Wednesday, June 20, 2007

Why Rent-to-Price ratios vary

I was talking to my colleague Tony Yezer yesterday about measuring rent-to-price ratios for different zip codes for Washington. It got me thinking about why rent-to-price ratios vary so much from place to place.

First is expected growth in prices. Places that are losing population (Detroit, Cleveland) will not see prices go up, because they have excess supply of housing, and will for the foreseable future. They must therefore have igh rent-to-price ratio (or low Price-Earning ratios for housing). Places that are gaining population but have no brakes on development will also not see prices go up, because house prices will not rise above replacement cost. For example, when house prices in Dallas go up a little bit, developers rush in to supply the market until prices fall back to construction cost. The only exception are places like the Park Cities, which have excellent schools that are not easily reproducible. Because prices don't go up in Dallas, the rent-to-price ratio is high.

Conversely, San Francisco and Maui are not replacable, so while they are somewhat volatile, the underlying house price trends are upward. As Gyourko, Sinai and Mayer point out, as people in the upper reaches of the income distribution get richer, they outbid each other for these unusual places: they can be viewed as the Monets of real estate. But these places are unusual.

The other thing that can influence rent-to-price ratios is the tax code. Because mortgage interest is deductible, owning is relatively more valuable in places with high federal and state marginal tax rates (i.e., Cailfornia, New York, New Jersey, Maryland). The large place with the highest combined Federal and State Tax Rate is likely San Jose; that with the lowest is El Paso. Sure enough, rent-to-price ratios in San Jose are very low; in El Paso they are very high.

No comments: