Occasionally, a company and its CEO become so inextricably linked that the image of the CEO becomes inseparable from the image of the company. Sometimes it is the CEO’s accomplishments that come to represent the organization, and other times it is his or her personality. In the case of Merck’s P. Roy Vagelos, his unusual blend of scientific knowledge and leadership abilities turned Merck into a pharmaceutical powerhouse. Vagelos began his career as a senior surgeon at the National Institutes of Health and later served as chairman of the department of biological chemistry at Washington University’s School of Medicine in St. Louis, Missouri. He joined Merck in 1975 as head of the research labs and became CEO in 1985. During his nearly 20 years with the company, Vagelos oversaw the development of a series of breakthrough drugs, including Timoptic, Primaxin, Vasotec, Mevacor, Zocor, Proscar, and the generic ivermectin—products with total annual sales of $ 6.1 billion. More recently, Vagelos shocked the pharmaceutical industry with Merck’s acquisition of the prescription-benefits-management company (PBM) Medco Containment Services, which provides prescription drugs to health maintenance organizations and the employees of large corporations. Some have called this move brilliant, while others have called it overpriced and ill advised. But one thing is certain: Merck’s acquisition of Medco set off a firestorm within the pharmaceutical industry when competitors followed Merck’s lead by acquiring their own PBMs. Eli Lilly bought PCS Health Systems, a PBM that was part of McKesson, for $ 4 billion, and SmithKline Beecham purchased Diversified Pharmaceutical Services for $ 2.3 billion. Together these deals have revamped this once conservative corner of the medical industry. Vagelos retired on November 1, 1994. In this interview with HBR senior editor Nancy A. Nichols at Merck’s Whitehouse Station, New Jersey, headquarters, Vagelos looks back on his career and forward into the future of the pharmaceutical industry. HBR: What was it like moving from academia to the corporate world? P. Roy Vagelos: When I told my closest friend at Washington University that I was going to Merck, he said, “Roy, you’re going to have to sell toothbrushes and combs.” I knew that wasn’t Merck’s business, but I wasn’t sure what the corporate setting would be like. When I arrived, Merck was launching a drug called Sinemet, which is still important in the treatment of Parkinson’s disease. I realized then that the kind of research I had been doing in a university setting could result in products that would have a major impact on the health of people around the world. I saw the opportunity to influence the course of medicine. Of course, I had no experience developing commercially viable drugs. Merck took a big risk in hiring me, and it turns out that I took on the challenge of a lifetime: managing the balance between the demands of science and the equally unrelenting demands of the marketplace. I needed to hold on to the values that were important to me as a physician and blend them with Merck’s need to remain profitable. In the years since then, Merck’s research laboratories have produced many vaccines and medicines—products for hypertension, heart failure, high cholesterol, prostate disease, and glaucoma, to name a few. We’ve had a tremendous impact on the lives of millions of people, and we’ve been successful as a company. Now it seems that the pharmaceutical industry is changing. How will Merck follow its own act in the new environment? It’s very difficult to change a successful company. Merck has a proven track record, but we know that we can’t escape the changes in the industry; we must embrace them. Holding on to old ways in a new world will put a company out of business. “It’s very difficult to change a successful company.” Health care reform, increasing competition, and an overwhelming desire for higher-quality health care at lower prices are all creating competitive pressures, which are transforming our industry. In the future, Merck will have to grow through increased volumes without considerable price increases. That seems obvious now, but it wasn’t six or seven years ago. Is that why Merck decided to acquire Medco? Yes. We saw a tremendous opportunity to create a new model for the pharmaceutical industry that would simultaneously improve the quality of health care, help contain costs, and increase Merck’s market share. Expanding our information base is critical to achieving these goals: Medco has data on 38 million patients, which allow us to learn a lot more about how our drugs are prescribed and used and, ultimately, how effective they are in fighting disease. Whether it is cutting-edge scientific information or the reams of data on how doctors prescribe Merck products, information lies at the heart of what the company does. Our ability to leverage information will set us apart. Our goal is to maximize the effectiveness of our drugs. First, we must develop the safest and most effective drugs possible in the labs. Then, once the drug is on the market and has been prescribed to a patient, we must be sure that the patient is taking the right drug, that he or she has the appropriate information to take the drug properly, and that the drug will not interfere with other medications the patient is taking. We can ensure all this by capturing information as it comes through the pharmacy and then putting it into a central data bank that feeds the information back to the physician, the plan sponsor, and ultimately the labs, where it can be used to create new drugs. Medco gives us the ability to link different parts of the health-care-delivery system. The result is what we call coordinated pharmaceutical care, which could save billions of dollars a year in health care costs in the United States through the prevention of inappropriate drug interactions, undermedication, and overmedication. A study by Sharon Willcox, David Himmelstein, and Steffie Woolhandler, published this summer in the Journal of the American Medical Association, shows that one in four elderly people is taking inappropriate medicines. Studies also indicate that one of every two patients who need prescription drugs for chronic conditions such as high blood pressure and elevated cholesterol simply stops taking medication after one year. We can help solve these problems. “At Merck, we must be able to condense reams of information and let our best judgement, not our worst fears, prevail.” One danger, of course, is that scientists and managers may become overwhelmed by all this information and the possibilities it presents. At Merck, we must be able to condense reams of information and let our best judgment, not our worst fears, prevail. At the end of the day, in pharmaceuticals or any other business, you’ve got to be able to place your bets. Merck has done that by buying Medco: we made a $ 6.6 billion bet on where the future of the industry lies. It’s an awfully big bet, but I believe that the company that best controls the information flow from doctor to patient to pharmacist to plan sponsor has the best chance of succeeding in this industry. How did you learn to bet like that? In the pharmaceutical business, sometimes success comes from investing more money in a few projects rather than less in many projects. As CEO, I have learned that the only way to make the kinds of complex bets that you must make in an information-intensive business like Merck’s is by balancing two metrics: risk and focus. I recognized this critical balance when I first headed up the research labs at Merck in 1975. At that time, our critics on Wall Street were saying that Merck would never become a growth company based on what its laboratories could produce. Some people were even suggesting that Merck should diversify out of pharmaceuticals, but our commitment to R&D never faltered. I saw my challenge as helping Merck’s scientists do their best work and focus on the big projects that could result in breakthrough drugs. During the so-called war on cancer, which had been declared by President Nixon, it was very popular to say that the United States was on the brink of finding a cure for cancer, although I and many others doubted that there was any scientific basis for this claim. People would come to the labs and ask us to find a cure for cancer, but that was too diffuse a goal for us to undertake. The science hadn’t evolved far enough, and the target itself was too broad—there are more than 100,000 genes in each human cell, and, to this day, nobody knows how many of them are cancer causing or cancer suppressing. The vast majority of basic research projects fail. The amazing thing is that some succeed. And since Merck didn’t have an unlimited amount of money to pour into research, the company really needed to focus its efforts. How did you redirect the research department? The scientists and I began to prioritize what the various research groups were doing. Each group was working on as many as ten projects at once. They were trying to hedge their bets, but they were almost guaranteeing that no single project would have the critical mass necessary to succeed. We began by setting out some basic criteria: we would look for areas where there were no therapies or drugs available, where the science was advanced enough for us to believe that we could make a breakthrough, and where we had enough knowledge of the disease to have some idea about how to arrest it. It is hard to tell at the very early stages whether a project will succeed. In order to focus, you have to let go of some projects that look very attractive: you can’t do everything at once. At that time, the labs were engaging in random screening of compounds, which meant that scientists were testing microorganisms, soil samples, or plant extracts, as well as chemicals made in our own laboratories to see if they caused any pharmacological activity. If they did, the scientists tried to create a drug. If they didn’t, the scientists tested more compounds. We introduced a more targeted method that focused on creating specific molecules to attack specific molecular targets, an approach now known as rational drug discovery. We knew that many great drugs, from aspirin to antibiotics, are enzyme inhibitors; that is, they block the action of particular enzymes involved in the disease process. I wasn’t the first to realize this—it was common knowledge among scientists—but Merck was the first company to adopt enzyme inhibition as its major approach to drug discovery and development. When we began to look for a drug to lower abnormally high cholesterol, for example, we focused on blocking the enzyme that controls the overproduction of cholesterol. I decided that we would use this project to test the feasibility of the rational-drug-discovery process. The result: two anticholesterol drugs, Mevacor and Zocor. We used the same approach for many projects, including the development of Vasotec and Proscar, which are drugs for high blood pressure and prostate disease, respectively, and we’re now using the approach in our AIDS antiviral research program. Today the entire industry uses the rational-drug-discovery process. Was it difficult to persuade the scientists to buy into this change? My experience in academia prepared me for that challenge. When you’re teaching people in a university, you are also learning from them; there’s enormous feedback in the interaction. Like people in the academic world, scientists need to talk shop, disagree, and even fight about the things they believe in. When I went into a lab and said, “Look, we want to take a molecular approach; once we select the targets, it’s up to you to figure out how to work on them,” enormous arguments resulted. But they were significant arguments based on science, and we were ultimately able to choose good targets that resulted in breakthrough products. Early on, I was working with the endocrinology group, which had a number of projects in progress. It was clear that the group had to focus its efforts, and discussions about how to achieve that focus went on for a while. One day in the midst of all this, I picked up a journal on my desk and saw a study by a group at Cornell University describing a rare congenital problem: Male infants were born with external genitalia that often looked female, and many were raised as girls. Then, as they went through puberty, their genitals became more clearly defined as male, their voices deepened, and they grew facial hair. The study in the journal showed that the adult males had very small prostates, and they did not develop male pattern baldness or acne. It turns out that two different groups, the one at Cornell University and another at the University of Texas, had discovered that these people had a hormonal deficiency caused by a defect in a single enzyme. I realized that if we could affect this enzyme in older males who had enlarged prostates, we might be able to stop further growth of the gland and maybe even shrink it. I knew immediately that this could be the key to a drug for the treatment of benign prostate enlargement. I walked down the hall carrying that journal, and the head of the chemistry section in the endocrinology group was walking out of his office with the same journal. We met in the hall, and I said, “This is it.” He said, “This is it.” The endocrinology group decided to work continuously until it either proved or disproved the hypothesis. The result was a drug called Proscar, which controls benign prostate enlargement. Research started in 1975, and Proscar came to the market in 1992. But the most remarkable part of the story is that the National Cancer Institute is studying Proscar to see if it delays the onset of prostate cancer. So our research is contributing to the war against cancer, not by making a frenetic, let’s-save-the-world effort, but by focusing on a specific problem. Did you bring the idea of focus with you when you became CEO? After managing the labs for ten years, I knew a lot about the trade-offs between risk and focus, and I was willing to make my bets. As CEO, instead of trying to get a lot of independent-minded scientists to focus on a big project, I tried to get people in all areas at Merck to focus on what was most important for the company. For example, I recently reviewed our strategic plan for the Far East. There are markets with tremendous potential for growth in South Korea, China, Singapore, and Taiwan, but we must make sure that we are committed to achieving specific goals in each country. We can’t focus on too many things at once. Whether it is in research or in marketing, the ability to focus on the two things out of ten that may affect the future of the company is so critical—and so hard to find. Most people work on ten projects at once and hope that one of them will succeed. And they are often not willing to put down one of the ten because that may be the successful one. That unwillingness is the fastest way to fail. But lack of focus isn’t the only way to fail? In the life of each drug-development project, there is always a crisis, a moment when it looks like years of research will go down the drain and the drug will never come to market. There are a million ways to fail in this business. We scientists flirt with each and every one. There is one sure road to failure that I have seen many wander down: some people become so afraid of failing that they are unable to do a critical experiment or take that first step into a market. Even though Merck has been bold in its strategies, the company has missed out on some major opportunities because people were unwilling to take that truth-telling step—to conduct the experiment that would show once and for all if what they had spent so many years studying would actually produce a new drug. “There are a million ways to fail in this business.” There are wonderful arguments for putting off experiments like that. As a leader, you can’t force people to do what you think is most critical; they have to do those things on their own. In the end, they must make the decisions. And a leader’s input must fade, so that when group members discuss their work, it is truly their work and their direction. How do you keep the company focused on what you think is important and, at the same time, give employees some pride of ownership in their work? No one has all the answers to business problems. When you work with knowledge professionals—experts in science, manufacturing, marketing, or administration—you are working with equals, people who excel in their disciplines. Working in a university setting taught me a lot about that, so much so that I think many industry leaders should spend some time working in academia. When I was the head of the biochemistry department at Washington University from 1966 to 1975, I couldn’t order tenured faculty members to work on specific projects. When they begin to work on one of your projects, they almost always have to give up one of theirs. So they must be invited to join a project as intellectual equals. “Research at Merck often begins as a very unofficial process of scientists recruiting one another to work on a promising project.” Research at Merck often begins as a very unofficial process of scientists recruiting one another to work on a promising project. Every substantial project starts with a zealot or a small group of zealots who entice others to join. Because people are attracted to the project and to the leader or leaders, they will work long hours. And the project becomes theirs. This unofficial drafting process may be somewhat slow in the beginning, but I believe that it pays off over the long run by creating a team with strong members. But team members and critical functions change over time. How do you manage that? Cross-functional interactions are crucial to drug development. It doesn’t matter if the world’s best biologists and chemists start the process—you must have the best people to carry it through every step of the way. The development process must be seamless or, I guarantee you, it will fail. Drug research projects take an average of 12 years and cost an average of $ 359 million. During that period, the disciplines critical to the project change. Projects often start with a biologist, and then a chemist will get involved. The drugs then have to be tested on animals and human beings, and, finally, the company has to manufacture them. A company cannot afford to lose a drug that has taken 10 years and cost millions of dollars just because there is a weak team member. Development projects need enormous amounts of expertise, and information must flow across disciplines. Years ago, for example, when we were animal testing a beta-blocker called timolol maleate—a very important drug for Merck—the early reports that came back were devastating. The drug was causing mammary tumors in animals. We had already invested 10 years and many millions of dollars, and it looked like we would have to scrap the whole project. Then a scientist from toxicology realized that the drug was causing the secretion of a hormone that was creating the tumors in animals, and he was able to demonstrate that the hormone secretion was not stimulated in human beings. The drug ultimately became Timoptic, a breakthrough drug for the treatment of glaucoma. In retrospect, it is clear that the project would have died had it not been for the toxicology department. It sounds like the kind of seamless information flow that Merck is hoping to achieve with Medco. The company that does the best job of understanding how information flows in the managed-care environment will ultimately prevail. The pharmaceutical industry is somewhat unique: the people who pay for our products—the plan sponsors—are not the people who prescribe the drugs—the doctors. And doctors are not the people who dispense them—the pharmacists. Moreover, the ultimate consumer is further removed because doctors and pharmacists don’t take the medication; patients do. So we have a complex marketplace with four customers to serve. We’re tapping into the information flow among all four in a way that allows us to serve one—the patient—better. About two years ago, I brought together eight of my senior staff members to look at the forces transforming the pharmaceutical industry. One of the most important changes we saw was the growing influence of prescription-benefits-management companies. PBMs are able to provide drugs to HMOs and large corporations at low costs because their high volume of sales and their ability to shift market share allow them to negotiate low prices with pharmaceutical companies. PBMs like Medco offer lower-priced drugs to plan sponsors by creating limited formularies, or lists of drugs that physicians are encouraged to prescribe based on therapeutic and cost factors. When medically appropriate, formularies favor less expensive drugs and encourage the substitution of generic pharmaceuticals for branded varieties. In classic terms of competition, we could see that the power of the buyers was growing: through their formularies, PBMs were changing the competitive dynamic of our marketplace by bringing together the person who chooses the drug and the person who pays for the drug. In the pharmaceutical industry, we have traditionally marketed our products by sending sales-people to doctors’ offices to tell them about new drugs. It became clear to us that the emergence of managed care and PBMs like Medco, with their ability to intervene in the prescribing and dispensing process, was introducing a whole new dimension to pharmaceutical marketing. Having salespeople visit doctors’ offices does not allow us to reach PBMs, HMOs, or plan sponsors—the major players in the emerging market. While salespeople remain critical for one-on-one interactions with doctors, we realized that we needed an entirely new method to deal with the changes we saw coming in the market. Our acquisition of Medco—and, in fact, our competitors’ new alliances with other PBMs—represents an attempt to gain additional market access and to focus our selling efforts. Let’s look at the numbers. Merck’s market share is currently about 9.5 % overall in the United States, but because some of the Medco formularies include Merck drugs, our products represent 11 % of the drugs that Medco sells. Our goal is to increase that figure. You can see that Medco gives us a tremendous opportunity. Don’t forget that the pharmaceutical industry is highly fragmented. No one has more than 5 % of the market worldwide, so the competition is fierce and the battlefield is crowded. When we acquired Medco in November 1993, it had $ 2.6 billion in revenue derived from 1,600 plans covering more than 30 million participants. With Medco’s information base and technology, we can track how these people are using their medicines. For example, we know that most people stop taking prescribed medication after a year, even if they have a chronic condition requiring continual therapy and even if someone else is paying for the medication. People with high blood pressure who are taking an antihypertensive, for example, will feel no different on or off medication, even though the drug may be saving their lives. “If patients use one of our drugs consistently, it’s as if we have created a breakthrough—that’s worth a lot to our shareholders.” With Medco’s extensive information systems, we can know when a patient has stopped taking medication and can alert his or her physician. Plan sponsors, who are paying for drugs, might figure that they will save money if people aren’t using the drugs. But in the instance of someone with high blood pressure, hospital bills could skyrocket if the patient stops taking medication, so sponsors are very happy to have us on the case. If a specific medication prevents certain patients from having strokes or other complications, the plan sponsor can save millions of dollars in health care costs. And the implications for Merck are enormous. If patients use one of our drugs consistently, it’s as if we have created a breakthrough—that’s worth a lot to our shareholders. But how will owning the distribution system, Medco, allow Merck to shift market share to its own products? I honestly don’t believe that anyone understands the effectiveness of Merck drugs on certain diseases or conditions and their safety profile as well as we do. Medco will help us understand the way our medicines are being used and will therefore allow us to manage their effectiveness and safety profile in a more systematic way. “Medco pharmacists make about 2 million calls a year to doctors…we can use these calls to ask doctors to choose our products.” Medco pharmacists make about 2 million calls a year to doctors either to tell them that the drugs they are prescribing for their patients conflict with other drugs the patients are taking or to ask them to switch to a less expensive alternative. This practice, along with the growth in managed care in general, has shaken the whole foundation of the industry and accelerated price competition. When it’s appropriate medically, we can use these calls to ask doctors to choose our products. How will you use the information you gather from Medco to compete differently? Medco positions us to move through four different stages of pharmaceutical-care management. The first is managed care. Through our managed-care division, we can offer plan sponsors a formulary of preferred products. For example, if a patient takes Zocor, a Merck medicine that lowers cholesterol, the copayment from the patient may be smaller and the cost to the plan sponsor will be lower than if the physician were to prescribe a competitor’s drug. The next step is risk sharing. For instance, we tell plan sponsors that if Proscar is offered to patients with benign prostate enlargement, we will guarantee that after a year, let’s say, their symptoms will either have improved or not gotten worse, or we will refund the cost of the drug. Sponsors and patients can’t lose. The third step is a capitation program. Instead of asking a company to pay for each drug its employees use, we offer to charge a flat rate of, for example, $ 10 per month per person. The price is based on an average cost of supplying drugs to a large population. Some patients require drugs that may cost up to $ 1,000 per dose, and some people use very few drugs. You have to have very good data to be able to project those costs, and we believe that Medco can provide those data for us. The fourth stage is disease management. We can tell plan sponsors that we will lower the overall costs of caring for certain patients by optimizing their drug usage. These patients may use more drugs than before, but at the end of the day, the total costs to the plan sponsor, including hospitals, doctors, emergency-room visits, and medication, will be lower. Can you give us an example of a disease-management program? Let’s look at just one disease: diabetes. Medco was successful as a supplier of mail-order drugs to diabetics who frequently needed drug supplies but also wanted to save money. A considerable cost of treating diabetics is insulin and other drugs, but there is a lot of evidence that if diabetics don’t also monitor their glucose levels and observe dietary guidelines, their other health care costs will increase dramatically. Using the information that Medco had about diabetes, and with the cooperation of plan sponsors and the permission of patients, we set out to identify the progress of some 100,000 diabetics who were part of Medco’s plan and to figure out how we could help them, their physicians, and their pharmacists minimize disease-related complications. Ninety percent of these patients agreed to enroll in a special program for diabetics. Once they enrolled, we began communicating with them about their disease. We sent out newsletters that focused on the importance of diet and exercise. We persuaded plan sponsors to cover the cost of the test strips used to ascertain blood-sugar levels, which are integral to controlling the disease but which most health care plans don’t cover. Then we set up an 800 number for diabetics to call, and we put together a group of six pharmacists who were trained to answer their questions. It turns out that three of those pharmacists are diabetics themselves. And now we’re putting together a risk-assessment study in which we ask these patients to tell us how they’re doing. The entire program is designed to minimize the risk that the disease will spin out of control in an individual patient as well as minimize the financial risk to Merck and to the plan sponsors. But this program will work only if we have the necessary patient data. Medco gives us access to the data of 38 million patients, and that’s a very good way to begin to see the patterns that will allow us to offer this service profitably. It sounds like the focus of your business is changing from research to information. Not necessarily. Information is critical, but now that we have Medco, our research arm takes on a whole new significance. We used to focus only on creating breakthrough drugs and getting to the market first with those drugs. If we were in a neck-and-neck research race for years but didn’t get our product to market first, it often wasn’t worthwhile to continue the project. But with our acquisition of Medco, all that changes. In today’s environment, developing a follow-on drug or a generic makes more sense. We will always be dedicated to achieving breakthrough drugs, but because we must now provide a broad range of medicines for our customers, there is new value for Merck in having a wide variety of products. Merck products are Medco’s house brand. Part of the secret to success will be to develop Merck medicines for categories in which we are not now represented—hence our new generics division and other initiatives. Consider what happens if instead of having 20 products on the formulary we have 40. The more Merck products on the formulary, the greater our power on the market and the more plan sponsors will save. “Now that we have Medco, our research arm takes on a whole new significance.” So Medco does change our focus slightly by making it attractive to continue research on products when it is clear that Merck’s entry will be the second one in a class. Yet there are industry observers who look at the state of the new-drug pipeline and see that it is a little bit dry and then look at the acquisition of Medco and say that the R&D focus of Merck could well get lost and that Medco could end up dominating the company. I don’t see that ever happening. First of all, I don’t agree that our pipeline is dry. We have major new-product candidates very close to market: Cozaar for the treatment of hypertension and heart failure and Fosamax for the treatment and prevention of osteoporosis. Following close behind them we have products for the treatment of asthma, irregular heartbeat, and glaucoma. I’m particularly excited about an oral drug that will stimulate the body to secrete its own growth hormone, potentially an important breakthrough for Merck. I am reminded of what people were saying when I joined Merck. Back then, some thought we should diversify out of pharmaceuticals, but we certainly wouldn’t be the company we are today if we had taken that path. Today Merck could focus just on distribution, but losing the R&D capability would put the company at a competitive disadvantage. Worst of all, such a move would rob society of its very best source of new medicines. Only a few pharmaceutical companies will survive the restructuring that has already begun, and the ones that do will have to excel at both research and the new distribution methods. You cannot succeed without mastering both. But you don’t master those competencies independently; you must get them to work together. How do you meld the entrepreneurial, information-driven culture of Medco with Merck’s more conservative scientific culture? So many mergers throughout business history have failed because managers either didn’t recognize the complexity of bringing two cultures together or they didn’t act quickly enough to fulfill the potential of the merger. Just as most research projects fail, most mergers don’t live up to expectations. With Medco, we must move quickly on at least three fronts simultaneously. First, we know that market opportunities won’t exactly follow our blueprint, so we have to be flexible. For example, when we announced our acquisition of Medco in July 1993, we thought that we’d leave Medco as a stand-alone Merck division, but we quickly changed our minds. By December, we decided to create a new division by integrating Merck’s managed-care business, which covers 40 million people, with Medco’s managed-care business. This new entity gives us tremendous market coverage and allows us to offer a wide range of services in the managed-care marketplace. “Most mergers don’t live up to expectations.” Second, we’re evaluating the most appropriate way to link Merck’s traditional marketing business with the new division so that there is a flow of information about Merck products, customer development, and pharmacy relations. Third, and most important, we are moving quickly to place some of Merck’s best executives in Medco. Medco people have a lot of valuable knowledge and experience to offer Merck people, and vice versa. We think we have all the elements in place, but the key to making the merger work will be speed, flexibility, and the right personnel. Can you give us the benefit of your experience as a CEO and tell us the most important lessons you have learned? As I look back, I can see that there are two issues every CEO must face: the first is recruiting the best people out there, and the second is adopting the right focus for the company. Let’s start with recruiting. I think it is the most important job any CEO does, and I’ve worked on it relentlessly. When you recruit the best people, you create a family. You want these people to take off on their own, and the faster they take off, the faster you should back away. The key is knowing when to intervene and when to step aside. I guess that’s where I have made the most mistakes. There were times when I could have told people what to do and I didn’t. And there were other times when I probably squelched people by stepping in too soon and telling them to do things my way. Striking that crucial balance is especially difficult when you are working with highly trained knowledge workers like scientists and the first-class executives we have at Merck. A second problem that can be tough for CEOs is finding the right kind of focus for their companies. At Merck, we’ve been praised for taking the long-term view consistently, and yet we may have abandoned some projects too soon. In some developing countries in the 1980s, for example, our prices were regulated, inflation was out of control, and our patent protection was weak. We were losing money, so in some cases we just got out. We sold some of our Mexican operations in 1986, for example, but we started buying our way back into that country last year when we realized that Mexico’s economic and political reforms offered a burgeoning market. In retrospect, we might have been better off if we had just stayed put and absorbed our losses. Given the benefit of that experience, Merck is now taking a much longer-term view in certain markets, such as China. One major focus there has been hepatitis B, an infection that is carried by 150 million people in China. Ten percent of hepatitis B cases lead to liver cancer, the country’s leading cause of death. Most often, the disease is transmitted from mothers to their newborn infants, and therefore the Chinese would like to immunize all newborns against the infection. Three doses of our recombinant hepatitis B vaccine are required to immunize a child, and roughly 20 million children are born each year. We expect that it would take 20 years to eradicate the disease. Selling 60 million doses of our vaccine every year for 20 years would represent a huge market, but the Chinese government could not afford to buy our vaccine, and the epidemic was running wild. So in 1989 we made a strategic and humanitarian decision to sell the Chinese the technology to make the vaccine and then teach them how to make it. There are many ways to think about a product in an information-intensive industry. The product is the vaccine, but the product can also be the knowledge needed to make it. In either case, the challenge is getting information to flow across borders, for example, from one lab to another in the same company or from a research to a marketing division. In this case, we needed to get information from Merck to China. So we sold the technology to China for several million dollars—a minimal amount compared to what we would have made if we had sold the finished product—and then we had to teach the Chinese how to produce the vaccine. Scientists and engineers from China came to our facility in West Point, Pennsylvania, and together we built two plants and operated them for three production cycles to ensure that the product was up to Merck’s standards. Then we dismantled the plants and sent them to China. Several of our engineers went to China and helped the Chinese get the plants running. One opened in Beijing in 1993 and the other in Shenzhen in June of this year. We got involved in this project because we recognized the seriousness of the health problem, and as a doctor it is very hard for me to turn my back on that. We at Merck try to live by that ethic. But we also know that we can use information to sell products, educate people, and, ultimately, build our brand in China. Our hope is that the long-term relationship we have built and the goodwill we have created will help our company there in the long run. If not, we’ve still done the right thing. But like everything else we’ve done, it’s a big bet. I hope that Merck will always have the courage to place big bets. I hope that’s my legacy.
