Policy 102: Science Policy and the Actors Involved

By Fred W. Weingarten

In my previous article, I described the "what" of science policy, the variety of issues and concerns of which it is comprised. Even when the basic goal is simply defined as federal funding for research, the range of debate can be surprisingly wide and complex. I'm going to talk about the "who" of science policy — who makes it, who influences it, who approves it, or (on the negative side) who messes it up? The picture is complicated, as one might expect when dealing with questions of influence and power in policy and political processes.

John Kennedy reportedly observed that when he was a senator, all power seemed to reside at the other end of Pennsylvania Avenue, in the White House. When he became president, however, it seemed that all power resided in the other direction, in the Congress. That observation captures a basic reality of policy and politics in Washington — nobody seems to be fully in charge of anything.

Of course, at any particular time, on any particular issue, someone or some organization may seem to be in control of the process, but power is nearly always in flux. This can be frustrating for those who want things to get done or decisions to be made quickly and efficiently, and it can be particularly frustrating for those pushing for major policy change. Once a consensus has been forged around a policy framework, there is not much interest in reopening negotiations, if for no other reason than someone is benefiting from that framework. It becomes easier to do things gradually, seeking incremental change within the old framework, because new decisions don't have to be made or old ones reopened. (Recall the failed attempt to completely restructure health care policy in the early days of this administration.) This Diffusion of power results in policy inertia.

Our political system was designed to be ambiguous and conflicting. There are good reasons for this to be so. The country is big, powerful, and incredibly diverse in its people's interests and values. It is best not to concentrate power too much, and it is necessary that any major policy changes be made only after broad consultation and deliberation. No doubt, though, we also pay a price for inertia.

Right or wrong, intended or not, diffusion of power has important consequences for people working in science policy, particularly if they are trying to promote change on the scale proposed in the Interim Report of the President's Information Technology Advisory Committee (PITAC). (See interview with Bill Joy, Co-chair of the PITAC Committee, page 1.) It means that in order to achieve the desired result of implementation of the full program, a lot of power centers need to be brought into some rough alignment, and marshalling of those who will promote the program and neutralizing those who might be inclined to oppose it must be done. How to do that is the subject of my next article. For now, let's look at the policy-making landscape.

Overview

Figure 1 (below) provides a top-level map of the policy environment. There are three basic power centers Congress, the Executive Branch, and a collection of private and quasi-private organizations and individuals that represent the community that is directly affected by or interested in science policy. I call this group the "Private Stakeholder."

Around these three groups is a penumbra of influence that I label "public interest." I assert that in the particular area of science policy, the public has, for the most part, not been a strong force one way or the other. The general public view that research is, by and large, a social good and that it's okay for Congress to spend some tax money on it — precisely how much and on what are decisions comfortably left to the system. This runs counter to occasional spurts of angst from some in the science community, who worry that the public is ignorant about science and contend that, if people just understood research better, they might support more funding. I'm skeptical about that particular argument.

That is not to say public opinion should be ignored. It can be a powerful force, good or bad. I certainly believe that it is important for the computer field to be sensitive to public concerns about potential negative effects of information technology. And, it would be a powerful impetus to realizing the PITAC's program were it to find its way in some form onto the presidential campaign platforms of one or both parties.

Nevertheless, my principal focus here is on the three "core" groups shown in figure 1. Discussing them is complicated enough, for each has a fairly elaborate structure of competing interests.

Executive Branch

Well, surely, the Executive Branch isn't complicated. "After all, it has a single boss and operates from a uniform policy," you might be tempted to say. "And pigs fly," I'd answer. Probably tougher even than negotiating with Congress are the daily negotiations the White House conducts within the labyrinths of executive agencies, trying to get the government reading off of the same page (or even out of the same book, or even the same library).

Each agency has its own legislated missions, most of which began long before the current president came into office and will last long after he is gone. Each one also has a set of interested constituencies that exert various degrees of influence on its policies and priorities. It may even have a formal or semiformal authority telling it what to do. For example, NSF operates under the oversight of the National Science Board, which may not always agree completely with the administration's priorities.

Each agency has its own budget issues. Although each year begins with a single presidential budget being sent to Congress, each agency must negotiate its own budget with its own authorization and appropriations committees. This means that even though an agency may want to cooperate within a multiagency program like the Next Generation Internet, it is forced to examine and define its potential role in the program through its own peculiar lens. That's why hammering complicated multiagency programs together can be both frustrating and time-consuming.

Furthermore, the White House itself has its own set of conflicting interests because the president has multiple roles to play and conflicting interests to serve. As a result, there are people in the White House who worry about the politics of decisions. Also, policy analysts in offices such as the Office of Science and Technology Policy look at the core substance of issues, and the inevitable keepers of the budget at the Office of Management and Budget are always saying, "Nice idea, but where is the money coming from?"

Sometimes on a particular issue, all three messages will be coming out of the White House at once. The president may express sympathy for the need to double research budgets. The budget director voices skepticism whether we can afford it, and the political staff ask what's the benefit and cost of spending limited political capital on this issue versus all other possible ones. (This situation is not unusual, nor is it specific to the current White House. They all act the same way, because all presidents face very similar dilemmas in running the government.)

Congress

Congress has its own morass of interests and power centers that have to be accommodated also. Just for starters, we have two houses, populated by 100 senators and 435 representatives. Each has been chosen in a separate election and considers himself or herself to be an autonomous decision-maker. It is a self-organizing group; the Constitution really says very little about day-to-day processes and structures. Members have personal offices with staff whose principal function is to serve their specific political interests and provide service to their constituents.

Each member also serves on committees, where the principal legislative work is done. So, adding to the complexity, we have several dozen committees and subcommittees to deal with. Science and technology policy, particularly when it involves computers, touches a surprisingly large number of them, including, of course, the Committee on Appropriations.

Finally, think of all the lines that partition these people in ways that affect their specific interests in issues - political party, regionality, ideology (not always party specific), ethnicity and gender, and so on.

The bottom line is that it is not easy to get a majority congressional vote for some new change in policy, although it can be done. It took many years and the efforts of countless people to get the High Performance Computing Act through congress, but it did pass and with a nearly unanimous and bipartisan vote at that.

Stakeholders

Finally, let's take a look at ourselves, the computing research community, we the people, the stakeholders, the special interest groups, as we are sometimes known to those who disagree with our views. (We have influence on science policy and our voices are heard through a surprisingly complex group of organizations and individuals. Each has its particular capabilities and strengths and each has its own role to play.)

Stakeholder voices include individual scientists, particularly (not exclusively,) those whose prominent awards, academy memberships, or titles lend credibility to their views. These include Academy members, Medal of Science and Technology winners, and, of course, Nobel Prize winners. Nobel Prize-winning physicist, Kenneth Wilson, was a major influence on the HPCC Program.

In addition to acting as individuals, scientists belong to influential groups. They sit on advisory committees to NSF and other agencies and they belong to study groups of the National Research Council, particularly the Computer Science and Telecommunications Board, and, of course, the President's Information Technology Advisory Committee — groups that conduct studies and publish reports urging policy actions.

The scientific societies also exert a strong policy voice in support of the fields their membership represents. Many are based or have offices in Washington specifically for this purpose. When CRA was created, it was purposefully located in Washington because its membership expected it to assume a major role in science policy, representing the needs and interests of computing research.

Coalitions of scientific organizations also form to advocate in support of particular issues such as NSF appropriations, the HPCC Initiative, and so on. Some of these groups can be ad-hoc, coming and going quickly as issues change; others have broader agendas that may persist over the long term.

Lastly, there are several organizations that, while not scientific societies themselves, have occasional interest in science policy. Most important for the computing community are industry groups and coalitions. They can be powerful voices for or against computing research initiatives. Although their energy can often be difficult to tap in to and channel.

Wrapping it all up

What does this enormously complex array of players in science policy mean for us? My thesis is as follows:

For major policy change to take place, there must be agreement among the three major sectors -executive, congressional, and stakeholders. In order for that to happen, communication among them must be strong. That is why I drew the lines connecting them as arrows. Cultural and communication barriers exist among the three groups that limit their ability to understand each other and reach consensus.

In fact, one of the major tasks of Washington science representatives is to monitor and strengthen conversation among them. I am not the only one, I am sure, who has been asked to carry a message from an executive agency to congressional staff and vice versa, and between both government sectors and the scientific community.

And, of course, the same need for convergence (and similar barriers) exists within each group. Agencies and the administration need to be aligned, as do the various interests in Congress. And, our own community, as well as the broader science policy community, needs to be brought together. HPCC was successful, not just because computer scientists wanted it, but because other disciplines and the computer industry pushed for it.

In these first two articles, I sketched briefly a structure of how science policy is made. Next time, I'll show what this structure implies for us in trying to fully implement the recommendations of the PITAC. We have a choice here. We can use the PITAC's recommendations simply as useful ammunition for an incremental gain over current funding. This requires no major change in strategy or effort on our part. On the other hand, what the Committee really proposed represents a major shift, both in funding and in the patterns of research support. If that is what we want to see happen, my thesis above suggests that we need to expend far more energy and develop and follow a sustained strategy for changing minds. It won't happen just because the PITAC, or even the President, asked for it!