#84 | PacLife PDX – Part 5 – More to the Story

When I wrapped up writing and publishing the first four parts of this series on PacLife PDX, I thought I had a pretty good feel for where this series was going to go. I thought I’d nailed down the salient elements of the product and was ready to move on. But I was wrong – sort of. I knew there were some questions to which I hadn’t found satisfactory answers, so I kept poking around and unfolding more and more of the product. What I found didn’t change the mechanics of the story as much as it changed the emphasis, especially as it relates to the relative weight of MX and QX. That bit at the end of the last post about discarding QX? Yeah, well, you can discard that. Things aren’t as simple as they once were.

The main piece of new information that I uncovered in the last few weeks was how PacLife determines the base charge that feeds into the MX portion of the formula. The fact that PDX has enormous charges is quite obvious, but what’s not so obvious is how PacLife arrived at a particular charge for a particular client or cell. This question had been gnawing at me since I first saw the product. There had to be a formula behind the scenes for “calibrating” the MX portion of PDX. I couldn’t imagine that an actuary had run an independent model for each and every cell to determine the appropriate base charge down to the dollar. There has to be some other number, some external, pre-calculated number that the base charge is related to.

So what number could that be? As it turns out, PacLife found a very convenient one with the Guideline Level Premium (GLP) calculation under the Guideline Premium Test (GPT), which is the original IRC 7702 test for the definition of life insurance. The GLP is a calculation derived from the guaranteed charges and statutory interest rate embedded in every product. It’s probably not unfair to think of GLP like the guaranteed premium for a non-guaranteed product – which means that the GLP is quite high relative to premium solves based on non-guaranteed charges and credits. For example, the GLP on PDX for a 45 year old Preferred male with a $1M DB is $18,895, compared to the level premium solve for this same cell of $7-9k. The base charge going to MX for this same cell in PDX is $8,744, which is about 46% of the GLP. As it turns out, that relationship generally holds across ages and face amounts. Voila. That’s probably the formula. Given that we are left to deduce the Performance Factor entirely on our own, you take what you can get.

The fact that PacLife pegs the base charge to GLP raises another strange and wonky issue that has decidedly real and important impacts on the product. The GLP calculation uses (mostly) guaranteed charges and credits, so it’s a concrete number for all rate classes at a given age, face amount and sex. The GLP is also more or less proportional across face amounts. Calculations related to GPT, including GLP, are very stable policy attributes in the multidimensional pricing space of life insurance policies. Until they’re not. Simply switching from a level death benefit (Option 1) to an increasing death benefit (Option 2) dramatically changes the GLP calculation and results. For example, the $18,895 GLP quoted above turns into $54,377 with an Option 2 DB, nearly a 3x increase, for the exact same age, sex and face. This is a fairly well-known (even if not well understood) phenomenon to anyone who sells life insurance for accumulation. If you want to take advantage of the narrower corridor in GPT, the only way to get a full slate of 7 maximum non-MEC premiums into a GPT policy is to use an Option 2 death benefit while you’re paying premiums because the GSP/GLP limits are so much higher.

What this means for PDX, then, is that switching from Option 1 to Option 2 materially changes the risk profile of the product because it dramatically increases the base charge that feeds into the MX portion of the product. It also changes the formula for setting the base charge. Whereas Option 1 PDX products have base charges that are pretty much always at 46% of GLP (+/-), Option 2 PDX products have base charges with ratios that float a bit more and generally go down as the client gets older. For example, a 35 year old Option 2 PDX has a base charge that’s 43% of the GLP. For a 65 year old Option 2 PDX, the ratio is just 25%. In either case, the actual base charge for the Option 2 PDX is still way higher than it would have been in an Option 1 PDX, but the difference isn’t quite as dramatic as the change in the GLP itself because the ratio adjusts. Why do you care? It’s certainly more evidence of the nearly paralyzing degree of complexity in this product. But it also blows apart any degree of generalizations you might want to make about the risk profile of the product. The size of MX is directly correlated to the riskiness of the product, and that riskiness can change with the GLP and with the formula for the ratio of the GLP that gets translated into the base charge. If your head is about to explode, just move on. Mine exploded a long time ago.

To break it down into simple terms – PDX with Option 1 is low-MX, PDX with Option 2 is high-MX. More MX means more illustrated return and more risk. More PDX, in other words. It would be a serious mistake to trivialize the difference. For example, on Option 1, the base charge of $8,744 going to MX in year 11 represents just over 2% of the account value at the end of the previous year (assuming maximum non-MEC funding and maximum AG49 illustrated Rate). Using Option 2, the base charge of $21,744 going to MX represents a little over 5% of the account value. Like I said, more MX. A lot more MX.

Strangely enough, switching to Option 2 actually means more QX as well. I do not have a good explanation for this. I really don’t. All I know is that QX dramatically increases when you switch to Option 2. For example, there is no QX for a 35 year old with Option 1 PDX, but QX gets as high as 11% (so a 1.11 index crediting multiplier) with Option 2 PDX. For a 55 year old, QX peaks at 1.21 with Option 1 but peaks at 1.47 with Option 2. Consequently, it’s not just that Option 2 increases the base charge which logically increases MX and therefore illustrated performance, but Option 2 also delivers more of the tontine-esque magical QX fairydust to boot. A lot more QX. But, as it turns out, that only applies to younger clients. QX for a 65 year old is basically the same under Option 1 or Option 2. All of this is yet another example of the confounding complexity of this product and, if you’re a skeptic, yet another example of all of the things that look great on the illustration but probably won’t play out in reality.

Now, from a design standpoint, getting more MX/QX juice courtesy of Option 2 is pretty easy. First of all, even though the entire formula is pegged to GLP, which is a calculation within GPT, you don’t actually have to use GPT for this product. Yes, you read that correctly – even when you use CVAT, you still get all of your MX-ness calibrated to GPT. In one respect, this makes sense because, as I said earlier, GPT calculations are very stable and provide a clean baseline for calibrating product attributes. But in other respects, it’s bizarre. I’m sure PacLife also pegged the base charge to GPT for legal/tax/pricing reasons in order to do the maximum amount without running afoul of some arcane GPT rule, but none of that applies under CVAT. So, in other words, if PacLife had only allowed CVAT, how big could the base charges going to MX be? I’m sure we’ll find out the answer when another company issues a competing contract only under CVAT with much bigger annual charges and a consequentially sexier illustration.

Second, PacLife permanently installs the higher Option 2 base charge even if you change the death benefit back to Option 1 after the first year. In other words, if you want more MX but don’t want an Option 2 DB forever, then just switch it back after the first year and you’ll get to keep the higher charge and more MX with the economics of an Option 1 DB. Magical, right? Yes, and bizarre, because this doesn’t fit how GLP is calculated. GLP/GSP changes if the face option changes in any year, but the base charge in PDX is only pegged to the initial GLP. Chalk this up to yet another confusing and capricious aspect of PDX.

I included a bit of PDX analysis in a recent presentation and briefly spoke about how the death benefit option in PDX serves to calibrate the MX and QX portions of the product. I brought up the fact that switching to Option 2 just for one year can increase illustrated cash value IRRs by as much as 1.5%. A producer approached me afterwards and said “so you’re telling me that if I want the product to illustrate better, all I have to do is start with Option 2 and then switch to Option 1?” And yes, he’s right, in the same way that it would be right to say that Lance Armstrong was a better bike racer when he was hopped up on EPO and doing 3 blood transfusions at the Tour de France rather than racing clean. He won. And then he lost. All of them. Reward always looks great on paper, but risk looms large in the real world.

In the next post, I’m going to show a statistical analysis of the risk in PDX and why I wasn’t deploying Trump-esque hyperbole when I said in a previous post that PDX is the riskiest and most leveraged IUL product in the industry. Statistically, it is – whether in its low-MX Option 1 form or its hopped up, high-MX Option 2 form. But, as you’ll see, there are a few more folds to the story than I anticipated.

Before I go there, though, I have to bring up one more issue that no one ever seems to want to talk about – compensation. PacLife is one of the few companies in the market that allows producers to adjust their compensation without changing the fundamental attributes of the product. Compensation is directly tied to policy charges in the first 10 years, that’s it. As a result, producers selling PacLife are always caught in a bit of a balancing act between illustrated performance and compensation. PDX takes away that dilemma and I mean that only in a negative sense. I think the best way to show you the problem is just to give you the numbers. You’ll see what I’m talking about. Below is a table of different coverage blends and Target premiums (compensation amount) for both death benefit options on a 45 year old Preferred male with $1M of DB.

So, which one do you sell to your client? It’s not so hard to see how producers – a lot of producers, from what I hear – have decided that PDX represents a nice way to retire early and sleep well. To put this in context, the average accumulation IUL product pays about $16,000 in Target for this cell and performs about as well as the Option 1, 50% base PDX product. PDX, therefore, allows producers to make three times as much compensation while delivering a product with equal illustrated performance. It seems too good to be true – because it is. Who pays for all of that compensation? The client does, in real dollars, because policy charges in the Option 2 PDX with 100% base are ridiculously high, as evidenced by the fact that the policy hasn’t even broken even by year 10 ($305k in CSV vs. $315k in premium) with a 6.21% illustrated rate. But MX pays for the compensation in terms of illustrated performance because the higher base charges associated with Option 2 generate so much extra return – and risk. By now, though, you’re well aware of the fact that the risk isn’t illustrated.

In a lot of ways, this compensation issue might be the most problematic aspect of PDX. Yes, the product is riskier than other Indexed ULs. Yes, PDX is absurdly, incomprehensibly complex. But if a producer who really believes in IUL can understand and get comfortable with this product (and carrier), they could find a role for it with some clients and in some situations. If the incentives for selling this product were the same as others, then it would have a rightful place in the market and would doubtlessly be a leading offering. But the incentives are not the same. A producer can make 3 times as much money selling PDX as any other IUL while showing equal long-term illustrated performance, albeit with significantly higher charges and dramatically increased risk to the client. All of the incentives are out of whack – including the incentive to find fault in the product. Upton Sinclair once said that “it is difficult to get a man to understand something, when his salary depends on him not understanding it.” Little wonder, then, that so few producers really understand this product and so many producers are selling so much of it. That’s exactly the way the incentives are aligned.