with Amir E. Khandani and Adlar J. Kim, Journal of Banking & Finance 34 (2010), 2767-2787.
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 R-squared'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 suggests that aggregated consumer-credit risk analytics may have important applications in forecasting systemic risk.
This document is the written testimony submitted to the House Financial Services Committe for its hearing on systemic risk regulation, held October 29, 2009, and it is not a formal academic research paper, but is intended for a broader audience of policymakers and regulators. Academic readers may be alarmed by the lack of comprehensive citations and literature review, the imprecise and qualitative nature of certain arguments, and the abundance of illustrative examples, analogies, and metaphors. Accordingly, such readers are hereby forewarned—this paper is not research, but is instead a summary of the policy implications that I have drawn from my interpretation of that research. This testimony focuses on three themes: (1) Establishing the means to measure and monitor systemic risk on an ongoing basis is the single-highest priority for financial regulation reform; (2) Systemic risk measurement and regulation will likely require new legislation compelling systemically important entities to provide more transparency on a confidential basis to regulators, e.g., information regarding their assets, liabilities, holdings, leverage, collateral, liquidity, counterparties, and aggregate exposures to key financial variables and other risks; and (3) Because systemic risk cuts across multiple regulatory bodies that do not necessarily share the same objectives and constraints, it may be more efficient to create an independent agency patterned after the National Transportation Safety Board (NTSB), solely devoted to measuring, tracking, and investigating systemic risk events in support of—not in competition with—all regulatory agencies.
Journal of Financial Economic Policy 1 (2009), 4-43
Financial crises are unavoidable when hardwired human behavior—fear and greed, or 'animal spirits—is combined with free enterprise, and cannot be legislated or regulated away. Like hurricanes and other forces of nature, market bubbles and crashes cannot be entirely eliminated, but their most destructive consequences can be greatly mitigated with proper preparation. In fact, the most damaging effects of financial crisis come not from loss of wealth, but rather from those who are unprepared for such losses and panic in response. This perspective has several implications for the types of regulatory reform needed in the wake of the Financial Crisis of 2007-2008, all centered around the need for greater transparency, improved measures of systemic risk, more adaptive regulations including counter-cyclical leverage constraints, and more emphasis on financial literacy starting in high school, including certifications for expertise in financial engineering for the senior management and directors of all financial institutions.
This document is the written testimony submitted to the House Oversight Committee for its hearing on hedge funds and the financial crisis, held November 13, 2008, and is not a formal academic research paper, but is intended for a broader audience of policymakers and regulators. Academic readers may be alarmed by the lack of comprehensive citations and literature review, the imprecise and qualitative nature of certain arguments, and the abundance of illustrative examples, analogies, and metaphors. Accordingly, such readers are hereby forewarned—this paper is not research but is instead a summary of the policy implications that I have drawn from my interpretation of that research I begin with a proposal to measure systemic risk and argue that this is the natural starting point for regulatory reform since it is impossible to manage something that cannot be measured. Then I review the relation between systemic risk and hedge funds, and show that early warning signs of the current crisis did exist in the hedge-fund industry as far back as 2004. However, I argue that financial crises may be an unavoidable aspect of human behavior, and the best we can do is acknowledge this tendency and be properly prepared. This behavioral pattern, as well as traditional economic motives for regulation—public goods, externalities, and incomplete markets—are relevant for systemic risk or its converse, 'systemic safety', and I suggest applying these concepts to the functions of the financial system to yield a rational process for regulatory reform. Also, I propose the formation of a new investigative office patterned after the National Transportation Safety Board (NTSB) to provide the kind of information aggregation and transparency that is called for in the previous sections. Another aspect of transparency involves fair-value accounting, and I review some of the recent arguments for its suspension and propose developing a new branch of accounting focusing exclusively on risk. I conclude with a discussion of the role of financial technology and education in the current crisis, and argue that more finance training is needed, not less.
with Nicholas Chan, Mila Getmansky, Shane M. Haas, The Risks of Financial Institutions and the Financial Sector, edited by M. Carey and R. Stulz, 2007. Chicago, IL: University of Chicago Press.
In this article, we attempt to quantify the potential impact of hedge funds on systemic risk by developing a number of new risk measures for hedge funds and applying them to individual and aggregate hedge-fund returns data. These measures include: illiquidity risk exposure, nonlinear factor models for hedge-fund and banking-sector indexes, logistic regression analysis of hedge-fund liquidation probabilities, and aggregate measures of volatility and distress based on regime-switching models. Our preliminary findings suggest that the hedge-fund industry may be heading into a challenging period of lower expected returns, and that systemic risk is currently on the rise.
Journal of Investment Management 5 (2007), 29-78.
During the week of August 6, 2007, a number of quantitative long/short equity hedge funds experienced unprecedented losses. Based on TASS hedge-fund data and simulations of a specific long/short equity strategy, we hypothesize that the losses were initiated by the rapid "unwind" of one or more sizable quantitative equity market-neutral portfolios. Given the speed and price impact with which this occurred, it was likely the result of a forced liquidation by a multi-strategy fund or proprietary-trading desk, possibly due to a margin call or a risk reduction. These initial losses then put pressure on a broader set of long/short and long-only equity portfolios, causing further losses by triggering stop/loss and de-leveraging policies. A significant rebound of these strategies occurred on August 10th, which is also consistent with the unwind hypothesis. This dislocation was apparently caused by forces outside the long/short equity sector—in a completely unrelated set of markets and instruments—suggesting that systemic risk in the hedge-fund industry may have increased in recent years.
with Nicholas Chan, Mila Getmansky, and Shane M. Haas, Federal Reserve Bank of Atlanta Economic Review 2006:Q4, 49–80.
In this article, we attempt to quantify the potential impact of hedge funds on systemic risk by developing a number of new risk measures for hedge funds and applying them to individual and aggregate hedge-fund returns data. These measures include: illiquidity risk exposure, nonlinear factor models for hedge-fund and banking-sector indexes, logistic regression analysis of hedge-fund liquidation probabilities, and aggregate measures of volatility and distress based on regime-switching models. Our preliminary findings suggest that the hedge-fund industry may be heading into a challenging period of lower expected returns, and that systemic risk is currently on the rise. This is a redacted version of our paper "Systemic Risk and Hedge Funds".
with Mila Getmansky and Shauna X. Mei, Journal of Investment Management 2 (2004), 6–38.
We document the empirical properties of a sample of 1,765 funds in the TASS Hedge Fund database from 1994 to 2004 that are no longer active. The TASS sample shows that attrition rates differ significantly across investment styles, from a low of 5.2% per year on average for convertible arbitrage funds to a high of 14.4% per year on average for managed futures funds. We relate a number of factors to these attrition rates, including past performance, volatility, and investment style, and also document differences in illiquidity risk between active and liquidated funds. We conclude with a proposal for the U.S. Securities and Exchange Commission to play a new role in promoting greater transparency and stability in the hedge-fund industry.
Commonfund Quarterly Fall 2002.
with Dmitry V. Repin, Journal of Cognitive Neuroscience 14 (2002), 323–339.
A longstanding controversy in economics and finance is whether financial markets are governed by rational forces or by emotional responses. We study the importance of emotion in the decisionmaking process of professional securities traders by measuring their physiological characteristics, e.g., skin conductance, blood volume pulse, etc., during live trading sessions while simultaneously capturing real-time prices from which market events can be detected. In a sample of 10 traders, we find significant correlation between electrodermal responses and transient market events, and between changes in cardiovascular variables and market volatility. We also observe differences in these correlations among the 10 traders which may be systematically related to the traders' levels of experience.