It was illegal, from about 2021 through about 2025, for employees of financial firms to text or WhatsApp about work on their personal phones. The theory was that US securities regulations require firms to keep copies of “all communications sent” by their employees, “including inter-office memoranda and communications,” that are “related to [their] business as such.” So if a financial services employee sends an email to a client, or to a colleague, her firm needs to retain a copy of that email. And if she texts a client or colleague from her personal phone, then the firm won’t have a copy, which violates the law.

This theory was essentially invented by Gary Gensler, who was the chair of the US Securities and Exchange Commission at the time, and I always thought it was a strange and aggressive interpretation of the rules. But you can understand why the SEC liked it:

1. The SEC is in the business of detecting actual fraud, and if banks and brokers make sure that all of their communications are preserved in searchable electronic form for years, then it will be relatively easy for the SEC to detect fraud. Ask for a pile of messages every once in a while, run a search on them for terms like “fraud” or “spoof” or “muppets” or “champagne,” and see what comes up. 
2. The SEC is in the business of extracting fines, and it extracted billions of dollars of fines, with very little effort, by charging dozens of financial firms with texting on personal phones.

Every time the SEC would announce a settlement, it would issue a press release in which Gensler would say stuff like “As technology changes, it’s even more important that registrants appropriately conduct their communications about business matters within only official channels,” even though that was obviously wrong. (As I once wrote: “Imagine if that was really the rule! You can’t have lunch with a client and talk about business, or have beers with your colleagues and gripe about work, because that does not create a searchable archive for the SEC to review.”)

This was a good fine-extracting business for the SEC because it is human nature, apparently, to text about work on your personal phone, and there was nothing banks could do to stop their employees from doing it. The SEC once found that Qatalyst Partners LP, the investment banking boutique, “willfully violated” the recordkeeping rules, even though (1) it had policies against texting from personal phones, (2) it frequently trained its employees on those policies, (3) it provided them with compliant text-messaging devices and told them to use them exclusively, (4) it captured and archived Slack and LinkedIn messages, etc. Qatalyst did everything it could to stop its employees from texting on their personal phones, but some of them did anyway, because that is the human condition. To the SEC, in 2024, that was illegal. [1]

This maximalist approach created an ironic problem for the SEC, though. The problem is:

• The SEC is a government agency and has its own recordkeeping obligations, and
• The SEC is an organization made up of humans, who definitely text about work on their personal phones.

Eventually people noticed this problem. First the SEC noticed: In April 2024, after a few years of fining banks billions of dollars because their employees used WhatsApp, the SEC finally blocked its own employees from using WhatsApp and Signal on their work phones. And then Coinbase Global Inc. noticed: Coinbase was in various disputes with the SEC, and in summer 2024 it asked “to search Gensler’s personal devices for any statements he's made on the cryptocurrency industry since being appointed leader of the agency.” 

The SEC’s response to this was, amazingly:

1. Actually Gensler was above reproach and communicated only on approved channels using his SEC-issued phone, which properly preserve and archived all of his communications, buuuuuuuuuuuuuuuuuut
2. We accidentally deleted them, sorry!

Try to imagine what Gary Gensler would have said if a bank tried that line on him.

The SEC and Coinbase have now settled their dispute, and Coinbase Chief Legal Officer Paul Grewal has an op-ed in the Wall Street Journal today gloating about it, as he absolutely should:

When Coinbase sought FOIA records to shed light on the agency’s interpretation of securities laws with respect to digital assets, the SEC refused—so we took it to court, and a judge ordered the SEC to produce them.

But there was a twist: The agency tasked with policing corporate record-keeping somehow lost reams of its own text messages between Mr. Gensler and other officials during the most intense period of the anti-crypto campaign. The SEC blamed a process that “automatically wiped” certain data—a hollow excuse from an agency that had extracted billions of dollars in penalties from financial services firms for nearly identical failures.

As part of our settlement, the SEC will pay $150,000 and fix its record-retention policies.

It is the only appropriate ending to the SEC’s crusade against texting from personal phones. I mean, I guess it’s the ending. The SEC hasn’t brought any texting cases in a while, and I don’t think they’re a priority any more. And as far as I know Coinbase is the only financial firm that has brought a texting case against the SEC. You could imagine others doing it now, though? All the banks that paid millions of dollars of texting settlements could demand to see what Gensler texted about them, force the SEC to be like “ahhh we lost those too sorry,” and demand partial refunds of their settlements. There were dozens and dozens of SEC texting cases for a few years; maybe now there will be dozens of SEC reverse texting cases.

An accidental but important feature of the US tax code is that exchange-traded funds don’t pay taxes when they sell stock. If you buy some stock for $100 and sell it a few years later for $500, you have $400 of capital gains and pay about $80 of taxes. If you are an investor in a mutual fund, and the mutual fund buys some stock for $100 and sells it for $500, you have $400 of capital gains and pay about $80 of taxes: The mutual fund’s taxable transactions are passed on to you. [2]

If you are an investor in an ETF, though, and the ETF buys some stock for $100 and sells it for $500, you pay no taxes. Instead, when you eventually sell your shares in the ETF, you pay taxes on all of the gains that have accumulated in the ETF. But that tax deferral is quite valuable: Paying $80 in 20 years is better than paying $80 now, and if you die and leave your appreciated ETF shares to your heirs, then nobody pays the tax at all.

Like I said, this is apparently an accident — it doesn’t seem as if Congress wrote the tax code quite intending this result — but it is a real and, at this point, quite beloved feature of ETFs. ETFs, I often write, just are mutual funds that don’t pay taxes.  There is not, as far as I can tell, much desire in government to change it. ETFs are popular, people like the fact that they don't have to pay taxes, it’s fine. 

It’s accidental from Congress’s perspective, but it is not accidental from the financial industry’s perspective. Quite a lot of work goes on beneath the surface of the simple story that I have laid out. The tax code doesn’t say “ETFs can sell stocks without triggering taxes.” What it says is that in-kind creation and redemption of ETF shares are not taxable transactions. Most stock ETFs have in-kind creation and redemption: A trading firm can acquire a basket of S&P 500 stocks and deliver them to an S&P 500 ETF to get back newly created shares of that ETF, or it can deliver shares of the ETF and redeem them for a basket of the underlying stocks. Because the ETF never sells the underlying stock — it delivers it in an in-kind exchange for its own shares — it doesn’t have to pay taxes; the in-kind exchange is not a taxable event. [3]

The ETF industry figured out that it could extend this logic to cover, not just inflows and outflows, but also trades. If an ETF wants to sell its ABC shares and buy XYZ shares, it won’t do that for cash. Instead, it will ask a trading firm to (1) hand it the XYZ shares in a “custom creation” transaction (trading firm gives ETF XYZ shares and gets back ETF shares) and (2) take back the ABC shares in a “custom redemption” transaction (trading firm gives ETF back its own shares and gets back ABC shares). The trading firm does the cash trading, but the ETF doesn’t, so it has no taxable transactions. These trades are often called “heartbeats,” because from the outside they look like money flowing into the ETF and quickly flowing back out again.

All of this suggests a question: Why should you ever pay taxes when you sell stock? ETF technology exists, and it is sort of backward and old-fashioned not to use it to defer taxes. Conceptually, what you should do is:

1. Get yourself an ETF.
2. Do an in-kind creation transaction, where you contribute your stock portfolio to the ETF. 
3. Now, instead of owning a bunch of stocks, you own one ETF, which owns the bunch of stocks.This exchange was a tax-free transaction: You contributed your stock portfolio and got back ETF shares, but it all happened in-kind, so you don’t pay any taxes. (It is called a 351 exchange, after the section of the tax code that says it’s not taxable.)
4. Now, if you want to sell one stock and buy another stock, you have the ETF do it. Or, rather, you have the ETF do a heartbeat transaction in which it gets rid of the old stock and acquires the new stock in in-kind redemptions and creations, so you don’t pay any taxes.
5. You manage your portfolio however you want: You sell stocks you don’t like and buy stocks you do, you sell down concentrated appreciated positions and diversify, whatever, all without paying taxes, because all of it happens in an ETF.
6. If you need money — not to buy stocks, but to buy houses or groceries or whatever — you sell some shares of the ETF, which does trigger capital gains taxes. But you pay taxes only when you take money out for consumption, not when you rebalance your portfolio.
7. Or not. Or you borrow against your portfolio (of ETF shares) to fund your consumption, never sell your shares, never pay taxes, and leave the ETF shares to your heirs with a basis step-up.

This is not tax or legal or investing advice, and there are some important nuances. I am casually assuming that you run the ETF, for instance, but in practice you’ll probably have some professional ETF manager running the ETF and doing the diversification. You’ll probably need a pretty big portfolio to get yourself an ETF; no one will do this for a $10,000 account. Also, Step 2 — transmuting your own portfolio into an ETF in a 351 exchange — is subject to some restrictions: You have to start with a somewhat diversified portfolio; you can’t just plop one giant appreciated stock position into an ETF and then heartbeat it away into a diversified S&P 500 fund. [4] This is a real restriction, as the most obvious use case for this technology is to diversify away your giant portfolio of low-basis appreciated stock that you got for founding a successful company. Still, if you’ve also been investing over the years, you might be able to create your personal ETF and then use it to diversify your holdings and trade tax-free.

That’s obviously a good trade for you, which means that it’s a bad trade for the US Treasury. A lot of people would prefer to defer taxes on their stock trades, this is a straightforward way to do that, so a lot of people will use it, so the Treasury will collect less tax money than it otherwise would. And, again, this is all basically accidental. Congress and the Treasury did not sit down and think “how can we encourage people to trade more stock by making some stock trades tax-free?” Treasury could change the rules, if it doesn’t like how this is going.

Bloomberg’s Justina Lee, Surya Mattu, Denitsa Tsekova, Isabelle Lee and Zachary Mider have a big story about how lots of people are doing 351 exchange ETFs because it’s a good trade:

The MIG Core ETF (ticker MIGO) … is what is known as a 351 conversion, a type of transaction that’s helping wealthy investors deal with years of market gains at the expense of government coffers – and attracting US Treasury Department attention as a result. They’re part of Wall Street’s booming tax alpha complex, which offers strategies that seek to beat the Internal Revenue Service as well as the market.

A 351 sees an investor turn a portfolio of assets into an ETF, typically so they can rebalance winning holdings without paying an immediate tax bill – a handy maneuver for anyone who has found themselves with a concentrated position in one company or a few stocks. Technically it’s a deferral, since tax will still be owed when any shares in the ETF are eventually sold. But it keeps more money invested for longer, and investors can pursue several strategies that reduce or even eliminate what is owed.

“The big benefit is taking a bunch of built-up gains in individual stocks and consolidating them into a diversified portfolio,” said Rory Riggs, a biotech investor whose firm Syntax Advisors helped create some of the first ETF conversions, including with his own assets. “It is about not having to realize those gains until you want to.”

Conversions are not easy to spot in the sprawling and saturated ETF world, but they’re proliferating fast. A Bloomberg News analysis of filings to the US Securities and Exchange Commission has identified 105 ETFs created through 351 exchanges. Collectively they boasted $22.1 billion assets at launch and helped defer at least $6.5 billion in embedded capital gains, the filings show. More than half listed last year, as major asset managers including AllianceBernstein LP and American Century Investments began creating them for clients. …

“High-net worth folks and their advisors see this as a slam dunk,” said Brent Sullivan, who runs the widely followed Tax Alpha Insider blog and recently wrote a paper on the legal issues arising from the phenomenon. “When you have a portfolio that is 50% unrealized capital gains, it starts to become a really attractive option.”

MIGO is an ETF seeded with the stock portfolio of the guy who invented the Hot Pocket. But Lee, Tsekova and Mider also have a story about how the Treasury is starting to wonder if it’s maybe too good a trade:

The US Treasury Department has expressed concern over a number of high-profile tax strategies touted by Wall Street that it says may be “too good to be true.” …

The strategies under scrutiny include so-called 351 conversions, box-spread exchange-traded funds, products that offset ordinary income, and funds that avoid dividend income by flipping between other ETFs. Speaking at a Wall Street Tax Association seminar, Kevin Salinger, deputy assistant secretary for tax policy at the Treasury, and Erika Nijenhuis, senior counsel, said the department has no wish to over-engineer rules, but it cannot ignore a market developing around transactions with results Congress did not appear to intend.

By the way. My description above suggests that these trades are essentially personal ETFs: You’re a person with a big appreciated stock portfolio, you plop it into an ETF, you diversify the ETF over time, etc. That is clearly how people think about these ETFs: Asset managers create them as tax trades for clients, not as broadly marketed retail products. Each wealthy family gets its own 351 ETF to deal with its own appreciated stock portfolio.

On the other hand, they are exchange-traded funds. They’re listed on the exchange, there’s a prospectus, anyone can buy them. You can just go on the stock exchange and buy a tiny slice of the stock portfolio of the guy who invented the Hot Pocket. I’m not sure why you’d want to — surely the point here is more to optimize taxes than it is to optimize performance? — but the people who do these transactions do still want their stock portfolios to go up, and they have sophisticated advisers. Maybe if it’s good enough for the Hot Pocket family it’s good enough for you? Someone should launch a diversified ETF of 351 exchange ETFs.

The smart way to think about stock prices is that they reflect the market’s expectations about future cash flows of the business. If SpaceX trades at $200 per share, that means the market thinks the present value of its future earnings is $200 per share. If it trades at $120, that means the market has a lower estimate.

The dumb way to think about stock prices is that rare Pokémons are more valuable than common Pokémons, and if there are 639 million shares of SpaceX available to buy then those shares will cost more than if there were 5.33 billion shares. You are not buying cash flows. You are buying an electronic token that is in greater or lesser supply; the more of the token that is available, the less you’ll have to pay for it.

Obviously both models are true to some extent, and probably the first model is more true in the long run, but, you know. Bloomberg’s Bailey Lipschultz reports that SpaceX has now set its first lockup release date:

SpaceX has set the stage for one of the largest share unlocks in capital markets history, with as much as $116 billion worth of stock becoming eligible for sale for the first time next month.

Restrictions preventing some insiders from selling as many as 911.5 million shares will lift on August 6, two days after the rocket, satellite and artificial intelligence company reports quarterly results for the first time. That’s just the start, with billions of shares set to be freshly eligible for trading by the end of the year. ...

Many [investors] will get their chance to exit by early December, when the number of shares available to trade will soar to 5.33 billion, up from around 639 million shares now, according to the initial public offering prospectus for Space Exploration Technologies Corp., as it’s formally known.

We have talked about this before. SpaceX’s initial public offering was the largest ever, but next month’s lockup release will be considerably bigger: Only 639 million shares became available in the IPO, but 911.5 million more will become available on Aug. 6, and billions more in December. The IPO was a carefully choreographed matching of supply and demand in which a fairly limited amount of SpaceX stock was sold to handpicked investors who were mostly expected to hold it for the long term. The lockup release is just, meh, here’s some stock, who wants it? Anyway the stock is down about 10% since the IPO.

A standard story that you used to hear about private equity went like this. There are a lot of good, profitable, unglamorous small and medium-sized businesses that were founded by Baby Boomers decades ago. Those businesses provided a good living for their founders, whose children grew up in luxury and became documentary filmmakers. As the founders approach retirement age, they have no one to take over their unglamorous businesses. There is a huge supply of good cash-flowing businesses with motivated sellers. Therefore, it is good to be a buyer: A private equity firm with a pile of cash and operational expertise can buy up lots of these aging-founder-owned businesses cheaply.

Even today you will occasionally hear some version of that story, but it’s a bit less popular than it was. Some of those founders have already retired, plus so many private equity funds and search funds had this idea that the supply is a bit picked over. If you are a plumber nearing retirement, you got 20 calls a day for years from people who wanted to buy your business; the buyers turned out to be at least as motivated as the sellers.

Now, though, there is a revised story that goes like this: There are a lot of good, profitable, unglamorous businesses that were founded by Baby Boomers decades ago and are now owned by private equity funds. Those businesses provide a good cash flow for their private equity funds, but private equity funds don’t want good cash flow; they want an exit. The private equity funds are reaching — or past — their expected lifespan, and as they approach retirement, they have no one to take over the business. The people running the funds are desperate to do something else; the investors in the funds are desperate to get their money back. There is a huge supply of good cash-flowing businesses with motivated sellers. Therefore, it is good to be a buyer.

Here’s an evocative Wall Street Journal article about the graying private equity funds whose children no longer want to run their plumbing businesses or whatever:

A record level of private-equity investments are stuck in funds limping along past their intended lifespans.

Often known as zombie funds, these funds are no longer raising money or making new acquisitions, in part because fund managers haven’t been able to sell their remaining assets. The net asset value of U.S. private-equity assets stuck in funds at least a decade old reached an all-time high of $348.5 billion at the end of 2025, according to PitchBook data. That is 3.5 times the amount in 2015 and more than 100 times that of 2005.

The slowdown in private-equity sales has fueled frustration among investors eager to cash out. Many managers of funds launched in the mid-to-late 2010s struck deals for their existing portfolio companies at the peak of the market in 2020 and 2021, when interest rates were nearly zero. Buyers are now unwilling to pay peak prices at higher borrowing rates, leaving those funds stranded past their typical lifespans.

“In some cases, it’s three guys and a Labrador running the last few assets of the fund,” said Finbarr O’Connor, chief investment officer and founding partner at Treo Asset Management, an outsourced manager for zombie funds. …

Francisco Alvarez-Demalde, managing partner of private-equity firm Riverwood Capital, said he isn’t seeing fire sales, or funds selling companies at deep discounts. But he said the industry will likely experience a correction.

“There’s going to be a cleansing between different funds, different assets,” said Alvarez-Demalde, who added that Riverwood is well-positioned. “Some assets will lose in that.”

I just feel like the Labrador could have been part of the motivated-Boomer-founder version of the story too. Anyway, I suppose this is the basic story behind the rise of secondary deals. Huge supply of motivated sellers! Good buying opportunity for private equity!

OpenAI Models Hacked Another Company’s Systems by Mistake. Goldman Sachs creates private markets platform as rich investors seek the next SpaceX and Stripe. The Startup Insiders Who Stash Huge Sums in Tax-Subsidized Retirement Accounts. The Bitcoin Slump Is Crushing Companies That Stockpiled Tokens. The US has collected about $13bn of Venezuela’s oil money: Where is it? OpenAI Is Adding Two Independent Board Members Ahead of IPO. A Luxury Taxi Company Wants to Stop New York From Tracking Its Wealthy Riders. Elon Musk says Grok Imagine will make ‘historically accurate’ AI adaptation of Homer’s Odyssey. Ethical fintech boss tried to set up arms deals with Wirecard’s Marsalek. Man Tied to $100 Million NJ Deli Scam Gets 21 Months in Prison.

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!