#227 | The New Pressure on Caps

It should come as little surprise that rates and caps continue to fall in Indexed UL products. Just one look at the IUL Benchmark Index will tell you why – caps are expensive (4.9% for a 10% cap) and yields are sub-3% for new money based on corporate bond yields. The dark skies over Indexed UL continue to loom and even rising equity prices and general optimism can’t slow them down. Equity volatility remains elevated, volatility skew is as nasty as ever and earned rates are lower now than they were in the height of the crisis.

Recent moves by Prudential and Securian highlight just how challenging today’s environment is for Indexed UL sellers. Prudential has been slicing caps since March. Index Advantage caps have fallen from 10% for the March sweep to 8.5% for the June sweep. Founders Plus, Prudential’s hot-selling death benefit oriented IUL, has seen its cap decrease from 9.5% for the December sweep to 6.5% for the July sweep. The curious thing about these rate changes is that both products share the same 4.6% fixed account rate. In theory, that should mean they have similar caps, but that’s never really been the case. Why not? Because, of course, there’s always more to the story. Founders Plus has lower policy charges (and lower compensation) than Index Advantage. It appears that, in some way at least, those policy charges are going to subsidize the cap in Index Advantage because that’s what matters the most for the competitive position for the product. Founders Plus, on the other hand, has other benefits that are not related to the cap, particularly its secondary guarantee provision.

But the more interesting cap decrease is Securian, which announced earlier this week that it would be dropping its S&P 500 point-to-point cap to 9% for the July sweep date. This move follows closely on the heels of the drop from 10.5% to 9.75% for the April sweep date. To put these rate reductions into practical terms, the illustrated rate on Securian’s products will have dropped from 6.4% to 5.7% in the space of 6 months and, need I mention it yet again, a very long fall from the 9%+ illustrated rate originally shown back in 2007 when Eclipse IUL had a 17% cap.

I can’t say that I blame Securian for dropping their cap yet again. In fact, I think it’s unequivocally the right move. Securian’s 10.5% cap trade back in March cost them a whopping 5.72% on $250 million of notional value. Assuming their option budget was right around 4.75%, which was the average cost of the 10.5% cap in the preceding months, Securian took a $2.5M loss on that trade – and they were thanking their lucky stars it was only that. Why? Because March has a relatively low account value being hedged. If the same trade had occurred in January on the same economics, they’d be staring down a $4 million hole from one trade. Not deadly, but it still stings, especially if it happens for more than a few months. As I’ve written before, the challenge of having a huge Indexed UL block is that, eventually, you have to price in real-time. You just can’t afford to do otherwise.

But that’s actually not the interesting thing about the Securian trade in March. I have access to institutional implied volatility data (courtesy of the work I do on annuity product development) and can therefore look back in time at how options would have been priced on a previous date. Usually, the estimates are very close what life insurers actually pay for their options – but not in March. Based on data from both Securian and Life of the Southwest, March was a very odd month. Take a look.

SecurianSecurianSecurianLSWLSWLSW
EstimatedActualGapEstimatedActualGap
January4.61%4.65%-0.04%4.50%4.49%0.01%
February4.64%4.64%0.00%4.55%4.55%0.00%
March5.36%5.72%-0.36%5.02%5.28%-0.26%
Securian has a 10.5% cap and LSW has a 9.75% cap

As you can see, the gap between the estimated price of the call spreads is vanishingly small in January and February, which you can reasonably chalk up to a rounding or vol reporting error. March, however, had an enormous gap. For Securian, the data would have predicted a cap price of 5.36% but the actual cap price was 5.72%, meaning that Securian had to fork out 0.36% more than the “fair price” of the option. Life of the Southwest similarly had to fork over an extra 0.26% more than the data would predict. What in the world is going on here? I’m not exactly sure, but I think I have a pretty good idea of what might have happened.

First, some context. Life insurers trade with bank counterparties in what’s referred to as an over-the-counter (OTC) trade, which is different than the exchange-traded options that normal folks can buy in their brokerage accounts. The advantage of OTC trades is that you can package them exactly the way you want them, which allows life insurers to perfectly match their indexed crediting liabilities with their purchased hedges. The bank then takes the option premium and dynamically hedges the exposure. Dynamic hedging is the process by which the bank attempts to replicate the exposure of the option by rapidly buying and selling shares of the underlying asset (in this case, S&P 500 futures). In theory, dynamic hedging perfectly offsets the liability created by the option. But in practice, very volatile markets coupled with thin liquidity can create situations where dynamic hedging can’t keep up with the value of the option. These situations present major risks for, say, a bank selling options to a life insurance company.

Now, to what I think happened in March. Markets were obviously extremely volatile, particularly towards the middle of the month when IUL companies were come to market to hedge. For dynamic hedgers, this means any trade executed could be a loser if it can’t be quickly and effectively executed. Therein lies the problem. This corner of the options market is very thin. Basically no one but life insurance companies hedging annuity and life insurance products buys 12 month at-the-money S&P 500 call options (which should tell you something, but that’s a conversation for another day). To make matters worse, they all come to market on nearly the same day. That’s a huge amount of demand for a relatively thin level of supply because of the inherent risk of dynamically hedging in a volatile and illiquid environment. So what happens? Well, obviously, the price goes up.

Why didn’t my estimates also see a price increase? Because exchange traded options are small and the risk is distributed across the exchange, but OTC options are huge and the risk is concentrated at the seller. Take Life of the Southwest’s trade for $180M of notional with Barclays on 3/20 for $15.6 million – that’s no small chunk of change. Why did LSW have to pony up an extra 0.26% to execute that trade? Volatility and illiquidity. They have to compensate their counterparty for the risk that the hedging will be less effective. And why did Securian have to pay an extra 0.36% on 3/19 for their trade with Barclays? Because they showed up with $250 million to hedge. It’s also probably not a coincidence that Barclays was the counterparty of choice for both LSW and Securian. That could very well indicate that other counterparties were less willing to step up to the plate when the world appeared to be falling apart, which meant Barclays had more pricing power than usual and could take the margin and the risk.

By this point, everyone should be well aware of the fact that interest rates, volatility and volatility skew have clear and attributable effects on the price of the options used to hedge indexed interest crediting. But March gave us a new threat – liquidity. As both FIA and IUL blocks have increased in size, hedging them necessarily relies on a deep enough market of counterparties to handle the demand. But it appears that in March, life insurers had to pay through the nose to get their trades executed in the midst of what felt like the end of the world, exactly the moment when they most needed to offload the risk. I’m guessing that this caught life insurers by surprise because back in 2009, FIA and IUL blocks were tiny compared to what they are today. The fact is that life insurers are increasingly dominant in the corner of the options market they rely on to hedge their products – and they’re paying for it.

So what? There’s a big implication. If life insurers have now realized that they are pushing up against liquidity constraints in the options market, then they need to set rates with the recognition that in the next crisis, they’re going to take a beating when they actually try to execute the hedges. Therefore, the ratesetting process at life insurers may need to start taking into account the slippage that they’ll experience when they try to hedge huge and ever-growing blocks of FIA and Indexed UL products in a highly volatile market. That should mean lower, but hopefully more stable, caps – or life insurers will simply ignore the risk and then have to pay the bill when it comes due.

7/16/20 Update – I confirmed that my suspicions were accurate with a top derivatives counterparty to life insurers. The market in April and March were extremely chaotic and counterparties were not stepping up to trade with insurers because there was too much risk in offloading the position, exactly as I’d suspected. Any counterparty trading in that period of time was pricing for a significant margin to offset the risk. Every life insurer trading any hedges in that period of time paid through the nose. The lack of counterparties and extreme pricing apparently came as a huge shock to the life insurers. So this is a real risk and it really is going to be a problem going forward in volatile markets, especially as the blocks of FIA, IUL and RILA get bigger.