A Reader Case Study: Whole Life Insurance

Welcome back! I hope everyone had a great 4th of July Holiday (U.S. Independence Day for non-U.S. readers)! Before we get started I have a small favor to ask: At the upcoming FinCon in Orlando in September, it’s time again for the Annual Plutus Awards. As you may recall, last year, my small blog was one of the finalists in the “Blog of the Year” category, thanks to the support of the many faithful readers. If you like what I’m doing here on the blog please nominate the ERN blog again in the relevant categories! Please head to the Plutus Award Nomination site and enter your ballot! You can nominate up to three choices per category and you don’t even have to fill out all categories. Only one submission per IP address, please! Thanks in advance for your support!

Today I have a case study about whole life insurance. Not the most popular investment vehicle among the FIRE enthusiasts, see, for example, an excellent summary of the disadvantages of Whole Life by White Coat Investor (though, for full disclosure, I don’t agree with all of his claims and calculations). But let’s face it: a lot of folks have policies and now wonder what to do about them. Here’s a case study about the tradeoffs when considering either cashing out the policy or keeping it intact. Let’s look at the numbers…

The background

Some details have been changed to preserve anonymity, but without changing the math of the case study!

Abe and Barb (not their real names) are fellow FIRE enthusiasts and reached out to me to answer a few questions related to their parents’ finances. Specifically, Abe’s parents, let’s call them Chris and Deb (also not their real names) signed up for a whole life insurance policy a number of years ago. The policy is a joint-life policy, so it pays out only after both spouses have passed. The policy is for a seven-figure sum, but to preserve some anonymity, I scale all numerical parameters to a policy death benefit of exactly $1,000,000. Specifically:

  • Death Benefit: $1,000,000
  • Current annual premium: $16,326 (ouch!) and that premium has to be paid every year for the duration of the policy up to age 100 (ouch again!).
  • Current cash value (if liquidated today): $105,281.
  • Age of both insured: 60
  • Abe and Barb are the beneficiaries of the policy. The policy is currently in an irrevocable trust with Abe and Barb as trustees.
  • A side note: The policy has been in effect for a number of years now and today’s cash value (the money you get back if you pull the plug on the policy) is actually less than the cumulative premiums over the last years.

My first question after seeing the request: Why the heck would anyone sign up for this kind of policy? It turns out that Chris and Deb own a family business and that business will be handed over to Abe and Barb after both Chris and Deb have passed away. At that time, a potentially sizable estate tax bill would be due. The business is a highly illiquid asset; it’s a privately-owned company (LLC or LP, I presume) and you can’t just sell a few shares at the NYSE to raise the cash for the tax bill. And so goes the rationale for the whole life policy that will pay out exactly when the last survivor passes away. I still don’t quite agree that this is a good rationale for buying a whole life policy, more on that later. But Chris and Deb have the policy now and whether or not it made sense at the time, it’s a sunk cost now. What can they do? The three obvious choices on the table are:

  1. Keep the policy
  2. Cut the losses and cash out the policy today
  3. Wait and see: Keep the policy for a few more years to at least “recover the losses” of the past few years.

There are actually two other suggestions and I will get into the details below. For now, let’s look at the two major tradeoffs: 1) cash out today vs. keep the policy and 2) cash out today vs. cash out in 5 years. The first calculation is quite a bit more involved, so let’s start with that one:

Tradeoff 1: Cash out the policy now vs. keep the policy

If Chris and Deb keep the policy alive they’d forego the cashout value today and the future premiums but they’d keep the death benefit. There’s also the additional complication about the uncertainty about when the death benefit is paid out. And then there’s an additional twist: the policy is held with a mutual life insurance company and as policyholders, Chris and Deb share some of the profits that their company generates. The default setting is for the annual dividends to be reinvested into the policy with the effect of increasing the death benefit. But we have no idea what the future will hold and how profitable that company will be. Therefore, projections about future death benefits are normally done for three scenarios:

  1. Extrapolate the current profitability and dividend rate for all future dates. The “non-guaranteed death benefit.”
  2. Assume no dividends at all, i.e., assume the death benefit doesn’t grow. This is the “guaranteed death benefit.”
  3. The midpoint between the two.

So, we’d have to calculate everything exactly three times because of the three alternative dividend assumptions. More work for me!

To warm up, let’s do one quick calculation and assume that the death benefit is paid out after exactly 30 years when the last survivor dies at age 90. Not too different from the joint survivor life expectancy as we shall see below. So, what are the tradeoffs? The incremental cash flows of keeping the policy are

  1. a negative flow of the cash value today,
  2. then 30 annual premium payments (also a negative flow)
  3. and a positive payment at the end equal to the death benefit.

Abe and Barb shared the numbers from the most recent policy statement:

  • A guaranteed benefit of $1,000,000.
  • A midpoint benefit of $1,375,785. It’s not exactly the arithmetic average of the guaranteed and optimistic benefit, so I can’t really tell what they mean by mid-point…
  • An optimistic benefit of $1,861,078.

Let’s calculate the internal rate of return (IRR) and the net present value (NPV) of the cash flows for a specific discount rate set to 4%, just like my blogging friend AoF did in his recent post. The results are in the table below. Not a bad return profile! The worst-case scenario (without any dividends) yields 3.06%, about as much as the 30-year Treasury bond right now. With the optimistic assumption of the current dividend yield, we get 6.23% and 4.72 for the midpoint. That’s worse than historical equity returns but still not too shabby! I personally doubt that even equities will easily get above 6% returns over the next ten years, but over 30 years you have a pretty good chance of beating that, even with today’s high CAPE ratio.

WholeLife Case Study Table01
IRR and net present value of cash flows (4% discount rate) if the policy pays out after 30 years (age 90).

Summary so far: it looks like once you disregard the sunk costs of bad returns over the initial few years of the policy, subsequent returns are definitely attractive relative to an investment in bonds.

So much for a warm-up! The actual problem at hand is slightly more complicated: There is the uncertainty about the timing of the death benefit because nobody knows when the last survivor passes away. So, let’s get an actuary to weigh in! Actuary on FIRE recommended the site Longevity Illustrator that helped me gauge the probabilities for when the last survivor passes away (I used two non-smokers, average health, both 60 years old). I calculated all the internal rates of returns IRRs and net present values (NPVs) for different ages, in steps of 5 years to save space (and computation effort). See table below.

I also calculated the IRRs and NPVs of the probability-weighted cash flows. For the math wonks, it’s important to note that I calculated the probability-weighted cash flows first and then the IRR, not the other way around. For the super-wonks, that makes a difference because the IRR calculation is “non-linear.” I confirmed with Actuary on FIRE that this is the proper, “actuary-approved” method in this context. The probability-weighted IRRs are pretty decent: between just under 4% (higher than current government bond yields) to 6.97% if we are optimistic and assume the insurance company keeps handing out dividends at the same rate as before. That’s a tax-free return, just to be sure!

WholeLife Case Study Table02
IRR and NPV at different ages of last spouse passing away. The bottom row uses the probability-weighted cash flows.

So, what’s the final recommendation? It depends! How much of a strain does the annual premium put on the parents’ cash flow? If they didn’t have to pay the premiums how would they use the money? Buy a government bond ETF yielding 2.8% to maybe 3.0%? Then it’s probably a better idea to just stick with the insurance policy. On the other hand, if they were to invest the freed-up cash flow in stocks or the business itself you might get better returns there! From Abe and Barb, I got the sense that tying up more money in the business might not be the most palatable option for the family right now. And stocks also seem a bit overvalued right now. One route that Abe floated and that would have been my suggestion, too, would be to keep the insurance policy intact and make this the “bond portion” in their personal portfolio. Then keep the remainder of their portfolio invested primarily in stocks. As long as U.S. bond yields are so low keeping the policy doesn’t seem so bad.

Tradeoff 2: Cash out the policy now vs. cash out the policy in 5 years

Another alternative: Why not “wait and see” for another 5 years? I think that’s an interesting route if done for the right reason. Of course, a bad reason would be to recover the losses from the last few years. They are sunk costs and should have no bearing on the decision today. Sometimes behavioral biases (loss aversion in this case) get in the way of rational decisions. But a good reason would be the following; I got the sense from Abe and Barb that the parent’s business is currently exploring new routes expanding their business. If this all turns out well then who cares? If the business becomes even more successful than it already is, there will be a lot less of a burden from having to pay the substantial premiums every year.

So, numerically, the tradeoff is between cashing out the $105,281 today vs. paying $16,368 for five more years and (according to the most recent statement) getting $204,637, $209,166, or $213,827 at the end as the surrender value of the policy for the guaranteed/mid-point/optimistic dividend scenario. The IRRs are now 2.43%, 3.03%, 3.63% in the three dividend scenarios. That’s really low compared to long-term equity returns but not too shabby compared bond or CD rates right now. And it’s tax-exempt!

And it gets even (slightly) better: In the calculations above I ignored the death benefit. It’s because the death of both parents is so unlikely (only about 0.28% cumulative probability over the five years). But a small probability multiplied with a large sum can still make a noticeable difference. So, if I factor in that there’s a 0.056% probability each year of a $1,000,000 insurance payout then the IRRs become 2.77%, 3.36%, and 3.88% under the three alternative dividend scenarios. Again, not that bad of an expected return for a low-volatility “investment” that’s also tax-free.

Side Note: Was this policy a bad idea from the start?

Apart from all the math, I’d like to share one or two additional thoughts on how much sense this particular whole life insurance policy made when Chris and Deb signed at the dotted line several years ago. I certainly understand the concern of the parents to care for the kids. But if you leave an estate big enough to generate a $1m estate tax bill one would think that the kids will be able to choke up enough cash to pay the tax bill themselves. Here’s a back-of-the-envelope calculation on how big an estate would have to be to generate a seven-figure estate tax bill:

  • Assume a $5m exemption. The 2018 exemption is actually now $11.18m, but a few years ago nobody would have known that 2017 tax reform bill would increase the exemption by that much. For most of the last decade, the exemption was at $5m and slightly above!
  • The first $1m above the exemption uses a sliding marginal tax scale, for a total of $345,800.
  • Above $1m, we face a marginal tax of 40%. So a $1m estate tax bill would come from an estate of $6,000,000+($1,000,000-$345,800)/0.4=$7,635,500.

So, to get a $1m estate tax bill, the business and all of Chris and Deb’s other assets would have to be assessed at over $7.6m. Illiquid or not, I would think there will some way to come up with a million bucks. Especially considering that Abe and Barb are FIRE enthusiasts who will likely have plenty of liquid financial wealth on their own two or three decades down the road. From an ex-ante point of view, I would have probably advised against signing up for the policy.

But there’s another concern! I have a fundamental problem with planning for an estate tax issue that many decades into the future. The main reason: businesses fail. They can fail for all sorts of reasons: macroeconomics (recession), sector effects (the specific sector goes through a tough time) or idiosyncratic risk (getting sued or regulated out of business, etc.). And with this expensive life insurance policy, you’d compound the business risk because if the business fails you’ll still be on the hook for the insurance premiums for as long as you live. Of course, you can always surrender the policy but now you have unnecessarily tied your personal investments to your business success. Why add more correlation to your investments?

But again, this is all “water under the bridge” now. But food for thought for others considering how to deal with the estate tax issue right now.

Other options to consider

Two other ideas I want to throw into the mix. First, before canceling the policy I’d shop around and see if someone wants to take over the policy from you. It’s called life settlement. Hedge funds, banks and other investors have gotten into the business of buying out existing life insurance policies. I can’t speak from experience here and I certainly can’t recommend or endorse a provider. You might get more than the surrender value that the insurance company offers, but there are also some drawbacks, see this article at AARP.

Whether you surrender the policy or sell it to a third party, I’d also explore the option of surrendering/selling a portion of the policy and keeping the rest intact. For the third party route that’s normally an option but I’d also check if the insurance company allows a partial surrender/cash-out.

Hope this helps and best of luck! And I like to invite other readers to weigh in with your comments and suggestions!

Picture Credit: pixabay.com

Advertisements

20 thoughts on “A Reader Case Study: Whole Life Insurance

  1. First known case in the wild of it maybe making sense to keep a whole life policy instead of cashing it in? 😉

    I like the wait and see for 5 years approach the best. Getting paid 3.36% (midpoint return w/ death benefit factored in) to wait and see isn’t bad vs 5 yr CD yields. Slightly better yield than 5 yr treasuries (2.73%) or the best NCUA/FDIC CD rates (3% at Penfed??). And the whole life cash out doesn’t come with interest rate risk or early termination penalties!

    If equities are in fact overvalued and we see a correction, then it might be a good time to cash out the policy and dump it in equities at that point. Otherwise holding to maturity seems to offer better than bonds (worst case) and equity-like returns (best case) so while you may not beat a stock index fund for 30 years you’ll probably come close.

    Liked by 2 people

    • Just remember if you hold on to it long-term, any gains above the basis (cost of premiums) is taxable at ordinary income tax rates. And you have to cash it out entirely in one year (or perhaps sell on the third party market as Ern points out) – ouch! Of course, you could take a loan against it, but that kind of defeats the value since you will pay at least the minimum interest rate noted in the policy – and if you don’t – it keeps accruing and will reduce the cash value and death benefit. Intriguing analysis and overall situation! Well done Ern.

      Liked by 1 person

  2. There’s really not much of return even if you keep it.

    My taken-out-in-1984-by-my-parents, plain-vanilla, $500,000 whole life policy now pays ~$600,000 death benefit.

    After nearly 35 years of ~$3000/year premium payments.

    Liked by 1 person

  3. Whole Life life insurance is such a bad “investment” and I struggle to even use that word. In this day and age where Term Life is SO cheap…it is a no brainer to take the cheap insurance and invest the rest (plus more obviously). Pretty soon you will be able to self-insure and you are set!

    Liked by 1 person

      • Very true. However, even in their case a term life insurance plan could have been part of a total “money/work/life”-plan, where for example the term life insurance pays until age 90; and the parents ALSO create a plan to either sell the business or transfer it to their children BEFORE they are 90 (let’s say by age 80, but give them the 10 extra years to have some time for the whole process of selling/transferring). So then the term life insurance would cover the period during which the business is intact and owned by the aging parents. And if the parents were to live beyond age 90 and not get the term life insurance payout, the business would either have been sold already or have been transferred to the kids already.

        Liked by 1 person

    • I’m struggling with that! On the one hand, it’s been a reliable indicator. On the other hand, I have pointed out scenarios how the YC could fail us next time:

      “A Type 1 Error, where the yield curve never inverts because the Federal Reserve is much more gradual in raising interest rates than in previous tightening cycles. Then the 10-year yield may never drop below the 2-year. But then some external catastrophic event, e.g., Rocket Man going crazy or a worldwide flu pandemic that paralyzes the world economy. The yield curve would likely still invert at that time (unless the flu wipes all the bond traders) but it would likely occur at the same time as the recession rather than before the recession.
      A Type 2 Error (false alarm), where the Federal Reserve is surprisingly aggressive in raising rates. The yield curve inverts because the 10-year yield is held down by demand for U.S. Treasury Bonds from abroad where interest rates remain artificially depressed. But the U.S. macroeconomy withstands the rate hikes and chugs along for another 5 years before going into another recession. I would probably not get too worried about just the yield curve inverting without any of the other indicators (see below) confirming the signal!”
      see: https://earlyretirementnow.com/2018/02/21/market-timing-and-risk-management-part-1-macroeconomics/

      Like

  4. If I’m not mistaken, you can add over 5M the estate calculation, for 1M in taxes. the 5M exemption, is per spouse. And transferable, in case not used by first death (between spouses, its totally exempt). So less to worry:)

    Liked by 1 person

    • True! But I think the set of people who truly benefit is much smaller than the industry wants us to believe.
      But for full disclosure, we both have a small whole life policy too. So, I don’t dislike the product as much as some most in the personal finance blogging world! 🙂

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s