We investigate how long an insolvent debtor can avoid default when survival benefits creditors collectively, but individual creditors gain from early repayment. Standard theory predicts that debt is not rolled over and default is immediate. Previous experimental studies of related environments suggest that liquidation is delayed, and we confirm that pattern here. We then study how secondary markets affect outcomes. The duration of the credit pool is unaffected, but average net payoffs increase because claims are held longer before redemption. With markets, realized payoff dispersion is significantly higher. Prices do not predict liquidation timing but correlate with realized payoffs.
Using administrative bank loan-level data matched to private debt deals, we study the interaction between private and bank debt for a large segment of dual borrowers. These firms use private debt for investment and bank credit lines for liquidity insurance. Private debt carries higher spreads and commonly features flexible, equity-like terms that are associated with greater reliance on bank credit lines. Relative to bank-only borrowers, dual borrowers obtain more and larger credit lines and draw them more intensively following liquidity shocks. Our findings suggest that private debt substitutes for bank term loans but complements banks’ provision of credit lines.
We document the capability of large language models (LLMs) like ChatGPT to predict stock market reactions from news headlines without direct financial training. Using post-knowledge-cutoff headlines, GPT-4 captures initial market responses, achieving approximately 90% portfolio-day hit rates for the non-tradable initial reaction. GPT-4 scores also significantly predict the subsequent drift, especially for small stocks and negative news. Forecasting ability generally increases with model size, suggesting that financial reasoning is an emerging capacity of complex LLMs. Strategy returns decline as LLM adoption rises, consistent with improved price efficiency. To rationalize these findings, we develop a theoretical model that incorporates LLM technology, information-processing capacity constraints, underreaction, and limits to arbitrage.
We examine the dramatic collapse of the Terra blockchain in May 2022. Using granular blockchain data, we show that the run resulted from the interaction of three economic forces: subsidized money creation, real-time observability of transactions, and investor concentration. The early success of Terra's algorithmic stablecoin, UST, was fueled by highly subsidized deposit rates, which attracted many investors but failed to engage them meaningfully with other services. Blockchain transparency allowed investors to observe Terra's worsening fundamentals and monitor each other's exits in real time. Larger and sophisticated investors reacted faster to adverse signals and served as catalysts for the run. These findings challenge the idea that blockchain transparency levels the playing field and highlight how greater observability and concentration can amplify financial fragility. Our results contribute to the understanding of the limits of private money creation and the dynamics of runs in fully digital financial systems.
We identify a new and important property of affine term structure models: at long maturities, the slope of the term structure of risk premia becomes “almost observable”. Exploiting this property by fitting models to long-maturity data dramatically improves the precision of estimates of risk premia and expected short rates across the term structure. In a Monte Carlo experiment we show that using rates with maturities up to 25 years, rather than 10 years as is standard in the literature, reduces the standard deviation of estimation errors in risk premia, expected short rates and term premia by over 70%. Applying this approach to US Treasury data, we find that variation in expected short rates accounts for a much larger share of variation in yields (and variation in term premia a correspondingly smaller fraction) than in the widely used estimates reported by the Federal Reserve. Accounting for time-variation in yield volatility is key to successful extraction of the risk premium information in long rates. Our approach therefore departs from much of the empirical literature in two important respects: it incorporates long-maturity yields and allows for time-varying yield volatility, whereas most existing studies ignore long-term yields and rely on constant volatility models.
