A Critical Review of Indexed Universal Life

Published on 04/02/2020

(This article was published in Joseph Belth's blog in April 2020 - you can see his summary here.)

Indexed Universal Life (IUL) has long been touted as a product that provides upside potential with downside protection. With the precipitous market decline in the first quarter of 2020 caused by COVID-19, this is no doubt a great opportunity for those who sold or purchased IUL products to take a victory lap. But should they be celebrating? And more broadly, are IUL products everything they are cracked up to be?

Understanding IUL Policy Mechanics

Generally speaking, IUL products allow a policyholder to select among a variety of indices that will serve as the benchmark for calculating a credited interest rate at the end of each period. The most popular index choice is the 1-year point-to-point S&P 500. With such a product, the S&P 500 price index (excluding dividends) on each anniversary serves as the starting point. At the end of each policy year, the percentage change in the index is calculated.

If the index has lost value, then the percentage change will be negative. The downside protection of the IUL provides that the credited interest rate will never be less than the floor, typically 0%, so even though the index lost value, the IUL policy will have a credited rate of 0% (or whatever the floor is) in that year.

On the other hand, if the index has gained value, then the percentage change will be positive. Typically, a policy will have a non-guaranteed cap rate that is associated with each indexing strategy. In today’s IUL products, it’s common to see cap rates in the high single digits – for the sake of argument let’s say the cap rate is 9%.

That means that the rate that will be credited on this particular IUL policy in that policy year will be no higher than 9% but no lower than 0%. If the price return happens to fall somewhere between 0% and 9%, then that will be the rate that is credited to the policy in that particular policy year.

Here are some critical observations to understand about the IUL policy mechanics:

  1. Again, the index does NOT include dividends. From 1955 through 2019, the S&P 500 provided a total annualized return of 10.4%. However, excluding dividends, the S&P 500 price index (which is what is used for IUL policies) provided a total annualized return of 7.2%. Consumers (and agents) are inclined to think of stock returns as averaging 10% historically; doing so overvalues the upside potential of IUL policies as “expected” returns for the index should be far lower, even if future performance doesn’t stray from the past.
  1. The cap rate is non-guaranteed and can be changed annually at the company’s discretion. As will be discussed later, there is significant cause for concern about what future cap rates will be.
  1. Even though the credited interest rate can never be less than 0%, that DOES NOT mean that a policy cannot lose value. In fact, given that expenses and cost of insurance charges will continue to come out of a policy, the account value of a policy will always decrease when the credited return is 0%. While there is no doubt some downside protection against market crashes, it does not provide protection against a policy lapsing if there is insufficient account value to sustain the policy.
  1. In times of high market volatility, there can be high sensitivity to the exact date that a policy was issued. For instance, let’s look at two policies with a cap rate of 9% - one policy has a March 4 anniversary date, and one policy has a March 11 anniversary date. For the policy with a March 4 anniversary date, the S&P 500 price index returned 12.1% for the period ending 3/4/2020. With a 9% cap rate, that means that the credited rate for that policy year would be 9%. In contrast, for the policy with a March 11 anniversary date, the S&P 500 price index returned -1.5% for the period ending 3/11/2020. With a floor of 0%, that means that the credited rate for that policy would be 0%. (Note that the March 11 policy would have a potential crediting advantage in the following policy year because the index would be starting at a lower point relative to the March 4 policy.)

Do IUL Policies Provide an Upgrade Over a Good Whole Life Policy?

In my opinion, no. They certainly look good on paper, and you can contrive situations where an IUL policy can perform well for short (or even relatively long) periods of time, but over the long haul I do not believe a compelling case can be made for an IUL policy outperforming a whole life policy from a good carrier (say a highly rated mutual insurance company) – particularly if the whole life policy were optimized to reduce agent compensation and maximize policy efficiency.

How is that possible? It’s important to understand that in order to provide the 0% crediting floor that is found in most IUL policies, an insurance company must invest almost all of its assets (think 95%+) in bonds in order to ensure no loss of principal. Conservatively ignoring all of the other expenses such as commissions, underwriting, etc., would leave roughly 5% of its assets to invest in derivatives and hedges to provide the upside potential if it materializes.

A common hedging strategy is to sell a call at a level that is consistent with the non-guaranteed cap rate, using those proceeds to partially offset the purchase of a call at the current market level. The insurance company backs into a cap rate that can be supported with the available budget (including proceeds from selling the call option), and that cap is in effect for the next policy year.

Unfortunately, the way things have unfolded in the IUL illustration battles, the numbers in an IUL illustration generally aren’t worth the paper they are printed on. Here are numerous issues with many IUL illustrations:

  1. The one thing that we can say with absolute certainty is that the early expenses on an IUL policy are massive, typically the most of any insurance policy I evaluate. Part of that is due to high agent compensation – which creates a powerful “win-win” perception with both high agent compensation and attractive illustrated values, and that no doubt fuels the popularity of these policies.
  1. Part of the massive first-year hit could be attributable to the insurance company deliberately suppressing the cash value and instead using some of that value to increase the options budget which allows for a higher cap rate. One very concerning observation is that it’s typical for the very same company that offers a 9% cap rate on an IUL policy to offer a 5% cap rate on an otherwise identical indexed annuity. The annuity doesn’t have as many moving pieces, and therefore the crediting methodology is more straightforward and transparent – and without fail that leads to a dramatically lower cap rate.
  1. Another part of the heavy acquisition expenses could be attributable to “lapse-supported pricing” that suppresses the early cash values with the intention of using profits derived from early surrenders to subsidize the long-term values of those who persist. This is not a technique that any insurer will readily admit, and in fact it should be essentially extinct now that the Illustrations Regulation has specified certain testing to limit the amount of lapse-supported pricing. Nonetheless, any time you see inexplicably high long-term policy values coupled with extremely low early policy values, one can’t help but be skeptical that there is an element of lapse support in play. And critically from the policyholder perspective: If lapses don’t come to fruition at the rate which is implicitly assumed within the illustration, then it’s less likely that the company will be able to deliver the illustrated values.
  1. Many companies have chosen to prop up the cap rate by punishing the policy in other areas such as higher cost of insurance (COI) rates, higher expenses, and/or higher surrender charges. Or more cynically, this could just be viewed as a form of “bait and switch” that some bad actors in the insurance industry have engaged in for decades – and indeed the industry has already seen IUL policies that have lowered cap rates after a policy has been in force. I strongly believe that there is going to be substantial downward pressure on cap rates on IUL policies in future years, and you are going to see many disgruntled policyholders as the cap rates decrease.
  1. As a timely aside, I suspect that the cost of options has dramatically skyrocketed due to increased volatility in the markets lately. I would not be surprised at all if cap rates drop sooner rather than later due to the higher hedging costs. Ironically, the very same crisis that companies and agents may use to try to sell more policies may make these policies less desirable if the increased volatility produces lower cap rates.
  1. Sadly, many illustrations contain inexplicable non-guaranteed bonuses. For instance, I recently reviewed a policy with a 9.25% cap rate that described using an illustrated rate of 5.76%, the maximum allowed due to regulatory guidelines (per the illustration). The illustration also notes that there is a non-guaranteed bonus that is included in the illustrated values.
  1. One basic rule of thumb for evaluating if an illustration passes the smell test is to look at the long-term rates of return and compare those to the assumed crediting rates. If you get a long-term IRR that is higher than the underlying crediting rate, then you know something is fishy. At age 100, the illustrated rate of return (for both cash value and death benefit since they are basically equal at that point) for this policy is 7.6%. When you consider that the illustrated rate is only 5.76%, you can see just how absurd this illustration is. Simple cash flow suggests that if you have a pot of money that is credited 5.76% - and from that pot of money you also have to pay cost of insurance charges, agent compensation (which has already been shown to be enormous), underwriting expenses, etc. – then there is no way that the long-term rate of return on the policy can possibly be higher than 5.76%.
  1. To put it bluntly, this illustration isn’t worth the paper it is printed on. The Illustrations Regulation was developed in the mid-1990s to curb rampant illustration abuse. Since IUL products did not yet exist, those products were not adequately addressed by that regulation. The industry developed Actuarial Guideline 49 (AG49) in 2015 in an effort to curb IUL abuses that had developed. Not surprisingly, the industry has responded by developing creative crediting and bonus strategies, many of which are reminiscent of the abuses that led the industry to adopt the Illustrations Regulation in the 1990s! While I’m surprised that this illustration is in compliance with AG49 – and if it is then it should be the leading exhibit why AG49 needs to be revised – what seems indisputable is that this illustration is not consistent with the spirit of the Illustrations Regulation that attempted to curb the illustration wars.
  1. Here’s another demonstration of how ridiculous this illustration is. For the illustration I reviewed, no money came in or out of the policy during the 16th policy year, so we can simply look at the change in the account value to see how the policy values are changing. Such a comparison reveals that the account value increases by 7.50% in that year! If we look even deeper, we can see that if we account for the cost of insurance charges that are described as coming out of the policy in that policy year, the actual crediting rate in the 16th policy year is 12.75%. Not bad for a policy with a cap rate of 9.25%!
  1. Another way that these IUL illustrations delve into the world of fantasy is when they illustrate heavy distributions. That’s because it’s common for these illustrations to show a permanent arbitrage associated with borrowing, and this illustration is no exception. Heavy borrowing begins in year 17 with an assumed policy loan rate of 4.76%, whereas the crediting rate on those borrowed funds is 5.76%. Here are some warning statements in the illustration: “This causes Alternate Loans to be significantly more volatile than Standard Loans.” “We reserve the right to increase the current loan interest charge rate. However, we will not increase the rate to more than the guaranteed rate (7.50%).” “Illustrating a hypothetical indexed interest rate greater than the loan interest rate over an extended period of time may not be realistic. If you change the assumption, the impact on your policy could be quite dramatic, even resulting in a policy lapse.” Nonetheless, this illustration plows ahead with a lifetime of borrowing arbitrage assumed – the more and earlier you borrow, the better the long-term returns will be!

Do IUL Policies Have an Investment Advantage Over Whole Life Policies?

It’s hard to see how an IUL policy would have an investment advantage over the general account of a quality mutual insurance company over the long haul. (UL policies in general can provide better access to policy cash value for those that are focused on distributions, but there is nothing unique to IUL policies – other than commonly understating the risk of borrowing by assuming that a positive arbitrage will last forever.)

Mutual insurance companies often invest 10-20% of their assets in equities or equity-like instruments, with the rest in bonds. That investment profile generally compares favorably with the IUL profile, where an estimated 95% of assets are invested in bonds with the rest in hedging instruments. Working backwards, one would have to believe that an IUL company could earn an incredibly high return on its options portfolio in order to prop up the returns to reach illustrated levels. Which begs a couple of questions: Why don’t companies simply skip messing around with insurance and just focus on the amazing returns seemingly available through these hedging strategies? And why don’t these insurance executives leave and form their own hedge funds that utilize these same investment strategies? Proponents of IUL policies say that these strategies are more realistic for a life insurance company to implement (as opposed to an individual investor or hedge fund) because of the tax ramifications of so much investment activity, but even taking that into account, something doesn’t add up.

It’s also helpful to understand that the current illustration rules dramatically favor IUL policies over whole life policies. IUL policies are permitted to back-test a current cap rate (which as I’ve pointed out above is far higher than a comparable cap rate on an indexed annuity and might be unsustainable) over a period of time when not only did these policies not exist but the options that would be needed to provide the hedging did not even exist. So these companies are saying that if we would have been able to offer this product over an “average” 25-year period and had we been able to set the cap rate at its current non-guaranteed level throughout that entire period, then these are the average credited rates we would have seen. On the other hand, whole life illustrations do not get to use a historical average of ACTUALLY credited rates; instead, those illustrations are limited to show only what is currently being credited on those policies, which in many situations are at 50-year lows. If whole life illustrations could simply use the benefit of the same historical averaging period that IUL policies are taking advantage of, then in many situations the average ACTUAL crediting rate for WL policies would be equal to or greater than the HYPOTHETICAL crediting rate for IUL policies that didn’t exist with cap rates that are artificially inflated.

[This is not to suggest that WL policies should be allowed to illustrate on this basis but rather to highlight the absurdity that IUL policies are able to illustrate on the basis of hypothetical lookbacks. Perhaps if cap rates were guaranteed then this lookback approach might have more credibility – but insurance companies have the ability to unilaterally lower the cap rate at their discretion all the way down to extremely unattractive levels.]

Much attention is given to the attractive (hypothetical) crediting rates available from an IUL policy. However, less attention is given to the fact that those crediting rates are largely irrelevant until many years into the policy. That is, even if great crediting rates materialize, the actual rate of return on monies invested is very poor for many years due to the overwhelmingly large expenses in the early years of a policy. For instance, in the illustration described here, it’s not until the 10th year that the accumulation value on the policy eclipses the sum of premiums paid into the policy. (And that is years earlier than other IUL illustrations I have reviewed.) So who really cares what the crediting rate (or cap rate) is in the first 10 years of the policy? What’s going to drive the long-term value of the policy is the level at which future cap rates will be set.

Premium Financing – Even More Leveraging

IUL is a favorite product with premium financing arrangements, where large sums of money are borrowed at currently low rates and invested into IUL policies.

One cannot overstate how attractive these arrangements look on paper. Consider that the illustration described here is already very aggressive and highly leveraged (because of the permanent borrowing arbitrage that was assumed with internal borrowing on the policy) with an extremely attractive illustrated return of 7.6%. One way we can make that look even better is to use premium financing to provide even more leverage.

Simply put, if we can borrow at less than 7.6%, then we are going to create value if we turn around and invest that money in a vehicle that is earning 7.6%.

The premium financing borrowing rate in this example is assumed to start at less than 4% and then steps up close to 5% over a 10-year period. The arrangement I reviewed assumed that the premium financing loan would be paid off with policy values after 16 years.

Not surprisingly, there are massive risks in this proposal that were not fully fleshed out. It’s not hard to envision a scenario where the perceived positive arbitrage turns into a negative arbitrage with dramatically negative consequences.

Conclusion

When you put it all together, many IUL illustrations resemble a house of cards. When you inject heavy internal borrowing into the illustration (using a favorable and potentially unsustainable arbitrage assumption), you now have another house of cards put on top of the first house of cards. (And for those that are so inclined, you can add another house of cards by introducing premium financing into the mix.)

Nonetheless, I do think it is possible for an IUL policy to be an acceptable purchase if the assumptions are set properly. I generally recommend staying away from policies that lack transparency (such as relying heavily on non-guaranteed bonus elements). I also suggest using an illustrated credited rate of 5% for evaluation purposes. (But recognize that with some policies, when they state that they are crediting 5%, they might be assuming a much higher effective crediting rate, based on black box bonuses built into the illustration.)

Finally, if you are going to assume borrowing internally from the policy, you should use a loan rate option that links (perhaps with a spread) the credited rate on borrowed funds with the loan rate charged on those borrowed funds – this lowers the risk dramatically and prevents reliance on an unsustainable arbitrage relationship. (Even if you intend to use the borrowing option that makes the credited rate on borrowed funds independent of the loan rate, you should initially ask for the other illustration to be sure that it’s not simply the loan assumptions that are driving the attractiveness of the illustration.)