Homebuilding is one of the most cyclical sectors in our economy. So much so that many economists look at this sector as one of the most predictable leading indicators of the broader economy, and for good reasons. Buying a new home is one of the largest purchases that a household makes. It is a purchase that is very sensitive to interest rate fluctuations since a minor rise interest rates can make that desired home unaffordable. As such, the degree of confidence that a household has in the outlook for the economy can play a major role in its decision to buy a home.

Housing’s reputation as an accurate leading indicator is well-earned. For the past four recessions, homebuilding has turned down anywhere between 20 and 46 months before the overall economy headed into recession. The magnitude of the housing decline often gives us a sense of how severe the ensuing economic downturn will be. For the relatively mild 2001 recession, housing starts fell a modest 6.4 percent, while for the devastating 2008-2009 Great Recession, housing starts fell a whopping 73.4 percent.

Currently, homebuilding levels have been relatively flat for the past year or so. There are several reasons why most housing forecasters feel that housing will not see much in the way of declines over the next several quarters. The employment market is very strong at present; the unemployment rate is near 50-year low and worker wages are increasing. Mortgage rates, which tend to rise during the tail-end of a business expansion instead have dropped by a full percentage point over the past year. As a result, consumer confidence levels are quite strong overall.

Is the housing industry telling us that an economic recession is not on the horizon, or alternatively that if we do see a recession over the next 12 to 18 months that it will be extremely mild by historical standards? Maybe, but we also have to consider that the current housing market has been facing some historic challenges over the past decade that have depressed this current housing recovery. In the three decades leading up to the last recession, our economy had produced an average of about 1.5 million housing starts a year. This past decade, we’ve averaged less than one million housing starts a year, with the high-water mark reached last year when 1.25 million new homes were started.

This level of homebuilding is not only well below the pace of recent decades, but also well below what is needed to house our population. The level of homebuilding needs to accommodate three sources of new demand: the increase in the number of new households that are formed; second homes, vacation homes, time-shares and other homes that don’t have regular occupant; and, homes that are lost to the housing inventory through natural disasters, or that are demolished to accommodate other activities on that property. Estimates from the Joint Center for Housing Studies at Harvard University suggest that over the coming decade we’re likely to need to average about 1.5 million new homes a year to meet these needs.

So, what is the homebuilding industry telling us about the condition of the broader economy? One message is that a key sector of the economy, one that traditionally has signaled impending upturns or downturns, is not currently signaling a broader downturn. However, a more accurate interpretation is that due to the rising costs of homebuilding, a serious labor shortage in the construction industry, and difficulties among younger households in generating sufficient income to afford to rent or buy a home, we’ve been underbuilding the number of homes that we ultimately need. That suggests that we have pent-up demand for housing that will generate healthy levels of homebuilding, even if the economy were to weaken.

Kermit Baker is the senior research fellow for the Joint Center of Housing Studies at Harvard University. He may be reached via e-mail at [email protected].