The combination of rising home prices, declining interest rates, and near-frictionless refinancing opportunities can create unintentional synchronization of homeowner leverage, leading to a “ratchet” effect on leverage because homes are indivisible and owner-occupants cannot raise equity to reduce leverage when home prices fall. Our simulation of the U.S. housing market yields potential losses of 1.7 trillion from June 2006 to December 2008 with cash-out refinancing vs. only 330 billion in the absence of cash-out refinancing. The refinancing ratchet effect is a new type of systemic risk in the financial system and does not rely on any dysfunctional behaviors.

We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R2’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk.

The financial industry relies on trade secrecy to protect its business processes and methods, which can obscure critical financial risk exposures from regulators and the public. Using results from cryptography, we develop computationally tractable protocols for sharing and aggregating such risk exposures that protect the privacy of all parties involved, without the need for trusted third parties. Financial institutions can share aggregate statistics such as Herfindahl indexes, variances, and correlations without revealing proprietary data. Potential applications include: privacy-preserving real-time indexes of bank capital and leverage ratios; monitoring delegated portfolio investments; financial audits; and public indexes of proprietary trading strategies.