Lessons Learned in Science and Technology Policy

The Potomac Institute for Policy Studies has, over its twenty-year history, conducted numerous studies on particular issues of science and technology, with recommendations to policymakers as to how to proceed in light of the implications of technical developments. In some cases, the recommendations amounted to investment decisions and policies. In other cases, the recommendations related to policies that are dominated by scientific or technological content. Further, some of the issues relate to national level organizations, while others affect one or more federal or state agencies. There were a variety of lessons learned and common themes that emerged over the years from these studies. This article uses selected examples to illustrate some of these lessons and themes.


Lessons Learned in Science and Technology Policy*

© Copyright, Potomac Institute Press 

“ ... the power of science to alter nature has reached such a state that society needs to have a much more fundamental place in considering its support.”

“... fundamental science is unpredictable, unavoidably sets its own agendas, and has an inherent timescale, both in its community structure and its execution, that is ill matched to the short-term perceptions of public opinion.”

These two conflicting passages, from the same Nature article1 allude to just one of the many complexities of making decisions on science and technology (S&T) matters. Yet, at every level of government such decisions and policies are formulated daily that not only affect investment for fundamental science, but also dictate or influence technology development and application (both of which may fall under the heading of “policy for S&T”). Additionally, decisions are made that apply science and technology content to engage national issues (“S&T for policy”).

In the past, the scientific community has often identified those issues that have high S&T content for special treatment. For example, policies that affect science receive extra scrutiny. But science and technology have become progressively more ubiquitous in our society and technical considerations are claiming a major role in nearly all policies, as even seemingly nontechnical issues demand science for policy considerations. This is particularly true of the data gathering, analytical, and enforcement facets of setting and managing policy. In this article, I will refer to these considerations as “technical aspects” of policy, which I believe seldom receive adequate attention.

Many books and papers have questioned the effectiveness of national processes to produce comprehensive policies that reflect accurate and actionable S&T, particularly in view of the increasingly powerful impacts of S&T on society. STEPS provides an opportunity to continue this discussion and to suggest changes needed to engage technical aspects of policy more effectively.

In this article, I offer a few insights into those aspects in the context of a series of studies conducted by the Potomac Institute for Policy Studies (the Institute) during the last two decades. These studies examined and assessed actual S&T programs and issues, with the help of the performing agencies. Although the experience provided insights into individual technical or programmatic problems, there was seldom time to consider how they affected or were affected by existing policies, or more especially to comprehend how the process for creating them could be improved. Writing this article only begins to make amends.

Levels of policymaking

Before turning to a summary of the Institute’s S&T studies from the perspective of policy we should mention the national and agency levels at which policy is generally made and managed. I will not cover this subject in detail in this article, but I would suggest that where policy responsibilities reside often determine their success or failure.

It is true that while most policies are fashioned at the agency level, they are often approved and modified at the national level. Regardless, they generally end up in an agency for management and enforcement. Further, a good deal of cycling from one level to the other may occur (or even from one agency to another, in the case of policies which may affect more than one agency’s mission). That is particularly true as the issues become better defined. This is important because each stage of policymaking, from concept to enforcement, can produce changes in the intended effects of the resulting policy, especially under the pressure of multiple political agendas. In complex S&T policies, these changes in effect may be quite counter-intuitive, or they may jeopardize the degree to which agencies and top-of-government are able to interact with one another. That is especially significant when there is political disagreement on the issues being addressed. Figure 1 diagrams typical policy issues according to the type of policy and the level at which it is handled.

Figure 1

Figure 1. Two levels of federal policy issues.

As an example of the migration of issues from one level to another, most government-sponsored work on human cloning, and the budgets that funded that work, are the responsibility of the National Institutes of Health (NIH) under the Department of Health and Human Services (DHHS). Originally, issues were primarily scientific and dwelled in the NIH community. Then, ideological and ethical issues surrounding cloning shifted the debate to the national level, leading to the question of whether all human cloning research should be prohibited, or at least tightly controlled. So, the debate on budget also shifted to the national level, in this case yielding reductions in S&T budget program element lines that affect cloning research, consequently affecting the pace of the work. It did not take long for this action to ripple into the commercial world, where it slowed US private investments in cloning. Similarly, human embryotic stem cell research that depends on the creation of, or at least access to, human embryos was affected by the decision to curtail that access. This demonstrates the breadth and complications of the questions that must be answered before policy is set, especially in the instance where perplexing social issues meet complex scientific concepts.

“Technical” policy challenges

The nation has many S&T-driven challenges and opportunities. Increasingly, we rely on the products of science and technology, guided by government decisions and policies to mitigate the former and take advantage of the latter. But, in the face of these expectations, S&T policies are too often fractional and narrowly focused, and issues are too quickly politicized. Consideration of S&T at the national level is often inadequate to support good decision-making on investments and other issues, even as the consequences of mishandling these decisions grow. To the degree that we correct this shortcoming, we will improve our prospects to gain the benefits of science without succumbing to its dangers.

We are warned daily about terrorist attacks; computer identity theft; new and frightening diseases; and a looming energy crisis that is creating global climate change, devastating oil spills and radioactive leakage. And, during the next few decades we will certainly uncover mechanisms more efficiently destructive than those that haunt our dreams today. Along with these threats are exciting opportunities offered by S&T, for even as scientific advancement creates problems, it is often the most effective means to solve them.

Over the past two decades, the Potomac Institute for Policy Studies investigated numerous issue areas that involved S&T. Most often the focus of the studies was on research priorities or approaches, or of forecasts, impacts, or applications of specific areas of science or technology. New ideas about transition or application of technologies, or shifts in funding between government and the private sector were also among the goals of these study projects. Whatever the specific goal however, the formulation and management of applicable S&T policy was always pertinent to study findings and featured large in our recommendations. Here are a few study subjects that were engaged and some policy issues that affected them:

Government’s role in S&T. As private sector investment overtook government funding in S&T, it became clear that new policies were needed to ensure that a balance was maintained between science and technology and that funding was prioritized among technical areas in a logical and purposeful way. Further, globalization magnified the effects of this shift, especially through the ready commercialization of products of research and development (R&D) programs with military potential, making it more difficult for the DoD to contract with a private sector seeking far more lucrative commercial markets. Yet few changes were made to accommodate the new investment landscape.

Dual Use S&T. As suggested above, a major challenge at this time was how to continue the output of technologically advanced military products when most of the world’s R&D was being conducted in the commercial sector for commercial purposes. In 1992 the Defense Advanced Research Projects Agency (DARPA) began a major thrust called the Technology Reinvestment Project (TRP) to design and manage the development of technology products that would simultaneously satisfy commercial and military ends. This was a profound effort and required questioning many hitherto conventional relationships with industry. For instance, DARPA’s policies encouraged consortia of government, industry, and academia to create new technologies and even allowed industry partners to retain product intellectual property for commercial applications.

As with many trial government programs, one of the TRP’s major selling points was productivity. Because of this, staffing was maintained at such a low level that an internal assessment of the project was impossible (a common policy weakness in programs dedicated to trying out new management ideas). Realizing this need, the project manager asked the Institute to initiate an evaluation of the TRP in 1998. A six-step assessment process was followed, employing carefully chosen metrics to judge the effectiveness of all policies and the success or failure of the products of 113 individual projects to address both the commercial and military marketplace (thereby assuring economies of scale sufficient to lower costs to the military). We presented our findings for both individual projects and for the entire program. Commercial success was evident in 33% of the individual programs and there was military value in 85% of these products.

These assessment results certainly warranted the adoption of the program’s governing policies. Instead, the principle DARPA program was canceled when Congress failed to continue its funding.2 Today, technology dual use is clearly in the DoD toolkit, but it took a long time to get it there and we are still grappling with fundamental questions, such as how to deal with world-wide access to the same dual use technologies for military use.

Shipbuilding in the United States. The US shipbuilding industry is an American anomaly. It is a fairly large employer, but its production levels are relatively low. It can produce remarkably sophisticated warships, but cannot compete in the global commercial marketplace. With this backdrop, the Maritech Program, a joint DARPA/Navy effort, was begun in 1993. The goals of the program were: making the DoD a better customer, injecting the latest technology into ship systems, and improving the commercial competitiveness of the US shipbuilding industry.

The Institute was asked to perform an analysis of the program in 1998. Using similar approaches as those described above, we found the program to be partially successful, making a distinct difference with American shipbuilders but failing to produce commercial competitiveness in the industry. Major difficulties discussed in the study report were embedded in the economic and technical relationships between the industry and the Navy acquisition community that remain mandated by legislation.

Commercial Management and Support of the International Space Station (ISS). At the invitation of NASA, the Institute examined the administration’s approach to attracting commercial businesses into the tasks of managing and utilizing the ISS. Issues identified included the prioritization of space transport versus ISS construction; NASA’s transparency and willingness to involve the commercial sector; and confusion over the kinds of commercialization that NASA would accept in space. Despite findings by the National Research Council (NRC) that scientific advancement through research under zero gravity conditions could not justify the ISS budget and the oft-cited dangers of relying on the shuttle for transport, the program continued. Our study illuminated many of these issues, but policy remained unchanged out of other considerations.

Managing government R&D programs and the transition of S&T products. The Institute conducted numerous studies of how and how well technologies were being developed and transitioned into use by the DoD. In particular, case studies and metrics were developed for several assessments of ongoing and past DARPA programs to provide a better understanding of the effectiveness of agency program policies. These policies incorporated a surprising number of management innovations, including the use of technology prototype development and testing; unique approaches to involve customers in product design; and ways to enhance transition potential.

During this period, 9/11 occurred and the Department of Homeland Security was formed, but anti-terrorism R&D was, by most estimates, poorly initiated, including the department’s “DARPA-like” S&T arm called HSARPA.

Managing Weapons of Mass Destruction (WMD). Mid-20th Century thermonuclear devices first provided mankind the potential to annihilate life on the planet. During the next 20 years, science may well move us down more paths toward apocalypse. Perhaps as troubling, the enabling technologies may well be in the private sector, not confined to government development and control. During the early 2000 decade, the Institute performed a DoD-sponsored search for emerging technologies that could facilitate the detection of WMD materials or devices. In the face of a considerable spectrum of threats, prevailing national policies were (and remain) unable to fashion a truly interagency approach to solving these problems.

Figure 2. Percent of US R&D Financed by the Federal Government v. Business Sector. Source Data: NSF, Science and Engineering Indicators, 2014, Figure: 4-6.

Forecasting the Impacts of Science and Technology on Society. One of the most difficult and yet fruitful study areas addressed by the Institute has been a series of attempts to predict the trajectory and impact of various areas of S&T. It became clear that, while frustrating and contentious, these projections afforded insights that could not be gained through other pursuits. While most of the issues and opportunities that we investigated had been previously noted by others, more in depth perspectives were afforded by many through our analyses. We emphasized the need for carefully sculpted policies. Typically, little is done collectively at the national policy level to prepare for, or even anticipate, the futures of S&T areas. Some of the trends that we have investigated are listed as follows.

  • Medical science has a list of its own as medicine moves from empirical to knowledge-based. Furthering and applying a knowledge base drawn from a diverse set of scientific fields to improve diagnostics and treatment has led to a better ratio of chronic versus acute diseases and longer lifetimes. For example, breakthroughs in brain and mind research seem likely to offer promise to help cure many mental disorders. The medical bag will contain even more tomorrow: nanotechnology quantum dots, DNA analyses that are becoming more informative and reliable, and so on. “Elder-technologies” will assume priority with the graying of America and most of the developed world. Specific issues include: the application of DNA knowledge to disease; the overuse of antibiotics fostering an increase in resistant diseases; and fending off new and virulent diseases and epidemics; creating haves and have-nots by limiting who receives expensive treatment (e.g., brain enhancement through drugs, implants, or DNA manipulation). All of these issues require better guiding policies for investment.
  • Feeding earth’s population. The world must be able to feed and clothe its eight or so billion citizens over the next thirty years without destroying the ecosystem. To do this, we will need the combined effects of breakthroughs in agronomy, agriculture, and resource management. Here, national policies must become international policies.
  • Mitigating or adapting to global climate change. Few technical issue areas produce as much conflict as global climate change. Despite the nearly unanimous agreement among scientists that it is real and largely man-made, and the support of a majority of Americans for that judgement, we cannot seem to form a cogent and workable climate change policy – certainly not one capable of surviving changes in administrations or Congress. Yet, it seems likely that with careful analysis and management, a basic policy could be developed that would at least allow progress to address the phenomena as statistically plausible. This is another case where a mere national policy is insufficient.
  • IT trends. Communication, computing, and robotics have been exhaustively examined and yet Institute studies found that national policies have continuously lagged behind issues that need addressing. This is particularly distressing when one looks, however myopically, into the future of IT. Just a few issues that beg for policy attention are:


  • Big Data technologies and impacts. As both Big Brother and Big Corporation focus multi-spectral sensors on society, we hear more and more concerns over the loss of privacy and identity as opposed to the benefits of wisdom gained from statistical studies. The number of articles and books published over the past two decades provide sufficient evidence that big data is a subject that begs for wise policies, yet there have been few serious attempts to create even a beginning structure.
  • Managing the effects of automation on society. Advancements in robotics and other forms of automation have paid large dividends in releasing mankind from dangerous or repetitive work, while maintaining a constant surveillance on a broad front. The outbreak of epidemics, security breaches, safety, production standards, etc., are amenable to surveillance through technology. These are obviously to our benefit, but downsides, such as the elimination of jobs, need attention. In a recent book3 Nicholas Carr suggests that robots are being regularly promoted in the workplace, while the people they replace descend the career ladder. This problem has spread from blue to white collar and professions (e.g., on average, airline pilots actually control their aircraft for only 4 to 6 minutes per flight). Today’s John Henry is losing the race. Another current issue, that of drone surveillance, poses still another blow to expectations of privacy.


Trends observed from the studies

Our studies, conducted over the past twenty years, have convinced us that making the right S&T decisions today is especially important and is becoming more so as time passes. This is because of six indisputable trends:

First, the rate of breakthroughs in S&T has increased dramatically. Our studies in IT demonstrate this most clearly. Moreover, the time between these breakthroughs and their application is decreasing. Again turning to the IT example, a new chip, architecture, or electronic component is barely out of prototyping stage before it is housed in a device and on the market.

The ubiquity and impact on our lives of these products of science is growing. Stunning pictures from the Hubble telescope, frightening revelations about the effects of climate change, and medical treatments unknown just a few months earlier are changing us dramatically. In another example, jurisdiction of computer-driven autonomy is no longer confined to small or unimportant decisions. In fact, defaults to computer judgment represents a significant part of the growing impact of S&T on our lives.

Technical acumen of policymakers has not kept pace with S&T progress. Of course, this is mostly due to the enormous growth in scientific breakthroughs and applications that one must understand, but nearly every Institute study discussed earlier demanded many hours of concentrated effort to understand the fundamentals of the subject matter. National policymakers have little time to devote to such an effort.

While decision processes at the agency level are usually consistent and thorough, they become extremely ad hoc at the national level. As issues move on to this higher level of government, technical questions are more likely to be viewed through the lens of political ideology than scientific metrics, and participants who may represent a broader array of interests than the agency mission dominate at that stage of consideration. In general, there is a lack of formal, objective, and logical process devoted to the issues at hand.

DARPA’s deft hand in establishing operational policies to guide their dual use initiative and to facilitate their innovation demonstrates government’s ability to make policy work at the agency level. The Institute found few equivalent successes at the national policy community.

The solicitation and use of outside scientific advice is generally poor at the executive or congressional level. For the most part, lack of technical understanding is not due to the unavailability of good scientific advice. Although the processes of forming and managing advisory bodies in the federal government needs some work, government receives much more good advice than it uses or even considers. Government must be able to take advantage of that advice: to understand it, weigh it against other factors, build upon or reject it, and then use it to develop a consistent and wise policy structure.

Concluding directions

The advisor is often at arms length from the advisee, who frequently lacks the time to iterate conclusions and recommendations. Advice prepared for the government is generally poorly aggregated and rarely applied to the problem at hand. It is also vitally important that advisory panels or committees interact with government sponsors as recommendations emerge from their findings. In our experience, there are often unexpected side effects from seemingly unrelated policies. Often the government sponsor will be more aware of these interactions than advisors from outside their organization.

In creating policy, broad issues are often addressed in narrow terms because of political expediencies or disagreement, or simply because of inadequate technical understanding by the policymakers. Essentially all of our studies revealed surprisingly broad aspects of the most esoteric science and technology. For instance, technical advancements in CAD/CAM systems have affected the entire process of building or repairing ships, including billing practices. In another example, the concept of dual use technology development opened new worlds of production and marketing considerations.

It also became obvious to us that timeliness is important. Emerging technologies, such as nanotechnology and genetics research, require early national policy consideration, not only to direct investment and ensure innovation, but also to prepare society for the direct and indirect impacts of the research and application. Rather than a proactive response to the implications of emerging technology, there is a large gap of time between the appearance of new products of technical innovation and the development of policies to guide them. It is becoming increasingly important to close this gap. We were slow to adopt dual use policies, and we continue to be late in addressing climate change. Again, as difficult as forecasting is, it is often necessary to “get ahead of the game,” whether the issue is business or S&T.

Finally, at the point where a policy makes good sense and is clearly beneficial, the government needs a wise ombudsman and a well-defined and fair path to acceptance as a guiding principle. This stage of policymaking suffers from a natural reluctance to adopt policies due to the difficulty of doing so in an always-contentious political climate. More important is the challenge of updating, or even eliminating, policies when circumstances change or contrary information becomes available.

Specific as they may be to Potomac Institute’s experience, the forgoing observations lead naturally to actionable suggestions and to recommendations of ways to make policymaking better, which would involve political will. Implementing those changes would be daunting, but might prevent some future crisis or provide massive societal benefits. In any case, exploring those changes and pathways by which they might be implemented is best left to another article.


*“By public policy is meant that which the law encourages for the promotion of the public good. That which is against public policy is generally unlawful.” Story, Eq. Jur. Sec. 274. See Newl. Contr. 472.

1.Dealing with Democracy,Nature, Volume 425, 25 September 2003.
2. It was assumed by political analysts that funding was withheld because a Congress dominated by one party feared its legacy value to a President who belonged to the other.
3. Nicholas Carr, The Glass Cage (W.W. Norton & Co., 2015).


*“By public policy is meant that which the law encourages for the promotion of the public good. That which is against public policy is generally unlawful.” Story, Eq. Jur. Sec. 274. See Newl. Contr. 472

©, 2016-2024, Potomac Institute for Policy Studies, All rights reserved.

Most Read

From the CEO

Potomac Institute For Policy Studies CEO


STEPS (Online) ISSN: 2333-3200

STEPS (Print) ISSN: 2333-3219

STEPS Archives