I used to read a lot of management books. Or rather, buy a lot of them, then take a quick look — and promise myself to get around to reading them someday. When (and if) I finally did, I’d often find the information dated and the advice stale.
Not so with most of Peter Drucker’s work. In Drucker I find things that sound timely — sometime prophetic — even today.
Drucker spoke of the productivity of knowledge, and how firms, industries, and even countries who managed to excel at it would over time come to dominate their respective sectors.
When I first heard him use that term, it sounded abstract — and not immediately relevant to my work in business strategy. That changed recently when I conducted a study of the innovation crisis in the US pharmaceutical industry (which represents 80% of the industry’s R&D worldwide.) This is widely known in the press as the patent cliff that affects most major pharma manufacturers to some extent.
In a nutshell, a significant portion of the industry’s revenues and profits depends on branded drugs — the more successful of which you see advertised on TV. Most of these are discovered by rapidly scanning lots of molecules, or computer models of the molecules, for potentially beneficial properties. The most promising of these are first matched with diseases for which they might be useful, then registered under patents. Then they are developed by being rigorously tested, first on animals, then on humans in a range of clinical trials.
The data from all these tests are presented to the Food and Drug Administration (FDA) for approval. Drugs that are approved, and that are added to the ‘formularies’ of insurance companies and government payers who will reimburse patients for their use, can becomes wildly successful. Lipitor, the statin used to control cholesterol in the blood, is the most successful drug to date, with total sales over $125 billion.
These blockbuster drugs (generally defined as those with annual sales over $1 billion) command prices in the marketplace much higher than their cost to manufacture. Their gross margins can be as high as 75-80 percent.
This all changes about 20 years later when the patent protection on a drug legally expires. Though sometimes ‘extensions’ can be filed to effectively lengthen the patented life, revenues eventually fall off a steep cliff as the drug typically becomes available as an equivalent ‘generic’. The manufacturer can lose as much as 20% of it total sales if several of its drugs come off patent in a given year. At this point the original developer often out-licenses the drug for distribution and sales by another company.
The branded drug industry depends on a huge ongoing R&D effort to discover and develop new drugs. This process can take more than a decade, and the average cost per drug is about $1.8 billion. But given the rich margins that can result, some pharma companies spend over 25% of their revenues on R&D — much higher than in other industries.
Many molecules drop out of development along the way, and even drugs that are introduced commercially are not guaranteed to generate large revenues.
The problem for the industry is that successful drugs coming off-patent are not being replaced by equally successful new drugs. One study found that, for each dollar of drug sales being lost due to patent expiration, only about 26 cents is being replaced by sales of new products.
This is in spite of the fact that the industry spends more every year on R&D. Innovation productivity in pharmaceuticals — and here we are very close to Drucker’s ‘knowledge productivity’ — has consistently declined over a period of more than 50 years. Not just a little, but drastically. If you measure FDA drug approvals per $1 billion (inflation-adjusted) of R&D spending, the figure has gone down from about 64 new drugs ‘bought’ by the billion in the early 1950s to a little less than 1/2 today. By this measure, pharma innovation productivity is less than one percent of what it was then.
Industry observers have dubbed this “Eroom’s Law”, since it seems exactly the converse of the productivity gains consistently seen in computer chip manufacturing (“Moore’s Law”).
There are many reasons for this, some not having to do with industry practices. Most notably, the FDA demands more information in order to assure the safety of the drug-taking public — especially in light of near-disasters like they experienced with Vioxx a few years ago. This takes more time and expense, and decreases the efficiency of the process.
Another factor that has been mentioned is that the industry has done much expanding by acquisition during the past decade — Pfizer alone bought Pharmacia, Warner-Lambert, and Wyeth — for the purpose of building a huge R&D effort (which now costs over $7 billion a year there, as it also does at Novartis, Merck, and Roche.)
But new studies have shown that some aspects of the discovery/development processes do not scale well — contrary to expectations, they become less efficient in larger operations. People are not able to meet as often to share information, and ideas are not passed as freely.
All of these ‘explanations’ stacked together seem insufficient to explain a 99-percent drop-off in fruitful innovation. There’s no punch line here — as there is no easy fix for the industry. With the downward pressure of the Affordable Care Act and Medicare pricing on health care prices in general, the challenge for the pharmaceutical industry lies in fundamentally rethinking their approaches to innovation.
There’s more on this in my recent article Innovation: The Moneyball Test in Pharmaceutical Executive.
As we begin our eighth continuous year of publication, we thank you for your readership and support. Please accept our best wishes for happiness, health, and success in the New Year.