Absolutely, if you're looking at all the drugs covered under Medicare Part D. It's happening right now with devices and with almost everything covered under Part D that's not a pharmaceutical product. So it could be your blood test strips for diabetes. It could be a wheelchair. It could be a bed for somebody who's at home being cared for. What the government has done is basically put everything into reference categories—just like they've done with drugs in Germany and Spain.

The first bids happened this year. Then they'll take the average price of the winning bid and say, "That's the price that we're going to reimburse at."

Now, somebody can choose to get a more expensive device, but they're going to have to pay for it out-of-pocket. As the baby boomers roll into retirement and the cost of Medicare Part D skyrockets, I can certainly imagine them doing something similar with drugs, where there's only one drug in each category and if you want any other drug, you pay the difference.

In this market, price has to be justified by economic value. And the economics of a drug for one indication is not necessarily consistent with the economics for another indication. And if you realize that from the get-go, there are a number of things you can do. You might kill one of the indications because it's not going to be economically viable. Or, you might lower the cost per unit, either modifying the drug itself—which would be very costly—or the way we package it—which could be less costly.

Let's consider Avastin, which was first launched for colon and rectal cancer. Genentech came up with a price that was pushing the limit but still justified from a value perspective, and that price was about $35,000 for the course of treatment. Everyone swallowed hard, but they accepted it.

A few years later, Avastin got approved for lung and breast cancer. Researchers had tested it using dosages that are twice as high for breast and lung cancer—the typical amount for those indications—and now a course of treatment costs $70,000. But the benefits are no greater than they are for colon and rectal cancer. It's going to add the same three months to your life, but you're getting charged twice as much.

Now from a value perspective, you can't tell that story. You can't explain why a month of life is worth twice as much when you have lung cancer than when you have colon cancer. But if Genentech had thought about this in advance, there are a couple of things it could have done.

For starters, it could have packaged a larger dose for lung and breast cancer and charged only 10 percent more. Unfortunately, in clinical trials, you're not only testing your drug's active ingredients, but you're also testing everything—right down to the packaging. So if you wait until the last minute to think about pricing, it's too late. If you decide you want to change the way you've packaged the drug, you've got to go back and do more tests. However, when you have pricing and reimbursement people involved early on—as you're designing your Phase III trials—they're able to make sure all the pieces fit together in a commercially logical way. And, amazingly, very few companies are doing that systematically.

What's striking is that you don't bring up the argument about recouping the cost of R&D, which was almost a mantra of the industry for the last several years.