A simple good trade is: 

1. Buy stock in a hot initial public offering at the IPO price. For instance, buy stock in SpaceX’s IPO at $135 per share.
2. When the stock starts trading publicly and immediately goes up 20% or so, sell it.

This trade is not perfect. For one thing, occasionally IPOs trade down, not up, so there is some risk. For another thing, you might be leaving money on the table if you sell after just one day. SpaceX went up another 20% on its second day of trading. “SpaceX shares have surged nearly 50% in their first days of trading and left a trail of remorseful traders in their wake: those who sold too early,” the Wall Street Journal reported this week. Plus maybe in 10 years it will be worth a quadrillion dollars.

All in all, though, this is an excellent trade; there are not a lot of trades that offer fairly reliable 20% one-day returns. And so the main problem with this trade is that nobody wants to let you do it. For one thing, it is hard to buy shares in a hot IPO. Retail investors often don’t have any opportunity to buy shares at the IPO price; even institutional investors who put in orders will not necessarily get any shares. Or, if you do get shares, you’ll get less than you hoped for. SpaceX’s $75 billion IPO “drew more than $350 billion in demand from institutions and retail investors,” Bloomberg reported: “Among the firms that placed orders, close to one-third of them didn’t receive any stock,” and retail investors asked for $100 billion of stock and got only $15 billion.

For another thing, if you do get shares in a hot IPO, it is probably because you pinky-swore not to sell them. When SpaceX and its bankers allocated shares to institutional investors in its IPO, “they assessed who would be committed long-term owners and weeded out those most likely to dump the stock for a quick profit,” the Financial Times reported, and most of the shares went to long-only investors rather than hedge funds. (Long-only holders could dump the stock for a quick profit, but if you do that too often you will get a bad reputation and won’t get stock in the next hot IPO.) And retail brokerages “typically have policies discouraging users from quickly selling, or ‘flipping,’ IPO shares,” and will also cut you off from the next IPO if you flip this one.

This is an annoying state of affairs: There’s a good, low-risk, high-return, short-term trade that is available on paper but that you can’t easily do. You want to do it. How?

The very best way to do it is probably to be a big hedge fund that pays a lot of fees to the big banks. Companies actually do want some IPO investors to flip their stock — if no one sold it, no one could buy it, and there’d be no IPO pop and no liquidity — but getting into that group is quite selective. If the banks like you — because you do a lot of trading and pay a lot of fees, or because you help them out by buying a lot of shares in IPOs that weren’t particularly hot — they are more likely to reward you with some shares to flip.

But that’s hard to do and still requires goodwill from banks, and you might want a more self-help approach. Here’s one, from Bloomberg’s Denitsa Tsekova:

A series of unusual multibillion-dollar flows has rippled through the exchange-traded fund market in the past week, suggesting some major investors used a contentious method to get exposure to the initial public offering of SpaceX. At least one fund manager placed temporary restrictions on its product in a bid to curb the practice.

The biggest flows were in Cathie Wood’s flagship strategy. The $6.9 billion ARK Innovation ETF (ticker ARKK) posted a record $4.6 billion inflow late last week before recording its largest-ever outflow of $6.2 billion in the following session, data compiled by Bloomberg show. That activity coincided with ARKK acquiring roughly 1.7 million shares of SpaceX on the day the rocket maker listed, and echoes similar patterns seen during a number of IPOs in the past year.

The theory among ETF specialists is that such flows represent an IPO arbitrage strategy. The tactic involves pouring cash into an ETF before an offering in which the fund is expected to receive shares, then pulling the money out once the stock begins trading. Such a maneuver could provide indirect, or synthetic, exposure to the IPO and would profit from any post-listing gain, though it risks distorting fund performance and disadvantaging longer-term holders.

“Investors appear to be using ARKK as a piggyback vehicle for access to opportunities they may not be able to easily buy directly,” said Bloomberg Intelligence ETF analyst Athanasios Psarofagis. “That trend has become more common in ARKK around major IPOs.”

We have discussed this trade before. If you know, or suspect, that ARKK will get a big SpaceX IPO allocation at $135, then you can buy shares of ARKK the day before the IPO and sell them the day after: The net asset value of ARKK will have been increased by the SpaceX IPO pop. Of course, the net asset value might also go up or down because of movements in the other stocks in the portfolio, but you can hedge those out: ARKK’s portfolio is public, so you can short the other stocks in ARKK as a hedge, leaving you with just SpaceX IPO pop exposure.

And, most neatly, you can do the whole trade with ARKK. The actual way that money flows into and out of ARKK is through in-kind creations and redemptions. You borrow the stocks in ARKK’s underlying portfolio, you deliver them (short) to ARKK, you get back shares of ARKK, you wait a day, you hand back the ARKK shares and get back the underlying portfolio, presto, now you own some SpaceX at its IPO price. “The scale of the flows,” reports Tsekova, “suggests the strategy uses a fund’s creation-and-redemption mechanism rather than buying and selling shares on an exchange, and that one or more large institution is behind the flows.”

People get mad about this, though oddly not Cathie Wood:

The potential losers are clear: existing ETF holders will see both any gains from the offering and their stake in the newly listed company reduced.

In a bid to protect holders of the $2.4 billion ERShares Private-Public Crossover ETF (XOVR), Joel Shulman, founder and chief investment officer of ERShares, restricted primary creations for the fund ahead of the SpaceX offering and imposed a 2% redemption fee. Investors in XOVR already had indirect exposure to the company before its public debut.

“We stopped the creations the week before the IPO, and the intent was to protect our long-term loyal shareholders from dilution,” Shulman said. “We implemented a redemption fee to cover any related costs associated with disruptive large inflows and outflows.” …

“We don’t participate in every IPO,” Wood said in a Bloomberg Television interview [in 2025]. “If they guess right, great. And if they don’t, well, then they have nothing. They’ve incurred some costs.”

There are other, fuzzier versions of this trade. Tsekova mentions “the Baron First Principles ETF (RONB) from Baron Capital” as another ETF that got inflows and outflows similar to ARKK’s. But Baron runs a couple of other funds that held SpaceX stock for a long time prior to the IPO. Those funds bought SpaceX at prices well below the IPO price, so if you were a long-term holder of those funds then you profited by SpaceX’s increasing valuation over the past few years. But, also, if you bought those Baron funds last month, you got a nice IPO boost. Barron’s (no relation) reported last week:

Baron Capital is one of the largest holders of SpaceX stock in the mutual fund industry and the firm wrote up the value of that stake a week before the coming IPO pricing, expected on Thursday.

The $17 billion Baron Partners fund and the $3.7 billion Baron Asset both got big gains last Thursday — almost 7% and 8%, respectively. As of March 31, Baron Partners had nearly 30% of its assets in SpaceX and Baron Asset had 23%.

The firm had said on its website that the SpaceX shares would be repriced on June 4, last Thursday. ...

The Partners fund was carrying its SpaceX stock at about $105 a share on March 31, compared with the planned IPO price of $135. The March 31 value appears to be the latest public mark on the SpaceX stock by the firm. …

Thursday's gains came a day after SpaceX filed its updated prospectus, which included the $135 targeted IPO price. 

Baron’s funds marked SpaceX at a $1.25 trillion valuation until June 4, “using prices of recent financing transactions.” But by, you know, June 3, it was widely reported that SpaceX’s IPO valuation would be in the $1.8 trillion area. Baron’s funds’ net asset values reflected historical prices of SpaceX stock, but you might reasonably have valued SpaceX based on its very near-future price. “SpaceX will go public next week at $135 and immediately trade up 20% to $162,” you might have thought, “so I should buy it from Ron Baron today at $105.” Morningstar analyst Jeffrey Ptak estimated last week that Baron Partners “received $1.8 billion in inflows” from June 1 through June 11, suggesting that a lot of people had a similar idea: Buy Baron Partners to get exposure to SpaceX at $105 or $135, and then — perhaps — sell when SpaceX predictably gets to $160.

This is a little bit like the opposite of the problems private credit had this year. Private credit funds had made a bunch of loans that they carried on their books at their original price, but the market was skeptical about those values. A fund might have a net asset value of $100 per share, but its investors thought it was worth $80 per share. Therefore, a lot of them wanted to redeem: If the fund was only worth $80 per share, but would pay you out at $100, that was a good trade. Baron arguably offered the reverse trade: It had SpaceX shares worth $135 or $160, but would let you buy them (along with the rest of its holdings) at $105. The inflows “likely diluted the position as a percentage weight,” Ptak noted: Redeeming private-credit investors at NAV is bad for other holders if the NAV is too high; letting in new mutual-fund investors at NAV is bad for existing holders if the NAV is too low.

Cathie Wood and Ron Baron are the classic buy-and-hold SpaceX true believers who of course are going to get shares in the IPO. But they run funds that are open to outside investors, and those outside investors might not all be buy-and-hold true believers. ARKK and Baron earn the IPO pop, because they buy shares at the IPO price (or better), but they do not immediately monetize it, because they don’t sell the shares the day after the IPO. But apparently some of their outside investors can synthetically monetize it; they can extract the IPO pop and capture it themselves.

We have talked a few times recently about the multistrategy hedge fund gig economy. The big multimanager multistrategy “pod shop” hedge funds are in the business of identifying talented portfolio managers, hiring them, allocating capital to them, supervising them, and scaling their capital up or down as their opportunities grow or shrink. In many ways, it makes sense to hire the portfolio managers as employees. You can supervise them more closely, risk-manage them more efficiently and yell at them more loudly if they are employees of your firm.

In other ways, though, they don’t really need to be employees. The typical reasons to hire an employee, rather than an outside contractor or gig worker or service provider, mostly don’t apply to hedge fund portfolio managers. You mostly don’t want centralized decisionmaking: Your risk-adjusted returns will be better if every portfolio manager is making her own trading decisions, not doing what she thinks the boss wants. Similarly, you mostly don’t want close collaboration, sharing of research and ideas, or serendipitous conversations by the water cooler. You don’t want a shared culture or esprit de corps; you want independent thinking, uncorrelated bets and vicious competition. Why do you need everyone to share an office and have the same name on their fleece vests?

And so we have talked about “buy-side alpha capture,” which all the multistrats seem to be discovering at the same time. Basically, the idea is that if you run a big hedge fund, instead of just allocating capital to your employees’ ideas, you can also pay outside hedge fund managers for some new ideas, and allocate capital to those. You apply your skills — at capital allocation and talent identification and idea vetting — to a wider assortment of managers and trades. You have more independent ideas to allocate capital to, and you don’t have to pay the outside managers a huge premium to come work for you exclusively. This month we discussed reports that Citadel and Point72 are getting into this business, and now Bloomberg’s Liza Tetley reports:

Chris Rokos’ eponymous firm is exploring starting a program to tap trading intelligence from other hedge funds, joining a slew of peers racing to sign up such contributors.

Rokos Capital Management is looking at establishing a so-called buyside alpha capture program that would receive stockpicking ideas from external equities managers, people familiar with the matter said, asking not to be identified discussing private information.

There is also a simpler, lower-tech way to achieve some of the same goals, which is just to take some of your own hedge fund’s money and invest it in other hedge funds. That happens too. Bloomberg’s Nishant Kumar reports:

Brevan Howard Asset Management is preparing to give money to stocks-focused hedge funds, adding to a wave of giant firms forking over cash to outside talent. ...

The new allocations will scale an existing part of Brevan Howard’s multistrategy Alpha Strategies hedge fund, the person added.

The biggest hedge funds are increasingly turning to external firms as a way to deploy the enormous assets they’ve accumulated in recent years, without the hassle and expense of hiring more traders. Goldman Sachs Group Inc. has estimated that more than 70% of multimanager platforms allocated some capital externally in 2025, up from 54% in 2022.

I wrote last week that “perhaps the next natural evolution” of multistrategy hedge funds “is a step back toward the fund-of-funds model,” and that does seem to be what is happening.

The other day I wrote some jokes about indexes of prediction markets, among them:

If the index was mostly high-probability bets (“bonds,” as they say), perhaps it would earn some expected return? (Those bets are arguably a sort of insurance, so you might expect to earn some premium.) ...

I will say, though, that “bet $1 million on an extreme favorite on Polymarket every month and, if you win, do it again” has almost the shape of a good financial product? Like, that pays a high bond-like return most of the time; it has some risk, but that risk is uncorrelated to other financial instruments. (It goes to zero when Spain draws with Cabo Verde.) Build an index, build a passive exchange-traded fund to track it, what is everyone waiting for?

A reader pointed out that there is a literature, sort of. Here’s an archived ESPN blog post about an unpublished paper by Mike Wohl:

Wohl found his advantage in betting the money lines for college football teams that were favored by 20-25 points. He wrote in his paper: “There were 376 games in the last six seasons (approximately 62 per season or approximately 4.5 per week) that had spreads of between 20.0 and 25.0. Of those 376 games, 94.95 percent of the favorites won the game outright. Investing equal amounts on all 376 contests produces six straight years of profitable returns with an average annual (non-compounded and non-annualized) return of 12.24 percent.”

That was in the early 2010s, and I have not examined the out-of-sample results, but it is plausible that money-line betting on 25-point favorites^[1] in college football still offers bond-like returns. People want to buy lottery tickets (bet on underdogs), so they pay for them; you are selling them the lottery tickets and collecting a positive expected return. (Or: You are taking a lot of risk, by betting $100 to win like $5, and you are getting paid for that risk in the form of positive expected return.)

It goes without saying that this is not investing or gambling advice. But I wrote in March 2025 that “I kind of think we are about two years away from a sports gambling ETF.” There is a whole genre of ETFs that provide steady bond-like income by doing weird trades; we have discussed BOXX (selling option box spreads for income) and CAIE (selling autocallables, that is, crash insurance on the stock market, for income). “Selling longshot football bets for income,” sure, why not. If the Securities and Exchange Commission starts allowing prediction-market funds, there’s going to be an ETF that advertises steady 12% annual returns achieved with a diversified portfolio of college football bets.

If you are a salesperson or trader or other revenue-generating employee, and you can demonstrably and reliably produce $1.8 million per year in additional gross profit for whatever company hires you, how much should you get paid? I feel like my rough generic guess is 50%, so like $900,000 per year. There are probably fields where the answer is more like 10% or 20%, so you’d get low-to-mid six figures. There is some theoretical argument that the right answer is, like, $1.79 million.

Here’s an organization that is hiring employees who on average generate $1.8 million in gross profits per year, but that pays them in kibble, dog treats, and pats on the head, because they are dogs:

A shortage of dogs used to sniff out drugs and other contraband are costing South Africa’s tax agency millions of rand in foregone customs revenue.

The agency’s Detector Dog Unit has 66 handlers, nine of whom are without dogs, and needs to acquire 14 additional canines to meet operational requirements, Finance Minister Enoch Godongwana said in a written reply to a lawmaker’s question on Thursday.

Each dog contributes about 29.7 million rand ($1.8 million) to the state’s coffers each year by detecting goods that would have otherwise evaded customs duty, indicating that the shortfall in animals is costing some 415 million rand annually in lost income, he said. 

The two obvious jokes here are “they should try offering higher salaries” and “when will these dogs be replaced by AI.” Also, though, good job dogs? “More than 60% of dogs assessed do not meet the testing requirements,” says Godongwana, but if the ones who do meet the standards are bringing in $1.8 million a year then maybe you should lower the standards? Like a $500,000-a-year dog is still a pretty good dog?

Or maybe I have it all wrong; maybe I’m deceived by that average revenue figure. Maybe this is a superstar economy, the top dogs are bringing in $20 million a year, and the marginal dog is barely breaking even after deducting the handler’s salary and the cost of kibble. Do they pay extra to retain the top dogs? I guess they can’t really quit to go work for Zimbabwe; not a lot of dog job mobility. Every so often I get emails from students saying “I am taking a finance class and need a topic for my final paper, any ideas,” and I only rarely have any ideas, but someone should definitely do an empirical study of compensation structures, revenue generation and incentive schemes for South African customs detector dogs.

Software Buyout King Orlando Bravo Attempts an AI-Era Reboot. Warsh Overhauls How the Fed Talks and Keeps Markets Guessing on Rates. Intel Shares Soar After Trump Says It Struck Apple Chip Deal. Trump’s ‘nuclear bros’: young entrepreneurs race to deliver US atomic revival. Billionaire John Paulson Scores Victory in Puerto Rico Legal Fight. CME Sues CFTC as Battle Over Perpetual Futures Trading Heats Up. JPMorgan Chase cuts off Anthropic access for its Hong Kong staff. Florida Pizza Mogul Played Quiet Role in Trump Sons’ Stock Deals.

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!