Roger Kay
Contributor
I cover endpoints and how they relate to the cloud.
Opinions expressed by Forbes Contributors are their own.
Is it better to start taking Social Security at 62, 66, or 70?
Work for a lifetime, and before you know it, you’re staring at your
final productive years. I’ve been a tech analyst for anywhere between
18 and 36 years, depending on how you look at it. I’ve been involved
with tech one way or another since 1970. I’ve covered companies like
Microsoft MSFT +1.59%,
Intel INTC +1.35%, and
Apple AAPL +0.44%
almost since they were founded. But all this tech talk gets old. You
see the same things, if not in circles, then in spirals; if not in
mirror images, then in familiar reflections.
We’re all in our 60s, the analysts I consider colleagues. Tim
Bajarin has been managing his blood sugar for years now, but as far as I
can tell, has hardly slowed down. Rob Enderle is selling his house in
Silicon Valley, having already moved to rural Oregon. He’s a force of
nature in our industry, writes more and talks more than anyone. But
still, such a move could easily presage a next phase of life. I still
have college bills to pay for a few more years, late family-starter that
I am. So, I’ll have to remain pretty active. But retirement is
somewhere out there on the horizon for all of us.
I found myself wondering, abstractly, when should I retire? And close
behind that thought came this one: when should I start taking Social
Security? Somewhere along the way, I got an MBA in finance and
marketing at Chicago, the quant school. We did a lot of modeling there,
and so I decided to model social security, using a wide range of
assumptions. I looked at three key ages: 62, 66, and 70; four interest
rate assumptions for cash-flow discounting: 1%, 2%, 5%, and 10%. And
two important assumptions about the solvency of the U.S. Social Security
Administration. My only tool was a simple Microsoft Excel spreadsheet.
The
ages are key because 62 — which I am now — is when you can actually
start taking Social Security. You get less forever after, but you can
start drawing. As of the moment, 66 represents the “official” 100% age,
when payments are what they’re supposed to be. That’s one year older
than it used to be, and younger than it will be for later cohorts. And
70 represents the age when payments reach their maximum. If you can
wait that long, you get more.
But this cheery scenario, which has served the next-older cohort, the group that’s now
65-74,
quite well, depends on there being money around in the future to make
those fatter payments. And, as I’m sure you know, that may not be the
case.
Addressing the end date for the stream of payments, I looked at two
different reports that each gave an estimate of when — 2031 and 3037 —
Social Security would run out of money if no reforms were enacted. Now,
I recognize that it is unlikely that there will be funds one day and
the spigot will clang shut the next. More probably, sometime before the
system, which has been in the red since 2010, actually runs out of
money, some brave Congresswoman, Senator, or even President will step up
and enact reforms that will allow payments to continue at reduced
rates. However, for purposes of this exercise, I made the simplifying
assumption that payments will continue as they are until they stop.
This assumption has the additional benefit that it can apply not just to
the system, but to the individual as well. That is, the system may
stop, but I may stop first. Even the greatest narcissist can’t truly
believe
après moi, le deluge. After all, as T.S. Eliot so eloquently and mystifyingly said, “This is the way the world will end … not with a bang, but a
whimper. I will cease, but something else will continue.
What is most interesting in the model, however, particularly to
quants, is the interest rate assumption. At Chicago, we plugged the
cost of capital into everything. It was always called Greek letter ρ
(“rho”), the cost of capital. Now, lazy people like me just used the
interest rate as a proxy for the cost of capital, and it’s way beyond
the scope of this column to talk about why that may or may not be a
solid move. But what even is the interest rate? Is it the lending
rate, the borrowing rate, the interbank rate, some other arbitrary rate
set in London, or what capital actually costs
you? Whatever.
You use the cost of capital/interest rate as a means to discount the
value of money between periods; that is, over time. The basic idea is
that a dollar today is worth more than the promise of a dollar
tomorrow. It’s completely intuitive. But how much more? To manage
this question, I used, as noted above, four assumptions. Those savers
among you may have noted that your bank accounts are yielding something
pretty close to zero. Paranoids in our Boomer ranks might think that
these “artificially” low interest rates are a grand conspiracy to
deprive us of our rightful retirement, but the fact is the world is
awash in money looking for a rate of return, and that’s the way markets
operate: all this demand for safe investments drives the price of money
(the interest rate) down.
I didn’t use a zero interest rate in the model because it would yield
an absurd result: $1,000 now would be worth the same as $1,000 20 years
from now, and you know that can’t be right. Perhaps the market
interest rate isn’t a good measure of an
individual’s cost of capital. For an old guy, having the money now while he’s still healthy enough to enjoy is worth a
lot
more than later, when he’s dead. So, I used the 1%, 2%, 5%, and 10%
rates. Folks in my generation can remember when interest rates were as
high as 20%, in the Volker/Reagan years.
So, here are the results. Using discounted cash flow analysis, and
starting the periods at the appropriate times — this year, in 2019, and
in 2023 — when I am/will be 62, 66, and 70, respectively, I ran all the
numbers, determining a net present value for each stream of payments, a
single number to represent that set of assumptions. What they show is
that you might be better off beginning to take benefits at each age,
depending on assumptions,. That is, the model is sensitive enough so
that the assumptions matter. For low interest rates and a longer-lived
system (and person), the recommendation is to wait until 70. For high
interest rates and a shorter-lived system/person, start at 62.
Net present value of the stream of payments
associated with taking Social Security at various ages using various
interest rates and assuming that the system goes bankrupt in 2031
Interestingly, under the model that shows the system crapping out in
2031, no assumption tells you to wait until 70, the maximum benefit
year. And even under the 2037 version, only the 1% and 2% models say
take it at 70. In fact, the 2037 model shows a perfect crossover. At
5%, the recommendation is 66, and at 10%, it’s clearly 62.
Same as above but with the assumption that the system will go bankrupt six years later, in 2037
So, where does that leave
us? I submit that the true cost of capital is higher for the individual
than the current market interest rate would suggest, and that
discounting of future payments at a reasonable rate — like 5% — makes
sense.
Thus, I’m likely to start claiming at age 66.
Good luck with your decision, fellow Boomers!