JV on the Rocks

Without plenty of attention and TLC, a joint venture may well run aground.
Roy HarrisJune 1, 2003

Trite as it may sound, a joint venture really is like a marriage. For both, good communication between partners is essential. Goals should be mutually agreed on. And a little passion helps. As for the darker side of the matrimonial comparison, many ventures, too, succumb to real-world temptations. Further, even if they don’t end outright in divorce (corporate-style), JVs often drag through years in an unproductive state.

There’s also a familiar ring to the reasons experts give for ventures failing or underperforming. Sometimes ventures just slide by for a while on good feelings from the wedding. “Many joint ventures have an initial endowment of independence from the parents, but then neither parent exercises the control they would exercise if the venture were an individual business unit,” says David Ernst, leader of McKinsey & Co.’s global alliance practice.

And in general, corporate partners simply do too little prenuptial planning. “Companies tend to be cavalier about alliances, and think of them as no big deal,” says Harbir Singh, a management professor and director of the strategic alliance program at the University of Pennsylvania’s Wharton School. A 2001 study he co-authored of 1,572 strategic alliances showed that while ventures are an increasingly popular way for companies to grow revenues, almost half fail.

There is much to be learned, though, from companies that understand the perils of venturing, and govern alliances with discipline and the desire to measure venture performance properly. The high level of planning that goes into Merck & Co.’s alliance structure, for example, assures that good decision-making “kind of cascades,” says executive vice president and CFO Judy Lewent, who has strung together successful drug ventures with Johnson & Johnson, AstraZeneca, Schering-Plough, and other firms. “As you think through the structuring of a joint venture, the structure enables the alignment of the partners to the best degree possible.” Merck wants the venture to be “as close to a wholly owned subsidiary as possible,” says Lewent, and that requires “the right kinds of performance measurements.”

The Unexamined Life

Inattention to measurement, however, seems endemic in most venturing. Mc-Kinsey, which counts itself something of a marriage counselor for troubled ventures, sees poor use of metrics at the root of JV problems. In a presentation titled “Measuring Alliance Performance: What CFOs Should Do to Avoid Job-Threatening Surprises,” McKinsey partner Ernst recently cited a 1999 Andersen Consulting survey showing that only 11 percent of alliance partners believe they have sufficient performance measurements in place. And a surprising 49 percent declared they had “essentially no performance measurements in place.”

What’s more, many agreements detailing the governance plan for JVs “specify pretty low levels at which parent or board approval is required for individual decisions. So from the outside, it looks like the board should have control,” says Ernst. The problem, though, is that the partners “tend to be invasive” and override board control “on individual expenditures or contracts that generate financial risk,” he says. “That means risk management is on a very ad hoc basis.”

By signing risk-sharing agreements and then inadequately monitoring performance, joint-venture partners are asking for trouble. Ernst lists such venture-related damage as earnings restatements caused by misrepresented venture revenues, discovery of hidden liabilities, and costs from duplicative staff or other inefficiencies. Poorly designed formulas for transfer-pricing structures can multiply costs quickly. And often the partnering companies have no organizational structure to fix things.

“Many ventures simply lack a challenge process,” says Ernst. “The analogy would be a business unit of a corporation that reports to a board but has no intermediate authority.” For such ventures, he suggests that a partner’s CFO “play the role of challenge agent,” or that a group be appointed to “aggressively set targets, ask probing questions, and press for more performance.” More broadly, Ernst suggests auditing the portfolio of alliances to see which ones are in need of restructuring.

Correcting underperformance problems could add at least $1 billion in shareholder value to Corporate America, according to McKinsey. And the study co-authored by Wharton’s Singh also found significantly higher stock-market prices accruing to companies with disciplined alliance-management processes. The gains after alliance announcements at those companies, in fact, add an average of $75 million to market capitalization, he says, compared with $20 million for nondisciplined companies.

Cultural Factors

Sometimes JV issues run deeper than matters of pure finance. When Delta Air Lines created an Atlanta-based air-cargo venture with Air France and Korean Air in November 2001, the partners set up a nine-member board to oversee airline-performance measures like freight sales and market share, along with indicators on the operations-support side, such as call-center performance and customer acceptance. The board meets quarterly, so partners can keep tabs on the venture’s progress.

The venture expected to encounter challenges from bringing together airline employees who are from different countries and who speak different languages. And it prepared for them. “Part of our mission was to define a new culture, something that people would identify with in this company,” says venture CEO Bernard Frattini. Also, the drive to create a single corporate structure out of three diverse airlines “was quite tough, and we had to change the business plan a few times” to reflect lessons learned. But customers “are now looking at the venture as one company,” he says, rather than trying to keep lines of communication open with their previous contacts. And morale is high.

Differences cropped up in accounting as well, although most seemed minor. “Delta called forecasts the ‘revised plan,’” notes Frattini. And the venture had to redefine such measures as net and gross revenue, which partner airlines calculated differently. The partners sought to use economies of scale to increase revenue beyond what the three airlines would have produced before. Frattini would like the venture’s scale to be boosted even more, by the addition of the two other Delta SkyTeam passenger-airline alliance partners — Alitalia and Aeromexico, which have chosen so far not to participate in the U.S. cargo venture. Adding another partner “will be a big signal that we have succeeded,” says Frattini. (CSA Czech Airlines recently joined the alliance.)

Merck’s Prescription

Merck likes to keep its JVs simple. “The most successful structures I’ve seen are 50/50 ventures, with no bells and no whistles, where the partners are in it together,” says CFO Lewent. The governance structure for its JVs generally involves a board and an executive committee, which Lewent usually co-chairs. “The [executive committee] has a broad charter, but the most important thing is that we talk frequently, and when we meet, we have the head of the venture with us,” she says.

Performance grids are set out for each venture, “with very clear expectations of what success means in a given year,” says Lewent. That covers revenue expectations, cash flow, and measures in such strategic areas as research-and-development productivity — a particularly important metric for Merck’s five-year-old animal-health venture, operated with Aventis.

At Merck, Lewent explains, venturing processes also reflect “the relative importance of a JV at different times in its development,” including when the JV might end, under a scenario that is part of the structure designed into the original agreement.

When a venture with DuPont Pharma ended in 1998, wrapping up an alliance that Merck counts as a success, it did bring with it one good lesson in venture management. By creating finite horizons for a decision on whether to continue a venture, the DuPont and Merck parties “hardwired” a deadline that changed the way the parties performed.

“I personally think that mechanism has some disadvantages to it. It’s a problematic dynamic,” says Lewent. Basically, the wind-down period to the end of the venture can encourage people to think more about their own benefit at the conclusion of the enterprise and less about the benefit to the organization — a phenomenon that happens less when the plans for ending a venture are based more on performance and the changing wishes of management.

In later ventures, she notes, Merck established more-flexible horizons for decisions that could mark the potential end of partnerships. The decisions, she says, now reflect “a shared understanding of what success is.”