数字银行与货币政策传导的演变

Digital banking and the evolvingmonetary policy transmission Abstract This paper maps the euro-area digital-banking segment and assesses how digital bankstransmit monetary policy relative to brick-and-mortar peers. I compile a hand-checked uni-verse of over 170 digital banks (2016–2025) from supervisory data, classifying institutionsby business model (e-retail, e-service, e-wholesale). Digital banks are small on average yetgrowing fast, rely more on household deposits—predominantly overnight—and hold largercash buffers and intangibles than traditional banks. Using a difference-in-differences designaround the ECB tightening cycle that began in July 2022 and the initial 2024 easing. Threeresults stand out. (i) The funding channel is stronger and faster at digital banks in tighten-ing: household deposit rates rise more and retail-funding spreads compress less, especiallyat overnight maturities and for stand-alone digital banks. Corporate-funding results are di-rectionally similar but weaker and less robust. (ii) Loan-rate pass-through is not stronger,implying margin compression and a later slowdown in lending growth at digital banks de-spite continued retail inflows.Household deposits are markedly more rate-sensitive thancorporate or unsecured funding.(iii) In early easing, digital banks cut new funding ratesrelatively quickly —particularly at longer maturities — yet effective deposit premia persistand retail inflows soften while margins begin to normalise. Policy implications concern theinteraction of market digital adoption and banks’ capacity to adjust balance-sheet durationthrough the monetary cycle, along with financial stability. Keywords:digital banks, neobanks, deposit rate pass-through, monetary policy transmission,deposit competition, retail funding, overnight deposits, household deposits, ECB tightening cycle JEL Classification:E52, G21, E51, E43, E58, O33 Disclaimer:The views expressed are those of the authors and do not necessarily reflect thoseof the European Central Bank (ECB). Non-technical summary Digitalisation has changed how banks in the euro area do banking. This paper maps the euro-area digital-banking segment and asks whether digital banks — institutions that deliver servicesalmost entirely online, with minimal physical presence — transmit monetary policy differentlyfrom traditional, branch-based banks.I compile a hand-checked sample of over 170 digitalbanks observed in supervisory data between 2016 and 2025, classify them by business model(retail, service, wholesale), and link this universe to microdata on balance sheets and interestrates. Identification relies on a difference-in-differences design around the ECB’s tightening thatbegan in July 2022 and the first cut in June 2024. Three findings stand out. First, the deposit channel is stronger and faster at digital banksduring tightening. Compared with traditional banks, digital banks raised household deposit ratesby roughly 15 bp more per 100 bp increase in the policy rate, and saw much smaller compressionin retail-funding spreads, with the largest effects at overnight maturities and among stand-aloneinstitutions.These differentials remain sizable even after accounting for how widely internetbanking is used in the countries where banks raise funds, whereas market digital adoptionitself makes pricing also more sensitive to policy. Corporate-funding results point in the samedirection but are less robust.Second, loan-rate pass-through is not stronger at digital banks:lending rates moved broadly in line with peers, implying margin compression and a weakermapping from higher funding costs into loan prices.Third, on quantities, resilience is price-driven:digital banks sustained pre-existing inflows of household deposits and retail fundingthrough the tightening phase, with household deposits proving much more rate-sensitive than atincumbents. By contrast, there is little evidence of comparable positive sensitivity for corporatedeposits or unsecured funding.Fourth, balance-sheet growth and non-financial private sectorlending decelerate more for digital banks late in the tightening, consistent with compressedmargins. An economic interpretation is consistent with competition on both sides of the balance sheet:arm’s-length, search-intensive deposit markets push digital banks to reprice more to retain retailfunding, while competitive lending conditions limit their ability to pass higher funding coststhrough to borrowers, compressing margins and franchise value. As the policy cycle turned, these asymmetries may gradually reverse. In the first quarters ofeasing, digital banks cut deposit prices more decisively, although the premium on outstanding deposits over traditional banks persisted.Unsecured-funding interest rate gaps narrowed aswell. At the same time, digital banks’ earlier advantage in retail-deposit inflows faded. Theseeasing-phase results are preliminary, given transmission lags and a short window, but they pointto weaker downwa