The Correlation between Unemployment and Foreclosures

By Charles Carter

RISMEDIA, QUESTION: I've noticed the number of residential foreclosures is still high. What is the relationship between states' unemployment rates and states' rates of foreclosures in the U.S.? Are the two connected?

—Jeff Tyler, Palo Alto, Calif.

ANSWER: We can use a statistic called Pearson's product-moment correlation coefficient, or Pearson's r, to see if the two are related.

Using the newest data available — states' foreclosure filings and states' unemployment rates for May — the result is 0.56. That means there's a high positive correlation. Pearson's r ranges between 1.0, which means a perfect correlation, and -1.0, a perfect negative correlation. No correlation would mean a Pearson's r of 0.

Karl Pearson, who invented this statistic, was a Cambridge-trained English academic who pioneered mathematical statistics before the turn of the 20th century. Pearson's r measures correlation between two variables, but there are better measurements that are more robust — that is, less susceptible to error due to outlying data points.

Before the housing crash and its aftermath, there were long-term factors that explained the much smaller range of states' unemployment rates. Americans were moving generally into the southern and western states from Midwest states. The 5.1 percent national unemployment rate in 2005 stood at about half of what is today. Two of the states with the highest unemployment rates today, Florida and Nevada, had unemployment rates of 3.8 percent and 4.1 percent respectively in 2005.

RealtyTrac reports foreclosures remain high in 2010, and the government's mortgage modification isn't making much of a dent in the problem. Of course a high, positive correlation between unemployment and residential foreclosures doesn't explain causation, but the two probably overlap somewhat.

(c) 2010, McClatchy-Tribune Information Services.