By Andres Drobny,Drobny Global Advisors
The consensus seems to believe that the ‘soft patch’ the US has experienced is now coming to an end. Many are now anticipating a good bounce in activity in the fourth quarter, with perhaps better news to come next year.
There are lots of reasons to believe this. The Fed did a big ease, equities have recovered, consumer confidence has rebounded a bit, and some weekly indicators (eg, unemployment claims) have improved recently. Also, with Mr Bush’s big win, there are mounting expectations of a fiscal boost, although the size is still a big question. Business spending is also expected to pick up next year, and perhaps into year-end as some tax breaks roll.
There is a problem with all this, though. In two words: autos and housing. The two sectors that held up US and global growth are now starting to work in the opposite direction. The auto story is becoming better understood. Production chased sales higher through this year and now sales have dropped back. But production remains close to all-time highs, and is in fact scheduled to increase early in 2003. There is thus an involuntary inventory build that has to be corrected by a further cut in production. Unless, that is, car sales return to boom levels, and in a hurry.
There is a similar story building with housing, though the lag looks likely to be a good deal longer. The amazing boom in the demand for housing has, like autos, also elicited a supply response. The stock of existing homes for sale has been increasing through this year, and is running some 10-20% above the levels during 2000 and 2001. Housing completions have also been rising steadily. Starts are now at the highest levels in more than 15 years.
Now that’s not necessarily a problem, especially given the extraordinarily high level of turnover. House sales are running at something like 50% above the levels reached during the last housing boom in the late 1980s, and are some 20% above late 1990s levels. So, there is nothing unusual or even troublesome about the pretty good supply response in the housing market. High turnover combined with sharp price increases is bringing out the sellers, but, as with autos, it leaves the market very exposed to even a modest drop in demand. In such circumstances, housing market conditions could turn quickly from a situation of excess demand to excess supply.
That’s precisely how bubbles pop, and that seems to be what happened in the housing crash that started in 1989-90. The data suggests that, contrary to popular thinking, it was in fact a sharp fall in demand that precipitated the early 1990s housing slump. At the worst point in early 1991, new home sales had fallen something like 30-40% from average levels that held during the late 1980s (existing home sales seem to have fallen only slightly).
But, the supply of houses available for sale took longer to adjust, despite a sharp and pretty quick drop in new housing starts. It takes time to build a house and the level of housing starts in early 1990 was driven, in part at least, by the high level of sales that took place during the late 1980s. A similar phenomenon occurred with tech inventories in 2000, after a steady period of phenomenal tech sales! This time around, housing sales are running at extraordinarily high levels. There’s a lot of potential downside to sales, which could quickly produce a housing inventory problem.
More ominous is that, unlike with cars, house prices have been shooting up. Any inventory problem that emerges in the housing sector has the potential to produce a faster drop in prices than is typical into a housing downturn. That could prove real, real messy, especially with all the leverage in the system. But, why should the demand for housing moderate? It certainly was surprisingly robust this year, when a weak economy was expected to drag housing demand down.
But that seems to have reflected very unusual circumstances. Mortgage rates came down harder than in previous cycles, but the unemployment rate went up by less, at least so far.
It was a knife-edge, with both factors serving to support the housing sector, but can’t the same pattern of weak growth and low interest rates continue? It might, but it is not a good bet. Compare housing to Government bonds, the other asset that has performed real well this year (and is believed by some to also be in a bubble).
Consider two extreme states of the world: boom and bust. In the former, bonds will get killed, in the latter, bonds do great. If you don’t have a view on the economic outlook, bonds are a 50/50 bet in these extreme cases; but not housing. If it’s boom, and interest rates rise (and bonds sell off), demand for housing seems likely to decelerate. It may hold up OK, but it’s hard to believe it will remain at such elevated levels.
That’s a problem. And if it’s an economic bust, demand for houses should also weaken, and could collapse, even if mortgage rates stay low. Bonds seem the better bet; they are 50/50 in this little exercise. Housing, in contrast, seems to do poorly in either scenario.
When does housing do particularly well? In that very special knife-edge environment, where growth is OK, but not strong enough to push up interest rates, just like 2002. But, even in this housing-positive scenario, long dated bonds should also perform well. So, they still seem to beat housing, unless you think boom is coming!
All this suggests that it is a long shot to assume that the current housing boom will continue unabated. It might just happen, of course, but it is not a good bet and this, in turn, suggests that there remains considerable hurdles facing the US and world economy next year. Yet, once again, the consensus still seems to assume the worst of the downturn is behind us.
Andres Drobny is founder of Drobny Global Advisors. This article was first published on www.drobny.com.