Chapter Three: The Decline Of Ontario's Occupational Pension System

3.1 Introduction

For at least 20 years — perhaps 30 — the percentage of the province’s workforce enrolled in occupational pension plans has been declining. This chapter analyses the extent and nature of that decline, its causes and its consequences. Such an analysis is essential not only in order to determine what, if anything, can or should be done to arrest or reverse it, but also to ensure that recommendations concerning other issues are made with a full appreciation of how they are likely to affect coverage.

Accurate data is particularly important in measuring and interpreting the decline in pension coverage. However, for reasons outlined in Chapter Two, this chapter must begin with a caution to readers: despite the extensive attempts of Commission staff and consultants to enhance its accuracy, all data cited in this and other chapters should be treated as somewhat suspect. A special problem in the current chapter is that, from time to time, I have had to rely on data concerning not just workers under Ontario jurisdiction, but all Ontario workers covered by any sort of pension plan (including those enrolled in plans under the jurisdiction of the federal government or other provinces). Occasionally, indeed, I have had to draw implications from Canada-wide statistics. When I had to use other than Ontario jurisdiction data, I have signalled this departure in the text.

3.2 The Extent and Nature of the Decline

3.2.1 The decline in occupational pension coverage: Some basic facts

As Ontario's workforce has expanded dramatically over the past 20-odd years, so too has the membership of occupational pension plans. In fact, as of 2005, about 300,000 more Ontarians had workplace pensions than in 1985. However, plan membership has grown much more slowly than the workforce itself and as a result, the percentage of workers covered by occupational pensions has fallen. Figure 1 shows the erosion of pension coverage over this period from a high point in 1985 of just under 40% of Ontario workers to about 34.7% by 2005 — noticeably below the Canadian average of about 38.5% in the same year (the Canadian average includes federal jurisdiction workers). While the decline in coverage has been gradual rather than sudden, it is still substantial. And it may get worse: many observers — rightly or wrongly — predict a further significant decline in defined benefit (DB) coverage in the near future.

Figure 1: Ontario-Canada; Occupational Pension Plan Coverage by Ontario Plans as a Percent of the Paid Labour Force, by Regulating Authority

Ontario-Canada; Occupational Pension Plan Coverage by Ontario Plans as a Percent of the Paid Labour Force, by Regulating Authority

Source: Informetrica, 2007 and OECD

Pension coverage is concentrated in specific sectors of the labour force and economy, especially in unionized workplaces. The Commission's research shows that some 76% of unionized workers are active members of occupational pension plans, as compared with 28% of non-union workers. And not only have unions generally gained pension coverage for their members, they have often secured better pensions and pension-related benefits for their members than workers in non-union enterprises typically enjoy. These benefits include indexing, portability and long-term disability, as well as coverage for both part- and full-time workers.

There is a similarly strong correlation between pension coverage and large workplaces. Large employers with the capacity and financial resources to initiate and administer sophisticated human resource policies often make use of pensions to recruit and retain workers. In workplaces with more than 1,000 employees, pension coverage runs at about 60%; at the other end of the continuum, for workers in small enterprises with fewer than 20 employees, the coverage rate is 10%.

Coverage rates also differ considerably between the private sector on the one hand, and the broader public sector (provincial public service, schools, colleges, universities, hospitals, municipalities and Crown agencies) on the other. As of 2005, about a million Ontario private sector workers were enrolled in pension plans regulated by the Financial Services Commission of Ontario (FSCO), a figure that has remained fairly steady for about 20 years. However, because of the substantial growth of the labour force, the extent of private sector pension coverage actually declined during this period — from about 32% in 1985, to 25% in 2005. By contrast, during the same 20-year period, plan membership in the broader public sector in Ontario grew from 480,000 to just over 700,000 members. The coverage rate in the public sector fluctuated during this period, but stood at about 78% in 2005.

As these figures suggest, the gap between public and private sector coverage has grown considerably, and public sector membership now accounts for a disproportionate share of the total. Although the private sector accounts for about 81% of all workers, it contains only about 60% of active plan members. On the other hand, the public sector, which accounts for some 19% of all workers, contains about 40% of active plan members. And one more fact whose significance appears in due course: public sector pension plans are relatively small in number (less than 2% of all plans), but they tend to be much larger than their private sector counterparts.

Linking all of these observations, it might fairly be said that the most financially secure workers — those who enjoy permanent, full-time employment, union representation and high earnings, and who work in large or public sector workplaces — are also most likely to belong to DB plans. Conversely, those who are the least secure — non-union workers who have part-time, temporary or contingent jobs, receive lower pay and are employed in small workplaces in the private sector — tend to have little or no access to occupational pensions.

While, for reasons discussed below, a decline in coverage has been the dominant tendency over the past 20 years, one important improvement in pension coverage stands out: coverage rates for men and women are now almost identical. However, significant disparities in coverage remain among other groups.

3.2.2 Changes in plan governance and design

The decline in overall occupational pension coverage has been accompanied by significant changes in plan governance and design. The general tendency of these changes has been in the direction of joint sponsorship and funding and, in some cases, joint governance as well, particularly in the public sector. The significance of this shift is canvassed in later chapters.

As of March 2007, just over 8,000 pension plans were regulated by FSCO. About 1,800 are so-called individual pension plans (IPPs) — that is, plans with fewer than 10 members. These plans primarily enrol senior executives and do not represent a significant proportion of overall coverage. Of the remainder, just over 6,000 plans are sponsored by a single employer (single-employer pension plans, or SEPPs), while 127 are multi-employer pension plans (MEPPs). Membership overall is almost evenly balanced between these two types of plans: 55% of plan members are enrolled in SEPPs, 45% in MEPPs.

The growing importance of MEPPs within the system partly reflects their historical origins in sectors such as construction and the garment trades, which were populated by a large number of small and sometimes ephemeral enterprises. In these sectors, workers tended to look to their union, rather than to their employer, for pensions and other benefits. Typically, MEPPs in these sectors were funded entirely by negotiated employer contributions, but governed either by unions alone, or by some combination of union and employer trustees. In more recent times, however, unions have managed to launch new MEPPs in sectors of private employment outside their traditional base, and even more importantly in terms of overall coverage, in the broader public sector.

MEPPs are not the only plans that contemplate the participation of worker representatives in governance. A 2005 amendment to the Pension Benefits Act authorized the establishment of jointly sponsored pension plans (JSPPs), most of which are also MEPPs in the colloquial if not the regulatory sense. Current JSPP enrolment is about 600,000 workers, or 35% of all active plan members. While in principle both MEPPs and SEPPs can be jointly sponsored, and may be located in either the public or private sector, all existing JSPPs are in fact extremely large multi-unit public sector plans.

Last but not least, an important shift in plan design has been the growth of defined contribution (DC) plans. As I explained in earlier chapters, the precise nature and extent of the growth of DC plans is difficult to track with available data. However, I can say that in 1985, about 8% of active members belonged to DC and composite plans — a figure that grew to 18% by 2005. While this represents a significant movement from DB to DC plans, it is far less dramatic than the movement that occurred during this period in the United States and the United Kingdom. Thus, DB plans continue to dominate the Ontario occupational pension landscape.

3.2.3 Voice and participation

As noted above, and as depicted in Figure 2, most DB plan members (80%) are represented by unions, and of those, a significant majority (69%) are enrolled in plans that are either jointly sponsored and governed (JSPPs), or are governed by boards of trustees, at least half of whose members must be union nominees (MEPPs). Consequently, either indirectly through the negotiation of a collective bargaining agreement (CBA), or directly through their representatives on boards of trustees, active plan members typically have an opportunity to influence plan design and decision-making. Only about 20% of DB active plan members are found in workplaces in which the employer both sponsors and administers the plan.

Figure 2:Active Members of Active Defined Benefit and MEPP Plans, by Collective Bargain Agent, 2006 (not including DC & Other)

Active Members of Active Defined Benefit and MEPP Plans, by Collective Bargain Agent, 2006 (not including DC & Other)

Source: AIR 2006, OECD

Moreover, as Figure 3 shows, some form of worker voice in governance remains a dominant characteristic of the entire occupational pension system, even when DC and other types of plans are taken into account. Some 71% of all active plan members are found in unionized workplaces; only 29% are found in workplaces where the sponsor unilaterally initiates and administers a DB, DC or other type of plan.

Figure 3: Active Members of Active Plan by Plan Type, Benefit Type (DB, DC & Other) and Collective Bargaining Agent, 2006

Active Members of Active Plan by Plan Type, Benefit Type (DB, DC & Other) and Collective Bargaining Agent, 2006

Source: AIR 2006, OECP

This distribution of active plan members is significant because, to a large extent, the PBA assumes a rather different set of arrangements. The paradigmatic pension plan under the PBA is the SEPP, which many assume is typically administered by employers themselves, with or without the assistance of professional advisors and service providers. This is the type of plan whose fate dominates debates over pension coverage, funding and regulatory strategies. Of course, the PBA regulates plans, not members, and the great majority of plans do conform to the PBA stereotype, or paradigm. Nonetheless, once it is understood that most workers are enrolled in plans where they have a voice (or the possibility of a voice) in its design and governance, the focus and tone of discussion change considerably. This is particularly true in relation to arguments over the decline of DB pensions and what can be done to arrest it.

Further, much commentary concerning the DB pension system — in both the public and private sectors — rests on the assumption that the cost of these pensions is borne entirely by the sponsor. To the contrary, in some private sector plans, in virtually all public sector plans and in all JSPPs (the latter two groups overlap), active members contribute to their own pension. Those who do so constitute a clear majority of the membership of all plans.

Finally, it is often asserted that sponsors will terminate their DB plans unless they can be assured of particular legislative changes. While any further contraction of the DB system would be very undesirable, and while many changes sought by sponsors may be justified on their merits, most plan sponsors are not free in a legal or practical sense to take such action unilaterally. Sponsors with organized workers — about 80% of all DB members (see Figure 2) — would have to overcome the resistance of the unions that, in many cases, won pension coverage through collective bargaining.

3.2.4 Conclusion

To sum up, Ontario’s occupational pension system has been undergoing gradual changes for 20 years or more. In certain respects — such as an increase in absolute numbers of workers covered, the avoidance of a sharp year-over-year drop in coverage rates, the achievement of gender equality in membership, and the introduction of innovative governance structures and plan designs — one might construe these changes as evidence that the system is adaptable and healthy. However, neither internal change nor external developments have so far succeeded in making occupational pensions significantly more attractive to sponsors, or in arresting or reversing the long-term downward trend in overall coverage. The next section seeks to explain why.

Back to top

3.3 Explaining the Decline in Coverage

Analysts and stakeholders have proposed many different explanations for the decline in pension coverage. Some suggest that changes in labour markets, driven by the changing structure of Ontario’s economy, account in large measure for reductions in pension coverage. Others suggest that changes in demography have placed the occupational pensions system under considerable stress. Some contend that excessive or inappropriate regulation encourages employers to defect from the occupational pension system. Others suggest that both workers and employers, for quite different reasons, have developed a less favourable view of DB pension plans in particular. Finally, and most drastically, some analysts contend that DB plans — the dominant form of occupational pension plan — were structurally flawed from the beginning, and that their flaws have simply become more obvious in the years immediately following 2000, during which the system was buffeted by the “perfect storm” of declining long-term interest rates and falling equity prices.

Each of these explanations for the declining coverage of occupational pension plans is reviewed below. So far as possible, each is assessed on the basis of evidence available to the Commission, though it must be said that in regard to some explanations, evidence is in very short supply.

3.3.1 Changes in Ontario’s economy and labour markets

Our research, and that of others, suggests that changes in union density — the proportion of workers represented by unions — correlates closely with pension coverage rates. This makes intuitive sense, given the higher rate of pension coverage in unionized workplaces noted in the previous section.

Figure 4 shows that the Ontario labour force grew from 4.1 to 5.8 million workers between 1985 and 2005 (right-hand axis). However, union density and pension plan coverage in Ontario both declined as a percentage of the workforce during the same period, more or less in tandem. Union density decreased from about 31% of the workforce in 1985 to 26.8% in 2005, while pension plan coverage decreased from about 39% to just under 35% (left-hand axis). In fact, neither unions nor pension plans suffered a significant loss of individual members during the period; declining density and coverage resulted from their failure to grow at the same rate as the workforce as a whole.

Figure 4: Ontario Labour Force Growth, Union Density and Pension Plan Coverage

Ontario Labour Force Growth, Union Density and Pension Plan Coverage

Source: PPIC, CANSIM 279-025 and 282-0078

The Commission’s research, and the research literature in general, provide a number of pertinent examples and explanations of how and why union density and pension coverage rates track each other so closely.

To take an obvious example, the public sector is more densely unionized than the private sector and pension coverage in that sector has been relatively stable. By contrast, in the private sector, union density and pension coverage rates have both fallen, more or less in tandem. Another example: between 1987 and 2003, union density dropped significantly in the manufacturing sector and rose slightly in the service sector so that the gap in unionization rates between the two sectors closed considerably. This shift was apparently accompanied by a similar convergence in rates of pension coverage. And another: the proportion of prime-aged women in both union membership and in pension plan membership has increased at pretty much the same rate.

Moreover, changes in union density and in pension rates seem to be closely correlated not only to each other but to employment trends in the manufacturing sector — a bulwark of both unions and DB pensions. Since 1988, Canadian manufacturing employment has seen two major periods of contraction, from 1990–1993 following a cyclical downturn and the advent of the original Canada–U.S. free trade agreement and, after an intervening period of recovery, from 2005 to the present. During this most recent contraction, Ontario alone has lost more than 180,000 — or one in six — manufacturing jobs. The full impact of this contraction has yet to be experienced, let alone documented. However, because it has hit the “big three” auto companies with particular severity, and because they have historically been associated with high rates of unionization and pension coverage, the next few years are likely to see a sharp drop in both.

Ontario has also experienced a similar, though less pronounced, decline in employment in the broader public sector, which accounted for approximately 25% of the provincial workforce in 1991, but only 21% in 2006. Once again, employment in an area of high union density and high pension coverage has been shrinking in relative, though not absolute, terms. And once again, the impact on overall rates of pension coverage has been negative.

Finally, changes in the duration and perquisites of job tenure seem likely to affect pension coverage adversely. Several decades ago, workers may have had one job for their whole working lives and, especially if they were unionized, a reasonable expectation of a pension thereafter. Now, it is not uncommon for workers to change jobs and even careers several times in their working lives and to arrive at retirement with inadequate pensions, one or more deferred pensions from previous employment or none at all. Nor is it uncommon for workers today to hold several part-time, contract, seasonal or temporary jobs at once. Whether or not these workers are unionized — they often are not — employers often decline to provide pensions for them, even though the employer’s “core” full-time workforce may enjoy pension coverage. For example, the Commission heard from university sector stakeholders that sessional faculty, who may work at several institutions and are often unionized, nevertheless lack access to pension plans at any of those institutions.

According to one recent study, declining union density, dis-employment in manufacturing and public employment, and new forms of job tenure may together account for as much as 90% of the decline in pension coverage of young men and 75% of prime-aged men during the period 1986–1997. Indeed, according to this study, declining union density alone seems to account for 40% of the decline in pension coverage for men and young women.

Future trends in pension coverage are no easier to predict than trends in the labour market or in the economy as a whole. However, one scenario comes to mind that might lead to increased rates of pension coverage. In sectors of the economy that are already experiencing, or will soon encounter, labour shortages, employers may feel obliged to offer workers inducements to remain in their present jobs. DB pensions might be one such inducement, along with innovative schemes to encourage phased retirement for key workers at the end of their careers. However, this scenario is largely conjectural and seems unlikely to occur in the near term.

3.3.2 Demographic changes

Demographic changes are often cited as contributing considerably to the decline in pension provision. Population aging, in particular, generates great pressure on pension systems in all industrialized welfare states. Indeed, the proportion of Canadians aged 65 or over is expected to double to approximately 25% of the total population within the next 25 years. This will likely pose two challenges for the occupational pension system. On the one hand — subject to the effects of immigration, discussed below — the ratio of active to retired plan members is likely to fall considerably. On the other, as the workforce ages and as longevity increases, more retirees will have to be supported for longer periods of time. While these developments ought to have been foreseeable, studies suggest that some plan sponsors and their actuaries using out-of-date mortality tables may have underestimated the risks they will likely confront in the years ahead. If actuarial underestimates ran as high as 15%, as some U.S. estimates suggest, the costs of funding plans will increase as more modern mortality tables are adopted. This is especially true for mature plans, where the number of retirees is already approaching or outstripping the number of active and deferred members.

Population aging is not the only demographic factor affecting pension coverage. Ontario has the most diverse population in Canada, largely as a result of immigration. According to recent statistics, almost 20% of its population belong to a visible minority group, and members of such groups comprise some 54% of all visible minority Canadians. Moreover, Ontario receives 57% of all immigrants to Canada, and 27% of all Ontario residents are foreign-born, a figure that rises to 44% in the Greater Toronto Area. If occupational pension coverage of members of visible minority groups and recent immigrants lags behind that of Canadian-born workers, as do their earnings in general, their presence in ever-increasing numbers will significantly affect the coverage and effectiveness of Ontario’s occupational pension system.

In fact, only one study has so far addressed the issue of pension coverage of immigrants and visible minorities. Its conclusions are unsurprising: immigrants — both male and female — have less pension coverage than Canadian-born workers. However, coverage increases with time spent in Canada, and coverage for male immigrants catches up with that of their Canadian-born co-workers within 10 years. This somewhat positive finding is qualified by three others, which are less so. First, coverage for female immigrants has not been catching up to Canadian-born female workers. Second, pension coverage has remained lower among visible minority immigrant men than among other immigrant men. And third, immigrants who arrive from abroad mid-career — even if they ultimately gain entry to an occupational pension plan — will obviously accumulate fewer years of pensionable service and consequently can expect lower pension benefits on retirement. These findings are of particular significance since immigrants will also likely receive lower benefits from the public pension systems (OAS/GIS and C/QPP) given their shorter residency in Canada, time in the workforce and period of contribution.

If Ontario continues to depend on immigrants to rejuvenate its aging workforce and to sustain its economy, as seems likely, it will ultimately have to address the issue of how its newest citizens will provide for themselves in retirement. And if Ontario continues to pride itself on the ethnic and racial diversity of its population, it will have to deal with the probability that many members of visible minorities who suffer well-documented disadvantages during their working lives will continue to do so after they retire.

Finally, gender is a demographic factor of undoubted significance. As mentioned, women have almost achieved parity with men in terms of pension coverage. However, the same cannot be said for pension adequacy. Parity in coverage is a formidable achievement, largely attributable to the arrival of large numbers of women in the public sector where pensions are widely available. However, the adequacy of pensions for female retirees raises three issues of concern. First, like immigrants who arrive in Ontario mid-career, women who take time out from paid employment for child-rearing and elder care will have fewer years of pensionable service. Second, like immigrants and members of visible minorities who experience discrimination at work, women often earn less than their male counterparts, a fact that diminishes their pensionable earnings. And third, women retirees can be expected to live longer than their male colleagues. If they do, they will likely experience more years during which their pensions will be eroded by inflation. Often beginning as they do with lower pensions, attributable to lower wages and fewer years of service, women are therefore likely to experience greater financial hardship in retirement.

In general, then, demographic factors provide several new insights into implications of the well-documented decline in occupational pension coverage. Increasing longevity will generate increasing pressures on pension costs. As immigrants and members of visible minorities in Ontario’s population join the labour force in ever-increasing numbers, pension coverage is likely to fall rather than rise. And coverage per se does not tell the whole story of decline in the occupational pension system: the level of pension benefits received by women, members of visible minorities and recent immigrant workers is likely to be lower than that of their white, male, native-born Canadian predecessors and colleagues.

3.3.3 The impact of regulation on pension coverage

Many sponsor representatives argued in their briefs and oral submissions to the Commission that the decline in pension coverage is directly or indirectly attributable to over-regulation.

They cited many significant examples: funding rules that force sponsors to make large, unpredictable and unsustainable contributions to maintain plan solvency without the prospect of swiftly recovering occasional surpluses as an offset; regulatory constraints on innovation in investments, plan design and administration; the need to comply with different regulatory requirements across the country; the imposition of professional standards such as “mark-to-market” accounting rules, which expose pension plans to additional pressures; and the application of trust law doctrines, which impose on sponsors, trustees and service providers tighter restrictions and greater responsibilities than were anticipated when the plans were first established. Added to these examples was a litany of complaints dealing with procedural or structural irritants: unnecessary reporting and procedural requirements; inefficient or unreasonable procedures adopted by the pension regulator; difficulties in dealing with lost beneficiaries and disputing spouses; and the complicating effects of pension plans on corporate financial reporting and restructuring.

The argument comes to this: employers have capped, closed, converted or declined to establish DB plans because the increasing burden of regulation has driven up the cost and trouble of maintaining them. And many more will do so — it is predicted — because regulatory burdens are becoming all the more insupportable in the context of difficult business conditions, an adverse investment climate, and growing costs associated with increasing longevity and other developments.

Given the frequency and force with which this argument was advanced, I take it very seriously indeed. However, the evidence does not seem to support over-regulation as a prime cause of declining pension coverage. The downward trend in coverage — evident in the mid-1970s and well-documented from 1985 onward — predates the introduction of most of the regulatory requirements complained of, and has not accelerated during periods of heightened regulatory or legal change. Moreover, the decline in pension coverage coincides with and is better explained by other factors — principally, the decline in union density and in employment in sectors where DB plans were most prevalent.

But that does not end the matter. On the one hand, legitimate complaints about regulatory laws or practices deserve to be addressed on their merits. On the other, if employers come to believe that the costs of providing pensions are too high, and if some of those costs are associated with the burdens of regulation, it would be prudent to do whatever can be done to ease those burdens. Finally, it is clear that changes in regulation should neither leave plan members without adequate protection nor expose communities to the financial and social costs that flow from plan failure.

This brings me to the argument advanced by those who contend that under-regulation has contributed to the decline in pension coverage. There are two senses in which this might conceivably be true. The first is that more vigorous and proactive oversight of the solvency of pension plans might avoid the need for expensive and, arguably, excessive funding requirements. The second is that past regulatory failures — especially those associated with the insolvency of sponsoring employers — have eroded confidence in the pension system. Better — smarter, more effective and cheaper — regulation might restore confidence and lead to higher coverage rates. However, for the same reasons as I was unconvinced by the argument that over-regulation caused the decline in coverage, I remain unpersuaded by the argument that under-regulation has done so.

Finally, the effects of the Income Tax Act (ITA) on pension coverage warrant comment. Initially, this legislation favoured the establishment of DB pensions by allowing sponsors to treat contributions to “registered” pension plans as a business expense, and by allowing workers to defer the payment of tax on this part of their compensation package until they retired and actually drew their pensions. Moreover, workers were permitted to save and shelter more income in a DB pension plan than in a DC plan or an RRSP. However, 1991 amendments to the ITA “leveled the playing field” by providing equal treatment for all three retirement savings vehicles. A study undertaken for the Commission by Jinyan Li concluded that this change in tax policy likely explains the rapid expansion of these alternative approaches to retirement savings. However it is less clear that this expansion occurred at the expense of the DB system, although no doubt some sponsors and some workers — especially younger workers — prefer DC plans and group RRSPs, which typically involve lower contribution levels.

The ITA has also perhaps exerted a drag on pension coverage by too narrowly defining which plans are eligible for registration, and hence for favourable tax treatment. These constraints appear to inhibit innovation in plan design. For example, cash balance plans, which attract extensive enrolment in the United States, cannot be established in Canada because they are not contemplated by the ITA. Or, to take another example, only pension plans whose members are “employees” are eligible for registration. Hence, plans that might recruit members based on some other affinity relationship — say, operators of restaurant or convenience store franchises — are ineligible. Indeed, even self-employed workers who are deemed to be “employees” under provincial labour legislation have been denied eligibility for coverage under the ITA, whose definition of “employee” is more limited. Given recent extensive changes in Canadian labour markets, with fewer and fewer people working under conventional, long-term contracts, these ITA constraints on plan design and membership should be reviewed at the earliest possible moment.

A final effect of the ITA has to do with the funding strategies and investment policies of DB plans. Under the ITA, sponsors are permitted to shelter their annual contributions only to a maximum of 110%, or 120% of plan liabilities (depending on some plan features and circumstances). In consequence, sponsors cannot over-contribute in good times, in order to under-contribute when times are difficult. It is quite possible that these provisions — especially in combination with the legal rules regarding surplus withdrawals — represent a deterrent for employers wishing to initiate or continue plans, and in that respect, they contribute to the decline of DB coverage. However, their more obvious effect is that they hinder the adequate funding of plans already in existence. Similar effects can plausibly be attributed to the federal investment rules, which constrain the scope of investments undertaken by registered pension plans. These concerns are dealt with in subsequent chapters.

3.3.4 Attitudinal shifts

Ontario’s occupational pension system ultimately rests, for better or worse, on the voluntarism principle mentioned in Chapter One, and thus on the decision of individual employers to establish or not establish a pension plan. That decision is likely to be influenced by the opinions of many different people: corporate executives such as Chief Financial Officers and Directors of Human Resources; union officials and staff members; professional advisors such as actuaries, accountants and lawyers; and insurance companies, bankers and investment counsellors. The more such people develop hostile attitudes to occupational pensions, the less likely it is that new plans will be established or old ones continued, and vice versa. Or to put the matter another way, shifts in the relative influence or power of those who favour or disfavour pensions may tilt the balance for or against them.

Of course attitudes are shaped by new facts, new perceptions of facts and new ways of analysing facts (not always the same thing), and sometimes they are shaped by new values and beliefs, including beliefs about what “relevant others” believe. It is entirely plausible that such shaping influences have indeed caused the attitudinal shifts that observers have detected with regard to occupational pensions. Some examples:

  • If surveys of service providers show that most expect DB plans to decline, they are likely to become a self-fulfilling prophecy.
  • If younger union members increasingly feel that they are better able to manage their retirement savings than the administrator of their pension plan, the priority accorded pension coverage may alter when union officials sit down to negotiate a new collective agreement.
  • If policy makers become convinced that DB pensions inhibit labour market mobility, they are unlikely to support initiatives to promote such pensions.
  • If a dominant view emerges among sponsor-side pension professionals that the DB system is in terminal decline, most will “cuddle up to consensus,” and few will advise clients to start a new plan for fear of the consequences if their advice turns out to be wrong; in fact, some will advise clients to avoid buying into companies that have such plans.
  • If from one perspective, having a pension plan is seen as a valued part of a successful human resources strategy, and from another as preventing optimal deployment of corporate capital, which of these two perspectives prevails will likely depend on whether pension policy is made by the Director of Human Resources or the Chief Financial Officer.

These are not hypothetical “ifs:” they are anecdotes gleaned from my encounters with pension stakeholders and professionals, which reveal increasingly negative attitudes about DB pensions in particular. These attitudes are being expressed not just at public hearings and in publications and speeches, but in union meetings and corporate offices, and in professional and policy-making circles. Cumulatively, they have almost certainly had an adverse effect on coverage.

What, if anything, can be done to change these attitudes?

To the extent that they represent reactions to real defects in the system, the best way to change them is to remedy those defects, and to make it clear that this has been done. To the extent that they are the result of the human tendency to see things in light of one’s own interests or through the optic of one’s own professional, political or theoretical stance, people with different beliefs, interests and perspectives must be brought together to exchange views concerning possible solutions, or at least to understand that there are many ways of looking at the pension system. And to the extent that they rest on misperceptions or inadequate information, the best way to change such attitudes is to put facts into the public domain. The Commission has attempted to do just that through its public hearings, stakeholder meetings and, especially, its research program.

To sum up, attitudes to occupational pensions will become more favourable and more supportive of increased coverage only if stakeholders and their advisors and advocates are provided with accurate and plausible information and given the opportunity to exchange views, and if ongoing efforts are made to respond to their legitimate concerns.

3.3.5 Inherent structural flaws and “the perfect storm”

A final — and, in some ways, more fundamental — explanation for the decline in coverage rates is that the current DB system was badly designed from its inception. Proponents of this view hold that its architects made erroneous assumptions about the system’s infinite sustainability. The key error, if I understand the critique, was undue optimism: the economy would continue to expand indefinitely, ever-rising equity prices and dependably high long-term interest rates would generate the resources needed to keep plans properly funded with relatively low contributions, and post-retirement longevity would remain constant.

The ups and downs of Ontario’s economy from the 1970s through the 1990s posed many challenges to pension plans, but also provided periodic opportunities for plans to accumulate large surpluses, and for employers to take contribution holidays and to finance benefit improvements. During this period, moreover, plans came increasingly to depend on equity investments, which produced relatively high returns, albeit with greater risk of periodic reversals. Furthermore, some plans adopted practices and policies that effectively masked the potential for a serious crisis in funding. While these strategies managed for some time to conceal the underlying flaws in the system, by the end of the 1990s the consequences of excessive optimism, the falsity of the underlying assumptions, and the masking effect of ill-considered funding strategies underlying DB plans started to become clear. Two developments, so the argument runs, made them glaringly evident. One was the maturing of many DB plans, with rising numbers of long-living retirees and dwindling numbers of active members. The other was the financial crisis of the early years of the new century when equity prices fell sharply and long-term interest rates declined from highs in the late 1980s to much more modest levels. These developments, of course, occurred against the background of perturbations in the economy and increasing concern for transparency and accountability.

The combination of these and other factors created what is called in the pension community “the perfect storm.” In that “perfect storm,” numerous plans foundered or feared that they might. As discount rates fell, previously affordable liabilities became much more expensive; sponsors — many of whom had recently taken contribution holidays or granted enhanced benefits — were having to make much higher contributions than they had anticipated, and in some cases, sponsors themselves veered toward insolvency.

Canada and Ontario escaped the worst of the “perfect storm,” which caused serious under-funding of pension plans in the United Kingdom and the United States, for three reasons. First, Canadian pension funds were not as heavily invested in equities as funds in some other jurisdictions; second, Canadian equity values increased considerably after 2002; and third, Ontario and other Canadian jurisdictions generally have more robust funding rules, as was pointed out by Colin Pugh in research for the Commission. Nevertheless, an air of crisis developed around the DB system, which — for some informed observers — persists to this day.

On the one hand, this far-ranging and fundamental critique of the DB system is very valuable, as it reminds us that not all problems are amenable to a “quick fix” and that the basic architecture and assumptions of the system may require scrutiny. To a limited extent that scrutiny is initiated in the concluding chapters of this report, but much more theoretical and empirical work remains to be done. On the other hand, the “fundamental flaws plus perfect storm” hypothesis may prove too much. It is tempting (and consistent with recent evidence) to suggest that the DB pension system may well be one of those institutions that is inoperable in theory, but works reasonably well in practice. Whichever is the proper characterization of this hypothesis, however, those who are persuaded of its soundness must surely be tempted to defect from the DB pension system at the first opportunity, if they have not already done so.

3.3.6 Why pension coverage has declined: Some guarded conclusions

The strongest empirical evidence available to the Commission suggests that labour market changes — especially diminished union density and decreasing employment in sectors where DB pensions were historically most common — have caused the largest part of the decline. If they were the only cause, no change I might recommend within the pension system itself would likely do much to improve coverage. However, large, long-term trends seldom have a single cause. The other factors explored in this chapter — demographic developments, regulatory issues, attitudinal change, and a firm view in some quarters that the design of the DB system is inherently flawed — have all likely made some contribution to the decline of the DB system. At the very least, the fact that people believe that they have done so has placed the system under stress. In consequence, however meagre the evidence to support them, none of the factors can be ignored in any attempt to account for the decline.

Back to top

3.4 Possible Consequences of a Decline in Coverage under Ontario’s Occupational Pension System

To recapitulate the themes developed in this chapter and the preceding one, the existing occupational pension system is not perfect. It has never covered more than half of Ontario’s workers, now covers only some 34.7% of them, and is at risk of covering an even lower percentage regardless of any improvements that I might recommend to the present system. Further, coverage is spread unevenly between the public and private sectors and, within the latter, among workers in different economic sectors, in enterprises of different sizes, and across class, ethnic and racial (but not gender) lines. Many of the workers who will most need pensions have little or no access to them. And benefit levels for some retirees (notably women and recent immigrants) are lower than for others.

But despite the shortcomings of Ontario’s occupational pension system, its decline should provoke great concern. Even in its present state, the pension system continues to ensure that millions of Ontarians will enjoy a decent retirement income, which will reduce or eliminate their reliance on needs-based public pensions, and enable them to buy goods and services from others. Moreover, the current system provides a convenient vehicle to encourage savings by workers who otherwise might be tempted to spend what they earn, and to ensure that their collective savings play an important part in sustaining Ontario’s economy and infrastructure.

What consequences will ensue if the decline in pension coverage continues? Some of those consequences were described in Chapter Two. To recapitulate: we can anticipate an increase in the number of older Ontarians living without income from occupational pension plans, declining markets for goods and services purchased by seniors, declining tax revenues and increasing public welfare costs, and a possible decrease in the large sums presently invested by pension funds in Canada’s and Ontario’s capital markets. Moreover, once overall pension enrolment drops below a certain level, the occupational pension system as a whole will become increasingly difficult to sustain; this is likely to happen sooner rather than later if enrolment in the public and private sectors continues to diverge so considerably.

Since these are all consequences Ontarians wish to avoid, if possible, it follows that public policy should attempt to arrest or reverse the decline as well as disparities in coverage and benefits.

This diagnosis of the decline of DB plans explains much of what follows later in this report by way of prescription. In Chapters Four through Seven, I recommend measures by which sponsors’ concerns, as well as the concerns of active and retired members, might be addressed. In Chapters Eight and Nine, I canvass a number of innovations in plan design, funding and governance that should help to arrest the decline in coverage, that may provide more workers with at least the functional equivalent of classic DB pensions, and that will hopefully provide new platforms for expanded coverage. In Chapter Ten, I suggest how government can respond in a more orderly and informed way to the challenges and opportunities that confront the pension system.

I conclude with a paradox: coverage should not be achieved by impairing the security of the pension promise for workers and retirees, but security cannot realistically be purchased if sponsors are required to pay too high a price for it. In short, a delicate balance has to be struck among policy goals that are all desirable but not always easily reconciled.

Back to top