AUGUST 2004

U.S. Housing Bubble?

Joseph E. Stiglitz, a 2001 Nobel Laureate in economics defined a bubble as: "[i]f the reason the price is high today is only because investors believe that the selling price will be high tomorrow—when 'fundamental' factors do not seem to justify such a price—then a bubble exists." Using this definition, Federal Reserve Bank of New York (FRBNY) senior economist Jonathan McCarthy and vice president Richard W. Peach examine data on U.S. house prices for evidence of a bubble. In their forthcoming FRBNY Economic Policy Review article, McCarthy and Peach also analyze the likelihood of a precipitous decline in housing prices and of a national economic downturn resulting from such a decline.

The most recent increase in U.S. house prices has been high by historical standards. The 36 percent average increase in real U.S. home prices since 1995 is more than twice that measured in the two most recent previous periods of appreciation (13 percent in the late 1970s and 17 percent in late 1980s). This upturn has continued—even during the recession of 2001, prompting fear of a bubble that is ready to burst. Given the definition of a bubble above, an increase in home prices by itself is insufficient to determine the existence of a bubble. The authors note that "[a]dditional evidence that relates current home prices to their fundamental determinants is required to solidify any claim of a bubble."

Part of the observed increase in home prices is due to the rise in quality of the housing stock. To assess the evidence regarding a bubble, it is necessary to take into account the "...physical and locational characteristics of homes sold... Otherwise, a perceived increase in home prices may reflect only the demand for more housing services that can be obtained through better quality homes with more amenities, even as the price of a standardized unit of housing services may stay the same." The authors use the constant-quality new home price index produced quarterly by the U.S. Bureau of the Census. It controls for changes in quality by accounting for a range of characteristics including square footage, number of bedrooms and bathrooms, the presence of a variety of amenities as well as the value of the lot. Comparisons of other home price series to the constant-quality new home price index show that "...a significant portion of price increases [found in the other series] can be attributed to increases in quality...and the quality of new homes sold—measured by size and amenities [by the U.S. Bureau of the Census]—has increased over time."

McCarthy and Peach assess the degree to which households are able to afford higher priced (and higher quality) houses. The authors analyze one of the commonly used benchmarks relating current home prices to their fundamental determinants, the home ownership affordability ratio, and find that this measure does not portend a significant decline in home prices. Considering the 425 basis point decline in fixed-rate conventional mortgage rates from 1990 to 2003 and the 50 percent increase in median family income since 1990, they conclude that a family earning the median income could qualify for a maximum mortgage 130 percent higher in 2003 versus 1990.

Declining mortgage interest rates also improved the ratio of annual principal and interest payments to median family income, an alternative measure of affordability. As of the third quarter of 2003, the ratio was at a 25-year low—it has remained at or below 15 percent since 1995. "This alternative affordability ratio suggests that a standard single-family home still remains quite affordable from a cash flow standpoint, even though home prices have increased rapidly in recent years. This in turn implies that home prices have risen in line with declines in mortgage interest rates and increases in median family income. Both of these conclusions argue against the existence of a home price bubble."

The equilibrium return on housing or the rent-to-price ratio is a measure of the rate of return on housing assets, a fundamental indicator of asset quality. McCarthy and Peach conclude that "[t]he rent-to-price ratio derived from the constant-quality new home price has fallen very little over the past few years and remains well within its historical range. Again, instead of providing evidence of a home price bubble, a consistently measured rent-to-price ratio indicates that home prices are not out of line with their fundamentals."

While McCarthy and Peach find no evidence of a national home price bubble, it is still possible that home prices could fall due to weakening economic conditions. However, past periods of slow income growth coupled with high nominal interest rates and high unemployment (such as in the early 1980s and 1990s) resulted in a 5 percent decline in real national home prices. The moderate nature of those national declines is due in part to the decentralized composition of the U.S. housing market. The national housing market is composed of "...many heterogeneous regional markets" and therefore, there can be "wide regional swings" in real home prices without serious consequence to the national statistics. "These wide regional swings may have been influenced by fluctuations in population and income growth that would not occur at the national level."

In order to explore the observed regional home price volatility, the authors analyzed state-level home price appreciation and personal income growth. Their analysis revealed that "...areas of rapid home price appreciation tend to be areas of rapid personal income growth, as one would expect." However, there were examples of states with similarly high growth rates in personal income experiencing widely differing rates of home price appreciation. The supply of new housing units accounted for this difference. States with a limited or fixed supply of housing stock "...because of population density and building restrictions...have tended to have the most volatile home prices as well as strong appreciation over recent years." Changing demand (in response to improving economic conditions) in combination with an inelastic supply of new housing units in states such as California, New York, and New Jersey led to recent strong home price appreciation in these areas, not home price bubbles. The inelastic supply makes home prices in such states more susceptible to changes in demand.

Ultimately, McCarthy and Peach find that "[t]he combination of the strong economic growth of the 1990s and the declines in interest rates is more than sufficient to explain the rise of home prices since the mid-1990s." While home price declines stemming from weaker fundamentals in states with inelastic supplies of new housing are possible, the likelihood of broader negative effects on national home prices and the U.S. economy is lessened due to "...the disjointed nature of the U.S. housing market."