In any given month, updates to a dizzying array of housing statistics wind up being analyzed and dissected for any hint of what might be in store for markets. Such statistics capture national and local house price movement as well as drivers of demand and supply for housing. The steady release of these measures in isolation from each other, however, makes it extremely difficult to gain a comprehensive understanding of housing market performance.
The point of such an exercise is to get ahead of extreme market movements before they create losses, but what exactly does extreme mean? Back in early 2004, for instance, near the beginning of what would become the largest rate of appreciation in Las Vegas home prices in any period since data for this market has been recorded, what information would have helped provide an early warning that home prices there would eventually pop?
Consistent with basic economic theory, home prices over the long-run are dictated by equilibrium between demand and supply. General macroeconomic conditions set the stage for housing including GDP and local market growth, mortgage rates, and unemployment rates, among others.
On the demand side, factors such as local market per capita income and population growth exhibit a strong statistical association with home price movement over time.
Likewise, various measures of housing stock are good indicators of the relative supply of housing in a market. Over the long-term, home prices exhibit an upward trend but at times can move well above or below that trend due to other factors.
To gain a bit of insight into how this plays out, consider the home price movement in a market well-documented for house price volatility during the boom and bust, namely, Las Vegas from 1978 to the present.
Looking at some combination of national, state and Las Vegas-specific market factors can explain the long-term trend. It is clear that the trend lines up reasonably well with home prices from 1978 through mid-2003, after which point home prices deviate substantially away from trend through the boom while falling well below trend during the bust period before reverting back to long-term market fundamental levels.
A couple of observations about home price movement are noteworthy from the Las Vegas data. First, home prices from mid-2003 to the present are significantly out of character from trend. Second, the boom and bust period reflects a tendency of many time series variables such as stock prices to eventually revert back to their trend levels after being too far above or below market fundamentals.
Understanding how far current price levels stray from their long-term trend would be more useful than just knowing that home prices increased some percentage from the previous period (as offered by most house price indexes).
But an even more useful measure of housing market performance would be to have some statistically reliable measure of whether the deviation from trend is significant or not and what factors specifically could be driving that outcome.
Deviations from long-term trends as reflected by post-2003 Las Vegas home prices can be explained by a set of factors driving speculation and contraction at times.
During speculative periods, excessive profit-seeking takes place. During times of contraction, loss minimization takes hold. Proxies of such behavior include changes in the investor-owned share of properties in a market, and local building activity.
Furthermore, growth of nontraditional mortgage products such as option ARMs or piggyback second lien mortgages fuel the potential for speculative behavior.
Another set of factors describing extra-normal deviation in home prices affect housing market velocity. Velocity in this case attempts to measure the degree of market frothiness which can influence and amplify behavior and reaction to market conditions.
Think of velocity as being comprised of a number of factors describing market momentum; driving up or down the market away from trend. Such indicators could include the average number of days on market for properties in an area, rent-to-price ratios, sale-to-list price ratios and vacancy rates, among others. Even foreclosure rates can explain deviations to the extent that they measure shadow inventory effects.
Finally, credit availability and usage may exacerbate speculative behavior during boom periods and contraction during busts.
Examples of such factors here include changes in underwriting standards and measures of leverage such as borrower debt-to-income ratio. These variables could be combined in such a way to explain deviations from trend from a set of regularly observed variables.
For instance, knowing that 25% of the deviation in home price in a given period was reported to be due to the change in market share of investor-owned properties versus other market factors provides a more holistic assessment of market movement than trying to distill meaning out of any single housing statistic upon its latest release.
Armed with an overall measure of housing market performance relative to long-term trend; an accompanying metric explaining whether that market is overheated or not; and importantly a way to attribute deviations in home prices precisely to selected market variables, market participants would be in a better position to take precautionary actions to limit their exposure in highly volatile markets.
Had such analyses been available in 2003, they could have introduced an important empirically founded sensibility check to market participants at a time when markets irrationally deviated from the long-term trend.
