Bubbles? What Bubbles?

Special Commentary from the Office of the Chief Economist
by Amy Crews Cutts, Freddie Mac's deputy chief economist, and
Frank E. Nothaft, Freddie Mac's chief economist
May 6, 2004

The strong run up in home prices over the last several years has many wondering whether the housing market may be a bubble waiting to burst. Many definitions of asset-price bubbles exist. Shiller (2000) describes a bubble as when an asset’s price is driven by speculation – the belief that prices can only go up without any supporting market fundamentals. Based on this, we believe that at the national level a house-price bubble does not exist, and, in nearly all metropolitan areas, rising home values are firmly rooted in economic fundamentals. Among them are: supply of housing, transactions costs, interest rates and user costs, rent versus price growth, and household incomes.

Inventories of homes-for-sale are near their lowest levels ever [Exhibit 1], and the supply elasticity in housing is highly inelastic due to zoning and environmental constraints; these are most binding in the fast growing coastal states. Additionally, the majority of new homes today are built by national builders who secure options on raw land rather than purchase it, obtain permits in advance, and move operations to growing areas opportunistically. This inventory management was not possible when builders were all locally based.

Exhibit: 1 Inventories of Homes-For-Sale Are Very Low

Unlike the tech-stock market of the late 1990s, which did exhibit bubble characteristics, transactions costs are an impediment to speculation. A $9.95 commission through an online broker is trivial relative to the $1000s of shares transacted. But average transaction costs in buying or selling a home easily add up to 10% or more of the asset value – $20,000 on an average home, or about one-third of the national median household income. Furthermore, while the average equity stock is held for 3 months and day-trading is possible, the average tenure in an owner-occupied home is 14 years and home sales from buyer contract to closing typically take more than 30 days.

Mortgage interest rates have been declining for the better part of 20 years. The past year witnessed 45-year lows; with the cumulative effect that mortgage rates have fallen 40% since 2000 [Exhibit 2]. The user-cost of housing is the total difference in costs from owning versus renting a similar property. A simple user-cost calculation is to compare the after-tax mortgage cost, assuming 100% financing, and property taxes and maintenance costs to price appreciation. With today’s mortgage rates near 6% (about 4.5% after tax), and assuming taxes and maintenance at 1% each, house price appreciation at or above 6.5% means the simple user cost is negative, that is, owners get to live in their houses for free. Simple user costs have been largely negative since 2000, and have been generally declining for more than 10 years. Thus, it is understandable that homeownership rates keep rising even with a weakened economy.

Exhibit: 2 Fixed Mortgage Rates Have Fallen by 40% Since 2000

Recent research by Cutts, Green and Chang (2003), decomposed the changes in rents and owner-occupied housing prices and examined the “P/E ratio of housing” a la Leamer (2002). They found that in 10 of 15 metro areas, quality-adjusted rents rose faster than quality-adjusted home values during the 1990s, and that in 9 of 10 areas, unadjusted rents rose faster than home values. In the subsequent three to four years, those same cities generally experienced faster growth in house-prices than in rents and price growth was supported by fundamentals, even in California markets. The run-up in prices could not be fully explained in only two cities: Boston and Providence.

Over the long run, household income must grow at least as fast as home prices. Due to the recession and post-recovery job losses, incomes have not been keeping pace with home prices over the past few years. But an examination of home-price growth and household incomes at similar points in the business cycle, for instance the trough of the 1990-1991 recession with the trough of the 2001 recession, indicates that home-price and income growth over that 10-year period are in line [Exhibit3]. Keeping in mind that mortgage interest rates have fallen 40% since 2000 and that the two federal tax reduction bills have increased disposable household incomes, net housing costs (what homeowners care about each month) and disposable household incomes (what households use to pay for net housing costs) are much more aligned than home prices and gross incomes.

Exhibit: 3 Home Price Growth Has Been in Line with Income Growth

Clearly, some localized housing markets are at risk – those with a highly inelastic housing supply, such as San Francisco, for which demand shifts translate into large price movements – and they could see a fall in home values if demand drops. But that is not a bubble, it’s just fundamental volatility.

References:

Cutts, Amy Crews, Richard K. Green and Yan Chang. 2003. “Did Changing Rents Explain Changing House Prices During the 1990s? Yes!” Paper presented at the 2004 American Real Estate and Urban Economics Association Annual Meeting, San Diego, CA January 4, 2004.

Leamer, Edward. 2002. “Bubble Trouble—Your Home Has a PE Ratio Too.” Working Paper, UCLA Anderson School of Business.

Shiller, Robert J. 2000. Irrational Exuberance. Princeton, NJ: Princeton University Press.