Still More on the Pensions Debate
Last week I attended what turned out to be an excellent pensions conference organized by the Institute for Research on Public Policy (IRPP). It was a behind closed doors, Chatham House rules event which stops me from blogging too freely, but several of the presentations have now been posted to the IRPP web site. The audience was mainly policy wonks, including several senior government officials working on the pensions policy file. As at the Calgary conference,Â the option of expanding public pensions to deal with the looming pensions crisis was very much in play.
The presentation by Keith Horner makes for interesting reading. He is a retired Department of Finance official who wrote a major research paper on income adequacy in retirement for the Mintz Report to the federal government and the provinces. (My earlier post on the Mintz Report is here:
Horner spoke favourably to the option of expanding the Canada Pension Plan,Â outlining his own specific proposal to increase the CPP replacement rate. Beginning from his research showing that 25-30% of middle-income earners are not saving enough, he suggests that a mandatory national defined benefit plan would be a better option that a national defined contribution plan, let alone the current default option of retail RRSPs for those who do not belong to pension plans.
Grant Schellenberg of StatsCan provided additional information from his earlier study, somewhat nuancing the message conveyed earlier. (MyÂ post on his report looking at the post retirement incomes of those belonging to and not belonging to pensions plansÂ while working is here
This presentation shows that the median income replacement rate for men who did not belong to pension plans while working is 7-9% lower than for those who did belong to plans, suggesting that additional private saving did not fully close the gap. It is interesting to note very high rates of pension coverage in hus sample above the bottom quintile of earners, underlining my earlier point that the income situation of the current retired who worked in a different era tells us rather little about the prospects for future retirees.
Frank Swedlove of the Canadian Life and Health Insurance Association spoke on a final panel on policy options, and spoke to the possibility of requiring all employers with more than 20 employees to support some kind of pension vehicle with auto enrollment of workers. That is far from my preferred option, but it does suggest that parts of the financial industry are very aware of the high cost and modest and uncertain returns of retail RRSPsÂ comapred to large pooled vehicles with fiduciary obligations to plan members. At any rate, it is a good debating point when the “nanny state” argument comes up. The insurance industry and the labour movement agree that voluntaristic, individual vehicles have failed…
Notes for my own presentation follow:
The Canadian Labour Congress proposes to phase in a doubling of Canada Pension Plan benefits from 25% to 50% of average earnings on a pre-funded basis; and to increase the Guaranteed Income Supplement by 15%.
1. Workers deserve a secure and decent income in retirement
Our retirement income system should give workers an income they can count on through the years they spend after leaving paid work. Most experts believe this means that they should get 60 to 70% of their previous earnings for life, fully indexed to inflation.
Public pensions â€” the Canada Pension Plan (CPP) and Old Age Security (OAS) â€” provide secure pensions which are fully indexed to inflation. All citizens (except recent immigrants) get OAS, paid for from taxes, and all workers (including self-employed workers) contribute to the CPP over their working lives. The CPP provides a fully portable, secure, inflation-indexed defined benefit pension at very low cost.
But public pensions provide only a bare-bones pension â€” a maximum of about 40% of previous earnings in theory, and about one-third on average. The CPP replaces a maximum of just 25% of insured earnings for those earning at or below maximum pensionable earnings of $47,200. That means that most workers must rely on the private part of our pension system â€” employer pension plans and RRSPs â€” for a big chunk of their retirement income.
Experience has shown that private pensions provide, at best, very uncertain returns. In the recent financial crisis, many RRSPs lost a huge chunk of their value, and many employers failed to deliver on pension promises that workers had paid for, and thought that they could count on.
2. Public pensions are far too low
Our system of public pensions works well, but replaces only a very modest proportion of previous earnings compared to most European countries, and even the United States.
An average worker will see only about one-third of their previous earnings replaced by public pensions. Old Age Security pays $516.96 per month, and the maximum CPP benefit is $934.17 per month, for a maximum of $1,451 per month or $17,414 per year.
But the average CPP benefit is just $472 per month, only about half of the maximum benefit. That is because only workers who have earned at or above maximum pensionable earnings (now $47,200 per year) for almost their entire working lives will qualify for the maximum. OAS plus the average CPP benefit comes to only $989 per month or less than $12,000 per year.
Seniors with very low incomes qualify for a public pension top-up through the Guaranteed Income Supplement. But the total amount falls short of raising them to a decent standard of living. The maximum amount of OAS plus GIS is only $1,169 for a single person, or $14,034 per year. That is why the CLC has called for a 15% increase in the GIS.
For better paid workers, public pensions replace far too low a proportion of earnings.
3. Our workplace pensions need support
Most unionized workers belong to employer-sponsored pension plans. But two-thirds of all workers, and 80% of private sector workers, do not belong to such a plan, and must save through high cost, very insecure, and unpredictable RRSPs. An expanded CPP is the best pension reform option for them since it is less costly and more secure.
There are major problems with our pension system even for workers who belong to employer pension plans. Many plans now provide a pension based on uncertain and fluctuating investment returns, not a defined benefit like the CPP. Most plans are not indexed to inflation, so retirement incomes fall a lot after retirement. In tough economic times, some employers will fail to pay the pensions they promised and that workers paid for. Small pension plans are much more costly to run than the CPP and other big pension plans.
Another big problem is that employer pensions work best for workers who stay with the same employer all of their working life. Workers who lose their job, or change jobs, usually lose out.
The great majority of employer pension plans are integrated with the CPP, meaning that the CPP benefit is taken into account when calculating how much the employer plan must pay to deliver a defined benefit. That means that the CLC proposal to double CPP benefits would gradually take the financial burden off employer pension plans. But our plan would still leave space for good pension plans to replace earnings above an average amount, and to pay for provisions such as early retirement.
4. RRSPs donâ€™t work
Workers forced to rely on RRSPs to fund their retirement save far too little, and pay far too much for very uncertain returns.
In big pension plans like the CPP, there is no danger of outliving a pension since the risk is pooled across large numbers of people. But people saving on their own may end up with very little retirement income. It is extremely expensive to convert RRSP savings into an annuity that will pay a guaranteed benefit indexed to inflation for life, something which the CPP does very effectively at low cost.
All of the evidence shows that people donâ€™t save enough in RRSPs. Only about 3% of income is saved through RRSPs each year, far below what is allowed, and there is now more than $500 billion of unused contribution room. Less than one in three workers pay into an RRSP each year, and the majority of those contributors earn more than $80,000 per year.
Average RRSP savings of an older worker nearing retirement are only about $60,000 â€” enough to pay for an annuity of less than $300 per month.
As was recognized in The Mintz Report on pensions for the federal government, RRSPs are much more costly than big pension plans. Most mutual funds have management expense ratios of between 2% and 3%, much higher than the CPP management cost of less than one-half of one perÂ cent. The greater rate of return to saving through the CPP of at least 2% per year makes a huge difference to how much pension can be earned over a working lifetime.
5. Too many seniors have low incomes
The Guaranteed Income Supplement (GIS) paid out on top of Old Age Security provides a guaranteed annual income to seniors, but at a very modest level of just over $14,000 for singles, and $23,000 for couples. That is not enough to meet basic needs, especially for singles in large cities.
Over one in three seniors (36%) qualify for the GIS, showing that the pension system is failing to provide adequate incomes to too many Canadians.
Claim 1: Canadaâ€™s pension system is working well
The Mintz Report to the federal government argued that our pension system is working well because a large majority of todayâ€™s seniors have incomes of 60% or more of their previous earnings.
The main flaw with this argument is that the situation of todayâ€™s more fortunate seniors tells us very little about the prospects for younger and middle-aged workers who are still many years from retirement.
Compared to todayâ€™s seniors who were often working 20 years and more ago, todayâ€™s younger workers are far less likely to belong to an employer pension plan. They are likely to find secure and well-paid jobs, if at all, much later in life. And they are likely to have much higher levels of student and mortgage debt than did earlier generations. In fact, Canadian households today owe 150% of their income in debt, an all-time high.
The CLC proposal to phase in a doubling of CPP benefits from 25% to 50% of average earnings will provide the greatest benefit to those who are many years from retirement. Changing times demand that we shift the balance from private savings to better public pensions.
Claim 2: Canadaâ€™s poverty rate for seniors is one of the lowest in the world
This claim, taken from an OECD report, is misleading on two counts.
First, it misses the fact that seniors in many European countries benefit from significant non-cash support from their governments, such as affordable housing, low cost home and residential care, cheap transit and very low cost drugs. Canadaâ€™s seniors have to meet many of these needs and more from their own pockets.
Second, while it is true that the poverty rate among Canadian seniors is relatively low, this is because poverty lines are set at low levels of just $12,000 to $15,000 for a single person, and $14,000 to $23,000 for couples. Even measured by these very low lines, one in seven single seniors live in poverty. And many seniors have incomes just above the poverty line.
If poverty was defined (as in many European countries) as having an income of less than 60% of a median household of the same size, the poverty rate for single seniors would be 20%, and it would be 15% for senior couples.
Claim 3: Increasing CPP means a costly tax increase for Canadian workers
The CLC plan to double CPP benefits over time would require, according to the former Chief Actuary of the plan, a premium increase of just under 6% of pensionable earnings, divided equally between workers and employers. (Doubling of benefits does not require a doubling of premiums since the current premium is set to cover a prior period of under-funding.) The proposal is to phase in this modest premium increase over a number of years.
This is an extremely modest increase in premiums from the current level of 9.9% divided between employers and workers. For an extra contribution of less than 3% of earnings, a younger worker earning the average wage would get an increased pension benefit of $934.17 per month, fully indexed to inflation. This is a far greater benefit than would be earned by making the same contribution to an RRSP.
CPP premiums are not taxes but premiums paid in return for a defined benefit at low cost.
The CPP premium increase would not affect the great majority of employers who now sponsor pension plans (since these are integrated with the CPP), and it would not affect the many responsible employers who help workers contribute to RRSPs. In fact, many of these employers support an expanded CPP since it would deliver better pensions at lower cost.
Claim 4: The CLC plan for CPP wonâ€™t benefit most workers
The full benefit of the CLCâ€™s plan for CPP would go to younger workers who spend their working lives in the new, improved system. But every worker will earn a doubled benefit for each year they are in the new system.
A worker aged about 45 today would not get the full increase from 25% to 50% of pensionable earnings, but they would still get a significant increase from 25% to 37%, worth over $450 per month.
Claim 5: Individuals should take responsibility for themselves
Some people have attacked expanding CPP and the GIS on the grounds that individuals should fend for themselves rather than be covered by â€œcompulsoryâ€ plans.
The reality is that pooling investment funds and risks produces much greater benefits, at much lower cost. That is why very few employer pension plans offer individuals the choice of belonging. And very few workers would turn down the offer of a good job, offering a decent pension, just because paying into the plan was â€œcompulsory.â€
In the final analysis, our society will not tolerate high rates of poverty among seniors. A bigger CPP and better pension incomes will result in significant savings to governments down the road.
Claim 6: A bigger CPP would hit financial industry profits
True. The end of high cost RRSPs and many inefficient, defined contribution pension plans would greatly reduce the huge profits earned by the financial sector from the pension savings of Canadian workers. Opposition by the banks and insurance companies is self-serving, but entirely understandable.
Thanks for the IRPP link. Note that the internal link at the site to Bob Baldwin’s presentation is broken: the correct URL is http://www.irpp.org/events/archive/20100504/baldwin.pdf
Meanwhile, a good place to test multi-employer supplemental pension reform would be at Ontario universities. At present, almost every Ontario university has it own little pension fund and plan. By merging these funds and selecting the best plans as models for a new province-wide pension plan for all university employees, Ontario could provide better pensions to more people at a much lower cost. Instead, every university in the “University of Ontario” system must struggle with its own pension solvency deficit, its own administative costs and its own risks. We need one DB pension plan for all Ontario university employees.
More on Rob Brown at
It remains to be seen what “modest, phased-in, and fully funded” means in this context.
Jon Kesselman has a paper comparing the different proposals to expand public pensions at
He comes out strongly in favour of an expanded mandatory plan, rather than a voluntary approach. Although there are parts of the paper that I disagree with, it’s pretty comprehensive.