Housing May Be Beginning a Decline

by Michael Bazdarich Senior Economist UCLA Anderson Forecast

 

            Our forecast had been looking for a decline in U.S. housing construction activity starting in mid-2005. In our September forecast, when that decline had not as yet begun, we pushed the starting date for the forecasted decline in housing out to early-2006. We think the logic for this coming decline is powerful, and we strongly believe it is going to occur, enough so to base our forecast around it. Nevertheless, such turns do not always take shape as soon as one might think they should, and so a little patience in calling and waiting for them is usually a wise take.

            Having said that, we will report that there are some signs of late that housing activity is starting to head lower. The indications are mainly anecdotal, and so it is still too early to call the turn. However, we can work through the details of these signals, as well as the more fundamental factors which have yet to reflect any downturn.

            The most widely-followed indicator of housing activity is housing starts. The chart above splits housing starts into single-family and multi-family components. Total starts in October were announced at 2.014 million units SAAR, down -5.6 percent from September and at the lowest rate since early this year. However, the sharpest decline was in multi-family starts, which have been trending down all year. Single-family starts dropped -3.7 percent, but this mainly offset a preceding increase.

            Single-family starts have been holding at the highest on record. Maybe the October decline is the beginning of a downtrend, but there is nothing in the behavior of this indicator to say so for sure. As for multi-family starts, these have been trending down since early this year, but the declines look to be driven more by market factors–the strength of single-family activity siphoning activity–than by any incipient signs of slowing.

            Meanwhile, new home sales held most of that decline in August and held near their lows in September. If these sustain their recent lows, that would point to a sustained decline in single-family housing starts in the months to come. In line with the recent decline in new home sales, applications for home-purchase mortgages have shown a downtrend since late-September. Then again, both home sales and mortgage applications have declined in the past, only to rebound in subsequent months, as the accompanying charts make clear.

            These declines, of course, come on the heels of over a year of rising short-term interest rates and an increase in mortgage rates since early-September. Maybe the declines are the beginning of a sustained downtrend. Again, this would merely be in line with our reading of the fundamentals. Nevertheless, it is still too early to state definitively that a decline has begun.

 

            Aside from these indications, a number of anecdotal factors are also pointing to a decline. Various regional markets are reporting softer home sales rates. At the same time, the equity markets have started to sell off the stocks of various homebuilders.

            As with the declines in home sales and purchase mortgage applications, these may all be indications of an incipient decline in the housing market. We will know more next time around. In the meantime, the only sensible thing to do is exert a little more patience and wait to see whether these signals are indeed the beginning of what we have been calling for.

            Sustained declines or not, it is clear that all of the growth in housing construction has evaporated over the last two years. After boosting GDP growth by nearly a full percentage point in late-2003 and early-2004, housing is barely contributing anything to economic growth at present.

            Why do we believe so strongly that housing is set for a decline? The next chart should drive the point home. Not only is housing construction activity holding at rates that in the past have been associated only with much faster population growth than is now being experienced and not only has housing activity been rising for 14 years after past expansion typically stalled out after four years, but population growth has been slowing for the last few years and is set to slow further in the years ahead.

            On top of the demographic influences, of course, are the facts that interest rates hit bottom two years ago and have headed up since, while the 2001 - 2003 tax cuts are fully two years in the past. Both fiscal and monetary stimuli had their full effects on housing quite some time ago, and those factors are working to restrain housing activity at present. Furthermore, our reading of the demographics suggests that home construction is above sustainable rates. Finally, there is evidence in a number of regional markets of outright “bubble” market conditions.

            On all these grounds, we believe housing is due for a sustained decline. The remaining questions are how hard the fall will be and when it will begin. Again, the recent anecdotal evidence suggests the decline may be beginning. We’ll soon see for sure.