Disclaimer: As a non-finance person who’s done a little work in crypto, I’ve come across a couple of items of concern that I think are worth sharing. Part of what we do as product managers is look around corners. We can’t help it. We try to understand what happens when systems interact, dependencies change, and something that looks beneficial in one place creates unintended consequences somewhere else. One of the things we learn in product management is that optimizing one part of a system doesn’t necessarily optimize the whole thing. In fact, sometimes making one product tremendously successful can create unintended consequences elsewhere. That’s especially interesting when the products themselves start becoming part of critical infrastructure.

What interests me here is that three worlds don’t always seem to get discussed together in general business or product management type circles. These are: government fiscal policy, traditional banking, and digital assets. There’s plenty of specialized research on these subjects, including their interactions, but I suspect most people don’t encounter it.

Most people still think of stablecoins as “crypto stuff” and U.S. debt as “government economics or politics.” But what happens when growing worldwide demand for digital dollars also creates demand for U.S. government debt? And what happens when that same process shifts funding away from traditional banks or creates new channels through which financial stress can spread? That’s what I want to explore. Not as a prediction of an impending crisis, but as a series of connections that I think deserve more attention, especially among those who create products or services in this area. So I’m going to try to untangle this.

What prompted me to write this was this post The Numbers Behind the Bond Bear Market (with a Look at Equities) from famed investor Ray Dalio. I’d read his book, Principles for Dealing with the Changing World Order: Why Nations Succeed and Fail, a couple of years ago, and it was disturbing. This article might be as well. Personally, I’m an optimist. However, some of what I’m going to present suggests some potential storm clouds. In any case, here we go.

Prerequisite Primer

If you’re a finance person or believe you remember your macroeconomics well, skip this section and head down to “Connecting the Dots”.

If you want a quick overview refresher, we’re going to go over a few macroeconomic literacy basics first. Just scroll down if you’re finding this part too much just general knowledge.

The U.S. government taxes, spends, and borrows money through decisions made by Congress and the executive branch. This is fiscal policy, with the Treasury Department managing the government’s financing and debt issuance. The Federal Reserve, meanwhile, conducts monetary policy. It influences interest rates, financial conditions, and the availability of money and credit, but it doesn’t directly control all the money in circulation. These institutions operate under different authorities and sometimes pursue objectives that create tension. One particularly concerning situation is called Fiscal Dominance: when government financing pressures become so great that they constrain the central bank’s ability to pursue its monetary objectives independently.

Treasury Bills, Notes, and Bonds: Three Ways to Borrow

When the government spends more money than it brings in through taxes, it borrows the difference by issuing Treasury debt. Think of this debt as a government IOU, split into three main categories:

Treasury Bills (T-Bills): Short-term government debt, generally issued with maturities of 4 to 52 weeks, with special cash-management bills available for other short durations.
Treasury Notes: Medium-term government debt, currently issued with maturities of 2, 3, 5, 7, or 10 years.
Treasury Bonds: Long-term government debt, currently issued with maturities of 20 or 30 years.

The Metaphor: Imagine financing a house. A 30-year Treasury bond resembles a long-term fixed-rate mortgage: the borrowing cost is locked in for decades. A Treasury note is more like a medium-term loan. A T-bill resembles a short-term loan that must be repaid or refinanced relatively soon. There’s nothing inherently wrong with using short-term debt. In fact, it can be cheaper, particularly when short-term interest rates are lower than long-term rates. Investors may demand a premium for lending over longer periods, so borrowing short term can reduce the government’s immediate financing costs. But that saving comes with a tradeoff: when the debt matures, the government must refinance it at whatever rates prevail. If rates fall, it may save even more. If they rise, the government can end up paying substantially more than if it had locked in a long-term rate. That’s a refinancing risk we’ll come back to.

Price and Yield: The “See-Saw” Relationship

For an existing fixed-rate bond, its market price and yield move in opposite directions. When its price rises, its yield falls; when its price falls, its yield rises. At a new Treasury auction, investors effectively determine the yield the government must offer to borrow money.

The Metaphor: Think of a see-saw on a playground. On one side is the Price of the bond; on the other side is the Yield (interest rate).

If a restaurant has a line around the block (high demand), it doesn’t need to offer discounts to get customers.

But if the tables are empty (weak demand), the restaurant has to put out a “50% Off / Free Appetizer” sign to get people through the door.

In the debt market, that “free appetizer” is a higher interest yield. When demand is weaker relative to the amount of debt being offered, investors generally demand a higher yield. That’s another way of saying the government must pay more to borrow.

Rollover Risk: The Refinancing Treadmill

When a government debt IOU matures, the government almost never pays it back using cash sitting in a bank account. Instead, it pays off old maturing debt by issuing new debt to new investors. This is called rolling over debt. The U.S. has a particular advantage because it borrows in its own currency, but that doesn’t mean Treasury can simply print whatever money it needs without economic consequences. And refinancing debt is different from creating new money.

Let’s keep in mind that the United States has two related advantages:

Monetary sovereignty: The U.S. issues debt denominated in a currency that its own central bank can create. This is not unique to the United States. Japan, the United Kingdom, and other countries that borrow in their own freely floating currencies have similar capabilities.
Dollar dominance: The U.S. dollar is the world’s principal reserve currency, widely used in international trade, banking, financial markets, and cross-border transactions. This gives the U.S. an unusually deep international market for its debt. This advantage is much more distinctive.

The second advantage is particularly relevant to this discussion because growing international demand for dollar-backed stablecoins could reinforce demand for dollar-denominated reserve assets, including Treasuries. Nevertheless, there’s still Rollover risk, which is the danger that when your old debt comes due, market conditions have worsened and you are forced to re-borrow that same money at much higher interest rates.

The Metaphor: Imagine taking out a credit card with a 0% introductory APR for 6 months. As long as you can keep transferring that balance to a new 0% card every 6 months, you’re fine. But what happens if, on month 6, no bank will offer you 0% anymore and the only card available charges 25% APR? You haven’t added any new spending, but your monthly interest bill just exploded. You probably shouldn’t have taken the offer. You should probably find another way. And you know it. And really, you maybe shouldn’t even be offered these deals. But it happens. It’s arguably predatory. That’s another topic.

The more heavily the U.S. relies on short-term debt, the more frequently it must refinance that portion of its borrowing at market rates. The government already rolls over enormous amounts of debt as a normal part of its operations. Our collective concern is what happens if short-term obligations become too large a share of its financing, particularly if interest rates rise or buyers become less willing to lend on favorable terms.

Another Example: We use heating oil at home. We just signed another heating oil price-protection contract for the coming year, with downward adjustment. We’re not oil traders. But we’re making a bet about an uncertain future. In our case, we’re spending about $400 to protect ourselves against potentially nasty price spikes over the coming year. Could we lose that bet? Maybe. But only in a small window. If oil prices fall just a little, but not more than our cost to lock in a capped contract that can still float down, we might have been better off buying at spot prices. But we’ve purchased something else as well. Predictability. We’ve limited our exposure to a spike risk that, right now, seems more consequential to us than potential savings.

Government debt management involves a similar tradeoff, although on a fantastically, enormously different scale. Borrowing short term may save money today, but it also means repeatedly exposing yourself to whatever rates the market demands tomorrow. Borrowing long term may cost more upfront, but it provides certainty about financing costs for years or decades.

Neither approach is automatically better. The question is how much refinancing risk you’re willing to carry, and what you’re getting in return.

Synthetic Buyers & Rehypothecation: The Digital “Shadow Engine”

To understand how stablecoins fit into this, we need to look at two addtional mechanisms:

Structural Buyers (Relatively Price-Insensitive Demand): Here’s an interesting kind of buyer. Normally, an investor decides whether to purchase an asset based on its expected return, price, risk, or some combination of those. But some investors buy certain assets because their business model requires them to hold them. Stablecoin issuers fit partly into this category. Companies like Tether and Circle issue digital tokens intended to maintain a one-dollar value. To support redemption, they hold reserves consisting of highly liquid assets, including short-term U.S. Treasury securities, cash, bank deposits, and other eligible instruments. So as demand for dollar-backed stablecoins grows, issuers generally need to acquire more reserve assets. This creates a source of demand for safe, short-term dollar assets, including Treasury bills, that is tied partly to demand for the stablecoins themselves rather than simply to investors deciding whether Treasury yields are attractive. That doesn’t make issuers completely indifferent to price or interest rates. They still make investment and liquidity-management decisions, and users can move in or out of stablecoins depending on market conditions. But here’s why I find the model so interesting: the issuer can earn income on reserve assets while the person holding the stablecoin typically receives no interest directly from the issuer. That’s potentially a tremendous business model at scale. (This is likely why so many want to issue stables. It’s like a gift card business that never expires. But there may be only room and marketplace trust for a few top players.) Of course, actual stability depends on redemption demand, the actual quality and liquidity of those reserves, operating expenses, and regulation. And that leads to another interesting question: what happens if people collectively want their dollars back?
Rehypothecation (Reusing Pledged Collateral): This is one of those financial concepts that has always bothered me a bit. And it’s worth distinguishing from ordinary lending and investing. Suppose I lend you $1,000 and you use it to buy a machine for your business. That’s lending. There’s a credit risk, of course, but the money is being put to work in a fairly understandable way. Now suppose you pledge securities as collateral for a loan, and the institution receiving that collateral is permitted to pledge those same securities again to secure a different obligation. That is rehypothecation. The original assets may now support a chain of financial transactions and obligations. This isn’t automatically fraudulent or even inappropriate. Collateral reuse is an established part of modern financial markets. But it can create dependencies and vulnerabilities that are difficult to appreciate by examining any individual transaction. This reminds me of aspects of the 2007–2008 Global Financial Crisis. That crisis was driven by a combination of poor mortgage underwriting, securitization, excessive leverage, fragile funding arrangements, and interconnected financial institutions. CDOs and collateral practices were parts of that larger structure, though CDOs themselves weren’t simply rehypothecation. My concern isn’t that stablecoins are recreating 2008. It’s that financial systems have a long history of building layers of obligations and leverage that appear manageable individually but become dangerous when confidence disappears and everyone needs liquidity at once.

The Analogy: Imagine you park your car with a valet. In a simplified stablecoin model, the issuer keeps enough qualifying assets available to cover the value of the tokens it has issued. Think of a valet keeping a car safely accounted for until its owner returns with the ticket. The analogy isn’t perfect, because the issuer may hold short-term investments rather than the original cash, but the basic obligation is to maintain redeemable backing.

Rehypothecation is like the valet taking your parked car out while you’re at dinner, renting it out as an Uber to make extra cash, and trying to park it back before you finish dessert. If nothing goes wrong, everyone makes money. But if the Uber gets into a crash before you leave the restaurant, tthe valet suddenly has a problem. One arrangement that looked safe in isolation has created a risk for someone who never agreed to participate in the second transaction.

I want to do one more metahphor, just because this is such an interesting topic. (To me anyway.) Imagine walking into a casino. You hand the cashier or table dealer $100 in cash, and they give you $100 in plastic chips. To play in the casino, you must use their chips. The cashier takes your real cash and locks it safely in the vault. This is how a basic stablecoin works. Your cash sits in reserve while you trade with digital tokens.

Now imagine the casino takes that $100 and buys a very short-term U.S. government security. It earns interest, but it still owes you $100 whenever you redeem your chips. That’s broadly analogous to how an issuer can earn income from stablecoin reserves. It isn’t, by itself, rehypothecation. Now a third party enters the picture. Someone offers to borrow your casino chips and pay you interest. They might use those chips as collateral for another loan, or lend them onward as part of a larger financial arrangement. Depending on how the transactions are structured, collateral may be pledged or reused multiple times. Suddenly we have different kinds of risk. The casino still owes the original value of its outstanding chips, while the third-party lender has created additional obligations involving those same chips. And if that lender fails, it doesn’t necessarily mean the casino itself has lost its reserves. But the people who lent their chips may have a very different problem. Now imagine a loss of confidence somewhere in that chain. Participants begin trying to unwind their positions, sell assets, repay loans, or redeem stablecoins. Some may discover that liquidity isn’t nearly as available as they assumed. That’s the part I find concerning. The risks aren’t necessarily located where the underlying money first entered the system. (Maybe my money. Or mom’s retirement money. Or a kid’s college fund.)

The risk often isn’t inside any one component; it’s in the handoffs between them, where one party’s reasonable optimization becomes another party’s unpriced dependency.

An important distinction: U.S. payment-stablecoin regulation under the GENIUS Act restricts issuers from rehypothecating required reserves, subject to specified exceptions. (Though enacted in 2025, it’s scheduled to become effective in January, 2027.) So I’m not suggesting that compliant issuers can freely lend out their backing assets. The potentially more complicated leverage and collateral risks can arise elsewhere, with exchanges, lending platforms, DeFi protocols, and other intermediaries using stablecoins after issuance. There’s a lot of folks out there offering yield of some sort on these assets via some means or another.

Some macroeconomic mechanics starting places:

Connecting the Dots

OK. With that out of the way, I’m going to try to connect some dots. This is where I’m heading further out of “my lane” and I’m really looking for some pushback or clarity. Think of the conclusions here as my assertions, not necessarily anything I’m claiming as absolute truth.

Dot 1: The Government’s Supply & Demand Problem

Traditionally, U.S. Treasury debt has attracted enormous demand from domestic and foreign investors, including commercial banks, pension funds, investment funds, private investors, and foreign central banks. Countries like Japan and China have been particularly important holders. But the picture is changing. China, for example, has substantially reduced its Treasury holdings over time. Other countries and private investors continue to buy, however, and aggregate foreign holdings haven’t simply collapsed. The more interesting concern may be how much debt the U.S. needs to issue, at what maturities, and what yields investors will demand to keep absorbing it. That distinction matters. A Treasury market can continue attracting plenty of buyers while the government still faces higher borrowing costs. That’s because investors don’t simply decide whether to buy government debt; they also decide what return they require for holding it. The U.S. might successfully sell every bond it offers, but at increasingly higher yields. And if the government is issuing more debt at the same time, those higher rates can become especially consequential. (See Treasury’s Major Foreign Holders of Treasury Securities.)

To adapt, the U.S. Treasury shifted its strategy. The Treasury finances government operations using a mix of bills, notes, and bonds. In periods of large deficits and changing market conditions, the mix of new issuance can shift, including greater reliance on short-term bills, which it seems to be doing. What I find interesting is that stablecoins are emerging as a source of demand concentrated in these very short-term instruments. They can potentially make that part of the market more attractive and easier to finance. Check out what the Federal Reserve Band of San Francisco has to say in Stablecoin Issuers’ Growing Appetite for Treasury Securities.

Wild Things

Want to know the really wild part? At least, the part that makes my head spin a bit? These are debts that might never be redeemed. Ever. Loans come due. Gift cards expire. Etc. etc. But a minted stablecoin for which an issuer is just holding funds in reserve? There’s a potentially powerful economic dynamic here: the stablecoin issuer has an ongoing redemption obligation, but no predetermined date when that obligation must be settled. As long as tokens remain outstanding, their backing can remain invested, although individual reserve securities will mature and need to be replaced.

But there’s a fascinating catch… Indefinite circulation doesn’t mean permanent funding. The issuer must remain prepared for redemptions, potentially on very short notice. Circle’s USDC terms, for example, provide qualifying customers with redemption rights and describe how the underlying reserves support those obligations. That’s an unusual arrangement. It’s a potentially long-lived source of reserve demand backed by liabilities that holders may want redeemed on very short notice. In normal times, it could create remarkably persistent demand for short-term government debt. But in a crisis, those same redemption rights could cause that demand to reverse abruptly.

So the question isn’t merely how much stablecoin issuers buy. It’s how stable that source of demand really is when conditions change. Now, I know what you might be thinking. Ordinary bank demand deposits also have no fixed maturity and may remain outstanding for decades while being withdrawable on demand. True enough. However, what makes stablecoins particularly interesting is the combination of ongoing circulation, usage, redemption-on-demand characteristics, and reserve portfolios heavily oriented toward short-term government securities.

Anyway… Back to our story…

A Quick Note on the Fed: Quantitative Easing (QE) involves the Federal Reserve purchasing securities, adding reserves to the banking system. Quantitative Tightening (QT) generally works in the opposite direction, reducing the Fed’s securities holdings, often by allowing assets to mature without fully reinvesting the proceeds. QT can increase the amount of Treasury debt that private-sector investors must absorb, although the actual market effects depend on Treasury issuance, the Fed’s operations, and other investors’ behavior. Stablecoin issuers are one increasingly significant source of demand in this broader market. To me, QE and QT still seem a bit like magic. The Fed can create or extinguish bank reserves electronically. Quite a neat trick. The creation and destruction of “That which is money” with just a few keystrokes on a computer. But I digress… the point is, stablecoin issuers are an interesting new sort of buyer.

Dot 2: Enter the “Structural Buyer” (Stablecoins)

The Treasury already has an enormous established market for short-term debt, including money-market funds, banks, and institutional investors. But another category of buyer is becoming increasingly interesting. Enter stablecoin issuers like Tether and Circle. The interesting thing about these buyers is that demand for their products can translate into additional demand for the assets they hold as reserves. In other words, someone buying digital dollars for use in crypto markets, international payments, or other applications may indirectly be supporting the market for U.S. government debt.

What are stablecoins again? Here’s a one sentence description: A stablecoin is a digital cryptocurrency whose value is pegged to a stable reserve asset; most commonly the U.S. dollar so we combine the speed and accessibility of blockchain networks with the price stability of traditional money. (It doesn’t have to be the U.S. dollar, but at least for now, it typically is.) If you want more in-depth on this, see Part 1: Stablecoin Flavors – What Are You Holding?

In a simplified example, when a customer buys, (or in this case mints), $1 of a dollar-backed stablecoin, the issuer receives $1 and assumes an obligation to redeem that token according to its terms. It must maintain appropriate backing, but it doesn’t necessarily buy a Treasury bill with that particular dollar. Its reserves may include cash, bank deposits, Treasury bills, qualifying repurchase agreements, and government money-market instruments, depending on the issuer and applicable rules. There’s another distinction worth making. Buying an existing stablecoin from someone else on an exchange doesn’t necessarily cause the issuer to receive new dollars or acquire new reserves. The relevant change occurs when the amount of stablecoins issued and outstanding increases. (I’ve written about how this happens in detail in a 2025 article: How Does Fiat Become Cryptocurrency?)

Stablecoin companies must hold ultra-safe, liquid assets to guarantee that users can redeem their digital tokens 1:1 for real dollars. As a result:

Stablecoin issuers collectively hold over $150–$200+ billion in U.S. Treasury debt. And this is likely to grow. This is just an opinion. But I think we’ll continue to see growth here as more folks move to crypto rails, even if mostly for institutional transfers. Now ok, ~$200 billion is maybe meaningful at the short end of the curve but small overall. Money market funds (~$8 trillion assets) absorb a lot of supply, and so on. Still, this has been growing fast and that’s the point here. It’s not about today. It’s about trajectory and trying to assess potential cascade type issues. And it’s not primarily about today’s size. It’s about whether a product category that started as a crypto settlement tool is wedging its way into becoming critical market infrastructure whose failure modes now touch traditional funding markets.
Stablecoin issuers now hold more U.S. short-term debt than entire sovereign nations like Germany or South Korea. (All since just 2014, although the real growth explosion is really just since 2020. See History of Stablecoins: From BitUSD to the $260 Billion Market.)
Stablecoin issuers are structural, relatively price-insensitive buyers of reserve assets. When new fiat-backed stablecoins are minted, issuers need corresponding backing regardless of whether Treasury yields happen to be attractive. They aren’t necessarily required to purchase T-bills specifically, but those securities are a major component of reserve portfolios. Consequently, growing stablecoin issuance can create additional Treasury demand driven by the issuer’s reserve obligations rather than by conventional investment decisions.conditions.

This Is No Longer Just a Theoretical Relationship

Here’s where this becomes especially interesting. Researchers at the Bank for International Settlements have studied the relationship between stablecoin demand and short-term Treasury yields. (See Stablecoins and safe asset prices.) In an updated study examining data from January 2021 through March 2026, they found that an inflow of approximately $3.5 billion into stablecoins reduced the yield on three-month Treasury bills by around 4 basis points within ten days. The effect was concentrated in short-term securities, with little spillover into longer maturities.

Four basis points might not sound like much. But in a market where governments borrow and refinance trillions of dollars, even small changes in financing costs can matter. We shouldn’t take that result and simply multiply it across the entire national debt, of course. Still, it demonstrates something important: demand for stablecoins is no longer just a crypto-market phenomenon. It’s measurable in conventional government debt markets. And there’s another wrinkle. The researchers found that the impact was stronger when Treasury-market intermediaries were under stress.

This is the sort of connection that interests me. A person deciding to hold digital dollars may have no interest whatsoever in U.S. fiscal policy. Yet collectively, those decisions can influence the pricing of government debt. As mentioned, I look at these things from a product manager’s perspective. So how seemingly separate things relate or cascade in what might not always be intuitively obvious ways is of interest.

A Product Perspective

Here’s another way to look at this from a product perspective. A stablecoin issuer wants adoption, liquidity, trust, and ultimately a sustainable business. The U.S. Treasury wants reliable financing at reasonable costs. Banks want deposits and lending relationships. These are three different sets of objectives, and nobody is necessarily managing the system as a whole. (Though ok, yes, that’s suppose to be the purview of regulations and laws and these are still nascent.)

Yet success for one participant can change the operating conditions for the others. It’s a fascinating example of how product-market fit, at sufficient scale, can produce consequences far outside the product’s intended market.

Dot 3: Are We Trading Cheaper Financing Today for Greater Dependence Tomorrow?

On the surface, this looks like a win-win. The world gets more access to digital dollars, stablecoin issuers earn income on their reserves, and the U.S. benefits from additional demand for government debt. This does concern other countries. As I’ve referenced before in my article on De-Dollarization Risks from Crypto and AI others are concerned with U.S. crypto mercantilism. (See Cryptomercantilism vs. Monetary Sovereignty-Dealing with the Challenge of US Stablecoins for the EU.) This is its own potentially massive subject of concern. It’s part of this story, but we’ll skip digressing into it for now.

But there are several complications worth considering…

First, the demand is concentrated at the short end. Stablecoin reserve portfolios tend to favor highly liquid, short-duration assets such as Treasury bills. This helps finance short-term government borrowing but does relatively little to directly support demand for 10-year or 30-year securities. The BIS research I mentioned earlier supports that distinction.

Second, more buyers don’t necessarily mean stronger underlying fiscal conditions. Additional stablecoin demand may help lower short-term borrowing costs, but it doesn’t reduce the government’s spending obligations or automatically solve its structural deficits. In fact, cheaper financing can sometimes reduce the immediate pressure to address longer-term problems.

Third, the composition of buyers matters. The Treasury has always depended on a diverse collection of investors, each with different motivations. Stablecoin issuers add another source of demand, but one increasingly connected to digital-asset adoption and redemption behavior. If that market grows substantially, its inflows and outflows could become more important to short-term Treasury conditions.

Fourth, short-term borrowing carries refinancing exposure. Treasury bills must be repaid or rolled over relatively frequently. If the government comes to rely more heavily on short-term financing, changes in interest rates can flow into its borrowing costs more quickly than they would with long-term fixed-rate debt.

I’m not saying stablecoins are hiding a collapse in traditional Treasury demand. I can’t support that claim. But they do raise a different question. Could a new and potentially volatile source of demand make short-term Treasury financing cheaper in normal times while introducing additional dependencies in periods of stress? I’d somewhat written about these issues before as well in Hidden Stablecoin Trade-Offs and we can maybe couple that with another concern about Will RWA Tokenization Growth Increase Systemic Risk?. Though I think the RWA risk is perhaps lesser as adoption has been slow for all manner of reasons.

One More Distinction: Buying Debt Isn’t Always Financing New Debt

There’s also a distinction worth making. When an issuer buys newly issued Treasury debt, it is directly participating in the financing of government borrowing. But when it buys an existing Treasury bill from another investor, that money goes to the investor selling the security, not directly to the Treasury. So how does that help the government? By increasing demand and liquidity in the Treasury market, secondary-market purchases can support prices and lower the yields investors require, which can improve the terms on which the government issues new debt.

That’s why I believe it’s fair to say stablecoin growth can help finance U.S. government borrowing, even though not every stablecoin purchase results in new cash going directly to the Treasury.

Dot 4: The New Systemic Risks

From a product-management perspective, there’s an important distinction here. We normally assess risks to our product. Such as outages, security breaches, liquidity problems, regulatory exposure, and customer losses. But as products become interconnected, there’s another question. What risks does our product introduce into systems that depend on it, or that it increasingly depends upon?

A stablecoin can work exactly as designed and still participate in a larger chain of financial dependencies that creates risks nobody intended. Conversely, a problem elsewhere in that chain can affect a perfectly well-managed stablecoin. That’s the sort of system-level risk that can be easy to overlook when everyone is focused on their own piece.

This arrangement isn’t just a quirk of accounting. It introduces genuine systemic risks across both digital and traditional finance:

The “Fire Sale” Threat: What happens if a major stablecoin experiences a crisis of confidence and large numbers of holders attempt to redeem their tokens? In principle, issuers maintain liquid reserves precisely to handle redemptions. Treasury bills are generally highly liquid, and some can simply mature and provide cash. Issuers may also hold cash or other assets that can be used for redemptions without selling securities. But if redemption demands become unusually large or rapid, an issuer may have to sell substantial amounts of reserve assets. And if several issuers or related institutions experience stress at the same time, those sales could put pressure on short-term Treasury markets. The risk isn’t that a stablecoin redemption automatically causes a Treasury crisis. It’s that a market originally developed for digital payments could become another channel through which sudden demands for liquidity propagate into the traditional financial system. In fact, the more stablecoins become integrated into global financial infrastructure, the more important it becomes to understand what happens when those flows reverse.
The “Shadow Bank” Loophole: Under the GENIUS Act, covered payment-stablecoin issuers are prohibited from paying holders interest or yield simply for holding their stablecoins. But the situation becomes more complicated once the tokens circulate outside the issuer. Exchanges, lending platforms, and DeFi protocols may offer opportunities to earn returns by lending stablecoins, supplying liquidity, or participating in other financial arrangements. Depending on the structure, these activities introduce credit, counterparty, liquidity, collateral, or smart-contract risks. And unlike an ordinary FDIC-insured bank deposit, a stablecoin balance or lending position generally doesn’t come with deposit insurance. This is where I think the shadow-banking comparison becomes more deeply interesting. The stablecoin issuer may be maintaining its reserves properly, while entirely separate institutions are constructing additional layers of lending and leverage around the tokens. The result could be a growing ecosystem of financial activity with some of the economic characteristics of banking, but different rules, protections, and failure mechanisms. There’s nothing automatically wrong with this. But we should be careful about assuming that because the underlying stablecoin is fully backed, every product built on top of it is equally safe.

And I wonder how many ordinary customers really appreciate these differences. A bank offering a negligible return on deposits isn’t terribly appealing when someone else advertises much higher yields on digital dollars. It’s not hard to imagine more consumers moving in that direction without fully appreciating where the additional return comes from or which protections they’ve surrendered.
Commercial Bank Depletion: Here’s another potential consequence. What happens if significant amounts of money move from traditional bank deposits into stablecoins? Part of the reason for laws restricting stablecoin yield are expressly to protect bank deposits. Banks rely heavily on deposits as a source of funding for loans and other assets. If customers shift money elsewhere, banks may face higher funding costs or need to change how they finance lending. But this isn’t as simple as saying every dollar entering a stablecoin disappears from the banking system. Depending on where issuers place their reserves and how Treasury securities are purchased, funds may reappear as deposits at other banks or move through other parts of the financial system. That might sound reassuring, but it raises an additional concern: the funding may not return to the same institutions, and it may not have the same characteristics. A relatively stable retail deposit at a community bank is not necessarily equivalent to a large institutional deposit at a money-center bank. A shift in funding composition could disadvantage smaller banks, increase their borrowing costs, or make it more difficult to support local lending. So the concern isn’t necessarily that stablecoins drain money out of the banking system altogether. It’s that they may change where the funding resides, how stable it is, and which institutions can access it. What’s holding people back here? It could be we’re just still in early adopter, early majority phases for this part of the market. And many may justifiably be making choices of apparent safety and security over the potential for yield. We’ll see how it goes.

The practical question for product teams is whether a stablecoin offering is mainly competing for payment utility or for deposit-like balances. And whether the funding that leaves a community bank reappears somewhere that still supports the same real-economy lending.
Concentration Risk: A New Kind of Financial Dependency. There’s one more dimension to this that concerns me, and that’s concentration. Much of the stablecoin market is dominated by a relatively small number of issuers, with Tether and Circle accounting for a substantial share of dollar-backed stablecoin activity. Concentration isn’t inherently bad. In fact, large issuers may benefit from scale, professional reserve management, and stronger operational infrastructure. But think about what happens as these companies become increasingly important buyers of short-term government debt, providers of digital-dollar liquidity, and infrastructure for other financial products. A problem at one institution might no longer be contained within the crypto ecosystem. It could affect exchanges, payment companies, counterparties, and traditional financial markets. There is a paradox here. The more successful stablecoins become at integrating with the existing financial system, the less isolated their risks become. This isn’t unique to stablecoins. Financial history is full of examples where concentrated intermediaries provided tremendous efficiency and liquidity right up until their problems became everyone else’s problems. Back in 2008, when Lehman Brothers failed, I was working in a building with a good view of the downtown NYC financial area. I remember the day it happened and discussing it with some of my colleagues. We were surprised government didn’t intervene sooner. I don’t know this. I’m not sure if anyone does. But maybe government folks were thinking, “ok, well, it’s just one.” But it quickly became an “oh s$#%” moment and scramble to contain contagion. There’s the old joke about “If you owe the bank $100,000 and can’t pay, you have a problem. If you owe the bank $1 billion, the bank has a problem.” If some as yet unknown issue crops up with one of these major players, is it potentially systemic?

I’d somewhat written about these issues before as well in Hidden Stablecoin Trade-Offs and we can maybe couple that with another concern about Will RWA Tokenization Growth Increase Systemic Risk?. Though I think the RWA risk is perhaps lesser as adoption has been slow for all manner of reasons.

In product terms, the issuer’s reserve design and the third-party products built on top of the tokens are separate surfaces with different failure modes. Users may experience them as one thing.

But Isn’t This Also a Good Thing?

Before concluding, I think there’s a reasonable argument on the other side. Stablecoins may be genuinely useful financial infrastructure. They make dollar-denominated value easier to transfer across some borders and networks. They can facilitate payments and settlement, and they create another way for people around the world to access digital dollars.

From the U.S. government’s perspective, there’s also something potentially attractive about growing global demand for dollar-backed assets. If stablecoin issuers buy more Treasury bills, that can support demand for government debt and potentially lower borrowing costs. And this isn’t entirely speculative. BIS researchers have found measurable effects on short-term Treasury yields.

So perhaps we should welcome all of this. A growing market for digital dollars could strengthen the dollar’s international role while providing new financing demand for U.S. government debt. But that brings me back to the original concern. What looks beneficial when examining the government, banks, or digital assets individually might look more complicated when examining all three together. While researching this article, I found a June 2026 BIS paper that modeled exactly this kind of tradeoff. (See The macroeconomics of stablecoins.) The researchers identified two competing economic effects: a fiscal benefit from greater Treasury demand and a potential contraction in lending as stablecoins compete with bank deposits.

In their model, the lending effect slightly outweighed the fiscal benefit over the long run. But the outcome depended heavily on regulatory design, foreign demand, and other assumptions. It wasn’t a prediction that stablecoins would inevitably damage the economy. That’s an important distinction. I’m not arguing that stablecoins are inherently bad. I’m suggesting that their benefits and risks may arise in different parts of the financial system, at different times, and for different participants.

And perhaps that makes the complete tradeoff harder to recognize.

The Bottom Line

Stablecoins are no longer just crypto vehicles. They’re becoming increasingly connected to the infrastructure of traditional finance, including the U.S. Treasury market. And there lies an interesting tension. Global demand for digital dollars can create additional demand for U.S. government debt, potentially lowering short-term borrowing costs. That sounds like a good thing, (if you’re the U.S. anyway), especially at a time when the United States faces substantial deficits and refinancing requirements.

But the same development may also redirect funding among banks, create new concentrations of financial power, and introduce additional channels through which liquidity problems in digital markets can affect traditional ones. These aren’t necessarily reasons to reject stablecoins. And I’m certainly not suggesting that they’re about to bring down the Treasury market. What concerns me is the growing interdependence of these systems, particularly when the benefits of a financial arrangement appear in one place while its risks accumulate somewhere else.

We’ve seen versions of this before. Not necessarily with stablecoins or even with the same financial mechanisms, but with innovations that make individual transactions easier or cheaper while creating dependencies that become apparent only under stress. Technology doesn’t necessarily eliminate financial risk. Sometimes it moves the risk, changes who holds it, or makes it harder to recognize.

So here’s my question: Are stablecoins primarily strengthening the global dollar system, or could they also be creating a new layer of shadow-banking dependencies that we don’t yet fully understand?

I suspect the answer is some of both. What I’m less certain about is whether we’re giving enough attention to how the pieces fit together. For those of us that might be crafting products at any place along this value chain, it’s probably worth thinking about the larger picture and if there’s any guardrails we might want to consider.

Research References

BIS — Stablecoins and Safe Asset Prices (2025, subsequently updated). Evidence on stablecoin flows and Treasury yields.
BIS — The Macroeconomics of Stablecoins (June 2026). Modeling fiscal benefits and bank-lending tradeoffs.
BIS — Are Stablecoins and Bank Deposits Substitutes? (October 2026). Evidence of competition between stablecoins and deposits.
GENIUS Act — Public Law 119-27 (July 2025). Legal framework for reserve backing, eligible investments, and issuer restrictions.

Digital Dollars, Real Debt: Hidden Connections Between Stablecoins, Treasuries, and Banks