Long-Term Trends in the Market for New Homes
|
|
|
|
|
|
|
Written by Tammy D. Kittle
|
|
Wednesday, 29 April 2009 |
|
In recent decades, it has become increasingly common for builders to sell a home before construction has actually begun that is, when only a developed lot is available. This practice gives the buyer the opportunity to customize the new home that is, to select a particular floor plan or special design features that can be implemented when the house is built. Chart 4 shows that the percentage of overall new home sales represented by the industry's "not started" category has risen steadily over the last twenty five years, from an average of 19 percent before 1981 to 36 percent averaged over the first three quarters of 1999.(n4) During the same period, both homes under construction and completed homes the industry's other two categories of new homes have accounted for a diminishing share of sales.(n5)
A second notable industry trend is the construction of higher quality homes. Houses today are typically much larger than houses built twenty years ago: average square footage for new single family homes has risen steadily from 1,700 in 1980 to 2,170 in 1998. In addition, new homes tend to have many more "extras," such as central air conditioning, fireplaces, and two or three car garages. As a result of these changes, new houses have become more expensive. The real value of construction put in place relative to the number of houses under construction is 34 percent higher in the 1990s than it was in the 1970s. Also noteworthy is that the average selling price of new homes has drifted upward relative to a fixed quality index, providing further evidence that the quality of new homes has risen.
The trend toward bigger and better homes, like the trend toward customization, has been fueled by demographic factors. The peak of the population distribution currently lies in the age range of forty to forty four, and household heads in this group are more likely to be moving up to larger homes than purchasing the smaller "starter" homes that dominated construction in the 1960s and 1970s. Moreover, income has risen for households headed by men and women in the broad home buying age range of thirty to forty nine. Between 1970 and 1990, the number of such households that reported income exceeding $50,000 (in 1990 dollars) rose from 2.77 million to 6.74 million, or from approximately 25 percent to 33 percent of all such households. Thus, more families can now afford large, customized houses.
How Housing Trends Affect the Inventory to Sales Ratio The growing practice of selling homes before construction has begun has contributed to the decline in the inventory to sales ratio through a simple compositional effect. Homes that are not yet started typically have low inventory to sales ratios. The reason is that each such home in essence, a finished lot is a largely generic product that can be easily replaced by another, enabling builders to limit the numbers they keep in stock. Since this category of new homes now accounts for a much larger proportion of new home sales, its low inventory to sales figures are driving the overall ratio down. If we were to adjust the months' supply series to make the composition of new home sales in terms of stage of construction more like it was through 1980,(n7) the aver age months' supply for the first three quarters of 1999 would have been 4.50 rather than 4.06 still low, but not so conspicuously out of line with earlier levels.
The trend toward building larger, higher quality homes has affected the inventory to-sales ratio through an increase in construction time.(n8) Average "time to build" for contractor built single family homes has risen from about four months in the early 1970s to nearly six months more recently. This increase in construction time appears to have lowered the inventory to sales ratio in the new home sector by slowing the response of inventories to cyclical fluctuations in sales.
How would such an effect come about? Longer construction time means that there are always more homes in the pipeline. Thus, when sales turn down, it takes longer for builders to sell off the stock in the pipeline. By the same token, when sales turn up, builders may be less quick to rebuild their inventories, knowing that if sales start to falter, unloading the stock will require more time than in the past.
Evidence that the supply side of the market is, in fact, reacting much more sluggishly to fluctuations in sales is presented in Chart 5. The chart shows the correlation between the stock of new homes and current and previous sales for two different periods: first quarter 1963 to fourth quarter 1984 and first quarter 1985 to third quarter 1999. In the earlier period, the correlation peaks at three quarters, meaning that inventories are most closely related to sales three quarters earlier.(n9) By contrast, in the later period, the correlation starts out in negative territory and only peaks at eight quarters, indicating that inventories are most closely linked to sales a full two years before.
The longer lag between a shift in sales and the inventory response sheds light on the decline in the inventory to sales ratio in recent years. The current low ratio may in part reflect a permanent change in the timing of the inventory response to a surge in sales.
|
|
Last Updated ( Wednesday, 29 April 2009 )
|