What the Social Security Administration knows
We’re going to get a little morbid here. For people nearing retirement age, they often evaluate whether to take their social security benefits early, on time or late. If they take them early, the benefit is they receive them early; the downside is they are permanently reduced by a factor set by the Social Security Administration (SSA). For those deferring retirement beyond the standard retirement age (currently 67 for those born after 1960), they get the benefits later, but they are permanently higher. So it really matters when you’re going to die when determining what’s the right choice.
Of course, the SSA knows this. And they backsolve the discounts for early retirement and bumps for deferred retirement based on your expected death date. But the SSA can’t very well discriminate by different categories, and for example, assume a death age that is different for males versus females.
For example, if you were born after 1960, your “full retirement” age designated by the SSA is 67 years old. If you take retirement early (late), there is a formula to discount (increase) your retirement benefits. If you retire at age 62, you’d get a 30% reduction in your benefits. Therefore, suppose you were due $1,000/month under full retirement age; then you’d only be getting $700/month. The benefit is you got the $700/month for 5 years (60 months) more than you would otherwise; that’s $42,000. The downside is you have permanently given up that $300/month. We can take this math and derive that if you died before age 78 years and 8 months, you’d be better off (you got a $42,000 benefit spread out over 5 years versus lost $3,600/yr for 11 years 8 months, which is also $42,000).
What you know
So the SSA is using that one assumed death date when deriving the discounts/bumps for early/deferred retirement, and they cannot take into account your gender, health, smoking habits, etc. In the above example, if you’re a male who drinks and smokes, you should probably take social security early because you will not make it to the backsolved death age. Here is an expected death calculator for those morbidly inclined. Apparently, I’m defying the odds as I should already be dead.
Then, you can take a look at the below table I’ve derived from their discount rates that shows implied death breakevens by your birth year and retirement age (that was an annoying exercise since SSA varies discounts/bumps by birth year). Compare that to when you expect to live, and see if there’s an opportunity to game the system using what you know about yourself.
1. If you take early or late retirement, it does not impact your spouse’s decision to take spousal benefits; that’s a separate decision. I’ll probably put out a post on that (separate) decision.
2. However, it does impact survivor benefits. So if you die, your spouse could choose to take your benefits (if they are higher than their own), and those benefits would be impacted by your decision. Factoring that in is quite simple, however. Just take the difference between your derived death breakeven date in the table versus the retirement age (e.g. if 77 years death breakeven at a retirement age of 62 years, that’s 15 years), and if you or the survivor collect less than that in the case of early retirement or more than that in the case of late retirement, then the decision makes sense.
For simplicity and to be conservative, I assume that your surviving spouse would collect survivor benefits no matter what. You may have a more complicated scenario whereby your survivor may choose to collect their own benefits rather than survivor benefits. For example, suppose you were due $1,000/month, but due to early retirement only collect $800/month. Your surviving spouse can collect $800/month or whatever they would get on their own. Perhaps they are able to collect $900/month on their own, so they’d choose that (they can only pick their own or survivor benefits, not both). We assume they collect your $800/month, as this makes the early retirement scenarios conservative, in that if we say you or a survivor need to collect for <12 years to make early retirement worthwile, it may be longer. Conversely, for late retirement, if even with your benefit increase, your spouse would fall somewhere between your full retirement pay and the higher amount, then the number of years would need to be greater than shown. So it is not conservative. However, these scenarios are not very likely. Trust your judgement.
3. There are some scenarios where you might want to take early retirement if you believe your spouse may pass before you (as you can then switch to survivor benefits). We’re trying to keep things simple, so we don’t calculate for that, but comment below if need help.
4. Social security has COLA (cost of living adjustments), so we are already adjusting for future rates of inflation in these calculations. This is all in present value space.
5. Regardless of what you choose, you should still likely take Medicare at age 65 (independent decision).