Everyone has heard the statistic that replacing 70 per cent of pre-retirement earnings will lead to retirement income adequacy. But where did that magic number come from? And is it actually a successful benchmark in practice and not just in theory?
At the 2016 DC Plan Summit, Bonnie-Jeanne MacDonald, an actuary and academic researcher at Dalhousie University, challenged those who use the 70-per-cent replacement rate assumption when determining whether people will have enough in their later years.
She noted financial planners, policy-makers and pension plans have been using that as a benchmark for a long time but questioned whether its accuracy has been tested. “What is retirement income adequacy? That…benchmark is really critical. If it’s not valid, then all that work you’ve done just kind of goes out the window. Not only would the results be wrong, they’d be misleading.”
So would a worker with a 70-per-cent replacement rate actually maintain the same standard of living before and after retirement?
“There has never been a clear demonstration that the conventional final employment earnings replacement rate is actually valid,” said MacDonald.
“No one’s ever actually looked at population data after the fact and said, ‘Hey, is this thing actually working? Will workers who hit this target actually maintain their standard of living after retirement?’ This is something that our industry is completely based on. No one’s ever adequately tested it.”
As part of her research, MacDonald set out to test whether a 70-per-cent replacement rate could maintain a person’s standard of living before and after retirement. Using a Statistics Canada simulation model of the Canadian population, she tried to capture those who had close to a 70-per-cent replacement rate to see if they’d be able to maintain their standard of living.
MacDonald found that people who hit a 70-per-cent replacement rate experienced a wide range of living standards continuity after retirement. She also found that when looking at those it did work well for, there were no particular identifiable subgroups.
MacDonald also found that people with higher replacement rates didn’t necessarily have better standards of living compared to before they retired and those with lower replacement rates didn’t necessarily have poorer standards of living compared to beforehand.
“The conventional replacement rate is not doing the job it’s supposed to do,” said MacDonald.
MacDonald said the real issue isn’t whether 70 per cent is the right number but whether people should be looking at employment earnings as the measurement. “Final employment earnings in a single year are not representative of someone’s standard of living,” she said.
Two people with identical incomes can have very different standards of living depending on their circumstances, including whether they have a family, are in a dual-income household, carry debt or own their home. If two people with identical incomes but different circumstances save the same amount, one person’s standard of living will go up after retirement while the other person’s will go down.
As a substitute for the 70-per-cent replacement rate, MacDonald proposed using an alternative measure as a basis for assessing how well people maintain their living standards after retirement: the living standards replacement rate.
She said the way to approach it is to look at how much people spend before retirement and how much they’ll need to spend afterwards and then try to make up the difference with savings.
“This is a paradigm shift,” said MacDonald.
“We’ve been using the same replacement rate forever. It doesn’t work.”