Delinquency and default, Housing markets, IRA,

Delinquency risk is often ignored in recent mortgage-related literature. However, it is postulated to be of sufficient influence to the yields of mortgages and the corresponding mortgagebacked securities (MBS) to warrant concern from mortgage lenders and issuers/underwriters of MBS. This paper provides the a rigorous analysis of the delinquency risk of residential adjustable mortgages (ARMs) by utilizing the 'time in delinquency' from delinquency in addition to the conventional delinquency incidence measure. Utilizing the Poisson and negative binomial models on 684 mortgage observations from 1980 to 1999, it is found that uncontrollable environmental factors are essentially responsible for delinquency incidence. However, controllable factors such as occupation of borrower, mortgage term and whether the loan consists of an initial preferential rate period are influential in determining the time in delinquency, if this situation arises. The implication is that although mortgage lenders and MBS underwriters may find that delinquency risk is largely dependent on macroeconomic trends, it may be possible to control and minimize the time in delinquency or the time before reinstatement when delinquency occurred. This is useful in reducing potential losses from default/foreclosure, as, although delinquent borrowers may not have default in mind when they initially miss an instalment, they may find default inevitable if the missed payments were to be allowed to accumulate.

Delinquency risk is a major area of concern to real estate mortgage lenders and potential issuers of Mortgage-backed Securities (MBS). This paper provides the first rigorous analysis of residential adjustable mortgage delinquency in Singapore. By studying 633 individual mortgages from 1980 to 1999, it is found that delinquency risk is dominated by macroeconomic factors and several mortgage loan specific factors. In particular, market sentiment, change in mortgage rate and the premium of mortgage over investment rates have a positive influence on delinquency, while change in unemployment rate, capital appreciation of residential properties and housing rentals exerts a negative impact. However, the direction of influence of the change in economic growth is unclear. Generally, property-specific and borrower-specific characteristics do not have a statistical significant impact on delinquency risks. It is also discovered that rate of delinquency and the performance of the economy have a consistently negative relationship. Consistent with the finding that macroeconomic factors exert the greatest impact on delinquency risk, lenders' abilities to reduce the overall risks of delinquency in their mortgage portfolio are limited. Potential issuers of MBS will also likely find it difficult to package the securities to reduce such risks.