Scientific Research

After World War II, the United States erected a structure of support for scientific research that has been remarkably successful in producing new science, technology, and medicine—and more than 100 Nobel prizes in chemistry, physics, and physiology or medicine.

Advances such as the polio vaccine or the transistor have reduced human suffering or contributed to dramatic increases in our Gross Domestic Product (GDP). Research support flowed to our most productive and promising scientists with a minimum of micro-management. The implicit directive was: “We provide the resources. You advance the science.”

In addition to remarkable science, this system has given us what should be a good problem: many promising leads that, if explored, would result in new technology or in amelioration or cure of ailments that still plague us. But pursuing more leads costs more money, and a successful and growing science enterprise has become harder to manage. Many promising research directions are not funded; these are lost opportunities. What is worse, in my opinion, is the resulting siege mentality among funders, which has led to excessive micro-management and extreme conservatism in supporting research. Agencies that have contributed mightily to fundamental science, and produced applications of great value, are under pressure to narrow their goals, focusing on research with immediately apparent results. All of this has put destructive stress on the system.

The very nature of the research enterprise is being questioned. It is suggested that science is nearing an end, all major discoveries have been made, and diseases like cancer or Alzheimer’s disease cannot be cured. To the working scientist, such complaints reflect bleak ignorance. Accuse us of “irrational exuberance,” but we believe that although some problems are extraordinarily difficult—and solving them may be frustratingly slow—we will succeed.

Some years ago, when I was at the Institute for Advanced Study in Princeton, we young guys (it was all guys in those days) would sit around grousing: “Einstein, Heisenberg, Dirac, they solved all of the easy problems and left the hard ones for us.” In years since, those hard problems have been solved, often by ourselves (leaving new hard problems, of course). The message is that we can do it. We must ask how we can ensure that research is (at least) as successful in producing results in the next 50 years as in the past 50.

Research Works

For all the grousing (and the prophecies), scientific research yields an intellectual harvest that is sometimes beautiful, often useful, usually expensive, but extraordinarily rewarding. I would argue, in addition, that it saves money and makes money.

Economic growth is fueled by technology rooted in research discoveries. The last 50 years have seen magnetic resonance imaging, laser surgery, computers, and modern telecommunications (to take just a few examples) emerge from prior basic research. Let us pursue a single instance. Computers and telecommunications, which make a gargantuan contribution to GDP, require integrated circuits based on transistors. The importance of the transistor is generally grasped, but it is less commonly appreciated that the transistor could not have been invented without fundamental advances in quantum physics just 10 or 20 years earlier.

On this spectrum, biomedical research occupies a special position. It has a profound impact on human suffering and, in addition to generating revenue it reduces costs by reducing or delaying chronic illnesses—among the most expensive components of the cost of health care. In the United States alone the annual direct and indirect costs of brain-related illnesses are an estimated $600 billion. Zach W. Hall, Ph.D., former director of the National Institute of Neurological Disorders and Stroke, has written:

[Until recently] there has been almost nothing that physicians could do for patients with stroke, Alzheimer’s disease, traumatic brain and spinal cord injury, multiple sclerosis, and a whole host of childhood neurological disorders ranging from cerebral palsy to autism. Those ominous words “There is nothing that can be done,” have echoed in neurologists’ offices for decades.¹

Now this is beginning to change, and dramatically. The annual Progress Reports on Brain Research from the Dana Alliance for Brain Initiatives show the gains neuroscience has made in just a few years, in areas as diverse as learning and memory, emotion and stress, behavior and pain. These gains that translate into new hope for treating developmental disorders, schizophrenia, addiction, alcoholism, manic-depressive illness, and other maladies that inflict misery on millions. This progress also translates into economic benefits. The figures are only suggestive (and more study would be welcome), but it is estimated by the Ad Hoc Group for Medical Research Funding that just delaying the onset of aging-dependent diseases like Alzheimer’s could reduce annual spending for nursing-home care by as much as $35 billion.²

True, new technologies can increase expenditures. It is expensive to use Magnetic Resonance Imaging to plan neurosurgery for epilepsy, or to implant the newly approved neural prosthesis that enables some quadriplegics to feed themselves, or to treat manic-depressive illness or multiple sclerosis with new drugs. But these expenditures extend lives that would have been shorter, save lives that would have been lost, and preserve a quality (and productivity) of life that would have been diminished. In a way, it is like comparing the cost of a Model T of Henry Ford’s day with the complex electronic device that we now call a car. It is simply not the same vehicle. And most of us, when the length and quality of our lives are at stake, prayerfully accept the new and more expensive technology.

Should Research “Pay for Itself?”

But why, if research is valuable, doesn’t it pay for itself? Why, in a market economy, must research that fuels economic growth be supported by taxpayers?

An interesting answer was offered in a February 5, 1998, article by New York Times economic columnist Peter Passell:

[There is a] systemic failure of free markets to allocate adequate resources to research and development. Study after study has found that corporations capture only about half of the gain from inhouse innovation, with the rest going to other businesses or to consumers. The late Edwin Mansfield estimated that the private annual rate of return on investment in corporate research was 25 percent, while the “social” rate of return (including the benefits to others) exceeded 50 percent. Left on their own, then, corporations have an incentive to invest less on research and development than is desirable from the perspective of the economy as a whole.³

Put another way, unless there is a monopoly (for example, AT&T before its breakup), a company has no incentive to unilaterally increase its overhead for research because a large percentage of it will benefit competitors. For research in universities (a major source of progress) the situation is even worse. In effect, our universities are subsidizing, out of current operating budgets, research that has high rates of “social” return. There is no alternative to public support if we wish to obtain the “social” return on research that we desire.

Underlying Principles

If public funds are appropriate and necessary, the question becomes: How much money should be allocated and how should it be distributed to ensure that in the future progress will be at least as robust as in the past?

In my opinion, we must “return to the future” by making these decisions based on the broad principles that contributed to past success:

1. Provide adequate funding to pursue promising directions in fundamental science as well as promising leads on specific diseases or technologies.
2. Trust our scientists to determine the best scientific directions with a minimum of micro-management.

The first principle is important because, without enough money to follow exciting leads, opportunities will be lost, and both technical advances and treatments and cures for diseases will be delayed. It is less obvious—and to many, less persuasive in times of budgetary stringency—that research to advance fundamental science (for example, insights not immediately linked to a particular disease) must also be adequately funded. Without this, the leads, and ultimately the progress, will dry up. For example, fundamental molecular research on neurotransmitters, pursued with no specific application in mind, has lead to a promising new understanding of neurodegenerative diseases, such as Alzheimer’s, that affect memory.

The second principle follows from the insight that good research decisions are difficult to make from above. Although successful broad programs have been created, the best actual ideas for research come from those closest to the firing line. We should make it fundamental to fund our most talented scientists and trust them to choose the most fruitful areas to explore.

If we are to continue a successful research enterprise, the system based on these principles must be restored. Above all, we need adequate funding, reasonable risk taking, and a reduction of bureaucracy and micro-management.

How We Might Proceed

What is the right amount to spend? How should we allocate it? As part of the answer to these complex problems, I would suggest three actions.

First, increase public research funding to bring it in line with current needs. J. Morton Davis, Chairman of the Board of the D. H. Blair Investment Bank Corporation, suggested recently that we might compare what is spent on this country’s research effort with the Pentagon budget. Is defending ourselves against disease as important as defending ourselves against foreign enemies?

In addition to funds allocated by Congress, a much larger role could be played by private foundations and individual giving. Foundations have the flexibility to be more adventurous than public funders, stepping into the breach opened by the conservatism of government funding. Individuals, who now direct 80 percent of their annual giving to organized religion (a total of $58.87 billion in 1994)⁴, might consider a diversification of their charitable portfolios.

Second, link future public funding of research to some index such as the GDP, or a phantom royalty calculated as a percentage of the value of the technology and commercial products that have grown out of fundamental research during a shifting time window (say the past 75 years). This royalty should be commensurate with that paid to license the technology. (The long time window is critical because fundamental research does not usually produce a return on investment in the next quarter.) After all, if research is a major contributor to the GDP, shouldn’t investment in research be commensurate with the benefits that it produces?

As important as the level of funding is the ability to anticipate the amount of future funding, so that we could plan more sensibly. At present, agency funding cycles are relatively short—a few years at best—while producing a research scientist takes many years. Likewise, the results of basic research are often long in coming, so year to year changes in funding can be disruptive. Much of the current effort to find causes and treatments of disease involves genetic technology built on the basic research on DNA by James D. Watson, Francis Crick, and their colleagues—work that goes back at least half a century. Today, identifying the genes involved in neurological conditions such as Huntington’s disease, manic-depressive illness, and some forms of alcoholism is a critical step toward diagnosis and effective treatments.

Third, maintain a basic, balanced division of research funds between support for pursuing specific leads promising treatment of disease (or production of new technologies) and fundamental research directed toward producing new science, and thus new leads. A research portfolio is not unlike an investment portfolio; we need diversification of areas and risks and an eye for the long-term, overall return. Because research in one area (or one science) so often has unexpected applications in other areas (and other sciences), we should forswear wild funding swings from one fashionable area to another, or declarations of war on first one disease and then another. We should focus on maintaining a healthy, diversified research portfolio in all the sciences.

Distributing the Resources

Agreed-upon public funds should be distributed by many agencies and offices. I can think of no worse situation than distribution by one or two superagencies. Between 1950 and the present, among our most productive research eras, support came from an array of civilian and military agencies. One of these, the Office of Naval Research, produced astonishing research results that have benefited us and increased the GDP. Today, unfortunately, military agencies are under great pressure to fund only research that is immediately relevant to the military mission. If the military cannot or will not fund its research agencies as it once did, we might consider separate line items within the military budget for those agencies to protect their vital roles.

Multiple agencies for the allocation of research funds can generate competition in the search for investigators most likely to hit it big. Such competition in funding fundamental research is extremely healthy and is one way to produce a diversity of ideas. As an additional incentive for creativity, one might reward a successful agency by increasing its funding in a manner somehow proportional to the success of the research it supported in the past. (The tricky part would be to avoid penalizing support for long-term, fundamental research. Can our political system respond to success in a perspective of five, 10, or even 20 years?)

The Stakes

We must renew our structure of funding to surmount obstacles to a healthy, growing research enterprise. If we succeed, we can expect new ways of alleviating, curing, or preventing maladies that have afflicted humanity for millennia. Just as important—or more so—we can expect to open new doors to understanding ourselves as human beings. And we can hope to be prepared to manage those new plagues that almost surely will emerge. (Compare our progress on HIV in the past 15 years—as painfully slow as it may seem—with the centuries it took to deal with smallpox or syphilis.)

Some advances will come with gratifying speed. Others will come more slowly. The problems are not easy. Sometimes the fundamental knowledge is just not in place; sometimes we are delayed in the most frustrating way by lack of resources. Some discoveries will come from extending existing science. Some will come from utterly unexpected directions, from research in seemingly unrelated fields: research supported by agencies, in addition to the National Institutes of Health, as diverse as the National Science Foundation, the Office of Naval Research, and the Defense Advanced Research Projects Agency. Think again of nuclear magnetic resonance (beloved of physicists) that led to modern brain imaging, or think of research into the stimulated emission of microwaves (an esoteric quantum effect) that led to the laser, and thence to its use in surgery.

Think of the increased cost and misery, or the lost revenue, if the polio vaccine or the transistor had been delayed by just 10 years. In addition to reducing human misery, our investment in a healthy and diversified portfolio of fundamental research in all sciences will more than repay its cost. We may not be aware of what we will lose if we do not continue to invest adequately, but lose we surely will.