By Mark Hulbert, MarketWatch
Everyone “knows” that you should defer receiving Social Security benefits until age 70.
But to what extent does this advice depend on the federal government making good on its Social Security commitments? The Social Security Administration trust fund is currently projected to run out of money in 15 years, and if no changes are made before then, in 2034 will be able to pay just 77% of scheduled benefits. Might it therefore be better to start receiving benefits at age 66—or even take early retirement at age 62?
There are at least two reasons to revisit this perennial question today. One is that Laurence Kotlikoff, a Boston University economics professor, recently took issue with my column on this subject. Another is that, given that the midterms have produced an even more gridlocked Congress, the already-slim chances of a solution to the eventual Social Security funding shortfall have become even more remote.
My column on this subject, you may recall, focused on an analysis by Richard Band, the then-editor of Richard Band’s Profitable Investing newsletter. Band had just decided to begin receiving Social Security benefits at 66 rather than following the conventional wisdom of waiting until 70. Taking Band’s lead, I showed that, under certain assumptions about how the Federal government will handle its Social Security obligations, the present value of a retiree’s future Social Security benefits is higher if he begins receiving benefits at age 66 than at age 70.
Band wrote: “If you believe that Uncle Sam, saddled with a $20 trillion—and rapidly rising—national debt, will keep his Social Security promises flawlessly, you might as well wait until age 70 to take your benefits (assuming you’re in normal health).” Because he was skeptical, however, Band decided to “take the money and run.”
Kotlikoff’s analysis used different assumptions than I used in my earlier column and, not surprisingly, reached different conclusions. One big difference is that I assumed that Social Security would begin reducing benefits gradually before 2034, rather than waiting until then and facing a 23% cut in benefit payments. To the extent cuts occur earlier rather than later, of course, those cuts will have a bigger impact on the present value of all future payments.
The gradual cuts I put into my analysis were that benefits would be cut 0.5% a year starting in 2023. The start date of those cuts is arbitrary, of course, but the notion of such cuts is very much being considered. Some in Congress, for example, are proposing that there be a change to how the Social Security Administration calculates annual cost of living increases, the impact of which would be to lower the annual COLA by an estimated half a percentage point a year.
Another assumption that has a big impact on whether or not to begin receiving Social Security at 66 or 70 is the rate at which future payments are reduced in order to represent their “present value.” To the extent this rate, known as the discount rate, is larger, then future years’ payments will be deemed to be worth less—and taking Social Security earlier becomes the preferred course of action.
In my January column I assumed the discount rate would be equal to the yield on the 10-year Treasury, which at the time was about a half percentage point higher than the inflation rate. I’ve been unable to determine the discount rate used by the software on Kotlikoff’s website, but many software packages assume that the discount rate is the same as inflation. If so, then my assumption of a higher discount rate would be another reason my analysis found that it might be better to begin receiving benefits earlier rather than later.
To be sure, Kotlikoff is entirely reasonable in basing his analysis on the assumption that there will be no cuts until 2034. Given the gridlock in Washington, it’s certainly possible that no changes to Social Security’s funding will get enacted before then, which is the point at which current law would mandate a 23% cut in benefit payments.
Likewise, reasonable people can and do disagree on what is the proper discount rate. ( Click here for a long and thoughtful discussion by retirement guru Michael Kitces on what discount rate to use in retirement planning .)
But the existence of such disagreements between reasonable people just reinforces my broader point: The question of when to begin receiving Social Security is not as settled as many assume it to be. Don’t let the apparent precision of the numerical conclusions reached by various analyses (including mine!) cause you to overlook the crucial and dominant role played by those analyses’ assumptions.