Analysts' Odd Coverage of Colocation Sector: The Not-So Invisible Hand

Includes: DFT, DLR, EQIX
by: Michael Collins

On December 16th, something odd happened concerning analysts' coverage of the co-location sector: 24 hours separated a new BUY to $64 price target recommendation on DLR from Goldman and a Sell to $40 price target sell recommendation by UBS. What was even odder was that UBS had previously pegged DLR with a BUY to $65.

What lay behind the stark UBS call on DLR and DFT? Was there a smoking gun that UBS saw, but the other analysts all missed?

A little background might help shed some light on how this divergence might have come about:

In early October, Equinix (NASDAQ:EQIX) surprised the market when it warned that its third quarter and 2010 revenue will miss expectations. Equinix said third quarter revenue will be $328 million to $330 million, down from its previous outlook (a roughly 2% top line miss). For the year, Equinix projected revenue of $1.21 billion, below the midpoint of the company’s previous outlook.

The company said it cut its outlook because of “underestimated churn assumptions” in North America, as well as “greater than expected discounting to secure longer term contract renewals.”

The stock took a 30% hit the next day and other players in the sector all took heavy losses.

Upon closer reading, it became clear that this was more of a company specific issue; 50 percent of the revenue shortfall was related to lower-than-expected income from the former Switch and Data facilities Equinix acquired earlier this year.

Most analysts took the view that this was a company specific event related to M&A issues. Before the October 21 earnings call, Benchmark Capital wrote:

We expect Equinix will provide more detail regarding the recently announced Switch & Data miss and US churn, both of which contributed to Equinix' reduced 2H10 guidance on October 5. We believe the macro drivers remain intact for the data center industry and that scale creates a competitive advantage. We maintain a Buy rating. We view the 3Q10 lowered guidance as largely Equinix-specific. Half the miss was attributed to Switch & Data, which seems an integration and forecasting miscue. Further description of the churn and pricing outlook is required if investors are to regain confidence in Equinix.

Furthermore, the Equinix earnings call also mentioned that they were losing business to the wholesalers (DLR and DFT).

The Shorts

Certain hedge funds took the Equinix news as a cue to short DLR and DFT. The general theory being that the space was overbuilt and was going to witness a repeat of the millennium collapse. As of December 15th, the hedge funds had shorted:

DLR: 90.4M Float, 18.7M Shares Short, or 20. 7% of the float

Average Volume of 1.66M: = 11.22 days to cover

Earnings Date: Feb 23

DFT: 54M float, 13 M short or 24.25% of the float,

Average volume of 656k = 19.8 days to cover

Earnings Date: Feb 10

These are pretty impressive short positions, even more so when one considers that until the UBS call on Friday, December 17th, no analyst was bearish on these stocks. Is this UBS analyst the Meredith Whitney of the Co-location sector? What great analytical insight could have led to such a drastic change of heart on DLR?

Firing Blanks

What was the smoking gun behind the BUY to $65, No! Sell to $40 call UBS call on DLR?

  • Was there an accounting scandal?
  • A massive unseen drop in topline growth like Equinix?
  • A Technological breakthrough?
  • Some other horror story not uncovered by Goldman Sachs and others the day before?

None of the above. UBS based this sector downgrade on roughly 3 points:

UBS analyst says, "We believe the wholesale data center REITs are trading at unsustainably high valuations, despite slowing growth profiles in 2011 (and beyond), accelerating competition from private players, and pricing pressure in the retail channel...Data center obsolescence risks have continued to rise over the past year, as technological innovation is driving the industry toward more power-efficient designs. With high power costs representing a key concern for data center tenants, we expect this to be an ongoing theme over the next several years..."

The notable analyst call also mentions the excess availability of financing as another reason that would lead to oversupply.

The point here is that there was nothing new, no smoking gun here that warranted the massive re-rating.

Now compare this to the DLR ‘Buy to $65’ by Goldman Sachs, which put DLR on their “America's Convictions List”. The Goldman analyst cited the exact opposite reasons from UBS, so the disparity becomes rather awkward. This is particularly true when you look at all the other buy recommendations and price targets on DFT and DLR.

The Squeeze?

We all know that we live in a world where hedge funds pay a lot of money for research and then even more in sales and trading commissions to large banks. It is not uncommon for hedge funds to exercise their influence over analysts and persuade them to present their thesis in a new piece of coverage. However, there was no smoking gun cited by UBS to account for the massive downward revisions of DLR and DFT. Hence, in my opinion, hedge fund ‘thesis pushing’ is the most credible explanation for the awkward call. It was quite unfortunate timing for the UBS analyst that the Goldman note preceded his call by 24 hours as it simply further exposed his hand. The fact that this happened on quad option witching Friday, where the most damage could be inflicted, only adds credibility to the macabre conspiracy theory. Well, if you buy the conspiracy theory that UBS may have been convinced by certain short funds to revise their ratings on DLR and DFT, and then UBS accidentally released the report on quad witching Friday.

The only conclusion can be that the shorts, lacking anything concrete, are desperate and throwing the kitchen sink at these stocks in an attempt to cover their positions before the earnings calls. Christmas trading was very light, and in the case of DFT, this leaves the shorts a little over one month to cover 13M shares, not a great prospect on a stock with a volume between 650k-800k per day. So if you’d like to start the New Year with a squeeze, this pair looks as good as any.

Disclosure: Long DFT