As we see further in this paper, HPIs behave like prices of dynamic assets, so the similarity between home prices and stocks was not mentioned by accident. In particular, HPI returns, which we call “HPI rates,” resemble those of stocks and bonds, and a parallel to the empirical (real-life) rates and the risk-neutral ones can also be drawn.

Since the fixed income universe revolves around the concept of interest rates, it is important to find a link, if any, between them and home prices. Given known interactions between prepayments and rates, such a discovery can alter key perceptions of both prepayment and default options embedded into MBS or ABS. If, for example, home prices are proven to have a negative correlation with rates, then defaults and prepayments will have a negative correlation to each other. This finding will enlarge expected losses, and hence, valuation effect of defaults.

Dynamics of the HPI Rates: An Empirical Look

There are only a few sources for home prices and we used OFHEO as our primary source. The HPI time series is reported quarterly for all 50 states separately and for the U.S. as a whole. We computed quarterly returns and annualized them. The resultant measure, the U.S. HPI rate (hereinafter, "HPI rate"), is shown in Exhibit 1 going back to 1975. Along with the HPI rate, Exhibit 1 depicts the behavior of the 10-yr Treasury rate, the longest interest rate index available. The purpose of showing both indices on the same plot is two-fold: to assess a visual correlation (if any), and to compare the two distinctly different types of dynamics.

 

>>>