Imagine this situation: A married couple has one spouse who is a stay-at-home parent (who generates no income). The income-generating spouse dies at age 30, with an appropriately-sized life insurance policy. How should the surviving spouse invest the life insurance proceeds?
It’s a tricky question, and I don’t have a perfect answer. (I’m not sure there is one.) Still, I think that anyone who could potentially find himself/herself in such a situation would be wise to make a plan ahead of time.
It’s akin to planning a super-long retirement. We have to determine how to invest a portfolio from which you want to take a stream of inflation-adjusted withdrawals over a very lengthy period (potentially more than 50 years).
- What asset allocation would you use?
- What rate of withdrawal would you be comfortable using?
A part of what makes these questions so difficult to answer is the fact that we can’t learn much from backtesting various asset allocation/withdrawal rate combinations to see how they’d hold up over historical 50-year distribution phases. After all, we only have two such independent 50-year data sets–not exactly a large sample size.
Asset Allocation
On the one hand, for such a lengthy period, it seems that it would be difficult to achieve the long-term returns necessary to sustain 50+ years of withdrawals without a hefty stock allocation.
On the other hand, the “sequence of returns risk” problem that plagues retirement planning becomes even worse when we’re looking at such a long period. If the investor uses a stock-heavy allocation and the first few years go particularly badly, the portfolio could easily fail to generate the desired income for another 50+ years.
Personally, I’d attempt to minimize sequence of returns risk by using a fairly conservative allocation–something like 40% stock, 60% bond (with a healthy portion of the bond allocation being invested in TIPS). But I’d only be comfortable using such a conservative allocation because I’d also make sure to…
Use a Low Withdrawal Rate.
The most important piece of the puzzle is to use a very low starting withdrawal rate (3% or lower). The goal is for the portfolio to last almost indefinitely. If you aim for the portfolio to be depleted at the end of the 50-ish-year expected time horizon, but you overestimate the sustainable withdrawal rate by even 1%, you could run out of money far earlier than desired.
What Would You Do?
As I mentioned above, there simply isn’t enough data to get a very conclusive idea of how well any given strategy would have held up historically over 50+ year periods. As such, the above thoughts are what I would do with the money, but I absolutely cannot say that there would be no better approach.
What would be your plan if you were faced with the prospect of having to draw from a portfolio for (potentially) more than five decades?
Hi. I'm Mike Piper, the author of this blog. I'm 
The scary thing as I look at your thoughts is I may not have enough insurance. Scarier, I was just telling my wife the other day where to find the insurance info.
My original goal was to have enough to allow her to pay off the house and then be able to have some income, but would still need to work. Running what I have through your scenario doesn’t look very pretty.
I was recently thinking of extending the time frame it is currently a 20-yr term that I set-up last year. But the 20-years would be cutting it short if something happened near the end.
A better goal may be to make sure the withdrawals can cover expenses and then some and not necessarily pay off the house. Basically, she could live off the withdrawal if desired, but she is certainly a smart cookie and could work. Our youngest is 8 though, so wouldn’t want her to feel like she has to work at least for a few years.
Although it is hard to think about these things, I certainly am glad for the post. I think I need to make some changes.
I would be concerned though with even 40% in stocks since a bad year or two could be disastrous. :O)
Hi Chris. I’m happy you found the post helpful.
One thing that I now realize I should point out is that in the article I’m assuming that the goal is to continue to allow the non-working spouse to not work.
As you mentioned, if that spouse is willing and able to go (back?) to work, then a smaller amount of insurance would be necessary, as the amount of required income per year would decline once he/she found a job.
The insurance proceeds don’t have to last indefinitely. The surviving spouse is only 30 (or thereabout). He or she can start working and only draw from the investments to supplement his/her own income. Single working moms/dads are not uncommon at all. He or she may remarry as well.
TFB: You’re absolutely right.
I hadn’t meant to imply that not remarrying and not going back to work necessarily should be anyone’s plan. Rather, I was simply attempting to address the question of what to do if that is the plan, as it’s a difficult investment question to answer.
I agree with TFB. These were exactly my thoughts as I read the article. I think it’s important to be realistic. In this unfortunate situation, this should not/would not be the plan.
Debbie: I actually wrote this article with 2 couples I know personally in mind. For each of them, it is the plan for the non-working spouse to continue not working.
I agree that that shouldn’t be the plan for everybody, and I admit that it certainly requires a lot of insurance and presents an investing challenge. But I feel that if that’s what they want to do (and they’re OK with making the necessary sacrifices now), the most helpful thing I can do is attempt to provide guidance on how to achieve that goal.
Edited to add: Or what about a scenario in which both spouses already are working, but one makes significantly more than the other? In that case, there isn’t really a “go back to work” option, as the surviving spouse is already doing his/her best to earn what he/she can. In such a scenario (unless the surviving spouse is OK with a decline in living standard), you’re again left with the “pile of money that needs to last a long time” outcome.
Of course, the surviving spouse can remarry, but I would think it’d be ideal for that person not to need to remarry for financial reasons.
I’d almost go more conservative than 40% stocks/60% bonds. I’d go either 30/70 or 20/80. Capital preservation is more important than growth for the surviving spouse.
Plus, as others mentioned, the surviving spouse could go get a job. If they did, the income generated from the insurance proceeds could be used as income supplement, rather than the primary source of income.
Allocation is not nearly so important as withdrawal rate. For that timeframe, you should go down to around 2% — the money needs to last essentially “indefinitely.” Withdrawal rate is the big determiner of probability of success. If you go over a sustainable rate by even a tenth or two, your probability of making it to age 95 drops significantly.
My favorite reference, currently, on the topic is Jim Otar’s Retirement Myth Unveiled. Personally, I’d probably pick an allocation something like 45% equity, 35% short-to-intermediate bonds, 15% tips, 5% money market. ETFs or cheap cheap index funds (because fees are akin to pushing up the withdrawal rate). But seriously, allocation is way less important than withdrawal rate.
Just to follow up, a 3% withdrawal rate is the kind of rate you’d pick somewhere between the ages of 50 and 55. Way to likely to fail at a young age like 30.
Don: I agree that a lower withdrawal rate is better, but I’m not sure I’d agree that a 3% starting withdrawal rate necessarily has a very high likelihood of failing.
If we look at historical scenarios starting in 1926, according to FireCalc, a 3% withdrawal rate (indexed to CPI) with a 40% equity portfolio would have been successful over 97% of the 50-year periods.
As I said above, I don’t place a ton of confidence in this, as a) there isn’t a ton of data, b) these 50-year periods aren’t independent data sets, and c) the future won’t look exactly like the past. But I don’t see any evidence indicating that 3% necessarily is high-risk.
What am I missing here?
It might be what you’re missing is that in my family people live a long time. A thirty-year old is going to want 65 years of income in my model. I ascribe to William Bernstein’s philosophy–the goal isn’t to retire rich, it’s to avoid dying poor.
It might just be that I’m pretty conservative.
Oh, and I actually think another difference is that I find Otar’s argument somewhat compelling that you can’t reasonably expect anything like the dividend return that people were accustomed to seeing 100 years ago. When I backtest, I assume a modern dividend rate, and that leads to more conservative conclusions.
Don: Both valid points. Thank you.
And I absolutely agree that we shouldn’t count on the same returns that the market was able to achieve over the last century.
Wouldn’t this be the perfect spot for the single premium immediate pay annuity? You’ve figured out the income you wanted to replace, you bought the term insurance, and now you have the funds in hand to replace that income.
DIY Investor: Yes, eventually. However, to the best of my knowledge, no insurance companies offer inflation-adjusted lifetime SPIAs starting as young as age 30. (And for that matter, the mortality credits that young are likely to be close to nill, so there wouldn’t be a very large advantage to annuitizing at such a young age.)
I see most people are focused on how to invest the money and withdrawal rates and I’m wondering if we should step back and think about what the money is for. Maybe the house should be paid off and 529 plans ramped up. Some might be used to go back to school for the surviving spouse if needed. For the investable portion I would agree that a 60% stocks/40% bonds allocation with a 4% withdrawal rate would be fine.
You’re right on the annuity. I found a quote starting at age 40.
David Trahair (Canadian author) had a great line in his book “Smoke & Mirrors”. He advised not going overboard with life insurance because “it will end up paying for the rich lifestyle of your spouse and his/her new partner after you are gone. 🙂
Seriously, I plan to get more life insurance. When I made our initial calculations – I got enough so that my wife would be ok for a while, but would probably have to go back to work eventually.
The problem is that she has now been out of the work force for about 6 years and will likely remain out for another 2 or 3.
While I’m sure she could get a job – I’m not sure how easy it would be or how much it would pay. Which is why I think I’ll increase my life insurance.
Interesting post. My spouse and I are both well past age 30 and, therefore, will never personally face the situation presented. That said, I feel now, and am pretty sure I felt this way 30 years ago, that I would not want to encumber my spouse with essentially staying married to someone who was deceased (the contract is “till death do us part”) for half a century or longer. I always assumed that either my spouse or I would get on with life, meaning certainly going back to work and probably re-marrying, if one of us died at a young age. The only circumstance, in my opinion, in which the scenario presented is realistic is if there are special circumstances, such as where the non-working spouse is handicapped.
MikeH – Although I wonder (if that were a true concern) if the insurance proceeds could be directed into a trust which would only be used for specific purposes such as current home payoff, college savings, “normal” living expenses, etc.
I teased my wife that I wanted a good figure that would provide, but not too high that would encourage my wife to take me out. :O)