The new retirement math

When people talk about retirement, the conversation generally drifts toward how much they have saved or think they need to save. Unfortunately, they often miss an important point — what happens when they actually retire. With their conversations focused on savings, I would wager not many are thinking about how much of their savings they will need to live on each year of their retirement.

I spend a lot of time speaking with investors across Canada about retirement. In these conversations, I ask them how much of their current income they think they will need to live the life they want in retirement. For the most part, they say somewhere in the range of 60 to 70 per cent of their current income (so if they are making $100,000 today, they think $60,000 to $70,000 should let them live “comfortably”). When I ask them what those numbers are based on, they invariably answer, “That is what everyone seems to suggest” and leave it at that.

Since almost everyone in the retirement planning field does suggest a retirement income replacement rate of 60 to 70 per cent, my team and I assumed there would be some solid research behind it. We were more than a little surprised to discover that there is actually very little. This dearth struck us as especially odd, given the importance — and widespread application — of the 60 to 70 per cent figure to many company pension plans as well as individual retirement plans.

We know that retirement is changing in Canada and around the world. People are retiring earlier, living longer and going about retirement in very different ways than in generations past. We also know that retirees do not necessarily reduce their consumption once they stop working, as was once assumed. In fact, data from Statistics Canada show that consumption in retirement declines much less rapidly than incomes and that it declines very little, if at all, in the early years of retirement. In addition, anecdotal evidence suggests that people are becoming more and more active in retirement, something that comes up increasingly in the conversations we have with advisors. Some even tell us they have clients whose retirement plans involve increased consumption.

If retirees are not reducing their consumption in retirement — or at least not reducing it very much — then the conventional wisdom of retirement income replacement is doubtful. That led us to another interesting question: should the retirement replacement rate dictate the level of consumption in retirement, or should the level of consumption in retirement dictate the retirement replacement rate? Or to put it another way, should retirees choose their consumption rate or have that choice made for them by financial constraints?

Put in those terms, our research worked to determine what replacement rate is necessary to allow retirees to maintain the same level of consumption they had before retirement. This doesn’t mean they won’t change their spending patterns — i.e., purchase different things in retirement than when they were working — but it will show what they need to live the retirement they want.

The methodology we used divided pre-retirement spending into four components: (1) personal income taxes, (2) contributions to public programs (such as Employment Insurance and Canada Pension Plan), (3) contributions to retirement savings and (4) consumption.

In retirement, we eliminate contributions to public programs and retirement savings, maintain consumption at the pre-retirement level, factor in age- and pension-related tax credits, and calculate the income necessary to cover consumption and taxes. The actual exercise calculated replacement rates at various levels of pre-retirement income, for singles vs. couples, for single vs. two-income earners, etc., and was much more complicated than I have described here.

The end result challenges conventional wisdom. Our research shows that if retirees want to maintain their lifestyle, they now need to replace 75 to 85 per cent of their pre-retirement income. Specific replacement rates depend on a number of variables, such as whether they are being calculated for a single person or a couple, how many income earners there are in the family, the levels of pre-retirement income and retirement savings, and the extent to which specific tax strategies are used.

Our research is igniting some lively discussion among analysts and advisors. It is also likely to cause some serious rethinking of retirement plans. We know that not everyone will use these new replacement rates. Nevertheless, we strongly believe they provide a new and useful benchmark for retirement planning.