John Marburger

Making Science Policy

June / July 2008 By: John Marburger Volume 6 Number 3

Print this Article

E-mail this Article

The concept of a policy context is deeper than you might imagine. The concept of a policy itself is deeper than you might imagine. As I mentioned in one of these talks years ago, my favorite science policy paper is one that Daniel Sarewitz wrote in 2003 called "Does Science Policy Exist, and If So Does it Matter?" One of the positive events in science policy scholarship during the past year is that Dan finally published his paper in the Summer 2007 issue of the National Academies journal Issues in Science and Technology.

The important insight of this paper is that try as one might—€”and generations of science advisors have labored on this—€”the process of policy implementation in American government is shared over so many uncoordinated actors that the policy behavior of the governmental science machinery is not only different from what policy makers intended, but sometimes difficult to trace at all. Anyone who wants to evaluate the effectiveness of a policy first has to decide what the policy is "on the ground" as opposed to some theory of it embedded in a law or a strategic planning document. The actual outcome of the process of policy proposal, authorization and appropriation rarely resembles what any particular stakeholder in the process hoped for at the beginning.

That has certainly been true for the past two years during which two different congresses failed to appropriate science funding in patterns that were worked out over months by their own authorization and appropriations committees. Most observers predict that once again the forthcoming fiscal year will begin with a Continuing Resolution (which translates as Continuing Irresolution). I think there is a danger of reading too much into this extraordinary failure of process. On Sarewitz's model of real world policy implementation, such things are to be expected in general.

Another insight of Sarewitz's paper, to which I have repeatedly called attention because of its importance for forecasting the outcomes of policy struggles in Congress, or between the executive and legislative branches, is the remarkable stability of federal R&D funding as a percentage of the domestic discretionary budget. Exactly how this happens each year is somewhat mysterious. It's like trying to infer the laws of thermodynamics from the behavior of the individual molecules in a gas. The small scale motion is chaotic and irreproducible, but the overall behavior always comes out the same.

The big picture of the national budget really matters. The discretionary budget is under constant and growing pressure from the large mandated portion of the budget. On top of this are huge fluctuations associated with events that may seem highly singular and catastrophic, such as war, hurricanes, and economic bubbles, but recur in nearly every administration. Regardless of the circumstances, the give and take of politics, including all the partisan dealing, all the lobbying, and all the local issues that intrude on the national scene, ends up giving research about the same fraction of the discretionary budget every year in administration after administration. The time series has bumps, but they rarely range outside a narrow band.

Many have complained about the impact of the 2008 omnibus bill on research. It did have negative effects, which I will mention in a moment, but the omnibus bill did provide increases for some important science areas, just not as much as the president's proposal, and much less than the appropriations bills being considered in the subcommittees. And of course it did not reflect the priorities for funding either in the president's American Competitiveness Initiative, or in Congress's America COMPETES authorization bill. As the dust settles, however, once again research will have received approximately its usual slice of the pie.

I think this pattern will likely persist in future administrations. I think it will actually be difficult to match the increases in research funding that have occurred during the Bush years. There is much mythology about this, and much quibbling about definitions and what the numbers really mean, but overall there is a much greater amount of research money on the table today than there was at the beginning of the administration. Everyone has their own ideas about how it could have been distributed differently, both among fields of science and over the years. But there cannot be any question that this country has significantly boosted spending on research during this administration. The myths of downward trends in science spending are based on measures other than actual dollars spent. Patterns of U.S. science funding do show some disturbing trends, and they need to be fixed. But if we dwell only on those trends I believe we indirectly raise false expectations that future administrations will be able to solve science funding problems simply by adding more funds to the pot. As always there will be winners and losers.

This is the right place to acknowledge and praise the actions taken by the AAAS this year to treat earmarks in a systematic and straightforward way in their budget analysis. Two years ago I challenged OMB and AAAS to come together and work out a common approach to accounting for and reporting the effect of earmarks on the science budget. Last year OMB introduced an earmark website full of fascinating information. This year AAAS has made a useful study of the issue and has incorporated a good treatment in their reports. They estimate that congressionally designated, performer specific R&D projects in 2008 total $4.5 billion, and concluded that "in a tight budget environment earmarks once again crowd out hoped-for increases in competitively awarded research programs."

These earmarks are not simply whims of Congress. They come as a direct result of members of the science community, contractors and others advocating for their projects. Many are worthy—€”let us stipulate that they are ALL worthy. But is this the most effective use of our resource commons? What guarantees the result is the best for the American people, or is even, in some sense, sustainable? What guarantees that the most essential research ever gets done? Enlightened organizations representing large numbers of constituents here in Washington have made some efforts over the years to persuade their members that they need to advocate for broad programs, or for agency funding, and not for specific projects that seek to avoid review before receiving funds. They have not had much success. The jobs of many in this audience are part of a large and growing machinery aimed at bringing home the bacon for your organizations.

Earmarks are not the only aspect of the exploitation of the science funding commons. Various set-asides and incursions into the agency research missions have a similar effect. Like the SBIR tax on every agency's science budget that narrowly escaped a costly increase earlier this month, thanks to a floor action initiated by Congressman Vern Ehlers, who is also a physicist. Or the addition of new dimensions to program requirements related to education or outreach. The research commons is shrinking because of these incursions. Let's assume here too that they are all worthy. Most taxpayers look at these agency budget—€”for NSF or the Department of Energy, for example—€”and assume the spending is for science. How is the science going to survive as the commons shrinks under these incursions?

I don't want to dwell at length on this problem, but it is very serious, and I think it helps to view it through the resource commons lens rather than as a problem of ethics. Curiously, Garrett Hardin's subtitle for his original "Tragedy of the Commons" paper is "The population problem has no technical solution; it requires a fundamental extension of morality." So he, at least thought the situation in extreme cases entailed moral considerations. Most social scientists who study these questions agree that commons must be managed under a consensual framework that regulates the behavior of the actors to avoid degradation of the resource.

During my tenure in Washington I have seen three remarkable issues around which consensus among stakeholders has sprung up with an almost religious fervor. The successful doubling of the NIH budget between 1998 and 2003, the successful establishment of the new area of homeland security research and the as yet unsuccessful introduction of a set of actions, including selected research increases, to bolster future national economic competitiveness. The lack of success of the latter is quite unusual given the very wide and intense support for it across parties and sectors. I have said enough about the process failure that led to this result, but it is one of the most serious pieces of unfinished policy business begun in this Administration.

There is no question that there is now a large and unhealthy imbalance among funds for various sectors of science, usually described as biomedical research versus physical science, engineering, math and computer science. Federal support for the latter has languished for several decades. Now federal support for biomedical research has languished for half a decade. The budgets are still out of balance. We would certainly be better off today if the 109th Congress had passed its appropriations bills before going home. They would have begun to restore balance and start the building up of a technical workforce in fields badly needed for future progress in all parts of science and technology. This year, for the third year in a row, President Bush has proposed a path forward that aims to shrink the gap for critical fields where global competition is very strong.

Dr. Marburger is director, Office of Science and Technology Policy,
Executive Office of the President. This article is excerpted from remarks given at the policy forum of the American Association for the Advancement of Science, May 2008.