07-041 April 30, 2007 Eli Lilly's Project Resilience: Anticipating the Future of the Pharmaceutical Industry Rebecca M. Henderson It was early May 2004, and Peter Johnson found himself looking forward to the Senior Management Forum that he was scheduled to moderate at the end of the month. As the Executive Director of Corporate Strategic Planning at Eli Lilly and Company (Lilly), it was his job to support the company's senior team in the kinds of in depth strategic discussions that were fundamental to maintaining Lilly's leadership position in the rapidly evolving pharmaceutical industry. Nearly twelve months before, in response to the widespread perception that the industry was facing an unprecedented set of challenges, Peter had been asked to undertake a comprehensive review of how the industry was likely to evolve - and whether Lilly should consider making significant changes in core strategy and/or its business model as a result. In response, Peter had pulled together a cross functional team of high profile people from across the company to look at the key environmental uncertainties facing the industry, and to answer the question, \"What might Lilly need to do to compete in the future?\" The team - calling itself Project Resilience after Gary Hamel's article \"The Quest for Resilience\" - had divided its work into three interrelated phases: scenario planning, business model evaluation, and core capabilities assessment. How might the future evolve? Scenarios Capabilities Required How could Lilly compete? Alternative Business Models Competitive Dynamics What should Lilly do? Alternatives for Lilly Recommendations The first two phases had taken months of work, but were now largely complete. Peter was fairly confident that the team had a good sense of the different directions in which the industry was likely to This case was prepared by Rebecca M. Henderson. Professor of Management. Professor Henderson is the Eastman Kodak Leaders for Manufacturing Copyright 2007, Rebecca M. Henderson. This work is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 Unported License. To view a copy of this license visit http://creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA. ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson evolve, and of the ways in which Lilly - and its competitors - might be able to compete under the different scenarios that the team had described. Now, however, the team was grappling with the last, and toughest phase - what should Lilly do? They needed to come up with some concrete recommendations for Lilly's senior management as to which business models the firm should explore and how Lilly should develop new capabilities to position the firm for the future. Peter sighed as he turned back to his desk. It was going to be a long night. Turmoil in the Pharmaceutical Industry? On the surface, the global pharmaceutical industry appeared to be in robust health. Global sales for the industry were approximately $550 billion in 2004 (Table A). About 42% of this total could be attributed to members of PhRMA, the Pharmaceutical Research and Manufacturers of America, a group that included the majority of all U.S. firms. Net incomes were also very robust, and in general pharmaceutical stocks had performed well since the stock market crash of 2000. Table A 2004 Worldwide Pharmaceutical Sales by Region Region Total Worldwide Market North America (U.S. & Canada) European Union Japan China All Other Sales (US$ billions) $550.0 $248.0 $144.0 $5.8 $9.5 $90.5 % Growth vs. 2003 7% 8% 6% 2% 28% NA % Share 100.0% 45.1% 26.2% 1.7% 1.7% 16.5% Source: MedAd News, May, 2005. At the same time the industry was coming under significant pressure on a variety of fronts. Most noticeably, there was some evidence that the research productivity of the industry was declining. Measuring pharmaceutical research productivity is notoriously difficult, since it can take as long as 10-12 years to bring a new drug to market and the number of drugs released in any given year reflects investments and actions made over many years. Nevertheless it appeared that while real research spending in the industry was accelerating dramatically, outputat least as crudely measured by the number of new drugs introduced every yearwas falling (Exhibit 1). For example, the cost of discovering, developing and launching a drug which was roughly $318 million in 1990 1 topped $1.7 billion in 2003. 2 Meanwhile, one in 13 drugs in the preclinical development stage reached the market 1 Bruce Rasmussen, Implications of the Business Strategies of Pharmaceutical Companies for Industry Developments in Australia, Working Paper No. 1, Center for Strategic Economic Studies, Victoria University of Technology, March. 2002. 2 \"Drug Development Costs Hits $1.7 billion,\" DrugResearcher.com, December 8, 2003. April 30, 2007 2 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson compared to one in eight from 1995-2000. 3 In turn, the industry was witnessing the emergence of smaller, more nimble companies which specialized in particular therapeutic areas and which, due to limited financial resources, built their pipelines by relying on products and molecules that were already available in the marketplace. 4 Simultaneously, advances in tools such as genetics and genomics had led to waves of entry by small, specialized \"biotech\" firms. While these firms had yet to yield significant returns to their investors according to some observers, aggregate returns to the venture capital that had been invested in the industry were barely positiveaccording to the Boston Consulting Group, while they represented only 3% of the drug industry's total R&D spending, 67% of the drugs in clinical trials in 2003 were from small biotech companies. 5 Industry growth was also under attack. In 2004, the global pharmaceutical market grew only 7%, marking the first time the industry had not reached double digit growth since 1995, 6 and over the next five years $40 billion of branded pharmaceuticals were expected to lose patent protection. 7 These trends were placing significant financial pressures on large pharmaceutical companies like Lilly. Average pharmaceutical revenues for a large firm were on the order of $10 billion/year, and their valuations implied that the equity markets were expecting them to grow on the order of 10% a year. Since the typical successful new product yielded revenues of around $300-$400 million a year, each company required roughly two to three new launches annually to meet expectations. Unfortunately, while success rates varied dramatically across companies, on average the large firms were introducing major products at a rate of roughly one every two years. At the same time, the conventional technologies of drug discovery were being challenged by an explosion of new science. Advances in genomics and genetics, imaging technology and fields such as protein chemistry, molecular biology and cellular mechanics were opening up promising new areas of enquiry. Some scientists were calling for an entirely new field of \"systems biology.\" In January 2003, for example, MIT's Computational and Systems Biology Initiative held its first conference, attracting 300 participants. There was also some evidence that scientific advances and the pressure on the industry were causing a move away from large, primary care blockbuster products. The heart of the \"Fully Integrated Pharmaceutical Company\" (FIPCO) model in which each company had its own discovery, development, manufacturing, sales and marketing functions for the majority of products in its pipeline, the blockbuster model relied on the development and distribution of a small number of drugs that could achieve global sales in excess of $1 billion annually by focusing on very large markets. 3 \"Drug Development Costs Hits $1.7 billion,\" DrugResearcher.com, December 8, 2003. 4 Celia M. Henry, \"Morphing the Model,\" Chemical and Engineering News, March 7, 2005. 5 Catherine Arnst, \"The Waning of the Blockbuster Drug,\" BusinessWeek, October 18, 2004. 6 Michael Rosen, \"Though Pharma Growth Slides, Blockbusters Reach New Records,\" Wisconsin Technology Network, June 6, 2005. 7 \"Evolving Generic Competition\" Global Competitive Intelligence report, Jan 2004. April 30, 2007 3 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson (See Exhibit 2 for a list of some blockbuster drugs.) In 2001, 35 blockbuster drugs accounted for on average 46% of the top 10 pharmaceutical companies' sales. 8 (Exhibit 3.) The success of the blockbuster model depended on achieving large returns from a small number of drugs in order to pay for the high cost of drug discovery and development. An aggressive sales and marketing strategy was paramount. The larger pharmaceutical companies relied on their 10,000 to 15,000 person sales forces to get products into the hands of doctors. At the same time, since the Food and Drug Administration (FDA) had eased direct-to-consumer (DTC) regulations in 1997, pharmaceutical companies were attempting to reach consumers through television and Internet advertising. Merck's VIOXX and Pfizer's Celebrex, a new class of prescription strength pain killers, were the first to ride this wave. In 2000, Merck's DTC budget for VIOXX reached $161 million, $40 million more than PepsiCo spent marketing its Pepsi product. 9 Within a year of being launched, VIOXX and Celebrex together had captured 40% of the market from traditional anti-inflammatories like ibuprofen. 10 Financially, the blockbuster model made pharmaceutical companies' share price particularly vulnerable. In 2001, Merck's announcement that sales of several of its blockbuster drugs would not meet expectations and that there was a 12-month gap in the blockbuster pipeline was met with a 15% reduction in the company's share price. 11 Revenue for VIOXX, however, topped $2.6 billion that same year, a 44% increase over 2000. When Merck withdrew VIOXX from the market in September 2004, because of concerns about patient safety, Merck's stock lost nearly 25% of its value. As in other industries, there was also a new and growing trend towards developing more \"personalized\" products and services. The success of Genentech's drug Herceptin had made this approach particularly salient. Herceptin was approved in 1998 for the treatment of metastatic breast cancer. For patients with the appropriate genotype (something that could be determined with the aid of an appropriate diagnostic test known as the HercepTest), Herceptin appeared to perform significantly better than common alternative treatments; for other women the drug performed no better than the standard treatment. The drug, therefore, served a smaller market than a more conventional therapy, by some estimates 15% to 20% of breast cancer patients. 12 Herceptin, which would not have reached the market without an accompanying diagnostic test enabling doctors to identify patients whose gene type made them eligible for the therapy, was one example of the synergies that could be created between a therapeutic and a diagnostic. 13 8 Bruce Rasmussen, Implications of the Business Strategies of Pharmaceutical Companies for Industry Developments in Australia, Working Paper No. 1, Center for Strategic Economic Studies, Victoria University of Technology, March. 2002. 9 Stephen P. Bradley and James Weber, \"The Pharmaceutical Industry: Challenges in the Next Century,\" Harvard Business School Case No. 703-489. 10 Barry Meier, \"Medicine Fueled by Marketing Intensified Trouble for Pain Pills,\" The New York Times, December 19, 2004. 11 Bruce Rasmussen, Implications of the Business Strategies of Pharmaceutical Companies for Industry Developments in Australia, Working Paper No. 1, Center for Strategic Economic Studies, Victoria University of Technology, March. 2002. 12 Matthew Bell, \"Unraveling the Pharmaceutical Industry,\" Arthur D. Little, 2002. 13 \"Theranostics: Guiding Therapy,\" Med Ad News, December 1, 2004. April 30, 2007 4 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson While Herceptin appeared to have been quite financially successful for its developer, Genentech (a full course of treatment is currently priced at about $70,000, and in 2006 total revenues from Herceptin were $1.2 billion), many industry observers were concerned that many \"personalized\" drugs were likely to be much less financially attractive than conventional blockbuster drugs. The trend of using biomarkers and diagnostics to better segment patient populations would inevitably result in fewer patients for a given therapy. Furthermore, many researchers were beginning to realize that complex illnesses such as cancer and cardiovascular diseases, required drugs that hit more than one target simultaneously. 14 There were also concerns that the conventional blockbuster \"FIPCO\" model was at risk due to further changes in the pharmaceutical marketplace. Here, two trends in particular were key. The first was the impact of patent expirations, as generic competition was continuing to raise the competitive hurdle for branded prescription (or \"Rx\") products, especially in large and crowded primary care markets. The $40 billion of branded pharmaceuticals that were expected to lose patent protection in the next five years would, for example, create significant competition for new drugs. The second was the increasing saturation of the main selling channel for the blockbuster FIPCO modelthe sales representative detailing individual physiciansresulting in declining returns on investments in this channel. Both physicians and payers were objecting with increasing volubility to the promotional practices required by the blockbuster FIPCO model. The industry was also coming under increasing political pressure. Health care expenditures as a percentage of GDP were rising dramatically, and pharmaceutical costs - which were sometimes not fully reimbursed by either federal or private health plans - were becoming increasingly visible. (See Exhibit 4.) At the same time, some of the practices of the industry were coming under scrutiny. For example, in \"The Truth about Drug Companies\" Marcia Angell - who had been Editor in Chief at the New England Journal of Medicine - made the following claims: 14 The pharmaceutical industry claims to be innovative, but only a small fraction of its drugs are truly new: most are simply variations on older drugs. Contrary to popular belief, big drug companies spend far less on research and development than on marketing. The pharmaceutical industry has an iron grip on Congress and the White House. It has the largest lobby in Washington ... and contributes heavily to political campaigns. Drug companies promote diseases to match their drugs. Millions of normal Americans have to come to believe that they have dubious or exaggerated ailments like \"generalized anxiety disorder.\" Drug companies have enormous influence over what doctors are taught about drugs and what they prescribe. Simon Frantz, \"Playing Dirty,\" Nature, October 13, 2005. April 30, 2007 5 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson Drug companies have substantial control over clinical trials of their drugs. There is good reason to believe that much of the company supported research on prescription drugs is biased as a result. These kinds of claims were hotly contested by the pharmaceutical companies themselves, but they nonetheless signaled a potentially troubling erosion of public support for the industry. Yet another threat on the horizon was the possibility of aggressive entry from India and China. Companies such as Ranbaxy Laboratories had set their sights on becoming \"International researchbased pharmaceutical companies\" and by some estimates, Chinese pharmaceutical firms sold over $19 billion worth of product in 2002. Eli Lilly: More than 125 years of history Eli Lilly and Company was founded in May 1876 by Colonel Eli Lilly in Indianapolis, Indiana. A 38year-old pharmaceutical chemist and a veteran of the U.S. Civil War, Colonel Lilly was frustrated by the poorly prepared, often ineffective medicines of his day. Consequently, he made these commitments to himself and to society: He would found a company that manufactured pharmaceutical products of the highest possible quality. (Lilly was the first pharmaceutical company to ask physicians and hospitals to give feedback on the efficacy and safety of its products. 15 ) His company would develop only medicines that would be dispensed at the suggestion of physicians rather than by eloquent sideshow hucksters. Lilly pharmaceuticals would be based on the best science of the day. Eventually, Colonel Lilly's son, Josiah K. Lilly Sr., and two grandsons, Eli Lilly and Josiah K. Lilly Jr., each served as president of the company. Each contributed a distinctive approach to management, and together, these management styles established a corporate culture in which Lilly employees were viewed as the company's most valuable assets, a belief that the company claimed was still the cornerstone of its corporate philosophy. While some analysts had speculated that Lilly's location in Indianapolis put it at something of a disadvantage, the company believed that its strong Midwestern roots and deep history gave it a unique competitive advantage. The company described its values as: Respect for people, which includes our concern for the interests of all people worldwide who touch or are touched by our company: customers, employees, shareholders, partners, suppliers, and communities; Integrity that embraces the very highest standards of honesty, ethical behavior and exemplary moral character; 15 Margaret L. Eaton, \"Developing and Marketing a Blockbuster Drug: Lessons from Eli Lilly's Experience with Prozac,\" Stanford Graduate School of Business Case No. BME-6. April 30, 2007 6 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson Excellence that is reflected in our continuous search for new ways to improve the performance of our business to become the best at what we do. With 2004 revenues of $13.9 billion, 42,000 employees worldwide and medicines marketed in 142 countries, Lilly was one of the world's 20 largest pharmaceutical companies. (Exhibit 5 gives key financial information for the firm.) The company, which prided itself on its strong record of sciencebased research productivity, spent more on R&D as a percentage of sales (21%) than any other major pharmaceutical company. (GSK spent 15% and Merck 14%. 16 ) It had major research and development facilities in nine countries and conducted clinical trials in more than 60 countries. Lilly's four therapeutic areas included neurosciences (44% of revenue), endocrinology (31%), oncology (10%) and cardiovascular (5%). 17 Zyprexa, a treatment for schizophrenia, bipolar mania and bipolar maintenance, accounted for 32% of Lilly's revenue topping $4.4 billion. Several lawsuits brought during 2004 contesting the validity of Zyprexa's patent and some of Lilly's marketing and sales efforts surrounding the drug caused the company's stock price to fall 19% during the year. Lilly was one of the only major pharmaceutical companies not caught up in the merger and acquisition activity of the late 1980s and 1990s. (See Exhibit 6.) The company was able to safeguard its independence by looking internally for core capabilities it could develop and exploit including improving speed to market, leveraging existing products, narrowing its R&D focus from eight to five therapeutic areas, spinning off its non-core medical device and diagnostic businesses, and creating multi-functional, product-focused teams (known internally as heavyweight product development teams) which focused exclusively on the development of a single compound. This included all activities related to drug discovery, manufacturing, sales, marketing and distribution. 18 Meanwhile, sales of the company's blockbuster anti-depressant Prozac, which topped $2.8 billion in 1996 (43% of revenue) despite bad press and lawsuits, enabled the company to heavily invest in new product development, further safeguarding its independence. These efforts appeared to pay off. In 2002, two years after Prozac's patent expiration, Lilly was immersed in the most productive new drug launch in its 127-year history and in 2004 Lilly launched five new products, of which three were \"first in class\" - or the first drugs to reach the market exploiting a particular \"mechanism of action.\" First-in-class drugs tended to be more difficult to manage than already established therapeutic classes due to the fact that knowledge about the disease, drug and market was far less certain. 19 (Exhibit 7 shows major introductions by year for the period 1983-2004.) 16 Jeff Swiatek, \"Eli Lilly Looking for Forumula to Cut Drug Costs,\" The Indianapolis Star, October 18, 2004. 17 Animal Health and \"other pharmaceuticals\" made up the remaining 7% of revenue. 18 Matthew C. Verlinden, \"Eli Lilly: The Evista Project,\" Harvard Business School, Case No. 699-016. 19 Margaret L. Eaton, \"Developing and Marketing a Blockbuster Drug: Lessons from Eli Lilly's Experience with Prozac,\" Stanford Graduate School of Business Case No. BME-6. April 30, 2007 7 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson Building the Scenarios for Project Resilience Scenario planning is a disciplined method for thinking about the future and its inherent uncertainties. Peter Schwartz, president of the Global Business Network and a leading expert on scenario planning, calls scenarios \"stories about how the world might turn out tomorrow,\" and adds that the purpose of scenario planning is \"not an accurate picture of tomorrow, but better decisions about the future.\" 20 To accomplish this, scenario planning creates a number of different future worlds that are both plausible and also sufficiently differentiated so as to enable decision makers to compare and contrast them and their strategic implications. For Project Resilience, the working team created four scenarios for the future of the research-based pharmaceutical industry that were set in 2020, approximately 15 years away. Given the industry's long R&D cycles, the team decided that 15 years was enough time to allow Lilly's senior managers to imagine future worlds that were substantially different from the current industry environment. The team also decided to focus their scenarios on the U.S. market, both because of its overriding size and importance to the sales and profits of the pharmaceutical industry, (in 2004, more than 50% of Lilly's sales were in the U.S., and some industry observers speculated that the U.S. market accounted for a disproportionate share of profits for every major pharmaceutical firm) and also because the external environment in the U.S., which for many years had been relatively stable and supportive of the industry's business model, was becoming increasingly turbulent. If the pharmaceutical business model was going to be forced to change, the team reasoned, it would be because of what happened in the U.S. market. Understanding how this market might evolve was a critical first step in assessing future strategic options for the industry in general and for Lilly in particular. The scenarios were developed by first selecting the two most critical and uncertain of the many external factors (called, in the language of scenario planning, \"driving forces\") that could impact the future evolution of the pharmaceutical industry. The project team tested a number of possible driving forces before ultimately selecting \"R&D Output\" and \"Rx Purchase and Prescribing Decisions\" as the two forces to use for their scenario planning exercise. They reviewed their choice with both senior Lilly colleagues and outside consultants who were well versed in both scenario planning and the pharmaceutical industry environment. These two \"driving forces\" became the scenario axes, with each end of an axis representing opposite outcomes of the driving force that the axis describes (see Figure 1). For example, one end of the \"R&D Output\" axis represents R&D output that produces breakthrough innovation, while the other end represents output resulting in incremental innovation. 20 For more on scenario planning, see, fro example, Peter Schwartz's book The Art of the Long View (Currency: 1996). April 30, 2007 8 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson Figure 1 R&D Output Rationing Innovation Breakthrough innovation Haves and Have-nots Tight formularies and utilization controls Innovation \"rationed\" to patients judged most likely to benefit; Restrictions on off-label prescribing by MDs Right drug to right patient thru accepted treatment algorithms developed by credible thought leaders Companies must demonstrate big improvements over current therapies to obtain price premium for new products Products in new therapy areas most likely to command price premiums HSAs in widespread use; patients pay out-of-pocket for many Rx drugs Middle class and rich can supplement employer HSA contributions (pre-tax and after tax) to ensure access to new therapies; poorer consumers cannot Increased consumer price sensitivity and access inequities create pricing pressures on new therapies At same time, very expensive biotech products covered thru high-deductible catastrophic insurance plans, so pricing pressure on these products is less acute Rx Purchase and Prescribing Decisions Centralized (public or private) Individual (patient/prescriber focus) Price Sensitive Patients Payers Rule Rx prices regulated by govt (price ceilings) Very tight formularies and utilization controls Even with price regulation, companies must offer big discounts to \"play\" on formularies Restrictions on promotional spending and programs FDA focus on safety increases cost of clinical trials Massive pharma industry consolidation Employers continue to pass ever larger share of healthcare costs onto employees, so patients are price sensitive Most seniors still have significant out-of-pocket drug costs Heavy use of generics, OTC products (FDA speeds Rx to OTC switching in many categories) and alternative medicines Small share of population willing to pay premium for branded products based on heavy DTC marketing efforts; most not Companies focus on NILEX and promotional efforts to drive sales Incremental Innovation The axes were defined as follows: Vertical Axis: R&D Output (measures both quantity and quality of industry innovation) Breakthrough Innovation: Pharmaceutical industry R&D is very productive, with 40-50 NMEs submitted annually for regulatory approval. Many of these NMEs \"New Molecular Entities\" (>30%) offer major advances in safety, efficacy, and/or \"customization\" (tailored therapeutics) vs. currently available therapies. (By contrast, the industry currently produces about 20 NMEs per year, with approximately 15% receiving priority consideration from the FDA.) Incremental Innovation: Pharmaceutical industry R&D productivity remains stagnant, with about 20 NMEs submitted annually for regulatory review. Most of these NMEs (approximately 90%) do not represent significant advances in safety, efficacy or \"customization\" (tailored therapeutics) vs. currently available therapies. April 30, 2007 9 ELI LILLY'S PROJECT RESILIENCE: ANTICIPATING THE FUTURE OF THE PHARMACEUTICAL INDUSTRY Rebecca M. Henderson Horizontal Axis: Rx Purchase and Prescribing Decisions Individual Decision Making: The decision about which Rx therapy will be used is determined primarily through interactions between individual patients, who are paying most of the cost of their medicines, and prescribers. Centralized Decision Making: The decision about which Rx therapy will be used is determined primarily by the government health system or by a consolidated group of private insurers who are paying most of the costs of Rx drugs. Putting the two axes together defined four future worlds: \"Haves and Have-nots\