The idea of expanding the Canada Pension Plan (CPP) resurfaced in December 2012 during a meeting of the federal and provincial finance ministers. The ministers agreed to explore possible reforms to the CPP at their next meeting expected sometime in the coming months.

The proposal getting the most attention is a mandatory expansion of the CPP. Such a change would require working Canadians to contribute more of their income today via payroll taxes with the promise of greater benefits in retirement. Unfortunately, the debate thus far has ignored a critical economic insight that leads proponents of an expanded CPP to overestimate the increase in overall retirement savings and therefore the benefits of this policy reform.

Economic theory tells us that higher forced savings for retirement through the CPP will lead Canadians to reduce their voluntary savings elsewhere. That means an expanded CPP would not increase overall retirement savings to the extent expected but it would change the mix with more going to CPP and less to other savings like Registered Retirement Savings Plans (RRSPs).

The economic framework for analyzing how people consume and save traces back to the work of Nobel economist Milton Friedman and others. Friedman’s work concluded that people choose how much to spend (consume) and save based on their expectations of income over the course of their lifetime. If people’s preferences for spending vs. savings don’t change, and if they don’t expect more income over their lifetime, they will simply offset increased government-mandated savings with less voluntary savings, leaving the overall amount saved largely unchanged.

Thankfully, Canadians don’t have to rely solely on theory since we have a natural experiment with mandatory increases to CPP contributions. Specifically, the CPP payroll tax increased to 9.9% in 2003 from 5.0% in 1993. Those increases give us an opportunity to observe the actual response of Canadians who save voluntarily in RRSPs.

Our recent study looked at CPP and RRSP contributions for two age groups: Canadians under 45 and aged 45 to 65. It further separated each age group into two income groups: $10,000 to $50,000 and $50,000 to $100,000. We particularly focused on the 45-65 age group making between $10,000 to $50,000 since this group is likely the most sensitive to changes in the CPP because of their age and income.

Using three different measures, our analysis consistently found that RRSP contributions declined as mandatory savings to the CPP increased.

For instance, the percentage of tax-filers aged 45 to 65 with income between $10,000 and $50,000 contributing to RRSPs declined between 1993 and 2003. Specifically, 40.2% of tax-filers in this group contributed to RRSPs in 1993 and the proportion fell to 33.0% by 2003.

We found similar results when examining the share of income contributed to RRSPs (see the figure elsewhere on this page). Again, for Canadians aged 45 to 65 with income between $10,000 and $50,000, the share of income contributed to RRSPs declined to 3.5% in 2003 from 4.4% in 1993. Meanwhile, the share contributed to CPP doubled to 3.0% from 1.5% of income as the CPP payroll tax was raised.

A third measure showed the dollar value of RRSP contributions per tax-filer also decreased as mandatory CPP contributions increased. Taken together, our findings strongly suggest a substitution between CPP and RRSPs occurred in the past when mandatory CPP contributions increased, as basic economic theory would predict.

The debate about the benefits of increasing the CPP contribution rate for all workers would then, at a minimum, account for the costs of reduced RRSP savings, which include a loss of flexibility and choice.

With RRSPs, the assets accumulated over time can be fully transferred to a beneficiary upon death (the CPP only offers scaled back benefits to survivors). Moreover, if you're young and interested in buying a house, RRSPs through the Home Buyers' Plan can help by allowing penalty and tax free withdrawals up to $25,000. Similarly, if you're middle-aged and looking to transition to a new field of work, the Lifelong Learning Plan allows you to withdraw RRSP savings up to $10,000 per year penalty and tax free. Finally, if you have a terminal illness or need emergency funds, you can use RRSP savings.

These benefits are lost when Canadians are forced to save more in CPP and then offset those increases with decreases in their RRSPs. Other aspects of this trade-off, such as the comparative benefits of the CPP (defined benefit in retirement) compared to the benefits of RRSPs (flexibility and choice), also need to be assessed and discussed.

The key to providing retirement income through savings is a set of rules that allows for an optimal mix of savings for different people in different stages of life and with different preferences. There may be benefits to a compulsory expansion of the CPP, but these benefits need to be weighed against the costs, which as our analysis shows could include a reduction in voluntary RRSP savings.