David Einhorn Comments on Green Mountain Coffee Roasters

Author's Avatar
Oct 16, 2013
We added to our short position in Green Mountain Coffee Roasters (GMCR, Financial). Although the company again missed the consensus estimate for sales, bullish analysts scrambled to lower forward revenue forecasts while insisting that all is well in mudville. Attention quickly shifted away from the results when new CEO Brian Kelley announced on the Q3 earnings call that GMCR would hold its first ever investor day in September. When asked what prompted the decision, Mr. Kelley said, "I think a number of people on our team have found that an investor day that is crisp, but thorough on the key issues can be very valuable to help people understand our company. And I think it's – that's the core purpose is to help you understand our company better.

"The evening before the invitation-only event on September 10, the New York Times reported that there was a large discrepancy between the number of K-Cups the company says it has sold and the numbers implied using data from the tracking firm IRI. For years there have been questions about misconduct within GMCR's distribution and accounting departments. This new information raised the possibility that this activity is continuing, with GMCR potentially booking hundreds of millions of dollars of non-existent K-Cup sales.

We watched the company's live investor day webcast, waiting to see how Mr. Kelley would fieldany inquiries on the matter. Despite a lengthy Q&A session, the questions never came. We later heard that Mr. Kelley was asked about it during a break and essentially told investors that for the New York Times to be right, then you'd have to believe that he was in on it.

That's a non-denial denial worthy of Nixon Attorney General John Mitchell. Mr. Kelley, a former Coke executive, was hired to replace Larry Blanford in late 2012. Prior to his tenure at Coke, Mr.Kelley spent five years as the CEO of SIRVA, Inc., a relocation services company whose shares_collapsed during his tenure and were delisted or 'relocated' from the NYSE shortly thereafter. A subsequent lawsuit that was settled alleged that when Mr. Kelley was told of a reserve shortfall at a subsidiary, he, together with other defendants, refused to take a charge to earnings and instead raised earnings guidance. The suit referred to Mr. Kelley as "a spin doctor due to his ability to twist negative information about SIRVA."

Three years ago, GMCR stopped disclosing the number of K-Cups sold, which is comparable to Ford not disclosing the number of cars and trucks it sells. One analyst (the only one who is publicly bearish on the stock) tried to derive K-Cup sales volume by multiplying the number of brewers GMCR says are in use by the number of K-Cups GMCR states are used per machine. At the investor day, Mr. Kelley derided this derivation: "So, you're applying straight math that we don't do." The analyst pressed and Mr. Kelley suggested that the investor day was neither the time nor the place: "We're not going to get into that here. That is not the intent, and we're not going to go into that, the model, in that kind of detail here." Not here? Then where? Apparently basic sales data is not GMCR's cup of tea and does not meet the criteria for an investor day Q&A that promised to be thorough on the key issues. It is impossible to reconcile GMCR's piecemeal disclosures with its financial statements, and the new CEO repudiates straight math. If the core purpose of the investor day was to help people understand the company better, perhaps the next question should have been, "Are you in on it, Mr.Kelley?" GMCR's investor day also highlighted several other holes in the bull case:

On the one hand, GMCR reports to have sold 30 million brewers into the at-home market.On the other hand, the company claims that the installed base is only 16 million brewers.According to the company, 26%-32% of all brewers ever sold are not in home use.

Mr.Kelley said many were "gifted to someone else." Insinuating that Keurig brewers are gifts that remain unopened is certainly a unique marketing strategy as the holiday season approaches, but it reinforces the small-installed-base narrative which in turn supports the bull story in two ways: (1) it implies that the market is not saturated; and (2) it counters analysis that would otherwise indicate that the attachment rate (estimated daily K-Cup consumption per brewer) is falling. Setting aside the implausibility that so many brewers are purchased and not used (our work suggests that many were never sold in the first place), if millions of machines are sitting in people's closets (who are unlikely to be future customers), the market is far more saturated than the company would like everyone to believe.

In 2011, GMCR management claimed that it couldn't meet K-Cup demand due to inadequate manufacturing capacity. At the investor day, management stated that it is using only 31% of its capacity in 2013 and used only 29% in 2012. We can't reconcile the company's prior claims that it had a capacity shortage in 2011 and then had so much surplus capacity in 2012. Between 2011 and 2013, gross margins expanded from 38% to 42%. How can a manufacturer go from operating at full capacity to only a small fraction of capacity without underutilization negatively impacting margins? Moreover, 31%utilization implies that GMCR has enough capacity to meet its growth plans through the balance of the decade. Given this, it is hard to see why management plans to add even more. Though GMCR has reduced its capital spending this year, it still spends almost twice as much as other packaged goods companies, and indicated that next year capital spending "might tick up a little bit" more than sales.

GMCR intends to offer a new brewer in late 2014. Management said that the new brewer would be able to discern between GMCR-manufactured K-Cups and non-licensed K-Cups,hinting that this would allow the company to close the system and freeze out the non-licensed competition. Management declined to say whether it would program the new brewers to do so. Closing the system would enable GMCR to reclaim its monopoly priceson K-Cups. However, it is unclear why consumers would want to switch from an opensystem offering more choices and lower prices to a closed system with fewer choices andhigher prices. Last year, GMCR introduced the Vue, which flopped for this reason.Conversely, if the new machine does not close the system, then it won't change thecompetitive dynamic where K-Cups are becoming commoditized and GMCR is losingmarket share. We have heard that GMCR management privately suggested to one largeshareholder that the threat to re-close the system is a bluff to try to convince competitors to become licensed partners.

From David Einhorn’s Greenlight Capital third quarter 2013 letter.