How can regulation help stabilize stablecoins?

by Tirupam Goel, Ulf Lewrick and Isha Agarwal Monetary and Economic Department June 2026 JEL classification: G2, G28, C6Keywords:capital regulation,liquidity regulation, BISWorking Papers are written by members of the Monetary and EconomicDepartment of the Bank for International Settlements, and from time to time by othereconomists, and are published by the Bank. The papers are on subjects of topical This publication is available on the BIS website (www.bis.org). ©Bank for International Settlements 2026. All rights reserved. Brief excerpts may be ISSN 1020-0959 (print)ISSN 1682-7678 (online) Making stablecoins stable(r): can regulation help? Tirupam Goel, Ulf Lewrick, and Isha Agarwal June 2026 Abstract We model a stablecoin issuer that optimises capital, cash and bond holdings under per-sistent stablecoin flows. Absent regulation, the issuer holds little capital and favoursinterest-bearing but less-liquid bonds over cash. This exposes coin-holders to defaultrisks and poses systemic spillovers via price impact of bond fire-sales. How can regula-tion mitigate these risks? We consider capital and liquidity thresholds as usable buffers.They can be breached in stress but discipline issuers by triggering additional redemp-tions, thus endogenising stablecoin flows.The thresholds work through asymmetricchannels. While the liquidity threshold only raises cash holdings, the capital threshold JEL Codes: G2, G28, C6 Keywords:Capital regulation; liquidity regulation; stablecoins; crypto; money mar-ket funds; financial stability; buffer usability. 1Introduction Stablecoins have grown rapidly over the past five years, with the two largest issuers nowrivalling major US Treasury money market funds in their holdings of short-dated govern-ment securities.Their deepening footprint in traditional financial markets has placed aspotlight on one of their main vulnerabilities: liquidity mismatch in reserve management.Fiat-backed stablecoin issuers hold demandable liabilities against interest-bearing but less Major jurisdictions are currently exploring a regulatory architecture that combines liq-uidity and capital thresholds for stablecoin issuers.1 The design and calibration of theseregulatory thresholds, however, is evolving differently across jurisdictions, reflecting a lackof consensus among policymakers. Two questions are unresolved. First, how do liquidity andcapital thresholdsinteractwhen applied to a non-bank issuer whose balance sheet adjustsmechanically with the demand for its liabilities?Second, what numerical thresholds areneeded to meet specific micro- and macro-prudential targets? This paper provides a frame- Our dynamic model features a fiat-backed stablecoin issuer that faces serially correlatedflow shocks.The issuer responds by choosing the privately optimal cash-bond mix of itsreserve assets and raises capital to absorb losses.Bonds earn a spread but are costly to interest. The unregulated model admits a closed-form solution under mild assumptions. Theoptimal cash policy as a function of flows mimics a “shark-fin” that sequences cash use acrossfour flow phases: (i) for large subscriptions, which render future outflows unlikely, the issuerholds no cash and only invests in bonds; (ii) for small to moderate subscriptions, it maintains a precautionary cash buffer; (iii) for small to moderate redemptions, it uses cash first to avoidcostly bond liquidations; and (iv) for large redemptions, it must sell bonds, with extreme Regulation seeks to address two wedges that a stablecoin issuer does not internalise in itsprivately optimal choices. First, limited liability induces risk-shifting. The issuer holds littlecapital and favours interest-bearing but less liquid bonds over cash. It thus underweights taillosses borne by coin holders (e.g. Li and Mayer [2026]). Second, forced liquidations transmitstress to money markets by depressing the prices of bonds and close substitutes (e.g. Ahmedand Aldasoro [2025]; d’Avernas and Vandeweyer [2024]). These externalities motivate a dualregulatory objective comprising a microprudential target—the issuer’s probability of default A novel feature of our framework is that regulatory thresholds operate asusable buffers,which makes stablecoin flows endogenous. Imposing regulatory thresholds as hard constraintsis common practice in the literature and is analytically convenient but can be self-defeating: for example, cash that cannot be drawn down when redemptions arrive is of little use.3We endogenously linked to the issuer’s balance-sheet strength. This design creates incentives forthe issuer to build buffers in normal times while preserving their usability during periods of Our model delivers three main results.First, the LR and CR thresholds shape issuerbehaviour throughasymmetricchannels. The LR threshold induces the issuer to hold morecash but has no direct effect on capital. By contrast, the CR threshold not only raises capital The second result concerns the ulti