#93 | Securian/Minnesota Life Orion IUL

Executive Summary

Despite its attempts to diversify, Minnesota Life has undeniably built its business on the back of Indexed UL. Orion IUL is an evolution on the long-running Eclipse IUL product. The biggest difference between the two is that Orion introduces the Annual Policy Credit (APC) – a virtually undisclosed bonus structure that is, to put it mildly, quite complex and convoluted. It appears to have 5 distinct phases where its values are driven by different things happening in the policy, everything from the premium pattern to the crediting rate to the COI slope. It is virtually impossible to model with any predictability. Because the inscrutable APC is the sole reason why Orion IUL performs well on the illustration, we recommend that practitioners stick with Eclipse IUL or sell another company’s product.

Securian is the holding company over Minnesota Life, but the Life division of Securian has historically been branded as Minnesota Life. I suspect that they’ll fully transition away from Minnesota Life and to Securian Life by the end of the year or so. I chose to refer to the company in this article as Minnesota Life to maintain historical continuity.

In a lot of ways, to trace the ascent of Indexed UL is to trace the ascent of Minnesota Life. In 2006, when the company rolled out its flagship Eclipse IUL, the Indexed UL market was a tiny sliver of what it is today. The same was true of Minnesota Life – its 2009 total Life reserves were just a little over $2B, which is a size befitting the small, provincial company that Minnesota Life was. But Eclipse IUL changed all of that. More than any other company, Minnesota Life rode the IUL wave from insignificance to towering heights. As of the end of 2017, Minnesota Life sported over $7B in Life reserves, the vast majority of which is tied up in Indexed UL products. In 2009, Minnesota Life took in just $200M in annual premiums. In 2017, that number was $1.2B. No rise has been as meteoric as Minnesota Life’s.

How did a little company with less brand recognition than a rural Chinese toy manufacturer grow to be one of the biggest names in Indexed UL? Simple. When Eclipse was rolled out in 2009, the cap on the one-year S&P 500 point-to-point account was an astonishing 17%. This was a gaudy outlier in a market defined by 12% and 13% caps – not so different than recent times – and it got a gaudy amount of attention. The illustrated rate on the product was right around 9.5%, if I remember correctly. On paper, it smashed everything in its path.

Minnesota Life had a couple of tricks up its sleeve to make Eclipse work. The first was that Minnesota Life credited indexed interest to the account value at the end of the year, after all of the charges have been deducted, rather than the industry-standard average (or mid-point) account value, which essentially deducts just half of the policy charges. Minnesota Life, in effect, was crediting a higher cap to a smaller number. As a result, the 17% cap in a fully funded Eclipse contract was really more like a 15% cap in terms of what Minnesota Life paid to hedge it and the actual dollars it credited to the contract in the first year. As more premiums are paid, though, the subsidy to the cap from crediting interest to the end of year value diminishes because charges as a percentage of the account value shrink. But, in 2009, all Minnesota Life had was policies in their first year and so 17% it was.

The second trick is simply that Minnesota Life was, and still is, driving at the ragged edge of its option budget. Carriers writing Indexed UL products have been forced to choose between sustainability and competitiveness of caps. Most companies, particularly some stalwarts like PacLife, tend to undershoot their caps a little bit so that they can ride out adverse environments without making changes. Not so with Minnesota Life. Take a look at the history of option prices and caps in Eclipse IUL over the years, which I pulled directly from their statutory filings. These are the real option trades that Minnesota Life executed to support their IUL products. Note that I only have option data for when Eclipse had a 16% cap in late 2009, which is when I presume that they actually started to hedge the equity exposure in the contract.

Every time the price of the option package meaningfully goes up, Minnesota Life’s cap goes down. That’s what I mean by driving at the ragged edge. There’s no margin for error. Along with the rest of the industry, Minnesota Life’s general account yield has been on a downward trend since 2009, so you can see how the pressure on caps was fierce in the first 3 years of the product when Eclipse’s cap fell like a stone from 17% to a rather pedestrian 13%. As of this writing, a recent uptick in option prices in the last couple of months has pushed the cap on Eclipse (and now their new product, Orion IUL) down to 11.75%, which is middle of the pack these days. You might also notice that Minnesota Life is quick to drop caps but was not so quick to raise them when the price of the cap spreads dropped precipitously in 2013 and 2014. Draw your own conclusions.

Throughout this entire period, the Eclipse chassis itself remained basically unchanged. It is classically simple. The star of the show was always high caps, illustrated rates and aggressive illustrated loan leverage. During the AG49 debate, I used an Eclipse illustration with regulators as an example of Indexed UL run amok. The illustration showed $100,000 premiums for 10 years and income starting in the 20^th year that, over the life of the policy, totaled more than $140 million. Yes, $1 million into $140 million. Eclipse’s bread and butter tactic was delivering illustrations that were just jaw dropping. No surprise, then, that Minnesota Life was one of the most militant and uncompromising supporters of the Indexed UL status quo during the AG49 discussions.

Over time, Minnesota Life added a few other indexed account options that were moderately successful. Only in the last few years did Minnesota Life add a bonus based on indexed interest that pays 1% of the total indexed interest earned over the last 10 years. In addition to moderate improvements to Eclipse, Minnesota Life also rolled out a comprehensive suite of other products to round out their portfolio. In just a few short years, they went from a one trick pony to a whole traveling circus.

And even the main event got an update in March of 2017 when Minnesota Life rolled out Orion IUL, the new flagship product to designed to succeed Eclipse. The two products share much in common. Same premium load. Same per policy charge. The base charges are both a 10 year duration and Orion’s are slightly less than Eclipse’s, which is fitting because Orion’s Target premium is slightly less than Eclipse’s. The COI slope in Orion is 10-20% higher than in Eclipse, with the gap widening as the policy gets more mature. In other words, these two products are pretty similar. All else being equal, Orion should be slightly less competitive than Eclipse thanks to the higher COI slope.

However, as you are probably already guessing, there’s a twist. And, of course, it’s a non-guaranteed bonus that is barely disclosed or mentioned in any of the illustrations, marketing material or even the policy filing. And, as you’re probably already guessing, it’s extremely complex and virtually impossible to decipher. And it makes all the difference between a competitive product and a so-so product. We’ve seen this movie way too many times by now, have we not?

I have to hand it to Minnesota Life on this one, though. The mechanism in Orion IUL is called the Annual Policy Credit (APC) and it’s a doozy. The description of how it works is as follows: Non-guaranteed values calculated based on certain factors, including but not limited to accumulation value, interest, index credits, mortality, persistency, policy duration, premiums, policy indebtedness, taxes, expenses, and additional agreements. As far as I can tell, that’s a pretty accurate list of all of the things that go into the formula for the APC. It is exceedingly complex. But I spent a few hours attempting to figure it out and I think I actually teased out some of these factors. I’ll break them down for you in a few simple graphs. I should also note that I can only show you correlations, not causal links. With bonuses as complex and poorly disclosed as this one, understanding them is an exercise in poking at them to see what they do and then trying to see patterns. It’s kind of unbelievable that we’ve devolved into selling products as inscrutable as this one to innocent customers, but I digress.

Also, there are a lot of ways to interpret Minnesota Life’s explanation that the APC is paid as “non-guaranteed values.” Does that mean a fixed dollar amount? Indexed multiplier? Bitcoin side accounts? It could really be anything. But what I found to be the only measurement that showed discernable, meaningful patterns was quoting the APC as a percentage of the Account Value. Think of it as something closer to an interest bonus than indexed multiplier. And with that, here are the factors that I found to impact the APC – and I’m sure I missed a few.

The first one was the easy one – illustrated rate. The most obvious place it shows up is in the very, very end of the contract, like just before age 121. At the maximum rate of 6.88%, the APC was 0.45%. At 5.88%, it was 0.40%. At 4.88%, it was 0.35%. See the pattern? Yes, you do. At 3.88%, it was 0.30% and at 2.88% it was 0.25%. For any given crediting rate, the final APC was the same regardless of funding pattern, so I’m pretty sure that this successfully isolates the portion of the APC related to the illustrated rate. I came to think of it as something like a baseline rate that operates under everything else in the APC. It’s the common denominator.

But, like any good algebra student, I also came to see that the best way to understand the APC for any particular funding pattern was to subtract out the baseline rate. And, as you can see from the chart below, you get remarkably similar results across all illustrated rates once you pull out the Illustrated Rate Factor. I only ran it from year 11 (when the APC really kicks in) to year 41. You’ll see why in a second.

Now, you might notice that years 11-16 have a bit of a slope to them. That lead me to the second identifiable factor for the APC, which is the premium funding pattern. The APC always converges to the numbers you see above after year 16, but the rate and shape of the convergence is dependent on the funding duration. Take a look at the chart below. All three were illustrated at 6.88%.

The curious thing about this pattern is that the total premium amount and account value isn’t what impacts the APC. The Single Pay and 7 Pay Account Values are virtually identical. I actually noticed this phenomenon when I was running illustrations at 2.88% and had to run a 10 pay in order to keep the policy in force. The APC credit had a lower slope than the 7 pay at 2.88%, which doesn’t make any sense and isn’t related to the Account Value because the AV in all years of the 2.88% is lower than at 6.88%. I wish I had an explanation but I don’t. This is just what happens when you poke at the APC.

The third factor is the most curious. Believe it or not, the COIs are also an input into the APC. It took me a while to figure this one out, but the graph below makes it so obvious that it’s hard to miss. I’ve quoted the COI charges as a percentage of Account Value as well so you can really see the correlation. The scenario is a 7 pay premium solve for $1 of CSV at age 100 at 6.88%. Notice that this graph goes all the way to age 121. You can clearly see how the APC moves almost in lock-step with the COI as a percentage of AV in later years. It is actively working to offset the impact of rising COI charges.

The same pattern shows up for a 7 pay, max funded scenario but the effect of the COIs is muted by the lower COI charges as a percentage of the AV. As a result, the APC in later years is also muted. You’ll also note that the APC is actually higher in the lower funded scenario than in the max funded scenario even earlier in the policy duration.

This graph is actually a perfect guide to the APC, which I actually see as 5 distinct phases that operate with different inputs that I’ve demarcated on the graph with the dashed gray lines. Phase 1 is when the APC is a trivial, token amount and wasn’t even worth graphing. Phase 2 is years 11-16, where the premium duration determines how the APC grades into its ultimate slope. Phase 3 is years 16-41 are an APC that gradually increases at a linear rate, usually 1-2bps per year. Phase 4 is years 41-45, which is where the APC adjusts from Phase 3 to Phase 5, which is the COI-driven APC amount. And, finally, Phase 5 is the APC that is based on the illustrated rate after all of the COIs have been eliminated because the policy AV is equal to the DB. Thinking of the APC in terms of these phases seems to be the truest way to understand, in basic terms, what it does. I’m sure that these five phases take different durations and magnitudes at different ages, face amounts, whatever. Working out all of its kinks would probably take a hundred hours and twenty pages worth of explanation. I’ve stared into the abyss with the APC and decided that it’s not worth the leap.

If you really want to be charitable about my analysis, you might say that I’ve outlined a reasonably logical explanation for the how the APC works, but you’d be going a bridge too far. I’ve outlined how the APC works on the illustration. I have no idea how it will play out in real life. It’s just too complex. I never thought I’d say this, but the APC is even more complex than the Performance Factor in PacLife’s PDX and it’s just as poorly disclosed. If you have to choose one of these two products to sell, I would probably tell you to sell PDX. I would choose PacLife over Minnesota Life in isolation and I would feel more confident attempting to explain the Performance Factor in PDX than the APC. There’s at least some inherent logic, consistency and even some stability in the PDX structure, for all of its vagaries and complexities. I’m not sure I can say the same for the APC. There are just too many inputs and too many moving parts.

But, fortunately, you don’t have to choose between these two products and you really shouldn’t choose either of them. There are plenty of simpler and more transparent options in the market that will perform just as well in the real world. And if you want to sell a product from Minnesota Life, you should go with the old one trick pony – Eclipse IUL. It’s a structurally better product than Orion IUL and it doesn’t saddle producers and their clients with something as ridiculously convoluted and inscrutable as the APC. Vote with your production, and vote for simplicity.

If your interest was piqued by the option trade data shown earlier in this article, then you’re going to love the upcoming article series on hedging Indexed UL. The Minnesota Life statutory filings were an absolute gold mine – 100 back-to-back trades every month from September of 2009 to December of 2017 with separate entries for the at-the-money call and the out-of-the-money call. I couldn’t have asked for more perfect documentation. And what it shows will blew me away. Stay tuned.