SubscribeThe bad news for the housing industry continues nearly unabated. While there are suggestions that the end of the down cycle may be in sight, Global Insight believes that the turnaround will wait until mid-2007.
The question is whether those active in the residential marketplace should just hunker down for a time, or whether opportunities exist that will keep work flowing and position companies for recovery when it arrives.
Housing starts fell 14.6% in October, their largest drop since March 2005. Year-over-year, starts are down over 27%. While single-family units have taken the biggest hit, down almost 16% last month, multi-family starts have struggled almost as much with a 9.1% decline. New permit data, a reliable indicator of future housing activity, indicate that the bottom has yet to be reached. Housing permits fell for the ninth straight month, by 6.3% in October, down 28% from their year-earlier levels.
Single-unit permits dropped 3.8% in October, to31.7% below October 2005 levels. Single-unit permits have been steadily falling in most regions with no signs of leveling off. Permits in the Northeast were up 2.1%, but regional permit data are subject to revision and can be volatile. One cannot assume the Northeast is leveling off based on one month's results. Indeed, the region's inventory of unsold homes is at near-record highs, making a near-term turnaround unlikely.
In Global Insight's latest forecast, housing starts drop 11.0% in 2006 and another 13.8% in 2007, cutting GDP growth by about 1.2 percentage points in the fourth quarter of 2006 and by another 0.8 percentage point during 2007.
Weakness in the housing market is also reflected in the Mortgage Bankers Association's Market Composite Index of mortgage applications. The index fell 3.7% on a seasonally adjusted basis in mid-November and was down almost 2% from its year-earlier level.
The purchases index, which excludes refinancing, fell almost 3% and is down 15% from last year at this time. The refinancing index dropped more steeply, by 4.3%, and is now 22.2% below its year-earlier value. Clearly, the impact of reduced home price appreciation and higher interest rates are being felt in the refinancing sector. On the positive side, the average contract interest rate for 30-year fixed-rate mortgages dropped two basis points to 6.13%, the lowest value since January.
Lower interest rates are one ingredient toward improving home affordability. Another is price. The Office of Federal Housing Enterprise Oversight (OFHEO) House Price Index increased an annualized 4.6% in the second quarter, the lowest quarterly growth rate since 1999 and a clear indication that home price appreciation is decelerating.
The OFHEO House Price Index is a repeat sales index, a better indicator of home price appreciation than the existing and new home sales estimates, which can be influenced by changes in the market sales mix. It is also available for all states and 275 metropolitan areas, which allows for better analysis of local market conditions on home prices.
Year-over-year, the index rose 10.1% in the second quarter, although the better purchase-only index, which eliminates refinancing valuations, grew 8.2%. Since the purchase-only index is not available for sub-national detail, the relevant comparison needs to be with the total index, recognizing that it overstates the true home-resale market.
On this basis, while home prices appreciated year-over-year in all 50 states, they fell in five states: Massachusetts, Maine, Indiana, Ohio, and Michigan. Even in states that were leaders in year-over-year appreciation rates -- Arizona (24%), Florida (20.1%), and Idaho (20.1%) -- prices decelerated sharply. Indeed, prices are decelerating the most in the very states that had the highest appreciation. The exceptions are Louisiana and Mississippi, where price gains remain strong as these states continue to recover from the 2005 hurricanes.
With the housing market continuing to weaken and inventories continuing to grow, price declines are likely across more states and metropolitan areas over the next few quarters. Indeed, a year-over-year price decline in the national index is likely.
More recent price data from the National Association of Realtors (NAR) confirms this story. Third-quarter data from this survey indicates that home prices fell in 45 metropolitan areas, up from 26 metropolitan areas with price declines in the second quarter. While home price declines in struggling industrial cities is not surprising, the interesting feature was that the real estate slowdown had extended to previously vibrant markets in Arizona, California, and Florida.
Detroit had the dubious distinction of seeing the largest home price decline, by some 11%; Sarasota, Florida, was more unexpected in second place with a 9% drop. Indeed, 5 Florida metropolitan areas were among the 10 areas with the largest year-over-year price declines. However, the most largest retreats in home prices have been in Michigan, Ohio, Indiana, and other areas exposed to significant losses of manufacturing jobs, particularly in the automotive sector. Job losses in these areas have also increased foreclosure rates, which has depressed home prices further.
The news is not universally bad. More than 100 metropolitan areas experienced price increases in the third quarter, and 21 of them witnessed double-digit gains. The largest home price increases were seen in Salem, Oregon (25%); Elmira, NY (19%); and Salt Lake City, Utah (17%).
Even so, Americans remain upbeat about home values. A recent National Association of Home Builders (NAHB) nationwide poll found that better than 80% of homeowners expected the value of their home to appreciate over the next five years. Only 13% felt their home would lose value.
This confidence in home values may explain why remodeling activity picked up in the third quarter. According to the NAHB Remodeling Market Index (RMI), a measurement of remodeler perceptions of market demand, current market conditions increased from 45.6 to 47.8 in November. Still, readings below 50 indicate market softness. Enthusiasm for remodeling is mixed across the country. Current market conditions in the South (51.1) and West (54.1) have moved back into the positive range, and future market conditions in the West increased to 56.8. The Northeast was virtually unchanged at 46.6. Conflicting information arises from the Midwest, as current conditions improved, while future conditions deteriorated from 40.6 to 38.3, likely due to uncertainty in the region's manufacturing sector.
The RMI also segments its findings into owner-occupied versus renter-occupied units. The RMI owner-occupied component moved into positive territory in the third quarter, while the renter-occupied segment retreated slightly. Yet, in the future expectations index, the renter-occupied component increased from 28.8 to 37.1. While still a relatively low value, this is a significant move and an indication that rental properties are of interest once more after a prolonged period of disfavor as low interest rates made home purchases more attractive. Rental property remodeling accounts for a third of remodeling expenditures, according to the NAHB.
Sales prices, of course, are partially a function of sales. One reason prices have declined in some markets is a glut of homes in inventory. Sales of existing homes fell in 38 states in the third quarter, according to the NAR. Hardest hit were Nevada (-38%), Arizona (-35%), Florida (-34%), California (-29%), and Hawaii. Nine states saw sales fall more than 20% compared to their year-earlier levels. Bright spots did exist in North Carolina, Texas, and Montana; not coincidentally, areas where home price appreciation has been temperate.
What does this mean for residential construction?
Builder's confidence, as measured by the NAHB, was up somewhat in November for the second consecutive month. Still, an index value over 50 indicates that more builders view business conditions positively than negatively. One year earlier, the index stood at 61, and it peaked for this housing cycle at 72 in June 2005. With the most recent index value at 33, it is clear that developers see improvement in the marketplace, but they need to see more positive data before their enthusiasm is piqued once more. All regions showed lackluster levels of confidence, with the South (40) nearest to equanimity and the Midwest (16) most pessimistic.
The confidence numbers are borne out by results for leading builders. Pulte Homes, the largest builder in terms of revenue, reported last month that third-quarter income missed forecasts, and trimmed its expectations for the current period to below consensus estimates. Centex, the second-largest builder, also reduced its outlook last month, while other leading builders -- DR Horton, Lennar, and KB Home -- are also expected to see earnings and sales fall.
Toll Brothers, a builder of luxury homes, cut its guidance recently and reported results for the most recent quarter that were down about 10%. The company attributed its diminished expectations to cancellations that equaled about 7% of beginning fourth-quarter backlog. The cancellations were equivalent to 37% of the orders signed in the most recent quarter, and fully a quarter of the cancellations were from the Orlando or Northern California markets.
Global Insight analyzes the construction markets in all metropolitan areas in the nation. Here, we pay particular attention to the residential market, especially the single-family home market, which comprises the majority of spending.
It is tempting to overreact to a down (or up) market. Looking at forecast information one year out can be misleading, particularly in smaller markets that can exhibit eccentricities in any one year. A better estimate of potential, and one ideal for middle-to long-range planning, is to examine a five-year forecast outlook. Global Insight uses a five-year compound annual growth rate -- essentially an average growth rate -- for this purpose. We can then rank the metros by their outlook in a disciplined fashion.
Residential construction is measured in dollars, so it is a function not just of starts activity, but also home prices. Over the past five years, these tended to be mutually reinforcing on the upward side; i.e., areas with heavy start activity also had the most rapid home price appreciation. In the current environment, these two drivers have become mutually reinforcing on the negative side. The areas with the weakest start activity also tend to experience the weakest home price appreciation; indeed, often there will be home price depreciation in these markets.
Consequently, the housing markets with the weakest outlook tend to be those that experienced the most rapid appreciation during the residential boom. Conversely, those with the strongest outlook had more subdued activity over recent history.
Florida is the state with the most challenges for builders over the next few years. It has 3 metropolitan areas in the 10 slowest-growing residential markets and none in the top 25. Regionally, the South Atlantic has the largest number of weak metros in the top 25, largely due to Florida, but also due to Virginia and Maryland metros around the larger District of Columbia metro area. Weakness will also be found in some of the overpriced metros of the Middle Atlantic region and some of the manufacturing-intensive portions of the West South Central region.
Positive developments are more evenly distributed across regions. We are optimistic on the East South Central region, particularly smaller metros in Alabama and Tennessee. Attractive candidates can also be found in Georgia, Pennsylvania, and Minnesota.
The nature of the fastest- and slowest-growing metros is that they tend to be small. Larger metros simply cannot achieve rapid growth, and it would take a seismic shift in employment to create negative growth. Of course, it is the larger metros that are of greatest interest to builders, building materials manufacturers, and investors. The table offers a sample of the outlook for selected major metropolitan areas. The historical growth is the compound annual growth rate over the past five years, while the forecast growth rate is the compound annual growth rate over the next five, with 2006 the base year for the computations.
| Historical Growth | Forecast Growth | |
|---|---|---|
| Atlanta | 4.00 | 0.28 |
| Boston | 6.63 | -3.03 |
| Chicago | 5.93 | 0.05 |
| Dallas | 5.13 | -0.13 |
| Detroit | -7.74 | .038 |
| Honolulu | 13.46 | -1.68 |
| Houston | 12.79 | -0.88 |
| Kansas City | 4.75 | 3.29 |
| Las Vegas | 19.51 | -0.95 |
| Minneapolis | 6.14 | -1.11 |
| New York | 12.92 | -2.98 |
| Orlando | 21.34 | -4,71 |
| Philadelphia | 10.39 | 0.72 |
| Phoenix | 17.52 | -1.43 |
| San Francisco | 10.74 | -0.42 |
| Washington, DC | 8.40 | -0.74 |
| U.S. Average | 11.21 | -1.69 |