Bloomberg Law
Free Newsletter Sign Up
Bloomberg Law
Free Newsletter Sign Up

Matt Levine’s Money Stuff: Who Can Resist the Crypto Boom? (Correct)

Jan. 11, 2022, 9:28 PM

Citadel Cryptosecurities

I guess the most straightforward interpretation of the news that Paradigm, a crypto venture capital firm, is investing in Citadel Securities, a market-making firm that does not trade crypto, is that Citadel Securities is going to start trading crypto? Like this seems unlikely:

Paradigm: We are a leading investor in crypto trading firms.

Citadel Securities: Well don’t look at us, we hate crypto, never traded it, never will.

Paradigm: I suppose we could use some diversification.

Maybe, who knows, but my guess is the obvious one: The crypto boom is creating enormous gushers of wealth for financial intermediaries, Citadel Securities is a skilled financial intermediary that sits atop various long-running gushers of wealth (stock trading, etc.) that are starting to look just the littlest bit depleted, and it is eyeing all those fresh bubbly crypto gushers with interest. It takes money to build out a crypto business, but fortunately it is also extremely easy to raise money to build a crypto business, so here we are, here we are.

Here is the press release about the deal, in which Sequoia Capital and Paradigm will invest a total of $1.15 billion in Citadel Securities at a valuation of about $22 billion. “Firm Will Continue to Leverage its Superior Technology to Accelerate Global Growth Across Products, Markets and Regions,” it says, and:

“Citadel Securities has developed software and algorithms that have driven substantial improvement to market structures for the benefit of institutional and retail investors everywhere,” said Matt Huang, Co-Founder and Managing Partner of Paradigm. “We look forward to partnering with the Citadel Securities team as they extend their technology and expertise to even more markets and asset classes, including crypto.”

I sometimes come across claims that crypto will somehow disintermediate finance, that the decentralization of crypto puts the power, and all of the economic benefits, of crypto in the hands of ordinary people and gets rid of the need for big middlemen like Citadel Securities. These claims are crazy. Look around! There are endless profiles of people who have become billionaires by starting crypto exchanges, trading platforms, market makers, derivatives businesses, etc. (Meanwhile I have never read a profile of someone who became a billionaire by using crypto to solve any problem other than trading more crypto but never mind!) Here’s one from two days ago about Changpeng Zhao of Binance, who’s worth $96 billion. Here’s a Forbes profile of Sam Bankman-Fried of FTX, “the world’s richest 29-year-old,” worth $15.2 billion (per Bloomberg). Here’s a profile of Coinbase CEO Brian Armstrong’s $133 million house; Bloomberg gives his net worth as $8.7 billion. Armstrong’s Coinbase co-founder Fred Ehrsam ($3 billion) is one of Paradigm’s founders. Here’s a profile of the two founders of OpenSea, the non-fungible-token trading platform, worth $2.2 billion each.

Meanwhile Ken Griffin, the founder of Citadel Securities (and also of Citadel, the associated giant hedge fund) is worth $21.1 billion, which is pretty good, but he’s been doing it a lot longer than these other guys. Being a stock-trading middleman was a good way to make a fortune, back when stock trading was dominated by big inefficient banks and there was lots of room for new competition. Now it’s, you know, fine, but it’s efficient and grueling and you fight for every basis point and also you get to be the subject of congressional hearings and weird conspiracy theories.

Meanwhile in crypto! Meanwhile in crypto people will pay millions of dollars for a JPEG of an ape and then hand over the private key to anyone who asks nicely! Meanwhile in crypto you can make huge profits by being the robot counterparty to fat-finger trades, and no exchange will break those trades! Meanwhile in crypto the transaction fee for a $75 transaction is $75! Meanwhile in crypto the take rate forexchanges is 0.57% of their transaction volume, and Robinhood Markets Inc. collects more payment for order flow for crypto orders than it does for stock orders. The sidewalks in crypto are carpeted in a layer of $100 bills three inches deep! What kind of lunatic would run a high-tech global multi-asset automated trading firm and not get into crypto?

“Ken Griffin” is I suppose a partial answer to that question, but just wait:

Paradigm is focused on crypto and Web3, a reimagining of the internet, areas Citadel Securities is likely to incorporate in the future as they become more regulated. To date, Mr. Griffin has been a crypto skeptic and avoided trading digital currencies in his businesses.

I look forward to the conspiracy theories when Citadel Securities starts making markets (and paying for order flow) in crypto. Sure it was funny when Ken Griffin bought that copy of the Constitution that ConstitutionDAO wanted to buy, but how much funnier would it have been if he also made money on every ConstitutionDAO trade?

I wrote last year about traditional-finance traders jumping into crypto, and I guess it bears repeating:

Obviously one aspect of this story is that if you are a futures trader or electronic market maker or derivatives structurer in traditional finance, you work very hard to scratch out a few basis points in a competitive industry. If you move to the much shaggier crypto industry, there is less competition — in the relevant senses of well-capitalized players with good technology and access to leverage and pricing models — and you can potentially make a lot more money a lot more easily. If you work on Wall Street, you probably like money, so moving to a place that gives you more money is pretty tempting.

But another aspect of it is that if you work on Wall Street — particularly as a derivatives structurer or high-frequency trading developer or securities lawyer — you actually like this stuff. You are there because you are fascinated by structure, because you enjoy puzzling out the technical details of these mechanisms and trying to find new ways to push on them. In traditional finance this is all sort of saturated and competitive: Not only is it hard to get rich by coming up with a new derivative structure or trading algorithm, but it’s hard to even get the intellectual satisfaction of doing that, because so much of what is possible has already been done, and innovation tends to be technical and incremental. Whereas in crypto it can feel like more of a blank slate, or at least, like a different foundation where it takes ingenuity to re-erect the old structures.

The crypto boom is a wild gold rush for money, but it is also in important ways an intellectual gold rush. If — like Citadel Securities — you have spent the last few decades building the intellectual and technological apparatus to trade financial assets quickly and efficiently across multiple venues in a complex global environment, getting handed a bunch of brand-new assets and venues and environments has to be a fun and exciting challenge, a way to flex your muscles and expand your imagination and do more creative thinking about the puzzles you love. The fact that it is comically lucrative is also nice!

Tax structuring

A basic rule of U.S. income tax is that a business pays tax on its net income, but a person pays tax on her gross income. A business takes its revenue, subtracts its expenses, and pays tax on the difference; business expenses are tax-deductible. A person takes her paycheck and pays taxes on that; personal expenses are not tax-deductible.

That is just the basic rule, though, and there are many exceptions. For instance there is a long list of personal expenses — mortgage interest and charitable donations are some big ones — that are more or less tax-deductible; you subtract them from your income to arrive at your taxable income. For a long time, state and local taxes (“SALT”) were on that list; you could deduct them from your income in computing your federal tax liability. In 2017, Congress changed that, capping the SALT deduction at $10,000. This is bad for high-income taxpayers in high-tax states, like, for instance, law-firm partners in New York, who now have to pay federal taxes on a larger gross income and so pay more in taxes. It is also arguably bad for those high-tax states: Income there is now taxed twice, as it were, so at some margin those states will face pressure to lower taxes and reduce services, or high-income people might move away to lower-tax states.

And so there is an incentive for tax structuring. The states want to continue collecting taxes from the high-income taxpayers, but both the states and the taxpayers would prefer for those taxes to remain deductible for federal tax purposes. The taxpayers say to the states “boy we’d like to pay you these taxes but state taxes are no longer deductible on our personal tax returns, is there something you can do?” And there is!

From my brief conceptual overview there are two obvious paths to explore. One is: Call state and local taxes something else, something that remains deductible. The most straightforward move here is to call them “charitable deductions.” Something like:

  1. The state charges income taxes.

  • 2. The state sets up a charity to fund state government, and you can donate to the charity.

  • 3. Any donations to the state charity fund get a state tax credit (not a deduction): If you donate $20,000 to the charity, your state tax payable (not your state taxable income) goes down by $20,000.

  • 4. Instead of paying $20,000 of state taxes, you donate $20,000 to the fund and pay zero in state taxes.

  • 5. You have a $20,000 charitable deduction for your federal taxes.

That’s a good one, we discussed it a couple of times in 2017 and 2018, and people (that is, states) definitely tried it. It did not really work, though; the Internal Revenue Service concluded that you don’t get the charitable deduction for that $20,000, because you got something of value (the tax credit) for your donation.

The other path is a bit subtler : Call the state and local taxes “state and local taxes,” but instead of making people pay them (people’s expenses are not generally deductible from their federal income taxes), make businesses pay them (state and local taxes, like other business expenses, remain deductible for businesses). It’s a bit tricky doing that in the general case, but here’s a Wall Street Journal article about one version that apparently works:

New York business owners are saving billions of dollars in taxes by using a state-approved system for sidestepping the $10,000 federal cap on state and local tax deductions.

Owners of closely held businesses paid the state $11 billion in pass-through entity taxes by the end of 2021, according to the state’s Department of Tax and Finance. Pass-through businesses usually don’t pay income taxes directly but pass income and deductions through to their owners’ individual returns.

By paying New York’s new pass-through entity tax, those business owners shifted their state income taxes from their individual tax returns—where the cap would pinch them—to their business filings, where the cap doesn’t affect them.

The total represents about one-sixth of New York’s projected personal income tax revenue for the fiscal year ending March 31. The result: If those business owners had an average federal tax rate of 32%, they would be saving more than $3 billion. …

Some of New York’s largest law, financial and accounting firms routed their partners’ income through the state’s workaround, which took effect last year. Nearly 96,000 filers used the program, the tax department said.

“That’s a big number,” said James Wetzler, a former New York tax commissioner. The take-up is much greater than other workarounds, he said, including a program that lets employers pay federally deductible payroll taxes to help their employees have lower income taxes. That program had 328 users in 2021, according to the state. The Trump administration nixed another workaround that relied on people making charitable contributions in lieu of taxes.

Basically if you are a partner at a law firm and your share of the firm’s profits is $1 million, you have personal income of $1 million: The law firm is a “pass-through entity” for tax purposes, and does not normally pay taxes itself; the partners treat its income as their income and pay taxes on it. But under the new New York system, the partnership can pay your share of the state income taxes, which is a business expense for the partnership and so reduces your income. If your state income tax would be $100,000, the partnership pays it, and you have $900,000 of taxable income. You’ve paid the state taxes out of the business’s income, before it became income to you, so you effectively get to deduct it from your federal taxes:

Details vary by state, but here’s generally how the workarounds operate. States impose taxes—often optional—on pass-through entities such as partnerships and S corporations, a tax designation for certain closely held businesses. Those taxes are paid and get subtracted before income flows to business owners, effectively creating an unlimited deduction.

The laws use tax credits or other mechanisms to absolve owners of individual income-tax liabilities from business income. Thus, they satisfy state income-tax obligations without generating income-tax deductions subject to the federal cap.

Good trade! In general I think of tax structuring as having two competing sides:

  1. Governments, who try to write and enforce tax rules in ways that are hard to game, and

  • 2. Sophisticated taxpayers, who try to game the rules anyway.

And there are various things that you can say about that competition, and about which side pays its advisers more, etc., but the point is here that you have tax-code writers and taxpayers teaming up to minimize the taxpayers’ other taxes. A nice little synergy.

Private markets are the new public markets

Some companies are public, meaning that they publish a lot of detailed financial and business information for everyone to read, and in exchange they are allowed to raise unlimited money from regular retail investors and mutual funds and so forth. Other companies are private, meaning they don’t have to publish much information, but are legally limited to raising money mostly from “accredited investors,” basically rich people and institutions.

Companies start out being private. Historically one reason that companies switched from being private to being public is that they wanted to: Being public is annoying, what with all of the detailed financial reports and shareholder lawsuits, but there’s more money in public markets than in private ones, so if you wanted to raise a lot of money you’d go public. Another reason that companies switched from being private to being public, though, is that they had to: U.S. securities laws had a rule saying that a company with more than 500 shareholders has to register its securities publicly, file reports, etc. The theory was basically, you know, if you have 500 shareholders, you are not a closely held family company or anything; you are pretty public, so you might as well send audited financial statements etc. to all those shareholders. And so if you were a private company and you got big enough and lasted long enough, your stock would sort of leak out to more and more people: You’d hire more employees and give them stock, your early investors would sell down their stock as its value went up, etc., until you ended up with more than 500 holders and had to be public. (This is, for instance, a popular explanation for why Facebook Inc. went public when it did.)

This is all old-timey stuff. In recent years, as I have often said around here, “private markets are the new public markets”: Big private companies can raise tons of capital quickly from private investors (venture capital funds, rich individuals), or from historically public-market investors (mutual funds, hedge funds) who now buy private shares, so there is less need to go public to get money. On the other hand, the 500-holder rule has also gone away. In the JOBS Act, Congress raised the threshold to 2,000 holders and excluded employee-compensation shares from the count. So now it is pretty improbable that a company would be forced to go public because it has gotten too big.

This annoys people. There are probably some positive externalities from big companies being public. Their shareholders can read their financial statements and dial in to their earnings calls, but so can journalists and activists and regulators and competitors and customers. If you want to know stuff about big companies, or if you want to be able to put pressure on them to do stuff, it is more convenient for you for those companies to be public.

Also, if you want to be able to invest in good companies, it is more convenient for you if more big good companies are public. Most people can’t invest in whatever private company they want. For one thing, private-company investments are legally limited to mostly accredited investors. More importantly, private companies mostly get to pick their investors; even if you’re a billionaire professional venture capitalist there’s no guarantee that you can buy shares in the private company you really want to own. Whereas pretty much anyone can buy shares of any public company. If you care about letting ordinary investors build wealth, giving them access to more companies is good, and making those companies be public and file financial statements and not do fraud is also good.

On the other hand you can’t force companies to go public if they don’t want to. Except that the U.S. kind of did, until the JOBS Act? So you could always do that again? Anyway:

The Securities and Exchange Commission is preparing to force more transparency from big private companies, as regulators grow concerned about the lack of oversight of the private fundraising that has fueled their rise. …

The SEC, Wall Street’s top regulator, has begun work on a plan to require more private companies to routinely disclose information about their finances and operations, according to a semiannual rule-making agenda and people familiar with the matter. It is also considering tightening the qualifications that investors must meet to access private markets, and increasing the amount of information that some nonpublic companies must file with the agency.

“When they’re big firms, they can have a huge impact on thousands of people’s lives with absolutely no visibility for investors, employees and their unions, regulators, or the public,” said Democratic SEC Commissioner Allison Lee, who has called for the change. “I’m not interested in forcing medium- and small-sized companies into the reporting regime.” ...

Federal statute requires companies with more than 2,000 shareholders “of record” to register their securities with the SEC and periodically disclose key information, whether or not they have conducted an IPO. But that threshold is rarely crossed because SEC rules allow an unlimited number of people to own shares in “street name”—through the same broker-dealer or investment vehicle—and be counted as one shareholder.

“The SEC has been intentionally undercounting shareholders for decades,” said Tyler Gellasch, executive director of the Healthy Markets Association, an investor trade group.

Now, the agency is working on a proposal that would enable regulators to look under the hood of such entities for a more complete shareholder tally. Its goal is to push large, private companies into the same disclosure regime that their publicly traded counterparts face. ...

One option would be to count each investor in a feeder fund—a vehicle set up by broker-dealers to pool assets from multiple clients for private-market deals—as distinct shareholders of record. A more aggressive path would be to count the number of people investing through a venture-capital fund or private-equity fund.

I don’t know how much this really matters — 2,000 is still a large number — but I suppose it could. (If every beneficiary of a state retirement system counts as a “shareholder” through a venture capital fund, then taking any VC investment will require companies to go public, which would be a weird result; I assume that’s not the plan.)

I have said occasionally that the SEC is sort of the all-purpose meta-regulator of the U.S., since public companies are so important and so much of what they do can be regulated through the lens of securities law. But this fits uneasily with private markets being the new public markets: The SEC does have power over private companies that sell securities, but that power is a lot more limited than it is with public companies. (Like, sending fake financial statements to investors will get a private company in trouble for securities fraud, whereas basically any business or political decision by a public company could raise securities-law issues.) The more the important companies are private, the less important the SEC is. You can see why the SEC might have incentives to fix that.

Web3 Stuff

Yesterday I wrote about Web3 and I wanted to follow up with a few things here. First I should link to two other excellent newsletters that wrote about similar Web3 issues, Can Duruk at the Margins and Ben Thompson at Stratechery last week and again yesterday.

Second, in writing about the practical centralization of Web3, I discussed ExcelCoin, my pretend cryptocurrency that I keep on a list in Microsoft Excel. “If you want to buy some of my quadrillion ExcelCoins, send me an email,” I wrote, like an idiot; “I will make a market at $0.99 bid / $1.01 offered.” A footnote to that sentence said “don’t actually” send me an email, but who reads footnotes. Anyway I got a number of emails asking to buy ExcelCoins. The number is large enough that it was not convenient for me to respond to each email individually, sorry, and yet it was also small enough that I am not planning to quit my job to live off sales of ExcelCoin. I believe I have orders for 393.01 ExcelCoins, for which I could get about $396.94. In a sense that’s a pretty good rate for typing a few numbers into a spreadsheet, but I worry about scalability.

Also, in writing about ExcelCoin, I mentioned that the technology (me typing in Excel) could be extended to track non-fungible tokens, not just fungible coins. (One ExcelCoin bidder also asked for an ExcelNFT of “a picture of a rocket” though we did not talk price.) Apparently Roy Keyes had already implemented a very similar idea, called CSVChain; the way it works is that you send Keyes an email asking to buy one of the NFTs listed on the site, and if you agree on a trade he updates the CSV file listing the owners of the NFTs. “All transactions are appended to the CSV file using the echo command and the >> operator, making the CSVCHAIN immutable. The CSVCHAIN is held in cold storage on this USB drive to ensure its security.” This is a good joke! We had a little exchange on Twitter and now I seem to own a CSVChain NFT named Scorp Scorpy Scorp. I would include a picture of it here but I have no idea about its copyright status!

Finally, one technical point. Yesterday, after joking about ExcelCoin, I wrote:

On the other hand millions and millions of people own U.S. dollars, which means precisely that some bank has a list and there is an entry for them on that list. It’s probably not in Excel, but same basic idea: The bank has a list of dollars in its accounts, the list is not kept via any sort of consensus mechanism or blockchain or whatever; the bank just keeps the list using, uh, Cobol.

A reader pointed out that really there is a consensus mechanism, insofar as banks usually keep multiple copies of the list in different systems and have some mechanism for making sure they agree. This is about redundancy and resiliency, not about decentralization — the bank keeps all the lists — but, still: consensus mechanism. The actual dollar financial system is better run than ExcelCoin.

Things happen

Fed Vice Chairman Richard Clarida to Resign. Take-Two Interactive to Buy FarmVille Maker Zynga in $11 Billion Deal. Tiger Global Lost 7% Last Year, First Annual Drop Since 2016. AQR quant fund kicks off the year with 10% gain after 2021 rebound. Retirement Funds Bet Bigger on Private Equity. Shimao Plans to Repay Yuan Bond Due Saturday After Cash Inflow. China Developer R&F Delays Bond Maturity as Buyback Falls Short. Top Chinese Tutoring Company Laid Off 60,000 Workers After Crackdown. Eli Manning Throws Himself Into Private Equity as Partner at BVP. Cryptocurrency Traders Move Into Carbon Markets. OHM is (-3, -3). Baby Doge is up. ‘MasterChef’ blasted for allegedly serving endangered species.

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!

(Corrects charitable deduction math in second item.)

To contact the author of this story:
Matt Levine at

To contact the editor responsible for this story:
Brooke Sample at

  1. I have not seen anyone try to call them “mortgage interest” but I feel like that is promising too? The bulk of local taxes are usually property taxes, lots of people have mortgages, maybe you can impose the tax on the mortgagee and get it back in interest, etc.
  2. We talked about this one in 2017 too, in discussing a paper that described it as a “franchise tax” on pass-through entities.
  3. Like, if my employer paid my state taxes for me, I’d presumably have additional federal taxable income for that amount. To the extent a state charges personal income tax, letting a business pay the tax for its employees doesn’t really help. What you need is to recharacterize the tax as something other than personal income tax. Charging businesses payroll taxes (instead of charging individuals income taxes) is promising, but complicated. Charging *pass-through* businesses an income tax on income that would otherwise be passed through to their partners is promising and relatively straightforward.
  4. Here I am counting people who sent emails or tweets like “I’ll buy 20 at the offer, what’s your Venmo,” not people who sent emails or tweets like “how do I buy ExcelCoins,” or “you should sell them for real, you’d make a lot of money,” or “I’d like a billion.” Some people sent in bids with higher limits than my $1.01 offer; I count those as bids at $1.01 (i.e. I assumed that my offer would not move despite the demand). Some people seem to have bid $1.00, technically below my offer, but I counted those because, I mean, this is all fake. Some people were flexible on size; I used the maximum size that they specified. A few people offered to exchange ExcelCoins for various coins they had invented and kept on their own spreadsheets (or, in one case, in their own memory; “It’s very fast”); I did not count these unless they also offered dollars as an alternative. And, yes, one person emailed asking to buy 0.01 ExcelCoins, explicitly to pump up the market cap of ExcelCoin: If someone buys 0.01 for $0.01, then my remaining 999,999,999,999,999.99 ExcelCoins should be worth $999,999,999,999,999.99, making me the world’s richest person, technically. Oh, also, fun fact about that last sentence: I went into my Excel ledger of ExcelCoins, where I have written down that I own 1,000,000,000,000,000.000000 ExcelCoins, and I took that number and subtracted 0.01, and Excel told me that I’d have 1,000,000,000,000,000.000000 ExcelCoins left. My own math does not agree with Excel, whoops. It is possible that ExcelCoin is too big a project to keep in Excel.
  5. To be clear, I did not send him money for Scorp Scorpy Scorp, I did not ask for it, and —despite its listing on his site with a purported minimum price of $7,117 —I am not looking to sell it. I assume that he, and I, are joking about all of this, and that nothing of value has been exchanged. I do think that the combination of (1) huge valuations forephemeral zero-costnonsense and (2) air-dropping can create some odd issues of journalistic (and other) ethics, though. Are you sure that Scorp Scorpy Scorp *isn’t* worth $7,117? If you minted an NFT of, like, spinning neon text saying “BRIBE” and burned it to the chair of the Securities and Exchange Commission, is that a thing?

© 2022 Bloomberg L.P. All rights reserved. Used with permission.