Bloomberg Chief Financial Writer: The Underlying Logic Behind U.S. Public Companies' Frenzied Purchases of Cryptocurrency

Original Title: "Sell Your Crypto on the Stock Exchange"

Author: Matt Levine

Compiled by: Odaily Planet Daily jk

Crypto Treasury Companies

Last Tuesday, SharpLink Gaming Inc. was still a company focused on online marketing for lottery games, with a stock price of about $2.91 per share and a market capitalization of only about $2 million. Although it is still listed on NASDAQ, it is actually on shaky ground. A few weeks ago, it had just undergone a reverse stock split to maintain a share price above the minimum requirement of $1 set by NASDAQ, while it also failed to meet NASDAQ's basic requirement of at least $2.5 million in shareholder equity.

So, SharpLink announced on that day a round of stock issuance to raise $4.5 million at a price of $2.94 per share. The official statement was that the funds would be used to "restore compliance with NASDAQ's minimum shareholder equity requirements." However, the company also added, "We may use some of the funds to purchase cryptocurrencies as part of a treasury management strategy we are considering."

To be honest, this is not surprising. SharpLink is technically a listed company, but by real-world standards, it resembles more of a "shell company"—with a market cap of $2 million and annual revenue of only a few million dollars, such a scale struggles to support the operational and compliance costs of being a listed company. In the past, this was a problem.

But in 2025, this has become an opportunity. SharpLink possesses two highly sought-after yet relatively scarce assets in the current market:

• It has a shell of a U.S. listed company;
• And it has basically done nothing with this shell.

This makes it an ideal candidate for "transforming into a crypto treasury." As I often say, the U.S. stock market is willing to pay more than $2 for a $1 crypto asset. This phenomenon has long been recognized by entrepreneurs in the crypto space. If you have a large amount of Bitcoin, Ethereum, Solana, Dogecoin, or even TRUMP, the best way is to put them into a U.S. listed company and then sell them at a higher price to investors in the secondary market.

But to do this, you first need a listed company. Such shell resources are not abundant, as most quality companies are already quite busy. If you call Apple Inc. and say, "We want to merge our Dogecoin holdings with you to make them more valuable," Apple will surely decline.

The real opportunity lies with those marginal listed companies: they are still listed but barely hanging on. The phones of these companies are being bombarded with calls every day.

So we saw the following press release:

SharpLink Gaming announces a $425 million private placement, officially launching its Ethereum treasury strategy…

SharpLink continues to operate as a company focused on providing performance-driven online marketing services for the U.S. lottery industry.

According to the announcement:

• After the private placement is completed, SharpLink will officially launch its Ethereum Treasury Strategy;
• Joseph Lubin, founder and CEO of Consensys and co-founder of Ethereum, will serve as chairman of SharpLink's board after the private placement transaction is completed;
• Investors in this round of private placement include several well-known crypto venture capital and infrastructure companies, such as ParaFi Capital, Electric Capital, Pantera Capital, Arrington Capital, Galaxy Digital, Ondo, White Star Capital, GSR, Hivemind Capital, Hypersphere, Primitive Ventures, and Republic Digital.

In other words, Consensys, a blockchain software company led by Ethereum co-founder, aims to operate a $425 million Ethereum asset pool, while the capital market values this asset far above its actual worth. SharpLink happens to be the ideal "shell resource" to achieve this goal. Thus, Consensys and its co-investors will invest $425 million to purchase SharpLink's stock at a price of $6.15 per share, and SharpLink will use these funds to buy Ethereum (ETH).

At the opening today, SharpLink's stock price was $33.93, and by 1:30 PM, the trading price was about $35, bringing the company's market capitalization to $2.5 billion. In other words, this $425 million worth of Ethereum assets has received a $2.5 billion valuation in the U.S. stock market.

It is important to note that SharpLink currently does not hold any Ethereum. The investors are providing dollars, not Ethereum. This is not "we already have a lot of ETH, so let's go public," but rather "since the U.S. stock market is willing to buy $1 of ETH for $2 or even $6, we should certainly take advantage of this arbitrage opportunity."

From a certain perspective, this is almost an open arbitrage opportunity. Theoretically, anyone with hundreds of millions of dollars in cash can buy cryptocurrencies in the market and then put them into a listed company shell, and the U.S. stock market will immediately give a paper profit of more than five times. What you really need, besides startup capital, is to find a small listed company that can "hold the coins."

Remember that little snack shop in New Jersey? It once had a fully diluted market cap of $2 billion. It just came a bit too early. In fact, this snack shop (or the shell company behind it) existed precisely for this kind of "trading model": a listed shell company paired with a transitional small business (like opening a snack shop), with the real purpose of completing a reverse merger with a private company—most likely a foreign enterprise—to go public. As for why the snack shop people manipulated the stock price and are now going to jail, I still haven't fully figured it out, but it actually has nothing to do with this trading logic itself. The core of this play is—finding a suitable merger target.

That snack shop unfortunately got shut down before the "crypto treasury company" model really took off, but goodness, if it could have held on until today, it would have been an incredible deal. Imagine if that New Jersey snack shop could merge with a $425 million Ethereum asset pool; its $2 billion market cap back then would actually make sense. Now, it only takes a few hundred million dollars in cryptocurrency, plus a micro listed company, to combine and achieve a valuation of billions in the capital market.

Back to SharpLink: last Thursday, its stock price rose by 35%, and on Friday, it rose another 79%, while the related transaction was only officially announced today. I suspect there may have been information leaks or insider trading, but I wouldn't jump to that conclusion easily. After all, SharpLink has long publicly stated that it is considering a crypto treasury strategy, and it is itself an ideal "candidate shell" company (listed but with little business). Even without insider information, it is entirely reasonable to speculate: "This small company is likely to issue a crypto-related announcement soon, and when that happens, the stock price may soar by hundreds of points, so I might as well buy a little first." Of course, this is not investment advice, and my use of "reasonable" is not in the traditional sense of "rational."

The entire event is quite absurd, but I want to highlight three particularly absurd things.

First: Is this trick still working?

I have written a lot about "crypto treasury companies" in the past few months—MicroStrategy Inc. is basically the earliest to play this game and has been doing it for many years. Recently, this model has suddenly exploded. Intuitively, it shouldn't all be successful.

After all, MicroStrategy is a large listed company with a mature investor relations team, effective promotional strategies for retail investors, and indeed holds a large amount of Bitcoin, along with first-mover advantages, diversified financing channels, and natural advantages of being included in leveraged ETFs and some indices. If some investors (like mutual fund managers or certain retail investors) want exposure to Bitcoin but cannot buy coins or ETFs directly, then MicroStrategy may indeed deserve a certain valuation premium.

But the problem is that now a whole bunch of "mini MicroStrategies" are being treated with crazy premiums by the market. The market's love for these "new crypto treasury companies" seems to have no end. I cannot explain this phenomenon at all.

A month ago, I wrote a line: "The current situation is like the crypto circle is continuously playing the U.S. stock market, and the U.S. stock market keeps falling for it." Now it feels even stronger.

Second: Are people still doing this?

This is actually not so surprising: I wrote last month that "if you run a crypto investment fund and haven't acquired a U.S. listed company with a stalled or thin business to play this arbitrage, that would be mismanagement."

For all companies related to the crypto field, the lowest capital cost globally right now is to acquire a listed company and transform it into a crypto treasury model. Therefore, we have seen players like Tether, SoftBank, Bitfinex, Nakamoto Holdings, and others join the fray. The Financial Times even reported that Trump Media & Technology Group is also getting involved—this is not surprising; honestly, it would be strange if it didn't join.

However, because of this, most of the listed companies participating in this game (besides MicroStrategy) are almost all small, semi-abandoned companies. Real companies like Apple, which have real businesses, cash flow, and operations, would certainly not get involved in this "weird operation to make stock prices soar" play.

For some entrepreneurs in the crypto space, the situation may be similar. We have reason to believe that Ethereum founder Vitalik Buterin is more concerned with optimizing the Ethereum protocol than thinking about how to package and sell ETH at a high price to stock investors. But for many, the valuation premium is simply too tempting to resist.

Third: How to cash out?

SharpLink "magically" created $2 billion in paper profits this morning. So what happens next?

Theoretically, this profit is generated by the investors participating in the private placement (like Consensys and its co-investment institutions). But the problem is, they likely cannot cash out immediately: typically, such private placement transactions have lock-up periods, and their shares need to be formally registered before they can be sold. Moreover, they collectively hold 97% of SharpLink's shares, and if they sell all, the stock price will surely crash.

In the entire year before the transaction was announced, SharpLink's average daily trading volume was only about 75,000 shares. Based on today's circulation, it would take more than three years to sell all their shares.

Although the stock market now values the $425 million worth of ETH they bought at $2.5 billion, they cannot "take out" this $2.5 billion. This paper profit is locked within the stock market valuation and cannot be extracted.

However, this is indeed a question worth studying. Modern finance seems to have found a way to steadily create billions in market value without much effort. While it’s not to say "anyone can do it in an hour," it is clear that many have discovered that the operational threshold is not high.

But if you cannot turn paper value into real cash, then it ultimately remains just a "show-off trick." Nominally, you have become a billionaire because you hold 97% of SharpLink Gaming's shares, but don't forget, just a week ago, this company's 100% valuation was only $2 million, and you would also worry about how long this bubble can last.

You would certainly think about "cashing out" a portion, but directly selling in the market does not seem like a viable path.

Of course, there are some "boring but realistic" answers: for example—"They now own a company valued at tens of billions with extremely low capital costs; they can continuously issue new shares to buy more Ethereum, thereby expanding their 'empire' and influence; when you control a company of such scale, you can pay yourself a high salary."

That sounds fine, but the problem is: these people originally had hundreds of millions in funds; they are not doing this to find a good job.

The real question is—how can they "cash out" that $2 billion?

I don't have a particularly good answer—if I did, I would probably be doing it. However, I want to point out that this question is very "crypto-like": it was originally a typical dilemma in the crypto industry, and now it has been brought into the stock market by a new generation of "crypto treasury companies."

This is a classic wealth story template in the crypto circle:

• You create some "magic beans" (Magic Beans)—like a new token—and then you hold most of them;
• There are not many beans actually traded in the market, but the transaction price is high, so the entire project's market cap looks very large;
• On the surface, you have become a billionaire, but once you really go to sell these beans, the market will crash, and you will get nothing;
• Holding "paper wealth" does bring some benefits, like prestige, resources, and superiority, but you also know that this "magic bean market" may not last long, so you are particularly eager to cash out.

The most famous case of this problem is probably the FTX collapse:

The FTX exchange and Alameda Research investment firm, controlled by Sam Bankman-Fried (SBF), appeared to be worth hundreds of billions on paper, but a large part of this valuation was supported by the crypto assets they created themselves. In November 2022, as the market lost confidence in FTX, these tokens quickly went to zero, and the company's valuation evaporated.

At that time, I wrote an article and quoted a conversation I had with SBF on a podcast. He mentioned a crypto token and the "box" model (Box Token) built around it:

"If everyone thinks the market value of this Box Token is around $1 billion, then it basically has that valuation. Everyone will account for it at that valuation. In fact, you can even use it for financing: pledge this token in a lending protocol to exchange for dollars. If you think its real value may not exceed two-thirds, you can also pledge part of it to take out money without paying it back—ultimately, you just get liquidated. In a sense, this is already something that can be monetized."

In the crypto circle, if you have a bunch of "magic beans" with a market valuation of $1 billion, perhaps someone would really be willing to lend you $500 million in "real cash," and these loans might even be non-recourse.

But in the stock market… even if you control a crypto treasury company whose market cap has risen by 100,000% and you hold 97% of the shares, it is very difficult to finance 50% or even 10% based on its paper market value.

But seriously, I would give it a try.

Applying Game Theory: Someone Really Cashed Out Successfully, and…

In the crypto world, there is another well-known case where someone successfully "monetized" a batch of "magic beans."

In October 2022, a trader who called himself an "Applied Game Theorist," Avi Eisenberg, applied game theory to a decentralized crypto perpetual contract exchange called Mango Markets and successfully executed a highly controversial arbitrage operation.

Mango Markets offers perpetual contract trading for various crypto assets, including futures for its own token MNGO. The contract prices are settled based on price oracles from several other crypto exchanges: your profit and loss on Mango depend on the price fluctuations of the corresponding spot assets on these external platforms.

Additionally, Mango allows users to borrow cryptocurrencies against the unrealized gains of their positions. For example, if you made $100 in a contract trade, the platform might let you pledge that unrealized gain to borrow $50 worth of cryptocurrency—and it's a non-recourse loan, meaning if you can't pay it back, you have no obligation to repay.

Eisenberg's operation was as follows:

1. He bought several million dollars' worth of long positions in MNGO perpetual contracts on Mango Markets;
2. At the same time, he opened an equal amount of short positions, resulting in a net position of zero (flat);
3. Then, he went to the corresponding "reference exchanges" and bought a large amount of MNGO spot;
4. Due to the low liquidity of MNGO, his buying significantly pushed up the market price of MNGO;
5. This caused the value of his long contract positions on Mango to rise rapidly;
6. He then used the "paper gains" from these long positions as collateral to borrow a large amount of cryptocurrency from Mango and withdrew it;
7. Then, he sold MNGO on the reference exchanges, driving down the spot price;
8. As a result, his short contract positions became more valuable;
9. He again used the unrealized gains from this short position as collateral to borrow more cryptocurrency from Mango.

Ultimately, according to official disclosures, Eisenberg borrowed and quickly withdrew over $100 million in crypto assets from Mango Markets.

In simple terms, this was almost like Eisenberg "stealing" $100 million from Mango Markets. He manipulated the price of MNGO, artificially inflated the value of his contract positions, and then used this inflated value as collateral to borrow a large amount of funds. Since these loans were non-recourse—in decentralized finance platforms, this is almost an industry norm—he had no obligation to repay.

Of course, he was eventually arrested.

We have discussed this case several times before, including:

• When he just completed the transaction;
• He subsequently posted a "Statement on Recent Events" on Twitter, explaining that he did indeed do this, but there was no problem because "all our actions were legitimate operations conducted in the open market, according to the design of the protocol, although the protocol development team may not have fully anticipated the consequences of setting these parameters";
• And when he was arrested, U.S. federal prosecutors clearly disagreed with his explanation.

Eisenberg was ultimately found guilty by a jury last April. However, last Friday, the judge overturned the conviction.

According to Bloomberg:

U.S. District Judge Arun Subramanian last Friday vacated Avraham Eisenberg's conviction for fraud and market manipulation, while also declaring him not guilty on a third charge. The judge ruled that the evidence presented during the trial was insufficient to support the jury's finding that Eisenberg made false statements to Mango Markets. And Mango Markets is a decentralized finance platform driven by smart contracts.

(This is the source of the original judgment opinion.)

This case exposes two key issues:

First, jurisdictional issues: Eisenberg was prosecuted in New York, but his so-called "applied game theory operation" took place in Puerto Rico, targeting some technically "borderless" crypto trading platforms.

The three reference exchanges he used to manipulate the MNGO price were:

• FTX, headquartered in the Bahamas;
• AscendEX, headquartered in Romania;
• Serum, a decentralized exchange that may not even have a headquarters.

And there is no evidence showing that the Mango Markets platform itself has any direct connection to New York.

There has always been a consensus that "if you commit a financial crime, it is likely to be linked to New York," so federal prosecutors in New York can almost reach globally. But this case indicates that cryptocurrency is approaching the limits of such jurisdictional boundaries.

In the crypto circle, there is a "stereotypical" belief: as long as you put things on the chain, you can evade the jurisdiction of various countries' laws. But the reality is not that simple.

For example, in Eisenberg's case: although he was sentenced in New York, he could theoretically still be prosecuted in Puerto Rico or even Romania. However, putting things on the blockchain could indeed allow you to escape the judicial reach of the U.S. Attorney's Office for the Southern District of New York (SDNY). In the crypto circle, this is already considered a fairly clever "operation."

In any case, this is the first key issue of the case: the reason Eisenberg's alleged "commodity manipulation charge" was overturned is that the prosecution chose the wrong venue for the trial. If the U.S. Department of Justice wishes, it could consider re-filing these charges in Puerto Rico.

But besides commodity manipulation, he was also convicted of wire fraud—which was also completely overturned by the judge, and the prosecution has no right to re-prosecute.

The second core issue involved here is: while Eisenberg's actions constituted market manipulation, whether they constituted "fraud" is actually not clear.

According to U.S. commodity law (which applies to crypto tokens including MNGO), as long as you use "any manipulative means" in derivative trading, you can be charged with commodity manipulation, and Eisenberg was charged for this reason. However, "wire fraud" is stricter; it requires the perpetrator to make false statements through a computer or communication system to gain financial benefits.

The court's ruling pointed out:

"To establish fraud, it must be proven that there was a material misrepresentation." The judge's conclusion was: no matter what Eisenberg did, he did not lie to anyone.

The government argued during the trial that Eisenberg's "fraud" was mainly reflected in two aspects (quoted from the judgment, omitting citations):

• He made Mango Markets believe he was applying for a legitimate cryptocurrency loan when he was actually trying to steal funds;
• He inflated the value of his collateral, leading the platform to believe it was valuable, while in reality, this value was artificially inflated and lacked actual support.

But none of these constitute lying.

Clicking the "borrow" button with no intention of repaying the loan may seem like fraud at first glance, but in the context of a crypto platform offering non-recourse loans, it does not hold.

Under the operating mechanism of such platforms, borrowers have no personal repayment obligation: the platform can only use the collateral as a means of recovery. If the value of the collateral falls below the loan amount, it is common practice to simply abandon the position. As the judge stated:

"If a user borrows funds, but the value of their collateral plummets, what happens? The system will liquidate it. There is no evidence showing that the 'borrow' function on Mango Markets implies that the user has an obligation to repay—even no other obligation—despite the fact that this term may have such implications in traditional contexts."

So, in other contexts, if someone deliberately conceals or distorts important information related to the terms or negotiations of a loan agreement when signing it, it may be considered fraud. But here, there are neither terms nor negotiation processes. There is only one word: "borrow."

Or to quote SBF: "You never have to pay back; you just get liquidated in the end."

As for "inflating the value of collateral," Eisenberg did not actually do this: Mango Markets calculates the value of his collateral based on market prices (which he manipulated).

Interestingly, this does not constitute fraud because an earlier case regarding LIBOR manipulation provided precedent support:

Of course, Eisenberg was well aware that the value of his portfolio was derived from market manipulation and knew that this valuation would not last long. So, although the valuation of his portfolio at the time of collateralized borrowing may technically be "accurate" (calculated based on the market price at that moment), the government believes that his statement about the value of the collateral was misleading. …

The government argued that when Eisenberg borrowed, he implicitly expressed two points to Mango Markets:

First, the value of the collateral in his account had not been manipulated;

Second, these collaterals were indeed valuable.

And both of these points, in the government's view, were false statements.

But this logic conflicts with the ruling of the U.S. Second Circuit Court of Appeals in United States v. Connolly.

In the Connolly case, Deutsche Bank (DB) reported daily to the British Bankers' Association (BBA) "DB's borrowing rates in the interbank market."

And the defendants—the traders at DB—sometimes requested LIBOR quote personnel to submit quotes favorable to their positions. Trial evidence showed that other DB employees and LIBOR quote personnel themselves admitted that "adjusting LIBOR quotes for the benefit of traders was, at the time, considered 'wrong.'"

But the court did not buy it. The court rejected the government's assertion—that these quotes implicitly confirmed "the quotes were not influenced by traders."

Even if market participants generally believe that traders interfering with LIBOR quotes is improper, the absence of explicit prohibitions or guidelines at that time is decisive. The court pointed out that although the BBA later did issue relevant prohibitory rules (just as Mango Markets updated its protocol after Eisenberg's operation), "there were no such rules or prohibitions in place during the early stages of the case."

We also discussed the Connolly case in 2022: LIBOR itself is a number "determined by guesswork," so Deutsche Bank's traders are unlikely to commit a crime for "getting this number wrong." It can now be seen that this has a similar logical analogy to the price of the MNGO token.

In summary, it should be emphasized that at least in terms of wire fraud, the platform's terms and conditions are indeed key. If Mango Markets had clearly told users: "If you want to borrow against your position, you must promise that you have not engaged in any market manipulation," then Eisenberg's transaction would constitute fraud. But it did not say this, nor did it say anything at all, so his actions do not constitute fraud.

Another typical creed in the crypto circle is: "Code is law": as long as a crypto system allows you to do something, you have the right to do it, even if the development team did not fully anticipate the consequences when setting parameters. Under this philosophy, traditional legal norms, background agreements, or user agreements are not important; the only thing that matters is what code is written in the system.

But the ruling in this case does not entirely mean that. Its actual implication is: code can become law. If you operate a crypto platform and tell users "please do not engage in manipulation, attacks, or other destructive behaviors," then when someone actually engages in manipulation, they may get into trouble. But if you operate a platform and say none of these things, just "this is how the platform operates, you figure it out," then even if someone finds a loophole in the system and manipulates it, that is legal, or at least does not constitute wire fraud.

This actually makes sense. I once wrote in an article discussing Eisenberg's operation: "You can imagine two different market systems and let users choose to join one": one called "Nice Market," with clear rules prohibiting manipulation and insider trading; the other called "Fun Market," where as long as you can find a way to profit, it's fair game, and the play is completely open. I even suggested that, given the relative lack of connection between crypto systems and real-world financial systems (although this situation is changing), it could serve as a testing ground for the "Fun Market," provided participation is entirely voluntary. This may be the little bit of "practical rules" conveyed by this case.

However, all of this does not help Eisenberg much. As Bloomberg pointed out, when he was arrested for this crypto case, U.S. law enforcement discovered that he had downloaded 1,274 images and videos of child pornography between 2017 and 2022, so he was sentenced to about four years this May for possession of child pornography.