Spending on basic research generates jobs, wealth and innovation with major impact on our life and future. But what is science really worth and how science contributes to economic growth are questions not easily answered. Even more, our thinking on this aspect may be flawed.

An international, interdisciplinary conference on “The Economics of Innovation: From Basic Research to the Market” was held at the Fondation Brocher in Geneva, in September 2017. The meeting brought together economists, basic research scientists, policy makers, science funders, venture capitalists, and industry in the uniquely inspiring setting of the Brocher Foundation, to promote cross-disciplinary dialogue and inductive reasoning. The general consensus was that the research being done at the major research universities drives discovery and development of new products and processes, which eventually reach the market, creating jobs, capital and are essential for addressing the major welfare challenges of our time. Basic research is the key driver of innovation, the results of which set the foundation for the world of tomorrow. But there are some basic questions that remain extremely challenging to address and are worth thinking about. The present paper jump starts the dialogue on this complex problematic and asks a first question: what is the return on the investment in basic research? Can we calculate a precise and all-inclusive rate of return on investment?

Research cannot pay itself

Why is this question so difficult to answer? Is it due to the often-lengthy time lag between discovery and application or a problem with measuring the outcome? A simple way to look at the problem is to ask how does the income from licensing compare to the investment in the universities. This is a very tangible and easily measured statistic. Unfortunately, at what many would consider the most successful research universities like Stanford, Harvard and MIT, licensing income surprisingly represents only a small fraction of the total research expenses. In other words, probably no one can expect even the best research universities to have their research pay for itself. On the other hand, where would we be without the major discoveries from the research universities? Clearly, health care and many other life-style elements would be worse off. What is the missing element in this conundrum?

Research outputs as return on investment

Research outputs often comprise both tangible and intangible aspects thus making it even more difficult, though maybe not impossible, to quantify the return on investment. For instance, basic researcher, Kai Johnsson, spoke of one of the startups he helped to create. Even though the startup provided employment for several people for a few years, it eventually went out of business. Johnsson’s conclusion was somehow paradoxical: he considered this “failed” start-up as a successful enterprise. People lost their jobs along the way but they all improved their entrepreneurial skills and, remarkably, found even better jobs afterwards. In a broader sense, this example illustrates the idea that academic research is of value not only for the potential material benefits it may yield (that could be quantified) but in addition it may improve lives in many different ways, difficult to evaluate beyond wealth generated by these companies.

Why is it important to invest in basic research ?

As a result, if we cannot quantify easily how much society profits economically from basic research, are there clear and important benefits from investments in basic science that would compel us to continue doing so? The president of the SNSF, Prof. Matthias Egger, provided an insightful vision on the topic. Firstly, basic science increases the amount of useful and sometimes marketable knowledge. As a possible readout of this parameter, Nature recently created the Nature Innovation Index. This traces the path from basic research findings to commercialization of new products and services, and accounts for the often long lag between discovery and application. Interestingly, the University of Geneva was ranked 21^st worldwide and 2^nd in Europe in this index, indicating that the knowledge generated from research at the University provided benefits to society even though many of these products might not have been developed directly by researchers at the University.

Another major output of basic research is the training of skilled graduates, who not only acquire highly specialized skills, but also gain soft skills such as the capacity for problem solving and persevering in the face of adverse conditions that allow them to adapt to new situations. Hiring these potential employees is clearly an attraction for enterprises, especially those depending on new technologies. These high-tech companies create employment, make benefits and pay taxes, which are rarely included in the economic output of basic research. It is purposefully that the major high-tech hubs, in Switzerland and elsewhere, are located nearby strong research centers, such are the Silicon Valley (near Stanford, UC Berkeley, UCSF, etc.), more Boston’s Route 128 (Harvard, MIT, Tufts, etc.), the Research Triangle (Duke, UNC, etc.) in the US, and Cambridge, Zurich Technopark (UNIZ, ETHZ) and innovation parks within the Lemanic Region (UNIGE, UNIL, EPFL, CERN, IMD). This proximity is attractive to high-tech industry due to the availability of highly trained personnel and connections to research centers where they can often gain access to sophisticated facilities and expertise.

A third argument is the fact that basic research bears innovation value in often foreseeable ways, i.e. research creates new scientific instrumentation and methodologies that benefit society in various ways. For example, an old discovery, that of radiation could be developed to be used in medicine, not only to visualize bones, but to treat certain cancers. Novel technologies to detect and treat cancers at early stages have vastly improved the prognosis for many cancer patients.

Investment in research also improves scientific networking and thus the ability to solve new problems. Having a certain concentration of scientific research in a region leads to increased interactions, knowledge transfer and an added-value through interdisciplinary collaboration. While one can easily find examples of this type of contribution, it is very difficult to quantify its importance on society.

Beyond the creation of start-ups, mentioned above, and the advantages to society that they bring, basic research improves the level of knowledge in the society. Many scientific studies are reported by journalists, covered by the media and explained to the general public in other forums. An increased public understanding is an intangible benefit of the investment in research.

But are all outputs of basic research beneficial to society?

This is an important question in the eyes of public policy makers that was brought up by Clare Craig, director of Science Policy at the Royal Society, UK. There clearly are some effects that can be considered negative. For example, the global economy is largely driven by new technologies, which have permitted an extension of the expected lifespan, but this has also increased the world population and together they probably both contribute in part to the increase in CO₂ emissions. While few would argue against increased life expectancy as a positive outcome, it does create new important problems that the society needs to deal with and there might be some costs to this. Another example pertains to the advances in technologies. While they clearly have led to an increase in general health, household income and general welfare, income inequality in the developed societies has increased at the same time. In addition, the benefits of increased life expectancy are not evenly distributed geographically, with poorer and less educated regions benefiting less. These are also issues that need to be addressed and are becoming important preoccupations of politicians.

How good are our mechanisms to allocate the funds?

Investments in basic research have an overall benefit to society but how good are we at allocating resources to the best research? Some interesting parallels appeared at the conference by two completely independent analyses, one presented by Maria Leptin, Director of EMBO, and the other by an economist, Hugo Hopenhayn, Professor of Economics, UCLA.

Current practice for the allocation of research funds uses peer review with the disturbing tendency to rely heavily on journal metrics and citations, even though many scientists have risen against this in the DORA declaration. There is also a tendency to favor what is termed high-risk, high-gain research, which also means that there should also be a high degree of failure. One can examine cases retrospectively and clearly find examples where publication metrics were not predictive of future success. Leptin put forth the example of Peter Hegemann (Professor and Head of the Department for Biophysics at the Humboldt University of Berlin) who was instrumental in development of optogenetics. His citation metrics were not very impressive before he made his crucial discoveries and even lagged behind the first years after receiving scientific prizes for his discoveries. Another example is Jacques Dubochet (Professor Emeritus at University of Lausanne), one of three recipients of the 2017 Nobel Prize in Chemistry. His crucial discovery that opened up new possibilities to examine protein structure in vitreous ice by electron microscopy was published in the Journal of Microscopy, hardly a member of the high impact group of journals. So, is it wise to finance exclusively what is perceived as high impact science when our judgements are not entirely accurate or would it be wiser to spread the funding more broadly? What is better for society and discovery in general?

Hopenhayn modeled this problem in an economic model. His analogy for the high-risk, high-gain project was high-hanging fruit and the model for other research was low hanging fruit. Using his model, he argued that market forces push towards the high-hanging fruit because the profit that can be made is larger. By attempting to reach many high-hanging fruits the subject has a higher chance of being successful getting one. The driving forces here would be the scientist or entrepreneur on one hand and the review panel or investor that allocates the funding both favoring the higher risk. But is this better for society?  The model would predict that a combination of high-risk, high-gain, and lower risk projects would give a better total productivity. Of course, in both, the pure model and the real-life situation of science funding, this is all a question of setting precise parameters for risk and gain. What we should therefore consider is whether our current system of research funding needs improvement. Leptin argued for creating an environment that allows the best scientists to take risks rather than favoring high risk project per se.

How can we improve the process of innovation?

When one discusses innovation, one usually puts most emphasis on the creation of companies that are “scalable” meaning that they will exhibit high growth and returns. This is often the case for the biomedical industry and development of new medicines. Typically, a small startup would be founded, usually based on IP, and would obtain seed-funding to begin the testing and development of their product. At this point, venture capital could enter the company, but this usually will substantially change the decision making with the venture capitalist now having a say and perhaps the decision-making power. As presented by VC Clare Fairfield, the interests of the venture capitalist and the entrepreneur do not necessarily coincide. For instance, the venture capitalist might want to retrieve his investment by selling off early, even though the best option for the innovation process would be to wait. Therefore, venture capitalists often have an important impact on the life of the company. Is this positive for the innovation process? This might depend on the venture capitalist. According to Fairfield, about 90% of venture capitalists produce negative returns on their investments, which suggests that sometimes it might not be a good idea to have them making the decisions. So, what makes for a successful venture? Education is important. Getting advice from successful mentors and coaches, learning from the mistakes of others, increasing education in entrepreneurship should help. Aebisher (former EPFL President) and Fairfield were clearly in agreement on this. Education in entrepreneurship, at all levels, from the scientist to the venture capitalist, should improve the success rate of the process, which for startups is currently quite low. Also, it is important to learn from other successes and failures and not try to copy other models such as the one represented by Silicon Valley. According to Fairfield, although there are incredibly successful companies in Silicon Valley, the model is flawed, because it accepts a very high failure rate, no longer focusses on early stage investing but on the money return, bypasses the entrepreneur and fails to treat the industry as a people business. Probably one key to success is to pay attention to the people at each stage of the innovation process, make sure that they are educated in entrepreneurship, particularly concerning the stages at which they are acting and ensure a systematic and careful attention to detail at all stages.

Perspective on the Economics of Innovation and basic research

It is clear that basic research, coming from public or private investment, plants the seeds for development of new technologies and innovation, but the road to the market is complex and still poorly understood. The basic question of what is the return on the investment in basic research is extremely difficult to address, although there is no doubt that there are many benefits to the economy and society. Perhaps our thinking is flawed and we should not be asking how much we will make with our investment in basic research, but think hard about what our lives would be like without it. With this perspective, thinking globally, it is clear that basic science has had a transformative influence on our society and is probably one of the major keys to solving our current problems.