Money Stuff: A Stock Price Is Not a Prediction

Stock prices sometimes tell you what the market thinks will happen. Consider a company that has signed a merger agreement to sell itself for $30 per share in cash. If the stock trades at $29, that suggests that the market thinks the deal will close. If the stock trades at $18, that suggests that the market thinks the deal won’t close. If the stock trades at $31, that suggests that the market thinks someone else will come in and offer more money to buy the company.

Suggests. None of these prices gives you an exact market-implied probability for any particular event. If the stock trades at $28, you might think “well $28 is about 93% of $30, so the market thinks the deal has a 93% chance of closing,” but that is not a very good interpretation. Instead, that price means something more like:

• There is some probability that the deal will close and pay $30 soon.
• There is some probability that the deal will be delayed; it will eventually close and pay $30, but the present value of that money is considerably less than $30 today.
• There is some probability that someone will come in and offer more money, and the company will end up selling for $32.
• There is some probability that the deal will fall apart and the stock will trade down to $23.
• Or to $18.
• Or it will fall apart because the company is a fraud and it will trade down to $0.
• Or it will fall apart because the company discovers a diamond mine under its headquarters, and the stock will trade up to $35.
• Or the US dollar will suffer hyperinflation and the stock will trade to $30,000.
• Etc.

There is a variety of possibilities, and each possibility has some range of possible effects on the stock’s value. (“The deal closes tomorrow” has a narrow range of effects — it’s worth $30 — while “the deal falls apart and the company stays independent” has a wide range.) 

In the general case, if you are a shareholder of Warner Bros. Discovery Inc., the thing that you want is for Netflix Inc. or Paramount Skydance Corp. or Comcast Corp. or, for that matter, Elon Musk or anyone else to pay a lot of money to buy Warner. [1]  If Netflix pays $40, that’s great for you. If Paramount pays $40, that’s equally great for you. If Warner announces “the deal is off because we have discovered a vastly more potent form of Harry Potter” and the stock trades to $40, also great! You might have some preferences about Netflix vs. Paramount, based on your analysis of their closing certainty or financing certainty or the value of the consideration they offer. But mostly you want as much money as possible. 

Warner’s stock closed at $29.53 per share yesterday. Warner has a signed deal with Netflix, in which Netflix has agreed to pay about $27.75 per share (in cash and stock) for most of Warner and leave Warner’s shareholders with a stub company worth perhaps $1 or $2 per share, or perhaps more. Paramount has launched a competing tender offer to pay $30 in cash per share. Neither deal will close all that soon, so just as a matter of the time value of money, it seems unlikely that either deal is worth $29.53 per share today. If you’re paying $29.53 today, you’re betting on more. The Wall Street Journal notes:

That price level suggests that investors expect a higher bid from one or both of the bidders, according to hedge-fund managers who bet on the probability of a takeover’s completion. If there was more skepticism of a deal being completed, Warner’s stock would trade at a chunky discount to the offers given the long time expected to close either of the offers and the antitrust risk.

But if you’re paying $29.53 today, that is not a bet on who will win. If there’s a bidding war, and Netflix wins with $35 per share, or Paramount wins with $35 per share, you will be equally happy. You might have a thesis; you might have a vision of how this will play out, and that vision might give you the confidence to buy now. But some other investor might pay the same price to buy the stock with an opposite vision of who will win, because the price does not depend on who will win. [2]

Don’t you want to know who will win, though? I mean, maybe you don’t; maybe you don’t care. But a lot of people seem to want to know who will win; there is a lot of media and public and political attention being paid to this fight. And the stock prices don’t tell you the answer.

The Journal adds:

Polymarket, the prediction platform, shows Paramount and Netflix are in a dead heat to close a transaction by the end of June 2027, with about a 42% chance each. About 16% of bettors are wagering that there will be no deal at all. 

When I looked at about 11 a.m. today, Polymarket had Paramount at about 49% (and Netflix at 36%); Kalshi, the other big prediction market, had Paramount at about 51%, though with a different contract specification. [3] Those are fairly clean probabilities. The “Paramount buys Warner” contract on Polymarket pays $1 if Paramount closes the deal and $0 if it doesn’t; paying $0.49 to buy that contract really does reflect a view that the probability is 49%. [4] All of the possible outcomes in which Paramount closes the deal — it pays $30, it pays $35 — pay out the same $1; all of the possible outcomes in which it doesn’t — Netflix pays $27.75, Netflix pays $35, Comcast pays $32, Elon Musk pays $42.69, Warner stays independent — pay out the same $0.

The prediction markets thus provide a pure prediction of an outcome that is of interest to a lot of people. “I think Paramount will win so I will buy Warner stock”: not a clean bet. “I think Paramount will win so I will buy Paramount stock,” or “sell Paramount stock,” or “buy Netflix stock”: also not clean bets. “I think Paramount will win so I will bet that Paramount will win”: Yes, perfect.

Well, perfect for journalists. Perfect for onlookers. If you are interested, as a curious bystander or a White Lotus fan or a believer in freedom of the press, in who will win the battle for Paramount, you might want to fast forward through the bidding war and regulatory approvals and just find out the answer. You are not interested in the price; you just want to know the winner. Today’s stock price gives only a limited and murky signal about who will win. Whereas the Polymarket market gives a clean and specific signal about everyone’s chances of winning. (Sadly the signal is roughly “it’s a coin flip,” but that’s how life is sometimes!) 

This is, of course, the point of prediction markets. Or rather there are two related points:

1. If you have insight about some event, or just an itch to gamble on the event, you can bet on that event specifically, disaggregated from its consequences. If you think Netflix will win, you do not need to model up the value of the stock consideration or of the stub Discovery Global equity, or revise your model if the mix of consideration changes. You just bet on “Netflix wins,” and everything else is irrelevant. 
2. If you want insight about some event, you can go to the prediction market website and read off the market’s prediction. You don’t have to back out various complicating factors or conditional values. You can write a newspaper article saying things like “Polymarket, the prediction platform, shows Paramount and Netflix are in a dead heat to close a transaction by the end of June 2027, with about a 42% chance each.” Which tells you something that the stock market does not.

There is one important caveat here, though. Roughly $3 billion worth of Warner stock traded yesterday. As of this morning, Polymarket reports about $27,000 of volume on the “Who will close Warner Bros. acquisition” market. The market for Warner stock is deep and liquid; people have real money at stake. The market for the who-will-win probabilities is … recreational. If you want to bet a few bucks on it, you can. But, like, BlackRock is not on the other side of that bet in size.

Why not? BlackRock is a Warner shareholder, and ultimately an owner of real economic activity; to a first approximation, it cares only about the economic consequences of the auction, and not at all about who wins. Buying or selling who-will-win contracts does not do much to hedge or manage its risk: Either bidder could win in a way that is good or bad for the stock, and BlackRock’s exposure is to the stock. People with real exposure to economic activity are not making clean bets on isolated events, disaggregated from their consequences. The consequences matter.

Time Magazine announces a “Person of the Year” each year. It’s often, though not always, a person. In 1982, it was “The Computer”; in 2006, it was “You.” In recent years it has been Taylor Swift, Donald Trump and Elon Musk, but also “The Guardians” and “The Silence Breakers.” “Person” is a broad category. 

This gets a lot of attention each year, and in 2025 that means you can gamble on it. Polymarket and Kalshi, the two big prediction markets, both listed markets on the Time 2025 Person of the Year, and they both attracted a lot of volume (some $55 million on Polymarket and almost $20 million on Kalshi).

This morning, Time announced that “The Architects of AI Are TIME’s 2025 Person of the Year.” Not a person. Some number of persons. At least eight: There’s a cover illustration featuring Mark Zuckerberg, Lisa Su,  Elon Musk, Sam Altman, Jensen Huang, Demis Hassabis, Dario Amodei and Fei-Fei Li, though I am not sure that the picture is meant to be exhaustive. On Polymarket, “Architects of AI” is now trading at 100 cents on the dollar, with other options — including Altman, Musk and Huang, who are actually on that cover — trading at zero. Kalshi’s resolution is different, and all of people on the cover — Altman, Musk, Huang and the rest — resolved to “Yes.” [5]

Nobody, I think it is safe to say, bet on “Architects of AI” before about, uh, last night? Polymarket shows that option being created last night. On Kalshi, many of the winners (Amodei, Su, Hassabis, even Musk) were trading below 5% until about yesterday evening, when they jumped to about 85%. 

But artificial intelligence has obviously been a dominant theme in news this year, and if “The Computer” could be Person of the Year in 1982, then surely AI could be Person of the Year in 2025. On both Kalshi and Polymarket, “AI” traded above a 20% probability for months. Yesterday, AI spiked as high as 73% on Kalshi and 71.5% on Polymarket. Apparently the news … leaked? Quasi-leaked? Leaked in a garbled fashion? Leaked in a way that people misinterpreted? Leaked in a way that people correctly interpreted, but then they misinterpreted the prediction markets’ rules? I don’t know. In any case, bets on AI spiked yesterday and cratered today: The Person of the Year was about AI, but it was not AI, and if you bet on AI you came close, but you lost. There are complaints online.

Incidentally there was also a Polymarket market on “Will Time 2025 Person of the Year be leaked?”:

This market will resolve to "Yes" if a "Yes" option within the "Time 2025 Person of the Year" market … is priced over 90 cents for a majority of minutes during any one-hour period ending at least 4 hours before the same option is officially revealed to be the Time person of the Year. Otherwise, this market will resolve to “No”.

As of noon today it had not resolved, but was trading at about 100%. We have talked a few times recently about how good prediction markets are at, uh, getting this sort of news (the Nobel Peace Prize, the Spotify Wrapped rankings) a few hours before it is officially announced. “Surely someone will get the Time announcement a few hours before it’s official” was a pretty good bet, though here they garbled it a bit.

Yesterday Bloomberg News published an investigation into Jeffrey Epstein’s investments, and the Wall Street firms that facilitated them even after he was first charged with sex crimes in 2006. The lead anecdote is about a hedge-fund salesperson at Smith Barney who sold Epstein on a hedge fund. The salesperson “said through a spokesperson that his interactions with Epstein were solely professional and ended in 2007, before … he learned of Epstein’s crimes.”

It is not, as far as I know, illegal to sell a hedge fund to a sex criminal. It is not illegal for a sex criminal to invest in a hedge fund, and I have not seen any allegations that Jeffrey Epstein did anything illegal with his hedge fund investments. (There are allegations that Epstein did illegal stuff with some of his banking relationships, like withdrawing money from the banks to pay for sex trafficking.) Nonetheless. I personally would prefer not to be the lead anecdote in a story about Jeffrey Epstein. I would prefer not to say through a spokesperson that my interactions with Epstein were solely professional, even if they were. “I had professional interactions with Jeffrey Epstein” is not, in 2025, a boast. [6]  Better, in hindsight, not to have had any interactions with Jeffrey Epstein. 

The technical term for this is “reputational risk.” When you go to work at a bank, at some point pretty early on you will go to a training in which someone says to you “don’t do anything that you wouldn’t want to appear on the front page of the newspaper.” It turns out that selling financial services to Jeffrey Epstein can get you on the front page of the newspaper, in the bad way. In hindsight, a lot of banks would prefer (now) not to have sold financial services to Jeffrey Epstein.

Reputational risk also tends to shade into legal risk. Sometimes it was, at least in hindsight, illegal to sell financial services to Jeffrey Epstein. Deutsche Bank AG and JPMorgan Chase & Co. have paid hundreds of millions of dollars of fines and settlements for doing it. They would really prefer not to have sold financial services to Jeffrey Epstein.

It is hard to know what will end up on the front page of the newspaper in 20 years, but one tries. “This guy has a lot of money and wants to use our financial services, but he seems super shady”: Sometimes that will get an account approved (money is good), but sometimes it won’t (shadiness is a good proxy for reputational risk). Ordinarily the way these decisions get made is:

• the relationship banker wants to open the account, because she gets credit for any money the shady client brings in, and probably won’t be around in 20 years; and
• the compliance department argues against opening the account, because its job is to manage reputational risk.

And the exact answer will depend on things like how much money the client brings in, how serious the compliance/reputational concerns are, how powerful the relationship banker is, etc. But the point is that ordinarily there are commercial considerations on one side and compliance considerations on the other. In particular, compliance does not generally come to the meeting and say “well if we don’t open this account, we’ll get in trouble.” Not opening the account is sometimes a bad commercial decision, but it is always the safe compliance decision. [7]

Until now:

Nine large US lenders made “inappropriate distinctions” among certain customers, the Office of the Comptroller of the Currency said Wednesday, a finding that falls in line with President Donald Trump’s calls to rein in what he sees as the practice of debanking.

Between 2020 and 2023, the OCC found that the lenders — JPMorgan Chase & Co., Bank of America Corp., Citigroup Inc., Wells Fargo & Co., U.S. Bancorp, Capital One Financial Corp., PNC Financial Services Group Inc., Toronto-Dominion Bank, and Bank of Montreal — maintained policies that restricted access to banking services or required the customers to go through escalated reviews and approvals.

The restricted access to firms involved a range of sectors, including oil and gas exploration, coal mining, firearms, private prisons, payday lending, tobacco and e-cigarette manufacturers, adult entertainment, political action committees and digital assets, the OCC found in its review. The OCC’s findings also confirm that “these or similar policies and practices were in place at each of the banks reviewed.”

“The OCC is committed to ending efforts – whether instigated by regulators or banks – that would weaponize finance,” said Jonathan Gould, the head of the OCC.

Here are the OCC’s announcement and the preliminary findings. Within living memory, a bank might think things like “if we finance a gun company and its guns are used in a school shooting, that will get us negative press, so let’s not do that.” But now the risk is reversed.

Sometimes in this column I have occasion to write hypothetically about individual investors who are fairly wealthy but not very financially sophisticated. When I do this, I often resort to the crude stereotype that these investors are dentists. “Dentists,” I have written, “can invest in any dumb thing someone can dream up and sell them.” I did not invent this stereotype. This is a thing. There is a popular notion, in the US financial industry, that dentists tend to invest a lot, manage their own investments, invest in pretty spicy stuff, and, uh, not always be the very very best at it. I do not know if this stereotype is true, and occasionally dentists have emailed me to say “hey come on” and I have apologized. But then I go and do it again.

Meanwhile in Germany:

Berlin prosecutors are investigating potential wrongdoing after bad investments led to a loss of more than €1 billion ($1.3 billion) at a pension fund for thousands of dentists in the region.

The General Prosecutor’s Office in the German capital is probing the issue for corruption-related crimes, a spokesman for the agency said in an email on Wednesday. He declined to provide more details. The pension fund known as VZB confirmed the losses.

“We currently expect to lose around half of our invested assets — approximately €1.1 billion” as of last year, VZB representative Thomas Schieritz said by email. “The complete revaluation of all investments has not yet been completed, so these are preliminary estimates. The final result is expected to be available in the first quarter.”

The case is the starkest example yet of the slow-burning crisis engulfing many German pension funds after they ventured into riskier corners of the market to prop up returns eroded by years of ultra-low interest rates. VZB’s investments included hotels and vacation resorts, startup companies including a digital insurer that went bust, and a now-insolvent shrimp farm, according to Schieritz.

Refire reported in August that “the fund's investment committee consisted of practicing dentists rather than financial professionals, making complex real estate and private equity investments without appropriate expertise or oversight.” I feel like, if you asked US financial firms to pitch products to an investment committee of dentists, they would rub their hands together and cackle with evil glee. Apparently that is true in Germany as well.

When I was a summer associate at a law firm, one of my summer classmates came in one Monday with his arm in a cast. “What happened,” everyone naturally asked. “I got in a proxy fight,” he told us, which I think is to this day the best corporate law joke I have ever heard. Anyway here’s this:

At least two fistfights broke out last week between PDD Holdings Inc. employees and Chinese regulators who were performing checks at the e-commerce company’s Shanghai premises, according to people familiar with the matter.

The altercations involved PDD staff and officials from the State Administration for Market Regulation, the people said, asking not to be identified discussing a sensitive issue. The SAMR officials were investigating reports of fraudulent deliveries on PDD’s platform, the people said, adding that police made several arrests in the aftermath.

“It’s unheard of for interactions between large Chinese companies and regulators to descend into physical confrontations,” adds Bloomberg News, and I wonder if that’s true in the US. When Neil Barofsky was the special inspector general for the US financial industry bailouts, he famously had his agents carry guns and badges to inspect banks. If you work in US financial services and have ever punched an SEC examiner, do let me know.

For Trump, the Warner Megadeal Talks Are All About CNN. Hedge Funds Are Reviving Once-Dormant Appraisal Arbitrage Play. BlackRock’s Crew of Quant PhDs Are On Track for a Record Year. Disney Invests $1 Billion in OpenAI, Strikes Licensing Deal. Fed to launch $40bn debt-buying scheme after money market strains. Apollo’s Rowan Sees Market Makers Coming for Private Credit. Bezos and Musk Race to Bring Data Centers to Space. Silicon Valley Is Racing to Make Critical Minerals — and Blunt China’s Dominance. Meta’s New A.I. Superstars Are Chafing Against the Rest of the Company. ECB considers ditching AT1s to ‘increase quality’ of banks’ capital. Berkshire’s Next Boss Confronts Making Buffett’s Empire His Own. Distressed Funds Snap Up First Brands’ Top Loans to Drive Debt Restructuring. “Everything these days is so impersonal: most people just gamble by staring at their phones.” 

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!