February 12, 2020
By: Glenn Blomquist
Outer space probes, radio telescopes, large particle accelerators, genomic platforms, and similar entities are fascinating focal points of intellectual curiosity and discovery. While we marvel at what we learn from them, we can be taken aback by their costs. The $150 billion cost for the International Space Station has been shared by taxpayers in the United States, Europe, Russia, Japan, and Canada. The Large Hadron Collider near Geneva, Switzerland that enabled the 2012 discovery of the elusive Higgs Boson is estimated to cost about 13.5 billion Euros over the 1993-2025 period for taxpayers in the participating countries. According to a recent article in The Economist, proposals are being made to build new infrastructures: A Future Circular Collider in Switzerland, an International Linear Collider in Japan, and a Circular Electron-Position Collider in China. Such are the topic of Massimo Florio’s book, Investing in Science: Social Cost-Benefit Analysis of Research Infrastructures. In it, he demonstrates that benefit-cost analysis (BCA) can be useful in answering the question: Are these costly research infrastructures worth it? He draws upon his substantial experience to adapt the traditional framework to the specific characteristics of research infrastructure (RI). He identifies elements common to RIs, describes how they can be measured and valued, and gives examples from work that he and others have done. This pioneering book fills a gap in that such large-scale investments in science only infrequently have been evaluated using BCA.
Florio begins by distinguishing between Big Science such as the Manhattan Project to develop nuclear weapons or the Apollo program to land astronauts on the moon that furthered US national goals during World War II and the Cold War and RIs that are collective, typically international, enterprises designed and managed to create knowledge and related services for multiple users. Given the public good nature of much of the output, wide dissemination is common, something that stands in contrast to the secret classifications given to output of Big Science associated with the Department of Defense. The public good nature also presents challenges for benefit estimation and funding, some of which are discussed later in the book. The first two chapters lay out the fundamentals of applying the BCA model and basics of estimating the costs of RIs. The book closes with a chapter about stochastic net present value and sensitivity analysis.
The primary contribution of the book, however, is found in the chapters about estimating benefits. Chapter 3 about benefits to scientists makes a good case that the net benefit is the value of influence beyond the RI and is measured by the value of the publications and citations by external scientists to RI publications. Their value is the opportunity cost of the time spent producing the publications or time spent reading and understanding for citations. Chapter 4 is about benefits in the form of human capital formation. RIs contribute to the experiential learning and productivity of students and post-docs and the value is measured by the earnings premiums the early career researchers receive over their work lives. Chapter 5 is about direct benefits to firms. They are measured by the value of the patents generated by the RI and, for non-patented innovations, the marginal shadow profits. Spillovers are indicated by backward citations by a new patent to an RI-generated patent and forward citations that are the number of times the new patent gets cited. The private values of the patents are the private returns gross of tax, interest, and depreciation. The external values are determined using the patents and values of patents in the relevant technological field.
Benefits to users of information technology in the Big Data era in the form of free data depositories and software are the focus of Chapter 6. Florio describes how the World Wide Web (WWW) was initially conceived at the European Organization for Nuclear Research (CERN) as a way to share information among scientists working on particle physics. Benefits are estimated by the value of time saved searching for information and, alternatively, by stated preference studies that elicit the value of open access. Chapter 7 deals with benefits to users of science-based innovations such as new cancer treatment therapy and Earth satellite systems. A formidable challenge can arise in determining how much quality and length of life are improved by the innovations (causality). However, for RI such as the Copernicus satellite system that monitors ice conditions on the Baltic Sea, reduced costs of shipping provide a conservative measure.
Benefits to visitors who consume the scientific culture at RI facilities are the subject of Chapter 8. For example, the National Aeronautics and Space Administration’s (NASA) Kennedy Space Center in Florida has 1.5 million visitors per year with adults paying $57 each. It has over 11 million Facebook followers and more than 28 million followers on Twitter. A travel cost method is used for valuing on-site visitation. The number of virtual visitors, time spent, and value of their time is used for valuing virtual visits. Stated preference studies complement the revealed preference methods. In addition, citizen science is valued as time spent by volunteers multiplied times their value of leisure time plus the value of time saved by professional scientists.
Valuation of science as a global public good is the topic of Chapter 9. Florio draws on theory and methods developed for BCA of environmental and resource policy proposals and regulations. The 2019 NASA budget of $20 billion dollars directly or indirectly costs every adult in the US roughly $95. Florio argues that intrinsic (nonuse or passive use) values of scientific discovery are relevant just as they are for valuation of environmental and resource regulation and policy. He suggests limiting the scope of intrinsic benefits to taxpayers of the countries that would actually finance the investment to yield a conservative estimate. He demonstrates feasibility by reporting results of a stated preference willingness to pay study for the Large Hadron Collider. This intrinsic value of knowledge per se is included in BCA of research infrastructure in addition to the benefits described already. The intuition is that some taxpayers support RI because it might turn out to be useful and they just want to know more.
The book is well written. Nearly every chapter has detailed examples and an appendix that contains suggestions for further reading. The style is appropriate for analysts who do BCA as well as individuals advising or deciding about science policy. Some may find bits of the benefit estimation provocative, but most everyone interested in science research infrastructure will find it informative and worth reading.
Glenn Blomquist is Emeritus Professor of Economics and Public Policy at the University of Kentucky and former editor of the Journal of Benefit-Cost Analysis.
- "Assembling the Future" The Economist (January 4-10, 2020)
- Massimo Florio. Investing in Science: Social Cost-Benefit Analysis of Research Infrastructures. Cambridge, Massachusetts and London, England: The MIT Press, 2019.