Oracle's Sun Deal: A Closer Lookby
Oracle's Apr. 20 conference call with Wall Street analysts to discuss its $7.4 billion buyout of Sun Microsystems (JAVA) was a straightforward affair. Oracle (ORCL) CEO Larry Ellison and his lieutenants ran though a quick synopsis of how they'd use the hardware and software assets they're buying, and Sun Chairman Scott McNealy read his canned statement in the most perfunctory manner possible. The companies brooked no questions, and analysts were off the phone in 15 minutes.
What remains are plenty of unanswered questions for investors and customers about how Oracle plans to turn Sun from a money-losing entity into a profitable asset. Deep job cuts seem inevitable; less clear is what role, if any, Sun's honchos will play at Oracle. Oracle is saddled with computer servers built on Sun's outdated Sparc chips, though the company hinted at plans to combine its software and Sun's hardware in new ways. It's also uncertain what Oracle's ownership of the widely used Java programming language will mean for the rest of the computer industry, particularly IBM (IBM). Herewith, the five biggest outstanding questions about the deal—and how they might be resolved.
Will Oracle sell Sun's hardware business? One of the biggest questions surrounding the deal is what Oracle will do with Sun's $9-billion-a-year server and storage business. Many observers expect Oracle to sell it to Fujitsu (FJTSY), IBM, or another computer maker. Sun holds just 4% of the server market, and most of its revenue comes from servers built with Sparc chips, which customers have rejected in favor of often cheaper industry-standard servers based on chips from Intel (INTC) and Advanced Micro Devices (AMD).
Slugging it out in the server business would also mar Oracle's healthy 46% operating margin. And a hardware foray by Oracle could push competing computer makers Hewlett-Packard (HPQ), Dell (DELL), and IBM into the arms of Microsoft (MSFT). Oracle would rather those companies sell machines that house its software. "I'd be surprised if Oracle tried to slant their software toward the hardware they just acquired," says Tod Nielsen, chief operating officer at VMware (VMW). Besides, says Gary Scholten, senior vice-president and chief investment officer at Principal Financial Group (PFG), the two companies already cooperate to ensure Oracle's software works with Sun's hardware. "They've done so many things so closely over the years, in some ways [the deal] shouldn't have been as much of a surprise as it seemed," says Scholten, whose firm manages retirement funds and pensions.
In the event Oracle keeps Sun's hardware business, it could try to bolster profit by shedding low-margin parts of Sun's server line and keeping more powerful computers with fatter margins. Pacific Crest Securities analyst Brent Bracelin estimates that Sun generates about $3 billion a year from supporting its Solaris Unix operating system. Most of that is tied to Sparc servers. And there's not a lot of expense associated with Sparc, since Sun outsourced its design and manufacturing.
Oracle executives told Wall Street they plan to engineer the company's software to work well with Sun equipment and offer specialized servers for industries, including banking and telecom. Oracle machines could be attractive to CIOs looking to pare down their list of suppliers.
How will Oracle steward the Java programming language? On the day he announced the deal, Ellison called Java "the single most important software asset we have ever acquired." It could be a prophetic statement, since Oracle's ownership of the programming language, a computer industry standard, would have wide-reaching effects.
Java runs on 800 million PCs and 2.1 billion cell phones, and Oracle hasn't been shy about raising prices on its products. That means it could pump up Java revenue from the $220 million Sun booked in fiscal 2008 to perhaps $1 billion over time, says Citigroup (C) analyst Brent Thill. Controlling Java could also help ensure that customers of the dozens of companies Oracle has acquired since 2005 have a smooth path to new versions of Oracle's applications written in the language.
Perhaps more important, Oracle would hold more sway over a foundation of much of the world's business software. Owning Java gives Oracle more components of the software "stack" that companies use to create applications, including the programming language, Java middleware, and Oracle's market-leading database. The combination would give Oracle's salespeople more ammunition against Microsoft. And while much of IBM's software is currently based on Java, Oracle could impose restrictions on its rival's use of the language in future products, says Forrester Research (FORR) analyst James Staten.
Will MySQL stay as compelling? Sun claimed more than 11 million users of its MySQL database, which it bought for $1 billion in 2008. Users include Internet powerhouses Google (GOOG) and Facebook, which favor the software's facility with Web applications. Taking ownership of the open-source software could give Oracle an entrÉe to Web 2.0 companies.
MySQL also could bolster Oracle's ties to the software developer community, helping it steal programming thunder from Microsoft's SQL Server. "Microsoft has always understood how to deal with developers and has good relationships with them," says one industry executive close to Sun. "Oracle doesn't have the same reach into the most innovative developers in the world. With MySQL, they get there."
Unfortunately for Oracle, most of MySQL's users don't pay for the software, opting for a free, unsupported version of it. Sun booked just $208 million in MySQL sales in fiscal 2008, vs. Oracle's $4.8 billion in expected database bookings for the fiscal year that ends in May. And Oracle's flagship database will pull along an additional $7.6 billion in support revenue. So MySQL could be more important strategically than monetarily to Oracle. "It was a huge source of pain at the low end of the market that they now don't have to worry about," says Citigroup's Thill.
How many Sun workers will lose their jobs? To get from Sun's $1.8 billion in operating losses for the six months ended Dec. 28 to Oracle's projected $1.5 billion in first-year profits, the cuts will have to be deep. Thill estimates that 40% to 70% of Sun's 33,000 employees will get the ax, the kind of massive cutbacks that Wall Street has wanted and that McNealy and Sun CEO Jonathan Schwartz resisted. "Sun is considered to have one of the most bloated cost structures in the tech industry," Piper Jaffray (PJC) analyst Mark Murphy said in an Apr. 21 research note.
Sun's Schwartz and McNealy are unlikely to take management jobs at Oracle, though McNealy could be a candidate for a board seat. He's only 53, close to Ellison's age, and deeply interested in the fate of Sun's technology. Ellison and McNealy had talked many times over the past month and a half and hashed out many details of the deal without the aid of bankers, according to a person familiar with the proceedings.
Schwartz is another story. He's young, a broad thinker, technically and financially savvy, a great communicator, and will likely be able to write his own ticket after selling troubled Sun at a healthy premium for investors. Sure, he's weathered criticism for giving away software and not taking a bigger whack to costs. But Schwartz could be a boon to a company with fewer strategic and operational problems than he inherited at Sun.
Is a culture clash coming? Oracle's take-no-prisoners sales tactics and bare-knuckled approach to career advancement don't appear to mesh with Sun's positive, can-do engineering culture. Yes, Schwartz and McNealy carried a certain swagger and ego, but Sun has always been about pushing technical boundaries. Oracle these days thrives on aggressive sales and smart financial bets.
Yet Sun's operations will likely benefit from Oracle's approach. "Sun was not a well-managed company and needed more discipline," says an industry executive familiar with its business. "They now will get it. If you don't take the medicine, someone will force it down your throat."