GM Confirms US Auto Sales Have Peaked, Avoid Auto Stocks – Forbes
One of the more annoying habits of today’s financial punditry is to insert the word “peak” in front of any industry that is facing the prospect of lower demand. I have heard peak oil, peak retail, and, of course, peak auto more times than I would like. When investing in stocks in a cyclical industry, though, gauging the peak is key to avoiding underperformance relative to the broad market indices.
On a conference call with analysts yesterday, GM’s Chief Financial Officer Chuck Stevens noted that the U.S.’ number one automaker had lowered its forecast for U.S. auto sales in 2017. From an earlier forecast for 2017 of “mid-17 million units,” which would have put 2017 at the level of 2016’s record total of 17.55 million units, Stevens guided to an industry volume of “low-17 million units.” He clarified later in the call that GM’s forecasters believe that industry volumes could fall 200,000-300,000 units in 2017.
So, that’s peak auto. Stevens’ forecast of a several hundred thousand unit decline, if correct, implies that the U.S auto market did indeed peak in 2016. After seven consecutive years of sales increases from the recent low posted in the post-Crisis year of 2009, the industry is going to have to adjust to lower demand.
The first sign of slowing demand is rising inventories, and Stevens noted that GM continues to work on its goal of reducing system-wide inventories to 70 days’ supply from the current level of 110 days’ supply. In my early Wall Street days as a cub autos analyst, I was tasked with tracking monthly inventories. I would have had to adjust the y-axis of my charts to show a figure of 110 days. That number is simply way too high. The ideal range of 60-70 days allows dealers to pre-finance their showroom inventory (floorplanning in industry parlance) and manage cash flows. At 110 days, that economic equation is out of whack, and, most obviously, dealers will actually run out of physical space to park all those unsold cars on their lots.
So, the solution is to cut production. Automakers book revenues when the vehicle is produced, not sold (they don’t own the dealerships,) so lower production must lead to reduced earnings estimates. That fact is not lost on analysts, as Zacks’ numbers show that the mean EPS estimate for GM for 2017 has fallen by a nickel to $6.18 in the past 30 days. The key takeaway from Stevens’ comments, though, is that those revisions are insufficient, and I believe GM’s estimates for 2017 will continue to be revised downward as we move through the summer.
It’s difficult to own any stock when estimates are falling, but is it wise to own auto stocks–GM and Ford both pay attractive dividends–past the peak year in auto sales for a cycle? The historical stock price performance couldn’t be any clearer: the answer is no. Auto stocks perform terribly in the first year after a peak is established.
To analyze this, I looked at the performance of Ford shares in the past three auto cycles. GM’s share price history has been scrubbed from many sources owing to the company’s bankruptcy and re-emergence in the past decade, and Fiat Chrysler America was part of DaimlerChrysler during the late-1990’s and early-2000’s, so that stock’s performance is not comparable to historical periods.
Ford has always been a good bellwether for the U.S. auto industry, though, and F stock’s performance during the past three post-peak years has been terrible. To analyze the figures I looked at the prior three years in which cyclical peaks for U.S. auto sales were established–1978, 1988 and 2000–and looked at Ford’s share price performance in the year following that peak. Note that I used 1988, not 1986, as the peak of the auto cycle for the 1980s. Tax reform in 1986 produced a spate of pre-legislation buying, and thus most industry observers (I started following the auto industry in 1992 and analysts were still talking about 1986) disregard that year’s sales figures.