Given all the hopes for medical progress that ride on biotech progress, one might assume that Congress and the administration would seek ways to encourage investment. One would be wrong.

In San Francisco this week, the Biotechnology Industry Organization (BIO) is holding its eighth annual investors’ forum. The relentless optimism usually characteristic of corporate communications is missing these days:

“We are coming out of period of intense pressure and conservatism in life sciences investment,” a BIO conference ^[1] press release said.

“Private equity and capital markets are increasingly failing to fund promising early-stage scientific research, primarily because it is viewed as too high-risk,” Martin Sabarsky, chief financial officer and chief operating officer of HR BioPetroleum, said on behalf of BIO before a congressional hearing. But:

“While today’s uncharted economic environment and evolving legislative landscape pose major challenges . . . we remain optimistic about long-term growth,” said John L. Craighead, BIO’s managing director of investor relations and business development.

Across the continent, Congress is also thinking about biotech. Given the concerns about investment and the depressed state of the industry (which usually lost $5 billion or so per year even before the crisis of 2008) and given all the hopes for medical progress and human betterment that ride on biotech progress, one might assume that Congress and the administration would be seeking ways to encourage investment.

Nah! Henry Waxman, chairman of the House Energy and Commerce Committee, and the administration want to reduce protection of the biotech industry’s intellectual property. As Theodoric of York, Medieval Barber (aka comedian Steve Martin) would have said, “what this industry needs is a good bloodletting!” which was Theodoric’s cure for everything. This prescription is likely to be about as successful as Theodoric’s ministrations.

Biotechnology is an evolution of the pharmaceutical industry. Basic pharmaceuticals are small-molecule chemical substances that can be manufactured by mass production techniques akin to those used in other large industries. Biologicals are composed of larger and more complex molecules based on Recombinant DNA, which involves “making proteins such as human insulin and other therapies in cultured cells under controlled manufacturing conditions.”

Since its origination in the 1970s, the biotech industry ^[2] has created 200-plus new therapies and vaccines, including products to treat cancer, diabetes, HIV/ AIDS, and autoimmune disorders. It has 400-plus substances currently in clinical trials, including treatments for heart disease, cancer, AIDS, Alzheimer’s, Parkinson’s, stroke, septic shock, diabetes, anemia, cystic fibrosis, multiple sclerosis, lupus, kidney disease, and liver disease. For many of these conditions, current treatment options are limited. The industry also claims credit for“hundreds of medical diagnostic tests that keep the blood supply safe from HIV and detect other conditions early enough to be successfully treated.”

The lifeblood of biotech is research and development (R&D), as embodied in protectable intellectual property. The Supreme Court’s 1980 decision in Chakrabarty was a key step in its development, because the case’s broad recognition of patentability ^[3] “provided a profound contribution to the growth of biotechnology . . . and enabled our country to become an international leader.”

Biotech, like the pharmaceutical industry in general, requires heavy up-front investment before the first unit of a product goes out the door and rather low marginal cost for every subsequent unit. It costs over a billion dollars to get a product to market, and fewer than 1 percent of candidates survive the obstacle course, and that only after about 10 years of development, so the overhead burden on the successes is very high. Nonetheless, the industry raised $100 billion between 2003 and 2007.

Obviously, these investments will not be made without a promise, or at least a hope, of financial return, which means that the intellectual property involved—the fundamental substances themselves and the expensively obtained clinical data that supports their safety and efficacy—must belong exclusively to the innovator. Without such protection, copycats would shun the failures, mimic the successes, and set prices at a level far below anything possible for the innovator, since the copier bears none of the upfront costs of research or testing.

The result, in relatively short order, would be greatly reduced R&D. Some extremists might like this outcome, since they oppose intellectual property as an institution. In this alternative world, medical R&D should be a public good produced by the government, or perhaps by government and academia. But this is not the system we have, which is a complex amalgam of private, academic, and government institutions working together, with intellectual property rights as an important part of the mix. The system presents many conundrums, but few outside the abstract world of communitarianist academia want biotech research dependent on the whims of congressional earmarkers.

The intellectual protection given basic pharmaceuticals (called “NCEs” for “new chemical entities”) takes two forms. Patents are crucial, but for NCEs the ordinary 20-year term is reduced by the 8 to 12 years that pass between the initial patenting and the completion of the testing and regulatory processes. The second protection is called “data exclusivity.” It is based on the fact that even after the expiration of a patent a generic manufacturer could not legally sell a drug without demonstrating its safety and efficacy, which would require replication of the clinical testing done by the original innovator.

Generic manufacturers objected on the ground that a generic’s active ingredients are chemically identical to the initial patented drug, so they should be able to free ride on the original tests. Pharmaceutical innovators objected to turning their hard-bought data into community property. In 1984, Congress enacted Hatch-Waxman, which allows the use of the data by the generic companies after a period of five years. This gives the innovators some protection in addition to patents.

For new biological entities (NBEs), the basic issues are the same, but some complications are added by differences between NCEs and NBEs that render patent protection of the latter somewhat less reliable. Thus, “data exclusivity provides investors with an ‘insurance policy’ against the potential failings of patent protection for biologics,” Henry Grabowski of Duke University writes, and assumes more importance in the context of follow-on biologicals than in the case of small molecule drugs.

Not surprisingly, the proper length for a period of data exclusivity is hotly debated. The issue was a focus of a Federal Trade Commission report, Emerging Health Care Issues: Follow-On Biologic Drug Competition (June 2009), and is a major issue in pending legislation that would create a pathway for follow-on biologicals. (For an outstanding explanation of the basic issues see Henry Grabowski’s 2008 Nature Reviews article.)

Several legislative proposals in the House opt for 12 to 16 years of exclusivity, and the Senate Health Committee has approved 12, but Chairman Waxman favors zero years and has introduced a bill providing for five. The administration, relying on the FTC Report, proposed a “generous compromise” of seven.

At bottom, of course, this debate is simply another manifestation of the never-ending debate over the proper level of protection that should be given intellectual property, and of the search to find a middle ground between the one pole of excessive deference to inventors and the other of killing R&D geese before they ever get a chance to lay golden eggs.

In the context of data exclusivity for NBEs, though, the controversy has an interesting twist. The debate is taking a highly technical form which assumes that econometric analysis can predict an optimum level of protection after looking at an entire rapidly developing industry over periods of decades. Thus Grabowski’s model, which is conceptually quite elegant, is disputed by the FTC report (pp. A-5/6) on the ground that he used a cost of capital of 11.5 percent and it should have been 10 percent, and that the correct figure for the contribution of operating profit to recoupment of investment should have been 60 percent rather than 50 percent, and so on. Using these revised figures, the break-even period for a portfolio of investments in biotech shrinks. The FTC report also has extensive predictions of the future development of the industry, including assessments of how long many kinds of market developments will take, all of which cut in favor of a shorter time period.

As we should have learned from the recent financial disasters, relying on such fine-grained predictive analysis is a road to disaster. Grabowski’s conclusion that data exclusivity should last at least 12 years is persuasive not only because of the elegance of the model, but because his analysis is tied tightly to known historical facts. When one gets into the realm of prediction, one’s willingness to stake the future of an important industry on being right on details of future developments should decline to zero.

Respect for uncertainty is especially necessary when the predictions are made by the FTC, which has, as an institution, the antitrust lawyers’ bizarre assumption that price should equal marginal cost, a view that has almost zero applicability to the real world. As economist William Baumol has noted:

Only this year I heard a conference presentation dealing with the economic and legal principles of copyright suggest that the innovating Schumpeterian entrepreneurs are automatically to be deemed proper subjects for antitrust attentions because in the period before imitators enter the market, they can charge prices that exceed the marginal-cost levels of perfect competition. Never mind that this is a prescription for undermining intertemporal efficiency. Never mind that marginal-cost pricing would generally preclude recoupment of the research and development (R&D) costs of the innovations at issue, costs that will have to be incurred many times again if innovation is to continue. And never mind that a world of perfect competition requires constant returns to scale and firms so small that they would never attract the attention of regulatory or antitrust personnel.

In the end, we will never be able to arrive at agreement on the perfect period for data protection for biological research. There is still no agreement on the proper period for patent protection, and that system has been around for hundreds of years. A 20-year period was picked for patents, and it seems to work well enough, so the failure to achieve perfection is not of great concern.

With regard to biotechnology and data exclusivity, we are and will remain similarly ignorant. So the question is actually quite simple: what are the costs of error in each direction?

If the protective period is too short, then innovation will suffer as the industry turns into a bunch of copycats, just waiting for someone else to be foolish enough to innovate, and investment will dry up.

If the period is too long, then the worst outcome will be an over-investment in biotech that leads to even more innovations and products than we can readily use. Prices of some drugs might stay higher than they would under a perfect system (whatever that is), but that is better than the infinite price of a drug that does not exist, and consumer welfare would be improved by the added choices available. Besides, the researchers will quickly snare any excess profits and recycle them into even more research, leaving the investors with an adequate but not exorbitant return.

Furthermore, since the alternatives under consideration are limited on the upside—no one is talking more than 16 years—any potential harm from being too generous is also limited. (Personally, I think biotech deserves far more protection and I suspect that the creator of the data should own it, period, and anyone who wants to use it should make a bid, but that option is not on the table.) On the downside, the damage that could be caused by too short a period is potentially great, and would be very difficult to undo.

The Office of Science and Technology Policy recently released “A Strategy for American Innovation: Driving Toward Sustainable Growth and Quality Jobs.” The report is written at too lofty a level of abstraction to grub around in the muck of actual laws and their impact on private sector investment and innovation, nor does it discuss biotech, which is one of our most innovative industries. But it does say: “Intellectual property is to the digital age what physical goods were to the industrial age. We must ensure that intellectual property is protected in foreign markets and promote greater cooperation on international standards that allow our technologies to compete everywhere.”

Even Theodoric of York could reach the right result on this one, but for Congress it’s a nail-biter.

James V. DeLong is vice president and senior analyst of the Convergence Law Institute, LLC, and special counsel in the Washington, D.C. office of Kamlet Reichert, LLP.