36. Life Settlements (part 2)
Securitization of life policies, STOLI fraud and questions around market size
We discussed in the last post that stranger-originated life policies (STOLI) are illegal… but stranger-owned policies are not.
If a policy is taken out with the intent of selling to an investor (often accompanied by fraud on the policy application), that is illegal and will be contested by the carriers.
Naturally, the life insurance industry contributed to the emergence of the life settlement industry. Life insurance has a long tail in history. In the 15th century, merchants in Genoa and other Mediterranean cities needed to purchase insurance on their ships and ship’s cargo.
Due to the expense of insuring one’s ship, merchants became creative and began taking out life insurance on their lives and naming their lenders as beneficiaries.
Lenders were OK with this because they would receive the death benefit from the life insurance on the merchants. The economic function was similar to if the cargo itself were insured.
Since then, life insurance has evolved, and so has the law that governs it. The foundational purpose of life insurance is to protect against the loss of the insured and provide the beneficiaries with monetary relief in the event the insured dies. This often helps mitigate death / funeral costs, loss of a household’s income provider, etc.
A term policy is what most might think of when they consider life insurance. It’s a basic policy which provides coverage for a set term with a set premium level. Once the term is up, the insured can renew the policy or let it lapse.
Other types of policies have since become mainstream, including whole life, universal life, variable universal life, etc.
The simple way to think about the evolution of life insurance is that it has morphed from a simple way to receive coverage over the risk of one’s death to becoming a savings and investment vehicle (as often many of these types of policies provide a savings vehicle and mechanism to choose investments).
Overview of Life Insurance Products
Let’s do a quick summary of difference products within this space.
At a high level, you have i) permanent life insurance and ii) term life insurance.
Permanent life insurance typically carries a cash value and includes whole life, universal life and variable universal life.
Most permanent policies let you build cash value, which can be accessed to pay for emergencies such as medical expenses or even college tuition. Anything drawn on the cash value usually nets against the death benefit that is ultimately paid.
Stepping back, there are a few key characteristics between different types of life insurance:
how long does coverage last?
what features and benefits are offered?
how are the premiums structured?
Whole Life - premiums are locked in at the time of purchase. Whole life typically starts as more expensive than term, but in the long run may prove to be a better value.
Universal Life - also known as adjustable life… you can modify premiums and the death benefit while the policy is active. Typically lower premiums than whole. The cash value may be in addition to the death benefit.
Life Settlements
As we previously discussed, the insurable interest doctrine became part of insurance law to counteract life insurance encouraging murder.
However, in the 1980s, a use of life insurance emerged that had not yet been fully considered.
The 80s were a time when AIDS swept across the U.S. People with AIDS were suddenly in precarious financial and medical circumstances.
Viatical settlements developed to provide financial relief to AIDS patients, which could be used to cover medical expenses. In turn, the investor covers the premiums, pays a cash payout to the insured and becomes the policy’s beneficiary.
The viatical industry eventually slowed as medical advances allowed AIDS patients to continue working and live longer. In the heyday of the viatical industry emergence, policyholder’s life expectancy was often estimated to be anywhere from several months to less than 2 years (when selling to the investor).
As the life expectancy expanded, that eroded the economics to the investor, which fueled the decline of viatical settlements.
But by this point there was a whole ecosystem of brokers, consultants, insurance agents, and they all sought to keep their fee streams alive by finding other routes in the life settlement space, despite the shifting landscape within the viatical sub-segment.
Ultimately, the industry expanded to seniors with other forms of illness. And then it expanded to those without illnesses who simply wanted / needed to sell their policy (that is dubbed life settlement rather than viatical settlement).
And next came STOLIs.
Legal life settlements and STOLIs share many similarities. Mechanically, the deals are structured identically. Both fail to meet the insurable interest provision. But where STOLIs turn to fraud is when the policy is originated with the intent of being sold to an investor.
An argument can be made that life settlements should be permitted only as an exception to insurable interest when the insured is suffering medically or financially… and can prove so.
Let’s summarize key differences between viatical and life settlement.
Viatical
Life expectancy: a few months to 2 years
Policy value: ~$80k
Life Settlement
Life expectancy: 11-12 years
Policy value: ~$1mm
Securitization of Life Settlements
Securitization took hold in the 1960s and 1970s. Banks looked to diversify their portfolios and began selling mortgages to investors who couldn’t originate them.
Instead of selling loans individually (which itself offers a host of challenges), the bankers pooled mortgage loans together into bundles, thereby spreading the risk of defaults across the entire loan package.
Then, the bankers would issue securities, such as bonds, backed by the cash flow from the loan-package mortgage payments.
Banks now made money from mortgage underwriting but also from securitizing and selling mortgage-backed securities (MBS) to investors.
The next innovation was tranching - dividing the bundles into different levels based on risk / return, so that defaults would hit the riskiest level first.
After 2008, when MBS blew up, bankers wanted to find another way to replace this business. So, much like the viatical industry turned to broader life settlements, these bankers turned to bundling life insurance policies into bonds… just like they did for mortgages many years earlier.
In securitizing these policies, there are many factors to consider as it relates to diversification and risk. Same goes for analyzing these securities.
If many underlying policyholders have illnesses, you’d want to spread the type of illness such that if a cure was found, the value of the bond wouldn’t collapse (as life expectancy would presumably then be pushed out much further). It’s an “icky” risk to talk about, but it’s material to this security type.
Further, you would want carrier diversification within each bond. Meaning, say, you put a cap of 20% of carrier exposure for each bond such that one carrier can’t hold more than 20% of the bond’s policy face value.
This reduces credit risk amongst the underlying carriers who serve as the counterparty to each individual policy. If a carrier blew up, you wouldn’t want them being the counterparty to, say, 50% of your face value…
Other risks include
Insurable Interest - the carriers would fight to avoid paying the death benefit if it goes to an investor without insurable interest. So expect litigation risk.
Sample Size - need a large number of policies within each bond as statistics about the insured are unlikely to be sufficiently credible with a small pool (say, less than 1,000 or 500 policies). Rating agencies say ~300 policies should be sufficient, but idk… that feels small.
So this market is quite interesting from many perspectives.
If you think about this from a carrier’s perspective, life settlement businesses are the enemy and impact not only the amount of death benefits you pay but also how you price new policies with declining lapse rates.
From an investors standpoint, it seems to offer investment risk uncorrelated to the broader market. There’s also asymmetries that exist to where investors can take advantage of mispriced policies.
And from the insured perspective, settlements allow you to get out of a policy for more than the surrender value or simply letting it lapse. On the extreme, some are enticed to participate in STOLI fraud to receive further lump sum compensation…
How about from the originator’s perspective? Meaning the one who’s securitizing these instruments.
Well, we mentioned how, much like mortgages, it allows the originator to capture additional fees by securitizing and selling these products.
But, one headwind is the capital required to enter this market as an originator. If you want to securitize life policies, you have to first own a bunch of policies.
You can accomplish this in 2 obvious ways.
you go out and buy a bunch of policies from the market. Let’s say 500 is the number to be properly diversified in one bond. You need 500 policies. Assume the face value of each policy is $1mm. That’s $500mm.
How do you fund it? Something like warehouse lending will do. Effectively use a revolver to acquire each policy and paydown the revolver when you sell the new securitized product
the alternative here is if you already own the policies! If you have the policies on your balance sheet, you then don’t need to finance the expensive purchase of 500 policies. You can simply package your policies up and effectively de-risk by selling the bundles to investors, and taking additional fees in doing so.
In 2009, AIG securitized 3,400 policies that it held on its balance sheet for $8.4bn
Pricing of Life Insurance Policies
Ok, we’ve gone in a few different directions throughout this post. Let’s wrap up with a discussion on how life settlements impact policy pricing.
We left the last post discussing how lapse rates effect pricing of policies. Because lapse rates impact pricing, those who can afford to keep paying the premiums the longest benefit from the lapsing of others.
Think of this as a subsidy.
Non-investors who lapse on their policies subsidize the investors who have deep liquidity and can afford to pay the premiums for longer.
It’s a subsidy because the lapsing policyholders allow the carriers to lower the pricing which benefits everyone who takes out a policy (especially those who ultimately receive the benefit).
Life insurance polices are priced based on a few key determinants
actuaries have a scary assessment of when you are likely to die… and so they can price your risk accordingly
the carrier has a very well informed estimate of when you are likely to stop paying premiums… and let your policy lapse
Let’s say that a carrier thinks there is a 30% chance that you let the policy lapse. The carrier will then adjust the price of the policy to reflect this chance… in simple terms let’s say it’s a 30% discount to the price otherwise stated if the lapse probability were 0%.
This is where there is an information asymmetry. On all of these policies, the carrier thinks there is a 30% chance of policy lapse. But the investor who buys the policy knows that this is really 0%… they bought the policy with the full intention of paying premiums until they receive the death benefit… and have plenty of liquidity to do so.
So, the 30% pricing discount translates to the investor’s profit, stemming from the information asymmetry inherent in this market. And that doesn’t include how investors purchase policies for pennies on the dollar.
Conclusion
How big is the life settlement market?
Here are some stats
2017: $2.8 billion face amount / 2,027 policies
2018: $3.8 billion face amount / 2,587 policies
2019: $4.4 billion face amount / 2,878 policies
2020: $4.6 billion face amount / 3,241 policies
2021: $3.9 billion face amount / 2,937 policies
2022: $4.5 billion face amount / 3,057 policies
So the market recovered in 2022 with $4.5bn of face value volume in the secondary market, compared to $3.9bn in 2021.
The total amount paid for 2022 policies amounted to $824mm, compared to $715mm in 2021.
So, in 2022, secondary buyers acquired $4.5bn of face value for $824mm… that’s 18 cents on the dollar!
In 2021, it cost $715mm to acquire $3.9bn of face value, also 18 cents on the dollar.
Ok, so the takeaway might be that this market is seeing some healthy growth after the 2021 shock.
But, my question is why aren’t these numbers higher? Why isn’t the settlement market 10x the size? 100x?
Between 1990 and 2010, $30.8 trillion of new life insurance coverage was issued in the U.S. Around $24 trillion of in-force coverage was dropped during this same period. That equates to an ~80% aggregate lapse rate!
$24 trillion lapsed over 20 years, which means that ~$1.2tn lapsed per year, on average. $1.2tn divided by the face value of $30.8tn is a ~4% lapse rate per year. Simpler way to do this is 80% divided by 20 years, but oh well.
That matches with other data points I saw around 3% lapse rates for traditional whole life and 4.6% for universal life. Variable life and variable universal life lapse around 5% per year.
25% of permanent insurance lapses in the first 3 years. 40% lapses within the first 10 years. And almost 85% of term policies fail to pay a death claim… similar to our 80% lapse rate above.
Ok, call it a 4% lapse rate per year. Statista shows that $3.3tn of aggregate face amount of life insurance was purchased in the U.S. in 2022.
If we assume 4% of 2022’s new policy amount lapses, that’s $132 billion of lapsed policy value per year.
The way to think about this is that 2022 addition to policies in-force of $3.3tn increases the TAM of the settlement market by $132bn.
So, the real question is, if $1.2 trillion of policy value lapses each year, why is the settlement market only doing $4.5 billion per year?
That’s 40 basis points of market share against the total addressable market…
We’ll dig into this market size question and lapse-based pricing in more detail next time.
Thank you for reading.
++
Until next time.
John Galt