Circle: A Milder Theory

Matt Taibbi wrote a long piece dissecting Circle which he concludes by saying:

If every crypto company will struggle this badly to answer basic questions like Where’s your money? or What’s your risk?, the storm hasn’t even started yet.

This is a pretty clear allusion to some entities he talks about in the article being insolvent, and some kind of 2008-esque broad solvency crisis afflicting the crypto industry in general and Circle in particular. There are quite a few other theories going around alleging misconduct and all manner of financial chicanery.

Here we offer a far milder theory of what is happening at Circle which does not allege or even suggest any real wrongdoing. It is, perhaps, an intimation some gaming of the system is going on but not in a way that feels outside the bounds of normal business conduct. The theory has 3 legs:

1. Much of the USD-to-USDC tokenization is economically meaningless
2. The Yield product is a small circular financing scheme at arbitrary rates
3. Pumping the SPAC IPO value is the only real profitable use case

We will take these in turn. Again there is no allegation of real wrongdoing and this is just a theory assembled from public info.

Meaningless Tokenization

Dollars in bank accounts are already held electronically. Most banks have some physical cash on hand — but the bulk of the funds have no physical form. In this Odd Lots interview the CEO of Silvergate Bank touts their “automated tokenization of bank deposits” functionality. This is simply bragging your database can synchronize records with another type of database. That is something — it is not nothing — but this is hardly groundbreaking technology.

IBM’s Information Management System offered a primitive version of this functionality in the 1960s and every modern database supports some form scripting that can achieve these results. We are talking about dozens of lines of code here. This is not a valuable service in and of itself and nobody will pay much for it. Similarly a bank can always offer internal transfers 24x7x365 — internal transfers concern only the bank’s own records.

But what if, instead, Circle was paying banks to tokenize their deposits and settle internal transactions with USDC? Economically this is meaningless — the bank could just use their own database — but it would generate a lot of USDC and some associated financial services employees to talk about how they use it. As a marketing expense this is not meaningless at all.

Circular Yield

The Yield product, for a time, offered returns far in excess of normal USD money-market funds. As Taibbi points out this is no longer the case. He also quotes a claim the program is on the order of a few hundred million dollars. If they have 50 billion in USDC outstanding and 300 million in Yield they can easily subsidize 5% or more return on the Yield product out of even a fraction of a percent on the main reserves. Or they could simply use investor money to subsidize the Yield product. They might even lend to another party and top-up the interest themselves. All manner of odd things could create this yield.

As a manner of banking-finance these make no sense and are just the sort of left-pocket/right-pocket action David Einhorn found at Allied Capital. But again, as a marketing expense, it may well be fine.

But why? Why would someone do this? Everyone knows the normal crypto KYC shell game. It’s a Hotel California setup where you can deposit easily but your funds are stuck once an endless cycle of KYC requests begin.

What if Circle isn’t playing Hotel California — they are playing Augusta National. Augusta Notional is a famous golf club that hosts the Masters golf tournament every year. It is a stunning, and stunningly well-kept, golf course with beauty that is the stuff of legend. It is also a golf club with members. But nobody seems to know exactly who they are, how one becomes a member, or much of anything else about it.

Circle could run Yield as a sort of Augusta National of subsidized rates and only invite investors and other related parties. Pitching someone “I want your funds to provide subsidized yield to the public to raise awareness of my product” does not sound quite as appealing as “We will use a portion of your investment to subsidize a deposit you place, handing back some of the investment in the form of a higher yield than a money market fund and then running ads to promote how much yield you received due to our ‘innovation’.”

If Circle only accepts deposits from people who are investing in the business anyway then the yield is just a marketing exercise. If I give you $100 as a straight equity investment that is straightforward. But if I give you $110 in equity alongside a $90 deposit that pays 10% interest? That’s still $100 in equity — but now I can run ads about how my super-efficient operation generated 10% yield without anybody loosing any money.

Pump It

Now you have to ask yourself: why would someone do this? Why does this make any sense? If you do these two things you can make two claims. First, you can claim a lot of traditional financial institutions are using your product and their executives will rave about it in the media. And second, you can point to your high deposit rates as evidence your “innovative financial products” are more efficient than dusty old TradFi. This might border on misrepresentation for a public company, or a company about to IPO, or really any entity with publicly traded securities.

But it is almost certainly fine for a company planning to list via SPAC merger! The lower standards applied to this process mean the company can tout it’s utility and efficiency without the sorts of detailed disclosures, including more stringent rules with respect to related-party dealing, that would be required in a normal IPO. Likely the real financial engineering is a bit more complex to provide cover — guaranteed returns associated with vesting schedules, or large fixed deposits or something. This is just a high level sketch of a theory.

In this scenario the holders of USDC are fine. There is no great financial tsunami about to destroy the stablecoin. The only group in the firing line here are the future buyers of the SPAC shares. Those folks — if the business is mainly an economically-meaningless illusion— will get hosed.

So?

We should emphasize that this is just speculation and may be completely incorrect. It is also most certainly not an allegation of any real wrongdoing! This is completely fine under the current rules. Even the SEC has publicly pointed out that the disclosure requirements for SPACs are far milder than traditional IPOs. We do have access to some financial statements for Circle and it is pretty clear their expenses exceed stablecoin reserve interest revenues by a lot:

And this theory fits the facts very well. Circle Yield was at one time offering incredibly attractive rates — but not many people got them. There is a lot of USDC outstanding — but it’s not used nearly as much as USDT or BUSD relative to their respective market caps. Banks talk up the utility of stablecoins including USDC for use cases that sound like existing database-driven products they’ve offered for decades.

But a lot of people, in finance and otherwise, find it hard to believe Circle’s management would be committing the sorts of outright fraud Taibbi alludes to in US-registered and managed entities. This theory fits within the many possible risks and market changes outlined in their disclosures. Further, that document makes clear that Yield represents a tiny fraction of Circle’s total liabilities. And as long as the Yield participants owned less than 5% of Circle’s equity they would not be mentioned in the related-party transaction section.

Rather, and this is perhaps the least surprising observation of all, maybe this tech startup is simply playing the usual tech startup games to pump their share price. Taibbi’s issues with clarity and documentation are surely real and reasonable points to raise. But suppose they are simply the sort of regulatory sloppiness that plagued AirBnB and Uber for years. Let’s not assert anyone is breaking the rules here — but let’s also not expect a group of tech executives to act like community bankers from the 1960s.

Quite a few EV companies came to market recently with, um, limited product offerings. But they had to spend money on engineers, factories, demonstrations and so on. In finance this is a lot easier — there is no real external cost at all. Circle would need to provide suitable financial incentives for a few institutions to write dozens of lines of code, give some interviews, and draft marginally-more-complex funding arrangements. That is likely not much money at all —and certainly nothing on the order of what Nikola spent to roll an electric truck chassis down a hill. That savings, however, may be the main innovation here.