Money Stuff: Hedge Fund Charged Clients for Art

I write from time to time about the economic structure of modern multi-strategy “pod shop” hedge funds. Stereotypically a hedge fund invests its clients’ money and gives the clients the returns, after taking some fees — classically 2% of assets and 20% of returns — for itself. The modern pod shops are not like that. They charge “pass-through fees,” meaning that they invest their clients’ money, earn returns, pay their expenses out of the returns, pay their employees large performance-based bonuses out of the returns, and then give most of what’s left to the clients. The deal with the clients is not “we get 20% of the returns and you get 80%.” The deal with clients is more like “we aim to meet our cost of capital by giving you a risk-adjusted return, net of fees, that you are generally happy with.”

In this, I have argued, the hedge funds are like banks, or really like any other business: They do not give their investors (bank shareholders, hedge fund limited partners) a fixed share of their revenue; instead, they (1) pay their employees what they need to pay in a competitive market for talent and (2) aim to compound returns for their investors at a rate that exceeds their cost of capital. The “cost of capital” is a somewhat hypothetical thing. [1] But in the long run, either you give your investors enough returns to make them happy, and your fund grows, or you don’t, and the investors take their money back. And unlike shareholders in a bank, hedge fund investors can — eventually — take their money back.

The analogy is not exact. The clients of a hedge fund are not actually shareholders of a business. They have a contract — a limited partnership agreement — with the hedge fund, and the contract lists what expenses the fund can pass through to the clients, and the list is not literally “whatever we want.” The hedge fund has to disclose in advance the specific categories of things that it might spend client money on. In February, Bloomberg News published a fun story combing through big hedge funds’ filings to find what they are allowed to spend pass-through fees on, and though the short answer is “pretty much everything,” the long answer is long. A partial list [2] :

… 401(k) matching, administrators, bonuses, buyouts of prior deferred compensation, certain employee perks, costs relating to resuming in-person work following the Covid-19 pandemic, employee gifts, industry conferences, professional development related expenses, severance arrangements, bank service fees, brokers, class action recovery and monitoring, fees and expenses paid to prime brokers, filing fees, general partner liability insurance, marketing, public relations, valuation agents, commuter reimbursement plans, cost of first-class airfare, drinks, gathering for employees and their guests, snacks, Bloomberg terminals, corporate access charges, research, software, trading systems, market data, internet, data center costs, rent, security deposits, utilities ...

It’s pretty much everything you might need to run a hedge fund business, but not literally everything. A bank’s managers use their business judgment to decide what to spend money on, and all of the money they spend is effectively charged to the shareholders. A pass-through hedge fund’s managers mostly do the same thing: They use their business judgment to decide what to spend money on, and they spend it, and then by default they charge it to the clients. But every once in a blue moon they will spend money on something arcane or weird or aggressive — a private jet, say — and they will go to charge it to the clients, and someone in the finance or compliance department will pipe up and say “uh actually we have a long list of stuff we can charge to our clients, but this isn’t on the list.” And then one or more of the following things will happen:

• They’ll charge it to the clients anyway, and the compliance person will grumble.
• They won’t charge it to the clients — the management company (which is in most cases mostly owned by the guy who founded the hedge fund) will pay for it — and the manager will grumble.
• They’ll put it on the list, for next time. And maybe the clients will be like “oh sure, we’re paying for hundreds of millions of dollars of portfolio manager bonuses, what’s one private jet.” But maybe the clients will grumble. “No come on we’re not buying you a jet.” 

The constraint on the hedge fund is not only that it has to earn its cost of capital; it also has to pass through expenses to clients that the clients can stomach. [3]  The clients can stomach a lot! But there might be limits.

Anyway:

Capula Investment Management’s former chief US compliance officer accused the hedge fund firm of wrongfully firing him for raising concerns about the expensing of artwork and private jet travel as well as certain trading activity.

Igor Abramov sued London-based Capula late Monday in Manhattan federal court, claiming he was subject to retaliation and terminated in July after suggesting the $32 billion firm might be violating Securities and Exchange Commission regulations on pass-through expense disclosure, conflicts of interest and risking investor funds.

Can a hedge fund charge clients for the cost of buying art to hang on its office walls? I mean, as a conceptual matter, sure, why not. (Banks buy office art with shareholder money.) But it has to be on the list, and if you put it on the list there is some chance that the clients will push back, saying “oh come on you can’t use our money to buy yourself art.” But you don’t want to come to work in an office with bare walls. And you don’t want to spend your own money if you can avoid it. It is a conundrum.

Here is Abramov’s complaint. Capula moved from more of a 2-and-20 model to more of a pass-through model, because pass-through fees have become the norm in recent years:

In December 2021, Capula launched its Capula Multi-Strategy Master Fund Limited (the “MS Fund”) which contained a “pass-through expense” (“PTE”) structure pursuant to which Capula would be able to charge, without any limit, directly to the MS fund’s underlying investors ... a broad range of its operating “overhead” costs. The expenses charged by Capula to the MS Fund investors via the PTE mechanism included multi-million “sign-on” and compensation packages of Capula’s portfolio managers, hundreds of thousands of dollars in placement agent costs for recruiting such portfolio managers, as well as significant compensation (totaling millions of dollars) for other Capula employees, partners and consultants who performed “work in relation to the MS Fund”. ...

Capula induced investors to invest in the MS Fund with the uncapped PTE structure on the basis that by agreeing to allow Capula to charge these additional uncapped multi-million dollar expenses to the MS Fund investors (in lieu of typical, capped management fees), Capula would be able to retain the most talented, and generally highly compensated, portfolio managers, which would ostensibly deliver stronger returns for the investors.

Right, that’s the deal: If you give us discretion to pay portfolio managers as much as we want, using your money, we will get better portfolio managers, who will make more money for you. But there are details to worry about:

In fulfilling his compliance officer duties, Plaintiff emphasized that, under the SEC’s federal regulations, Capula’s expenses must be clearly and specifically disclosed to the investors in the relevant fund’s offering prospectus prior to Capula initiating such charges. As part of his compliance communications, Plaintiff pointed out that the SEC had recently adopted even more stringent regulations regarding fund expenses and disclosures to ensure greater investor protection and prevention of inappropriate conduct by investment managers that would breach their fiduciary duties to investors.

He claims that at some point he found out that “Artwork and Private Air Jet Travel expenses had been charged to the MS Fund investors” without disclosure. So he “instructed” his bosses that, “pursuant to Capula’s fiduciary duties under the SEC regulations, Capula must now notify the MS Fund investors regarding any expenses that had previously been charged to them by Capula without their knowledge, including Artwork and Private Air Jet Travel,” and that it would have to reimburse the investors for those charges. And they said no:

Mr. Corsalini strongly disagreed with Plaintiff and indicated that Plaintiff was not sufficiently “commercial” in his U.S. compliance role and did not understand the current “market practices” with respect to the PTE issue …

Plaintiff was told he was raising immaterial issues and was referred to on multiple occasions as not “commercial.” …

During an annual performance review conducted by [Capula Chief Legal Officer Jonathan] Bloom with Plaintiff via telephone on October 30, 2024, Mr. Bloom indicated to Plaintiff that management was quite unhappy with Plaintiff’s handling of the PTE issues and that new management, specifically, [Chief Executive Officer Enrico] Corsalini, believed that Compliance must be more “commercial.” Mr. Bloom informed Plaintiff that if Plaintiff failed to dramatically change his “inflexible” compliance approach, Mr. Corsalini would restructure U.S. Compliance to get Compliance in the U.S. “off his back”, that it would impact Plaintiff future- and compensation-wise, and that Plaintiff’s role and compensation would be severely diminished.

I have to say that I kind of sympathize with the managers here. Generally speaking I see their point that “current ‘market practices’ with respect to the PTE issue” are kind of “ehhh pass through whatever you want,” and it is not especially commercial for your compliance officer to insist that you carefully disclose everything. But he’s a compliance officer! That kind of is his job.

One casual way to describe the modern economy is that:

1. Artificial intelligence labs like OpenAI and tech “hyperscalers” like Meta Platforms Inc. have committed to spend, collectively, trillions of dollars to build out AI infrastructure.
2. They don’t have trillions of dollars.
3. But they will pay for it over time with their hundreds of billions of dollars of AI-related revenue.
4. They don’t have hundreds of billions of dollars of AI-related revenue either.
5. But they will.
6. You can tell because of their valuations. OpenAI’s equity is worth $500 billion. The only way its equity is worth $500 billion is if its future cash flows are sufficient to (1) cover its hundreds of billions of dollars of spending commitments and (2) leave hundreds of billions of dollars left over for its equity investors.
7. That is, all of the future AI cash flows are not directly observable — you can’t extrapolate them from current AI cash flows or anything — but they can be inferred from equity-market enthusiasm for AI.

This is, you know. This is fine. This is the point of financial markets, to tell you what the future will be like. Modern AI has a science fiction flavor to it, and its future economic impact is, at this point, necessarily speculative. “It may be difficult to know what role money will play in a post-[artificial general intelligence] world,” says OpenAI, but they would say that. The market is telling you that, in a post-AGI world, a lot of money is going to flow to AI equity investors. Which also implies that there will be a lot of cushion for AI creditors. If you lend money to finance a data center for the AI boom, you can get an A+ rated bond and a 6.6% coupon, not because AI data centers have a long track record of paying back their debt but because the companies here are so giant that surely the debt is safe. 

Still it could make you nervous. It could make you nervous if you are an AI skeptic. But it could also make you nervous if you are an AI lender. If you are basically in the business of lending billions of dollars to build data centers and earn 6.6% returns, then your upside is, you know, 6.6% returns, and your downside is that all of this is massively overblown and the data centers will be full of tumbleweeds in three years. You might think about hedging. The Financial Times reports:

Deutsche Bank is exploring ways to hedge its exposure to data centres after extending billions of dollars in debt to the sector to keep up with demand for artificial intelligence and cloud computing. 

Executives inside the bank have discussed ways to manage its exposure to the booming industry as so-called hyperscalers pour hundreds of billions of dollars into building infrastructure for their AI needs that is increasingly funded by debt. ...

Sceptics have pointed out that billions of dollars have been deployed in an untested industry with assets that quickly depreciate in value due to the rapid change in technology.

Deutsche has lent predominantly to businesses that service hyperscalers such as Alphabet, Microsoft and Amazon, and the debt is secured against long term contracts that promise steady returns, according to two people familiar with the bank. 

Right, yes, the debt is secured by long-term contracts with the hyperscalers, but your confidence in those contracts will depend a bit on your views of how well AI will work out for them. But how do you hedge?

The German lender is looking at options including shorting a basket of AI-related stocks that would help mitigate downside risk by betting against companies in the sector. It is also considering buying default protection on some of the debt using derivatives through a transaction known as synthetic risk transfer (SRT). 

SRT, sure, fine, that’s just economically transferring the credit risk of the debt to other investors. But shorting AI stocks is the interesting hedge. On the one hand it is terrifying, and you have to size it right: If the AI boom is bigger than people expect, then those stocks will go up a ton (and you’ll lose a ton on your hedge), while the debt will still just pay 6.6%. On the other hand, there is a logic there. All of this AI debt is in some sense predicated on AI equity; the equity values of AI companies are how you know that the debt is sustainable. So if the debt makes you nervous you can short the stocks.

Some people buy stocks, hoping that they will go up. Other people sell stocks short, hoping that they will go down. To sell stocks short, you borrow some stock from somebody who owns it, you sell it, you wait for it to go down, you buy it back, and you return it to your stock lender, paying the lender a fee for the loan. The stock lenders are essential to the transaction.

Which is maybe a little weird. You are betting against the stock, you are hoping it will go down, you are perhaps even trying to drive it down. But the stock lenders are people who own the stock, and presumably they want it to go up. You are, in some sense, enemies. If there were no short sellers, presumably the stock’s price would be higher, [4] which would be good for its owners, including the stock lenders. Why should the stock lenders lend you their stock so that you can do your dirty little short selling?

And in fact, in various meme-stock episodes, you will sometimes see efforts to get people to stop lending their stock to short sellers, with the explicit goal of forcing them to buy back their shorts and push the stock price higher. [5]

But for the most part, if you want to short a stock, someone will lend you the stock. Why? There are various possible answers, [6] but an important one is: Some stock lenders don’t actually care if the stock goes up. In particular, stock index funds are not exactly in the business of buying stocks that will go up. Index funds are in the business of (1) giving investors the performance of the stock index and (2) charging extremely low fees. If an index fund can lend out its stock holdings, it will collect some money from the borrowers, which can cover some of its expenses or improve the returns to its investors. If its stock lending makes the stocks go down, that’s not the index fund’s problem: The stocks go down, so the index goes down, but the index fund’s investors get exactly what they expected, which is the performance of the index. 

And so, customarily, stocks with a lot of index-fund ownership are relatively easy to short (the index funds will cheerfully lend you their shares), while stocks with a lot of, for instance, founder-CEO ownership are hard to short (the founder-CEO probably won’t lend you her shares, and will get mad if you ask).

Incidentally this is a nice counterpoint to the argument that you sometimes hear that index funds make stock prices less efficient. That argument goes something like: “Index funds buy stocks indiscriminately, and in fact they tend to buy high (as companies get big and are added to indexes) and sell low (as they get smaller and drop out). As markets are dominated by index funds, there will be no one left to make prices correct.” But the point here is that index funds reduce the constraints on short sellers, which has a tendency to make prices more efficient: If some stock is overvalued, short sellers can step in to sell it, because they can borrow the stock easily.

Still, this might annoy you a little bit if you are an investor in index funds. (You don’t just want the performance of the index; you also want the index to go up.) But not as much as it might annoy the Bank of Japan, which spent most of the last 15 years buying equity exchange-traded funds to push up Japanese stock prices, lower the cost of capital and stimulate the economy. (This program started in 2010, and the BOJ announced in September that it would start unwinding it.) Here is a fun paper from Mitsuru Katagiri, Junnosuke Shino and Koji Takahashi, titled “To Lend or Not to Lend: The Bank of Japan’s ETF Purchase Program and Securities Lending”:

This study investigates the role of passive investors in the equity lending market by utilizing the expansion of exchange-traded fund (ETF) markets due to the Bank of Japan’s (BOJ) ETF purchasing program. We find that the BOJ’s purchases increase equity prices particularly for stocks with limited availability in the equity lending market. However, over the longer term, the BOJ’s cumulative purchases reduce lending fees, thus weakening the program’s effects. These findings suggest that ETF managers supply stocks that constitute ETFs to the equity lending market, and the lending behavior of ETFs, influenced by the BOJ’s program, alleviates short-selling constraints.

I mean they definitely pushed the prices of stocks up, by buying half a trillion dollars of ETFs. But they pushed them up by less than they would have if they had not loaned those stocks out. From the paper:

Our empirical analysis finds that an increase in the share of passive investors raises the supply of lendable stocks. In addition, we find that the flow of the BOJ’s purchases has an insignificant effect on lending fees in the short run, suggesting that the upward shift in the demand curve caused by the BOJ’s purchases offsets the increase in the supply of lending stocks. … We further show that the cumulative increase in passive investment significantly reduces lending fees over longer horizons, thereby helping to ease short-selling constraints.

It apparently got easier to short Japanese stocks as the Bank of Japan started buying all of those stocks.

The US Supreme Court held oral arguments on President Donald Trump’s blanket tariffs today. I have written about the case a few times, and it has always seemed very clear to me. The US Constitution says that only Congress can impose taxes. Tariffs are taxes. These taxes were imposed by the president, not by Congress. There is a statute — IEEPA, the International Emergency Economic Powers Act of 1977 — in which Congress delegated to the president some power to regulate international commerce in an emergency. But:

1. The power to regulate international commerce does not necessarily include the power to impose taxes. (The Securities and Exchange Commission regulates stock trading, but can’t just unilaterally impose taxes on it.)
2. IEEPA requires an emergency, an “unusual and extraordinary threat” to US national security. Surely all trade with all foreign countries cannot be an unusual and extraordinary threat. 

So IEEPA does not authorize tariffs, if it did authorize tariffs it would not authorize these tariffs, and even if it did that might be an unconstitutional delegation of Congress’s taxing power. It just seems like a very easy case!

On the other hand, if the Supreme Court strikes down the tariffs, Trump will post mean things on Truth Social, which is also an important part of the legal realist analysis. The Constitution does not operate itself, and the Supreme Court is “not final because we are infallible.”

But it still seems like an easy case to me, and from what I have seen it seems like an easy case to the Supreme Court as well. “Supreme Court Appears Skeptical of Trump’s Tariffs” and “Key Justices Show Early Skepticism of Trump’s Tariff Power” are some of the headlines this afternoon, and the Kalshi and Polymarket probabilities of a Trump win fell from about 40% to about 30% during the argument. 

Trump does love tariffs, though, and there are other statutes that do more clearly give him the power to impose tariffs. These statutes have more restrictions; they are limited in size or time or require fact-finding and due process. A possible long-term outcome here is that Trump will mostly get to impose his tariffs, but he will have to do it by following the law. 

Wall Street Bonuses Projected to Jump for Second Straight Year. Equity Traders Seen Grabbing This Year’s Biggest Bonus Hikes. JPMorgan discloses US inquiry into alleged debanking practices. Carl Icahn, Nearing 90, Is Still Trying to Right His Empire. SBF Faces Skeptical Judges at Appeal of FTX Fraud Verdict. Ripple Says Fortress, Citadel Securities Invest $500 Million. ‘Casino-Like’ Market Powers Robinhood to 450% Gain in Trump Era. Does momentum exist in prediction markets? Eighteen arrested in connection with alleged global online fraud network. Wall Street Drops Fear of ‘Hot Commie Summer’ in Overture to Mamdani.  Trader Jeff Yass Is Giving $100 Million to ‘Anti-Woke’ University of Austin. Burger King Braces for the Demise of the Penny. UK Police Recover £12 Million in Dinosaur Bones From Laundering Suspect. The Great British Bake Off Is a Great British Betting Zone. “Lairmore said co-workers gave him a sandwich-shaped plush toy and a patch reading ‘felony footlong’ as gag gifts.”

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!