One of the key features of the current financial crisis has been a substantial breakdown in the historical relationship between the performance of key economic measures and delinquency rates on residential mortgages and many types of consumer loans. The unemployment rate, for example, has long been believed to be highly correlated with consumer loan delinquency rates. Although this statement is widely expected to be true, it does not seem to have held true in the current business cycle, which saw consumer delinquency rates begin rising well before the unemployment rate. In fact, the onset of the financial crisis, which was largely brought about by rising delinquency rates on subprime mortgages, occurred at a time when the unemployment rate was low and declining.

Unemployment Rate and Delinquencies

On the surface, the relationship between the unemployment rate and delinquency rates seems straightforward. If someone loses their job and does not find another one, their income is probably reduced and they are more likely to either delay or stop paying the mortgage, credit card bills and other consumer loans on time. When this happens in the context of a business cycle and lots of people lose their jobs, the loss of income results in rising delinquency rates. Although this analysis seems straightforward, there are some unanswered questions such as—what is the precise relationship between the unemployment rate and the delinquency rate and are there other variables that provide even stronger relationship to the delinquency rate and have more predictive power than the unemployment rate?

To answer these questions, we conducted basic econometric analyses between mortgage delinquency rates and various measures of the unemployment rate, unemployment claims and income.1 Overall, our analysis shows initial unemployment claims and continuing unemployment claims are better predictors of future mortgage delinquency rates than the unemployment rate, personal income or wages and salaries. The unemployment rate is still useful, but claims for unemployment insurance explain more of the change in mortgage delinquency rates.

We ran a different set of models to find the best predictor for credit card delinquency rates. Of all the independent variables we tried, the only one that proved to be statistically significant was nominal personal income. The unemployment rate, with a one-quarter lag, also has a statistically significant relationship with credit card delinquency rates but its relationship is much weaker than nominal personal income growth.