Exit Timing Still Vexes VCs

When 3Par went public in November, 2007, investors in the venture funds backing it already stood to gain a solid return.

The Fremont, Calif.-based maker of enterprise storage products, which had raised $177 million over the prior eight years, ended first-day trading with a market capitalization around $840 million, after pricing above the expected range. Its largest VC investors—Mayfield Fund, Menlo Ventures and Worldview Technology Partners—owned more than half of the company’s shares.

But rather than liquidate their holdings or distribute shares back to limited partners, 3Par’s backers sold only a small portion of their stakes. They held through the financial crisis, during which 3Par shed more than half its value, and, come summer, still owned about a third of outstanding stock.

Keeping their stakes turned out to be a fortuitous move. When a bidding war over the company erupted between Hewlett-Packard and Dell in August, 3Par shares shot up to nearly triple their prior value. As of mid-September, the company was seeking shareholder approval for a $2.4 billion acquisition offer from HP. Menlo still held a 15% stake in the company (valued at $309 million), followed by Worldview (13.4%, valued at $276 million) and Mayfield (9.9%, valued at $257 million.) Within the space of a few weeks, 3Par went from respectable exit to certified home run.

The apparent moral of the story—for those willing to seek morals in stock market fluctuations¬—seems to be that good things come to those who wait. That’s particularly true for investors in companies that went public on the eve of the credit crunch. While venture investors typically look to sell shares as quickly as possible following the requisite 180-day lockup, holding on appeared a more prudent tactic in the wake of a bear market contraction.

An overview of ownership filings for venture-backed companies that went public shortly before the financial crisis shows that VCs often still hold significant stakes today. Kleiner Perkins Caufield & Byers, for example, is still hanging on to 10%, or about half its pre-IPO stake, in ArcSight, a security and compliance software provider that went public in early 2008. The decision not to sell earlier worked out well for Kleiner. ArcSight’s stock has quadrupled since its bear market lows, and it got another boost last month when Hewlett-Packard said it would buy the company for $1.5 billion—a 24% premium over ArcSight’s pre-announcement closing price. Kleiner’s stake in the company, which raised about $15 million total venture funding before its IPO, is worth about $152 million.

Aruba Networks, a provider of services for remote access to enterprise networks that went public in early 2007, also continues to have significant VC shareholders: Sequoia Capital and Sutter Hill Ventures still control 15.5% and 7.9%, respectively. That, too, should boost fund returns, as Aruba shares, after a steep tumble, are now back above their offering price. (For others, see table.) With the vast majority of technology stocks up dramatically from the lows hit early last year, wait-and-see has largely been a winning strategy.

Often the hard decision is when you have a company in that class, where, say, you’ve got a 10x in three years. Do you hold, or do you exit at the IPO? In many cases, the right decision is to hold.”

Todd Chaffee

Still, it doesn’t always work out that way. Venture backers in two other companies that went public over three years ago—BigBand Networks and AuthenTec—have found buy-and-hold a punishing approach. BigBand, a provider of digital video networking technology whose stock peaked above $20, is now under $3, a poor showing for Redpoint Ventures, which still owns about 19% of the company, according to a recent securities filing. Meanwhile, AuthenTec, a provider of fingerprint security systems, has seen its stock fall below $2 per share (from a peak above $17), which is terrible news for Crosslink Capital, which owned 6.5% of the company as of this summer.

Thus, moral of the story is unsatisfying: Good things or bad things come to those wait. The challenge is figuring out when patience is warranted, and when it’s preferable to opt for a quick exit.

When Good Isn’t Enough

Determining when to sell public shares is a complicated matter. For instance, just because a company did well in its IPO, it doesn’t follow that a venture backer should sell as soon as possible.

“Often the hard decision is when you have a company in that class, where, say, you’ve got a 10x in three years,” says Todd Chaffee, general partner at Institutional Venture Partners. “Do you hold, or do you exit at the IPO? In many cases, the right decision is to hold.”

For the later stage and growth companies that IVP targets, Chaffee says, “the exit decision is actually harder than the entry decision.” And while in theory it’s hard to sniff at a return of several multiples, in practice, when the stock price of a fully exited portfolio company continues to skyrocket, there are apt to be regrets.

ArcSight is a case in point. IVP had a 12% stake in ArcSight at the time of its IPO. In recent filings, neither it, nor venture backers New Enterprise Associates and Integral Capital Partners, is listed as a major stockholder. Spokespeople for NEA and IVP declined to discuss their ArcSight stock holdings. A spokesperson for Integral didn’t respond to a request for comment before VCJ went to press.

While there are cases when [VCs] would do better holding, they are — even with inside information as board members and large shareholders — just as bad at market timing as the rest of us.”

Paul Kedrosky

The venture business is rife with stories of VCs who ponder the question: “If only I’d hung on.” It’s something Chaffee recalls Sequoia’s Don Valentine ruminating about Cisco, which the firm exited prior to its meteoric rise in the late 1990s. Chaffee also recalls a too-early exit when he was an executive vice president at Visa International, prior to joining IVP. Visa made something like a 220x return on its investment in Yahoo, he says, but it actually could have made more by riding the stock to its peak.

However, VCs are often disinclined to wait too long, in part because long-term buy-and-hold investment in public equities is not part of their core mission. Limited partners already have plenty of exposure to technology stocks through their public market holdings. They invest in VC for the potential of larger, albeit riskier, returns in early stage private companies.

“It’s not about how good of a ride or how long of a ride,” says Dan Gravelle, a partner at FLG Partners, a CFO consulting service that works with venture-backed companies. “It’s about when do you think you’ll be able to make a better investment somewhere else?”

It’s easy, for instance, for Google backers to kick themselves for selling the stock at $300 rather than $700. But even that, Gravelle says, was only about a 2x return—and a far-from-guaranteed one at that. VCs ought to be aiming for something bigger.

Overall, research shows VCs have historically had good timing in distributing shares to LPs, says Josh Lerner, a Harvard Business School professor who specializes in venture capital and entrepreneurship. “They seem to have good timing at taking things public,” Lerner says, “and the more established groups seem to have better timing than the less-established groups.” There are exceptions, of course, such as the venture investors who hung on to online retailer eToys as share prices plunged from more than $50 to mere pennies.

Generally speaking, studies have shown that VCs should be selling as soon as they come out of lockup, says Paul Kedrosky, a senior fellow at the Kauffman Foundation. “While there are cases when they would do better holding, they are—even with inside information as board members and large shareholders—just as bad at market timing as the rest of us,” Kedrosky says.

Taking Credit

They [VCs] seem to have good timing at taking things public, and the more established groups seem to have better timing than the less-established groups.”

Josh Lerner

There’s also pressure from limited partners to consider. It’s particularly pronounced in the current environment, where companies typically need upwards of seven years to generate the $50 million-plus in revenue preferred by IPO investors. If an early stage investor holds on to a late-blooming portfolio company for several years after an IPO, the entire process of returning capital to investors could conceivably exceed a fund’s expected 10-year lifespan.

But a quick exit isn’t always possible, particularly if venture firm partners remain directors.

“To the extent venture capital firms are still on the board and have very significant stakes, they often have limited windows for liquidity even after the company goes public,” says Victor Hwang, managing director at Industry Ventures, a secondary investor. Still, from personal experience, Hwang has learned that there’s value in a timely exit. Hwang was managing director of acquisitions at Internet Capital Group, the top-performing IPO of 1999, where he says he was able to sell some shares at a favorable valuation.

Another option is that once a firm takes a portfolio company public, it needn’t sell shares at all. It can simply distribute the shares to LPs, which can sell them as they see fit. That’s a strategy IVP has sometimes followed, Chaffee says.

The trouble with this strategy, however, is if a VC distributed during a down market, and the shares later go up in value, the VC can’t record the gains as part of its fund returns. Thus, a home-run exit for an LP may go down in VC’s books as a mediocre one.

But it also works the other way, too—perhaps more often, says Lerner. That’s because when firms such as Cambridge Associates evaluate venture exits, they look at the price of shares when they’re distributed to LPs. Commonly, however, LPs are unable to actually sell at those levels, because if they try to unload a large number of shares at one time, prices will drop.

Still, the costs for either error can be quite high. VCs left billions on the table, for instance, for bailing too soon on medical device developer Intuitive Surgical. The Sunnyvale, Calif.-based company had a market cap just over $300 million when it went public in 2000, and dipped to about half that over the following two years.

A series of breakthroughs and key sales wins put the stock on a tear a few years later. Currently, Intuitive Surgical is valued at nearly $12 billion, but venture backers Mayfield, Morgan Stanley Dean Witter Venture Partners and Sierra Ventures didn’t stick around for the steep ascent. Spokespeople for Mayfield, Morgan Stanley and Sierra Ventures did not respond to requests for comment before VCJ went to press.

Perhaps the experience taught Mayfield a lesson about patience—one that worked out quite well in the case of 3Par.