Price-volume Correlation in the Housing Market: Causality and Co-movements (P.18)

Overall, we find that, first, exogenous variables, such as employment, household
income, the mortgage rate, etc., play significant roles in determining prices and trading
volume in the housing market, which supports the theories (e.g. Wheaton 1990) that
argue for the effects of exogenous variables as a possible explanation of the price–
volume correlation. Second, our results reject the hypothesis that prices do not Granger
cause trading volume at the 1% level, and reject the hypothesis that trading volume
does not Granger cause prices at the 10% level (at the 5% level for markets with low
supply elasticity). The Granger causality tests provide strong evidence for Stein (1995)
etc. and some evidence for Wheaton (1990). Third, we find decreases in house prices
reduce trading volume while increases in house prices do not affect trading volume,
which is a strong evidence supporting Stein (1995) etc. Fourth, we break down the
MSAs into two groups with high and low supply elasticity respectively, and find very
interesting heterogeneity. We find that trading volume Granger cause prices in tight
markets, but not in loose markets. We also find that house prices in tight markets are
less sensitive to variables related to financial constraints on homebuyers, including
mortgage interest rates and average household income. Moreover, we find growth in
employment has stronger effects on house prices in tight markets.

Next Page 19



Leave a Reply

You must be logged in to post a comment.