13 Jan 2003  Research & Ideas

Making Biotech Work as a Business

What will it take for biotechnology to fulfill its economic potential? Participants need to think twice about the strategies and assumptions that are driving the industry, says HBS professor Gary P. Pisano.


No matter how you slice the numbers, all arrows in biotechnology's financial performance after twenty years are still pointing downward, said HBS professor Gary P. Pisano recently. Despite the great promise of the sector—to improve the understanding of disease, accelerate the pace of new therapies, create safer and more effective treatments and, of course, reap profits—industry executives, like hopeful patients, still restlessly wait for relief.

Given its scientific potential, said Pisano, biotech continues to bump against several economic barriers that need to be confronted if the industry is ever going to leap ahead. The main barrier, according to Pisano, is the lack of integration among costs, rules, technologies, and disciplines. In a presentation at the Harvard Business School Alumni Healthcare Conference, held last November, he discussed the biotech paradox of high optimism and low return. He also offered strategic pointers for companies immersed in the field.

Suddenly, with the ability to monetize IP, there was a new freedom.
— Gary P. Pisano

At moments during Pisano's talk, titled "The Biomedical Revolution: From Scientific Promise to Commercial Reality," he expressed the frustration that many people in the biotech industry feel as he clicked through the depressing data in gloomy graphs on a big screen. "It's not a very profitable sector of the economy" based on the performance of new entrants, he deadpanned to the attendees. The fact that a sector is not profitable and yet continues to survive, he observed, is itself fascinating.

"Never before in the history of any industrial transformation or revolution have we seen [the investment community] consistently sustain such losses over a long period of time," Pisano said. The year 2002, for example, was a banner year for bad news, with thirty late-stage clinical failures of biotech drugs, he said. The online magazine Signals, which is published by Recombinant Capital, a consulting firm that works on biotechnology alliances, opined that even though 2002 was bad, it was not as bad as 1994: That year was known among insiders as "The Year of the Placebo."

What the data says

The good news: One hundred thirty drugs can be traced to biotech, said Pisano. (For purposes of discussion during his presentation, he defined "biotech" as a general term encompassing new technologies for drug discovery.) And patients with previously unmet medical needs can now get treatments that were not available before. That should be inspiring, he reminded the audience.

Economic frustrations remain, however. Pisano cautioned that current data on research and development productivity are hellishly difficult to sort out. This is due to the difficulty of measuring such amorphous factors as intellectual property creation and the long time lag before any expected return on investment. The sector's tendency to form alliances between firms also clouds the picture, Pisano said.

Caveats aside, his own research and that of the Tufts [University] Center for the Study of Drug Development was not encouraging. A Tufts study he cited, for example, found that R&D lead times—the time from clinical development to trials—increased from slightly less than thirty months in 1989 to around seventy months in 2002. The length of time it takes to develop a drug is obviously a major driver of the cost of developing the drug, Pisano said.

"The productivity of R&D biotech is certainly not increasing," he observed. "It seems to be quite convincingly decreasing over time. So that's also puzzling and somewhat troubling."

Early promise

In its first years of life twenty or so years ago, the biotechnology sector seemed healthy and its future looked bright. One early pioneer was the company Genentech. In addition to its scientific advances, the company has also influenced the industry by proving that intellectual property could be monetized through alliances with big pharmaceutical partners, according to Pisano. Genentech cut its first deal with Eli Lilly in 1982 and new entrants were excited.

The monetization of IP created what Pisano and his colleagues dubbed "a market for know-how." The new entrants became sellers; established pharmaceutical companies became buyers of the technology. This market for know-how reduced the barriers to entry.

"Suddenly, with the ability to monetize IP, there was a new freedom," Pisano said. "You could get into this industry and have someone basically fund some of your R&D. You could capture a return much faster. This contributed to the flood of new entrants that we have seen and which has continued pretty consistently over the twenty-year period. There have been ups and downs, but by and large the industry is still characterized by significant rates of entry."

The monetization of IP and the success of Genentech had another effect for business: a powerful impact on the new firms' strategies, Pisano added. Companies realized that they could focus on just one component of the R&D supply chain; they did not have to enter the industry as a drug company.

What this has added up to is a sector filled with "islands of expertise," Pisano said, with every idea almost seeming to merit its own firm.

Key problem: how to integrate

If biotech executives are going to link the islands of expertise together, they need to look at two broad kinds of mechanisms, according to Pisano: organizational—"so firms can expand their boundaries to include multiple capabilities. Virtual integration is a form of doing that"—and contractual, the creation of alliances and collaborations.

"What I want to suggest to you is that what should drive the nature of this choice is the nature of the underlying technology," he said, "in particular, the degree to which you can clearly specify, for lack of a better term, interfaces between the different tools."

In some sense, we have a mismatch between the technical needs for integration and the organizational structures which have emerged.
— Gary P. Pisano

Interfaces are either messy or clean. The electronics industry, for example, has devised standards and defined interfaces between components of various systems so companies can work together relatively smoothly. Intellectual property boundaries are clear. Biotechnology, on the other hand, is "definitely" over on the messy end of the interface spectrum, observed Pisano. Biotech firms develop their own approaches to problem solving based on their own tacit knowledge, and intellectual property rights are vague. And because firms are usually so small, they are very dependent on alliances, but each firm joins an alliance via its own little island.

How to win?

In his view, alliances are a very cumbersome way to organize an innovation system in biotech. Parties entering into agreements bring different levels of information, causing valuation problems and high governance costs. IP rights are also tricky. The lawsuits that crop up in biotech are usually about IP, said Pisano. And the relationships are very complex, much more so even than in the electronics business.

"These are very difficult relationships to unwind. They're not just ones you can pop in and out of easily. So I think this creates a kind of drag on the system," he said.

"In some sense, we have a mismatch between the technical needs for integration and the organizational structures which have emerged."

His suggestion for companies going forward: Think twice about the strategies and assumptions that are driving your industry. Universities, venture capitalists, public investors and the firms themselves should take a step back and invest more thought to what Pisano calls systems knowledge: how to integrate the costs, rules, technologies, and disciplines.

"I think the evidence is overwhelming that one tool doesn't do it. Yet time and time again, we see new entrants coming into this business saying, 'This tool will revolutionize the discovery process,' when it's much more likely that the integration of tools, of how they work, will have a much more powerful effect."

Companies could also rethink their partner strategies. They could consider whether they have enough knowledge on the inside to complement outside partnerships.

And, they should revisit their acquisition strategies. Some firms are acquired for the sake of scale, but the scale argument did not persuade Pisano.

Biotech companies should also think about their internal organizational structures and processes that help or hinder integration. An individual company may boast many capabilities in-house, but in reality the company may be stocked with islands of expertise that aren't working together at all.

In terms of R&D as a system, Pisano said, the productivity impact of any one tool or technology can be small. A new tool may be terrific, but it may just create a bottleneck somewhere else. Real gains in productivity, he suggested, will probably come from integrated improvements across multiple components of the system.

Finally, he offered, the economic barriers in biotech once again illuminate the problems of fragmentation that industry types have complained about for almost two decades. Every year like clockwork, biotech executives warn that the industry is on the verge of consolidating. So far, that hasn't happened.

To Pisano, the issue of fragmentation is not as simple as having too many small firms. "Small firms are extremely important to this whole industry," he stressed. Rather, the issue is one of having too many small firms with narrow competencies. Large firms can also focus too narrowly, he added.

A smoother way to integrate multiple tools and skills, then, might just be the key to biotech success.