Here is their thesis:
1. Geographic footprint may limit early cycle recovery.
2. Detroit restructuring could bring near-term disruption.
3. Sizable premium valuation appears unwarranted,
Let's address them individually:
1. Footprint, for this they conclude:
Here is the data on AutoNation vs. housing locations they provide:
In short, the company’s geographic footprint virtually mirrors the housing boom, which we believe boosted highly profitable truck sales and historical earnings well above realistic levels for any 2010 or 2011 recovery. While housing sales may experience a “V” shaped recovery in some of these markets – we do not believe that truck sales to contractors and housing professionals will immediately follow.
I am not sure what this proves, as while 20% of locations are in high foreclosure areas, 80% are not.
Also, I think the numbers are off. If I go to Realtytrac.com, I see the foreclosure rate for Orlando, Fla is far lower than the Weisel numbers (latest data used):
Click to enlarge
Here is the problem with the data. They are using "total foreclosures" which is not accurate because it assumes those homes are still on the market (not resold). It also inflates the data to make it worse than the reality. Now, I am not saying the above areas are not worse off than the national average, I am saying that they are Let's look at the correlation between housing and auto.
If we look at sales since the last recession '01-'02, we find that while housing sales increased 64%, auto sales stayed relatively flat. If we go back to the '90-'91 recession we find auto sales increased roughly 40% vs an over 120% gain for housing between recessions. Far from "boosting sales" the housing boom from 2001 to 2007 seems to have no effect on sales at all. This tells us the housing / auto sales link is a suspect one at best. What one should think is that it is economic activity that effects both, not one leading to the other as both will rise and fall into and out of recessions.
That being said, because the housing boom was so dislocated from reality and so severe, so has the downturn been. There is still significant downside to housing still as inventories continue to grow and millions more foreclosures loom. Autos, however, seem to have stabilized at 9 million units. It appears based on all evidence auto sales will bottom and climb before housing does.
Why? Asset life. I can live in my home for 30 years or more. In that time frame I will own an average 5 vehicles. I do not need to sell my home if I do not want to barring unforeseen circumstances. I will need a new vehicle in a few years no matter what I do. Population growth also bodes for auto sales. As our children age, they need vehicles well before they need a home and when they do have the option of renting.
What Weisel misses is that a return to 11-12 million units a year for autos (33% market growth from the current 9m) is necessary just to replace what is coming out of the market due to age demographics. Housing does not have this variable. What they also miss is the near 20% reduction in dealer ranks that will happen before this is all over. They briefly acknowledge this but give it little credence. It also means that AutoNation will have a far larger piece of that pie that is again growing.
Any disruption would be welcomed as AutoNation has made no secret of its desire to lower its exposure to domestic brands. A Chapter 11 by either of the large automakers (GM, Chrysler) would allow that process to proceed far easier than current.
Is it a value? After a 140% run, not really. Does it deserve a premium to the industry. Without question. When we consider competitor Sonic (NYSE:SAH) received a "going concern" notice and is trying to sell dealerships that no one wants to stave off a Chapter 11 filing, Penske (NYSE:PAG), Group 1 (NYSE:GPI) and CarMart (NASDAQ:CRMT) were barely profitable in 2008 at the 11 million units the industry sold. Because of this, a prolonged 9 million unit pace will eliminate many of these competitors.
The effect of all this will further expand AutoNation's market share without them having to expend a single dime to do so. Along with the additional sales, a little talked about effect will be the pricing power and margin expansion fewer competitors will provide.
I cannot understand the logic for this. Yes, 35% of revenues are from domestic brands as of 12/31. BUT, AN is also the #1 BMW dealer in the U.S. and soon to be the top Mercedes dealer. How they conclude this mix is "inferior" to other dealer groups makes little sense to me. Additionally, the "geographic mix" argument falls flat when one takes into account that as of 12/31, AN suffered sales declines in all sales categories LESS than the national average. Were Weisel's geographic mix scenario true, these numbers would have been worse.
We believe AN should trade in-line with the group based on what we view as an inferior brand and geographic mix and already lean cost structure, offset by relatively less leverage (more owned properties) and better stock liquidity.