When the press release came in announcing that housing starts rose 8.4 percent last year to the highest level in 25 years, I was reminded of an old maxim: The only people who can be wrong more often than economists and still keep their jobs are weather forecasters.

It's a good thing that economists, who by and large predicted that starts would decline last year, don't run home building companies. They probably would have clamped down on operations to protect against a decline in starts. They might have trimmed staff, passed on land deals, and delayed computer purchases.

They would have missed out on an awful lot of opportunity.

The consensus forecast among housing economists last year—you can read it by clicking on articles from our January 2003 issue—was that starts 1.65 million. Instead, they rose to 1.85 million. The economic community missed by 12 percent.

The pack worried about a bunch of things that don't seem such a big deal to the economy now—public accounting irregularities, a lack of consumer confidence, and war with Iraq, among others. Their arguments were, of course, elegant, analytical, and quotable.

A Dismal Science Housing economists always seem to underestimate the strength of the market. One reason they tend to be so conservative is that there's more risk in not calling for a downturn than missing an upturn. If you don't see the end coming, and it comes, business people want your hide. People are a lot less angry when you fail to identify an opportunity.

How did the economists miss the mark by so much? For one thing, few of them expected the economy to remain weak for as long as it did. That led to a further decline in interest rates; 30-year fixed rates fell to 5.8 percent by the end of the year. The consensus forecast was that interest rates would be in the mid 6s by the fourth quarter.

The other unexpected factor that turned up last year was that the Census Department has chronically underestimated immigration. It has been off by as many as 200,000 people a year for a decade. Those people formed households that need a place to live.

A Bright Outlook The good news is that immigrant households tend to buy homes on average 10 years after they arrive in this country. The people who came here during the last decade will propel new-home construction in the coming decade. The Harvard University Joint Center for Housing Studies recently forecasted the need for 1.9 million to 2.0 million new-housing starts during the 2005 to 2015 period. The best may be yet to come.

All national housing forecasts, of course, should be taken with a grain of salt. While low interest rates seem to have lifted nearly all markets, home building is still a local business that depends on local economies. Additionally, the industry remains so fragmented that the best builders—the ones who can deliver most value to customers—can always grow and take market share, even during a market decline.

The final thing that's missing in a lot of the cool analytical formulas that economists use is the incredible desire in this country to live in a better place. When the economic stars align to promote homeownership, Americans rush in to take advantage. It's a dynamic that can't be measured. It can only be admired.

Boyce Thompson, Editorial Director

e-mail: bthompson@hanleywood.com