How does the unemployment rate affect mortgage rates?

Question

from regression analysis I found that there is a strong positive relationship between the unemployment rate and mortgage rates. I can't figure out why. Any thoughts?

Answer 1

Your result most likely stems from an omitted variable. What other variables are you including in your regression. Have you tried lagged variables of unemployment?

Also what data are you using. If you are using data from the past 10 years following the 2001 recession unemployment starting declining. During the same time period the large demand for US assets caused 10-year bond rates to fall to extremely low rates. In turn mortgage rates also fell. Again I'm wondering if you are missing a couple of key variables.

Answer 2

You need to be careful, mortgage rates are prospective rates and unemployment data is retrospective data.  Data collected at time t may in fact reflect time t-1 and forward rates at time t+359.  Further, the mortgage market has itself changed over time being deposit funded and insurance reserve funded twenty years ago and mutual fund owned today.  That creates different owners with different liabilities.

Finally, time series regressions are very difficult to do correctly.  It is an entire field in itself.

Unemployment is related to bond prices because higher unemployment levels tend to result in lower inflation, which makes bonds safer and permits higher bond prices, so there should be a positive relationship with prices but a negative relation with rates.  However, a mortgage could be thought of as 360 forward obligations and the current unemployment level does not reflect future beliefs about the economy in a direct manner.

If you find a positive correlation that is very strong, there is also a possibility that you have a unit root problem and your t-tests are misspecified.  The significance could be spurious.  It partly depends upon whether the relationship is stationary or not.  If you are running your tests using an ordinary statistics package, it is likely your correlation method is invalid.

Answer 3

It does seem logical to me that higher unemployment rates would  decrease mortgage rates.  However, mortgage behavior is extremely complex, and rates are influenced by wany factors. Other factors could be moving to increase interest rates at the same time unemployment is moving to reduce rates.

Answer 4

I dont think there is any direct cause and effect explanation for this correlation.

Rather, both of these factors are part of broader economic forces which are surely related, but not related just to each other.

In general, the economic is going through a slump right now. That will force the unemployment rate to go up because employers are cutting jobs because they are not selling as much. Mortgage rates are going up because the banks and lenders are facing a liquidity crisis, and because the housing market is in a slump. For too long, lenders and borrowers alike assumed that housing prices would keep going up, and this led to an abuse of credit. Now this easy access to credit is being curtailed, and mortgage rates are going up to compensate.