This paper surveys the literature on racial disparities in mortgage lending over the past twenty five years, including a discussion of the theoretical models, empirical methods, and data used to test for discrimination. Weaknesses in the models, methods, and data typically used together with...

We develop a unique paired loan dataset containing information on multiple conventional conforming mortgage loans of households to examine home equity extraction decisions over the period 2000-2006. The main question addressed is how much households borrow when refinancing their current mortgage...

We extend the work of Ambrose and LaCour-Little (2001) on traditional one-year adjustable rate mortgages by analyzing the performance of 3/27 hybrid instruments. Under this contract innovation, which first appeared in the mid-1990s, note rates are fixed for three years after which they convert...

The public release of the 2004-2005 Home Mortgage Disclosure Act data raised a number of questions given the increase in the number and percentage of higher-priced home mortgage loans and continued differentials across demographic groups. Here we assess three possible explanations for the...

The release of the 2004-2005 Home Mortgage Disclosure Act data raised a number of questions given the increase in the number and percentage of higher-priced home mortgage loans and continued differentials across demographic groups. This paper assesses three possible explanations for the observed...

Developing a good prepayment model is a central task in the valuation of mortgages and mortgage-backed securities but conventional parametric models often have bad out-of-sample predictive ability. A likely explanation is the highly non-linear nature of the prepayment function. Non-parametric...

Home equity lending grew rapidly from 2000 to 2008 with balances more than tripling. In this paper, we examine the role this phenomenon may have played in increasing aggregate default risk during the mortgage crisis. We also document a relationship between growth in home equity lending and high...

We develop estimates of risk-based capital requirements for single-family mortgage loans held in portfolio by financial intermediaries. Our method relies on simulation of default and loss probability distributions via simulation of changes in economic variables with conditional default...

"We assess nonparametric kernel density regression as a technique for estimating mortgage loan prepayments - one of the key components in pricing highly volatile mortgage-backed securities and their derivatives. The highly non-linear and so-called """"irr ational"""" behavior of the prepayment...

We present a simple theoretical model of adverse selection when lenders allow reduced documentation. The model shows how reduced documentation attracts both riskier borrowers and larger size loans. We then empirically test implications of the model using stated income loans originated during the...