After 21Vianet (VNET) completed its IPO at $15 ($2 above the high end of the proposed range) and its shares jumped to over $21 on their first trading day, I wrote a brief article wondering if 21Vianet's multiples were sustainable.
To my surprise (I'm really very bad at predicting short term stock movements), the stock declined since its peak and is now trading closer to its IPO price, at about $ 15.50.
The article, however, attracted several negative comments, so that it might be useful to deepen the analysis and try to better explain my position. I'll do it both through a check of the Chinese Internet market and current stock valuations, and by a deeper look at the company's fundamentals, in a subsequent article.
Some readers correctly mentioned, in the comment section, other Chinese Internet stocks that enjoy very high multiples right now.
The landscape for these stocks is indeed interesting.
As Screener.co underlines in today's article on Seeking Alpha, there are several examples of companies achieving very high EV/Revenue multiples, with YouKu (YOKU) getting the leading place in this kind of analysis:
At an enterprise value of almost 100x its revenue, YouKu (YOKU) is the clear leader of the pack in terms of valuation multiples. It did grow revenue 152% YoY, but the "Chinese YouTube" faces high video streaming costs just like its U.S. counterpart.
While the company's gross margin has been increasing, it was a mere 33% last quarter. That compares with a gross margin of over 65% for Google (GOOG) and over 72% for Baidu (BIDU). Google was scrutinized for paying $1.65B for YouTube back in 2006, while investors are valuing YouKu at an enterprise value of a whopping $5.8B less than five years later. It is informative to compare the aggregate global traffic of YouTube and YouKu; while Alexa reports ~27% of global Internet users visit YouTube, only ~1.7% visit YouKu. Furthermore, the Western traffic of YouTube likely monetizes better than the East Asian traffic of YouKu.
Screener.co defines this as “a strange phenomenon where a number of high growth tech companies in emerging markets are trading at sky high valuations,” and goes on examining Baidu multiples as opposed to Google.
While Google's EV/Revenue ratio is 4.5x, Baidu's is 36.2x; that is quite a premium! In fact, Baidu's enterprise value of $50.4B is more than 1/3 the enterprise value attributed to Google ($141.8B) despite Google posting more than 24x the revenue in the last full year ($29.3B of revenue relative to Baidu's $1.2B). Again, nobody doubts that Baidu is a high-margin business that leads its category in a rapidly growing emerging market... but at what price is it an attractive investment?
There is no doubt that 21Vianet is now enjoying similar market expectations.
As the leader in the network-neutral Chinese data center sector, the company is positioned to enjoy great growth in the next few years. The IPO filing reports an interesting forecast for its market:
While carrier-operated data centers historically have held dominant positions in the data center industry in China, the demand for carrier-neutral data center services has experienced significant growth. According to IDC, in 2009, carriers occupied 64.9% of the data center services market, while carrier-neutral data center providers shared the remaining 35.1% of the market, an increase from 32.1% in 2008.
According to IDC, the total data center services market in China was US$667.1 million in 2009, a 22.7% increase over 2008, and is expected to reach US$1.9 billion by 2014, representing a five-year CAGR of 23.8%. The following chart sets forth historical and projected revenues of the data center services market in China:
(Click chart to expand)
The market is always right, so Chinese Internet stocks might be entitled to different (much higher) multiples than their western counterparts, right now. This situation, however, might evolve every trading day, and is probably justified, in part, by higher growth expectations. It's a bit hard, in 21Vianet's case, to justify much higher multiples when the forecasted growth, for this specific sector, is very similar to the North American market (expected to grow at around 20% going forward) or even inferior to the average Asia Pacific growth expectation for next few years.
Going back to 21Vianet's specific data, it may be interesting to look at its 2010 acquisition of Beijing Chengyishidai Network Technology Co., Ltd. and Zhiboxintong (Beijing) Network Technology Co., Ltd. (CYSD) and (ZBXT), or the Managed Network Entities, as an example of today's “exuberance” valuing these Chinese Internet companies.
On September 30, 2010, 21Vianet acquired a 51% equity interest in CYSD and ZBXT, with the option to acquire the remaining 49% equity interests in the Managed Network Entities before December 31, 2011.
On page 60 of the IPO filing we have more details about the deal:
As part of our business expansion strategy to expand our managed network services, we acquired a 51% equity interest in the Managed Network Entities in September 2010, with an option to acquire the remaining 49% equity interest before December 31, 2011. We have paid RMB50 million [i.e. $7.6 million] consideration to the seller of the Managed Network Entities and agreed to pay additional contingent consideration based on certain financial and operating metrics of the Managed Network Entities. We currently intend to exercise the option before its expiration to acquire the remaining 49% interest in the Managed Network Entities. According to the purchase agreement we entered into in September 2010, in order to acquire the remaining 49% equity interest, we will need to pay consideration determined using the proportionate amount of the finalized cash consideration for the initial 51% acquisition. We intend to pay cash consideration ranging from RMB60.0 million (US$9.1 million) to RMB73.5 million (US$11.1 million). In addition, we plan to issue 9,665,540 shares to 11,835,360 shares based on the 2011 operating results of Managed Network Entities to the management of Managed Network Entities under our 2010 share incentive plan. We are discussing with the sellers of the Managed Network Entities with respect to the timing, forms and final terms of the payment.
In a different part of the filings, there is also a mention of an option to acquire shares of the company as part of the transaction:
The Company issued an option to the Seller to acquire RMB25,000,000 [i.e. $3,8 million] of its ordinary shares at a fixed exercise price of US$8.61 per share, exercisable at any time through June 2012.
Keeping it simple, the price for the Managed Network Entities is about $ 18.7 million, plus some options. In an attempt to quantify the cost, the company is indicating a “total purchase consideration of RMB172,439,000 [i.e. $ 26.1 million]”.
On page 73 of the IPO filing we find more info about the revenues derived from the acquired companies:
We started to recognize revenues from the Managed Network Entities in the fourth quarter of 2010, and recognized RMB60.2 million (US$9.1 million) during the quarter.
Total revenues for Q4 2010 were RMB197,312,000, or about $29,9 million.
The acquired entities represented about 30% of total revenues.
Some readers will criticize my approach, but I can only congratulate a company that spends about $20 to $25 million to buy two companies achieving combined revenues totaling 50% of its own (valuing them less than 1 time their annualized Q4 revenues), and succeeds in a US IPO valuing the combined company almost $1 billion (or 8 times Q4 annualized revenues), just 6 months afterwords.
It definitely used very conservative multiples while buying, and enjoyed rich valuations when floating the company.
In my next article, I'll try to analyze more specifically 21Vianet's data, and explain why a company with only 2,645 cabinets hosted in its own data centers may be considered a little overvalued, at least in my humble opinion.