Banks mediate short-term deposits into long-term investments through a process called maturity transformation. This structure is efficient, but it is fragile because a bank run can suddenly drain the liquidity that keeps the system alive. In 1983, Douglas Diamond and Philip Dybvig built a mathematical model of this fragility and showed that deposit insurance can act as an effective safety device. Ben Bernanke, using historical statistics, demonstrated that the Great Depression of the 1930s became severe largely because of widespread bank failures. Their work overturned the view that banking crises are merely a symptom of recessions and established that they can actively deepen downturns. The research also implies that a central bank acting as lender of last resort and a well-designed deposit insurance scheme are crucial for society. Modern financial regulations and crisis-management tools are built on these theoretical and empirical foundations.

The Diamond–Dybvig (DD) model is a general-equilibrium framework that combines many depositors with uncertain liquidity needs and long-term investment projects under incomplete contracts. The model features multiple equilibria, revealing that even a solvent bank can fail purely because beliefs shift. Deposit insurance and central-bank liquidity support eliminate the crisis equilibrium and yield a Pareto improvement, though at the cost of potential moral hazard. Diamond (1984) developed the economics of diversification and monitoring, showing that a bank which centrally observes borrowers is more efficient than a multitude of isolated depositors each doing their own monitoring. Bernanke (1983) combined VAR-style methods with archival sources to quantify how disruptions in bank credit channels affect total factor productivity and investment. He found that the widening of credit spreads was a dominant driver of the collapse in investment and employment, complementing monetary explanations based on money supply. Subsequent literature embedded these insights into New Keynesian and financial accelerator models, making financial intermediation frictions a standard ingredient of DSGE analysis. During the 2008 global financial crisis, runs in the shadow banking system and the drying up of liquidity matched the DD mechanism, and Bernanke as Fed Chair implemented policies derived from the theory. These developments informed Basel III capital rules and liquidity coverage ratios. Debates continue on how to apply the framework to crypto platforms and money-market funds.

Diamond–Dybvig (1983) formalised a three-period model with self-selection of liquidity shocks and a bank that implements a closed-end mutual liquidity-insurance scheme. The combination of deterministic production technology and stochastic withdrawal demand generates multiple equilibria, including a payoff-insolvent run equilibrium and a functioning one. Policy implications show that with full information a deposit-insurance backstop is Pareto improving, whereas under asymmetric information the problem reduces to optimal regulatory design. Diamond (1984) solved the delegated monitoring problem in a multi-period diversified game, proved a law-of-large-numbers property for loan portfolios created by bank diversification, and derived the shape of the optimal debt contract. Bernanke (1983) identified credit-supply shocks via a synthetic indicator method and estimated propagation coefficients from bank credit allocation to real activity. His evidence extended the Friedman–Schwartz quantity-theory view, quantifying how intermediary destruction raises the cost of reallocating funds. Subsequent literature integrated these insights into Kiyotaki–Moore credit chains and Gertler–Kiyotaki accelerators, forming modern macro-finance. Liquidity evaporations in opaque repo markets and SIVs during crises motivated facilities such as the Primary Dealer Credit Facility and TLGP. Basel III’s LCR/NSFR and total loss-absorbing capacity requirements internalise twin risks of liquidity and solvency under tail events. Current research frontiers explore how CBDCs might alter the stability equilibrium of commercial banks and whether bank-run analogues apply to DeFi protocols.