The debt threshold: How political deadlock triggers bank repression
Olivier Jeanne develops a theoretical model analyzing three sectors—households, banks, and government—to study optimal financial repression when fiscal adjustment is delayed by political deadlock. The government faces three core options: pursue fiscal adjustment (preferred but politically difficult), use financial repression via the banking sector to avoid default, or default outright. Financial repression is modeled as a multi-dimensional policy controlling: a) quantity of government debt banks must purchase, b) how these purchases are financed (deposit expansion versus crowding out private lending); and c) whether the government extracts quasi-fiscal revenue by suppressing interest rates below market levels
The model assumes quasi-fiscal repression is more economically distortionary than conventional taxation, the government never defaults on banks (due to banking crisis risks), and fiscal adjustment occurs with constant probability over time. Empirically, Jeanne uses the Arslanalp-Tsuda database covering 23 advanced economies from 1989-2023, running threshold regressions to identify debt levels triggering bank absorption of government debt.
Three-stage progression of financial repression:
- Stage 1—No repression (debt below ~110% GDP): Government debt held entirely by non-bank investors; banking sector maintains efficient balance sheet with no government pressure
- Stage 2—Balance-sheet repression (debt 110-220% GDP): Banks absorb all incremental government debt beyond the threshold, financed primarily by expanding deposits rather than reducing private lending; no quasi-fiscal revenue extraction yet occurs
- Stage 3—Quasi-fiscal repression (debt above ~220% GDP): Government stabilizes debt by extracting direct revenue from banks through suppressed interest rates on reserves or mandatory below-market holdings
Empirical validation: The 100-120% debt-to-GDP threshold triggers balance-sheet repression across advanced economies. Below this threshold, banks absorb 26-36% of incremental debt; above it, they absorb 99-112%—statistically indistinguishable from 100%. All G7 countries except Germany have crossed this threshold, with Japan entering earliest (1990s). Japan’s trajectory (122% to 220% debt/GDP from 2001-2023) demonstrates banks can absorb massive debt increases through deposit expansion rather than crowding out private lending.
Policy implications: Balance-sheet repression has relatively low costs compared to alternatives (default or quasi-fiscal extraction) and can serve as an “effective bridge to fiscal adjustment.” Countries like the U.S. have “considerable leeway” before resorting to quasi-fiscal repression—using Japan’s experience as benchmark, if U.S. debt grows at 2.8% annually, it would take 37 years to reach Japan’s 2023 level, yielding welfare 97% of immediate fiscal adjustment versus only 3% under default.
From Fiscal Deadlock to Financial Repression: Anatomy of a Fall
Author: Olivier Jeanne
From: Hopkins University
Currency repression as fiscal tool
Oleg Itskhoki and Dmitry Mukhin extend the financial repression literature beyond traditional domestic debt markets to analyze currency-market interventions—policies that deliberately suppress returns on foreign-currency assets held by residents. This “expanded view” encompasses:
- Taxes or restrictions on foreign currency holdings by domestic agents
- Surrender requirements forcing exporters to sell foreign exchange at below-market rates
- Quantitative restrictions on household foreign currency demand
- Differential treatment between residents and non-residents in FX markets
The authors develop a small open economy model with sticky prices and segmented currency markets. Households value foreign currency for liquidity and hedging purposes, creating downward-sloping FX demand.
Findings:
- Household heterogeneity: When the economy contains hand-to-mouth workers who earn only domestic wages and cannot save, increased FX demand by wealthy savers depreciates the exchange rate, raising import prices and reducing real wages for constrained households. Financial repression that limits FX hoarding by savers leaves more foreign currency available for imports, redistributing real income toward liquidity-constrained workers. A utilitarian planner placing sufficient weight on these households finds positive repression welfare-improving despite aggregate quantities remaining unchanged.
- Fiscal revenue extraction: With downward-sloping household demand for FX deposits, the government acts as monopolistic supplier and can extract seigniorage by lowering returns below the world rate. The authors characterize the seigniorage-maximizing degree of repression and show temporary easing engineers real depreciation that boosts fiscal revenues from exports, while tightening has opposite effects. These intertemporal trade-offs affect timing but not present value of fiscal revenues.
The authors conclude: repression is a redistributive/fiscal tool, not a stabilization instrument, and effectiveness increases perversely with greater international isolation.
Sanctions and Financial Repression in the Currency Market
Authors: Oleg Itskhoki, Dmitry Mukhin
From: Harvard University, London School of Economics
The illusion of financial repression
Ricardo Reis, in this paper, addresses the twin debt problem (public debt at 95.1% global GDP, external liabilities at 173% for major economies) and identifies three financial repression channels:
- Direct taxation of financial institutions and capital flows,
- Devaluation through unexpected inflation or currency depreciation, and
- Yield discounts from artificially scarce government debt that provides services (liquidity, duration, currency denomination).
Reis provides the conceptual framework unifying the above two papers. Olivier Jeanne’s domestic debt mechanism (banks absorbing sovereigns at 100-120% debt threshold) maps to Ricardo Reis’s yield-discount component: Jeanne formalizes how repression works, Reis quantifies revenue. On the other hand, Oleg Itskhoki and Dmitry Mukhin’s currency repression (FX controls, capital restrictions) corresponds to Ricardo Reis’s devaluation component: both papers show that currency repression is dominated by FX interventions and only justified for redistribution. Reis’s empirical examples (Brazil capital taxes, Turkey liraization, China onshore-offshore spreads) validate the theoretical predictions.
His core finding: repression is illusive—it generates large temporary revenues but averages to zero as markets adapt, producing substantial efficiency losses. Post-pandemic inflation devalued U.S. debt by 2% GDP annually (2021-25), but pledgeability discounts collapsed from 3.9% GDP (2015-19) to 0.26% (2021-25).
Financial Repression in the XXIst Century
Authors: Ricardo Reis
From: London School of Economics
