On August 6, The Capitol Forum held a conference call with Andrew Granato, Assistant Professor at the University of Texas School of Law, and Pranjal Drall, a J.D.-Ph.D. candidate in Financial Economics at Yale University, to discuss their paper, “Private Credit’s State Backstop: How Private Equity Socializes Risk Through Insurers,” and The Capitol Forum’s reporting on private equity-backed life insurers’ exposure to private credit and the regulatory risks associated with those investments. The full transcript, which has been modified slightly for accuracy, can be found below.
TEDDY DOWNEY: Hello, everyone. Welcome. I’m Teddy Downey, Executive Editor here at The Capitol Forum. Today, I’m pleased to be joined by Andrew Granato, Assistant Professor at the University of Texas School of Law and Pranjal Drall, a JD, Ph.D. candidate in Financial Economics at Yale University.
We’re going to be talking about their paper, “Private Credit’s State Backstop: How Private Equity Socializes Risk Through Insurers,” which examines the growing role of private equity on life insurers in private credit markets, the risks associated with those investments, and how the state insurance guaranty system can shift potential losses onto competing insurers and taxpayers.
The paper is so, so, so good. It’s 60 pages. It’s an easy read though. It actually is an easy read. I loved every minute of it. Please read it. If you have not read it, please, please, please read it. You’re doing yourself a disservice if you don’t read it.
But Andrew and Pranjal, thank you so much for doing this today.
ANDREW GRANATO & PRANJAL DRALL: Thanks for having us.
TEDDY DOWNEY: So, I always like to start off—particularly in a paper like this where I think the perspective is really interesting. You take a different lens from a lot of the other people who are writing about private credit and private equity and insurance. And I would love to get a little bit of your background, how you came to write this together, and how you came to look at it from the perspective that you did.
ANDREW GRANATO: Yeah. So, I used to work at the Federal Reserve Bank of Chicago in their Insurance Initiative Unit before I went to grad school. So, I did the same program that the Pranjal is in right now. I’m a couple of years ahead. And so, I’m starting as a professor. But I have a lot of background in insurance, and, in particular, like the tax law of insurance.
PRANJAL DRALL: And so, recently I’ve been working on the consequences of informational frictions in asset valuation. So, basically, how problems of valuation create various kinds of problems for the financial system. And part of that is studying private credit in different contexts.
Andrew and I started talking about this project like six months ago, like February-ish. And we were like I do private credit. Andrew does insurance. And there’s like a nice paper waiting to be written here where we noticed the trend of PE buying insurance companies and thought there’s like a paper here somewhere. And we started working on it and we were able to finish in like a six month span.
TEDDY DOWNEY: It’s a really complex space. And I don’t know how much of your time you’re able to spend on it. But to get this much clarity in a short amount of time, I think it was really impressive. And in your paper, you talk about how places like McKinsey and just generally people who like this industry talk about how private equity buying insurance companies creates a flywheel. It makes a lot of sense to match up illiquid assets from private equity with these insurance policies that don’t need to be paid out until much later on.
But to me, reading your paper, I’ll be honest, it felt like a very complex scheme to defraud the public—which I would include as taxpayers, policy holders, and maybe even investors in some of these vehicles—through private equity and private credit self-dealing and these elaborate ways of hiding risk from regulators and investors. Obviously, those are two diametrically opposed ways of looking at what has happened here. But for our listeners, maybe you could walk us through how private equity decided to buy life insurers and how that changed incentives in the industry.
ANDREW GRANATO: Yeah. So, this story, I think, starts in the aftermath of the 2008 financial crisis. That’s also the story of a start of like the private credit industry as we know it today. Dodd-Frank gets passed. It gets more difficult for banks to make these kinds of loans themselves. Other investors step into this area. There’s potentially also increasing demand from the borrower side as well for these bespoke loans that have particularized loan conditions that are not publicly traded, that are often made directly, often held to maturity.
And so, there is, like in the telling of the flywheel metaphor of all of these, as you say, great synergies between that asset management model and the liability side structure of insurers, where most of insurers’ liabilities are the actuarial reserves for policyholders. You can think of policyholders to an insurer being people who’ve made quasi-loans to an insurer. They are paying premiums to that insurer upfront, often for very long periods of time. And they may or may not actually get payouts from that insurer.
What they’re doing is they are covering their risk. And so, they are transferring the risk of having to make payouts to that insurer. And the insurer pools these risks. And so, at an expectation, it has to make some certain amount of payments, but it doesn’t know to whom.
And those payments are often very far in the future. When people buy a life insurance policy, potentially they say like, oh, I buy a 10-year term life policy. If I die in the next ten years, the insurer makes a payout to my beneficiary. Or they potentially buy an annuity. They contribute all this money upfront to the insurer, and then the insurer will make slow payouts to the policyholder over time for the duration of that policyholder’s life.
So, that story of the virtuous synergies of the flywheel, we want to emphasize that we think it’s not wrong. We agree that there are these natural synergies between the long-term liquid loan side of making private credit and the relatively stable long-term liability side of the insurer. What we want to say is that if the story ended there, that that would be a deeply incomplete telling of what is really going on, which, as you know, is the rise of this entire model in which the private equity buyout funds, the private credit funds, and the life insurer are all under the control of the same entity.
And so, when you have that situation, there’s many differences between holding the private credit loans on the balance sheet of the private credit fund versus holding it on the balance sheet of the life insurer. And so, I’ll turn it over to Pranjal to talk about why those differences are so crucial.
PRAJANJAL DRALL: Yeah. I mean, another way to think about is that an insurance company takes a bunch of money from policyholders, invests it, earns a spread, and then gives back the policy, whatever they promised, 20, 30 years down the line.
In order to make investments, MetLife can hire a bunch of people to make investments on their behalf. But as you can imagine, a lot of 25-year-old Harvard MBAs don’t really want to work at MetLife. They’d rather work at Apollo or like a fancy PE shop.
So, insurance companies are able to outsource this investment decision-making to private equity in some way. And there’s like various sorts of arrangements you can have between the insurance company and the private equity company. And so, a very benign way to think about it is instead of having the insurance company make investments, you can have the PE company make investments on your behalf.
Now, that sounds fine. There’s economies of scale here as PE shops have built great businesses investing people’s money. Why don’t we let them invest insurance order premiums so they can make more money for the insurance company? And in expectation, also offer better deals to the policyholders? Because you can reduce prices and give them higher payouts if you’re able to make more money.
TEDDY DOWNEY: And there’s competition on that bid potentially for that work.
PRANJAL DRALL: Exactly. Like insurance companies, some of them are holding $300 billion in assets and that can generate a lot of money in fees. So, Blackstone and Apollo and KKR are competing for that business. So, you can imagine like the fees are fairly high in that setting.
A different way to think about it is Apollo, instead of competing in that business, would rather own an insurance company, have it all be internal to the asset manager. And that’s the trend we highlight as more complicated or problematic in that the insurance company—when they lose bargaining power to bargain on behalf of policyholders and they become part of this bigger asset management umbrella—then they cede more and more control to the PE shop and have them make not only investment decisions, but also how much fees to pay. So, they can just pay themselves higher fees. Or when they buy assets, they can buy assets at a markup when they’re buying from their own asset management’s different business lines versus when they buy loans from a different non-PE portfolio.
So, that’s the problematic end where just by ceding control of the insurer, the policyholders lose out in expectation.
TEDDY DOWNEY: Your paper actually has an example of when it is to the parent-company PE firm’s advantage to make the insurer insolvent. That seems like a deeply problematic, concerning conflict of interest. But you very explicitly lay out five ways that private equity companies can effectively loot their insurers to the detriment of the insurer.
And you don’t talk about this, but obviously, everyone that works at the insurer essentially is harmed by that. The policyholders and taxpayers who effectively rescue the insurers when the insurers go belly up. And the five ways you list, you say PE firms may drain an insurer by assigning the insurers asset management or other tasks of fee charging affiliates.
So, basically just extracting a bunch of fees. PE firms may strategically offload poorly performing or excessively borrower friendly loans from affiliated private credit funds onto the insurer at inflated prices. PE firms may reorient insurer portfolios toward affiliated structured and private credit that offers higher promised yield, but creates greater loss and severe stress. PE firms may engage in valuation arbitrage with NAIC’s risk-based capital regulatory regime to make their private credit assets look systemically less risky than they actually are. PE firms may expand the use of opaque shadow reinsurance that conceals risk.
This looting of the insurance company seems deeply problematic to me. I would love to go through these onebyone. Tell me about these fees and how these fees add up. Where do these fees come from? What are the problematic fee-charging things that you see?
PRANJAL DRALL: Great. So again, going back to the hypothetical, say an insurance company is fully owned by a PE company, you can make the insurance company pay higher fees to yourself. So, the basic idea is every dollar you invest for the policyholder, you can say make 20 basis points on that. That’s the most direct form of extraction in that the fees are getting paid out today just for investing.
And obviously the investment—and in the paper we cite some other empirical work that shows once a PE company buys an insurance company, they offload investment staff at the insurance company. But whatever money they save, they pay out in fees to the broader asset manager.
So, again, you might think that’s benign in that someone has to invest the money. Investing money costs money and you can pay it out in fees. But the point is like the bargaining power where a PE company, when they own insurance, just can pay themselves way more.
The other more implicit and hard to empirically study fee extraction would be: an insurance company pays out other costs. So, like IT and HR and a range of other services that they get. A PE company might have a bunch of portfolio companies that they make equity investments in and they can provide those services to the insurer.
So, some of these insurance companies are gigantic and can support other portfolio companies that are owned by PE in different lines of business.
So, there’s like the direct fee extraction. There’s also this support of other PE businesses by just using their services. And again, there’s not a lot of bargaining. So, you can just take a lot of the policyholder money and pay inflated fees.
ANDREW GRANATO: Yeah, and the background here is you might think like, well, if the private equity firm, if they own these other companies, they also own the insurer and they’re conducting these internal transactions. Maybe the insurer relatively loses, but why does that matter? Because they’re just all just part of one big company.
And the reason why is that it is only the insurer part of the private equity firm that is subject to this unique insurance regulatory regime called an insurance guaranty fund. An insurance guaranty fund is a public backstop for the policyholders of an insurer.
I think listeners will likely have heard of federal deposit insurance. This is a regime that functions somewhat similarly though it differs in important ways for bank depositors. So, the reason why you might be very confident that your checking account is secure—even in the event if your bank goes insolvent—is because you are insured on that money through the Federal Deposit Insurance Corporation.
And similarly, insurance policyholders are insured on their insurance through state level guaranty funds up to certain statutory caps. And so, what this ends up meaning is that I think you should consider who are the funders of each of these different businesses within the PE firm’s portfolio.
So, if you have the buyout funds where the private equity firm is the general partner, and then they have LPs who are often these large institutional investor clients’ pensions, family offices, university endowments. These are a relatively small number of investors who have a ton of capital each and who are repeat players in this market, who the private equity firm has incentive to keep them happy because they want to be able to continue to invest on their behalf and earn fees on those investments.
The life insurer, on the other side, who funds the life insurer? It’s the premiums of this dispersed retail base of policyholders. It is random people all across the U.S., none of whom individually is particularly important to the insurer and none of whom has strong incentive to monitor what the insurer is doing, partially because they are a relatively small creditor to the insurer and partially because they have this public backstop.
So, I would imagine that a lot of people in the audience who have life insurance policies, like has anyone ever thought to check what does my life insurer do with this money? I certainly haven’t. And so, when you combine that with the fact that the insurance guaranty fund enables a public backstop for the policyholder, that creates this like consistent asymmetry in incentives for who the private equity firm has an incentive to favor when conducting these internal transactions within the firm.
TEDDY DOWNEY: Yeah. So, basically you’re saying that, look, if the parent company was going to have to pay for this insolvency, it wouldn’t be as big of a deal that they’re feeing themselves to death or whatever. They’re taking money from one pot and putting it in another. But what you’re saying is, actually, if the insurer goes belly up, the people that get left holding the bag are the policyholders, the amount that is not guaranty of the policy.
So, let’s say you have a million dollar policy, only $300,000—you lose $700,000 for your spouse who’s trying to collect after the other spouse has died or whatever. As you say, a very sympathetic customer potentially, they’re losing $700,000 or whatever they’re losing if they have in excess of $300,000. But then the taxpayer—in 40 states—is actually the one that ends up subsidizing the vast majority of the rest. Because the remaining insurers that pay into the guaranty fund get a tax credit against that. So, the bag holder is the policyholder and the taxpayer in each of these states, not the parent company.
PRANJAL DRALL: Not the parent, yeah.
ANDREW GRANATO: Yes. So, the way that insurance guaranty funds work is that if a life insurer goes insolvent, we have this guaranty to the policyholders that—depending on the state—maybe something like the first $300,000 of their insurance policies are guaranteed. So, the state regulator has to take over the insurer and then decide like, okay. Well, first we look at how many assets are left over in the insurer. Are they enough to make these like expected payouts?
Well, the insurer is insolvent. So, probably not. And so, then they have to go and they have to find a way to raise money to make policyholders whole up to these caps. And the initial way that they do that is they actually bill every surviving life insurer in the state.
So, if you’re some other random life insurer and you happen to sell policies in Ohio, and an insurer that operates in Ohio goes down, well, congratulations. You just got a bill for your proportionate share, based on premium volume, to bail out the policyholders of this other insurer. So, initially the burden falls on the rest of the industry to bail out the policyholders. But then, as you know, in the vast majority of states in the United States, state law provides insurers a tax credit for these assessments that can be taken over a period of roughly five to ten years.
So, because there is a tax credit that offsets this burden to the insurers, the burden actually shifts to the taxpayers almost completely. The insurer has to front the money and so they lose that time value of money. But the actual financial liability is ultimately placed on the taxpayer.
So, in 2008, when taxpayers had to bail out financial institutions, the legislators had to vote on it. It was extremely important. This was a signal moment. But in the insurance context, this taxpayer bailout actually occurs automatically in the most obscure way possible that minimizes, I think, the probability that anybody would even notice that it’s happening.
TEDDY DOWNEY: Yeah, yeah. I mean, you mentioned in the paper AIG didn’t have to go through this process because they got bailed out by TARP. I want to come back to the insurance. I want to come back to the guaranty stuff later. I just want to get back to all the different ways that PE is looting the insurance company.
But I want to make it very clear. Basically, you’re saying that the bag holders are the policyholders, and in many cases, in the end, the taxpayer. So, you may not know it, but 20 years from now, ten years from now, five years from now, two years from now, if any of these go belly up, your park isn’t getting fixed. Your trash is not getting picked up. Your public school’s not getting funded. I mean, your services in your state may be harmed by this looting that is currently going on.
Let’s go to the second one here. PE firms may strategically offload poorly performing or excessively borrower friendly loans from affiliated private credit funds onto the insurer at inflated prices. Tell me about that. Because I see that and I’m like, so, you’re a PE firm. Let’s say you can’t sell off your firm that you’re buying out.
PRANJAL DRALL: Yeah.
TEDDY DOWNEY: You’re like, hey. Let’s just make loans to either the fund itself or the struggling companies. And like, there’s not a good market for that. So, let’s just load it onto our insurance company asset balance sheet because we can’t sell it anywhere else. Am I oversimplifying it? Tell me about what’s going on here.
ANDREW GRANATO: So, I think the intuition is exactly right. These are all loans. So, there’s no like equity, but the idea is the same. Where recently there was this PE company called Blue Owl. And Blue Owl has a private credit fund that was in talks of being merged after they got a lot of redemption requests. And they were in the news for like failing to meet redemption requests of very sophisticated investors. Think your endowment, pension fund, those kinds of LPs.
And in light of that, instead of—so, there are two kinds of funds. Ones that are completely private, that do not have any—whose valuations are basically reported by the PE shop saying our loans are worth X amount of dollars. Then there’s like a publicly traded fund, owned by the same—run by the same PE company. That fund happened to trade at 20 percent discount.
So, the public market said that the value of their loans—and the two funds have 99 percent overlap. That’s maybe slight differences, but like it’s substantially the same portfolio of loans. The public market said the loans are worth actually 80 cents on a dollar. This was after a new GPT model came out and there was a concern that software service companies were struggling. And PE had made a bunch of these loans to like middle-market companies in Chicago. And the idea was like, why would I pay the Chicago company? I can just do everything inhouse using ChatGPT.
And at this time—this is like March of this year—all of these loans started trading down significantly, but the privately held valuations of these loans did not change or did not change enough. So, everyone saw this as like a private fund that is trading at a hundred cents a dollar on the private market. Same assets are trading 80 cents a dollar in the public market. And the investors want out, the sophisticated investors in the private fund.
So, Blue Owl tries to go on the market and a transaction is announced where Blue Owl’s own insurance company called Kuvare and they’d buy the loans at 99 cents a dollar from Blue Owl. Again, the same loan is trading at 80 cents a dollar in the public market. And the insurance company bought the loans at 99 cents a dollar. And the LPs who exited were endowments and sophisticated LPs.
So, that’s one example of a transaction where there’s some proxy price. We all in the public market thought the price was 20 percent discount to the loans. But the insurance company is buying at an inflated price compared to that public market benchmark. Again, that’s one example.
There’s a paper we cited, an empirical paper, that shows when PE companies buy affiliated company assets, they’re paying something like 10 to 40 basis points higher for the same asset on the same day when it sells to one time when the insurance company is buying and when like a non-controlled buyer is buying.
So, you can see that like broad based in the data as like an average, but it’s also happening at like this Blue Owl example is very stark. But the basic idea is like Yale is very sophisticated. They’re monitoring really well. The policyholders aren’t monitoring. And so, the PE company is able to like pay a higher price to Yale because, again, I don’t know what insurance company is buying what loans. Like the NAIC is supposed to be looking into these affiliate transactions. But I’m sure we’ll get to that later. But the rough story is that they’re not doing a good enough job of policing these kinds of affiliate transactions.
TEDDY DOWNEY: Andrew, anything else on that?
ANDREW GRANATO: Yeah. So, I think with the Blue Owl transaction and with all of this in general, the extent to which this is a problem, I think, is endlessly empirically contestable because the valuations of these assets are difficult to do. These are often bespoke individualized loans.
So, what exactly is this loan worth? Well, that’s very hard. It’s very hard to figure that out in an objective sense. I don’t even—like, how would you even begin to do that? And it’s exactly that opacity that enables actors who are closest to the ground. So, in this case, that would be the private equity firm that owns these loans to be able to leverage information asymmetries.
Like you could have potentially, in theory, a perfectly clairvoyant regulatory regime that would be able to crack down on this. But like, obviously, that doesn’t exist in the real world. We’re stuck dealing with the limited information that we have.
And so, while the extent to which this is happening—we could have a fight about that forever—what we want to point out is just that the incentive to conduct these internally favorable transactions is just permanently there whenever you have a private equity firm that owns, both the company that is on the receiving end of the loan and the life insurer that is holding the loan itself, or that could be the holder of the loan if transferred by a different affiliated private credit fund that is also part of the same platform.
TEDDY DOWNEY: Yeah, they have a real incentive to lie as a PE company. But if you’re looking at it from the standpoint of the policyholder or the taxpayer, you want there to be an incentive to tell the truth because you want the proper amount of capital in the insurance company. So, it doesn’t go belly up. So, these seem very much at odds here. And we can come back to who could be looking at what these are worth when we get to the rating agencies, but we’ll get there momentarily.
You’ve got a third way PE firms may be looting their insurers. “PE firms may reorient insurer portfolios toward affiliated structured and private credit that offers higher promise yield but creates greater loss and severe stress.” So, I think this gets at a little bit of maybe two things I’d like to get more clarity on this.
One is just like this isn’t the Warren Buffett or the classic insurer model. Yeah, we’re going to invest these premiums and this money that we’re getting from the policyholders and we’re going to invest it in AAA assets or money good at whatever very highly rated assets, safe assets. We’re going into risky assets, high yield assets.
And you also mentioned structured. Is there anything about how they structure these and the high yield?
PRANJAL DRALL: Yeah. So, the basic point is that an insurance company makes a bunch of kinds of investments. Sometimes they hold cash. They make some AAA rated safe bond to say AT&T, which is in expectation super safe. Then they invest some in Treasuries, different parts of the yield curve. And then they hold equity positions in like a CLO, which is more risky.
And what PE has done is that they’ve transformed that by investing in private credit, which is generally higher risk in that if you just look at how much private credit moves out to the market, it moves more like risky debt than it does safe AAA-rated bonds. And you don’t need to do a lot of sophisticated analysis, the median loan in private credit (at least, non investment grade private credit) is usually to a smaller company with more contingent cash flows that is usually more exposed to certain industries (say, software). And so, it’s very different from vanilla publicly traded bonds. There is of course “investment grade” private credit which you can argue is safer and behaves more like the kind of assets that insurers should hold.
Now, again, conceptually, this would be okay if we knew the true riskiness of an investment. So, the idea is the insurance company holds all kinds of different assets that carry different kinds of risk. And private credit is just one asset that is more risky than other parts of the portfolio. So, if there’s like a clairvoyant person who can see all risk, you would just say, “hey insurance company, you’re holding this risky loan so now hold more of a safe asset.”
But the problem is that since valuation of these loans and assets is difficult, especially difficult for the regulator which has to trust the ratings that the insurance company provides. And there’s a thing called private letter ratings where only the insurance company and the regulator see the rating and it’s not public. So, it’s very hard to even track what rating was this given. How do you check whether these ratings are benign or good? Because you can observe loan defaults. So, you can check, when a thing gets an AA rating, what is the probability of default historically and in certain kinds of downturns? But with private ratings, you don’t know that. You don’t know how safe/risky the assets are. So, it’s hard to implement the risk-based capital regime.
So, the basic point is that private credit is hard to value. Structured securities are also hard to value. And by loading the balance sheet with these assets, what you’re doing is you’re getting the same rating as hypothetically a AAA, publicly traded, AT&T style bond, but it’s higher risk.
So, functionally it’s higher risk, but it looks like it’s as safe as a publicly traded bond. And that creates this incentive where in the upside case, if markets do well, all the loans get paid on time, the insurer/PE parent is earning higher profits because they’re holding higher yielding assets. But in the case of a downturn, you’re marginally increasing probability of insolvency and those costs are socialized. The point is that due to the regime, the policyholders lose out in the story we’ve already talked about.
So, the basic idea is that by holding risky or hard to value loans, insurers have higher upside but the downside risk is protected. So, you’re incentivized to hold riskier assets.
One more way to think about it is, it also penalizes good insurance companies. So, like say there’s two types of PE insurance companies—without even thinking of the Warren Buffett example. So, there’s like one insurance company that’s owned by a PE company that has access to really nice private credit, that is actually AA and very safe and also yields higher because they have better lending technology. And that’s a plausible argument. So, they invest in—I don’t know, Anheuser-Busch—in this massive private credit loan that is the size of a publicly traded loan, but more bespoke.
But there’s like a crappy private equity owned insurance company that doesn’t have access to make these like super big and safe loans. And they invest in these crappier loans. Because of the ratings opacity, their loan looks similarly risky as the safer loan.
So, in equilibrium, the companies have an incentive to move up the risk chain. Because the idea is like, oh, I can just look like I’m making a safe loan. So, what’s the point of making a safe loan? You always want to make the risky loan that looks safe as opposed to making the truly safe loan. So, valuation opacity also penalizes a really sophisticated PE shop that has access to really good loans that are smaller or like a crappier PE company might not.
So, in the paper, we don’t talk about this right now. But our hunch is that the most concerned regulators should be are these smaller PE-owned insurance companies that are like more flybynight, as they have a greater incentive to take more risk because they have less to lose. And we see this in the data in different contexts. Where like firms that are smaller tend to be more willing to take on more risk. So, if they go into stress, it’s less costly.
So, you might imagine like a smaller PE-owned insurance company, that doesn’t have access to all these like high flying good assets, will just invest in crappy private credit, over-inflate the risk. And then like, if they go under, they go under.
TEDDY DOWNEY: I guess my only pushback would be, I don’t understand why a big one would necessarily be any different. They have the same exact incentives for the most part, except maybe they don’t have access to these special things, you know, special great deals. But they’re doing the same types of deals. And if they didn’t have these problems, they wouldn’t need to buy the insurance because they could sell the private equity thing.
PRANJAL DRALL: Yeah.
TEDDY DOWNEY: Anyway, I’m a little bit more skeptical that the big ones are any better than the small ones. I would say the safe ones are also being harmed by unfair method of competition.
PRANJAL DRALL: Right. Yeah, but there’s also large non-PE insurers that increasingly hold private credit assets so it’s hard to make a general point along this PE/non-PE dimension alone.
TEDDY DOWNEY: Because in those ten states where it’s not offloaded on the taxpayer, they take it on the chin, right?
PRANJAL DRALL: Yeah.
TEDDY DOWNEY: They’re going to be the ones bailing these guys out in California, Maryland, and whatever the other states that don’t have the tax credits.
Andrew, anything else to add?
ANDREW GRANATO: Yeah. So, the private equity-owned life insurers have been in the vanguard of this transformation of the balance sheet of the insurers. They have been really pushing into private credit. They’ve also really been pushing into asset-backed securities, CLOs. But we’ve seen non-private, equity-owned life insurers in recent years also increase their investments in these more opaque types of asset classes. I think following the lead of these PE-owned ones who have shown the proof of concept, which is that it turns out you actually can just invest in these much higher risk assets and you won’t be penalized for it by the regulators.
TEDDY DOWNEY: Yeah, that seems problematic. I think in the paper you still mentioned it’s a lot lower in terms of like the aggregate volume.
PRANJAL DRALL: Yeah.
TEDDY DOWNEY: And the other thing I think you’ve mentioned before is you’re not just competing on the asset side. If all of a sudden, you’re taking higher fees and taking higher rates and getting higher returns and things like that, you can offer better terms in the near term —
PRANJAL DRALL: Yeah, get more market share.
TEDDY DOWNEY:—for the annuities, get more annuities in. But are you going to be around in 10 years, 15 years, 20 years? We’ll see. And by the way, if you’re not, it’s the taxpayer and the policyholder who hold the bag.
All right, fourth looting. And by the way, we’ll get to questions. We’ve got some listener questions already. If you have questions, please put them in the chat or in the Q&A panel. We’ll get to them in a few minutes.
I just want to get through the rest of these looting strategies. PE firms may engage in valuation arbitrage with NAIC’s risk-based capital regulatory regime to make their private credit assets look systemically less risky than they actually are. What are the strategies here?
ANDREW GRANATO: Yeah., So, three and four are a pair. So, we’ve been talking about them together. But the point I would add here is this mechanism called a private letter rating that has been the subject of a lot of controversy and the subject of a lot of recent literature in economics. A private letter rating is—as the name suggests—a rating that is itself private.
So, when insurers have to go to the regulators and make attestations about the level of risk that they’re holding on their balance sheets, what they do is they go to a credit ratings agency, which is a private company. It’s just a different private company that does this sort of work. And they say, please give us a letter rating for the amount of risk that this asset entails. That agency gives a rating. And then the insurer reports that rating to the NAIC.
And the private letter rating is private to the ratings agency, the insurer, and the NAIC. So, no one else is observing what these private letter ratings are. And these have been the subject of so much controversy, in particular, because of a ratings agency called Egan-Jones, which there’s been a lot of reporting on Egan-Jones. This is a very small rating shop that has churned out a vast quantity of these private letter ratings. And to the extent that people have been able to do empirical studies on them, they continually, over and over, find wild levels of optimism.
Some ratings studies estimate that there’s maybe three notches on average worth of optimism. And so, Egan-Jones is quite controversial. The Bermuda Monetary Authority has data that they will no longer use them. They’re no longer acceptable for use under their jurisdiction.
But I think it would also be incorrect to lay the blame at the foot of a single ratings agency. All ratings agencies suffer from that agency problem that was talked about a lot in 2008. Because banks have a somewhat similar system. They also have credit ratings agencies. And when the bank—or in this case the insurer—is the one paying you have incentive to keep them happy. You, as the ratings agency, have incentive to keep them happy. And what would make the insurer happy is if you said there’s very little risk in this rating.
There are incentives pushing in the other direction, of course, if your credibility over the long run matters. And if you seem like you’re just not credible, there is a threshold upon which people will cut you off entirely, as Bermuda has done with Egan-Jones. But in these cases of highly opaque assets, where valuation is just inherently very difficult, that creates a lot of room to maneuver in terms of pushing these ratings up.
TEDDY DOWNEY: And you don’t get this in your paper, but NRSROs are regulated, I think, by the SEC. And that came up a lot, that the SEC should do some reform. And Dodd-Frank. That hasn’t occurred in any meaningful way really. Obviously, with Egan-Jones running amok here, as you pointed out, there is—and I wouldn’t be too enthusiastic about the SEC’s likelihood of doing anything now, just given the staffing levels there and just the willingness to deregulate or otherwise look the other way when it comes to accounting issues. But do you think NRSRO oversight, you know?
PRANJAL DRALL: Yeah.
TEDDY DOWNEY: You mentioned federal policymaking could get involved more broadly if there’s lack of scrutiny at the state level. Could even states that do securities regulation—I mean, anyone that looks at ratings, at the state or federal level, could poke around here in terms of what the quality of these ratings are and whether or not these NRSROs are doing something in the public interest or what have you.
PRANJAL DRALL: Yeah, that would be a very reasonable fix. The problem is just that any time the NRSROs exclude a rating agency or include an additional one, it’s not sufficient. The fundamental agency problem is that the insurance company/investment manager is paying them. So, they have an incentive to produce good ratings. So, the solution would be either auditing them or randomly sending some loans to the big three.
TEDDY DOWNEY: Encouraging other models. There’s investor pay models. There’s subscription models. I mean, yeah, this has been – basically, what I remember about Dodd-Frank, they threw out all these interesting proposals and did none of them. They just said, yeah, we’re actually okay. We’ll have a little bit more competition.
PRANJAL DRALL: And the problem with the competition idea is that like competition actually is the problem in some ways here, where like Demotech and Egan-Jones, their business model is we use AI to rate a ton of loans, faster than the big 3 and we specialize in private assets. So, since they rate tons of loans, fairly quickly, with probably less diligence than the big three do, that is a source of the problem.
TEDDY DOWNEY: It’s a race to the bottom. It’s not competition. It’s a race to the bottom. Yeah, competition without rules, you get a race to the bottom. I want to get to listener questions.
Let’s get to my last ways that PE firms loot their insurers. PE firms may expand the use of opaque shadow reinsurance that conceals risk. This really got me. This really got me. Because, in the paper, you show, look, these NAIC, these state regulators, they’re not really doing anything. They’re not really doing good oversight. I mean, they’re making reforms on the margins. They’re too little, too late. They’re just not effective. They don’t have enough staff. They don’t have enough resources. It’s a one at a time way that they look at things. There’s so much more sophistication on the side of the PE firm and the insurers. They’re just running circles around these state regulators.
But then in the paper, you say—and we’ve written a little bit about this as well—well, that’s not enough. The way that they’re engaging in valuation arbitrage—AKA duping the state regulators and NAIC capital requirements—that’s not enough. They need to go one step further and use shadow reinsurance. Have these sometimes captive reinsurers offload some of that some of these holdings, some of these policies, to a reinsurer and then book that as like an asset or whatever.
I’m not 100 percent sure I understand it. But the idea that you need to go even another layer further to trick the state regulator just seems mind boggling to me. Please tell us about this shadow reinsurance and whether or not you were surprised by how much of it was going on.
ANDREW GRANATO: Yeah, this, to me, is really egregious. I want to say I have a lot of sympathy for the problem that the NAIC has to deal with. I think it’s a very tough job trying to manage this situation. But I think the fundamental structure of insurance regulation in the U.S., where under the McCarranFerguson Act, this has like devolved to the states and then they’re trying to coordinate through the NAIC, is just a very difficult structure for trying to deal with the scale of the problem.
And shadow reinsurance is a very clear example of risk shifting and hiding risk. So, a reinsurance transaction just means that an insurer takes some of its liabilities, which are these actuarially estimated reserves, that it expects that it will have to pay out to some policyholders because some policyholders are going to die and then they’re going to have to make payouts to the beneficiaries.
So, we have some of these liabilities. And then we also have some assets on the asset side of the insurer balance sheet. And the insurer can transfer risk to just a different insurer. And in it of itself, that’s a totally legitimate transaction. The insurer might totally reasonably decide, well, I don’t particularly want to hold this exact risk. I’m just going to pay someone else to take on that risk for me. And if you were doing this in a competitive marketplace where the reinsurer is a separate company and they’re negotiating—we reach some negotiated transaction to value these contingent liabilities. We make that transaction—that’s all fine and good.
Captive reinsurance involves reinsurance to a subsidiary of the same company. So, if you have a family, you’re moving risk around within the family. But it’s still fundamentally all under the same ultimate corporate entity.
And shadow reinsurance, what that refers to is it refers to captive reinsurance to certain jurisdictions, because you can have the reinsurer be in a different jurisdiction than the insurer is, where there is systematic opacity. This involves most prominently Bermuda. States can also compete for reinsurance business by offering these sorts of terms. States like Iowa and Vermont notably do this.
And so, even though the data that you get on an insurer’s balance sheet is very strong, if you look at the insurer itself, once you reinsure to a shadow reinsurer, all of that data goes away. And so, you really have no idea then what the reinsurance company is doing with its assets. You have no idea if they have enough money to be able to pay for the expected liabilities that they are going to incur. And this I think is a really stark example of being able to just shift risk off of balance sheets in a way that makes it almost impossible for any external observer to be able to tell what’s going on.
TEDDY DOWNEY: I mean, it’s shocking to read about, to see how much is going on, that it’s allowed. I mean, there’s so many problems with this shadow reinsurance. I know that’s something we’re going to focus on writing about.
I want to get to listener questions really quickly. But before we do that, I wanted to ask, can you give us a few examples of reforms that you would like to see that you think would nip this type of conduct in the bud or address some of the problems that you’ve seen from what’s going on? And then also, have you gotten feedback from regulators, interested people, to reform this?
Because it seems like this could be a really big problem. I mean, I was saying before we started talking, it reminds me a lot of 2008. It’s a lot of assets we’re talking about. Yeah, maybe it’s not in the banking system, but seems like there could be a lot of contagion. It’s not exactly potential for run on the bank, but there are some potential elements like that, cascading issues. Maybe they couldn’t deal with multiple failures at once. Maybe they couldn’t deal with a big national failure. You get bailouts. It’s a sympathetic customer. I mean, you could see this going pretty haywire, similar to 2008.
So, would love to get your recommendations for reform and if you’ve started to hear any feedback from people who want to improve this situation.
PRANJAL DRALL: I can talk about one set of reforms. And I think we just got an email from a regulator. So, the timing there was great. So, basically there’s two sets of reforms.
One is meant to improve things pre-insolvency. So, what to do before any bad things happen. And then the second set of reforms would be, how do you make the guaranty fund system better that deals with insolvency? So, I’ll do the first part and Andrew can do the guaranty fund stuff.
So, the basic core problem, again, is that private credit is opaque, hard to value. So, a lot of the energy has already been spent by the NAIC to improve the situation. This sets of reforms are basically intended to create a better valuation regime.
Now, what does that mean? Well narrowly, it means more scrutiny of private letter ratings. More conceptually, there’s a fundamental difficulty in valuing illiquid assets so the problem is actually quite difficult and likely not fully solvable. You might imagine a better regime that like audits of certain rating agencies or maybe all the insurance companies pay into a fund that the state regulator then uses to get their own ratings from a different rating agency, like a big three or something.
Separately, the regulator can also just penalize insurance companies for holding opaque assets where the idea is that since it’s very hard to value illiquid things and complicated assets and the regulator is always behind the eight ball as they have to police each transaction and each loan. That’s very costly. So, instead of the regulator doing loanbyloan, they just say, look. If you hold X percent of your assets in opaque private credit, you’ll pay a premium into the deposit insurance fund today. Here, you do not care about how risky the assets truly are; it’s just a penalty for holding opaque assets in of themselves. This is meant to be a Pigouvian tax on opacity itself.
ANDREW GRANATO: Yeah, and another bucket of reforms, just to mention briefly, is aligning incentives in insolvency. So, one thing that you could do is you could say, well, if a life insurer goes bankrupt, you could give the receiver a claim against the insurance holding company, if there is one, for some percentage of the guaranty fund assessment.
So, to say that in English, if the rest of us have to pay for the policyholder bailout, the controlling company of the insurer should also be on the hook for at least part of that policyholder bailout. And so, what that does is it means that, ex-ante, prior to the insolvency, the controlling entity—say it could be like a private equity firm or there’s many other insurance holding companies—will be making decisions knowing that if that insurer were to go down, that they would be partially on the hook for the liabilities that are also spilling out to everybody else.
TEDDY DOWNEY: I want to have you guys back after your next paper so we can talk more about solutions. We’ve got some listener questions to get to. First one here, Brighthouse Financial, have you looked at that insurance company at all? It just says Brighthouse Financial question mark.
ANDREW GRANATO: Yeah. So, we actually spoke with the Charlotte Business Journal about the pending acquisition of Brighthouse Financial. I think it’s always very hard to comment on individual transactions because the facts in any individual transaction are always highly disputable.
And so, I think we would just note the incentives at play in that transaction or the incentives that are at play broadly speaking whenever a private equity firm acquires a life insurer and also has these buyout funds and private credit funds.
TEDDY DOWNEY: Another question here. Yesterday, Andrew tweeted information that expressed concerns about insurance companies’ use of leverage. Twenty to one and fifty to one leverage was mentioned. Can you please explain how this leverage is achieved?
ANDREW GRANATO: Yeah. So, insurance companies tend to be pretty highly leveraged, as are banks. What that entails from a financial stability perspective is that the higher the ratio of your debt is to your assets, the more fragile your company is in terms of the risk that you would ultimately go insolvent. Like, the less it takes for a decline in your asset value to push you into a technical insolvency where you simply do not have enough assets to cover your liabilities. And so, leverage here, I think, is quite important. Particularly as the asset side of these life insurers gets more and more risky, you start pushing these bounds more and more intensely.
TEDDY DOWNEY: And when you think about leverage, how are they getting so much leverage? Is it because they’re keeping as little capital against their liabilities? And why is it bigger now? Is that not regulated? Or how do they end up being more levered than typical? And yeah, we’ve got a question here. How does that leverage ratio compare to 2008?
ANDREW GRANATO: I looked at this a while ago. I don’t want to say specific numbers because I don’t want to get it wrong. But the idea here generally is that if you are a life insurer, your liability side is actually unique.
So, you do have direct creditors. So, investors can make direct loans to a life insurer. But the bulk of most life insurers’ liabilities are these actuarial reserves, which are an accounting construction where we’re trying to estimate, to the best of our ability, what the liabilities of a life insurer are. And you get these through the classical insurer function of risk pooling. Where if I’m an insurer, I sell 100 life insurance policies. Let’s say they last like ten years. Like some number of those people who I issued policies to are going to pass away. And then I’ll be on the hook for the payout to their beneficiary. But I don’t even know who those people are.
I’m making an estimate and that estimate gets more precise as the number of policies that I issue increases. And so, the calculation of those liabilities is a technical matter of actuarial science and accounting. But, in general, it is true that the equity on the balance sheet, which is just mechanically the estimated asset value of the insurer minus that liability calculation for the insurer. There’s not that much left over.
TEDDY DOWNEY: Got it. And the last question, then we’ll let you go. I think this question gets at the urgency of the problem, like how things could go wrong. What level of bond impairment needs to happen for it to trigger a financial liquidity issue? If we go into a recession and these loans marked at par go to 70 percent, how much systemic risk would that cause for the industry?
PRANJAL DRALL: I was just typing an answer to that. But the basic idea is it’s complicated. So, I there’s this paper from Michael Ohlrogge at NYU which isn’t public/finalized but the basic point is that they estimate that in under low/moderate stress scenarios, insurers would be fine given the current level of private credit exposure. But in extreme distress, say like impairments of 15, 20 percent, that would wipe out all the insurer equity. But it’s a very complicated question because you need to understand the correlation between private credit and different asset classes that the insurance company holds.
So, any sort of credit impairment will then affect both AAA safe bonds and private credit, which is inherently more risky. So, you just need to understand how they move together, which is a hard empirical question. And sophisticated researchers can disagree on that front.
So, it’s hard to give a great answer. But one can hand wavily think about the problem in the following way: 15 percent of the insurer assets are in private credit, and there’s leverage, and then under certain scenarios, a three or four percent impairment at the balance sheet level might be enough to wipe out equity. But again, this is hard to do empirically.
TEDDY DOWNEY I mean, my thing, it feels a lot more fragile to me. Just because you’re already bumping up against these NAIC risk capital rules. You’ve got the shadow insurance. All it takes is for some law enforcement person to come along and say, actually, you’ve got a lot of affiliated things here. Those are risky. You need more assets. That’s what’s happening with the guy who owns the Dodgers and the Guggenheim and Walters, whatever.
Or actually, these assets are marked wrong. You’ve got to make this more accurate. This is what happens with Blue Owl. You can see that, just the transparency alone, actually resulting in a lot of risk to the current model. Like no credit impairment needs to necessarily happen for some of these guys. Is that the wrong way to think about this? It seems very fragile.
PRANJAL DRALL: No, that’s right. So yeah, like in the Dodgers case, you need more nonaffiliated assets. You have to go sell the affiliated-Dodgers assets and then do something else. Again, this is where the insurance liabilities are in some ways a benefit because they’re so long dated. If a regulator spots problems early, it’s not like the problem is unsolvable. We have time to fix it. And an insurance company can slowly offload the bad asset or whatever the regulator is asking them to do. Unlike in other contexts where there’s more runnable deposits. Like in insurance, since there’s less runnability, you have more time to fix problems if there’s no credit impairment.
TEDDY DOWNEY: Oh, well, actually, I’ve got one last question. I promise I’ll let you go. The runnability. I actually did have a question about this. Because you mentioned three kinds of life insurance. You mentioned term. You mentioned annuities. And then there was this other one that involves investment-like accounts.
ANDREW GRANATO: Yeah, so cash value life.
TEDDY DOWNEY: Cash value life. And you can call that back. Like, you can get that back, right? If you’re a life insurance. So, if the life insurance policyholders get wind of you doing this stuff, and they’re like, actually, I have like a 401k-like asset in there. Just give it back to me. Does that mean that there could be a run-like situation at some of these guys? Or is that amount of money in the insurance company not enough to come into play in that respect?
ANDREW GRANATO: Yeah. So, this is the next paper.
TEDDY DOWNEY: Ahhh.
PRANJAL DRALL: Yeah. So, how permanent is permanent capital? It depends. I think it depends a lot on the kinds of product, like the different types of product lines, what percentage they are of the insurer.
And then here’s just one example. So, if you have an insurer that sells a lot of low-value policies. Let’s say almost all of your life insurance policies are $150,000 to $200,000 of death benefit or maybe much lower. Maybe you do just like $10,000, $20,000 death benefit policies. All of those policies are going to be fully covered by every state’s guaranty fund. So, there’s no incentive for any policyholder to ever run from that.
But if you have a ton of very high-value life insurance policies, $1 million, $2 million, more, in death benefit, well, then that 200K, 300K of protection starts to look paltry. And so, those two life insurers, I think, are in a very different position when it comes to what incentives their policyholders have to try to retain attachment to that insurer as that insurer hits financial distress. So, we’re going to have another 60 pages about that, hopefully, in the next few months.
PRANJAL DRALL: And I’d also answer Ben’s question who asked about the government-run life insurance program. So, I think, as Andrew was saying, there’s these policies where the policyholder is basically getting the returns of a 30-year Treasury bond. Maybe the government can run well and they already do that through the life insurance problem for federal employees which is run by the government and all the investments are in government bonds and it seems to work well.
But it doesn’t work well if you want like these more exotic policies that invest in like high-yielding assets, so you have a higher payout at the end. There I don’t really think the government could run like a sophisticated life insurer. Because even the federal employee one is partially outsourced to MetLife.
So, yeah, in short, I’m skeptical of an idea like that. Additionally, the other consideration is that there’s a lot of competition in life insurance. In the grand scheme of things, minus all the concerns above, the industry runs ok. It’s just we need to figure out the solvency regime and deal with this problem more directly.
TEDDY DOWNEY: Yeah, it runs fairly well until it creates a credit crunch and financial collapse. But I am excited for your next paper. I would love to have you back. This was incredible. Again, one of the most informative, interesting, readable for an academic paper. Just so great. I can’t recommend it enough. Please go read it. Please, please, please read it.
And I can’t thank you enough for joining us. Pranjal, Andrew, this was a true pleasure.
ANDREW GRANATO: Thank you so much for having us.
PRANJAL DRALL: Thank you for having us on.
TEDDY DOWNEY: All right, and thanks everyone for joining us. This concludes the call. Bye-bye. Bye.