#7 | IUL – Higher caps, lower charges?
A truism in the Indexed UL market is that higher caps are provided by higher policy charges. Carriers generally can’t pick up more than a few basis points of efficiency in their hedging platforms despite what Allianz would have you believe. Yields across carrier general accounts are more or less equal. Mortality, overhead expenses and targeted profitability are also largely similar. The only variable left to tinker with, then, is the policy charge structure. Increasing policy charges allows for more of the investment return to be passed to the policyholder because the carrier essentially accepts no asset spread (which turns into a larger options budget and higher cap) in exchange for other policy charges. The graph below shows a strong, direct relationship between policy expenses over the first 10 policy years and the average hypothetical historical illustrated rate which reflects the relative value of each policy’s cap/floor combination.
So, in short, be prepared to accept higher policy charges if you want a higher cap. That’s the story I’ve been telling for a while now and felt fairly comfortable with until I really started to think about it. Policy charges in the first ten years are indisputably correlated with higher caps. But what about after ten years? Most of these contracts have large, fixed charges that burn off between 5 and 10 years into the policy. Furthermore, overfunding means that the cash value will likely match or exceed the originally underwritten face so the amount of COI charges versus account value will be miniscule. Charges are a big differentiator in the first ten years but not nearly as much in the next ten years. And if policy charges are the reason for the higher cap, is it reasonable to assume that the carrier can maintain a relative advantage when the effect of charges is limited?
A nice, neat story would be that products with high policy charges in the first 10 years also have high policy charges in the next 10 years so that we could assume maintenance of the current cap differential. But, as the graph below shows, that’s simply not the case.
A linear Least Squares Regression line for the data has a slight positive slope and a paltry 0.08 R-Squared. In statistical lingo, that means that the LSR can predict only 8% of the data shown. In other words, the points are statistically scattered around without much pattern. If we extend the analysis to all 45 years (in this case), the correlation is only slightly better with a 0.14 R-Squared. For the sake of comparison, the R-Squared figure for the 10 year charge and cap graph is 0.49, signifying a much tighter pattern. It appears that carriers have consensus on how 10 year charges should compare to caps and not nearly so much after that.
Why might a carrier want to have higher policy charges early on and lower charges later? The actuarial argument might have two prongs. First, higher early policy charges means that the carrier takes its profit (and then some) and won’t need to charge nearly as much after that to provide a higher cap than its peers. Account value will be diminished, NAR will be higher and therefore the carrier will be able to extract higher COI charges. But that difference is marginal, at best. COI charges for young clients are miniscule and the dollar difference in the annual COI deduction between $950,000 of NAR and $900,000 of NAR is probably less than a couple hundred dollars against $50,000 of account value. Second, policyholders won’t overfund as illustrated and will likely dump the policies after 10 years or so. I don’t buy either of these arguments. If the logic was consistent, there would a clear correlation between caps and charges in years 11-20. The proof’s in the pudding – there isn’t.
The cynic’s argument is that charging early for a cap shown for the life of the contract is a type of illustration arbitrage. The illustrated rate is fixed at a certain level assuming the cap never changes and the last 30 years of history repeats itself (courtesy of the lookback methodology). A product with a 13% cap has a higher illustrated rate than a product with a 10% cap. But let’s say the 13% cap is predicated on a 10 year fixed charge and the pricing for the two products is identical after year 10. Will the high cap product still have a high cap? Maybe, maybe not. But will it illustrate at issue as if it still maintains a 13% cap against its 10% cap peer? Definitely. So carriers can essentially use short-term policy charges to fund long-term results. This argument isn’t perfect by a long shot, either. Actuaries have to sign a statement saying that the policy shows supportable pricing, meaning that the current cap is theoretically supportable for the life of the contract. What we don’t know, however, is what kind of spread and interest rate assumption the actuary is making to back up the sustainability claim. Assuming rates increase, a 13% cap might be totally supportable but we can also assume that the other carrier would also have a 13% cap at that point. The issue here is not projected sustainability because that’s contingent on a whole host of assumptions. The issue is whether or not the high cap product will maintain its cap relative to its peers. I don’t think the evidence that it will is particularly compelling.
How does this impact producers? Profoundly. Indexed UL product selection is much more nuanced that industry practice makes it out to be. But, again, lowering illustrated rates can cover a multitude of other errors. The price of being wrong if you illustrate a product at 5% at onset for the 10% cap product and 5.5% for the 13% product is trivial. But showing the industry practice spread (6.50% versus 7.75%) is a much bigger deal. I’m also in favor of carriers switching to asset based charges even though it makes IUL smell like a securities product because asset based charges are the only way to ensure a consistent cap differential between products holding everything else constant. Asset based charges also reduce illustrated arbitrage to its usual flavor in Indexed UL – beating the options market by turning a 5% options budget into an 8% illustrated rate. But I digress. Producers and brokerages should think twice before selling a product for its current cap and realize that the long term cap will vary unpredictably both in absolute level and relative to its peers in a pattern that’s impossible to predict.