Document(s) 7 of 14

Current Social Policy Review Process

The setting for the current debate on social policy reform in Canada (including what some refer to as the crisis of the welfare state) parallels that of other industrialized countries in many respects. Common features include a secular growth slowdown topped by a recession in the early 1990s, an aging population requiring higher transfer and health-care expenditures, rising needs and costs associated with the increasing prevalence of the single-parent family, and public pressure to widen the coverage of various social services to new categories of recipients.

These factors, affecting the demand for, and availability of, public funds, have contributed to the near-universal fiscal crises in which industrialized countries find themselves. By the end of the 1980s, despite some austerity measures, budget deficits remained high, with Canada one of the more extreme cases. Policy reform is not just a matter of controlling social spending while maintaining a high level and quality of services. It is also a redesigning of the system to meet the changing needs of the new prevalence of temporary and insecure work from more frequent job and career changes, more self-employment, the single-parent family, and the more complicated family earning patterns associated with the multiple-earner family. These features all present serious challenges to the traditional social security structures.

Other common contextual factors are the rapid technological change, especially in the information area, and the related pattern of increasing globalization. Most of the industrialized countries are also involved in economic blocks, which imply regional economic integration going beyond the world trend in that direction. For Canada, the major event was the signing of the Canada–U.S. free trade agreement of 1989, followed soon by the creation of the North American free trade agreement (NAFTA) in 1993. Although for Canada, the free trade agreements do not represent as drastic a change as they would, or will, for many countries of Latin America and the Caribbean (LAC), they are certainly important enough for its lessons to be potentially instructive. Will they produce gains from trade that translate quickly into fuller and better employment of the labour force, or will they lead to widespread downscaling of work forces, wider earnings differentials, and jobs that are less secure because of the flexibility required to compete effectively in the international market?

Several features are special to Canada's social policy crisis/challenge. A number of these relate to its being small compared to the U.S., its main trading partner, although relatively open (high export and import to gross domestic product (GDP) ratios) and integrated with that partner, but having social goals, as revealed by past and current policies and structures, which differ significantly from those of the U.S. No other member of a major trading block is in such a delicate situation. To none is it likely to be so important that social policymaking be kept "national" even as the economy becomes increasingly international or intertwined with that of the U.S. Available data seem to confirm that the widening income disparities that afflicted the U.S and some other industrialized countries in the 1980s were significantly less marked in Canada (Berry and Stewart n.d.), and this difference is often interpreted as a result of the different social policy systems.

There has been widespread fear in Canada that competitive pressures implicit in a closer economic integration with the U.S. will undermine Canada's social programs (Banting 1992, p. 22). Social programs that raise the costs of production would push economic activity to the south, whereas ones that lower business costs by socializing some of them might be interpreted in the U.S. as subsidies and subjected to countervail or other trade reprisal. Optimists on this count argue that our distinctive social programs evolved during a period of increasing integration, that taxes reducing net incomes would be no problem, whereas those that raised costs would simply trigger an offsetting exchange rate adjustment. Regional development grants to industry might be challenged but generally available social programs would not.

Two other features of Canada's political landscape warrant attention: the importance of the provinces and the presence of a significant socialist party impact on social policy. Although the socialist New Democratic Party (NDP) has never achieved power at the national level, it has at one time or another governed four provinces and, over the years, has made major policy innovations in areas of provincial responsibility (like health) and introduced policy issues into the national political discourse. The current experience of governing at the provincial level under conditions of a serious recession and severe fiscal limits, has led to conflict (especially within the Ontario party) around the difficult question of unconditional support for traditional union/labour positions and has forced the beginning (or continuation) of what may be the ideological modernization of Canada's socialist party in the light of the more difficult fiscal constraints and the more open economy of today.

The regional and provincial dimensions of Canada's social, political, and economic structure have historically been very important. The relatively flexible federation has been a response to the large geographic dimensions of the country, the cultural division between Quebec and English-speaking Canada, and the varying economic conditions and structures across the four main regions: the Atlantic provinces, Quebec, Ontario, and the west. During the last two decades, regional strains have been intense, both because of the secessionist movement in Quebec (where a second referendum on this issue is expected in 1995) and because of the resentment from the west, especially felt in Alberta because of the national energy policy that, during the years of high oil prices, transferred much of the associated rents from producing to consuming provinces.

In important areas of social policy like health and education, the role of the national government in Canada's federalist system has been to participate in the design of systems and to use federal–provincial transfers in such a way as to induce the provinces to create what is in effect a national system. In the early stages of this process, the federal government used open-ended cost sharing arrangements with the provinces, thus creating strong incentives for the poorer ones to improve their systems in these areas, but also failing to build in strong cost-control incentives. Because the deceleration was less marked in Canada than in some other industrialized countries and because military spending was low, it was fiscally easier to expand social spending than would otherwise have been the case.

In the heartland province of Ontario, provincial spending grew at 13.3% per year in nominal terms and more than 6% in real terms over 1981–91, with an increasing share going to the social side, especially community and social services (Graham and Lightman 1992, p. 88). Earlier, the province had seen a major burst of spending on education.

In recent years, each element of the social spending package has come under scrutiny. This scrutiny has involved the recognition that:

• Policy effectiveness cannot be taken for granted and spending programs should not be judged by their size but by their product; hence, new monitoring and evaluation instruments need to be brought into play.

• There are too many negative incentives and lack of internal coherence built into elements of, or the overall system of, social policy.

• It is important to take account of the economic efficiency and growth implications of social policy, especially in the area of education/training but also in that of health, child-care, and others.

• With the scarcity of financial resources it is always necessary to ask how the costs should and can effectively be shared by the public sector and by private individuals or groups.

• Targeting may be the most efficient way to handle some services previously provided on a universal basis.

• It may be efficient to separate the funding and the delivery of certain services, with the private sector more often an effective delivery mechanism than an appropriate financier.

Evolution of the Social Support System

Nearly all the elements of the current structure of Canada's social support system have been put in place during the postwar period. Beginning with the unemployment insurance program in 1941, the federal government took the lead in redesigning and extending the system of social security. It is less complete than those of many European countries but more extensive than the American one. Most comparisons with other industrial countries put Canada toward the low end in public sector social spending.

The Organisation for Economic Co-operation and Development (OECD) figures for 1990 on what they refer to as "public social-protection expenditures," including health care, unemployment insurance, family transfers, disability pensions and services, and old age pensions, but excluding education (in which Canada is a relatively high spender) show expenditures of 18.8% of GDP. This significantly exceeds only those of Japan (11.8%), Australia (13.0%), and the U.S. (14.8%), and puts Canada in the same range as such low-income European countries as Greece, Ireland, and Spain. This falls substantially short of the higher income European countries, where the figures range from 23.4% (Italy) to 33.9% (Sweden).

Total social spending by Canadian governments shows a sharp upward trend between 1966 (when it was 14.9% of GDP) and a peak in the early 1980s (over 24% in 1982 and 1983), after which it fluctuated in the 22–24% range over the rest of the 1980s (Hunsley 1992, p. 106).

To these figures should be added indirect "costs," which take the form of tax rebates and the like and the subsidization or direct delivery of goods and services. Hunsley (1992, p. 109) estimates tax subsidies for 1988 of 53.8 billion. Together with direct expenditures of 140.8 billion in that year, this implies total direct plus indirect public spending in this area of about 30% of GDP, still without taking account of private expenditures not fully reflected in the tax rebates. After declining a bit from the peak in 1982–83, direct social spending rose again with the recession of the early 1990s and, barring major policy changes (which may be in the works), would seem likely to settle at a level several percentage points above that of the 1980s. In the narrower category of social security expenditures, there has been an increase from about 9% of GDP in 1989 to 13% in 1993 (Toronto Globe and Mail, 19 January 1994, p. A4, citing Statistics Canada).

Over the 1980s, social security costs for the elderly and health costs rose, whereas expenditures on education, unemployment insurance, and federal family benefits decreased. During the recent recession, social security and unemployment insurance expenditures rose rapidly (Hunsley 1992, p. 107).

As of the late 1980s, health and education accounted for about half of total public direct spending, and for 10.6% of GDP. Social security accounted for 43%, the chief elements being the old-age pension (8.4%), other pension plans (8.8%), and unemployment insurance (8.4%). Welfare (5.3%) and worker's compensation (2.7%) were smaller and the other categories relatively unimportant (Table 1).

Table 1. Total government direct spending in social policy fields, 1988, by program area.
Program	Spending (millions)*	Percentage of total social spending*	Percentage of GDP
Health	36,558	25.97	5.71
Education	31,295	22.23	4.89
Recreation and culture	5,341	3.79	0.83
Housing	2,947	2.09	0.46
Labour	3,765	2.67	0.59
Social security	60,891	43.25	9.52
Income supplements	16,154**	11.47	2.52
Pension plans	12,419	8.82	1.94
Unemployment insurance	11,818	8.39	1.85
Social assistance (welfare)	7,481	5.31	1.17
Workers' compensation	3,857	2.74	0.60
Family allowances	2,654	1.88	0.41
Child tax credit	2,064	1.47	0.32
Other	4,444	3.17	0.71
Total	140,797	100.00	22.00
*Based on Hunsley (1992, pp. 107–108). **Of which the old-age pension is $11,804 million.

As noted in the foregoing, the federal government encouraged expansion of various social programs in the 1960s and 1970s. The Canada Assistance Plan (CAP), put in place in 1966–67, replaced a group of categorical programs for people in need. Provinces were required to define in legislation what "need" meant, could no longer impose a period of residency as requirement, and were required to establish an objective appeal system for anyone denied assistance. As of 1989–90 these expenditures accounted for about 5.3% of total government spending in the social sector (Hunsley 1992, pp. 107–108). Although no upper limits were set on cost sharing, a team reviewing CAP in 1985 for the new Conservative government concluded that provincial government determination to keep welfare entitlements well below minimum wages (and to keep these latter low) had effectively controlled spending, better than the federal government could have achieved by itself.

The fiscal equalization system, begun in 1957 and later extended and constitutionally entrenched in 1982, supports the weaker provinces by providing unconditional cash transfers designed to allow all provinces to provide comparable levels of public services at comparable levels of taxation. The overall social support system has become highly redistributive across regions through the fiscal equalization, unemployment insurance and other federal income security/support programs (Leslie 1992, p. 48). The cost-sharing apparatus implied that federal payments would continue to expand as long as the provincial programs did and would also rise if interprovincial disparities did.

When the federally operated unemployment insurance program underwent a de facto conversion to a scheme of income support for the intermittently employed in 1971, outlays rose from $542 million in 1970 to more than $11 billion in 1985. Beginning in 1974, basic exemptions, including those for children and other dependents, were indexed thus removing a source of elasticity in the government revenue system.

When the upward pressure on spending came up against the economic downturn of the 1970s, the federal government began to look for ways to limit its fiscal obligations to the provinces. In 1977 the Established Programs Financing (EPF) system was set up. Under it the federal government agreed to make an annual fiscal transfer consisting of "tax points" (the estimated yield of a part of the income tax) and cash transfers, whose level was based on 1976 levels of cost sharing, adjusted for increases in GDP (Leslie 1992, p. 49). The EPF was the first of a long series of federal actions to transfer fiscal and program responsibility to the provinces, to limit federal fiscal responsibilities and to make the transfers more redistributive. In 1990 Parliament amended the original law such that cost-sharing payments under the CAP could not increase faster than 5% per year for provinces not receiving equalization. As of 1992, only 28% of social assistance outlays came from the federal government, compared to 50% in 1989–90. The third major form of fiscal transfer, the equalization payments, has also moved from an open-ended to a capped program, and is likely to shrink in future.

As noted in the foregoing, the pull-back of federal funding does not appear to have led directly to a deceleration of overall social spending, although it may be necessary to wait a few years until the dust has settled after the early 1990s recession to see how much of the growth in those years is attributable to spending caused by the recession. As awareness of resource stringency became greater in the 1980s, governments tried to target benefits with greater precision. This resulted in a number of deductions that had benefited people in higher tax brackets being converted to tax credits of equal value to anyone paying taxes but of no value to those not doing so. Hence, distribution impacts were not predictable in the absence of detailed information on where in the distribution the impact is felt.

There has been a tendency for the total social spending ratio, and some significant elements of it like unemployment insurance, to rise sharply in recessions (especially the last two) but not to fall significantly during the booms. Such a pattern could simply reflect the superimposing of cyclical movements on a positive secular trend, or it could reflect an asymmetric downward rigidity of certain types of expenditures. Certainly, the pull-back has put the provinces under intense fiscal strain, contributed to mounting budget deficits, and forced the onus onto the provinces for making hard decisions in the social area.

Amid the obvious need for fiscal restraint and care, new needs and social demands continue to appear. Gender pay equity, pioneered by the province of Ontario, has had a significant impact on both private and public costs, as the equalizing has tended to be in the upward direction. The "spouse-in-the-house" rule of eligibility for social assistance was dropped following a legislative review to determine which rules were contrary to the Charter of Rights and Freedoms (the constitution promulgated in 1982). The province adopted the spouse definition current in the common law and in use by Revenue Canada (Hunsley 1992, p. 123). This change substantially widened eligibility. The province has also expanded community based services (especially in-home care and home support services) and financial assistance to top up incomes for the working poor. These changes led to rapid growth in expenditures and caseloads. When the recession and industrial dislocation followed and the federal government limited growth in CAP contributions to Ontario to 5% per year in 1990, a sharp fiscal scissors effect was felt.

Under current circumstances, virtually all provinces are unhappy for one reason or another. Quebec has traditionally insisted on comprehensive responsibility for social policy, with federal involvement in this field viewed as inappropriate and inconsistent with the original terms of Confederation. As long as the federal transfers were being expanded and the associated objectives were largely consistent with the province's preferences, the system was tolerable to Quebec. Under retrenchment, however, and with the federal government withdrawing from previous commitments, there has been pressure for change (Leslie 1992, p. 51). The western provinces that would gain from a transfer of fiscal responsibilities have also been pushing for decentralization. Ontario's position on continued heavy federal involvement has been ambivalent, but the province has suffered severely during the recession and is quite unhappy about recent federal pullbacks. The Atlantic provinces are worried that a dismantling of the transfer mechanisms will harm them.

The opting out clause in the still-born Meech Lake Accord (whose purpose was to redefine the Canadian constitution in a way satisfactory to all of the provinces as well as to the federal government) drew much attention, mostly negative. The idea is that provinces would undertake their own programs in a variety of areas, including those under discussion here, consistent with "the national objectives." This concept, although implicit in many earlier arrangements for, and discussions of, elements of social policy (e.g., the setting of minimum wages) was, at least in this very explicit sense, new to the federal–provincial forum and never clearly defined during the Meech Lake discussion. It could be the peg on which a useful concept of responsibility sharing might be hung.

Implications of Globalization and Regional Integration

Canada has been a relatively open economy, with export/GDP and import/GDP ratios in the 25% range for several decades. Trade has been concentrated, with the U.S. share accounting for about 75% of the total. Tariffs and quantitative restrictions on trade between the two countries have been on the low side.

Canada's main motives for initiating a free trade agreement (FTA) with the U.S., which perhaps were not self-evident, included: (a) concern that the world was evolving in the direction of major trading blocks and fear of being left outside, and (b) a desire to avoid the uncertainty and risks created for Canadian exporters by unilateral and unpredictable imposition of trade sanctions by the U.S. on Canadian industries whose incursions in the U.S. market led to complaints and political pressure there.

Because expected trade creation from the FTA was not particularly large, some estimates of the static gains to Canada were on the low side, say 1–2% of GDP. Estimates that included gains from rationalization to take full advantage of assumed economies of scale but disregarded possible losses through firm migration to the U.S. (like Wonnacott 1975; Harris 1984) could be substantially greater (e.g., in the range 5–10% of GDP). If losses from firm emigration, for whatever reasons, including lower taxes, lower labour costs, etc., were large and gains from economies of scale were small, it appeared (to this author at least, although I did not see such calculations) that a net overall loss to Canada was a possible outcome.

Many of the structural changes showing up in the Canadian economy over the last decade or so appear to have come primarily from the nature of technological progress. This has destroyed middle-income, middle-skill jobs while creating high-skill, high-income jobs in smaller numbers and from the massive entry of the low-income (mainly Asian) countries into the world trading system to a degree that destroys the competitiveness of many industrial country industries in the more labour-intensive segment of the economy. It is evident that no plausible level of barriers to trade and investment flows would isolate Canada from these effects entirely, nor would the country wish to eschew the benefits that those processes bring. But it is probable, both for Canada and for other countries of the hemisphere, that the degree of integration into world markets is a determinant of the size of both the benefits and the costs of those phenomena.

Another general concern with integration into the world economy is that it increases the relative role of large firms, whether multinational companies or large national firms. Given that large firms tend to employ atypically capital intensive technologies, a shrinking role for small and medium enterprise could be expected to decrease the number of "good" jobs and worsen income distribution (Berry 1992).

Many Canadians believe that tighter economic integration with the U.S. implies an inevitable shift toward the U.S.-preferred market system, with its limited role for government planning and direction of the economy, loss of instruments for regional policy, and so on. Some (including many economists) view this with approval, whereas, for others (probably the majority), it is a source of concern.

Labour in both Canada and the U.S. tended to attack the FTA and NAFTA on the grounds that they would reduce wages, union power, and worker welfare. In effect, they represented a victory of capital over labour. How valid this fear will turn out to be remains unclear, partly because it has naturally been hard to sort out the effects of the early 1990s recession in Canada from those of the Canada–U.S. free trade agreement. In a recent econometric study, Gaston and Trefler (1994, pp. 21–23) conclude that the FTA "does appear to have contributed" to job loss. Their estimates imply that, when fully implemented, the FTA tariff cuts will ultimately reduce Canadian employment by 3.3%. Other important factors contributing to what they describe as the "devastating" labour market experience of 1989–91 were the recession, the strong Canadian dollar and the high interest rates associated with the fight against inflation. The authors note with some surprise that in the midst of dramatic job losses real wages were virtually stable. Thus the full response to the labour market disequilibrium took the form of job loss.

Social Policy Reform Process

Because the provinces are responsible for most of the spending areas on the social side, including education and health, they have historically had considerable scope for innovation. (Thus, the now national health system originated in the province of Saskatchewan under the administration of the socialist government of Premier Douglas. A burst of spending on education was undertaken in Ontario in the 1960s–70s under a centrist Conservative government.) This is a policy space that is likely to be increasingly ceded to them again, although now in the context of very serious financial limitations and with some continuing constraints on decoupling from the national patterns. Experiments considered successful have tended to be copied by other provinces or to work their way into a national system encouraged by the federal government through its conditional transfers.

Background analyses undertaken by academics, Royal Commissions, interest groups and others make an important contribution to policy discussion. Before the signing of the FTA a massive and in many ways in-depth study of its likely impacts on Canada was undertaken under the leadership of former Liberal Minister of Finance, Donald McDonald, commissioned by the Conservative government of Brian Mulroney.

In the social area, at least a half dozen important and usually influential studies of this sort have been undertaken in recent years, including the Task Force on Child Care (1986), the Program Review Task Force on the Canadian Assistance Plan (1985), and others. Private institutions organize policy research covering a wide range of the political and ideological spectrum, from the left-of-centre advocacy groups to the more conservative C.D. Howe Institute (publishing a range of policy studies in the social area this year) and the right-wing Fraser Institute. Currently, Human Resources Minister Lloyd Axworthy is struggling with the task of reforming the country's social programs and substantially reducing their cost (the objective is to save $1.5 billion in federal government transfers for social spending, in line with the government's pledge to cut the budget deficit to $25 billion within 3 years), a difficult challenge that, however, appears to be high on the list of policy priorities of the current Liberal government. Axworthy's social policy reform group confronts, among other political problems, the risks involved in producing a harsh-looking reform proposal just as the Quebec sovereignty referendum campaign is getting under way.

Under the Canadian parliamentary system, decision-making power is less widely shared than in a congressional system like the U.S. and civil servants (as in the "Ottawa mandarins") have traditionally been quite important. The last word tends to rest with a fairly narrow circle of politicians: the prime minister, the responsible cabinet minister, the minister of finance, and a few other influential cabinet members. Whatever its weaknesses, this system is in some ways less vulnerable to incoherent policy design caused by pressures from too many different directions.

Social Policy Under Regional Integration

The broad challenge for social policy at this time is simultaneously to meet both the social goals and the associated economic goals and constraints. Meeting this challenge will call for a more effective and subtle decision-making process than before. One element is that social welfare institutions may be particularly inflexible and resistant to change, because their past experience is one of growth, they provide what the citizenry gradually comes to think of as "rights," and some of the providers also see it that way. The current consensus or near-consensus on the performance of the main components of the social system would seem to be that:

• The health system is effective both in the results it achieves and in the security it provides, but is becoming excessively expensive. Cost-saving modifications will be necessary if it is to retain the current broad outlines.

• The formal educational/training system, although reasonably strong in certain respects is unnecessarily expensive at some levels and is weak in responding to rapidly changing market demand and in facilitating the education/work transition.

• The unemployment insurance system has evolved into an income maintenance system. It is partly for individuals and partly for regions, with negative incentive effects on job search, no significant capacity to foster retraining or effective job search, and little or no integration with the social assistance programs. The system has become part of Canada's permanent interregional redistribution system. The 1991 figure of unemployment insurance payments amounts to about $1,500 per employed Canadian (Courchene and Stewart 1992, p. 146).

• Social assistance, mainly in the form of income support, has been quite successful in reducing poverty rates for the aged, but less successful for other groups. Although poverty has been reduced over the last quarter-century, CAP seems not to have been a major factor, in comparison with the establishment and improvement of national standards for senior's benefits and the shift of married women into the labour force (Hunsley 1992, p. 122).

  Child poverty and poverty in single parent families remains high by comparison with European countries. The Luxembourg income studies put the rate of child poverty in Canada at 16.8% in 1981 compared to 24.1% in the United States, and 5–6% in each of Norway and Sweden. The differences are even bigger with respect to child poverty in one-parent families: 60% in the U.S., 51.2% in Canada, 13.7% in Norway, and 9.8% in Sweden (Ross and Shillington 1989, p. 85). Child-care funding has changed little since 1970 (Baker 1992, p. 68). More people, especially women, are taking more time off from work (absenteeism) at considerable cost to their employers (Akyeampong 1992).

• The ineffective interface between social assistance, insurance against short-term unemployment, and reentry into employment (including retraining activities, assistance with job search, etc.) remains a major flaw in the system. Too many elements of the system create antiadjustment entitlements that exacerbate the adverse shocks; the substantial spending in the areas of welfare, unemployment insurance, and training contributes little if any benefit to the country's long-run development prospects. The changing character of the labour market and the shrinking points of easy entry make the challenge in this area especially difficult.

  The internal incentives among the welfare, unemployment insurance, and training programs are very bad. Welfare is provincial, unemployment insurance is federal (although the federal government has backed out of financing except on an interim basis). Training has typically been federally financed, but provided, and sometimes guided, at the provincial level with policy decisions frequently driven politically. The average tax rate on the transition from welfare to work at the minium wage is generally above 100% (Courchene and Stewart 1992, p. 147).

  The same goes for the step function that links unemployment insurance benefits and earnings. Provinces are induced to create 10–12 week jobs to shift people from provincial welfare to federal unemployment insurance. Meanwhile, exit from both welfare and unemployment insurance is also discouraged because procedures for getting back on welfare or unemployment insurance are time-consuming and complex (Baker 1992, p. 75). The transition from school to the workforce is also clearly inadequate. One possible approach is the European apprenticeship system, but its merits in the Canadian context remain to be tested, and other options should receive serious thought as well.

• On matters of discrimination, by gender, race, age, etc., there has been considerable advance, especially in some provinces, although here too change in the labour market has made the job in some ways harder, especially for older people and the poorly educated. For women, however, the information technologies may have been a boon.

• Although a major increase in spending in the social areas will not be possible in the foreseeable future, this does not necessarily imply that major improvements cannot be made. Some past spending has probably had a low payoff, so major improvements in overall performance are not impossible.

The Competitiveness Question

The widespread argument that an increased need to focus on international competitiveness constrains what a country can do in the social area is based on two propositions that:

• A country must invest to achieve competitiveness and that such investment (e.g., in physical infrastructure, human capital, etc.) competes for funds with more "social" uses; and

• High taxes, which might otherwise be the way to achieve both objectives simultaneously, may be ruled out by the needs of harmonization under tighter economic integration.

The Harmonization/Convergence Question

Tax harmonization could oblige Canada to adopt lower rates and a lower level of progressivity than would otherwise be chosen. This will increase post tax and transfer inequality, putting downward pressure on the availability of public resources, and possibly forcing Canada's health and educational systems toward the substantially private and high-quality variance systems that characterize the U.S. Such a sequence would contrast sharply with the pattern in the European economic community, where the European social charter reflects to some extent, the preferences of Germany, which has used its political power to impose elements of its costly high welfare system on such entrants as Portugal and Greece (O'Higgins 1992, p. 4).

Two related issues fall within this convergence/harmonization area. First, how much convergence of economic outcomes can be expected from a tighter degree of economic integration, policies held constant, i.e., with such policy differences as currently exist more or less continuing on into the future. The answer to this question involves both the extent of factor mobility and whether it increases as a natural correlate of integration, and the impact on economic outcomes of the trade and investment integration itself.

Where factor mobility is high, convergence of wages (more precisely of labour costs) and returns to capital is almost a foregone conclusion. Where it is not high, the issue is more complicated, since trade and some degree of capital movement can also produce that result but are less guaranteed to do so. Where the nature of the convergence is undesirable, e.g., a fall in the wages of unskilled workers or a widening of earnings gaps to socially undesirable levels, the next question is whether policy interventions, allowable under the new agreements, can be used to offset such effects. Looking toward a future in which Mexico and probably other countries of the hemisphere will be part of the same economic block, the key problem for both Canada and the U.S. is the extent to which there will be wage convergence in which our wages adjust downward significantly. It must be assumed that both will attempt to impose some labour market restrictions if they become concerned that such downward pressures might become too strong.

The second issue is to what degree policy harmonization will be necessitated as a defense against factor mobility or other negative impacts, even when the nature of the harmonization goes against the preference of the policy-adjusting country, in this case Canada. (Policy harmonization that unequivocally benefits both countries need not be discussed further.) Where harmonization is resisted, new costs can arise.

Much convergence may already have occurred between Canada and the U.S. over their long period of close economic ties. It is unclear whether the FTA will lead to a significant further convergence, but the possibility warrants consideration. The economic integration implied by both the FTA and NAFTA relates to goods and services markets and to capital markets but not to the labour market. Even the already existing mobility of certain occupational categories, however, is high enough to imply significant costs if Canada loses a lot of expensively trained people to the U.S. as a result of holding earnings differentials lower (as with doctors, for example) than their high, and in some cases, rising levels in that country. There is a general and continuing risk of losing the most talented people because of smaller intraoccupational category differentials.

Capital mobility between Canada and the U.S. is also high enough to contribute to convergence. The lower labour costs in some parts of the U.S., due sometimes to lower base wages, but more systematically to lower nonwage costs associated with weaker unions and less payroll taxes, have induced some Canada-based firms to move to the U.S. Some movement of this sort would be expected independently of any new trade deal, but one effect of the FTA was to ensure continued access to the Canadian market even for a firm based in the U.S. Such access no longer constituted a significant reason for location in Canada. Firms planning to sell substantially in both markets were now more likely to want to locate in the U.S. because, as the usually larger market, the advantage of proximity would be significant in some industries. The issue under debate is whether these effects have been or will be quantitatively significant.

Harmonization has been discussed in several contexts including the tax system and the extent of proactive policy and nonmarket incentive systems. It has been viewed as most essential in the areas of corporate and personal income taxes. It is widely felt that higher corporate taxes could lead to quick loss of firms. Higher personal income rates run the risk mainly at the upper end, especially if our system is also more progressive. There are at least two arguments suggesting a shift to more indirect taxes.

The first, is that the system may be able to achieve greater overall tax/expenditure progressivity if more financing is done that way. The second is that consumption taxes may have less impact on plant or enterprise location and on mobile skills than do direct taxes. On this second point, however, Canada may be an exception because such a high share of the population lives near the U.S. border (O'Higgins 1992, p. 18). Property taxes might appear to have potential but some recent moves to the U.S. seem to have been influenced by these (Courchene and Stewart 1992, p. 141). A value added tax (VAT) seems the best bet as it is competitively neutral, exports being exempt and imports covered.

The Record on Harmonization and Convergence

The existing similarities in economic structure, average income, occupation-specific earnings, and income distribution between Canada and the U.S. might be taken as evidence of convergence over past decades, during which the two countries became more closely entwined economically. But the broad similarities in factor proportions, in particular the large endowments of natural resources per capita, undoubtedly account for the bulk of the similarity making it hard to judge whether economic integration is responsible for a significant part of it or not. It is generally accepted that convergence has occurred in Europe during the European Economic Community period, although the mechanisms remain to be worked out and a significant competing hypothesis is simply that middle-income countries often grow faster than high-income ones whether in such trading arrangements or not. One of the important questions to be answered in the next couple of decades is the extent of convergence between Mexico and Canada–U.S.

On harmonization, Banting makes a persuasive case that, so far at least, the record suggests considerable continuing independence of Canadian social policy from that of the U.S. in spite of convergence in some elements of the tax system. The history of tax reform in the 1980s demonstrates a conviction within the Canadian government that to impose higher corporate taxes than the U.S. would risk serious erosion of its tax base and revenues (Alpert et al. 1992). It is not clear, however, whether the new FTAs will have a greater effect in the direction of harmonization than did the gradual process of economic integration in the past.

As of the early 1970s, comprehensiveness of social programs was generally wider in Canada. Harrington (1984, p. 85) described the welfare state in the United States as "primarily for people over sixty-five"; pension payments were similar in the two countries. Canada has also had a stronger set of targeted programs. Despite the rhetoric of the War on Poverty in the U.S., political support for vertical redistribution through selective programs has been weak (Banting 1992, p. 25), benefits of the programs are low and limited resources move through them. Canada has had a wider set of income supplements and a stronger system of social assistance at the provincial and local levels.

Although in 1960 selective programs were a little over 20% of all total income security expenditures in both countries, in Canada they rose to 37.4% by 1980 and stayed around 35% through 1987, whereas in the U.S. they remained constant to 1980, then fell to 17.5%. As of 1979–81, the impact of the system on poverty (defined as 50% of median disposable income in each country) was quite a bit larger in Canada (52.7%) than in the U.S. (38.1%) (Banting 1992, p. 27).

Over the last couple of decades, the debates on education, training and child care have been quite similar in the two countries. Canadian concern, however, focuses especially on health care, where the gap between the two has been widening. In Canada, private expenditure represents about a quarter of the total. In the U.S. Medicare provides insurance for the elderly and Medicade, a means-tested program delivered by state governments provides care to 42% of the nation's poor.

The rest of the system mainly relies on the private sector. About 13% (mostly low-income workers and their families) are uninsured and another 8% seriously underinsured (Banting 1992, p. 28). In 1971, the two nations devoted similar shares of GDP to health care but, by 1989, the American system was considerably more expensive at 13% of GDP. In the U.S., health insurance coverage has eroded over the last decade as employers have pushed more of the burden onto employees and, in many small firms, the coverage has collapsed, exacerbated by the trend from manufacturing to services and by the recession (Banting 1992, p. 28). Despite federal initiatives to expand the coverage of Medicade (e.g., to pregnant women and children), increasingly restrictive state rules seem to be reducing its overall reach, and sharp limits on reimbursement rates lower the willingness of doctors and hospitals to accept large numbers of such patients.

The pattern of public spending on the elderly is similar in the two countries, with the biggest difference lying in the more redistributive character of the Canadian program. In the 1980s and 1990s, although political attacks on major benefits were generally turned back in both countries, the changes that did result heightened the contrast on that front as the low-income groups in the U.S. suffered more from the cuts, whereas the Canadian program became more redistributive than before.

In the U.S., unemployment insurance is a federal–state program, funded by both federal and state payroll taxes, federal legislation generally determining the employment covered but the states determining the qualification periods and the level and duration of benefits. Canada's program has been exclusively federal, with contributions supplemented until 1991 by general federal revenues. There are regional differences in the qualification periods and the duration of benefits, leading to a substantial regional redistribution to the poorer parts of the country. It has provided greater income protection, and consumed a large 3.2% of GDP, versus only 0.4% in the U.S. (Banting 1992, p. 30). Although the Canadian system saw little change in the 1980s, the Reagan administration substantially remodelled the American one; this, however, tended to go unnoticed in the early part of the decade.

The 1991 Canadian reform brought the two closer again and the U.S., after long battles between the President and Congress, temporarily relaxed some of the harsher steps taken in the 1980s by extending benefits for victims of the recent recession (Banting 1992, p. 31). The gaps remain large, however, both in terms of the ratio of unemployment insurance payments to average weekly earnings (in 1991 about 45% in Canada and 35% in the U.S.) and especially in terms of the share of unemployed who are unemployment insurance recipients (around 90% versus around 40%).

Convergence is clearest in the child benefits area. The gap was large in the mid-1970s when Canada had a universal system of family allowances as well as the tax exemption for dependent children, and the U.S. was the only industrial nation with no general child allowance program other than such tax exemptions (Banting 1992, p. 31). Since then, both countries have moved toward more targeted systems. In Canada, beginning in 1993, the system is one single income-tested child benefit of $1,020 per child plus certain supplements (Banting 1992, p. 33). Differences will erode further in future, although Canada's system remains substantially more redistributive, and initiatives in the U.S. seem not to have fared well.

The pattern of stronger redistribution toward the poor in Canada is repeated in the area of public assistance. Although support to single individuals is pretty limited in some provinces, support is provided at some level to all persons in need. In the U.S., AFDC (Aid to Families with Dependant Children) was traditionally restricted to single-parent families, although, by the late 1980s, about half of the states had extended it to two-parent families when the principal wage earner was unemployed. Since 1990, all states must do this but only for 6 months of the year (Banting 1992, p. 34).

Levels of assistance are generally considerably higher in Canada. The fact that 49 states are not allowed to have fiscal deficits resulted in a wave of assistance reductions in 1991. Canada does not seem to be experiencing the wave of cuts like the U.S. in the early 1990s.

The difference between the two countries in the redistributive role of the state increased in the 1980s and probably the early 1990s, with the tax/transfer system's impact on child poverty rising in Canada between 1981 and 1987 from 34% to 41%, whereas that in the U.S. fell from 24% to 8.5% (Banting 1992, p. 37). The weakened redistributive role of government in the U.S. contributed to increasing pre tax/transfer inequality, whereas in Canada that role remained strong enough to largely offset that trend (Banting 1992, p. 38).

Overall, the pattern is one of divergence, with increased gaps in health care, pensions, public assistance, and the redistributive impact of state activity in general, resulting mainly from the erosion of coverage and benefits during the Reagan–Bush years. Convergence appeared in the case of child benefits (especially) and of unemployment insurance. It did not flow unambiguously from closer integration of the two economies, however, but rather from changes occurring in both countries but not apparently driven by the process of integration. On the child benefits front, the explanation seems to be common ideological trends.

In the unemployment insurance case, the changes in the early 1990s narrow the differences. The Canadian changes seem to flow from Canadian conservatism, with the focus on deficit control and shifting from income support to retraining (Banting 1992, p. 39). This agenda is also rooted in a view of Canada's position in the international economy and the pressures on Canadian labour markets and educational policy. The apparently limited harmonization impact of NAFTA at this time may be misleading because most of its effects may yet to be felt.

Rationalization, Privatization, and Cost Saving

Maintenance and strengthening of Canada's social sectors will require significant efficiency improvements in all of the major ones. It appears inevitable that provinces will begin to privatize various parts of the delivery system for social programs.

Although provincial experimentation is already ongoing, many Canadians express concern that as soon as completely free access to medical attention is compromised, a central feature of the health system is changed. This view is mirrored in the federal Ministry of Health's announcement that Quebec was in violation of the underlying principles of Medicare when it introduced very small user fees for hospitals when clinics are free nearby (Courchene and Stewart 1992, p. 144). More seriously, Quebec has proposed a user fee where a record would be kept of the value of medical services rendered to each person, and higher income individuals would pay something for services. This system could be quite progressive in its distribution impact, and the private risk it entails might be handled by putting some sort or ceiling on the charge over one or a few years.

Others suspect that free access is the source of much waste of services, but evidence from Canada and cross-country comparisons suggest that modest fees do not contribute importantly to cost control. The main sources of the cost explosion, apart from the aging of the population, are the high cost of modern medical technology and the high earnings levels of most medical personnel. The British and Japanese systems, for example, warrant attention as positive examples of achieving adequate to very good performance at much lower cost than either the Canadian or the American systems.

During periods of easy funding, either in the form of the open-ended, cost-sharing federal–provincial system or of access to natural resource rents (oil in Alberta, phosphates in Saskatchewan, etc.), provincial governments have tended to expand the hospital system to rather small communities. Several provinces (most notably Alberta, but also Ontario and others) have moved to close a number of these facilities. Rational management of the hospital system can doubtless be improved, but community interests and needs should also be weighed in the balance. Communities willing to raise enough money locally to support hospitals, presumably should have them. My own suspicion is that much of the economic analysis undertaken around this issue is of doubtful quality and does not thoroughly probe the costing and interface options with larger hospitals.

Tensions around the question of the earnings levels of medical personnel are inevitable in Canada, given the proximity to the U.S., the history of high physician incomes with origins in a time when their supply was very scarce, and the imperfect market which exists in medical services. (During the recent shift back toward the central role for family doctors in the U.S. system, the lack of output there has been made up by wholesale hiring from Canadian medical schools.)

The extra billing question has been debated for several years in Canada. The main intent of the proposal was to raise doctors' incomes and to allow the market to remunerate differently according to the skills or reputation of the physician. It creates two major dangers. First, the more widely recognized was that it could lead to a two-tier (higher and lower quality) health system. Second, it could also raise the total social cost of medical services, even to the point of exceeding those of a purely market system because, at the limit, it could lead to substantial market-set markups over and above a large subsidy from the state. The argument that universality is the only guarantee that the poor will get decent quality is not straightforward. O'Higgins (1992, p. 9) points out that affluence leads to a demand for more individually differentiated services and higher quality.

Universal provision might tend to diminish rather than increase equity if it subsidizes the rich by freeing resources for discretionary spending and places a quality limit on the poor. O'Higgins notes that the British National Health Service has been a remarkably effective way of holding down health costs, with tight controls on spending, probably suppressing consumer choice, and with rigid patterns of service provision, heavily dictated by medical consultants. Some indication of how high such costs could go is provided by current dental fees, clearly buoyed by the prevalence of insurance systems that make cost largely irrelevant to the buyer. The main intent of the Canada Health Act of 1984 was to eliminate extra billing by physicians and user fees, viewed as an impediment to access for lower income Canadians. But if the federal government does not continue to provide cash payments it will lose its capacity to enforce national health standards. The federal government has announced its intention to withhold other transfers if provinces contravene the Canada Health Act (Graham and Lightman 1992, p. 96).

It is clear that Canada's health system will have to struggle with cost control for many years. There will be no easy answers, and the quality of analysis around the issue will be very important. The American Health Management Organisations (HMOs) will no doubt provide relevant insights. The United Kingdom, has made innovations that also deserve study. There, some elements of the public service have been transformed into executive agencies with performance contracts and business plans, and the purchase of services has been separated from their provision, so purchasers could switch providers for cost or quality reasons (O'Higgins 1992, p. 7).

Policy issues in the education field are even more complicated, partly because the measurement of performance is harder and because there is on balance less public confidence in the quality of the people managing this system or, at least, the public component of it. Public concerns are fuelled by Canada's very expensive educational system failing to produce students who do well in internationally standardized tests of mathematics, science, or other fields. Some blame the curriculum. Many blame a complacent, ineffectively managed public education system in which, as in the health field, it is arguable that some employees are paid excessively.

There is a disconcerting amount of evidence that the well-funded public school system of Ontario has high rents that attract many individuals with limited interest or skills in the field of education. They control too many important administrative positions, in which they constitute an impediment to innovation, progress, and general performance. The Ontario system seems to demonstrate the adage that quality education cannot be produced by throwing money at a system, and may even demonstrate that too much money is an ingredient in a recipe for failure. The system appears to have many high-quality, dedicated teachers, but some are frustrated and ineffectively used.

The issue of the appropriate level of subsidy for private education has naturally surfaced. For many years, in provinces like Ontario, much of the nonpublic primary and secondary education was for a long time that of the Roman Catholic schools. In the early 1980s, the province agreed to help finance these schools.

Proponents of a freer market system would argue that the playing field should be levelled further with parents receiving vouchers usable in the schools of their choice, with the per student transfer equal to or near the average cost per student in the public system. This would raise the range of educational options available to parents, within some prescribed limits of curriculum, etc.

The slippery slope leading to the underfunded and neglected public schools of the American urban slums is a stark cautionary. Yet, in an age where considerable innovative capacity seems important, and where the sort of discipline and emphasis on learning that the public schools have had difficultly instilling, the potential value of a fairly widely accessible system of private schools is clear.

It will be difficult to put in place an effective regulatory apparatus capable of identifying the sins profit-oriented private schools are most likely to commit (taking advantage of uninformed parents through unprofessional, high-pressure sales techniques, excessive cost cutting in some areas of the curriculum, etc.), while matching respect for innovation with clarity of rules, etc. How difficult this job can be is well exemplified by the rancorous debate around the success or failure of new approaches to inner-city schooling in the U.S.

With respect to curriculum, there are many calls for a more vigorous response to the technological society that is increasingly dominating our lives. Some have highlighted the need for greater focus on the training and use of "para-technicians" (Economic Council of Canada 1990). Like the U.S., we have given too little respect to para-medics, -lawyers, and -engineers or technologists. (One of the sources of the high cost and low efficiency of the American health system is the lack of family doctors and the tendency for people to go directly to, or quickly be referred to, specialists, an advantage of the Canadian system where there are more family doctors.)

As Courchene and Stewart (1992, p. 145) put it, we have been rich enough to be a professional society, while abandoning the middle-skills range and skill-upgrading. Especially for small, medium, and fledgling firms the problem of underemploying engineers to do things that in Europe would be handled by technologists can be serious. One source of the problem is the tendency for the high-skill professions to cut the para's out of the market by limiting their functions through legislation. The recent revival of midwifery involved a battle on this front. But the more open the market is to alternative service providers the more society must worry about the control of quackery and incompetence.

Providing effective assistance to working-age people involves helping people with long-run causes of nonemployment or low income, or both; helping people with usually short-duration unemployment; and assisting reentry for both by retraining where necessary and by incentives to reenter. The distribution of the responsibilities for social assistance, unemployment insurance, and training/retraining among different institutions, different levels of government, and persons with different economic perspectives has, in most countries, led to incoherence in the responses to these needs. When inertia is added to the list of causes, a formidable challenge to effective reform is present. Canada's system has serious problems in each of the three areas.

Although there has been reasonable professional input in the design of the social assistance and unemployment insurance systems, the latter has been dangerously broad in the context of Canada's economic and political structure, whereas the training system has never been adequate, and its nonintegration with the other two systems has been striking. In some, perhaps all of these areas, the current design was better suited to the conditions of an earlier period, when a higher share of unemployment was indeed cyclical and essentially short term (making the concept of "insurance" the relevant one), and when most families had one earner with a permanent job as contrasted to today where many have more than one earner but many more jobs are part time or temporary. A system that worked fairly well under those earlier circumstances could not be expected to work well now.

An issue of some importance in the area of social assistance is whether people should have the right to a reasonable level of income even if they opt not to work. Some incline to say yes partly out of fear that the process of technological change is pushing the marginal product of unskilled labour very low leading to a permanent job shortage for the unskilled. Perhaps the only way to match high levels of labour force participation for the unskilled with reasonable income levels is to provide a guaranteed minimum income. Many of the problems of designing an effective combination of income security, incentive to work, and training would be simplified by such a step. But most estimates of the tax levels required to create this option at a clearly adequate income level put them significantly above those currently observed in OECD countries (O'Higgins 1992, p. 16).

Document(s) 7 of 14