Would an Oracle Deal to Buy EMC Pass the Success Test?

On Thursday, The Wall Street Journal reported on speculation that Oracle may be considering a purchase of EMC, which saw its stock rise 4.48%, likely due to investors' belief that the deal may actually happen. The Journal quotes Wedbush analyst Kaushik Roy who puts the probability of an acquisition at 25%, noting that the price tag for the data storage giant would be $50 billion.

There has been a significant amount of merger activity lately in the external data storage industry. Hewlett-Packard (HPQ) just bought 3Par for $2.35 billion after a bidding war with Dell (DELL), and IBM (IBM) inked a contract to buy Netezza for $1.7 billion. But Roy thinks it would be better for Oracle to acquire smaller companies such as NetApp (NTAP) or Brocade Communications (BRCD), according to the Journal.

I disagree: Oracle (ORCL) should acquire EMC (EMC), if only for its 85% stake in VMWare (VMW).

EMC consists of two businesses -- a thinly profitable external data storage product line and a more profitable virtualization software business that helps companies get more data storage and retrieval power with less hardware. If Oracle buys EMC, it should keep the software -- EMC's VMWare stake is worth $28 billion -- and spin off the less profitable hardware for its estimated $28.7 billion value.

The Four Tests of a Successful Acquisition

On Sept. 22, I assigned my business students at Babson College the task of analyzing whether an Oracle/EMC merger would make sense. I've been teaching them that most acquisitions fail. To be one of the minority that succeed, a deal must, at a minimum, pass four tests:

  1. Industry Attractiveness: Does the industry have high profit potential?

  2. Better Off: Will the combined companies be able to get a significant share of that industry due to their combined skills?

  3. Integration: Can the two companies merge their organizations smoothly so the deal is seamless to customers?

  4. Price: Is the price paid for the target low enough so shareholders will get a return on the investment?

EMC, as I just noted, works in two different industries: external data storage and virtualization software. The virtualization software industry is attractive, while the external data storage industry is not. This suggests a number of options for Oracle:

  • Not acquire EMC, but partner with it to get a better deal on its products when Oracle brings in a customer that wants what EMC sells;

  • Acquire EMC and keep it as is;

  • Acquire EMC and sell or spin off the hardware side of the business;

  • Acquire EMC and sell or spin off the virtualization software side of the business.

In applying these four tests to the acquisition options, my conclusion is that the deal could pass all the tests -- depending on how Oracle splits EMC up, the industry is attractive, the two companies could be better off, and the integration of the two companies would likely go well. But given the $50 billion price tag, it might be hard for to pass muster. Let's look at the details:

Industry Attractiveness: Pass/Fail.

The data storage industry's five-year average return on equity (ROE) is much lower -- 10.3% -- than that of the average industry in the S&P 500: 16.1%. Although the technical and system software industry where VMWare competes is more attractive -- it earns a 16.4% ROE, the application software industry to which Oracle belongs earned a 27.3% ROE. And at its 23.5% annual rate, EMC has been growing more slowly than VMWare's 48%. In short, data storage is not an attractive industry to get into, while virtualization software is more so -- although both are less attractive than application software.

Better Off: Pass.

Success in the data storage industry depends largely on aggressive sales and innovative product development. Both EMC and Oracle are known for hardball selling to companies, and EMC's products have kept pace by helping customers save money. Most importantly, combining the companies would give Oracle/EMC an even greater market share lead in external data storage: Sun Microsystems had 5.2% of the market in 2009 before Oracle bought it, while EMC is already the market leader with 25.7%. And, if properly managed, the two companies could sell to each other's customer bases.

Integration: Pass.

Oracle has been making acquisitions for the last several years and it continues to generate excellent financial results. This suggests that Oracle knows how to integrate its acquisitions well. But an EMC acquisition would pose a big challenge because both companies are known for aggressive sales so Oracle would need to make EMC's sales force an offer they couldn't refuse.

Price: Incomplete.

Oracle's market capitalization of $142 billion suggests that it could acquire EMC for $50 billion, a 15% premium over its current $43.6 billion value. Oracle would probably use a mixture of stock, debt, and cash. The big question is whether the additional profit generated by the deal -- when taking into account cost savings -- would more than exceed that $50 billion price tag. To defray some of that purchase price, it might make sense for Oracle to keep VMWare and spin off EMC's hardware business -- a sale of 80% of which could yield around $28.9 billion. (I derived that figure by assigning a data storage device industry price/sales ratio of 3.06 to its $11.8 billion in 2010 hardware sales, which I estimated by doubling its first half $5.9 billion in information storage sales.)

Many more questions will need to be answered before it becomes clear whether this deal will happen, and if so, what form it will take. But one thing seems clear: Consolidation in the data storage industry is inevitable, and there may still be opportunities for you to profit from it.