New home sales are down a staggering 63% from peak levels of 1.4 million. Housing starts have
fallen more than 50%, and, adjusted for population growth, are back to the trough levels of 1982.
Furthermore, residential construction is close to 15-year lows at 3.8% of GDP; by the fourth
quarter of this year, it will probably hit the lowest level ever. So what's going to stop the housing
decline? Very simply, the same thing that caused the bust: affordability.
The boom made housing unaffordable for many American families, especially first-time home
buyers. During the 1990s and early 2000s, it took 19% of average monthly income to service a
conforming mortgage on the average home purchased. By 2005 and 2006, it was absorbing 25%
of monthly income. For first time buyers, it went from 29% of income to 37%. That just proved to
be too much.
Prices got so high that people who intended to actually live in the houses they purchased (as
opposed to speculators) stopped buying. This caused the bubble to burst.
Since then, house prices have fallen 10%-15%, while incomes have kept growing (albeit more
slowly recently) and mortgage rates have come down 70 basis points from their highs. As a
result, it now takes 19% of monthly income for the average home buyer, and 31% of monthly
income for the first-time home buyer, to purchase a house. In other words, homes on average are
back to being as affordable as during the best of times in the 1990s. Numerous households that
had been priced out of the market can now afford to get in.
The next question is: Even if home sales pick up, how can home prices stop falling with so many
houses vacant and unsold? The flip but true answer: because they always do.
In the past five major housing market corrections (and there were some big ones, such as in the
early 1980s when home sales also fell by 50%-60% and prices fell 12%-15% in real terms), every
time home sales bottomed, the pace of house-price declines halved within one or two months.
The explanation is that by the time home sales stop declining, inventories of unsold homes have
usually already started falling in absolute terms and begin to peak out in "months of supply"
terms. That's the case right now: New home inventories peaked at 598,000 homes in July 2006,
and stand at 482,000 homes as of the end of March. This inventory is equivalent to 11 months of
supply, a 25-year high -- but it is similar to 1974, 1982 and 1991 levels, which saw a subsequent
slowing in home-price declines within the next six months.
Inventories are declining because construction activity has been falling for such a long time that
home completions are now just about undershooting new home sales. In a few months,
completions of new homes for sale could be undershooting new home sales by 50,000-100,000
annually.
Inventories will drop even faster to 400,000 -- or seven months of supply -- by the end of 2008.
This shift in inventories will have a significant impact on prices, although house prices won't stop
falling entirely until inventories reach five months of supply sometime in 2009. A five-month
supply has historically signaled tightness in the housing market.
Many pundits claim that house prices need to fall another 30% to bring them back in line with
where they've been historically. This is usually based on an analysis of house prices adjusted for
inflation: Real house prices are 30% above their 40-year, inflation-adjusted average, so they must
fall 30%. This simplistic analysis is appealing on the surface, but is flawed for a variety of
reasons.
Most importantly, it neglects the fact that a great majority of Americans buy their houses with
mortgages.
And if one buys a house with a mortgage, the most important factor in deciding what to pay for
the house is how much of one's income is required to be able to make the mortgage payments on
the house. Today the rate on a 30-year, fixed-rate mortgage is 5.7%. Back in 1981, the rate hit
18.5%. Comparing today's house prices to the 1970s or 1980s, when mortgage rates were
stratospheric, is misguided and misleading.
This is all good news for the broader economy. The housing bust has been subtracting a full
percentage point from GDP for almost two years now, which is very large for a sector that
represents less than 5% of economic activity.
When the rate of house-price declines halves, there will be a wholesale shift in markets'
perceptions. All of a sudden, the expected value of the collateral (i.e. houses) for much of the
lending that went on for the past decade will change. Right now, when valuing the collateral,
market participants including banks are extrapolating the current pace of house price declines for
another two to three years; this has a significant impact on the amount of delinquencies,
foreclosures and credit losses that lenders are expected to face.
More home sales and smaller price declines means fewer homeowners will be underwater on
their mortgages. They will thus have less incentive to walk away and opt for foreclosure.
A milder house-price decline scenario could lead to increases in the market value of a lot of the
securitized mortgages that have been responsible for $300 billion of write-downs in the past year.
Even if write-backs do not occur, stabilizing collateral values will have a huge impact on the
markets' perception of risk related to housing, the financial system, and the economy.
We are of course experiencing a serious housing bust, with serious economic consequences that
are still unfolding. The odds are that the reverberations will lead to sub-trend growth for a couple
of years.
Nonetheless, housing led us into this credit crisis and this recession. It is likely to lead us out. And
that process is underway, right now.
Mr. Moulle-Berteaux is managing partner of Traxis Partners LP, a hedge fund firm based in New
York.
