Merck's Recall of Rofecoxib — A Strategic Perspective
http://www.100md.com
《新英格兰医药杂志》
When Merck announced the voluntary withdrawal of its acute-pain medication, rofecoxib (Vioxx), on September 30, 2004, its stock price collapsed, wiping out more than a quarter of the company's market value in a single day. A second blow came on November 1, when an article in the Wall Street Journal suggested that Merck had known about elevated risks of heart attack and stroke for years and had tried to intimidate scientists who questioned the safety of rofecoxib. On that day, Merck's stock price fell by another 10 percent. The recall came as a shock not only to patients and their physicians; the debacle was also terrible news for those awaiting new medical therapies, because the dramatic financial consequences of withdrawing rofecoxib might well hamper the ability of a major pharmaceutical company to invest in new drugs that could help patients in the future.
(Figure)
Sales and Market Share of Rofecoxib and Its Primary Competitors.
Data are from IMS Health. Data for 2004 are through July 31.
Investors reacted harshly to the allegations in the Wall Street Journal because a pharmaceutical company that puts profits before patients' safety does not deserve the public's trust. And without this trust, the company is not viable as a business. Although it is easy to understand Wall Street's reactions to the newspaper's allegations, it is much more difficult to see why Merck's market value fell so drastically when it first announced the withdrawal of rofecoxib. The decline in Merck's market value on September 30 was much larger than estimates of the financial and legal cost of withdrawing rofecoxib. Using a cost of capital of 11.25 percent to discount future earnings, we estimate that the withdrawal of rofecoxib will cost Merck about $9 billion in forgone profits and up to $5 billion in future litigation costs. Rofecoxib was crucial for Merck's bottom line — the high-margin blockbuster prescription drug contributed a fifth of Merck's profits in 2003 — but Wall Street's reaction was surprisingly strong, even if the economic significance of rofecoxib is taken into account. To gauge the reaction of financial investors, we conducted an event study estimating what Merck's stock price would have been if September 30 had been a regular trading day. Comparing this prediction with the actual stock price, we find that Merck lost more than $28 billion in market value. What explains the large difference between the expected cost of withdrawing rofecoxib ($14 billion) and the much more severe market reaction ($28 billion)?
In part, we believe, investors reacted to growing concerns about Merck's business strategy. Wall Street viewed the rofecoxib announcement as news about Merck's diminishing ability to come up with safe and commercially successful blockbuster drugs. Of the 11 new products Merck planned to sell or file for Food and Drug Administration (FDA) approval by 2006, 4 had been canceled in the past two years. Blockbusters — drugs with sales in excess of $1 billion per year — make or break pharmaceutical companies. The top 10 percent of drugs is responsible for more than 50 percent of the value created by pharmaceutical research and development. Only one third of new chemical entities introduced by the industry eventually recoup the investment that was necessary to develop them.1
In recent years, research-and-development productivity has declined precipitously. Bain and Company, a management consulting firm, estimates that the investment required for one successful drug launch increased from $1.1 billion in the late 1990s to $1.7 billion in 2002, reducing the expected rate of return to 5 percent, a figure that is clearly below the cost of capital for pharmaceutical companies. These cost increases largely reflect the difficulty of discovering new compounds: whereas between 1995 and 2000, 1 in 8 drugs entering preclinical trials was ultimately launched, the ratio now stands at 1 in 13. It is particularly worrisome that attrition rates are on the rise among drugs in phase 2 and phase 3 clinical trials, after drug makers have sunk considerable amounts of money into promising-looking compounds. The decline in research-and-development productivity is not only a concern for financial investors; increasing difficulty in coming up with new drugs directly affects patients and physicians by limiting the number of new pharmaceutical therapies.
Pharmaceutical companies pursued two strategies in the hope of stemming the decline in research-and-development productivity. First, they started merging with other firms. The idea behind these mergers is that research and development are subject to economies of scale; the average cost of developing new drugs, the companies argued, is smaller for larger firms. This belief fundamentally changed the structure of the pharmaceutical industry. After a series of mergers, the 10 largest firms now account for more than one half of worldwide sales. In comparison, the 10 largest firms were responsible for about 12 percent of worldwide sales in 1987. While the rest of the industry was in the throes of merger fever, Merck's management remained on the sidelines, arguing that mergers would not increase research-and-development productivity or profitability. This attitude persists even after the rofecoxib debacle.
Is Wall Street punishing Merck for not recognizing economies of scale? If so, the punishment is undeserved. Although recent studies show that pharmaceutical companies are likely to merge whenever their pipeline is in danger of running dry — which is exactly the situation at Merck today — mergers in the years 1988 to 2000 did not increase the market value of firms, nor did mergers have a positive effect on sales growth or research-and-development expenditures.2 Irrespective of the appetite of the business press for yet another pharmaceutical megamerger, in this industry, bigger is not better.
In a second move designed to counter the decline in research-and-development productivity, pharmaceutical companies sought to replenish the pool of important patent-protected drugs by forming alliances with biotechnology companies. These alliances exploit economies of specialization, with biotechnology firms delivering new scientific discoveries and pharmaceutical companies providing expertise in drug testing, regulatory approval, and marketing. Although Merck entered numerous alliances, its strategy again differed markedly from those of important competitors. Rather than seeking promising substitutes for in-house research, Merck engaged in deals designed primarily to provide expertise that supported the work of Merck scientists. Whereas other companies filled their product pipelines with drugs they had licensed from biotechnology firms, Merck's alliances largely complemented the company's research activities.
The long-term strategic value of biotechnology alliances is hard to ascertain. If the market for new scientific discoveries works well, biotechnology firms will reap most of the benefits, because pharmaceutical companies have to compete for promising compounds, driving the price of licenses up near the expected value of the new drugs. There is some evidence that competition for licenses is intensifying. Bain and Company reports that the average return on investment for licensed compounds in phase 3 has decreased from 12 percent in the 1990s to 6 percent today. The higher cost of licensing compounds from biotechnology companies is one reason for this decline. Another reason is that collaborative arrangements are less likely than in-house projects to result in a marketable product. The chance of achieving an average return on a licensed compound has dropped from 40 percent in the late 1990s to 15 percent today.
The mediocre quality of many collaborative projects is hardly due to chance. The market for scientific research suffers from the familiar "lemons" problem, whereby higher-value products are driven out of the market by lower-value goods because buyers have difficulty distinguishing the two products and are willing to pay only an average price. This price does not fully compensate the sellers of high-value products, who therefore withdraw from the market. In the drug industry, pharmaceutical companies buy licenses at a discount because it is difficult to judge the market prospects of new compounds before signing a deal. Expecting such discounts, biotechnology companies have incentives to keep their most promising projects for in-house development, but they are happy to market projects of more uncertain quality. The lemons problem is real. Having studied 260 biotechnology projects during the past 20 years, Gary Pisano of the Harvard Business School has documented that collaborative projects in the biopharmaceutical industry are twice as likely to fail as projects that are developed in-house.3
Declining to merge and pursuing an alliance strategy unlike those of other firms, Merck stands out as unusual in today's pharmaceutical industry. The company has placed a huge bet on the success of its scientists. The rofecoxib fiasco serves as a harsh reminder that this gamble bears substantial risk. Yet we think Merck's strategy is fundamentally sound. The company's stubborn refusal to merge is the right business decision. Even its insistence on investing heavily in scientific expertise makes sense. This expertise, however, must not lead managers to blindly favor in-house development over external collaborative projects. Rather, Merck could use its scientific knowledge to develop a competitive advantage in evaluating external scientific opportunities. Strong internal capabilities and alliances with biotechnology firms complement one another. If Merck were better than its competitors at assessing biotechnology, the drug maker could afford to be more selective in choosing partners, betting on fewer but more-promising compounds, effectively reversing the lemons dynamic to its advantage. This outcome would be appealing to Wall Street and patients awaiting new drugs. Turning scientific expertise into competitive advantage is Merck's principal long-term strategic challenge. The task at hand, however, is for Merck's management to convince the public that it cares about patients' safety.
Source Information
From the Harvard Business School, Boston.
References
Grabowski H, Vernon J, DiMasi JA. Returns on research and development for 1990s new drug introductions. Pharmacoeconomics 2002;20:Suppl 3:11-29.
Danzon PM, Epstein A, Nicholson S. Mergers and acquisitions in the pharmaceutical and biotech industries. NBER working paper no. W10536 . (Accessed October 29, 2004, at http://hc.wharton.upenn.edu/nicholson/pdf/merger_Sept03.pdf.)
Pisano GP. R&D performance, collaborative arrangements and the market for know-how: a test of the "lemons" hypothesis in biotechnology. Harvard Business School working paper no. 97-105. (Accessed October 29, 2004, at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=41980.)(Felix Oberholzer-Gee, Ph.)
(Figure)
Sales and Market Share of Rofecoxib and Its Primary Competitors.
Data are from IMS Health. Data for 2004 are through July 31.
Investors reacted harshly to the allegations in the Wall Street Journal because a pharmaceutical company that puts profits before patients' safety does not deserve the public's trust. And without this trust, the company is not viable as a business. Although it is easy to understand Wall Street's reactions to the newspaper's allegations, it is much more difficult to see why Merck's market value fell so drastically when it first announced the withdrawal of rofecoxib. The decline in Merck's market value on September 30 was much larger than estimates of the financial and legal cost of withdrawing rofecoxib. Using a cost of capital of 11.25 percent to discount future earnings, we estimate that the withdrawal of rofecoxib will cost Merck about $9 billion in forgone profits and up to $5 billion in future litigation costs. Rofecoxib was crucial for Merck's bottom line — the high-margin blockbuster prescription drug contributed a fifth of Merck's profits in 2003 — but Wall Street's reaction was surprisingly strong, even if the economic significance of rofecoxib is taken into account. To gauge the reaction of financial investors, we conducted an event study estimating what Merck's stock price would have been if September 30 had been a regular trading day. Comparing this prediction with the actual stock price, we find that Merck lost more than $28 billion in market value. What explains the large difference between the expected cost of withdrawing rofecoxib ($14 billion) and the much more severe market reaction ($28 billion)?
In part, we believe, investors reacted to growing concerns about Merck's business strategy. Wall Street viewed the rofecoxib announcement as news about Merck's diminishing ability to come up with safe and commercially successful blockbuster drugs. Of the 11 new products Merck planned to sell or file for Food and Drug Administration (FDA) approval by 2006, 4 had been canceled in the past two years. Blockbusters — drugs with sales in excess of $1 billion per year — make or break pharmaceutical companies. The top 10 percent of drugs is responsible for more than 50 percent of the value created by pharmaceutical research and development. Only one third of new chemical entities introduced by the industry eventually recoup the investment that was necessary to develop them.1
In recent years, research-and-development productivity has declined precipitously. Bain and Company, a management consulting firm, estimates that the investment required for one successful drug launch increased from $1.1 billion in the late 1990s to $1.7 billion in 2002, reducing the expected rate of return to 5 percent, a figure that is clearly below the cost of capital for pharmaceutical companies. These cost increases largely reflect the difficulty of discovering new compounds: whereas between 1995 and 2000, 1 in 8 drugs entering preclinical trials was ultimately launched, the ratio now stands at 1 in 13. It is particularly worrisome that attrition rates are on the rise among drugs in phase 2 and phase 3 clinical trials, after drug makers have sunk considerable amounts of money into promising-looking compounds. The decline in research-and-development productivity is not only a concern for financial investors; increasing difficulty in coming up with new drugs directly affects patients and physicians by limiting the number of new pharmaceutical therapies.
Pharmaceutical companies pursued two strategies in the hope of stemming the decline in research-and-development productivity. First, they started merging with other firms. The idea behind these mergers is that research and development are subject to economies of scale; the average cost of developing new drugs, the companies argued, is smaller for larger firms. This belief fundamentally changed the structure of the pharmaceutical industry. After a series of mergers, the 10 largest firms now account for more than one half of worldwide sales. In comparison, the 10 largest firms were responsible for about 12 percent of worldwide sales in 1987. While the rest of the industry was in the throes of merger fever, Merck's management remained on the sidelines, arguing that mergers would not increase research-and-development productivity or profitability. This attitude persists even after the rofecoxib debacle.
Is Wall Street punishing Merck for not recognizing economies of scale? If so, the punishment is undeserved. Although recent studies show that pharmaceutical companies are likely to merge whenever their pipeline is in danger of running dry — which is exactly the situation at Merck today — mergers in the years 1988 to 2000 did not increase the market value of firms, nor did mergers have a positive effect on sales growth or research-and-development expenditures.2 Irrespective of the appetite of the business press for yet another pharmaceutical megamerger, in this industry, bigger is not better.
In a second move designed to counter the decline in research-and-development productivity, pharmaceutical companies sought to replenish the pool of important patent-protected drugs by forming alliances with biotechnology companies. These alliances exploit economies of specialization, with biotechnology firms delivering new scientific discoveries and pharmaceutical companies providing expertise in drug testing, regulatory approval, and marketing. Although Merck entered numerous alliances, its strategy again differed markedly from those of important competitors. Rather than seeking promising substitutes for in-house research, Merck engaged in deals designed primarily to provide expertise that supported the work of Merck scientists. Whereas other companies filled their product pipelines with drugs they had licensed from biotechnology firms, Merck's alliances largely complemented the company's research activities.
The long-term strategic value of biotechnology alliances is hard to ascertain. If the market for new scientific discoveries works well, biotechnology firms will reap most of the benefits, because pharmaceutical companies have to compete for promising compounds, driving the price of licenses up near the expected value of the new drugs. There is some evidence that competition for licenses is intensifying. Bain and Company reports that the average return on investment for licensed compounds in phase 3 has decreased from 12 percent in the 1990s to 6 percent today. The higher cost of licensing compounds from biotechnology companies is one reason for this decline. Another reason is that collaborative arrangements are less likely than in-house projects to result in a marketable product. The chance of achieving an average return on a licensed compound has dropped from 40 percent in the late 1990s to 15 percent today.
The mediocre quality of many collaborative projects is hardly due to chance. The market for scientific research suffers from the familiar "lemons" problem, whereby higher-value products are driven out of the market by lower-value goods because buyers have difficulty distinguishing the two products and are willing to pay only an average price. This price does not fully compensate the sellers of high-value products, who therefore withdraw from the market. In the drug industry, pharmaceutical companies buy licenses at a discount because it is difficult to judge the market prospects of new compounds before signing a deal. Expecting such discounts, biotechnology companies have incentives to keep their most promising projects for in-house development, but they are happy to market projects of more uncertain quality. The lemons problem is real. Having studied 260 biotechnology projects during the past 20 years, Gary Pisano of the Harvard Business School has documented that collaborative projects in the biopharmaceutical industry are twice as likely to fail as projects that are developed in-house.3
Declining to merge and pursuing an alliance strategy unlike those of other firms, Merck stands out as unusual in today's pharmaceutical industry. The company has placed a huge bet on the success of its scientists. The rofecoxib fiasco serves as a harsh reminder that this gamble bears substantial risk. Yet we think Merck's strategy is fundamentally sound. The company's stubborn refusal to merge is the right business decision. Even its insistence on investing heavily in scientific expertise makes sense. This expertise, however, must not lead managers to blindly favor in-house development over external collaborative projects. Rather, Merck could use its scientific knowledge to develop a competitive advantage in evaluating external scientific opportunities. Strong internal capabilities and alliances with biotechnology firms complement one another. If Merck were better than its competitors at assessing biotechnology, the drug maker could afford to be more selective in choosing partners, betting on fewer but more-promising compounds, effectively reversing the lemons dynamic to its advantage. This outcome would be appealing to Wall Street and patients awaiting new drugs. Turning scientific expertise into competitive advantage is Merck's principal long-term strategic challenge. The task at hand, however, is for Merck's management to convince the public that it cares about patients' safety.
Source Information
From the Harvard Business School, Boston.
References
Grabowski H, Vernon J, DiMasi JA. Returns on research and development for 1990s new drug introductions. Pharmacoeconomics 2002;20:Suppl 3:11-29.
Danzon PM, Epstein A, Nicholson S. Mergers and acquisitions in the pharmaceutical and biotech industries. NBER working paper no. W10536 . (Accessed October 29, 2004, at http://hc.wharton.upenn.edu/nicholson/pdf/merger_Sept03.pdf.)
Pisano GP. R&D performance, collaborative arrangements and the market for know-how: a test of the "lemons" hypothesis in biotechnology. Harvard Business School working paper no. 97-105. (Accessed October 29, 2004, at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=41980.)(Felix Oberholzer-Gee, Ph.)