by Calculated Risk on 1/20/2009 07:55:00 PM

David Rosenberg at Merrill Lynch wrote a research piece last week: "Not Your Father’s Recession ...(But Maybe Your Grandfather’s)" (no link)

Needless to say, the piece wasn't too upbeat.

But I was intrigued by some of the comments on vehicle sales. Rosenberg presented the following data (these are my graphs of the same data):

Click on graph for larger image in new window.

The first graph shows monthly vehicle sales (autos and trucks) as reported by the BEA at a Seasonally Adjusted Annual Rate (SAAR).

This shows that sales have plunged to just over a 10 million annual rate - the lowest rate since the early '80s recession.

The next graph shows the ratio of registered vehicles in the U.S. to the number of licensed drivers.

Rosenberg only plotted the data from 1975 through 2006 and the Y-axis was scaled from 1.0 to 1.2. This gave the appearance of a rapid increase in the ratio.

Looking at the raw data reveals that this includes private and commercial vehicles, so if someone drives a car to work - and then a different vehicle at work - that is counted as 2 vehicles per licensed driver. This would include truck drivers, police officers, and others. So a ratio of 1.2 vehicles per driver doesn't seem so high, although Rosenberg asks "Too many vehicles?"

This led me to plot the third graph (not in research piece).

This graph shows the total number of registered vehicles in the U.S. divided by the sales rate - and gives a turnover ratio for the U.S. fleet (this doesn't tell you the age of the fleet).

Currently this ratio is at 23.9 years, the highest ever. This is an unsustainable level (I doubt most vehicles will last 24 years!), and the ratio will probably decline over the next few years. This could happen with vehicles being removed from the fleet, but more likely because of a sales increase.

If the ratio of vehicles to licensed drivers declined to 1.1 (last seen in the early '90s recession), and the turnover ratio declined to 15 years, this would suggest sales would increase to 15 million vehicles per year. Although not as high as the recent boom years - this is still a sales increase of more than 40% above current levels.

Sales won't increase right away (look at the depressed sales during the early '80s), but this does suggest that auto sales are closer to the bottom than the top, and that auto sales will increase significantly in the future - although sales in 2009 will probably be dismal.