It's time for consolidation in the semiconductor industry. Whether chip company CEOs recognize it or not, they'd better move aggressively to eat, or they will be eaten.

That's the message many analysts drew from two recent high-profile leveraged buyouts (LBOs) in the semiconductor industry. In August the majority of Philips Semiconductor (now NXP) was purchased for about $10 billion by a group of private equity firms—Apax, AlpInvest Partners, Bain Capital, Kohlberg Kravis Roberts & Co. and Silver Lake Partners. (Just a year earlier, the same group had taken private the chip unit of Agilent Technologies for $2.7 billion.) Then in September, Freescale Semiconductor tentatively accepted an offer to be acquired for $17.6 billion by a private equity consortium comprising The Blackstone Group, The Carlyle Group, Permira Funds and Texas Pacific Group.

With two deals that together amount to investments of more than $27 billion, the industry is rife with speculation about which chip company might be next. Among potential targets are Analog Devices, Atmel, Cypress Semiconductor, Intersil, Linear Technology, Micrel, Silicon Laboratories and STMicroelectronics. Analysts speculate that analog chip companies, with their high margins and hoards of cash, would be particularly attractive.

In an LBO, private equity firms typically buy a stable, mature company that's generating cash. It can be a company that is underperforming or a corporate orphan—a division that is no longer part of the parent's core business. The point is that the company has lots of cash and little or no debt; that way the private equity investors can easily borrow money to finance the acquisition and can even use borrowed funds for more acquisitions once the company is private. In other words, the private equity firms leverage the strong cash position of the target company to borrow money. Then they restructure and improve the company's bottom line and later either sell it or take it public. Such deals can produce returns of 30 percent to 40 percent.

Private equity has been a growing source of funding for tech companies for several years. (See "Sampling of LBO activity in the technology industry" below. For more background, see "Gaining Leverage," January 2005.) Firms have been attracted as the chip industry has matured, becoming less cyclical and therefore a more reliable generator of cash. Several factors have contributed to this trend, says Paul McWilliams, editor at an investment advisory service called Next Inning Technology Research.

First, the chip industry—which was originally driven by military needs and later by the business market—is now driven primarily by the consumer market, which is less cyclical. Second, as the costs of building fabs have skyrocketed, chip companies have switched to fab-lite or fabless strategies, reducing their capital expenditures. And third, chip companies have gotten better at managing their inventories to avoid wide imbalances between supply and demand, McWilliams says.

Because of these changes, many chip companies have become flush with cash, says Suji De Silva, an analyst at Cathay Financial. The cash-to-sales ratio of some of these companies is greater than 30 percent, he notes. What's more, the private equity firms are flush as well.

"There's a lot of money available in private equity now, and they need to put that to good use," says Frans van Houten, president and CEO of NXP Semiconductors, the new company formed from the spinout of Philips' chip division. According to Thomson Financial, private equity firms had spent $421 billion this year on buying companies, through mid-October, up from $253 billion in all of 2005.

In addition, going private has become a more attractive option because of increased regulation and scrutiny of public companies, says McWilliams. Private companies don't have to spend the time and money to comply with Sarbanes-Oxley and SEC regulations, he notes.

When they look at the semiconductor industry, private equity firms see an industry sorely in need of consolidation. "We have too many semiconductor companies," says Derek Lidow, president and CEO of market research firm iSuppli.

Lidow counts about 450 chip companies around the world, noting that many of them are based on a business model that requires high growth rates. But growth in the chip industry has slowed from a historic rate of 15 percent to 20 percent to about 7 percent to 8 percent today, he says. "It's harder and harder to come out with a breakthrough product—every new product just cannibalizes an old one."

But the chip industry itself has been unable or unwilling to take the hard steps necessary for consolidation. Many chip companies are run by engineers, who tend to think in terms of technology rather than of how best to manage a product portfolio, Lidow says.

"The real leverage in this kind of a deal is to do portfolio management," he notes, which means managing groups of products by market segment or geographic region, for example, rather than by technology category. "That's usually not done in the semiconductor industry," he adds. "It's a very big shift, and I think it could be hard to internalize for quite a few CEOs.

"That's really what the private equity companies are up to. Although they may be temporarily creating new companies (by buying the chip unit of Philips, for example), the grand plan may be to then acquire and merge companies that have similar product portfolios.

"The people that can figure out how to create larger semiconductor companies that are still very efficient will reap huge benefits," Lidow says.

Indeed, merging it with another company was one of three options—the other two being an IPO and private equity—under serious consideration when Philips was deciding what to do with its chip division, says van Houten. But Philips couldn't convince potential merger partners to act.

"We talked with some companies, and rationally everyone agreed that there was synergy value in a merger," van Houten says. "But if other companies are not mentally ready to consider this, then merging becomes a less viable route."

One of those companies was presumably Freescale. According to public documents detailing its LBO deal, Freescale discussed merging with NXP and an unnamed third company but decided on the private equity deal in the end.

For Philips an IPO for the chip unit wasn't particularly attractive, because it would have taken longer for the parent company to reduce its stake. Another potential drawback was that the public markets tend to be less patient, wanting to see consistently increasing profits every quarter. Private equity, on the other hand, operates on an investment horizon of four to seven years, which would give the company more time to implement its strategy for operational improvement, says van Houten.

Private equity does have drawbacks, however. It can be disruptive if the investors come in with a different operational strategy. And all that borrowing can threaten the fiscal health of the company. Van Houten insists that NXP's investors not only bought into its current and future strategy but also agreed to a conservative balance sheet structure. Specifically, they agreed to a debt-to-equity ratio of 0.9 -- that is, 47 percent debt to 53 percent equity. "You'll find that is conservative compared to other LBOs," he notes.

One thing that seems clear is that private equity is bringing some dramatic changes to the chip industry. Private equity investors will acquire companies and juggle them to form new entities. In some cases, they will build out and grow a company; in others, they may break it up and sell off the parts. "I think that, going forward, we will see both cases, and rightly so," says van Houten. "If a company has very dispersed activities that are not synergistic, then maybe splitting it up is the right thing to do."

These private equity developments could light a fire under other chip companies, forcing them to restructure and consolidate as a defense, says De Silva. "All this private equity activity is sort of a nudge," he says.