After the release of the weak January U.S. housing starts report on Feb. 16, we at Action Economics decided to take a look at major housing sector indicators to see where we currently stand with the ongoing housing market correction. Our compilation of major indicators shows a bounce in real estate activity since last year's big June-August downsizing, but with more weakness in the pipeline for home prices, starts, and construction.
Fair Weather Trends
First, a look at the geographic mix of housing starts data for January shows a steady downtrend from the beginning of 2006 in all but the Northeast, a small but seasonally sensitive sector, where warm weather through January boosted activity after a lull from June to September. Disruptive weather in the remaining regions probably aggravated the downtrend in January, which will likely extend through February with a "bad weather" reading for most regions.
If we look at national totals for the starts and permits data, and throw in completions and new home sales, we see that completions and new home sales are showing a less dramatic downtrend. Completions data lag starts, so these figures should continue to trend down through 2007, even if starts bounce through the year. Cancellations embedded in the new home sales figures may be dampening the cycle for this report, with less weakness as sales fall, but less bounce as sales improve. Canceled home sales are not subtracted from sales when they occur, but the later resale isn't counted as a new home sale either, hence dampening the swing.
The Mortgage Bankers Assn. purchase index data has bounced substantially since October and has since outpaced the other housing indicators. We are seeing a seasonal correction in February, but we will need further weakness through March and April if this index is to drop back to levels consistent with new home sales.
The pending home sales index bounced sharply in December to come back in line with prevailing levels of existing home sales. We expect a drop in both of these measures over the coming months, as recent gains likely overstate strength due to warm weather in the Northeast and Midwest.
Industry Insights
The National Association of Home Builders' composite index and its components have bounced significantly from the lows set in September, as this highly volatile and erratic index starts its cyclical uptrend. Though this index provides directional clues for the sector, the magnitude of the swings varies over time relative to other indicators. And the index tends to be a contemporaneous rather than leading indicator, despite the seemingly plausible assumption that industry "insiders" have special insights into future trends. Note that troughs in the index tend to presage sizable and prolonged building booms that you might expect insiders to have insights on in advance.
The home price data for 2006 showed a clear deceleration, though repeat-sale data like the numbers from the Office of Federal Housing Enterprise Oversight (OFHEO) and S&P Case-Shiller data have outperformed most expectations for the year as a whole with an 8.5% gain for the OFHEO index and a 7.6% increase for the S&P Case-Shiller index. These repeat-sale price measures tend to lag the housing cycle, however, and we expect 1% to 3% declines for these measures in 2007, as the effect of the housing correction takes hold.
Even with the declines we expect in the repeat sale indexes in 2007 and early 2008, home prices will have averaged 6% to 8% growth over the five-year boom-bust interval of 2003 2008 that is the same as the long-run average for these measures.
Private and Commercial Construction
Our observation about the "averaging" of home-price swings through the cycle likely explains the lack of observable impact of the housing cycle on the various measures of consumer confidence, such as the Conference Board's consumer confidence index or the University of Michigan sentiment index. Most households did not conduct real estate transactions within the period, most transactions of long-held properties occurred at normal annualized appreciation rates, and some survey responses likely moved in the opposite direction due to associated swings in building material prices, commodity prices, and interest rates over the period.
A lagging effect of the housing cycle should also be found in the construction measures, as starts lead construction, and hence the residential component of real gross domestic product. We expect continued moderation in residential construction spending through the middle of the year, even if the various real estate market measures continue to show stabilization.
But we also expect continued robust growth in nonresidential construction, as employment levels for the economy as a whole continue to post growth past the prior cyclical peak, which leaves demand for new office space. This is unfolding alongside accelerating tax receipts at state and local governments, rapid growth in global travel and tourism, and ongoing strength in nominal health-care expenditures that drive up demand for new infrastructure investment in roads, malls, hotels, and hospitals that are captured in this measure as well as the public construction component.
Sideways Trend
And, as we frequently note, the "ex-housing" economy is showing virtually no sympathetic downturn outside of housing, aside from the factory sector correction that started in the fourth quarter of 2006 and is extending into the first quarter of 2007. The year-over-year GDP data excluding housing is actually accelerating through December, though the downward revision we expect in the fourth-quarter GDP gain to the 2.0% area from 3.5% will largely remove the recent sharp uptick in the year-over-year measure and replace it with a sideways trend. In either case, it is clear that the ex-housing economy is thus far not following the housing sector downward at all.
In total, the weak January housing starts report reminded the market that we are not out of the woods with regard to the housing sector correction. Though all major real estate indicators are showing improved conditions since the big June-August downsizing, high inventory levels will likely continue to put downward pressure on prices, starts, construction, and completions through the middle of the year. As such, housing should continue to weigh on GDP growth through midyear.
With all that, we are about 20 months into the housing correction, and we have yet to see any meaningful net wealth or industry contagion effects of the real estate market adjustment. If the seasonally important spring real estate market shows stable conditions, the risks that the Federal Reserve will have to move on interest rates as a result of volatility in this sector will diminish considerably.
Click here to join a debate about housing prices.