressure to use new instruments to target spending to low-income Canadians rather than making
universal payments.
Such criticisms helped to legitimate the fourth instrument of social policy: tax-based income supplementation.
Not much has been heard about this instrument thus far in this chapter, but John Myles and Paul Pierson (1997,
446) had this to say about the past three decades: “In Canada, the NIT/GI [negative income tax/guaranteed
income] model has been rmly imprinted as the policy paradigm of choice among legislators and of cials.”
What is this NIT/GI model? It is based on a speci c instrument: the negative income tax. It provides a
guaranteed, albeit always low, income to a targeted category of the population, such as seniors, families with
children younger than eighteen, or the working poor:
The original idea, proposed by Milton Friedman in 1943, was fairly simple: In good times, workers
would pay taxes to governments; in bad times governments would pay taxes to workers. Eligibility is
determined exclusively by income reported in a tax return. There is no surveillance of bene ciaries or
administrative discretion beyond that normally associated with the auditing of tax returns. Tax-back
rates on earnings and other sources of income are always much less than 100 percent. One result is
that bene ts can reach into the ranks of the middle class, albeit at a diminishing rate. (Myles and
Pierson 1997, 447 [emphasis in the original])
Canada is, of course, not alone in relying on NIT instruments in the realm of income security; however, in
Canada, the NIT instrument has been used broadly to manage the income of retired seniors as well as within
the unemployment insurance regime. It also was deployed to construct the replacement for the rst program of
social citizenship: family allowances.
 The 1966 GIS was the rst experiment with an NIT in Canada, and the instrument continued to be favored in
the area of pension incomes. In 1975, an income-tested spouse's allowance was added to the OAS/GIS to cover
those, usually women, between age sixty and sixty-four whose spouse received the minimum bene t and they
themselves had no earned income. The allowance was later extended to widows and widowers of OAS/GIS
recipients between sixty and sixty-four years of age. After failing to do so in 1985, because of a revolt of “gray
power” well covered in the media, the federal government in 1989 nally succeeded in imposing income
testing on the then-universal OAS (Battle 1998, 333).

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

The next big step toward a fully NIT system was supposed to be a seniors' bene t to replace OAS, GIS, and two
other tax credits. This major reform, announced in 1996 and scheduled to take effect in 2001, would have
“completed the transition” (Myles 1998, 448), however, the proposal came to naught and the promise the next
year to extend income testing foundered, as well, on opposition from the voters and the business community.
p. 426 The latter feared a contraction in their markets, if individuals lost the incentive to save for their retirement
in private pension plans and especially RRSPs.
Therefore, income security after retirement continues to rest on multiple instruments: the CPP/QPP
contributory pension plans, the income-tested OAS/GIS, and private pensions either negotiated via collective
agreements or from tax-sheltered savings. In other words, the family and the market sectors of the welfare
diamond are both signi cantly involved in the ways Canadians obtain their retirement incomes, as is the state,
which has its own programs and shapes the market conditions of the other two.
After UI was eliminated and EI took its place, a measure of income testing was introduced into the new
unemployment program as well. First, the maximum bene t was capped at a quite low level of earnings; the
reimbursement rate is 55% of earnings but only up to a maximum of $41,100 in earnings, although low-income
unemployed workers with children (earning less than $25,921) are eligible to receive the Employment Insurance
Family Supplement, which can raise the reimbursement rate to as much as 80%. In this family supplement,
which uses an NIT instrument, we see the intersection of two areas of income security—that of families and
that of the unemployed.
The family supplement in the unemployment insurance regime was only one example of reliance on income
testing as an instrument to supplement the income of families with children. Indeed, after the rst experiment
with the GIS for seniors in 1966, incomes of family with dependent children emerged as the next frontier for
experimenting with the NIT in Canada's social architecture. In 1970, the federal government suggested it would
institute a Family Income Supplement Plan. The plan would have transformed family allowances from a
universal to an income-tested program. Opposition before and during the 1972 election campaign caused the
government to take another track; allowances were substantially raised, but were made subject to taxation
(Banting 1987, 112–113). Subsequently, the universal family allowances were left to languish, by being only
partially indexed to in ation, and new income-tested family bene ts were introduced (Stroick and Jenson
1999, 79).
In subsequent periods of social policy redesign, when the social role of government was being restructured, the
NIT instrument was applied without hindrance. As the federal government moved into its phase labeled that of
“repairing the social union” (Prince 2003, 127), child bene ts became the terrain on which experimentation
occurred. Immediately after the earthquake budget of 1995, the provinces and territories announced at their
annual conference (which does not include the federal government) a major process of rethinking social policy.
This led the next year to the creation of the Provincial/Territorial Council on Social Policy Renewal, which the
federal government was invited to join. Over the following years, until the Social Union Framework Agreement
(SUFA) was signed in 1999, a number of important initiatives in the area of child and family bene ts emerged
from these processes of intergovernmental negotiations (Fortin et al. 2003).
 Chief among these was the National Child Bene t (NCB) initiative. Its 1998 self-description clearly locates the
NCB as an innovation both in the way that governments would work together and in the objectives they would
p. 427 pursue.
The National Child Bene t is a joint initiative of the Government of Canada and provincial and
territorial governments. Goals of the initiative are to:
• help prevent and reduce the depth of child poverty;

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

• promote attachment to the work force; and
• reduce overlap and duplication between Canadian and provincial/territorial programs.
The National Child Bene t will begin to remove child bene ts from welfare, assist parents with the
cost of raising children, and make it easier for low income parents to support their families through
employment without resorting to welfare.5
The third goal announces that the NCB is about intergovernmental relations and, in particular, new methods
for coordination. It includes something we can call asterisk federalism (Jenson 2001), in which Quebec simply
does not participate in initiatives undertaken by the other twelve governments.6 Beyond that, however, the
initiative as a whole reveals a consolidation of perspectives on the objectives of a redesigned social architecture.
The NCB marked the unveiling of a social investment perspective on social architecture (Banting 2006; Jenson
and Saint-Martin 2003). Along with a number of other actions within the area of child and family policy, this
perspective shaped most of the innovations in social policy in the last years of the twentieth century and the
rst half of this one.

The Social Investment Perspective: Achievements and Challenges

For more than a decade, welfare regimes, especially liberal and social–democratic ones, have displayed
convergence around ideas for modernization of social models via labor market involvement of all adults and
new forms of investment, especially in human capital and early childhood education and care, and in “children”
more generally (Esping-Andersen et al. 2002). International organizations have also been the carriers; a social
investment emphasis was present in the analyses of the Organisation of Economic Cooperation and
Development (OECD) by the mid 1990s and in “third-way” Britain by the late 1990s (Dobrowolsky and Jenson
2005). Three key features of the perspective can be identi ed. The rst is the emphasis on education and
learning to ensure that adults today and children in the future will be able to adapt to the labor market
conditions of a knowledge-based economy, which demands exibility in employment relations and supplies
many precarious forms of jobs, such as part-time, temporary, and self-employment. The second is an
orientation to the future. This means that there is greater concern for setting the conditions for future success,
for individuals and countries as a whole, than in achieving equality in the present. Third, and last, there is the
idea that successful individuals enrich our common future and that ensuring success in the present is
p. 428 bene cial for the community as a whole, now and into the future. These three ideas can be seen as three
principles, the translation of which into social policy has consequences.
Canada's version of the social investment perspective brought an intensi cation of policy attention to children
as well as an emphasis on ghting poverty, especially child poverty, by ensuring that as many parents as
possible among low-income families would be in the paid labor force (Jenson 2004b). It also has brought little
attention to improving the income security of adults without children younger than eighteen who are
considered “employable.” When not in the labor force, they have seen their income supports substantially
reduced and bene ts made much more dif cult to access (Banting 2006, 424ff.).
 The view of adults is also different in the social investment perspective. Adults raising children gain access to a
wide range of bene ts and services that adults without young children cannot access. At the same time, these
programs are founded on the assumption that all adults should participate in the paid labor force and, therefore,
they seek to promote employability and sustain participation. Caring for children as a lone mother is no longer
accepted as a substitute for labor force participation.
It is in this context that the new initiative to relaunch the social union needs to be interpreted. The NCB was
composed of a set of interrelated parts. The federal government's portion consists of two bene ts paid directly

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

to families with children. The rst is the basic Canada Child Tax Bene t (CCTB). It is income tested and on a
sliding scale, with the maximum available to those with a net income less than $37,178. The cutoff was high,
with the intent being to provide some CCTB to 80% of families. The second bene t paid by the federal
government is the National Child Bene t Supplement (NCBS), going to very low-income families (an annual
income of less than $20,883). The result is an NIT structure that reaches both middle- and low-income
families.
In some ways, however, the fanfare that accompanied the announcement of the NCB would involve a large dose
of exaggeration, if it were assessed only as an income transfer. By themselves, the CCTB and NCBS did little to
“reduce the depths of child poverty.” Initial amounts paid out as the basic bene t and the various cutoff points
for employed families were exactly the same as those made by earlier bene ts (Jenson 2000, 17). Although the
names changed, the emphasis on favoring employment income over that from social assistance remained, both
in the of cial presentation of the program (quoted earlier) and in the details of its design. In the name of
“removing child bene ts from welfare,” the NCB design legitimated provinces and territories simply treating
the CCTB and NCBS as “income” and reducing social assistance payments accordingly (Banting 2006, 428).
This “clawback” provoked a storm of protest among advocates for the poor and poor families.7
Where the NCB initiative was innovative, however, was in the notion that it would take down the “welfare wall.”
The basic idea is that social assistance during the years of CAP had been structured such that accepting a low-
paying job and going off social assistance could be very costly for families, because they would lose valuable
p. 429 bene ts beyond income. For example, few low-paying jobs provide supplementary health and dental
bene ts, but social assistance did. Therefore, the provincial portion of the NCB initiative was designed around
the idea that provinces would take the money saved by not paying social assistance to cover children (now
replaced by the NCBS) and “reinvest” those dollars in extended bene ts to ease the transition from welfare to
work. And, indeed, many of them did just that.
The social investment perspective brought several other innovations as well, in the area of child health, early
learning, and so on (Jenson 2004b). The structure of intergovernmental relations in each was quite similar. The
federal government would offer additional funding to the provinces to develop programs in speci c areas. In
particular, the federal government worked for half a decade to induce the provinces and territories to improve
child care services, both in terms of quantity and quality. In 2005, just before the Liberal government was
defeated by the Conservatives in January 2006, it managed to convince all ten provinces to sign agreements for
using new federal funds to generate spaces for early childhood learning and care. This success was seen as a
major victory by advocates for child care as well as the federal government.
The social investment perspective, with its emphasis on “making work pay,” is also now the driver behind the
fall in income of social assistance recipients. Although much attention went to smoothing the transition of
parents from social assistance to employment, those who did not make that transition saw their incomes
reduced over time, as we have seen. The gap between average incomes and those of social assistance recipients,
including those with children, widened. This loss of income affected “single employables” even more than
lone-parent families on social assistance (Banting 2006, 426–429). In part, these declines were the result of
governments engaged in retrenchment—that is, simply reducing their expenditures—and in part from the
strategy of the NCB, which was to ensure that any net bene t from the CCTB and the NCBS went only to
 families not receiving social assistance. Although earlier ideas about social policy had accepted “need” as a
basis for receiving income transfer, by the rst half of the rst decade of this century, need was not suf cient.
Adults had to demonstrate a willingness to engage in employment by taking a job before the advantages of this
postde cit spending would reach them.
We see in the design of the NCB and its surrounding child-focused initiatives a redesign of the welfare diamond
in Canada's liberal welfare regime. The market sector has gained in importance. It is now expected to be the
source of earned income for all working-age adults, with few exceptions. Only exempted from this requirement

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

are citizens with a disability, whose social bene ts have fared better in the years of cost cutting, and adults who
live with another whose income is suf ciently high that they can choose to remain out of employment. Income
security programs, and especially employment insurance, have been designed to ensure that there are
signi cant incentives to stay in the labor force. In a positive direction, longer parental leaves coupled with
guarantees of a return to their job give parents more time with a newborn, thereby reducing the probability
that they will exit the labor force to care for the child. More dif cult for many, however, have been the large-
p. 430 scale reductions in unemployment bene ts and other work-related supports. The cuts make it much more
likely for many people that a job loss or layoff will leave them without any replacement income.
At the same time, the state sector has redesigned its relationship to the market. Bene ts such as the CCTB and
NCBS derive from the recognition that market incomes are frequently inadequate. People may be employed
full-time and full year and still be living in poverty. People who are able to nd only part-time or part-year
work are in a more precarious situation. Indeed, the number of working poor in Canada increased dramatically
between 1981 and 1997. Working poor families went up 66% and working poor “unattached individuals” (the
Statistics Canada label for working-age adults not living with family members) increased 153% (Rice 2002,
110–111). Income inequalities before and, increasingly, after taxes and transfers have increased into this decade,
such that economic security has dropped, despite the increase in wealth and consumption (Banting 2006, 432).
When the federal government set about to “repair the social union” after 1997, it displayed some willingness to
spend anew. However, it was only willing to commit funds to supplement the salaries of low-income earners
when they had children as well as to promote provision of some low-cost services (subsidized child care, for
example) so that parents could remain in the paid labor force. Although family allowances after 1945 were
meant to redistribute income from those without to those with children, child bene ts now directly address
weak points in markets, such as low wages and high cost services. For the rest, little has changed from the
choices made during the previous decade to trust in markets, families, and communities more than the state
sector.
In many domains, family responsibility remains high or has even increased. Parents who follow the injunction
to seek employment nd themselves facing bills for child care, transportation, and other costs of employment
that weigh heavy. Canadian parents outside Quebec pay a signi cant portion of the costs of child care, because
government assumes a much smaller share than in many countries. Moreover, parents have dif culty nding
reliable and quality care. Almost the rst act of the Harper government in 2006 was to announce the
termination of the federal–provincial accords on child care signed in 2005 and their replacement by an
allowance to families with children younger than six years of age. This taxable bene t did little to cover the
costs of nonparental child care while signi cantly slowing the opening of new services, which is what the
accords had been intended to do. The year 2006 saw the smallest absolute increase in child care spaces in
several years, lagging well behind the number created each year since 2001 (Childcare Research and Resource
Unit 2008). In addition, families with responsibilities for caring for family members with disabilities or
vulnerable elderly family have very little in the way of public support to do so.
The result is that many low-income Canadians have little choice but to turn for support to their communities.
For example, reliance on food banks (in other words, free food donated for distribution by volunteers to needy
p. 431 persons) has grown exponentially in the past decade, increasing 91% between 1989 and 2006 (Canadian
Association of Food Banks 2007). This means that many are receiving charity in the same way they relied on it
before 1945.
Some of these patterns, such as favoring families with children for income transfer, follow directly from the
social investment perspective. However, the other patterns just described reveal the limit of its design.
Although after 1945 the goal was income security of all, the objective in the social investment perspective is
narrower and its outcomes more limited. Income security in retirement meets international standards, but the
social investment emphasis on children is not generating the promised results. On international measures

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

Canada remains a laggard on child poverty and well-being, and dead last in the public provision of what a
major international economic organization, the OECD, sees as a pillar of the new economy—high-quality
preschool child care services (OECD 2006, 105). With these gaps in putting the model in place, Canada is at risk
of having let its “preference for market solutions to welfare problems” tilt too much toward the market, with
insuf cient attention to solving the real welfare problems that continue to exist in any modern society.

Notes
1. Throughout this chapter, the term welfare will be used in its generic sense—as a synonym for well-being and in the sense
of the expression “welfare state” and “welfare regime.” It should never be construed, as in U.S. usage, as a synonym for
social assistance, as in the expression “welfare recipient.”

2. Similar use of public funds for tax expenditures are o ered to families saving for postsecondary education with a
Registered Education Savings Plan (RESP).

3. Quebec, in the midst of its post-1960 nationalist Quiet Revolution, opted out of the Canada Pension Plan, establishing an
independent but coordinated provincial system.

4. The Canada Assistance Plan transfers came with three conditions: provinces provide income assistance only on the basis
of need, they have an appeals system in place for the use of applicants and recipients, and no minimum residence
requirements would be imposed. Only the third was carried over to the CHST (Battle 1997, 331).

5. This description is from the original presentation of the NCB, June 15, 1998.
[Link]/ncb/ncbfaq_e.shtml.

6. This is “asterisk federalism” because key documents include a footnote (originally in the form of an asterisk) that,
although Quebec shares most of the goals of the initiative, it refuses to participate because to do so would be to accept an
intrusion into areas of provincial competence (Jenson 2000, 16, 22). See, for example, the description of the National Child
Benefit at [Link]

7. All but New Brunswick and Newfoundland and Labrador applied the clawback. For one of many critical discussions of it,
see National Council of Welfare (2006, 56 .).
p. 432
References

Banting, Keith. 1987. The Welfare State and Canadian Federalism, 2nd ed. Montreal:McGill-Queen's University Press.
Google Scholar Google Preview WorldCat COPAC

—— . 2006. Dis-embedding liberalism? The new social policy paradigm in Canada. In Dimensions of Inequality in Canada, ed.
David A. Green and Jonathan R. Kesselman, 417–452. Vancouver: University of British Columbia Press.

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

Google Scholar Google Preview WorldCat COPAC

Battle, Ken. 1998. Transformation: Canadian social policy since 1985. Social Policy and Administration 32(4): 321–
340. 10.1111/1467-9515.00119
Google Scholar WorldCat Crossref

Boychuk, Gerald W. 1998. Patchworks of Purpose. The Development of Social Assistance Regimes in Canada. Toronto: University
of Toronto Press.
Google Scholar Google Preview WorldCat COPAC

—— . 2004. A History of the Canadian Social Architecture: The Logics of Policy Development. Report F|36. Ottawa: CPRN.
[Link].
Google Scholar Google Preview WorldCat COPAC

Campbell, Robert. 1992. Jobs … Job … Jo … J . . .: The Conservatives and the unemployed. In How Ottawa Spends 1992–93. The
Politics of Competitiveness, ed. Frances Abele, 23–56. Ottawa: Carleton University Press.
Google Scholar Google Preview WorldCat COPAC

Canadian Association of Food Banks. 2007. Hunger Count 2007. [Link]

WorldCat

Childcare Research and Resource Unit. 2008. Child Care Statistics 2007. [Link].
WorldCat

Dobrowolsky, Alexandra, and Jane Jenson. 2005. Social investment perspectives and practices: A decade in British politics. In
Social Policy Review #17, ed. Martin Powell, Linda Bauld, and Karen Clarke, 203–230. Bristol: The Policy Press.
Google Scholar Google Preview WorldCat COPAC

Esping-Andersen, Gøsta. 1999. Social Foundations of Industrial Economies. Oxford: Oxford University Press.
Google Scholar Google Preview WorldCat COPAC

Esping-Andersen, Gøsta, Duncan Gallie, Anton Hemerijck, and John Myles. 2002. Why We Need a New Welfare State. Oxford:
Oxford University Press. 10.1093/0199256438.001.0001
Google Scholar Google Preview WorldCat COPAC Crossref

Evers, Adalbert. 1998. Part of the welfare mix: The third sector as an intermediate area. Voluntas 6(2): 159–
182. 10.1007/BF02353995
Google Scholar WorldCat Crossref

Evers, Adalbert, Marja Pilj, and Clare Ungerson, eds. 1994. Payments for Care. Aldershot, UK: Avebury.
Google Scholar Google Preview WorldCat COPAC

Fortin, Sarah, Alain Noël, and France St-Hilaire, eds. 2003. Forging the Canadian Social Union: SUFA and Beyond. Montreal:
Institute for Research on Public Policy.
Google Scholar Google Preview WorldCat COPAC

Goodin, Robert, Bruce Headey, Ruud Mu els, and Henk-Jan Dirven. 1999. The Real Worlds of Welfare Capitalism. Cambridge:
 Cambridge University Press. 10.1017/CBO9780511490927
Google Scholar Google Preview WorldCat COPAC Crossref

Jenson, Jane. 2000. Le nouveau régime de citoyenneté du Canada: investir dans l'enfance. Lien social et Politiques 44: 11–23.
Google Scholar WorldCat

—— . 2001. Canada's shi ing citizenship regime: Investing in Canada's children. In The Dynamics of Decentralization: Canadian
Federalism and British Devolution, ed. Trevor C. Salmon and Michael Keating, 107–124. Montreal: McGill-Queen's University Press.

Downloaded from [Link] by University of Western Ontario user on 14 December 2025

Google Scholar Google Preview WorldCat COPAC

—— . 2002. Against the current: Child care and family policy in Quebec. In Child Care Policy at the Crossroads: Gender and Welfare
State Restructuring, ed. Sonya Michel and Rianne Mahon, 309–332. New York: Routledge.
Google Scholar Google Preview WorldCat COPAC

—— . 2004a. Canada's New Social Risks: Directions for a New Social Architecture. Report F|43. Ottawa: [Link].
Google Scholar Google Preview WorldCat COPAC

—— . 2004b. Changing the paradigm: Family responsibility or investing in children. Canadian Journal of Sociology 29(2): 169–
192. 10.1353/cjs.2004.0025
WorldCat Crossref

p. 433 Jenson, Jane, and D. Saint Martin. 2003. New routes to social cohesion? Citizenship and the social investment state. Canadian
Journal of Sociology 28(1): 77–99.
Google Scholar WorldCat

Mahon, Rianne. 2000. The never-ending story: The struggle for universal child care policy in the 1970s. Canadian Historical
Review 81(4): 582–615. 10.3138/CHR.81.4.582
Crossref

Myles, John. 1998. How to design a “liberal” welfare state: A comparison of Canada and the United States. Social Policy and
Administration 32(4): 341–364. 10.1111/1467-9515.00120
Google Scholar WorldCat Crossref

Myles, John, and Paul Pierson. 1997. Friedman's revenge: The reform of “liberal” welfare states in Canada and the United States.
Politics and Society 25(4): 443–472. 10.1177/0032329297025004004
Google Scholar WorldCat Crossref

National Child Benefit. 1998. [Link]/ncb/faq_e.shtml.

WorldCat

National Council of Welfare. 2006. Welfare Incomes 2005. Ottawa: National Council of Welfare.
[Link]
Google Scholar Google Preview WorldCat COPAC

Organisation of Economic Cooperation and Development. 2006. Starting Strong II. Paris: Organisation of Economic Cooperation
and Development.
Google Scholar Google Preview WorldCat COPAC

Prince, Michael. 2003. SUFA: Sea change or mere ripple for Canadian social policy? In Forging the Canadian Social Union: SUFA
and Beyond, ed. Sarah Fortin, Alain Noël, and France St-Hilaire, 125–156. Montreal: Institute for Research on Public Policy.
Google Scholar Google Preview WorldCat COPAC

Rice, James J. 2002. Being poor in the best of times. In How Ottawa Spends 2002–03: The Security A ermath and National
Priorities, ed. G. Bruce Doern, 102–120. Don Mills, Ont.: Oxford University Press.
Google Scholar Google Preview WorldCat COPAC
Rice, James J., and Michael J. Prince. 2000. The Changing Politics of Canadian Social Policy. Toronto: University of Toronto Press.
Google Scholar Google Preview WorldCat COPAC

Stroick, Sharon M., and Jane Jenson. 1999. What Is the Best Policy Mix for Canada's Young Children? Report F|09. Ottawa: CPRN.
[Link]
Google Scholar Google Preview WorldCat COPAC

Downloaded from [Link] by University of Western Ontario user on 14 December 2025