| This is not investment advice, but these days there are a number of platforms where you can bet on the US presidential election. Offering election contracts used to be a violation of US law, but a recent court case seems to have changed that, though this is also not legal advice. On the other hand, these platforms do seem to violate the law of one price. As of about noon today, you could buy a Donald Trump contract — one that pays off $1 if he wins the election — for about $0.49 on Kalshi, and you could sell it for about $0.53 on Polymarket, collecting a free four cents. I am sure there are some fees that I am not accounting for, and I don't know how good liquidity is on any of these places; I am not telling you about this because arbitraging prediction markets on the most high-profile event in the world, after reading about it in a newsletter, is likely to be a good trade. I am telling you about it because (1) people keep emailing me and (2) there are Elon Musk angles. Forbes reports: Polymarket priced in a 53.3% chance of a Trump win and 46.1% of a Harris victory shortly after 6 p.m. EDT [Monday], up from 50.9% to 48.2% split in Trump's favor as recently as 9:30 a.m., extending Trump's Polymarket odds to their highest level since early August—shortly after Harris took over the Democratic nomination from President Joe Biden. … But Polymarket competitors moved far less Monday, as PredictIt favored Harris over Trump at what equates to a 51.4% to 48.6% margin, roughly flat over the last 24 hours, and Kalshi odds sat flat at 51% to 49% in Harris' favor. … Musk, who endorsed Trump in July and appeared at a rally with the Republican candidate Saturday, shared several posts on his X social media platform touting the Polymarket shift. Musk shared a post connecting the jump to his appearance at Trump's Pennsylvania rally, wrote betting odds are "more accurate than polls as actual money is on the line" and reposted a screenshot emphasizing the "collapse" of Harris' election odds.
Here is a post on X about a whale on Polymarket who has been betting a lot of money on Trump, and who might be driving the price moves. The post mentions some speculation that the whale — who goes by "Fredi9999" — is actually Musk, though I would not take that too seriously. Musk is into crypto and Trump, but any time there's a crypto whale doing something Musk-adjacent people speculate that it's his account. My assumption is that he has better things to do with his time, though that assumption is weakly held and he does a lot of dumb stuff with his time. The post also points out that this does not look like market manipulation: Fredi9999 is not buying sloppily with the effect of pushing up the price, but rather buying carefully, in a manner that seems designed to get him a lot of Trump contracts for his money. In other words, this looks like someone with a lot of money who thinks Trump will win (or, at least, that he's underpriced), not someone trying to push up the price of Trump contracts. Also, as Nate Silver points out, it's not clear why you'd want to push up the price of Trump contracts — for instance, it's not at all clear that doing so would help him win the election. Still one can speculate about some different facts: - Instead of Polymarket, what if this was happening on a registered US commodities exchange? The US Commodity Futures Trading Commission previously fined Polymarket (and shut it to US investors) because it was illegally offering event contracts, but since then a US court has ruled that the CFTC can't stop markets from offering election contracts, and mainstream traditional financial firms are getting into the business.
- Instead of buying efficiently, what if the whale looked more like he was trying to push up the prices? We have talked before about potential election-contract manipulation, and it seems like these markets are not so deep and liquid that a big whale couldn't push up Trump's market-implied probabilities materially.
- Instead of being a weird curiosity, what if election prediction markets were widely followed and influenced public perceptions of elections? What if the whale pushed Trump into the lead in prediction markets, and Trump's lead caused donors to give him more money, and the media to cover him more favorably, and Kamala Harris voters to become demoralized and not go to the polls, and thus caused him to win the election?
None of this is what happened, but none of it feels all that far-fetched, either. And … then what? What if the CFTC was pretty sure about all of this? Would it bring a commodities fraud case? Against the big Trump buyer? For trying to improve his implied election odds? What if the Trump buyer said "well, you can't prove that I did any commodities manipulation, because it's not like I sold the contracts at a profit, or even really had a plan to sell them at a profit; I just thought it was a good bet. Or maybe I was willing to spend money to help Trump win, but you can't prove which it is, and neither case involves a financial payoff for me." I don't know; in some ways that feels like a standard "open market manipulation" case (like the Bill Hwang case), where the hard thing is proving intent. In other ways it feels like a horrible mess? The CFTC getting into elections? When we talked about regulated election contracts in May, I quoted CFTC Chair Rostin Behnam saying that "such contracts would put the CFTC in the role of an election cop." We're getting closer to that happening. In the olden days, when a company sold itself, there were two sorts of acquirers, financial and strategic. A strategic acquirer was another company: If you were selling a widget company, and another widget company wanted to buy it, that was a strategic deal. A financial buyer was a private equity firm: If you were selling a widget company, and a group of ex-bankers financed by investors and junk bonds wanted to buy it, that was a financial deal. In a strategic deal, the buyer would integrate the seller's operations with its own, and get synergies by combining departments and cross-selling products. In a financial deal, the buyer would hope to make money from cost cuts and leverage and giving management a lot of equity to align incentives. One typical problem with a strategic deal was antitrust: Antitrust regulators are in the business of blocking anti-competitive mergers, and when two widget companies merge there is always a risk of a monopoly. So when you did a strategic merger, you had to file forms with the antitrust regulators — the main form is usually called "HSR," a premerger notification under the Hart-Scott-Rodino Antitrust Improvements Act of 1976 — and wait for regulatory approval, and sometimes the regulators would push back and ask for more information and maybe even try to block the deal. Financial deals had less of that problem. Private equity firms were in the business of leveraged buyouts, not the business of making widgets. If they bought a widget company it was because they did not already own a widget company, so there was no real antitrust issue. Because of the way the rules work, the private equity firms still had to fill out the forms, but they easily sailed through because there were no actual concerns. In 2024, the biggest private equity firms are gigantic world-spanning conglomerates with many portfolio companies in each sector, often doing rollups of industries that raise serious competition and consumer-welfare concerns. Two things that this means are: - It is more complicated for private equity buyers to fill out the forms. It's not just "we're six ex-bankers in an office who borrowed some money to buy a widget company"; now they have to go out and survey their entire gigantic conglomerate to see what other interests they might have in the widget industry.
- The forms matter more: Unlike in the olden days, the regulators are more likely to want to block a private equity acquisition on antitrust grounds.
I grew up in the olden days, though, and I still have an instinctive bias like "ehh a financial deal won't raise as many antitrust issues as a strategic one." It is just possible that some private equity firms, which also grew up in the olden days, have the same bias? Here's a fun story about HSR filings from Bloomberg's Leah Nylen, Ryan Gould and Allison McNeely: KKR & Co. and the Justice Department are sparring over a push to hold the firm's top executives more accountable for disclosures related to takeovers, as US authorities deepen their scrutiny of the private equity industry. The high-stakes fight stems from a federal investigation into whether KKR withheld information in its filings to government agencies about the competitive impact of mergers and acquisitions. To settle the probe, antitrust enforcers want the famed buyout firm to agree that their co-chief executive officers could be held personally liable for future lapses, according to people familiar with the matter. … The Justice Department has two parallel investigations into KKR's disclosures, one civil and one criminal, which the firm has previously disclosed. The company has said it received a grand jury subpoena over the accuracy of its filings. The criminal inquiry focuses on whether individuals who signed the forms were aware of any omissions. … The Justice Department rejected an earlier KKR offer to pay around $100 million, the people said. Perhaps more contentious than any settlement price, however, is the requirement floated by antitrust enforcers that Co-Chief Executive Officers Scott Nuttall and Joe Bae be held personally liable if more lapses occur within the next five years.
I have to say, I sympathize with both sides here. The basic ask appears to be "if KKR's deal teams and lawyers fill out some forms wrong, then we can charge its CEOs with a crime." Like: - If I were KKR's co-CEOs, I would never agree to that! It's not like they are filling out the forms; why should they be liable if some of their underlings fill out the forms wrong?
- I get why the Justice Department wants it? The point of the settlement is to make KKR fill out the forms right, and threatening the CEOs with personal liability is a way to get KKR to take the forms seriously.
A story that I like to tell about private credit goes like this. Once upon a time, banks made loans and held them on their balance sheets: Companies would come to a bank and ask to borrow money, and the bank, which had money, would lend it to them. For a variety of reasons, over the years, banks got more and more into "the moving business, not the storage business": They would increasingly syndicate their loans or package them into securities, selling the loan to some other ultimate investor. This has some advantages: It moves risk off the banks' balance sheet, making the banking system safer; it also might get companies cheaper loans because there is a bigger pool of ultimate lenders. It also has some disadvantages: The companies get less speed and certainty, because instead of just giving them the money the banks have to go out and find investors, plus if anything goes wrong for a company it's dealing with a faceless blob of adversarial creditors rather than a single bank it knows well. And so private credit came along, in which big asset managers, funded with long-term money, would make loans directly to companies (or to private equity buyers) using their investors' money. Private credit could offer speed and certainty and a real relationship, though sometimes at the cost of higher prices. And this has been a lucrative and fast-growing business for the asset managers, so much so that their big problem is how to expand it. This means simultaneously: - Finding new investors for private credit assets: Private credit managers would traditionally raise funds from insurance companies, pensions, rich people, etc., the usual investors in alternative assets, but they are increasingly looking to raise money from public markets to fund their growth. But it can also make sense for them to get into the syndication business, just like banks did: You can arrange more and bigger loans, with less risk, if you are selling the loans to someone else rather than holding them all yourself.
- Finding new assets to put into the funds: There are only so many private equity sponsors doing only so many leveraged buyouts, so if you want to build an enormous private credit business you need to find other borrowers. Lend to investment-grade companies, make mortgage and credit-card and car loans, etc.; the world is full of people who want money. Here much of the game is about hiring people to originate loans: Every big private credit firm employs people who talk to every big private equity sponsor — many of the private credit firms are also private equity firms — and will get called to do buyout loans, but if you want to make asset-backed loans against aircraft, you will want to hire some former bankers who know the chief financial officers of airlines. Finding new borrowers is not so much an intellectual exercise as it is a sales exercise, so you need salespeople.
Last year Apollo acquired a securitized products group, called Atlas SP Partners, from Credit Suisse Group AG, which addresses both of those issues: Atlas is in the business of originating types of loans ("asset-backed warehouse financing, forward flow and asset purchases," among others, in "a wide range of asset classes within residential and commercial real estate, and corporate and consumer debt") that Apollo previously didn't do much of. Today Bloomberg's Allison McNeely reports on how that's going: Atlas SP Partners — the structured finance business that's key to Apollo Chief Executive Officer Marc Rowan's plan to become a lending machine — has struggled under its new owner since last year's acquisition, according to interviews with almost a dozen people with knowledge of the unit. Cultures have clashed, business has slowed and a raft of senior departures culminated in the abrupt exit of Atlas SP's longtime head, Jay Kim, in August.
One problem is that it is not trivial to get into the moving business instead of the storage business. Traditionally the private credit business is about making loans and then holding onto them until maturity. But Atlas is in the business of securitizing stuff into bonds to sell to investors, and if you do that, it is helpful to have a trading desk that will make a market for the bonds. (Investors are more likely to buy the bonds if they know that they can sell them.) Credit Suisse had a trading desk; Apollo, less so: Atlas no longer had a major trading desk making markets in the securities it structured, and it didn't benefit from banks' low funding costs to compete in the most vanilla assets. … Atlas was operating without a broker-dealer license for well over a year. Without it, Atlas has struggled to generate business because — unlike competitors such as big banks — it couldn't easily syndicate deals for clients. Its residential mortgage business was hit hardest by difficulty originating new deals, largely because of the firm's lack of trading capabilities, some of the people said.
But "the firm finally secured a broker-dealer license and has begun plans to build a trading operation, with the hope that more deals will come its way if it can show there's a market for investors to sell the assets if needed." Another problem is … well, I mean, private credit has gone from roughly nothing to a huge, fast-growing, incredibly lucrative area of finance, taking away market share and revenue and talent and glamour from the big banks. How did private credit firms do that? You can point to various structural factors, but a dumb simple answer would be "they are better at their jobs than the banks are." Why are they better at their jobs? Again, there are various possible answers, but a dumb simple one would be "they work harder." This can make it hard to integrate people from a big bank: The melding of teams caused friction, particularly in the early days of the deal. When Apollo executives tried to dig into the granular details of the portfolio they'd bought as the deal was closing, they struggled to get timely information out of Credit Suisse, some of the people said. Meanwhile, Apollo's hard-driving culture — with its late-night phone calls — spurred disagreements with the ex-Credit Suisse staff over expectations that some people perceived as unreasonable.
Here's a Wall Street Journal article from last year about how Apollo and other private credit firms are "the new kings of Wall Street." Here's a Financial Times profile of Rowan from last week, arguing that he is working "to re-engineer finance according to his own designs." Credit Suisse disappeared a month after Apollo bought Atlas. I also do not like late-night phone calls, and have arranged my life to avoid them. But it's possible that Apollo's culture is not the unreasonable one. I guess if I started a new stock exchange, and on my first day I had zero companies listed on my exchange, I'd go around being like "you know all those companies that are listed on the New York Stock Exchange and Nasdaq? They're just not good enough to list on my exchange." And so: The head of the fledgling Texas Stock Exchange has pledged tougher listing standards than his New York rivals as part of his state's bold attempt to establish Dallas as a financial challenger to east coast dominance. Jim Lee, chief executive of the TXSE, told the Financial Times the new exchange's standards, including earnings tests, minimum prices and other unspecified measures, would be stringent enough to in effect exclude more than a third of the companies listed on Nasdaq and the New York Stock Exchange. The comments push back against early expectations that the nascent bourse would adopt looser rules in its attempt to break New York's pre-eminence. Its emphasis on "predictability" in its initial June launch was seen as a riposte to a controversial 2021 board diversity disclosure rule Nasdaq introduced that is being challenged in court. "Ours are going to be the tightest [quantitative standards] inside of the strike zone," he said in an interview. "Our qualitative standards will be tighter, not on every element, but in total, such that about 1,500 Nasdaq companies [would] fall out and about 200 NYSE companies [would] fall out — that would not qualify for continued listing on our exchange."
We have talked about TXSE before, and there are different ways you could go with listings standards. If you have tougher standards than NYSE and Nasdaq, maybe you develop a reputation as a premium venue and attract listings, though that seems hard to do starting from scratch. If you have easier standards than NYSE and Nasdaq, you can probably attract listings (from companies that can't list on NYSE or Nasdaq), but if you then develop a reputation as a venue for shady companies you'll never attract that many listings. My impression was always that TXSE's standards were going to be roughly "the same as NYSE and Nasdaq, except we don't care about board diversity," which is an obviously good pitch? The goal is not to send a signal of high or low quality, but to send a signal of "anti-woke," so that companies that want to send their own anti-woke signal will know where to go. It would be a real error for TXSE's quantitative standards to be so tough that Elon Musk can't list his companies there. I'm sorry I just have to put this here: Daniela Amodei, co-founder and president of model developer Anthropic, said the startup's developers have been using its Claude chatbot to help them code. It can't fully replace engineers and needs "some coaching," but many Anthropic developers have dramatically increased their productivity with Claude, Amodei said at The Information's Women in Tech, Media and Finance conference on Tuesday. It's at the "point that we've even sort of said, as we're doing head count next year, how should we think about that?" The potential economic returns "could just be incredibly high," she said.
I feel like the most obvious and hackneyed possible dystopian sci-fi plot at this point is "an artificial intelligence company realized that it could ask its current AI model to code its next-generation AI model, so it fired all its engineers and just let the AI program itself, yadda yadda yadda Skynet." Also I love that the motivation here appears to be cost savings: Human engineers are expensive, but you can improve your margins by getting rid of them and letting the AI program itself. Meanwhile my dystopian sci-fi novel is going to be about the disappearance of accountants: The head of KPMG US has said the industry urgently needs to make it easier and cheaper to become an accountant, to head off a "brewing crisis" in the profession. Paul Knopp became the first head of a Big Four firm to publicly back scrapping the requirement for a fifth year of higher education on top of the typical four-year undergraduate degree, as the numbers of US students taking accounting courses have fallen sharply. Knopp backed an "apprenticeship" model that could replace the fifth year of education. "We have a brewing crisis right now, with the number of students going to college and the number going into accounting, and we need to absolutely address it in the very near term," he said in an interview with the Financial Times. "I can't over-emphasise, it's not just the Big Four. We need more accountants in corporations and outside of the Big Four. The industry that we are in is systemically important to the functioning of the capital markets."
He's right, right? Financial statements are the raw material of finance. There are lots of people who want to be somewhere in the lucrative superstructure built on top of accounting, who want to be hedge fund analysts or private equity dealmakers or ESG Consultants But Evil. But nobody wants to be an accountant. You could imagine a story in which accountants get both scarcer and less qualified, so public companies' financial statements become less reliable, so markets become less efficient at allocating capital, so there are higher returns to hedge-fund analysts who can sniff out accounting problems, except those analysts are rarer too because the way they used to be trained was mostly by qualifying as accountants. I have written and thought a lot about meme stocks over the last few years, and it has made me a bit nihilistic. Sometimes I suggest that perhaps fundamental analysis — the idea that the prices of financial assets reflect the present value of their future cash flows — is a temporary phenomenon; people were trading stocks based on vibes and gamesmanship long before anyone built discounted cash flow models, and perhaps they'll keep doing so long after. I mostly think of that as a social phenomenon: The available information about companies keeps getting better, but there is no law of nature that forces people to pay attention to that information. You can just buy GameStop because your friends are buying GameStop. But what if the information also gets worse? The obvious solution is to have the AI models do the accounting, but that's what makes my novel fun. Wall Street Salaries Decline Further From Pandemic Heights. US Weighs Google Breakup in Historic Big Tech Antitrust Case. Treasury market volatility surges as investors rethink interest rate bets. AI's Thirst for Power Turns Utility Stocks Into Big Tech Proxies. Samsung Apologizes for Falling Behind in AI Chips Race. X Can Resume Operations in Brazil After Paying Millions in Fines. KKR-Owned BMC to Separate Into Two Private Software Companies. Billionaire Backs Space-Laser Solar Startup After Robinhood Exit. Bitcoin creator is Peter Todd, HBO film says. Cameroon takes unusual step of insisting its president has not died. 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! | 
No comments:
Post a Comment