January 23, 2012
By Nevin Adams, EBRI
There is an old adage that cautions about the consequences “when you assume…”
And yet, the business of retirement planning is replete with any number of so-called “common wisdom” rules of thumb. Doubtless many have well-intentioned origins—to make complicated concepts easier to grasp, and thus to address.
One of the more pervasive notions is that a realistic target for retirement savings can be determined by accumulating a sum that will provide an income stream equal to a percentage of one’s preretirement earnings—a sum that is generally expressed as 70–80 percent of what you earn prior to retirement. This starting point—generally referred to as a “replacement ratio”—includes any number of embedded assumptions, perhaps most significantly that the individual will need to spend less postretirement, generally understood to be on things such as taxes, housing, and various work-related expenses (including saving for retirement).
Moreover, the replacement rate approach represents, at best, an indirect approach to evaluating whether retired workers can maintain their standard of living in retirement—because what matters is not how much you have to spend, but how much you need to spend. A recent research report sponsored by the Society of Actuaries’ Pension Section, “Moving Beyond the Limitations of Traditional Replacement Rates,” also highlights the limitations of relying on replacement rates. A recent paper published by the Center for Retirement Research at Boston College (“How Much to Save for a Secure Retirement”) acknowledges that “the most direct approach would be a comparison of household consumption while working with consumption after retirement”—before launching into a discussion that instead draws on a relatively simplistic series of assumptions, not the least of which is that the goal of retirement saving is a replacement rate of 80 percent of one’s preretirement income.
The problem is that these assumptions are just that. As a result, in some lucky cases that 80 percent will be more than is required, while for others, unfortunately, it will be less. Furthermore, most of the assumptions underpinning such replacement ratio targets are implicitly using a 50 percent probability of success. How many people want to go into retirement with a 50−50 chance they will run short of money?
Additionally, these replacement rate models tend to ignore one (or more) of the most important retirement risks: investment risk (bad investments or a declining equities market), longevity risk (outliving your money), and the risk of potentially catastrophic health care costs.
The reality is that there is no “correct” single replacement rate. But the factors that undermine those simplistic rules of thumb are quantifiable. Those factors, and the importance of probabilities in retirement planning, are detailed in “Measuring Retirement Income Adequacy: Calculating Realistic Income Replacement Rates” (EBRI Issue Brief, no. 297). After all, it isn’t what you have accumulated at retirement that matters, it’s how much you have left at the end of it.