How Will the Housing Slump
Affect the U.S. Economy?

Editor's Note: Some links may require registration or subscription.

Earlier this month, Federal Reserve Chairman Ben Bernanke highlighted the risk that the sinking housing market could put a dent in U.S. economic growth in the second half of the year. Mr. Bernanke, answering questions after a speech in Washington, said the housing sector is undergoing a "substantial correction" that will likely shave about one percentage point off of the nation's growth in the second half, "and probably something going into next year as well," estimates close to private forecasters' views. But he also said it is tough to predict how the weakness will impact the overall economy.

The Online Journal asked Celia Chen, director of housing economics for Moody's Economy.com; Christopher Mayer, a Columbia University housing economist; and Susan Wachter, a professor of real estate, finance and city and regional planning at the University of Pennsylvania's Wharton School, to discuss their views on how the housing market could shake out and how big a ding it could put in economic growth.

What do you think? Share your comments on our discussion board.

* * *

Celia Chen writes: Housing markets are sliding fast. Home sales are well off of last summer's peak, house prices are down on a year-ago basis, inventories are mounting, and leading indicators of housing activity suggest that the market will weaken further before it turns up. Indeed, by at least one indicator, conditions have sunk to the depths hit during the last housing bust in the early 1990s.

Housing markets clearly need to correct to offset some of the excesses that have built up during the exceptionally strong boom of the last several years. The bad news is that the correction will take about one half of one percentage point off of GDP growth this year and another three quarters of a percentage point off of growth next year, as the slowing in housing hurts employment, construction activity and reverses the wealth effect.

The good news is that the market is correcting, not crashing -- and other economic drivers are strong enough to withstand the hit.

* * *

Christopher Mayer writes: I generally agree with Celia. Yet a longer historical perspective is also necessary. The housing market is still in reasonable shape, even if housing indicators are below their recent historical highs. Existing home sales and housing starts are above their levels a decade earlier and more than 50% above their cyclical lows in the early 1990s.

The recent 0.50 percentage point decline in mortgage rates, falling construction costs and increasing commercial construction will likely ease the macro effects of a housing slowdown.

Despite the hype about adjustable rate mortgages, statistically, housing prices in the last three decades have been much more responsive to changes in long-term real rates than to inflation or short-term rates. Housing did not stop rising when the Fed started raising rates two years ago, but it did slow when long-term real rates started their sustained rise at the end of 2006. If long rates increase again, the housing market may be in for much more trouble.

Of course, there is no national housing market, and individual markets respond very differently to national trends. Newspapers and commentators have been calling this a bubble for more than four years.

The evidence just does not support this claim, as I noted last year in the Journal. However, some markets are clearly in trouble (South Florida, Phoenix, Las Vegas and, to a lesser extent, California). Condominiums are facing much more difficulty than single-family homes.

* * *

Susan Wachter writes: We were asked as part of this project to respond to the question: What's next for the U.S. real estate markets: Soft landing? Bubble bursting? Crash-and-burn? My answer is all of the above. As Celia points out, due to fairly strong economic fundamentals, the majority of U.S. real-estate markets are experiencing -- and are only likely to experience -- a modest correction.

I also agree with Chris that long-term rates matter nationally and long-term interest rates, though rising, are still historically low. But in markets with disproportionate exposure to adjustable rate mortgages, short rates matter as well.

Moreover, some markets are at risk of more severe price declines, specifically because of their exposure to the new exotic mortgage instruments. For the vulnerable markets with high proportion of teaser-rate ARMs, the potential fallout in terms of deteriorating mortgage credit and the Fed response in its recently issued guidance could have large impacts.

My recent research, co-authored with Andrey Pavlov, clearly shows that aggressive and overly abundant bank lending in certain markets exacerbates the boom and bust cycles there. Internationally, and in particular in Asia, we have seen the effects of lending booms magnifying the effects of real-estate crashes. In the U.S., some markets are more vulnerable to a potential real-estate crash.

As Celia's data demonstrate, affordability is at a more than 15-year low. However, affordability constraints are worse in states with the highest house price gains, and these are the areas where the new exotic mortgage instruments have expanded most. Teaser-rate ARMs have expanded most where prices have increased the most. Historically in the U.S. we have seen the regional consequences of concentrations of riskier mortgage lending.

For instance, in Los Angeles during the real estate declines of the early 1990s, neighborhoods that had large concentrations of ARMs in 1990 suffered some of the biggest declines between 1990 and 1995. Interestingly, these same sub-markets saw the smallest increase in use of ARMs and other aggressive instruments between 1990 and 1995, which suggests that ARM funding reacts relatively quickly to feared deterioration in market conditions, contributing to further market deterioration.

Similarly, Philadelphia neighborhoods that had an unusual concentration of high-risk mortgages in 1999 saw the lowest price appreciation in the recent years, even when controlling for median income.

The more recent, unprecedented availability and popularity of aggressive lending instruments -- such as interest-only mortgages and even reverse amortization mortgages -- has likely put many markets at risk. Cities and neighborhoods that have enjoyed a rapid appreciation due in part to the availability of inexpensive financing are likely to lose these funding sources in the face of even modest negative demand shocks. Such financing restrictions have the potential to magnify the effect of rising interest rates and devastate these markets.

* * *

Celia Chen writes: Chris and Susan bring up great points. Real estate is definitely all about location, so while the nation will not suffer terribly from the downturn in housing, there are regions of the nation that will be more vulnerable.

The Northeast urban centers and the smaller metro areas on the perimeters of the Bay Area and Los Angeles come to mind as areas where prices are highly elevated, and where the local economic outlooks are somewhat weaker than average. As Chris mentioned, the condo market is even more overdone than the market for single-family homes, and this imbalance adds downside risks to places like Florida despite its strong economic fundamentals. While we are not expecting any of these areas to fall into a recession, the risks on the downside are high.

A housing slowdown will impact the local economies directly through weaker employment in housing-related industries. Indeed, these industries have grown quite a bit as a share of total employment through the housing boom.

Additionally, the wealth effect from gains in home equity will dissipate. Together, these two forces helped to add one percentage point to real output growth last year, and their unraveling will make times harder for homebuilders and retailers this year.

* * *

Christopher Mayer writes: It seems as if we all agree that the risk of a national housing crisis is still low. However, there is disagreement about the impact of ARMs and exotic mortgages and what markets are most vulnerable. Consider the impact of so-called risky lending on housing markets. The evidence suggests that fears of a big increase in defaults and a spiraling down of house prices are overblown. ARM use in the recent boom was below previous peaks in 1988-89 and 1994-95.

Previous periods of high ARM use did not lead to a large increase in defaults. A recent study by Christopher Cagan used data from nearly 20 million loans originated in the last two years to suggest that the adverse macro impact of rate increases on ARMs and the more exotic loans has been overstated. He concludes that "mortgage payment resets will represent a far smaller drag on the market than what occurred in the early 1990s."

Recently, many households have returned to fixed-rate mortgages, as happened after the previous peaks in ARM use. What markets are at risk? Joseph Gyourko, Todd Sinai and I identified a handful of "superstar cities" (e.g. San Francisco, Boston, New York, and L.A.) where supply constraints, combined with a growing population and higher income for the most skilled households have led to a 50-year pattern of above-average house price appreciation. These superstar cities have seen their house prices increase most quickly in the recent boom but are also at greatest risk if interest rates rise or the economy falls. Updating the data from my study of bubbles with Charlie Himmelberg and Todd Sinai suggests risks in many superstar and non-superstar markets.

While it's possible to overstate current risks, given the recent decline in interest rates, I expect significant price declines in non-supply constrained markets like Miami, Phoenix, and Las Vegas. So far, house prices in the superstar markets appear to be destined for a soft landing, with modest house price decreases. Recent Mortgage Bankers Association data show that mortgage delinquencies are highest in the Midwest and South and lowest in the Northeast and West. Christopher Cagan also identified similar markets as facing the greatest risks. The Federal Home Loan Bank data shows that borrowers in superstar markets have lower loan-to-value ratios.

* * *

Susan Wachter writes: I agree with Chris's point that the 2005 data suggest risk in many superstar and non-superstar markets. Chris attributes price rises in superstar city markets to supply constraints and growing demand, and states that these cities are at greatest risk if interest rates rise or the economy falls. We also agree that these are the cities where housing markets have increased most. But why is it that this past year has seen prices rise to the point that there are significant risks in many superstar as well as non-superstar markets? It's not supply constraints; they are always with us. Rather, these markets are exposed to these new riskier lending instruments. For example, the Federal Housing Finance Board reports that in 2005 ARMs increased in L.A. from 52% to 63% and in San Diego from 61% to 71%, even in the face of increasing short rates and a flattening yield curve.

The likely outcome in terms of price declines will, of course, depend on the overall economy as well as regional conditions. Regional outcomes will be affected by the macro interest rate and growth environment as well, but the markets exposed to risky lending instruments are more at risk if our current national economic environment becomes less benign.

* * *

Christopher Mayer writes: One quick response to Susan's post: I agree that lending may play a role in the housing boom, but the current housing boom is truly global -- low real interest rates seem to be the common factor across all countries.

Now, let me comment on a point that Celia made in an earlier posting -- I think too much has been made of the link between housing wealth and consumer spending. Firm evidence is hard to find. Karl Case, John Quigley and Robert Shiller use data from 14 countries and find a "...rather large effect of housing wealth upon household consumption."

Alan Greenspan argues that "discretionary extractions" account for much of the rise in mortgage debt. Yet it is hard to disentangle the effects of house prices from other fundamentals that drive consumer spending. Erik Hurst uses data on U.S. households to show that most of the impact of house price changes on consumption appears to come from a minority of (liquidity-constrained) households who are house rich and cash poor.

The large rise in household mortgage borrowing may be driven by other factors besides the tendency of households to use their home as a new source of income. Lenders have increasingly moved away from making unsecured loans (e.g., credit cards) and toward making more home-equity loans. Consumers are much less willing to default on their mortgage and lose their home than to declare bankruptcy to avoid paying a credit card. In addition, home-equity borrowing has financed many housing-related investments such as home improvements or second homes. While the purchases of second homes in some markets -- Las Vegas and Miami come to mind -- have clearly been speculative, increased wealth and the aging of the baby boom must also play an important role.

To conclude, I think housing plays an enormously important role in the economy and in household balance sheets. Yet that role can be overstated. I agree with Celia's previous posting suggesting that the largest immediate impact of a housing slowdown will come in reduced employment for housing-related industries. I am less sure that the current housing slowdown will push the economy into a recession or lead to a large increase in consumer defaults or big decreases in consumer spending. However, I think the causality is likely the other way: The housing market has never been more at risk of a crash if the economy falters or interest rates rise.

* * *

Susan Wachter writes: I agree with both Chris and Celia's points that the impact of the housing slowdown on the overall economy is contained, impacting mostly the housing-related industries. The interest rate increases, the numerous warnings about the new aggressive lending instruments, and the new federal guidelines have likely prevented a widespread market crash similar to the Asian financial crisis, which caused major declines in housing prices as well as economy-wide slowdowns due to the impacts on the banking sector.

Nonetheless, in the U.S., markets that are disproportionately exposed to these instruments are still at risk, especially if the national economic policy requires further interest rate increases to combat inflation. Any sizable correction in real-estate prices in these regions does have the potential to spiral down, through employment effects or through mechanisms directly or indirectly affecting financing availability. One such mechanism could be through the household balance sheet, but it can also come in the form of overreaction of lenders against creative and flexible financing instruments in general.

To summarize, I do not anticipate a nation-wide crash in real-estate markets. However, some local markets that have enjoyed a large availability of aggressive lending instruments, such as interest-only or negative amortization mortgages, are at risk for two reasons. First, the increased affordability due to these instruments has likely pushed prices above their fundamental levels. Second, as lenders restrict the use of these lending instruments, these same markets are likely to see a more sizable adjustment going forward.

* * *

Celia Chen writes: It looks like we are all in agreement on the contours of the outlook, that the unwinding of the housing market will depress, but not derail, the broader economic expansion. With a record one in 10 jobs now in housing-related industries, the first impact will be on employment. Not only will residential construction employment decline, but industries ranging from mortgage brokers to furniture retailing to interior design and landscaping will take a hit. Indeed, this impact is already evident, with housing-related industries shedding an average of 10,000 jobs a month since March. And the losses will only accelerate as we move into next year.

The negative wealth effect will take a bit longer to kick in, but by next year it too will be a drag on the economy by removing a source of spending power for consumers. While we can, as Chris has, argue about the mechanism by which rising house prices affect consumption, there is no doubt that the increase in home equity has generated a cash cow for homeowners. Even only including active mortgage equity extraction, homeowners were able to extract an annualized $500 billion of additional spending money -- 5% of all disposable income -- in the first half of this year alone. As house price gains dissipate and mortgage credit quality deteriorates, this extraction will fade away, taking a bite out of consumer spending.

Other aspects of the U.S. economy remain quite healthy and will be able to power their way through the housing downturn. Businesses remain flush with financial resources to continue hiring and expanding, and are unlikely to pull back much over the next year. There are numerous downside risks, however, to this fairly sanguine outlook that my co-bloggers have mentioned and that may provide fodder for future forums.

Email your comments to rjeditor@dowjones.com.