THE MAGAZINE FOR FINANCIAL DIRECTORS AND TREASURERS
  Home | Free email newsletter | Site map | Contact us 
 

CFO PROFILES June 2002

MICROSOFTER...OR NOT
As Microsoft's growth slows, CFO John Connors tries to keep its profits up.
By Kris Frieswick

When John Connors became CFO of Microsoft in January 2000, the job was still a pretty sweet gig: sweep up the mountains of revenue generated by Windows and Office software, report fabulous earnings growth every quarter, do deals with the ever-more-humongous cash hoard, and laugh all the way to the bank. An exaggeration? Maybe, but at the turn of the millennium, the CFO spot must have felt like a nice reward for a 40-year-old CPA from Montana who had paid his dues for 11 years at eight different jobs in just about every department in the company.

John Connors has bad timing. That's because he became CFO just as it started to turn into one of the most challenging jobs in the company: keeping returns high in a depressed technology market while Microsoft completes the transition from headstrong adolescent to Fortune 100 adult. It's Connors who now has the task of persuading investors that shares of the US software giant are still worth the heady multiples of a growth stock, despite evidence to the contrary.

Start, for example, with Microsoft's revenue growth rate in fiscal 2001 - a paltry 10 percent, down from 16 percent a year earlier. Over the same period, earnings declined 23 percent, thanks largely to a US$4.8 billion impairment charge on investments. Cost of revenue increased to 17.5 percent for the first six months of fiscal 2002, up from 13.7 percent during the year-earlier period. (Microsoft's fiscal year ends June 30.) Even with last fall's successful launches of Windows XP and Xbox, revenues were up only 12 percent for the first half of fiscal 2002, compared with the first half of the previous fiscal year.

The market for its primary revenue driver, the Windows operating systems for personal computers, is saturated, and PC market growth has slowed dramatically - down 5.2 percent worldwide in 2001, and predicted to grow only 3 percent in 2002. And although the company says that the introduction of Windows XP in October 2001 was the most successful launch of any of its operating systems, a big part of XP sales was attributable to existing Windows customers rushing to upgrade to the latest operating system in time to qualify for a new software maintenance program.

Then there's that pesky antitrust case, whose settlement is still pending. Microsoft faces potential penalties that may change the way it sells its flagship Windows product, and could alter the way it is allowed to pursue new markets going forward. All in all, since late 2000 declining growth and increasing uncertainty have pushed Microsoft's stock down to around US$60, a level not seen since 1998 and half its December 1999 peak.

That's not to say the company is in trouble - far from it. Its cash hoard now tops US$40 billion, and it has nary a penny of debt. Although slowing, Microsoft's revenues are still at record levels. And the stock currently sells at 55 times earnings, a 100 percent premium over the Dow Jones Technology/Software Index. But the go-go growth of the mid-1990s is a thing of the past. Chairman Bill Gates and CEO Steve Ballmer know that if the company wants to retain anything near that level of growth, it is going to have to find new markets.

"When you have US$27 billion in revenue," says Robert Herbold, COO of Microsoft from 1994 until last year, "your ability to grow by whopping percentages is less than when you're a US$5 billion company. Every business has to assess its challenges and think, where do we go from here?" Joseph Beaulieu, senior analyst with US bond ratings agency Morningstar, agrees. "In the past the company had the luxury of milking the cash-cow PC business. They had the luxury of doing things that were far-forward-looking," says Beaulieu. "Now they have to focus on making their numbers from quarter to quarter."

Right Man, Right Time

Microsoft hopes Connors, its fourth CFO, is the right man at the right time. He shrugs off the notion that his challenges are harder than his predecessors' were. "Any job you go into, you've got to deal with the hand you're dealt," he says.

Microsoft's previous CFOs played with very different hands. Its first CFO was Frank Gaudette, the self-described "grandfather" of Microsoft from 1984 to 1993, who guided the company through its initial growth spurt and IPO. Gaudette's successor, Michael Brown, took over just as revenue growth started to slow, from an average of 45 percent between 1988 and 1993 to 33 percent between 1993 and 1997, when Brown retired. Among his accomplishments, Brown imposed more discipline on the company's internal processes.

After Brown came Greg Maffei. An ex-investment banker and consummate dealmaker, Maffei was a logical choice during a red-hot stock market. In 1994, as Microsoft's treasurer, he came up with the idea of selling puts against the company's stock to offset the costs of its aggressive employee stock-option program. The bull market kept the puts out of the money, and Microsoft reaped more than US$2 billion in premiums between 1994 and 1999. As CFO, Maffei spent more than US$20 billion on private and public investments. By the end of fiscal 2000, the company's stock and warrant portfolio was valued at 68 percent more than Microsoft paid for it.

But Maffei was unable to secure the bigger operational role in Microsoft that he reportedly wanted, and he left in December 1999 to become CEO of 360networks. Connors, then head of Microsoft's Worldwide Enterprise Group, became CFO. A few months later, the stock market and the economy headed south, and in fiscal 2001 Connors had the unhappy task of writing off US$4.8 billion in impaired investments. That kicked income into negative-growth territory for the first time in company history. (He wrote down another US$1.95 billion in the first six months of fiscal 2002.) Connors was also forced to dismantle the put program, at a cost of US$1.4 billion.

In short, Connors's first two years on the job have largely involved cleaning up somebody else's messes, an observation he characteristically dismisses. "It's just part of the job that some things that you get are good, some things that you get ain't so good," says Connors. "You just have to work through them," he adds.

Mr Fix-it

This low-key, intensely practical approach is Connors's stock-in-trade. Those around him say he has a flair for boiling a problem down to its lowest common denominator, getting everyone to understand its scope, then setting out to methodically correct it.

As corporate controller between 1994 and 1996, Connors helped modernize the company's antiquated financial reporting system - at the time the subject of some internal embarrassment. "When I came to Microsoft [in 1994], I was shocked how far behind we were as a company using our own technology," recalls Scott Boggs, who was assistant corporate controller under Connors and is now vice-president and corporate controller. Working with Connors, Boggs spearheaded the creation of a state-of-the-art web-based financial system that provides near-real-time access to a variety of financial metrics.

Connors continued his efficiency campaign when he took over as CIO in 1996. "We were in the embarrassing situation of not having a good set of applications to run the company, and some [applications] weren't running on Microsoft software," says Herbold. "We wanted to make Microsoft a showcase relative to its systems," he says.

In mid-1999, to prepare for the future, Microsoft unveiled a new organizational structure built along customer segments instead of product lines - thus taking the focus off of individual products, like the slower growth Windows. Considering the slowing PC market, the company began seeking entry into new markets, such as computer games and game hardware, software for personal digital assistants (PDAs) and cell phones, and, most important, a new Internet application platform - .NET. To fund these investments while keeping operating profits up, Microsoft had to become leaner and more efficient.

Indeed, managing costs has always been a personal crusade for Connors. Microsoft avoided layoffs and was able to continue to increase salaries, he says, thanks in part to "the extreme focus we've had on gaining cost efficiencies." He claims that to date in fiscal 2002, this focus has saved more than US$600 million against budget amounts.

Saving the Margins

As a result of his cost-cutting evangelism, Connors routinely receives unsolicited ideas from employees via email. He proudly recounts one initiative that came after he delivered a rousing talk at the administrative group's annual conference. "One [administrative assistant] went back to her [R&D] group, and convinced the execs to cut their lab spending by US$13 million," he says.

Procurement is another focus for Connors, who as corporate controller helped replace a paper-based invoice and accounts payable system with on-line procurement software. In one of his first major initiatives as CFO, Connors threw his weight behind an aggressive procurement program that standardizes purchases across the organization, finding incremental savings where it can.

He set fiscal 2002's savings target at US$250 million, a sum that may seem small for a company with US$40 billion in cash and a dominant software franchise. But analysts say Microsoft's move into computer games, via its Xbox game console and gaming software, makes Connors's cost-cutting focus a crucial part of the company's long-term vision. "They've never before been a hardware maker of any real scale," says Mark Specker, a managing director of research at US-based SoundView Technology. And while the hardware business may drive top-line growth, its margins are much lower than software's. "John's goal has to be to take a piece of the body blow that [hardware] inflicts on net margins and to do his part to add as many basis points as he can," says Specker. "The only line on the income statement he has to play with is general and administrative, because they don't want to mess with sales and research."

So far, Connors has played well. General and administrative costs for the first half of fiscal 2002 were US$1 billion, including US$660 million to settle consumer class-action suits. Less that holdback, G&A spending was 2.6 percent of sales, compared with 3.1 percent for the first half of fiscal 2001.

Unlike Maffei, Connors isn't a deal guy, nor does he wish to be. In November 2000, Connors recruited investment banker Richard Emerson of US-based Lazard Freres to take over investment and acquisition deal-making. And as the stock market has fallen from its late 1990s peak, so has the pace of Microsoft's dealmaking, from US$20.7 billion in 1998 and 1999 to US$8.6 billion in 2000 and 2001. Most investments today are made to gain entry into established markets where Microsoft isn't a serious player.

Recurring Complaints

About the only thing that Connors catches flak for is the company's efforts to free revenues from dependency on the PC market, by changing from a perpetual to a recurring software licensing model. Under the old model, enterprise software customers bought Microsoft software (usually Windows and Office) with a perpetual-use license when they bought PCs or upgraded. When new versions came out, customers could either buy the upgrade for 50 to 70 percent of the full purchase price or just keep using the older version. Under a new program, Microsoft will allow all enterprise software customers to join Software Assurance, an ongoing software maintenance program. The program gives participants ongoing upgrades as they become available, but the annual fee is 25 to 29 percent of the cost of a full license.

Designed to produce a stable revenue stream, the recurring model is "very similar to the software licensing model that other large enterprise providers have always had, whether it's Oracle, SAP, et cetera," points out Connors. "The difference is that we had a broad range of other ways you could buy our software, many of which proved to be quite popular. So for our move to a larger percentage of subscription-oriented licensing [to succeed], the first thing is: you have to have value. If subscribers don't think there's value they won't upgrade, and the programs will be a disaster," he says.

Microsoft had originally given customers until October 1, 2001, to upgrade to the latest versions of Windows and Office in order to be eligible for the new program, or face paying full price for the newest versions after that date. But D-day was rolled forward to this summer after considerable customer outcry. Customers complained that Microsoft wasn't giving them enough time to prepare for the new plan, and some, especially those customers who don't upgrade frequently, continue to grumble that the program will increase their software costs.

The company is also unveiling a licensing model in which customers "rent" enterprise software licenses for three years, but then must renew licenses when they end. Analysts are still debating whether or not Microsoft's new "subscribers" will choose to participate in either new program, and whether the programs will have the hoped-for effect on revenue flow.

A possible hitch with Software Assurance and rental models is that they will increase the company's unearned revenues, which stood at US$6.6 billion as of December 31, 2001. Those revenues are already under scrutiny by the US Securities and Exchange Commission (SEC). Connors insists that Microsoft recognizes the unearned revenue in later periods in full accordance with GAAP, but the SEC is investigating whether the revenues served as a "cookie jar" that the company used to smooth earnings in the mid-1990s. In 1999, the company's former head of internal audit alleged in a wrongful termination lawsuit that such smoothing occurred, but the suit was settled out of court, and the SEC has yet to rule on the allegations. (A recent US$1 billion restatement of first- and second-quarter fiscal 2002 unearned revenue - necessitated because of a "clerical error" - would seem to undercut Connors's claim that Microsoft's accounting practices in this area are scrupulous.)

"[Microsoft's] policies regarding revenue recognition are thoroughly documented and remain consistent as much as they can," says SoundView's Specker. "If you document it and stick to it, then you can't be accused of changing the rules because you needed the extra revenue," he says. "But the goal of a P&L is to provide a current view of changes in business conditions," he adds. Even if Microsoft consistently applies its policy for recognizing unearned revenues, he says, the view afforded by the income statement is distorted.

Connors scoffs at such criticism. "I think the most important thing about this place is that a long-term view is taken," he says. "Bill (Gates) and Steve (Ballmer) pay virtually no attention to quarterly financial results. We set an annual plan for the company; we then set the expectations of the annual plan with the investment community." .

Hold or Pay?

Microsoft has also come under fire from the likes of Ralph Nader for amassing US$40 billion in cash instead of paying a dividend. Connors says the board of directors periodically reviews the question of paying a dividend, and "when the board, and our executive team, feel like the best way to maximize shareholder value is a payout, we will do that. But if you look at where we are today, we face big risks, big bets. And while US$40 billion is a substantial amount of money, given the current environment I think it's kind of an odd thing for your company to get beat up for having a healthy balance sheet."

Healthy, indeed. Analysts expect Microsoft will hold on to the cash until the antitrust penalty phase is complete, then go on a spending spree to buy into more markets, create more partnerships, and push shared technology platforms with other technology providers.

But the old days when Microsoft could conquer markets without much help are gone, so the company has a lot of fence-mending to do. Last March, CEO Ballmer said that Microsoft knows that it has to be "a respectful, open and appropriate competitor." For the first time, the company started opening parts of its software source code to competitors. And a recent company-wide memo from Gates refocused Microsoft's efforts on improving the security of its products - crucial to acceptance of its new .Net platform for Internet-based applications.

Who better to represent a more open Microsoft than John Connors? He's earned respect inside the company that is matched by praise from analysts. He's "the Eagle Scout of CFOs," declares David Readerman, at US merchant bank Thomas Weisel Partners.

Connors's role model is his father. Dan Connors coached high school basketball and football until John became a freshman, then quit so he wouldn't be John's coach. "He could have stayed another year or two [and broken some records]," recalls the son. "But he left the new coach with a great hand - and that's what I would like to do." He may succeed. But given the profound changes facing a maturing Microsoft, it could be a very different hand from the one Connors now plays.

Kris Frieswick is a staff writer at CFO in the US, CFO Asia's sister publication.