ChartingTheEconomy.Com

Unemployment Shows Housing has Farther to Fall

February 21, 2009

By: ChartingTheEconomy.Com

The purpose of this paper is to:  1) show a correlation between employment and housing during prior economic crises, and 2) use that correlation to help predict where housing prices are headed in coming years.  On a national basis there has not been a decline in home prices with which to compare today’s situation.  Because of this, I examine several regional markets which have experienced large housing price declines in prior economic slowdowns.  Markets which only had modest housing price declines in prior economic slowdowns were not used because they do not accurately reflect today’s situation.

This analysis clearly shows a correlation between employment and housing in the examined markets.  It also shows that unemployment peaks prior to when home prices bottom.  In other words, home prices continue to fall even after employment begins to recover.  This is bad news for today’s market given that most economists believe that unemployment will continue to rise.  The minutes from the January 27-28, 2009, Federal Open Market Committee meeting state, “staff again expected that unemployment would rise substantially through the beginning of 2010.”1  If today’s national housing market follows the examples of prior regional housing market collapses, we will not see the bottom in prices for at least a few years.

California in the Early 1990’s

Many who live in California are familiar with housing price declines.  During the early 1990’s many homeowners in California saw substantial declines in the prices of their homes.  These declines were not on par with what is happening today, but they provide a relevant point of comparison.  The question is:  How did unemployment change in relation to these housing price declines?  Let’s take a look at charts for the Los Angeles, San Diego, and San Francisco metropolitan statistical areas (MSAs).  Charts 1-3 show the annual unemployment rate graphed with the annual price change in housing for each of these MSAs.  Chart 4, shows the annual unemployment rate graphed with the cumulative price change in housing for Los Angeles during the 1990’s.

Charts 1-4 show some interesting correlations between unemployment and housing prices for each MSA.  First, in each of these major MSAs housing prices began to decline as the unemployment rate increased during the early 1990’s.  Second, in each case the unemployment rate peaked in 1993, however, the housing markets remained weak for several more years.  In L.A. and San Diego, housing prices actually continued to fall for three more years after the unemployment rate peaked.  In San Francisco, housing prices stopped falling with the peak in the unemployment rate, however, prices remained flat for several more years.

Washington, DC, in the Early 1990’s

Let’s take a look at Chart 5.  In the Washington, D.C. MSA we see a similar pattern of continued weakness in home prices long after the peak in the unemployment rate.  In this case, the unemployment rate peaked in D.C. in 1992 yet home prices remained at their bottom for at least five more years.  While not as bad as L.A. and San Diego, where home prices continued to decline after employment improved, this is another good example of how the recovery in housing lags the recovery in employment.

Finally, Chart #6 provides a broader picture.  In this case, the national unemployment rate is graphed with the housing price changes for the Case-Schiller Composite 10.  Note that Case-Schiller did not track the top 20 markets at this time. Therefore, this chart represents their broadest measure at the time.  As you can see from the chart, the national unemployment rate peaked in 1992 and began to improve steadily over the coming years.  However, housing prices in the top ten (10) MSAs continued to decline in 1993 and showed only slight gains in the next three years.

Conclusion

The purpose of this paper is not to compare today’s economic crisis to that of the early 1990’s.  It is to:  1) understand the correlation that occurred between unemployment and housing prices during the economic slowdown of the early 1990’s in several regional MSAs, and 2) to use this correlation to help determine where housing prices may be headed today.

Some clear conclusions can be drawn from this analysis.  First, this analysis clearly shows a correlation between employment and housing in the examined markets.  Next, it shows that unemployment peaked before home prices bottomed.  Given that most economists believe that unemployment has not peaked in the current crisis, this analysis tells us that housing prices most likely have farther to fall.

There are many factors influencing housing prices in today’s market.  These factors include government tax credits, mortgage relief programs, high household debt service ratios, adjustable rate mortgage resets, and poor loan quality. As much as anything, all these competing factors just confuse the housing picture.  To clarify the picture a historical perspective can be helpful.  If history proves to be a good indictor in this case, then we are in for several more years of falling housing prices.

Source:

Charts 1-6:  Unemployment data is from the Bureau of Labor Statistics. Housing Data is from the S&P, Case-Shiller Index.

Endnote:
1  Federal Open Market Committee, “Minutes of the Federal Open Market Committee January 27-28, 2009,”  p. 12