Chipotle Stock Repurchases Tell You Why ShopHouse Has Only 15 Units

| About: Chipotle Mexican (CMG)


Do ShopHouse and Pizzeria Locale really have the potential that investors hope for?

Management has allocated $2 billion into stock repurchases instead of opening at least 2,000 new units.

I estimate that ShopHouse AUR’s would probably be between 60 percent and 80 percent of Chipotle AURs.

Pizza and Burger categories also already very crowded.


An interesting aspect of a lot of comments about articles is that apparently someone does a few "back of the envelope" calculations in their head and is then ready to state that an author's projections "must be wrong" without having any deeper analysis of how all the numbers actually work. My article about CMG's Q2 earnings report also received such comments and so I'm writing this article to hopefully allow Chipotle (NYSE: CMG) investors to more thoroughly understand how the company's reported financial results will work in the future.

Another topic of this article also includes comments on the expansion opportunities for other formats using CMG's still substantial free cash flows. As you will see in my comments, I believe CMG management has already provided interesting "guidance" about what they think about the potential for their new concepts - which is that they have preferred to allocate $2 billion for repurchasing their stock instead of using the cash to expand ShopHouse or Pizzeria Locale.

To further quantify what could have been done with the $2 billion of cash, that could have opened at least 2,000 ShopHouse or Pizzeria Locale units which would have resulted in CMG having almost double the units that are currently open. I believe actions speak much louder than words or opinions and so I hope readers understand why I don't think either ShopHouse or Pizzeria Locale will ever provide significant future growth for CMG relative to the currently existing number of units.

A few additional comments first about Chipotle

Comments about my CMG income statement model which has been included in two of my articles (here and here) mainly seem to be along the lines of "you don't know how the numbers work" which is a very interesting comment given my almost 30 years of experience creating thousands of detailed financial models while analyzing rapidly growing companies.

As someone who also has extensive short selling experience from having managed a long/short hedge fund for nine years, I'll also mention that an almost endless list of short opportunities was always from the universe of rapidly growing "consumer services" companies (retailers and restaurants). Investor enthusiasm from previously large stock price gains seemed to create blinders which resulted in not fully appreciating when the inflection point of market saturation finally occurs and comp store sales in the future are typically then flat to negative.

Chipotle is a classic example of investor enthusiasm still lagging behind reality as even prior to the food safety issues, overall unit growth had slowed to around ten percent and comps had slowed to low single digits. Even with such already slowing growth, any dissenting perspectives about future prospects are greeted with lots of negative comments suggesting that an author has misunderstood something.

I have enough experience that I don't think I have missed anything, and I will be providing more details in the following. The critical variable is that new units usually open at levels at least 20 percent below overall AURs at any given time. As you will see in this July 22nd article in Restaurant News, one analyst is estimating that newer units are opening at revenue levels 30 percent below overall AURs.

Such a comment is in line with management's comments on the Q1 earnings conference call which was that new units in newer markets were being affected even more than units in more developed markets. Management confirmed in the Q2 call that such trends are probably continuing in describing that future openings will be re-focused on "higher performing" markets. Opening new units in more established markets with higher AUR's has an additional risk, however, of cannibalizing existing units in such markets which would then have a dampening effect on overall comps.

An underappreciated subtlety in how all this works for CMG is that there is actually a negative comp effect from the high unit AURs that they have in developed markets. High AUR units in general, although very productive economically, already have a tendency to "max out" in terms of comp growth as there is only so much throughput that can go through one unit. The still high overall AURs in a specific market due to the positive economic characteristics of such a market then do create a potentially attractive ROI for additional Chipotle units in the same market. Opening new units in such a developed market will still have an attractive ROI profile but overall comps will be depressed by cannibalizing the revenue potential of already opened units.

Regardless of opening new units in either developed or newer markets, new units open at significantly lower AURs than the overall AURs at any given time. As such, 20 percent lower AURs for new units are initially projected in my model even though units in developing markets are apparently opening at even lower AURs. There is also usually a "marginal location" effect with any expansion plan as well. It should be obvious that companies open the higher potential locations first and then to fulfill growth plans and expectations eventually start to take more chances with new locations and new markets.

All the factors described above affect why a ten percent growth rate in units does not immediately result in a ten percent growth in revenues. New units usually take three to four years to ramp up to corporate averages and a reasonable number may never get there given the "marginal location" effect described above. At the same time, mature units are also usually experiencing flat comps or, in the case of opening new units in more developed markets which cannibalize existing units, some mature units will start to experience negative comps.

ShopHouse and Pizzeria Locale potential

As for the future potential of ShopHouse or Pizzeria Locale, people are ignoring that ShopHouse has been in development for five years and has only 15 units and that Pizzeria Locale has only four units (and this article may help explain why). As for other possible formats, the burger category is already very crowded and my own opinion is that burger eaters are probably a lot less interested in a "food with integrity" and a "locally sourced ingredients" message. I also think Chipotle has a "long tail" in general concerning the food safety issues that will likely affect public perceptions about any new format introduced by the company.

As for ShopHouse specifically, another interesting data point is that AURs are currently estimated to be around $2.4 million. That may sound attractive compared to Chipotle AURs but those units are also only in very high potential markets where Chipotle units in those markets are probably averaging $3 to $3.5 million per restaurant. CMG management also has enough experience to understand what I am describing and so seeing that ShopHouse units may only have 60 to 80 percent of the revenue potential of a Chipotle unit is probably a reason why the format has so few units.

At least one analyst also believes that the overall market for ShopHouse units is much less than for Chipotle units (estimated at only 300 to 400 units over a ten year period) given the demographic apparently targeted by the concept and his comments are in this article. There are also other factors that suggest ShopHouse probably has much less potential than Chipotle. There is an already very well established competitor in Panda Express which has 1,800 units (after over 30 years of development). Panda Express AURs are also much less than Chipotle AURs at around $1.4 million per unit (which you can see in this table). CMG management would also be very knowledgeable about the much lower AURs of a "casual Asian" format and so that is probably another significant reason why ShopHouse has seen almost no expansion.

Aside my own comments about the potential of other formats possible opened by CMG, I believe the most telling comment about the growth potential of the other formats is that CMG could have opened 2,000 new units for the two other formats with the amount of money that the company has spent buying back its stock. Management apparently put its money where its mouth is on that topic by buying back the stock instead. As I described above about the much lower AUR of Panda Express, CMG management's capital allocation decision about that may be pretty rational.

One final thought about the actual importance of the ShopHouse and Pizzeria Locale formats for CMG may be from what I somewhat laughingly might call the "Tesla (NASDAQ:TSLA) effect."

For readers who are familiar with some of my articles about Tesla, I have a very dim view of the farcical situation at Tesla where the efforts to keep the stock levitated are based on there always being some sort of hoped for future development - but usually with no useful details or actual plans. I will apologize to the CMG management in even making such a comparison as they at least appear sincere in what they are trying to do as opposed to the manipulative and dissembling horror show unfolding at Tesla (and CMG is at least not taking deposits on future ShopHouse franchises!) but having a few "place holder" formats at CMG does have the effect of creating hopes for additional future growth although there is no visibility that any of that growth will occur.

There is no free lunch

The main thesis for those positive about CMG is that the company will return to previous AUR levels once memories fade about the food safety issues and that the stock will race back to new highs. Sara Senatore from Bernstein Research asked an interesting question on the Q2 conference call that suggested that such hopes are irrational.

Aside from a debate about whether a format that appears to have lost between ten and 20 percent of their customers will ever return to previous AUR levels, Ms. Senatore asked what are the cost structure differences between now and 2011 when a $2 million AUR produced a 26 percent operating margin (versus 15 percent in the latest quarter). The company gave a pretty good answer to the question listing five different things and, if you understand restaurant operations, you will understand that about 80 percent of the increased costs are now permanent.

Increased costs now being a permanent part of the expense structure is then what brings in the old economic concept about there being "no free lunch." Anything that appears free (including large stock gains!) has a cost that is borne somewhere and a tradeoff in the short-term about how to provide such an apparently "free" good. The applicability to CMG is that the positives cited by investors about why the company will fully recover (and hopefully their investments will recover too) are things that all had deferred costs that are now currently being paid.

A primary Chipotle attribute cited by CMG bulls is the "lines out the door" indicator that signified the previously broad popularity of the chain. Since there are trade-offs with everything, that also signified to me that restaurants experiencing such business levels were at full capacity and could not expand further (flat comps at best in the future) and that there also could be possible operational problems attempting to serve such a large number of customers.

I witnessed such operational problems first hand last September while standing in line at a Chipotle in my area. Right before reaching the front of the line I suddenly heard an already hectic server in the front line yell back to the food prep area "this chicken isn't done; you need to make sure that everything is cooked properly before bringing it out."

I might have been able to deal with one tray of chicken not being fully done (I was going to order Barbacoa that day!) but the additional comment that everything needed to be cooked properly suggested to me a restaurant that was being pushed beyond its capacity. In this case, the apparent "free lunch" (as apart from free burritos now being given out while attempting to rebuild traffic), had a hidden cost in corners being cut, employees being overworked and susceptible to sacrificing quality and standards, and overall consistent management of individual restaurants being compromised to achieve revenue targets.

While I don't really know what a sustainable capacity might be for a Chipotle unit, I believe that previous capacity levels are probably unsustainable to maintain consistently safe and high quality operations. Such a topic may be relatively irrelevant, however, as I believe pretty strongly based on the continued level of comps still down in the 20 percent range that a reasonable amount of previous customers will never return and that the previously very frequent Chipotle customer has now found some other dining options as well.

Some additional numbers behind my assumptions

Since various commenters have claimed my numbers "must be wrong," additional details about my assumptions are as follows:

Those assumptions result in there being around 3,150 units by the end of 2020.

As already described earlier in this article, new units open at significantly lower AURs than do overall AURs at any given time and so 20 percent lower AURs for new units are initially projected in my model. As such, and as I've also already mentioned, a ten percent growth rate in units does not then immediately result in a ten percent growth in revenues as new units usually take three to four years to ramp up to corporate averages. At the same time, mature units are also usually experiencing flat comps or, in the case of opening new units in more developed markets which cannibalize existing units, some mature units will start to experience negative comps.

In my overall model, I have incorporated all these factors but also actually have pretty positive comp assumptions which are as follows:

Since I'm sure that there will be some that think those assumptions are on the low side, keep in mind that the U.S. is now a very mature economy with very low nominal GDP growth.

In conclusion about the assumptions in my financial model, which projects $12.79 in 2018 EPS, all of the things that commenters claim I've not considered are all already in my model. Consensus estimates for 2018 have also already dropped quite a lot in just two days and are now $15.90 versus $16.43 prior to the Q2 earnings report. At that rate, consensus 2018 estimates will be close to my numbers by this time next year - which will still be 18 months prior to the end of 2018.

Spending money like a drunken sailor

I've already commented on the dynamic between allocating capital for share repurchases versus opening new units of both ShopHouse and Pizzeria Locale but there are other aspects of the share repurchases which I believe are worth mentioning. Putting it bluntly, I think CMG management has squandered a HUGE amount of cash (over $1 billion in the last three quarters) in repurchasing the company's shares at what I believe are very expensive valuations (at an average price of $488 a share).

Given that the cash balance at the end of the second quarter is now down to $565 million (combining cash and long-term investments), the company in three quarters has now allocated over 60 percent of its previous cash buying a stock at roughly 40-times likely earnings two years from now. Another way of looking at such purchases is that given the total assets of the company of $2.1 billion at the end of the quarter, the company has invested half of its assets over the last three quarters in a stock that they purchased at roughly 40-times earnings. No one - except someone who is either crazy or irresponsible - would make such an investment decision and so that is another aspect of why I still don't believe the company fully understands how its future prospects have changed.

There is one other unfortunate data point about such behavior on the part of the company and that is the recent allegations about Mark Crumpacker, CMG's Chief Creative and Development Officer who is now on administrative leave. While there is no evidence that any other senior executives of the company are engaging in similar activities, I am struck by the huge amount of stock recently repurchased as being similar to that of the spending of a drunken sailor on a binge.


What makes investing so interesting is that there are many different factors to consider when analyzing companies but that there will be no certainty to such analysis until at least one to two years into the future.

Based on my own fundamental analysis experience, extensive knowledge of a lot of different industries, and almost 30 years of very detailed financial analysis and modeling experience, my CMG financial model and EPS projections are my own realistic view of the company's likely future operating environment. As I've described earlier in this article, my overall model - not just the income statement model that I've presented in articles - does include very reasonable assumptions of all important variables such as comp store sales growth, additional revenues from new units, and also some future operating leverage even though future openings will not have the revenue potential of the existing store base.

Based on that overall model, I don't see some sort of disaster looming as CMG does generate a lot of excess cash in a more stable environment (between $300 million and $350 million a year is projected between 2018 and 2020). Those levels of projected free cash flows are also very interesting from the perspective of $1 billion being spent on share repurchases in just the last three quarters. As such, the company spent three years of free cash flow in only three quarters buying back its stock at 40-times my earnings estimate for 2018!

I also continue to believe that a future P/E multiple of 25-times earnings is appropriate for a company whose organic growth has now slowed to ten percent and which will be declining further from there. Commenters will say that is too low but even with a P/E multiple of 30-times earnings, that would result in a target price of $380 a share two years from now. What also influences my P/E multiple assumptions for all stocks is that I believe it is too risky to assume the current Federal Reserve "funny money" environment which has inflated the values of all financial assets will last forever.

For reference concerning that perspective, the Federal Reserve's current balance sheet has assets which are 22 percent of GDP versus historical averages of between six to eight percent of GDP. Those additional assets circulating through financial markets have inflated overall equity values by probably at least $3 trillion, which has resulted in current P/E multiples being 15 to 20 percent above historical averages.

Disclosure: I am/we are short CMG.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.