Digital Realty Trust: Bound for a Correction

Digital Realty hitting the hard reality

“The true art of memory is the art of attention.” (Samuel Johnson)

Remember datacenters? In 2000, many of these operators crashed with the tech bust, but like the phoenix they made have made a comeback again with the real estate bubble.

One of them in particullar, Digital Realty Trust (DLR), is still flying high.

At about $31 per share, it is about 40% down from its all time high of $50 and has rebounded from its November 21 low at $18. It currently trades at:

-142 times TTM earnings

- 15.1 times EBITDA (or a cap rate of about 7%)

-2.7 P/B

-8.1 X TTM Revenue

At these lofty prices the stock is, in my opinion, priced for perfection and bound to disappoint. Below are the four reasons why I think this stock is bound for a correction:

Primo: Sub-market cash returns on declining asset values. It generates a value destroying, miniscule, return on book equity of about 1%, and is paying a dividend yield of 4%. It’s like getting paid 4% on a bullet that will return less than the principal. For this type of return you’d be better off buying 10 year bonds and getting close to 3%.

Secundo: increasing costs of capital. It’s a capital intensive business (structured as a REIT) which requires constant tapping into the capital markets for debt and equity. Investment banks and their analysts love it, but investors are much less receptive now and with no appetite recovery in sight. Debt refinancing is much more expensive and equity issuances are more dilutive. This means less growth and lower earnings per share in the futures.

Tertio: Vulnerable client portfolio. Looking at their client portfolio, the largest client responsible for 10% or so of rental revenue is Savvis a $300 million market cap tech company providing “cloud” applications. Not a rock solid company by any means, but it does not seem to be in immediate survival danger either. The second largest customer is Qwest (Q), the telecom company formerly lead by Joe Naccio; the third, Equinix (EQIX), is another tech company, with a respectable $2 billion market cap, offering products which are rather similar to DLR: connectivity and collocation, etc. The fourth largest customer, called Telx Group is a company also offering fairly similar collocations services controlled by private equity firm GI partners. DLR is a portfolio company of GI partners and the CEO of GI partners is the Chairman of DLR, (interesting…). About 15% of the business comes from Silicon Valley.

Bankruptcies, consolidations and business relocations are likely to impact the portfolio, as does the fact that many of their clients have businesses that are remarkably close to DLR, and in many cases in addition to leasing from DLR, they own their own datacenter facilities, so they can always build and operate their own centers if they get cheap enough.

Quatro: inflation risk. The average life left in the leases is about 7 years. According to their filings “many of the leases provide for fixed rate increases”- many does not mean most, and a fixed rate increase means many that are not indexed to inflation.

Conclusion: In my view this Phoenix will soon turn into…an Icarus.

(The phoenix from the Aberdeen Bestiary.)

Disclosure: short DLR.

Comments

[bemink:]

I am in this industry everyday, through the good times and through the bust period, so lets try to correct a few inaccuracies....

1) The biggest customer is "Savvis", not Navvis. In addition to cloud computing, they also perform managed services, apllications and do some systems intergration. The do not compete with DLR.

2) One of Digital Realty's (DLR) largest investors is Global Innovation Partners or "GI Partners", not GL. GI also owns Telx. Telx main business is cross-connections and access to multiple networks. They also offer colocation. They do not compete with DLR, and actually enhance DLR's offerings.

3) Qwest is a customer of DLR's and recently paid DLR $14m to get out of a Chicago lease in a building that DLR didn't have any space to offer. Now DLR will build out space in a very hot market and offer at a higher rent. DLR will start with the $14m they just received as a termination payout....not a bad deal. Also, Nacchio has not been associated with Q for many years (7+?), so not sure why the reference is here...too many other things to discuss with Q.

4) As a former employee of Equinix, I can tell you that EQIX does not offer the same services as DLR. If you knew this industry, this is probably the biggest stretch in this slanted article. EQIX offers managed colocation, the best "peering" in the world and at a level no other firm can do. EQIX and DLR do not compete.

5) The comment "so they can always build and operate their own centers if they get cheap enough" is out of touch with the current capital markets. This facilities are not cheap ($1000-1500 per sf). Not many firms can secure this much capital, and at a reasonable price. DLR secured a substantial amount prior to the current liquidity/capital crises. This crises will actually drive more business to DLR and others like it due to a lack of reasonably priced capital to build these facilities. Also, unlike 2000-2003, companies with data centers are not going bankrupt and putting facilities on the market for pennies on the dollar. This is because this time around, these companies have many paying customers and have occupancy levels that are in the 50-70% average, and need only approx. 32-35% to reach CF break-even. In the current environment, most of these firms offer more than basic colocation, thus layering services, peering and applications and making more per sf, and per customer.

Although the current markets keep many industries in turmoil, this data center/managed services space is one that is somewhat insulated because the investment is high, leases long and operations critical. I would rate DLR as an outperform in the sector, as they are the leader in the industry. At $51, I would also rate EQIX a buy. Q would be a hold and SVVS a hold or sell.

Disclosure: Long on EQIX, Q, SVVS. Do not own DLR currently, but have been long in the past - owned from $18 to $42. If it goes down much lower, I will probably buy it again as I like their model and how conservatively they are approaching the current market.

Feb 16 01:43 PM

[Charlie Bottle:]

Thank you for your comments and additional insights. I'm getting the typos corrected.

Feb 16 02:41 PM