Richard Stanton

Estimation of Dynamic Term Structure Models

We study the finite-sample properties of some of the standard techniques used to estimate modern term structure models. For sample sizes and models similar to those used in most empirical work, we reach three surprising conclusions. First, while maximum likelihood works well for simple models, it produces strongly biased parameter estimates when the model includes a flexible specification of the dynamics of interest rate risk. Second, despite having the same asymptotic efficiency as maximum likelihood, the small-sample performance of Efficient Method of Moments (a commonly used method for estimating complicated models) is unacceptable even in the simplest term structure settings. Third, the linearized Kalman filter is a tractable and reasonably accurate estimation technique, which we recommend in settings where maximum likelihood is impractical.

Pricing Continuously Resettled Contingent Claims

Abstract This paper is a study of continuously resettled contingent claims prices in a stochastic economy. As special cases, the relationship between futures and forward prices is analyzed, and a preference-free expression is derived for these prices, as well as the price of a continuously resettled futures option, whose formula differs from Blacks futures option pricing formula due to the effects of marking-to-market the changes in the futures option premium.

The Bear's Lair: Index Credit Default Swaps and the Subprime Mortgage Crisis

During the recent financial crisis, ABX.HE index credit default swaps (CDS) on baskets of mortgage-backed securities were a benchmark widely used by financial institutions to mark their subprime mortgage portfolios to market. However, we find that prices for the AAA ABX.HE index CDS during the crisis were inconsistent with any reasonable assumption for mortgage default rates, and that these price changes are only weakly correlated with observed changes in the credit performance of the underlying loans in the index, casting serious doubt on the suitability of these CDS as valuation benchmarks. We also find that the AAA ABX.HE index CDS price changes are related to short-sale activity for publicly traded investment banks with significant mortgage market exposure. This suggests that capital constraints, limiting the supply of mortgage-bond insurance, may be playing a role here similar to that identified by Froot (2001) in the market for catastrophe insurance. The Author 2011. Published by Oxford University Press on behalf of The Society for Financial Studies. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com., Oxford University Press.

Anatomy of an ARM: The Interest-Rate Risk of Adjustable-Rate Mortgages

This article analyzes the dynamics of the commonly used indices for adjustable rate mortgages and systematically compares the effects of their time-series properties on the interest-rate sensitivity of adjustable-rate mortgages. Our ARM valuation methodology allows us simultaneously to capture the effects of index dynamics, discrete coupon adjustment, mortgage prepayment, and both lifetime and periodic caps and floors. We can, moreover, either calculate an optimal prepayment strategy for mortgage holders or use an empirical prepayment function. We find that the different dynamics of the major ARM indices lead to significant variation in the interest-rate sensitivities of loans based on different indices. We also find that changing assumptions about contract features, such as loan caps and coupon reset frequency, has a significant, and in some cases unexpected, impact on our results.

An Empirical Test of a Two-Factor Mortgage Valuation Model: How Much Do House Prices Matter?

Mortgage-backed securities, with their relative structural simplicity and their lack of recovery rate uncertainty if default occurs, are particularly suitable for developing and testing risky debt valuation models. In this paper, we develop a two-factor structural mortgage pricing model in which rational mortgage-holders endogenously choose when to prepay and default subject to i. explicit frictions (transaction costs) payable when terminating their mortgages, ii. exogenous background terminations, and iii. a credit related impact of the loan-to-value ratio (LTV) on prepayment. We estimate the model using pool-level mortgage termination data for Freddie Mac Participation Certificates, and find that the effect of the house price factor on the results is both statistically and economically significant. Out-of-sample estimates of MBS prices produce option adjusted spreads of between 5 and 25 basis points, well within quoted values for these securities.

Unobservable heterogeneity and rational learning: Pool-specific versus generic mortgage-backed security prices

Previous mortgage prepayment and valuation models assume that two mortgage pools with the same observable characteristics should behave indistinguishably. However, even pools with apparently identical characteristics often exhibit markedly different prepayment behavior. The sources of this heterogeneity may be unobservable, but we can infer information about a pool from its prepayment behavior over time. This article develops a methodology for using this information to calculate pool-specific mortgage-backed security prices. Knowledge of these prices is important both for portfolio valuation and for determining the cheapest pool to deliver when selling mortgagebacked securities. We find that unobservable heterogeneity between mortgage pools is statistically significant, and that pool-specific prices, calculated for a sample of outwardly identical mortgage pools between 1983 and 1989, may differ greatly from any single representative price.

A New Strategy for Dynamically Hedging Mortgage-Backed Securities

This paper develops a new strategy for dynamically hedging mortgage-backed securities (MBSs). The approach involves estimating the joint distribution of returns on MBSs and T-note futures, conditional on current economic conditions. We show that our approach has a simple intuitive interpretation of forming a hedge ratio by differentially weighting past pairs of MBS and T-note futures returns. An out-of-sample hedging exercise is performed for 8%, 9% and 10% GNMAs over the 1990-1994 period for weekly and monthly return horizons. The dynamic approach is very successful at hedging out the interest rate risk inherent in all of the GNMAs. For example, in hedging weekly returns on 10% GNMAs, our dynamic method reduces the volatility of the GNMA return from 41 to 24 basis points, whereas a static method manages only 29 basis points of residual volatility. Moreover, only 1 basis point of the volatility of the dynamically hedged return can be attributed to risk associated with U.S. Treasuries, which is in contrast to 14 basis points of interest rate risk in the statically hedged return.

Data Science Challenges in Real Estate Asset and Capital Markets

The real estate financial markets are complex supply chains. Understanding their behavior is limited by a lack of data that would capture the richly interconnected networks of financial institutions and complex financial products, e.g., asset backed securities. This lack of transparency is further compounded by limited knowledge of the contractual rules that control the flow of funds from mortgage pools to securities, as well as the financial events that regulate these flows. In this project, we will use the IBM Midas framework and tools to extract entities, relationships, events, contractual rules and risk profiles for financial institutions. Our source of information will be the MBS prospectus documents that are public and are filed with the Securities and Exchange Commission. We will describe the data management needs of the Haas Real Estate and Financial Markets (REFM) Lab and presents some recent REFM analytics that highlight the importance of these markets and the impact on systemic risk. We use excerpts extracted from the prospectus of a mortgage backed security (MBS) to illustrate the information extraction challenges and outline our approach to address these challenges.

Handbook of Financial Data and Risk Information II: US residential-mortgage transfer systems: a data-management crisis

This paper reviews the current state of residential-mortgage data structures from origination through the securitization supply chain. We discuss the various uses of these data, their limitations in mortgage-risk management, and the current lack of transparency in important segments of the mortgage market. We conclude that despite the size and importance of the mortgage market in the overall U.S. economy, current data-management practices make it difficult or impossible for borrowers, lenders, investors and government regulators to perform the oversight and analysis functions necessary to maintain an orderly market and to ensure fair pricing of securities backed by those mortgages. ∗The authors gratefully acknowledge helpful contributions from Rob Freund, Patrick Greenfield, and Dwight Jaffee. †U.C. Davis School of Law, U.C. Davis, jphunt@ucdavis.edu. ‡Haas School of Business, U.C. Berkeley, stanton@haas.berkeley.edu. §Haas School of Business, U.C. Berkeley, wallace@haas.berkeley.edu.

Consumer Lending Discrimination in the FinTech Era

Discrimination in lending can occur either in face-to-face decisions or in algorithmic scoring. We provide a workable interpretation of the courts’ legitimate-business-necessity defense of statistical discrimination. We then estimate the extent of racial/ethnic discrimination in the largest consumer-lending market using an identification afforded by the pricing of mortgage credit risk by Fannie Mae and Freddie Mac. We find that lenders charge Latinx/African-American borrowers 7.9 and 3.6 basis points more for purchase and refinance mortgages respectively, costing them