BIS Research: How Stablecoin Reserves Move Treasury Yields

Stablecoins are no longer a niche corner of the crypto ecosystem. With total supply exceeding $300 billion in 2026, issuers like Tether and Circle have become significant participants in the US Treasury bill market. A BIS Working Paper (No. 1270), authored by Rashad Ahmed and Iñaki Aldasoro, puts empirical weight behind what market observers had suspected: dollar-backed stablecoin reserve flows detectably move short-term US Treasury yields. The implications extend beyond crypto into monetary policy transmission, financial stability, and the structural demand for government debt.

What the Paper Studies

The paper, first published in May 2025 and revised in June 2026, examines the relationship between stablecoin minting and redemption activity and the pricing of short-term US government securities. The authors use daily data from January 2021 through March 2025, a period that spans the post-COVID Treasury bill supply expansion, the 2022 crypto winter, multiple stablecoin depeg events, and the Federal Reserve's aggressive rate-hiking cycle.

The core question is straightforward: when stablecoin issuers mint new tokens (receiving dollars and buying T-bills) or redeem tokens (selling T-bills and returning dollars), does this activity leave a measurable footprint on Treasury yields?

The Transmission Mechanism

To understand why stablecoin flows would affect Treasury pricing, consider what happens mechanically when someone mints a fiat-backed stablecoin. A user sends US dollars to the issuer (Tether, Circle, or another entity). The issuer deposits those dollars and, per its reserve policy, purchases short-dated Treasury bills. This creates incremental buy-side demand for bills, which pushes prices up and yields down.

Redemption works in reverse. A holder returns stablecoins to the issuer, which burns the tokens and must liquidate reserve assets to return dollars. If those reserves are concentrated in T-bills, the issuer sells bills into the secondary market, pushing prices down and yields up.

Why bills specifically: Both Tether and Circle hold the vast majority of their reserves in Treasury bills with maturities under 90 days, not longer-dated notes or bonds. This concentration means stablecoin flows primarily affect the short end of the yield curve, with minimal spillover to 2-year, 5-year, or 10-year maturities. The BIS paper confirms this: the yield effect is statistically significant for 3-month bills but not for longer tenors.

A One-Directional Pipe

The mechanism is not bidirectional in equal measure. Minting new stablecoins is often a deliberate, planned process: an institutional counterparty wires dollars to the issuer, which then places orders for bills on a regular schedule. Redemptions, by contrast, can spike suddenly during periods of market stress, as holders rush to exit stablecoins for fiat. This asymmetry is central to the paper's findings.

Methodology: Isolating Crypto-Specific Shocks

The empirical challenge is establishing causation rather than mere correlation. Stablecoin flows and Treasury yields both respond to macroeconomic conditions: a Federal Reserve rate decision could simultaneously drive demand for stablecoins and repricing of T-bills, without one causing the other.

Ahmed and Aldasoro address this with instrumented local projection regressions. The identification strategy relies on crypto-specific shocks: events that affect stablecoin minting and redemption behavior but have no direct channel through which they would independently affect Treasury yields. Exchange hacks, protocol failures, and abrupt shifts in DeFi yields are examples of such instruments. These events drive stablecoin flows without constituting macroeconomic news that reprices government debt.

This approach allows the authors to estimate the causal effect of stablecoin flows on Treasury yields, not just their co-movement.

Key Findings

The paper reports three principal results.

Inflows Suppress Yields

A two-standard-deviation inflow into dollar-backed stablecoins (corresponding to a surge in minting activity) lowers 3-month Treasury bill yields by approximately 2 to 2.5 basis points within 10 days. The effect is concentrated at the short end: longer maturities show little or no statistically significant response.

Outflows Hit Harder

The impact is asymmetric. Stablecoin redemptions (outflows) raise Treasury yields by roughly two to three times as much as equivalent-sized inflows lower them. A two-standard-deviation outflow pushes 3-month bill yields up by approximately 5 to 7 basis points.

This asymmetry makes intuitive sense. Minting is typically spread over time as issuers accumulate reserves through scheduled purchases. Redemptions, especially during stress episodes, are concentrated and urgent: issuers must liquidate holdings quickly to meet outflows, creating sharp sell-side pressure on a narrow slice of the bill market.

Supply Conditions Matter

The yield impact is not constant. During periods of ample T-bill supply (for example, when the Treasury is actively issuing new bills), stablecoin flows have no statistically significant effect. But when bill supply is scarce, the estimated yield impact rises to 5 to 8 basis points. In other words, stablecoin flows matter most precisely when the market is already tight.

Issuer Decomposition

When the authors decompose the yield effect by stablecoin issuer, USDT (Tether) accounts for the largest share of the estimated impact, followed by USDC (Circle). This ranking tracks their relative market shares: Tether holds approximately 59% of stablecoin supply and Circle about 23 to 24%, together representing roughly 83% of the market.

How Large Are Stablecoin T-Bill Holdings?

To appreciate the scale of the mechanism the BIS paper identifies, consider the current size of stablecoin reserve portfolios relative to the broader Treasury market.

As of early 2026, stablecoin issuers collectively hold an estimated $120 billion to $165 billion in US Treasury bills, depending on the source and date. This represents roughly 1.7% to 2.5% of the approximately $6.2 trillion outstanding T-bill market. For comparison, money market funds hold approximately $2.5 trillion in T-bills, corresponding to about 40% of the total market.

Stablecoin issuers are small relative to money market funds, but they are growing fast. Standard Chartered projects that stablecoin market capitalization could reach $2 trillion by end of 2028, which would generate roughly $800 billion to $1 trillion in fresh T-bill demand. At that scale, stablecoin flows would rival money market funds as a structural force in the bill market.

Reserve Composition by Issuer

The concentration of reserves in T-bills is a deliberate choice shaped by risk management, regulatory pressure, and the GENIUS Act, which requires regulated stablecoin issuers to hold reserves in eligible assets including Treasury bills maturing within 93 days, Treasury-backed repurchase agreements, and cash at insured depositories.

Tether

Tether's Q4 2025 attestation, signed by BDO Italy, reported $186.5 billion in liabilities with $6.3 billion in excess reserves. US Treasuries make up approximately 80% of Tether's reserve portfolio as of Q1 2026, with the balance in overnight repo, cash, a modest allocation to gold (~$8 billion), and bitcoin (~$7 billion). Tether has never published a full annual audit by a Big Four accounting firm: its disclosures remain point-in-time attestations rather than comprehensive audits.

Circle

Circle's USDC reserves are managed by BlackRock and invested primarily in the Circle Reserve Fund, a SEC-registered government money market fund. As of April 2026, roughly 75 to 80% of USDC reserves were held in short-dated Treasury bills, with 15 to 20% in Treasury overnight repo and the remainder in insured bank deposits. Circle publishes a monthly, line-by-line reserve report with a weighted average maturity near 30 days.

Regulatory convergence: The GENIUS Act explicitly excludes commercial paper, corporate bonds, gold, and cryptocurrency from eligible reserve assets. This means any issuer seeking federal regulatory approval must concentrate reserves even further in T-bills and equivalent instruments, likely amplifying the transmission mechanism the BIS paper documents.

The Run-Risk Problem

The BIS paper's most concerning finding is the asymmetry: outflows hit yields harder than inflows suppress them. This has direct implications for financial stability. If a stablecoin run forces issuers to liquidate tens of billions of dollars in T-bills quickly, the resulting sell pressure could spike short-term yields at exactly the wrong moment.

The Money Market Fund Parallel

This dynamic is not hypothetical. It has played out twice in money market funds, which share a structural similarity with stablecoins: both promise on-demand, par-value redemption backed by a portfolio of short-term assets.

In September 2008, the Reserve Primary Fund held Lehman Brothers commercial paper representing just 1.2% of its total assets. When Lehman collapsed, the fund “broke the buck”, falling below the $1.00 NAV threshold. The fund lost two-thirds of its assets within 24 hours as investors rushed to redeem. The panic spread to other prime money market funds, and the run stopped only when the US Treasury issued an extraordinary blanket guarantee covering money market fund liabilities.

In March 2020, the COVID-19 market shock triggered a similar flight from prime money market funds. The Federal Reserve responded by establishing the Money Market Mutual Fund Liquidity Facility (MMLF), effectively backstopping the sector for the second time in twelve years.

Stablecoins Have No Backstop

Money market funds, after two systemic runs, now benefit from regulatory reforms (swing pricing, liquidity fees, gates) and the implicit expectation of Federal Reserve intervention. Stablecoins have none of these protections. There is no Federal Reserve facility for stablecoin issuers. There is no deposit insurance. There is no lender of last resort.

The May 2022 collapse of the algorithmic stablecoin UST, which destroyed approximately $45 billion in value, demonstrated that contagion risk in stablecoins is real. Research from the NBER found that blockchain transparency actually amplified the run: investors could monitor each other's redemptions in real time, accelerating the stampede. While UST was algorithmic rather than fiat-backed, the episode triggered significant outflows from USDT and USDC as well, showing that depeg fears can spread across stablecoin types regardless of reserve backing.

Systemic Implications at Scale

Today, stablecoin issuers hold roughly 2% of the T-bill market: a meaningful but manageable share. The BIS paper shows that even at this scale, flows leave a detectable footprint on yields. The question is what happens if stablecoins grow to 5% or 10% of the bill market.

Scenario: A Run on a $500 Billion Stablecoin Sector

Consider a future where stablecoins collectively hold $400 billion in T-bills, roughly 6% of a $6.5 trillion bill market. A stress event triggers rapid redemptions amounting to 15% of outstanding supply ($75 billion) over five days. Using the BIS paper's estimates, scaled linearly (an approximation, since the relationship may be nonlinear), the resulting bill sell-off could push 3-month yields up by 15 to 25 basis points in under two weeks.

For context, a 20-basis-point spike in T-bill yields would tighten short-term funding conditions across the financial system. Money market funds, repo markets, and commercial paper pricing all reference T-bill rates. A stablecoin-driven yield spike would propagate outward into the broader short-term funding complex, potentially triggering a feedback loop: higher yields make stablecoins less attractive relative to money market funds, driving further redemptions, which drive further bill sales, which push yields higher still.

The Procyclicality Problem

This procyclicality is the central concern raised by the BIS research. Stablecoin reserves are not passively held: they are actively created and destroyed in response to market conditions. During crypto bull markets, stablecoin minting surges, pouring demand into the bill market and suppressing yields. During downturns, redemptions spike, pulling demand away and pushing yields up. This makes stablecoins a procyclical force in the Treasury market, amplifying rather than dampening yield volatility.

Compare this with central bank reserve managers, who typically adjust their portfolios gradually and countercyclically. Or with pension funds and insurance companies, whose Treasury demand is driven by liability matching rather than market sentiment. Stablecoins behave more like hot-money flows: rapid in, rapid out, concentrated in a single maturity segment.

Implications for Monetary Policy

The Federal Reserve has separately acknowledged that stablecoin growth has implications for monetary policy implementation. If stablecoin-driven T-bill demand becomes large enough, it could affect the transmission of the federal funds rate to broader financial conditions. The Fed controls the overnight rate, but market rates at slightly longer maturities (1-month, 3-month bills) are shaped by supply and demand dynamics in which stablecoin issuers are an increasingly significant participant.

The BIS paper's finding that effects intensify during periods of bill scarcity is particularly relevant for quantitative tightening. As the Fed reduces its balance sheet, the private sector must absorb more government debt. If stablecoin flows create additional volatility in this absorption process, they could complicate the Fed's efforts to normalize monetary conditions smoothly.

What Regulators Are Doing

The GENIUS Act, signed into law in 2026, directly addresses reserve quality by requiring that regulated stablecoin issuers hold eligible assets including Treasury bills maturing within 93 days, Treasury-backed repurchase agreements, insured deposits, and qualifying money market fund shares. Commercial paper, corporate bonds, and crypto assets are excluded.

The intent is to ensure that reserves are liquid and low-risk. But as the BIS paper implicitly warns, concentrating all stablecoin reserves in the same narrow instrument (short T-bills) creates a crowded-trade risk. If every issuer holds the same asset and must sell it under the same stress conditions, the resulting market impact is amplified.

Several policy analysts have suggested that stablecoin issuers should have access to the Fed's reverse repo facility (RRP) or a dedicated liquidity window, analogous to what money market funds received after 2008 and 2020. Without such a backstop, the next stablecoin stress event could force issuers into fire sales of T-bills, with the yield consequences the BIS paper documents.

Implications for Stablecoin-Integrated Protocols

For protocols that integrate stablecoins, the BIS findings underscore the importance of understanding what backs the dollars flowing through the system. A stablecoin is only as stable as its reserve management and redemption infrastructure. If a run on stablecoins can move Treasury yields, then the reverse is also true: Treasury market dislocations can create stress on stablecoin pegs.

This is directly relevant for platforms like Spark, which integrates the USDB stablecoin via Flashnet and Brale. Understanding the reserve dynamics behind any stablecoin integration is not optional: it is a prerequisite for assessing the systemic risk profile of the broader payment stack. For a deeper analysis of how stablecoin reserves function at the operational level, see our research on stablecoin Treasury bill reserve mechanics.

For developers building on Spark or integrating stablecoins into payment flows, the Spark documentation provides technical details on stablecoin token issuance and transfer mechanics. The stablecoin run risk analysis offers additional context on how redemption dynamics affect integrated protocols.

What to Watch

The BIS paper opens several threads worth monitoring as the stablecoin sector grows.

• Stablecoin supply growth relative to T-bill issuance: as the ratio increases, the yield sensitivity the paper documents will likely intensify
• GENIUS Act implementation: final regulations will determine whether all major issuers concentrate reserves identically or whether some diversification is permitted
• Fed facility access: whether stablecoin issuers gain access to the reverse repo facility or a dedicated liquidity backstop will determine the severity of future stress scenarios
• Non-dollar stablecoins: the paper focuses on USD-backed tokens, but euro-denominated stablecoins could eventually create similar dynamics in European government bond markets
• Tokenized Treasury products: if T-bill exposure migrates from stablecoins to tokenized money market funds, the run-risk profile may change but the underlying Treasury market exposure remains

Conclusion

The BIS Working Paper No. 1270 provides the first rigorous, causal evidence that stablecoin flows affect Treasury yields. The effects are modest today: a few basis points on 3-month bills. But they are asymmetric (outflows hit harder), state-dependent (effects intensify when bill supply is scarce), and growing with the sector.

The parallel to money market funds is instructive. MMFs grew for decades before their run risk materialized in 2008 and again in 2020. Each episode required emergency government intervention. Stablecoins are on a similar trajectory but without the regulatory infrastructure or central bank backstops that now exist for money market funds. The BIS paper is, at its core, an early warning: the same structural vulnerabilities that made MMFs a source of systemic risk are being replicated in a new, less-regulated, faster-growing vehicle.

Understanding these dynamics is essential for anyone building or using stablecoin infrastructure, whether as an issuer managing reserves, a protocol integrating stablecoin payments, or a policymaker designing the rules that will govern the next generation of digital dollars.

This article is for educational purposes only. It does not constitute financial or investment advice. Bitcoin and Layer 2 protocols involve technical and financial risk. Always do your own research and understand the tradeoffs before using any protocol.