The effect of income and collateral constraints on residential mortgage terminations

The prepayment behavior of home mortgage borrowers has been widely observed to be inconsistent with behavior implied by classical option theory. A substantial literature has emerged examining the problem, focusing on the characteristics of the mortgage and on the historic path of interest rates in attempting to explain the anomaly. This paper offers contributions to the literature in three respects. First, it explores the influence of household-level characteristics upon prepayment behavior, using both householder characteristics and collateral (house) value. Secondly, it empirically recognizes important interactions between the status of the prepayment option and the influence of income and collateral constraints upon prepayment behavior. Thirdly, it uses a major source of data that has not previously been used to examine the prepayment anomaly: the American Housing Survey. Among the findings are the following: when the household is either collateral constrained or income constrained, or the option is likely to be out of the money, the influence of the option value upon prepayment behavior is less by half. When the status of the option and the influence of potential household constraints are more appropriately recognized, these factors account for nearly all explanatory power otherwise attributable to household demographic characteristics.