How Unemployment Affects Mortgage Delinquency

I have had on my to-do list determining the relationship between unemployment and delinquency rates for mortgages. Calculated Risk has beaten me to it. Here is their graph using annual data for 2009:

Click on graph for larger image


They do not provide the equation and the correlation coefficient, so I'll add that.

y = 1.317 x + 0.017 and R-squared = 0.44

Delinquency rates in 2009 were, on average, 1.3 times larger than unemployment rates. The correlation coefficient is weak but definitely shows a tendency.

Note: R-squared = 1.0 is perfect correlation and a value of 0 indicates no correlation, i.e. random data. So the linear representation is almost halfway between no correlation and perfect correlation.

I have experimented with eliminating 2, 3, and 5 outliers, and the equation and R-squared values are not significantly affected. I have also experimented with polynomial data fits, as well as exponential and logarithmic, with no major changes in correlation.

The relationship between mortgage delinquencies and unemployment may be biased by the large number of mortgagors who are underwater. In the current situation, the number of delinquencies is most likely higher than in a situation where mortgagors have positive equity in their homes.

All that being said, at the present time a 10% unemployment rate corresponds to a 13% to 15% mortgage delinquency rate.

It has been said that the economy can not get on a good footing until housing stabilzes. It is clear from this data that housing is in trouble until employment improves. It can also be argued that employment will not improve until the economy improves.

Can you follow that circle? Or is it a spiral?


Disclosure: No stocks mentioned.