Money Stuff: Insider Trading Is Not About Fairness

The basic story of Archegos Capital Management is that it borrowed billions of dollars from banks to buy huge positions in like seven stocks, pushing up the prices of those stocks and creating huge paper gains for Archegos, which it then used to borrow more money to buy more of the stocks, pushing up the prices some more, etc. This was nice while it lasted, but it couldn’t last. Eventually, in March 2021, one of the stocks went down a bit, the banks sent Archegos some margin calls, it had no extra money, the banks foreclosed, the stocks went down and the whole thing collapsed. Archegos went to zero, its founder got 18 years in prison, and some of the banks lost billions of dollars.

Not all of them. At some point, the banks all discovered that Archegos (1) owned huge levered positions in like seven stocks (with exposure to “anywhere from 30-70%” of each stock), (2) had borrowed from multiple banks to buy those positions and (3) was in the process of collapsing. They arguably did not know any of those things until the collapse was well underway. But there was a brief window of time in which:

1. The banks knew this, but
2. The market did not.

In particular, the banks held lots of shares of Archegos’s seven stocks (ViacomCBS Inc., Baidu Inc., Discovery Inc., etc.), which served as collateral for their loans to Archegos. [1]  They knew that Archegos would not pay back their loans, so each bank knew that it — and every other bank — would have to sell the collateral. They knew that all of this selling would crash the prices of the stocks. If news came out like “hey, Archegos is collapsing, it owns a zillion dollars worth of Viacom and Baidu and its banks are going to be liquidating those positions,” then the stocks would drop before the banks sold: Everyone would know that big sales were coming, so nobody would want to buy the shares.

But if the banks sold before the news came out, they might get away with it: They could sell quietly before the market caught on to the problem, and they could perhaps sell the shares for more than Archegos owed them. And in fact some banks moved quickly and did fine, and other banks moved slowly and lost billions of dollars. [2] (Basically the fast banks’ sales alerted the market that something was going on, and the market caught on before the slow banks could sell.)

This might trouble you. When the banks were selling out of their Archegos positions, they knew some pretty important information that the buyers didn’t know. (They knew about Archegos’s positions, and its collapse.) When the news came out, those stocks fell; the people who bought the stocks from the fast banks lost a lot of money, while the fast banks avoided those losses. Is that … insider trading?

Well! Insider trading, I like to say around here, is not about fairness; it’s about theft. It is generally legal, in the US, to trade on information that no one else has. What is illegal is misusing someone else’s information. A chief executive officer who trades on inside information about her own stock is misappropriating that information from her shareholders [3] ; a therapist who trades on what the CEO tells him in a therapy session is misappropriating that information from his patient. But if Warren Buffett knows that his purchases of a stock will move the stock up, he is allowed to buy the stock without first disclosing his plans to buy it: Trading on your own secret information is fine. (Not legal advice!)

What about here? The banks were trading on material nonpublic information about Archegos. Did they have a duty not to trade on it? Some investors sued to find out, “alleging that they traded in the Issuers’ stocks at the same time the [banks] were selling their Archegos-related positions,” that they lost money to the banks, and that the banks were doing insider trading.

Today they lost in a federal appeals court. Here is the opinion. To be insider trading, the court writes, there has to be evidence that “either (1) Archegos owed a fiduciary or fiduciary-like duty to the Issuers’ shareholders or (2) [the banks] owed a fiduciary or fiduciary-like duty to Archegos.” The first is clearly not true: Archegos was an outside shareholder of its companies, and had no special relationship or inside information.

The second possibility — that the banks had some duty to Archegos not to trade on their information about its collapse — is more plausible, but the court rejected that too:

Even assuming, without deciding, that the news of Archegos’ collapse constituted material non-public information, Appellants’ claim still fails because [the banks] did not owe any fiduciary duties to Archegos. …

The [complaint] does not allege that [the banks] entered into an agreement with Archegos to act in its best interest or that [the banks] ever agreed to serve as Archegos’ fiduciaries. Rather, the [complaint] alleges only that [the banks] offered Archegos various brokerage services. More importantly, the [complaint] acknowledges that [the banks] were contractually entitled to sell their Archegos-related positions upon Archegos’ default. This type of commercial arrangement between adverse parties indicates that [the banks] negotiated with Archegos at arm’s length. 

This strikes me as completely correct. The whole point of lending money to a hedge fund collateralized by its stock positions is that, if you send the hedge fund a margin call and it doesn’t post more money, you can blow out of the stock before it collapses. If you had to disclose “hey our customer is collapsing and we gotta sell its stock,” the stock would be worthless as collateral.

Still it is quite harsh on the buyers! They bought stock for way more than it was worth, because the banks knew something that they didn’t. Doesn’t seem fair. But the lesson is that insider trading isn’t about fairness.

My high-level model of private credit is:

1. Banks have short-term, unpredictable liabilities. (They get money from depositors, who can ask for it back at any time.) Banks also make long-term, illiquid loans: They make business loans with terms of years, or mortgage loans with terms of decades. That’s traditional, but also kind of weird and risky: If the depositors all ask for their money back, the bank will have a hard time getting it. 
2. Life insurance companies, meanwhile, have long-lived, fairly predictable liabilities: They sell life insurance (they get money now and pay it out in a lump sum when the customer dies) and annuities (they get money now and pay it out over time, perhaps until some fixed end date, or perhaps until the customer dies).
3. So shouldn’t the life insurance companies make the loans?
4. “Private credit” is the hot modern way to say “the life insurance companies make the loans.” Sometimes this takes the form of an alternative asset manager like Blackstone Inc. raising money from life insurance companies (in funds or separately managed accounts) and making long-term, illiquid loans. Sometimes it takes the form of an alternative asset manager like Apollo Global Management owning a life insurance company (Athene, in Apollo’s case) and using its money to make long-term, illiquid loans. But the basic point of private credit is to make the long-term illiquid loans with life-insurance premiums rather than bank deposits.

This is all very sloppy and oversimplified in ways that I will summarize in a footnote, [4] but I do think it captures the essential points.

One limit on this, for an insurance company, is that you can only invest as much money as you have. If you have $100 of policyholder cash to invest, you might invest $100 in private credit (ignoring capital requirements, etc.). If you earn 10% on your private credit and have a 4% cost of capital on your policies, that’s a great trade. You might want to scale it up. Why not get $500 of policyholder cash, so you can invest $500 in private credit and earn more money? But that has problems: You have to hire salespeople to go out and find customers to buy annuities, etc.; your costs scale up with your assets.

What you really want is to sell a huge annuities: If you sell a billion-dollar annuity, then you can massively scale up this trade. Probably none of your customers need a billion-dollar annuity. But what if you found institutional investors to buy annuities? Sell the billion-dollar annuity to, like, Pimco.Of course you will tweak the terms of the annuity for Pimco (it’s not going to die?), but the basic structure — “you give us money today, we give you back money over time for some fixed period” — is annuity-like. Say, “you give us $1 billion today, we give you back $50 million per year for five years and then $1 billion at the end.” [5] That kind of sounds like an annuity! You sell a big annuity to Pimco and you get back a billion dollars to invest in private credit. If the annuity costs you 5% and the private credit yields 10%, it’s a great trade.

Of course, that trade also sounds like a bond, which is why Pimco will buy it. To Pimco, paying $1 billion today to get back 5% per year for five years, and the principal at maturity, is a pretty normal trade. (A bond trade.) And if you are a good stable highly-rated insurance company, it will be happy to do that trade with you.

And then there are some regulatory boxes to check. It turns out that, for insurance regulatory and ratings purposes, it is better to “sell annuities to customers” than it is to “sell bonds to investors,” so you want this trade to be classified as an insurance liability (like an annuity). Pimco’s preferences are more subtle: It wants to own a bond, not an insurance policy, but it also wants a senior claim on your assets, and it turns out that your insurance policyholders get repaid before your bondholders. So Pimco wants something like “an insurance policy wrapped in a bond.”

At the Financial Times, Lee Harris reports on funding agreements:

Enter funding agreement-backed notes (FABNs), the largest of a booming class of so-called institutional spread products with their roots in guarantees pioneered by the co-founder of Apollo’s in-house insurer, Athene.

At its heart, a funding agreement is a simple arbitrage: it is treated like an insurance liability for ratings purposes but raises capital like a bond.

Insurers such as Athene use FABNs to offer a guaranteed return to big investors such as Pimco. Asset managers such as Apollo, which fully merged with Athene in 2022, can increase the assets on which they earn management fees, and write more loans. In June, Apollo chief financial officer Martin Kelly praised funding agreements as a source of “really cheap financing”.

“It’s Apollo’s asset sourcing engine that’s driving its FABN issuance,” said Thomas Gallagher, an insurance analyst at Evercore. 

The “asset sourcing engine” means that it makes a lot of private-credit loans, and to get the money to make those loans it sells FABNs. Harris explains:

To issue an FABN, a life insurance company sets up a shell company and sells it a “funding agreement” which provides for principal and interest payments. 

The shell company then sells smaller notes to institutional investors such as banks. The notes promise a modest rate of return and mature after a fixed period, typically three to five years. 

The funding agreements that back the notes are unsecured obligations of the insurance company, meaning the insurer typically does not set aside collateral to back them. Investor prospectuses also stipulate that the notes are not “guarantees”.

But rating agencies have taken the view that, in the event of default, the funding agreement would have a senior claim on insurance company assets — equivalent to an insurance policyholder, and ahead of general creditors. … Moody’s has treated the agreements as equivalent to insurance obligations — in effect rating the insurers as though they were selling more policies, rather than issuing corporate debt.

Right, it’s just selling an annuity to Pimco. Wrapped in a bond.

I wrote the other day that “the basic situation in asset management is that the alternatives managers have the products and the traditional managers have the customers.” If you manage private credit assets, you are pretty happy with your business, which is quite hot and charges high fees, but you would like to grow by selling to more individual investors. If you manage traditional mutual funds, you have millions of individual investors as customers, which is nice, but they are all getting sick of paying for mutual funds and you’re looking for some way to keep them (and charge them high fees). The benefits of a combination are obvious — the alternatives managers have the products people want, the traditional managers have the people to buy the products — and so there have been many combinations. Traditional managers have started private-assets businesses, or acquired private-asset managers, or signed up for distribution deals with private-asset managers.

They don’t always work. Bloomberg’s Loukia Gyftopoulou and Silla Brush report on the problems:

Traditional fund houses such as Franklin Resources Inc., Invesco Ltd. and State Street Corp. also are trying to find ways to sell hotter products that generate higher fees, through acquisitions or otherwise — but it’s not easy. Some have been left contending with wide gaps in executive pay or cultural divides in the ranks. Salesforces built to pitch plain-vanilla products proved slow to move riskier private-market offerings. And the competition is formidable, with incumbents such as Blackstone Inc., Apollo Global Management Inc. and Ares Management Corp.

For instance, Franklin Resources now owns alternatives managers including Benefit Street Partners, but the combination is awkward:

Franklin lets the boutiques it buys operate independently. Most still have their own CEOs, and some their own offices. In interviews, people who have worked in those outposts said they had little to no contact with Franklin, and a couple couldn’t remember the name of its CEO, Jenny Johnson, of the founding family.

In tense meetings attended by Franklin staffers, executives from Benefit Street openly expressed dissatisfaction over the pace of progress in the first few years after their merger, people who were present said. Benefit Street executives complained that Franklin’s sales fleet wasn’t doing enough to sell their funds to the multitrillion-dollar market of wealthy clients. Some executives even mentioned the millions of dollars they were earning a year, underscoring their clout to Franklin managers earning much less, witnesses said.

Right, obviously the Benefit Street people would make even more millions of dollars if the Franklin salespeople could find more investors for them, but I’m not sure why that would motivate the salespeople. “To spur sales, Franklin at one point upped the commissions for those funds,” which I guess makes more sense.

I wrote yesterday about Elon Musk (1) buying $1 billion worth of Tesla Inc. stock and (2) thereby increasing the value of his Tesla stake by $17 billion. Nice trade! I added that Musk probably isn’t cashing out, though: “He moves his stock price for fun, not for profit.” But I said two careless things about this.

First, I wrote: “If Musk files a disclosure tomorrow saying that he sold $17 billion of stock today, I’ll be impressed.” To be clear, I was saying that he wouldn’t do that, but if he did it would be funny and cool. But of course that’s wrong: Musk is a Section 16 insider of Tesla, meaning that if he went and sold stock so soon after buying it, he would have to disgorge his profits. So it would not actually be an impressive trade.

Second, I wrote that “Musk has pledged some of his shares to secure personal borrowing, and I guess today’s stock pop increases his borrowing capacity.” Again, I was saying that Musk probably didn’t plan to use that increased borrowing capacity, but in fact it seems that his capacity did not increase. Tesla “has a policy that limits pledging of Tesla stock by our directors and executive officers,” and it limits Musk’s borrowing to “the lesser of $3.5 billion or twenty-five percent (25%) of the total value of the pledged stock.” I don’t know how much Musk has borrowed against his Tesla stock, but the point is he’s not allowed to borrow more than $3.5 billion against his $170+ billion of Tesla stock, so an extra $17 billion doesn’t actually increase his borrowing capacity.

Franklin Templeton predicts slowdown in investment management deals. Fed ‘ Third Mandate’ Forces Bond Traders to Rethink Age-Old Rules. Robotics Startup Figure AI Valued at $39 Billion in New Funding. Can Tokyo tempt finance workers to manage its trillions? 'Culling of the herd': Consulting could be headed for a decadelong shift that'll make it harder to climb its career ladder. Tesla Faces US Auto Safety Investigation Over Door Handles. Bessent Jokes About Hamilton-Burr Duel When Asked About Pulte.

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Or you can subscribe to Money Stuff and other great Bloomberg newsletters here. Thanks!