Issues Magazine

Do Patents Alter the Direction of Scientific Inquiry? Evidence from a Survey of Academic Scientists

By Paul H. Jensen and Elizabeth Webster

Nearly half of the academic scientists in a recent study reported that their choice of research projects had been affected by the presence of other parties’ patents. Issues with patent permissions and the culture of the workplace have the largest influence over whether or not patents affect the direction of research.

As universities around the world have pursued more aggressive commercialisation strategies, debate has intensified about the potentially deleterious effects of patenting on the scientific community. In particular, concern has been expressed that patents may increase secrecy, hinder informal knowledge transfer and alter the trajectory of scientific inquiry. Although there is empirical evidence from surveys of academics in support of these claims, others have questioned the contention that patenting has a damaging effect on academic endeavours.

Recent research has found that patenting has a positive effect on both the quantity and quality of academic patentees’ publications. However, the analysis largely ignores one of the major concerns about patents in academe: the effects of patents on third parties. That is, they focus primarily on the direct effect of patenting on a scientist’s own subsequent academic publication rate.

Here, we focus on the potential negative externalities associated with patenting in countries without a statutory research exemption. In particular, we examine whether patents alter the direction of scientific inquiry in Australia via their decisions about which research projects to undertake.

Background and Hypotheses
Historically, the separation between “science” and “technology” has been quite clear: “science” addressed basic questions about the laws of the universe and made their findings publicly available, while “technology” addressed practical problems and found proprietary solutions to these problems using knowledge often created by science. As a corollary of this, the boundaries between public and private knowledge were also fairly distinct.

Scientists worked in a domain where ideas, research tools and results were freely shared with colleagues via journal publications and academic fora. Technologists apply knowledge to create products and processes that are valuable to consumers, and in the process earn profits for their company. This binary taxonomy overlooks important areas where science and technology overlap.

In fact, an increasing proportion of science is now conducted in the domain – known as Pasteur’s Quadrant – where demand-inspired basic research takes place. This has effectively blurred the boundary between science and technology, and has led to a dramatic increase in the extent of patenting in the traditional academic realm. Hand-in-hand with this transformation has been an increased reliance on private sector funding of academic research.

The wisdom of increased patent intensity (and private sector funding) in academic research has been questioned by many. Other evidence suggests that we should care about restricting the degree of openness in academic endeavours.

Whether patents can alter the trajectory of scientific progress is partly a function of whether patent law applies to researchers. Historically, academics and public sector scientists have believed that as long as they do not commercialise their inventions they are exempt from patent law.

Although the degree to which a research exemption exists in other countries is unclear, there is widespread concern in jurisdictions without a statutory research exemption that permission to use patented technology is required. The concern is that, in the absence of a research exemption, patenting in the public sector will divert research away from areas that are intrinsically interesting but heavily patented into areas that are less interesting and less intensively patented.

In Australia there is no statutory research exemption, and there is widespread concern that the common law research exemption has been all but extinguished. In this context, our contention is that the individual scientist’s belief regarding the existence of a research exemption determines their behaviour regarding the choice of scientific research project.

Hypothesis 1: Patents are more likely to alter the direction of scientific inquiry if public sector scientists believe that there is no research exemption in patent law.

Over and above their beliefs about the existence of a research exemption, the direction of scientists’ research is likely to be influenced by their recent experience of being refused a request to use patented research technologies. Hence we conjecture that:

Hypothesis 2: Patents are more likely to alter the direction of scientific inquiry if public sector scientists have been denied access to use patent research tools in the recent past.

Even if a patent owner grants permission to use the patented technology to another scientist, the direct cost (i.e. the licence fee) or the indirect costs (i.e. the transaction costs associated with negotiating the contractual terms of the deal) may be substantial enough to make the scientist consider other alternative research questions. This may lead to the creation of what has been referred to as an “anti-commons”. This leads us to the development of our next hypothesis.

Hypothesis 3: Patents are more likely to alter the direction of scientific inquiry if public sector scientists believe that the transaction costs associated with negotiating a deal with patent owners are high.

Attitudes at the research workplace towards proprietary research, the need for secrecy and non-disclosure may also affect how willing scientists are to change their own research program in response to others’ patents. We argue that scientists working in environments where certain traditional values are very strong will be less likely to change their research inquiry as a result of others’ patents – partly because they will ignore them. This culture may also be influenced by the source of funds that the laboratory relies on to finance its experimentation: if there is a greater reliance on external industry funds, where there is probably a stronger emphasis on commercial application of the technology, researchers may be more careful about selecting topics that are heavily patented. This leads to the creation of our two other hypotheses.

Hypothesis 4: Patents are less likely to alter the direction of scientific inquiry if public sector scientists work in an environment (laboratory or department) that embraces the culture of “open science”.

Hypothesis 5: Patents are more likely to alter the direction of scientific inquiry if public sector scientists rely on contract research, commissioned research or industry funds.

Survey Data
The survey specifically targeted public sector researchers in the physical, agricultural and health sciences – the most patent-intensive areas of research – across the top eight Australian research universities and major public research institutes in Australia (such as the CSIRO). Together, these publicly funded research organisations account for the overwhelming majority of Australian public sector research activities.

An examination of responses indicates that senior, junior and pre-PhD researchers were more likely to respond to the first wave. Junior, pre-PhD and administrators were more likely to respond to the second wave (however, the last two groups were small in number). Three-quarters of the population surveyed were either senior or junior researchers, with the bulk of the reminder being professorial or head researchers.

Background data on important characteristics such as the scientist’s age, gender, department, university, years since completing their PhD, source of research funds, whether they worked in a research-only environment, and research team size were also collected. In addition, both waves of the survey asked questions on the following topics: the understanding of existing patent law; sources of knowledge; patent and licensing behaviour; materials transfer agreements; effects on research and workplace culture.

We opted for a broad question that did not limit itself either to anti-commons effects, projects that had already commenced or specific types of effects (such as restrictions on publication, excessive royalties). Moreover, we did not ask the respondent to assess whether the effect was positive or negative. Our question asked respondents to rate on a 1 (“no effect”) to 7 (“major effect”) Likert scale their position with respect to two contrasting statements: (a) “Other researchers’ patents have no effect on which research projects we chose to undertake”; and (b) “Other researchers’ patents have had a major effect on our choice of research projects”.

Table 1

Our data showed that 53.3 per cent of respondents believe that other researchers’ patents had no effect on their choice of research projects (Table 1). The remainder indicated that their choice of research projects had been shaped to varying degrees by the existence of other scientists’ patents. Although only 2.8 per cent of respondents stated that patents had a major effect on the choice of research project, a total of 24.4 per cent reported a score of between 4 and 7. Engineering, followed by science, indicated the greatest effects.

At least at the prima facie level, this suggests that patents are playing a non-negligible role in the direction of scientific progress in Australia. This question forms the dependent variable used in the regression analyses.

We used data collected from two waves of a survey of Australian scientists administered in 2007 and 2009. In the survey, we asked a specific question about whether the presence of patents in a particular research area shapes the scientist’s decision about the direction of their research. Our survey data – which contains 4513 observations from a mix of patenting and non-patenting scientists in Australian universities and public sector research institutes (“academic scientists”) – suggest that patents have directly affected the choice of research project (to varying degrees) for approximately 47 per cent of all respondents.

Research-only scientists, middle-aged scientists (in the 36–45 age bracket) and those from the engineering field of research were significantly more likely to indicate that their direction of research had been affected by other researchers’ patents. There was no significant difference between the 2007 respondents and 2009 respondents (both with the full sample and a matched sample). Gender had no discernible effect.

Table 2

Respondents were asked to indicate whether or not they thought a specific statement regarding the need to acquire permission to use patented research technologies was true or false, or whether they were unsure (Table 2). These questions capture Australian scientists’ beliefs about the existence of a research exemption, which are of obvious importance to our study since scientists will only modify their research direction if they believe that they are required, by law, to obtain permission (or pay a licence fee) for the use of patented technologies.

The responses reveal that 34 per cent of respondents were unsure about whether or not they needed permission to use patented research tools and techniques if that use is purely for research, and 22.5 per cent believed that they did not need permission. Respondents were more likely to believe that researchers needed permission if the researcher was employed in the for-profit sector (even if the use was for research) or if the use had commercial intent (regardless of sector of employment). However, there were high levels of uncertainty about the need for permission if the patent was owned by a public organisation.

Modelling and Results
To test our hypotheses, we modelled the determinants of the decision to change research direction due to the presence of patents. The decision to change direction is a function of the following factors:

• the scientist’s beliefs about the prevailing law regarding the existence of a research exemption;

• the scientist’s recent experience with regard to requests for permission to use a patented research tool;

• the degree of difficulty associated with negotiating with patent owners in relation to permission to use a patented research tool;

• the source of funding the scientist relies on;

• the culture of their department; and

• a set of other background factors including age, gender, years since completing PhD, faculty and survey year.

Overall, we found strong evidence supporting our hypotheses across the different models. The results reveal that scientists’ beliefs about patent law does have a significant effect on whether they have changed direction because of other researchers’ patents.

Across all four models, legal understanding is positive and statistically significant. Thus we conclude that there is strong support for Hypothesis 1. This result is consistent with our expectation since it suggests that researchers who believe that a research exemption exists would simply choose to ignore the presence of patents. Such a conclusion resonates with previous findings that the dominant behaviour among scientists is to “ignore patents”. This observation is also backed up by evidence in the US that some scientists believe they wouldn’t be sued, even if they infringe. In this instance, patents would not have any effect on scientists’ research trajectory.

There is also some fairly strong evidence supporting Hypothesis 2, which implies that recent refusals to use patented research technologies will subsequently lead scientists to believe that their choice of research projects has been affected. The variable in our model that captures this effect – experience – is positive and statistically significant in all of the estimations. Once again, this result has some intuitive appeal.

There was support for Hypothesis 3 that high transaction costs will alter the choice of research program. The more researchers’ departments relied upon commercial funding, the more likely they were to change their research in response to patents. In addition, research-only scientists were also more likely to report having changed their research direction, possibly because they work in a more applied area. The degree to which the scientist’s department embraced the open science ethos was highly significant: scientists were less likely to change the course of their research the more open their departments.

The results on the control variables are also worth noting. Engineering was most likely of all faculties to indicate a change in research direction, and architecture the least, all other things being equal. Larger research teams were more likely to have changed direction and young people were more likely to have been affected. However, there was no significant difference by gender, year or time since completing PhD.

The statistically significant variables noted above are not all of equal impact: negotiating with patent owners and department culture had the largest impact, and faculty was moderately important. The size of the research team was the other notable factor, with larger teams more likely to have had their research direction affected.

Nearly half of all scientists reported that their choice of research projects had been affected, to some degree, by the presence of other parties’ patents. The closest comparator we have for our question comes from biomedical surveys in the US. Respondents indicated that “too many patents” was rated most highly as a reason for not pursuing projects in 3 per cent of cases, whereas the comparative highest rating for our question was also 3 per cent. It is not possible to quantitatively compare our findings with these studies due to differences in question format.

It seems likely that patents lead non-patenting scientists into areas of inquiry that are free(er) of patents. Thus non-patenting scientists are pushed further into the realm of esoteric, abstract science where there are no (or fewer) patents, while the patenting scientists are being pulled further into research with potential commercial value.

Our results suggest that scientists are affected by the presence of patents, which is a cause for some concern if we accept that decision rights over research projects should be vested in the individual scientist. Although we cannot quantify the effects of this, or assess whether the effects are positive or negative, we believe that it provides some support for the idea that the government should re-evaluate whether or not to implement a statutory research exemption.

One implication of our results is that scientists may be choosing to move into more “basic” areas of research, since these are more likely to be patent-free environments. This suggests that the use of patents in the public sector may have two countervailing effects. The first is the conventional argument that patenting encourages scientists towards more “applied” research. The second is that, as the amount of patenting in the public sector increases, scientists may be forced further “upstream” into more basic research realms in order to find patent-free environments in which to conduct their research.

There are some valid reasons to argue that the impediments to free choice of research project that are imposed by patenting in the public sector research environment may come at some cost. For instance, it is well-known that scientific progress is uncertain and non-linear: even if there is a clear definition of the technological problem at hand, it can sometimes take numerous failed attempts by many different scholars (or research teams) to come up with a solution. In this uncertain environment, one way of pooling the risk is to let many different scientists work on a problem simultaneously.

This competition among different scientists can generate important benefits – although it is important to note that it can also lead to wasteful duplication of resources. One often-neglected cost of protecting IP in academia is that it can reduce the diversity of experimentation.

This article is adapted with permission from Intellectual Property Research Institute of Australia Working Paper No. 5/10, November 2010. The full paper is at www.ipria.org/publications/Articles.html