Szilard Point

In cost-benefit analyses of scientific grant proposals, the Szilard Point refers to the scenario in which the expenses incurred in obtaining a grant (E) are equal to the value of the awarded grant (V). Beyond this point, where the costs outweigh the benefits (E > V), the return on investment for pursuing the grant becomes negative.

Because the Szilard Point is more of a qualitative construct than a quantitative measure, and because the concept is rooted in a fictional story, the term has acquired several divergent meanings, such as the point at which tasks other than research, such as applying for grants, peer review, and handling administrative chores, would consume more time and resources than research itself. All definitions, however, allude to the decrease in the probability of obtaining research funding and the increase in bureaucratic burdens encountered in academia.

Origin of the term
The term Szilard Point was coined in 1988 by Richard Smith (then assistant editor of British Medical Journal), and independently in 1990 by Daryl E. Chubin (then staff member at the Science, Education, and Transportation Program, Office of Technology Assessment, Congress of the United States) and Edward J. Hackett (then associate professor, Department of Science and Technology Studies, Rensselaer Polytechnic Institute).

Szilard Point was named after physicist Leo Szilard, who in 1948 wrote a short story entitled “The Mark Gable Foundation.” The story was published in a collection of short stories entitled The Voice of the Dolphins in 1961.

“The Mark Gable Foundation” explores several issues related to technological advancements, in particular socioeconomic changes, such as mass unemployment, as well as changes in societal norms and cultural practices, such as the widespread practice of teeth removal and the use of chewing plates. A major issue in the story is the necessity of scientific progress. The eponymous founder of the Mark Gable philanthropic foundation is asked by the unnamed narrator if he would like to use his considerable resources to advance science?” “No,” Mark Gable replies. “I believe scientific progress is too fast as it is.”

The narrator, then, suggests the foundation's resources be used to retard scientific progress. Mark Gable agrees, and the narrator tells him that retarding scientific progress is quite easy. All is needed is a foundation with a large annual endowment to which researchers in need of funds could apply for grants. Decisions will be made by ten committees, each composed of twelve scientists. The scientists on the committees will be the most active and productive scientists and they will be remunerated handsomely. This part of the plan will, in itself, contribute to the retardation of science by replacing productive scientific work with unproductive bureaucratic activities. The second part of the plan is to award huge yearly prizes for published scientific papers. This part will steer scientists towards problems that are considered promising in terms of publishable results instead of addressing questions that are difficult and important.

“For a few years there might be a great increase in scientific output; but by going after the obvious, soon science would dry out. Science would become something like a parlor game. Some things would be considered interesting, others not. There would be fashions. Those who followed the fashion would get grants. Those who wouldn't wouldn't, and soon they would learn to follow the fashion, too.”

Szilard Point, Grant Proposals, and Peer Review
Szilard Point (whether by name or by implication) has frequently been used in arguments against the practice of peer review in judging grant proposals. In 1984, Rostum Roy, a material scientist at Penn State University, used the following real-life example to illustrate the wastefulness of this system.

In 1983 the Department of Defense started a new program making available $30,000,000 annually to provide some large items of research equipment to universities. The Department of Defense issued an invitation for proposals to all universities. Over 2,200 proposals were received for a sum of $625 million. The success ratio was less than 1:60. The time required for the preparation and submission of each proposal was estimated at one month's work of one person. Thus, more than 200 years of scientific work were diverted from research and devoted to the writing of grant proposal. If we include an average figure of 100% for overhead costs, the cost would be equal to 400 yearly salaries. Given the success ratio of 1:60, each scientific group would need to write on average 60 proposals to get one grant. A back-of-the-envelope calculation would yield a cost of about $480,000,000 for an allocation of $30,000,000. This cost does not include any administrative costs. This is an extreme example concerning a methodology that was fortunately not repeated, however, with 15–20% percent success rates, a “Szilard point” where waste exceeds benefit is no longer a frivolous speculation but a real possibility.

In the late 1990s, Robert Mullan Cook-Deegan, then at National Academies of Science, Engineering and Medicine noted that while administrative review costs at NIH or NSF rise arithmetically with the number of applications, external costs rise much faster as the percentage of proposals that are funded falls. If half of all proposals result in funding, which was the case at NIH in the 1960s, one unfunded grant proposal was prepared for each one funded. When success rates fall to one in five or one in six, as they have in several areas, four or five proposals are wasted for every one funded. The Szilard point is reached when the success rate of grant applications is one in 10—a ratio that has been reached in many subjects. Indeed, it may be worse: some requests for applications produce 30 applications and only one grant, meaning that the money spent on designing and reviewing the applications far exceeds the time spent doing the research. Preparing a grant proposal is a substantial effort, and the total external costs for all applicants may approach or even exceed the amount awarded to the successful one.

The implicit use of Szilard Point
The study of the bureaucratic factors retarding scientific progress, has become quite fashionable after 2000. Sadly, the term Szilard Point isn't mentioned in most such studies.