Plans to expand government pensions based on faulty assumptions

With talks to expand the Canada Pension Plan having stalled, the Ontario government has pledged to roll out its own provincial version. The impulse for a ‘big CPP’ hinges on the assumption that Canadians are too ignorant or misguided to plan for retirement themselves and the meddling hand of government bureaucrats can help them. In a new study published by the Fraser Institute, I question many of the assumptions behind the drive for ‘big CPP’ in Ontario and elsewhere.

To begin, Canada’s current cohort of retirees aren’t facing a retirement income crisis. People are living longer, healthier and wealthier lives in retirement. The few pockets of poverty among seniors, such as single elderly women who have never worked, are best addressed by better targeting government benefits, not a wholesale expansion of the CPP.

The crisis publicized by the pension ‘industry’ resides in the future and relies on assumptions and projections in models which are questionable. The model results are based on the traditional three pillars of Canada’s pension system—social security payments from government, the mandatory CPP, and voluntary pensions like RRSPs. This downplays the role of assets people hold in a fourth pillar outside the pension system, which total $8.6 trillion including real estate and various saving and investments, compared with $2.6 trillion held inside the pension system. And it completely ignores a largely undocumented but vital fifth pillar of support to retirees from family and friends; for example, 10 per cent of seniors live with their families, with unknown amounts of money and in-kind support flowing back and forth across generations, including inheritances.

There are many problems with model-based extrapolations of pension incomes years in the future. For example, models assume that the replacement rate of working income with pension income is fixed over time, when it more likely declines as older age curtails spending on travel and entertainment. Banks routinely exhort retirees to replace 70 per cent or more of their working income, when some experts find 50 per cent would be adequate for most. But the fundamental problem with targeting replacement rates is that they are an opinion, not an observable fact. A prospective retiree can rationally choose to retire early, accepting a lower standard of living to spend more time pursuing leisure activities or with family.

Another challenge with model-based projections results from a growing number of older Canadians staying in the labour force. Nearly half of Canadians over the age of 55 are still in the labour force, including one quarter aged 65 to 69 (a near doubling of the rate over a short period). Canadians are increasingly working past what used to be the traditional (and often mandatory) age of retirement and this shift is playing havoc with forecasts of the labour force. This should give pause to anyone basing policy prescriptions that increase payroll taxes for virtually all working Canadians today on model simulations of the distant future. Every extra year elderly Canadians spend working generates more income and reduces the time savings are withdrawn for retirement.

However, a major problem with using models to simulate the future of retirement is the underlying assumption that prospective retirees don’t understand their financial circumstances. In models, Canadians march towards retirement either blissfully unaware of the lower standard of living waiting for them or utterly incapable of altering their behaviour by saving more or working longer in response to that knowledge.

In real life, there’s ample evidence that Canadians alter their behaviour in response to a keen awareness of their circumstances and act decisively and rationally to control them. Some accept lower incomes in order to retire early while others work longer when circumstances dictate; they save less voluntarily when government increases mandatory saving; they save more in their own pension accounts when employer-based pension plan benefits erode; they elect to receive C/QPP benefits earlier or later than the traditional 65 years as they see fit; they shift consumption between the early and later stages of retirement; they save more in their later years to leave an inheritance; and they understand government will provide support as their health deteriorates in their final years.

Canadians are anything but the robotic automatons portrayed in models, doomed to endlessly repeat past patterns of behaviour, incapable of learning and adapting their lifestyle to the changing world around them. They are actively involved in making the myriad of decisions that affect their pensions and their retirement. If there is an expanded role for government to play in the future retirement system, it’s filling in the few cracks through which pensioners can fall into poverty.