#300 | Get Ready for AG 49-B

green and yellow printed textile

At the NAIC Life Actuarial Task Force (LATF) virtual meeting in early December, Fred Andersen (MN), the chair of the IUL Illustration Subgroup, read a prepared statement asking for comments on a particular nuance related to AG 49-A and the use of so-called “volatility-controlled funds.” His statement doesn’t mince words in saying that “our research found that some insurers are or are potentially planning…to only use a portion of the hedge budget…on providing the upside potential and using the remainder to fund a fixed bonus to the policyholders. In some insurers’ minds, this allows them to illustrate volatility-controlled funds plus the fixed bonus more favorably than a traditional, capped S&P 500 index.” You can find the full statement here.

It should come as a surprise to no one – and certainly not subscribers to The Life Product Review – that regulators have opened an inquiry into the effectiveness of AG 49-A. The guideline was fatally flawed from the start. During the drafting process, I signed a letter along with 12 other non-carrier insurance folks that listed 5 ways that life insurers could maximize illustrated performance under AG 49-A that were not consistent with the spirit of the guideline. Item #1 was the combination of proprietary indices* and fixed interest bonuses.

Since AG 49-A has been adopted, we’ve seen a proliferation of these types of accounts in the Indexed UL market. The effect, as I’ve written previously, has been astounding. Illustrated income can be augmented by as much as 65% for some products simply by switching the allocation out of the base S&P 500 account and into a proprietary index account with a fixed interest bonus. Life insurers have clearly found a simple, straightforward and cheap way to illustrate outsized performance. The question now is what can and should be done about it.

Let’s start with the second part – what should be done? At first blush, it looks like life insurers have simply exploited a vulnerability and the vulnerability needs to be closed. That’s certainly the argument that some regulators and some insurers will make, but Fred’s statement already hints at the other side of the story when he writes that “in certain subsets of history, the uncapped volatility-controlled funds performed better than capped S&P 500 funds.” If those kinds of results have actually been realized for real customers, then what’s the problem with illustrating them?

We’ve been down this road before. Recall that AG 49 was created in part because life insurers were using exotic crediting strategies that showed fantastic hypothetical historical backtested performance, often north of 9% and as high as 11%. How did those strategies work? By combining extremely strong historical index performance with relatively cheap options. These strategies were, essentially, natural diamonds. Carriers had to find them out in the wild, dig them up and then cut them into something beautiful.

Modern proprietary indices are lab-grown diamonds. They’re controlled, targeted and engineered. The net result is a more perfect solution that is far more accessible. And just like in the real diamond market, the indexed crediting landscape is now flooded with the engineered, lab-grown solution – more than 110 of them at last count. These indices are simply more perfect versions of what came before.

Both can be used to exploit the vulnerability in AG 49-A. As we’ve already seen, some life insurers are using both proprietary indices and exotic crediting strategies with fixed interest bonuses to augment performance. The issue with AG 49-A isn’t about proprietary indices – it’s about any crediting strategy that, based on the hypothetical historical lookback methodology, generates high returns with low option costs. Any strategy works as long as it has a lookback high enough that the carrier can lower the option budget and deploy the savings into a fixed interest bonus that can be tacked on top of the maximum illustrated rate. The net result is an account option that illustrates better than anything else under AG 49-A without increasing the total cost to the insurer.

The answer, then, is that yes – something should be done about AG 49-A. The vulnerability should be closed. Life insurers should not be able to use any indexed crediting strategy that illustrates higher option profits than the S&P 500 to illustrate better than the S&P 500. That was the goal of AG 49 as it targeted exotic crediting methodologies. That was the goal of AG 49-A as it targeted multipliers and buy-up caps. Now, it seems, the next logical step is to target AG 49-B to eliminate the illustrated benefits of account-specific fixed interest bonuses.

That brings us to the question of whether or not anything actually can be done. As I’ve written in previous articles, as AG 49-A took shape, it began to incorporate certain suggestions from life insurers that – in retrospect – look as though life insurers were writing the vulnerability into place and solidifying their inevitable defense of what they were planning to do afterwards. Just take a look at some of the new language and concepts used for AG 49-A:

  1. Switch to a single Benchmark Index Account while maintaining the ability for the life insurer to use a hypothetical BIA as the maximum rate for the product (Sections 3D and 4A)
  2. Applying the 145% NIER illustrated rate restriction only to the BIA, not all accounts (Section 4B)
  3. Requiring the actuary to use the lookback methodology for a non-BIA account (Section 4Cii), whereas previous language was based on “fundamental characteristics”
  4. Allowing the life insurer aggregate indexed accounts for illustration actuary testing (Section 5A)
  5. Not specifically addressing fixed interest being added to the illustrated loan arbitrage (Section 6)

Taken together, these 5 things provide a stout defense for what life insurers are doing with fixed interest bonuses. The single BIA requires using the maximum illustrated rate for all accounts, even accounts that would have arguably received a separate BIA with a different (and lower) option budget under the original AG 49. Not restricting any account other than the BIA by the 145% factor allows other accounts to illustrate “option profits” well beyond 45%, which are now easily defended by the required lookback methodology to determine the rate. And now the life insurer has plenty of air cover for setting rates with different levels of profitability or supportability as long as the aggregate product is supportable. And, finally, any fixed interest bonus can be tacked on to not just the illustrated rate but also illustrated loan arbitrage, which every carrier knows is the real honeypot because virtually every Indexed UL illustration shows income from indexed loans.

Removing the vulnerability, therefore, requires undoing some of the new language that was introduced into AG 49-A. There are a wide variety of ways to do it, each one generally relating to the previously modified code sections:

  1. Require that each indexed account be restricted by the illustrated rate on its own BIA based on the account’s Option Budget
  2. Ditch the 145% NIER restriction applied to the BIA and, instead, calculate the actual ratio used by the BIA at its current rate and apply that ratio to the Option Budget for each indexed account
  3. Allow lookbacks only for the BIA. Every other account must use the original language – “The Annual Rate of Indexed Credits [reflects] the fundamental characteristics of the Index Account and the appropriate relationship to the expected risk and return of the Benchmark Index Account.” – plus additional language detailing that the actuary must also take into account the volatility of the underlying options
  4. Requiring the insurer to test and report each account individually, including the BIA, while using the “option profit” ratio calculated based on the current BIA illustrated rate in 4A and the Option Budget/NIER
  5. Eliminate account-specific or allocation dependent fixed interest bonuses from illustrated loan arbitrage

I can say, with confidence, that a few of these changes, or maybe even all of them together, would go a very long way in eliminating the vulnerability in AG 49-A for proprietary indices and fixed interest bonuses. They’d also go a step further towards eliminating some of the less dramatic ways that life insurers are using proprietary indices to juice illustrated performance, particularly the idea that “profits” from proprietary index allocations can be used to prop up S&P 500 rates that would otherwise be unattainable.

What I can’t say with confidence, however, is that all of the vulnerabilities in Indexed UL illustrations are gone. Think about it this way: imagine you’re a cyber security specialist and you’re being asked to certify that a program is completely secure and impossible to hack. How many lines of code could that program contain before you lose your ability to make that certification? A thousand? Ten thousand? A hundred thousand?

This exact test was performed back in the 1990s at a government institution and professional, elite-level cyber security specialists couldn’t spot bugs in just 2,000 lines of code. Windows 11 has approximately 50 million lines of code. Can anyone ever be entirely sure that Windows is completely secure? Of course not. Modern enterprise software is simply too big and too complex for anyone to guarantee complete security.

The same goes for Indexed UL illustrations. The products and the regulations are simply too complex to guarantee that this round of regulation will, in fact, be secure. And to make matters worse, life insurers have a massive financial incentive to find vulnerabilities and exploit them. But how do you turn a vulnerability into an exploit? You need a vehicle to do it.

Read up on some of the biggest hacks in the last decade and you’ll see a common theme – the user almost has to do something to allow the hacker to take advantage of the vulnerability. The vulnerability needs a vehicle to be exploited. That’s why your IT Admin is constantly telling not to click on suspicious emails. That’s why core infrastructure is (supposed to be) “air gapped,” meaning it has no connection to the internet. That’s why secure institutions don’t use or allow USB drives, some literally putting glue in the ports. If you can never guarantee that the software itself is perfectly secure, then the only permanent solution is to remove the vehicle for the exploit. Without the vehicle, a vulnerability can’t sink its teeth into a system.

This brings us around to the distinct possibility that AG 49-B will be unlike the guidelines that preceded it. Both AG 49 and AG 49-A made a point to preserve the hypothetical historical lookback methodology (HHLM) as the basis for illustrated rates in Indexed UL – a methodology, it should be remembered, that was developed by the life insurance industry for specifically for use in Indexed UL illustrations. The HHLM is the vehicle for each and every one of the vulnerabilities that have been exploited in Indexed UL illustrations, without exception. It is what created the environment of exotic indexed strategies illustrating 10%+ that led to AG 49. It is what created the illustrated benefits of multipliers and buy-up caps that led to AG 49-A. It is what has led to the current vulnerability of combining proprietary indices and fixed interest bonuses that will most likely lead us to AG 49-B.

We’ve already been through two painful rounds of crafting regulation for Indexed UL illustrations and we’re about to go into a third. It should be obvious to everyone, at this point, that there is literally no way to guarantee that the regulation is secure and that there are not vulnerabilities. It seems as though there is also no way to guarantee that life insurers will not continue to try to hack the system. The only long-term solution, then, is to remove the vehicle for the vulnerabilities to be turned into exploits. If you remove the vehicle, then you remove the exploit.

If that’s what happens with AG 49-B, then quite literally everything we’ve come to understand about how Indexed UL “should” illustrate would be upended. The HHLM would be gone in favor of a fair-market valuation of the indexed exposure – in other words, it would illustrate consistently with how option strategies are valued for every other financial institution the world over. And, similarly, life insurers would have the opportunity to show the mechanics and potential upside and downside scenarios elsewhere in the illustration, without ledger values. That would also be consistent with how option strategies are discussed outside of insurance applications.

Parts of the Indexed UL market will, understandably, howl and scream if this happens. But is it really so bad? Think about where we are today. Currently, illustrated rates are nearing 5% anyway because of high option prices and falling portfolio yields. An 8% Cap at Pacific Life, for example, would switch from a 5.18% illustrated rate to something like a 4% illustrated rate. Is it really worth burning the house down over 120bps? Recall that during the original AG 49 discussions, most companies had 12% caps with illustrated rates of 7%. Falling caps have inflicted far more damage thanks to the HHLM than if the illustrated rates had been set appropriately from the beginning using a fair-market value methodology.

And that’s not even the worst of it. Right now, life insurance producers and distributors are left to fend for themselves (or subscribe to this publication) to figure out how they’re being duped into thinking they’re illustrating one thing when they’re actually illustrating something entirely different. The Indexed UL market, as it stands today, is not sustainable. We cannot continue to move from vulnerability to vulnerability, leaving behind us a trail of disappointed consumers who are collateral damage.

This brings me back to the analogy between Indexed UL regulation and cybersecurity. The pace of technology continues unabated because it generally appears to do more good than it does harm, despite the incredible security risks that more technology brings. The same goes for Indexed UL – it is a product that, fundamentally, does more good than harm. The core idea behind the product is solid. People like the idea of downside protection with upside potential. There is something powerful and intuitively appealing about the story.

But, tragically, the history of Indexed UL is marred by aggressive illustration practices. If we want a world that doesn’t destroy itself with its own technology, then we must be vigilant. We must remove both the vehicles and the vulnerabilities. That goes for technology and it also goes for Indexed UL. If we want a product category that is fundamentally good to stay, to proliferate and to serve clients well, then we must remove the vehicle that turns vulnerabilities into exploits. The HHLM – and its bizarre assumption of perpetual option profits – must go. For good.

* I use the term “proprietary indices” because I think it’s a more accurate description than “volatility-controlled funds.” The strategies in IUL/FIA are not funds, they’re indices, and most of the ones being used are proprietary to the issuer and may or may not have an explicitly stated volatility target (at least one notable and very popular proprietary index doesn’t). But as this article notes, the issue isn’t actually proprietary indices or “volatility-controlled funds” – it’s any strategy that backtests exceedingly well with a low current options cost. That just happens to be exactly what proprietary indices are designed to do.