People Will Pay for Illiquidity

Illiquidity provision

One sort of financial innovation is about adding liquidity. There is some class of thing that does not trade very much for some reason, and you find a way to make it trade a lot. Perhaps the thing is very big and not many people can afford to buy it, so you split it into small pieces so people can trade the pieces. This basically describes the stock market: If you like Tesla Inc. as a company, you probably can’t go buy all of it, for a bunch of reasons of which the most important is that it costs $725 billion. But Tesla is split up into billions of shares, and you can go buy a share of Tesla for about $230. 

Or perhaps the things are very different and non-fungible, making them hard to trade, so you smush lots of them into a big standardized package that is easier to trade. This is roughly the idea behind mortgage bonds, or bond exchange-traded funds, or we talked the other day about a guy who wants to do it for diamonds. There is no visible trading market price for a 1.53-carat VVS1 diamond, because there aren’t that many diamonds with exactly those characteristics, but if you can build some sort of standardized diamond basket then maybe you can create a market price for that diamond, and thus a market.

Adding liquidity is, conventionally, desirable. It reduces risk: If you can sell a thing easily, that makes it less risky to buy it, so you are more likely to commit capital to the thing. It increases demand: If only a few rich people can buy a thing with great difficulty, it will probably have a lower price than if everyone can buy a share of it easily. It improves transparency and makes prices more efficient. Also, financial innovation tends to be done by banks and other financial intermediaries, and their goal is pretty much to do more intermediation. More liquidity means more trading, which means more profits for banks.

Another, funnier sort of financial innovation is about subtracting liquidity. If you can buy and sell something whenever you want at a clearly observable market price, that is efficient, sure, but it can also be annoying. Consider the following financial product:

1. You give me the password to your brokerage account.
2. I change it.
3. You can’t look at your brokerage account for one year, because you don’t have the password.
4. At the end of the year, I give you back your password and you pay me $5.

Is this a good product? For me, sure, I got $5 for like one minute of work. 1  For you, I would argue, it’s also pretty good. For one thing, you avoid the stress of looking at your brokerage account all the time and worrying when it goes down. For another thing, you avoid the popular temptation of bad market timing: You can’t panic and sell stocks after they fall, or get greedy and buy more after they rise, because I have your password. “It is well known that one of the best services a retail broker can provide is not answering the phones during a crash,” I once wrote; in this product I am charging you for that service. Your mileage will vary — perhaps you are at market timing — but this service might well be worth more than $5 to you.

There are variations. For instance, if you have borrowed money to buy a stock, and the stock’s market price drops a lot, you will get a margin call from your broker asking you to put up more money. If you don’t have the money handy, the broker will sell the stock, and you will have lost money. If the market price then recovers, you will regret it. If you could buy a service that was, like, “if the price of my stock falls by more than 10%, we shut down the stock market and wait until the stock goes up again,” then that would probably be useful to you. 2  In general that is not a service that you can buy, 3  but it is kind of a service that people can buy. Xiang Guangda is a big metals tycoon, and he shorted a lot of nickel on the London Metals Exchange, and the price of nickel went up and he had huge losses, and the London Metals Exchange shut down nickel trading for a week or so so that the price could go down and he could avoid most of those losses. That was a very helpful service that the LME provided to him; they helped him out a lot, on his nickel trades, by shutting down nickel trading.

Or we have talked about a fun post from Cliff Asness titled “The Illiquidity Discount,” in which he argues that private equity is essentially in the business of selling illiquidity. If you are a big institution and you buy stocks in public companies, the stocks might go down, and you will be sad for various reasons. You might be tempted to sell at the wrong time. You will have to report your results to your stakeholders, and if the stocks went down those results will be bad and you will get yelled at or fired. Whereas if you put your money in a private equity fund, it will buy whole public companies and take them private, and then you won’t know what the stock price is and won’t be able to sell. The private equity fund will send you periodic reports about the values of your investments, but those values won’t necessarily move that much with public-market stock prices: The fund will base its valuations on its estimates of long-term cash flows, and those will not change from day to day. By illiquid, the private equity fund can less volatile. Getting similar returns with less volatility is good; getting similar returns and feeling like you have less volatility also might be good. 4 Asness writes:

If people get that PE is truly volatile but you just don’t see it, what’s all the excitement about? Well, big time multi-year illiquidity and its oft-accompanying pricing opacity may actually be a feature not a bug! Liquid, accurately priced investments let you know precisely how volatile they are and they smack you in the face with it. What if many investors actually realize that this accurate and timely information will make them worse investors as they’ll use that liquidity to panic and redeem at the worst times? What if illiquid, very infrequently and inaccurately priced investments made them better investors as essentially it allows them to ignore such investments given low measured volatility and very modest paper drawdowns? “Ignore” in this case equals “stick with through harrowing times when you might sell if you had to face up to the full losses.” What if investors are simply smart enough to know that they can take on a lot more risk (true long-term risk) if it’s simply not shoved in their face every day (or multi-year period!)? 

One objection to this sort of financial product — illiquidity provision — is that it does not generate a lot of transactions. If you work at a bank and you think of a product that will cause customers to trade bonds or houses or diamonds more often, then it is pretty easy to figure out how to make money from that product. (Do the trades for the customers, and take a commission.) If you work at a bank and you think of a product that will cause them to trade often, it is harder. Basically you have to charge them a bigger fee for doing less work. “We’ll give you a bond fund that offers instantaneous liquidity and transparent market prices and charge you five basis points a year, or we can give you a different fund that offers no liquidity and only updates prices when we feel like it, and that’s gonna be 200 basis points.” 

Anyway here is a fun Wall Street Journal article about private REITs

It has been a terrible year for many publicly traded real-estate investments as rising interest rates and falling property prices hit the market. The MSCI US REIT Index, which tracks publicly traded REITs, is down about 26% this year.

But it has been a strong year for a type of investment especially popular with individuals: nontraded real-estate investment trusts. Some of these funds have returned about 10%. The difference worries some investors—and it could cause losses for those who buy these now thinking that nontraded REITs are immune to the market selloff.

“With nontraded REITs increasing their valuations while markets are punishing public REITs, I’d run for the hills,” said Allan Roth, founder of Wealth Logic LLC, a financial planning firm based in Colorado Springs, Colo. …

The valuations differ because public REITs are valued at whatever their shares are trading for on the stock market. Nontraded REITs are valued monthly by their sponsors working with independent appraisers analyzing how much the commercial property they own is worth. …

Sponsors of nontraded REITs argue that their net asset values continue to rise because they are still seeing strong cash flows. Some have taken steps to protect against a rise in interest rates; Blackstone says its rate hedges have added $4.4 billion this year to the value of its nontraded REIT. And the Blackstone and Starwood funds focus on some of the strongest sectors such as apartments, where the housing market is strong, and industrial properties that profit from the growth of e-commerce. ...

Sponsors note that the stock market is almost always more volatile than the value of the properties they own and that publicly traded REITs began the year at excessive prices, a reason they have fallen so much. Blackstone executives have been buying shares of their fund recently. 

If you can buy a real estate investment that goes up when public real estate investments goes down, that’s good! That’s valuable diversification! You are paying for illiquidity, and getting what you paid for.

At a high level, to be guilty of fraud, you need to (1) say something false (2) know that it is false, or at least, have some sort of bad state of mind about what you are saying. (Not legal advice and not quite right, but a good rough intuition.) When people get prosecuted for fraud, often they will fight over both elements. “Everything I said was more or less true, but if it wasn’t then that was an honest mistake,” they will say. But sometimes it is fairly obvious that what they said was , and so all of the arguments are over their state of mind: When they said these untrue things, were they trying to trick people by knowingly lying, or was it forgivable carelessness?

It is rare, though, to have the opposite. It is rare for someone to be accused of fraud and say “well yes obviously I intended to lie, but what I said was , so it’s not fraud.” It is not theoretically impossible, but it is weird. Usually when people try to lie they succeed, at least in the narrow sense of “they try to say something false and in fact say something false.” On the rare occasions when they fail — when they try to lie and say something true instead — generally no one is tricked, and so there is no prosecution for fraud.

The state of fraud prosecutions for Libor, the London interbank offered rate, is incredibly weird. Libor was a benchmark of banks’ borrowing costs that was used to set rates on trillions of dollars of loans and derivatives. The way it was set was that the administrator of Libor would call up a bunch of banks every day and say “how much would you pay right now to borrow dollars for one month? Two months? Three months? What about yen?” and so forth, through a range of currencies and tenors. And the banks would answer, and for each currency and tenor the administrator would throw out the extreme answers and average the rest, and that trimmed average would be two-month dollar Libor or whatever. And occasionally a bank would get that call 30 seconds after it had negotiated to borrow dollars for one month, and its Libor submitter — the person who answered the phone when the Libor administrator called — would confidently say “our one-month dollar borrowing cost is 2.7539%,” and that was that. But most of the time the bank would not have taken out a two-month yen loan in the last hour, or the last day, or perhaps even the last week, and it would have no directly observable answer to the question. So it would estimate the answer. “We borrowed yen for one month yesterday at 0.25%, and since then rates seem to have moved by about 0.02%, and the one-to-two-month curve is about 0.07%, so I’ll say 0.34%.” And maybe that was actually the rate that the bank would have paid to borrow yen for two months at the moment it answered the question, and maybe it wasn’t.

These sorts of interpolations are not exactly rocket science, and the numbers the banks said were probably pretty close to what their actual borrowing costs would be. But you could not say with much confidence that they were exactly right.

And then people at a bunch of banks got really into doing fraud on Libor. The idea is that sometimes a derivatives trader had a billion dollars of derivatives that were indexed to three-month dollar Libor, and if Libor was 0.01% higher he would make an extra $100,000, and so he would call up his bank’s Libor submitter — and sometimes his buddies at other banks — to try to nudge Libor higher. And the submitter would be all ready to submit a Libor rate of 2.75%, and the trader would instant message him and say “help me out I need it higher,” and the submitter would submit 2.76% instead.

And all of this would happen on permanently preserved electronic chats, and all of the chats were as . Basically the chats were like the traders saying “I will pay you a bribe if you criminally manipulate Libor by lying in your submission,” and the submitters saying “okay I will do that because I am a dishonest liar, but I hope we do not get caught doing these blatant crimes,” except with a lot more misspellings. And so the banks got in trouble with regulators and quickly settled and paid billions of dollars of fines, and the individual traders who sent those chats were prosecuted for fraud.

Most famously, Tom Hayes, a former UBS Group AG derivatives trader, was prosecuted in the UK for manipulating Libor; he sat for “82 hours of interviews with investigators” in which he said things like “I probably deserve to be sitting here because, you know, I made concerted efforts to influence Libor,” and “ultimately I was someone who was a serial offender.” Unsurprisingly, he was convicted.

He was also charged with crimes in the US, but now those charges have been thrown out

A New York judge has thrown out criminal charges against Tom Hayes, the former UBS and Citigroup trader who served more than five years in prison in the UK for conspiring to manipulate the Libor benchmark interest rate. ...

The decision to drop US charges against Hayes follows an appeals court ruling in a separate US case, which overturned the Libor-rigging convictions of two former Deutsche Bank traders. That ruling found the government had “failed to show that any of the trader-influenced submissions were false, fraudulent or misleading”.

talked about that earlier ruling in January. The problem with the Libor prosecutions is that nobody ever proved that the Libor submissions were . Obviously the traders and submitters were to lie: The chats are terrible and show their state of mind pretty clearly, and anyway Hayes spent hours confessing to manipulating Libor. But there is no proof that they succeeded in lying: If a Libor submitter thought that his bank could borrow at 2.75%, and instead told the Libor administrator that it could borrow at 2.76%, either number might have been right! The appeals court pointed out that banks regularly borrowed different amounts of money from different counterparties at different rates at around the same time; it simply is not true that there was One Correct Rate that a bank would have paid to borrow money on any particular day. There is probably some range, and if you pick a number in the range you are telling the truth, even if you are consciously to pick the wrong number.

More to the point, it is very hard to that you are lying. I mean, if you borrowed a lot of dollars for three months at 2.75% right before the Libor administrators called, and then borrowed more at 2.75% right after, and you said “2.76%” when they called and asked for your three-month dollar borrowing rate, then that seems bad. But that is probably pretty rare, and ordinarily there will be some room for judgment. And the burden of proof is on the prosecution: They need to prove that your Libor submission was false, rather than you proving that it was true. That will generally be hard, and in any case US prosecutors never even tried. They looked at all those chats, all those confessions of guilt, and figured that they were good enough to prove that the traders were lying. But they never quite proved that they weren’t also telling the truth.

I have basically enjoyed writing about the saga of Elon Musk’s acquisition of Twitter Inc., because there was, you know, a plot. Musk said he wanted to buy Twitter, and there were interesting questions like “will he really” and “where will he get the money” and “will they stop him.” Then he signed a deal to buy Twitter and almost immediately pivoted to trying to get out of the deal, and there were interesting questions like “can he do that” and “in what circumstances will a Delaware court order specific performance if a buyer undermines his own financing.” Then he pivoted back to saying he would close, and there were interesting questions like “really?” 

But then the deal closed last week, and now it is just a stream of exhausting miscellanea. For instance, Musk wants to start charging people to have a little blue check mark next to their names on Twitter. I wrote yesterday about reports that the price will be $19.99 per month, but that seems not to be a final decision, and other numbers have been suggested. Also last night Musk was personally negotiating the price with Stephen King. “$20 a month to keep my blue check?” tweeted King. “[No], they should pay me. If that gets instituted, I’m gone like Enron.” Musk replied: “We need to pay the bills somehow! Twitter cannot rely entirely on advertisers. How about $8?” I absolutely love that, in between his busy schedule of reading printouts of 50 pages of code per Twitter employee to decide who to fire, Musk is personally going to negotiate commercial terms with each of Twitter’s hundreds of thousands of verified users. I have a blue check, I’m gonna tweet “I’ll pay $7.69” and see what he says.

Meanwhile Musk borrowed money to buy Twitter from his banks. Ordinarily the banks would have sold this debt to investors before closing, but given the process here — where Musk was trying to get out of the deal until the last minute — there was no time to do that, and they’re not about to do it now. For one thing the debt market has gotten worse and they would take a huge loss on the deal. 5  For another thing, traditionally if you are going to sell the debt you send investors an offering memorandum describing the company, its business plan, its financials, etc., and how do you do that here? The Financial Times reports

Banks that lent $12.7bn to Elon Musk for his $44bn Twitter takeover are preparing to hold the debt until early next year as they wait for the billionaire to unveil a clearer business plan they can market to investors, according to three people with knowledge of the plans.

Barring an unexpected rally in credit markets this year, the group of lenders, led by Morgan Stanley, Bank of America and Barclays, have conceded they will be stuck holding the debt on their books for months or even longer and will probably end up incurring huge losses on the financing package.

The banks have in recent weeks held short discussions with several large credit investors as they attempt to gauge the demand for the debt and the discounts they will ultimately need to offer to offload it. The conversations have been informal and some investors said they were given the impression the deal would not come to market quickly.

The seven lenders are wagering it will be easier to appeal to creditors after Musk presents a clear strategy for Twitter, including the size of cost cuts and estimates for the company’s financial performance in 2023 and 2024.

Some analyst has built the first version of the model, and there’s a line for blue-check revenue, and it was built around a $19.99 monthly price, and now she has to update it for that tweet to Stephen King.

Meme liquidity

There was a weird, like, weekend last month when people on the internet went around saying that Credit Suisse Group AG was going bankrupt. Credit Suisse had kind of a tough year, but there was no real indication that it was in any sort of acute financial distress; this seems to have just been a matter of unfounded rumors spread by retail investors on Twitter. A “reverse meme stock,” I called it, and wrote

It is theoretically possible to meme a bank into bankruptcy, in a way that it is not possible to meme, say, Tesla Inc. into bankruptcy. You get enough people worried, they stop funding the bank, and it blows up, even if its business was otherwise sound. That obvious Bagehot line is “Every banker knows that if he has to prove that he is worthy of credit, however good may be his arguments, in fact his credit is gone.” 

Of course in modern international banking the people you need to get worried are, like, wholesale funding markets and credit-default swap traders and derivatives counterparties, not retail depositors or shareholders. 

Risk Quantum reports

Credit Suisse’s average liquidity coverage ratio (LCR) dropped by a fifth in October, after a run on deposits during the first two weeks of the month.

Fuelled by what the bank called “negative press and social media coverage based on incorrect rumours”, the daily LCR averaged 154% between October 1 and October 25, compared with 192% through the three months to September 30. The bank said this was the result of a significant withdrawal of cash deposits and the non-renewal of maturing time deposits.

regulatory minimum is 100%, meaning that a bank needs to have liquid assets equal to at least 100% of its expected cash drain in “a 30 calendar day liquidity stress scenario,” so Credit Suisse didn’t get all that close to the brink. Also you do not have to accept Credit Suisse’s claim that the drop was due to “incorrect rumors” rather than clients’ reasonable response to Credit Suisse’s real financial troubles. Still there is at least an arguable transmission mechanism between people posting mean stuff on Twitter and an actual run on the bank.

Laws of Tax

I often quote Michael Graetz’s two laws of tax: It is always better (for tax purposes) to make more money than less money, and it is always better (for tax purposes) to die later than sooner. The Covid-19 pandemic was good for the first goal but bad for the second

US estate tax collections in 2021 hit their highest levels since former President Donald Trump’s tax cuts took hold, a stark trend reversal that experts attribute to one-off pandemic factors.

Nearly 2,600 estates paid $18.4 billion in taxes last year, almost double the $9.3 billion collected in 2020, according to data from the Internal Revenue Service. The spike is likely a temporary blip as higher-than-normal death levels and stock and real estate gains meant that more people were rich enough to be subject to the tax.

During the pandemic, more wealthy Americans “may have been caught with an unplanned estate,” said Jay Soled, a Rutgers University professor and estate-planning attorney. 

I suppose that, if you become rich enough to be subject to the estate tax, that is good news. But actually becoming subject to the estate tax is always bad news.

Things happen

Credit Suisse Is Not For Sale, Chairman Says. Jack Dorsey rolls his stake into Elon Musk-owned Twitter. After Record Year, University-Endowment Returns Drop Into Negative Territory. Penguin Random House From Acquiring Rival Publisher Simon & Schuster. Carlyle seeks $700mn over insurers’ failure to pay for Russian jet seizuresCFA Final Exam Pass Rate Slips to 48% While Remaining Above Pandemic Lows.  Ether miners repurpose tools following the ‘Merge.’ Hong Kong Considers Opening Crypto Retail Trading. Leon Black Renews Legal Fight With Russian Model Over Affair. Dogecoin Jump Fueled by Musk Enlivens Crypto’s Beaten-Down Bulls. After DAO Frenzy Last Year, Another Constitution Heads to Auction. Excuses, Excuses: Web Searches For Reasons to Skip Work Soar in 2022.

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 . Thanks!

1. Oh, come on, I do not steal your money, I am very honest in this hypothetical product. I just get the $5.

2. Sometimes the stock would keep falling and you’d regret using this service.

3. Though many markets have circuit breakers, trading halts for volatility, etc., which are more limited in effect and time but not totally unrelated.

4. This is also a service provided by, say, big hedge funds that have large private investments. If you are half in public tech stocks and half in private tech stocks, and public tech stocks fall by 30% in a quarter, then in some rough economic sense the value of your fund has probably gone down by about 30%. But in some rough accounting sense maybe you only mark it down by 15%. And then if values come back next quarter, hey that’s great.

5. “Isn’t that true whether or not they sell the debt,” you might ask, but see the first section about illiquidity provision.

To contact the author of this story:

Matt Levine[email protected]

To contact the editor responsible for this story:

Brooke Sample[email protected]