Last week the widely reported headline was “Housing Starts Climb Almost 6 Percent in April” yet many in the transportation marketplace openly questioned how that was possible. Whether it is the railroads moving the lumber to build houses or the flatbed carriers moving the lumber and other building materials to build houses, the widely held view from the transportation industry is that the house construction industry is ‘softer’ or ‘less robust’ than it was last year.
Generally, I have little use for seasonal adjustments, and throughout the first three months of 2019 I listened patiently to those who blamed the ‘softer than expected’ housing starts on the weather. Using their seasonal adjustments, April was “almost 6 percent” higher than March, but transportation companies don’t move ‘seasonal adjustments’ and construction workers don’t build ‘seasonal adjustments.’ What was is really happening in housing?
All we know is that the number of housing starts on a non-seasonally adjusted basis in the first three months of 2019 was -8.9 percent compared with the number of housing starts on a non-seasonally adjusted basis in the first three months of 2018. While the number of housing starts on a non-seasonally adjusted basis was only down -2.9 percent in the month of April on a year-over-year basis, the number of housing starts on a non-seasonally adjusted basis is still -7.2 percent year-to-date. This is much more in line with the anecdotal and hard data provided by transports.
What does the rail volume on SONAR say?
Although it may seem at first to be an over-simplified way of looking at it, in the U.S. we still build houses out of wood. Long before going to the county clerk’s office to obtain a building permit, builders of all sizes are communicating with their suppliers/lumber yards about how many houses they intend to build in the coming months. Those suppliers are in turn communicating with sawmills who in turn communicate with loggers who are harvesting trees. When the volume of lumber shipped via rail is down -5.6 percent on a three-week moving average basis, down -3.2 percent on a 12-week moving average basis, and down -2.7 percent year-to-date, I am willing to accept that a portion of the negative housing start volume is due to the weather but would also quickly assert that volumes would still be down even if the weather was mild. Even if the weather was perfect, there hasn’t been enough lumber shipped to build more houses than last year.
Why is this important? Beyond the obvious direct economic inputs that result from housing construction (construction workers employed, railroads and flatbed trucks moving building materials, and all of the HVAC, carpet/flooring, appliances and other items that go into a new house), there are several significant waves of secondary and tertiary economic effects. Not only do banks benefit from the increased mortgage lending activity but the likelihood of consumers buying a new car goes up dramatically after they buy a new house. Large home improvement retailers such as Home Depot and Lowe’s estimate that over 60 percent of their sales are driven by consumers that are within 12 months of a housing change (you fix your existing house up to maximize the sales price and you make the new house your home after you move in).
Bottom line – the volume of housing starts and the volume of lumber shipped to make those housing starts are down on a year-to-date basis and suggest a more bearish/less optimistic view of the U.S. domestic economy than the headline “Housing Starts Climb Almost 6 Percent in April.” The long-term track record of freight flows to predict the economy leaves us more confident in the lumber volume than the seasonal adjustments.