Uncertain IP leads to less innovation

We've been debating patents on the blog recently. Charlotte wrote about a really cool experiment that appeared to show that IP limits follow-on innovation.

I previously wrote that the follow-on benefits of innovation were on net positive, because the effect of bringing what would otherwise be business secrets into the open outweighs the cost of firms not being able to build on other firms innovations for free.

A newly published paper takes on another angle: collaboration. Entitled "R&D Collaboration with Uncertain Intellectual Property Rights" (full pdf of 2011 version) and by authors Dirk Czarnitzki, Katrin Hussinger, and Cédric Schneider it argues that firms shy away from working with other businesses when their intellectual property rights are less clear.

Patent pendencies create uncertainty in research and development (R&D) collaboration, which can result in a threat of expropriation of unprotected knowledge, reduced bargaining power and enhanced search costs. We show that—depending of the type of collaboration partner and the size of the company—uncertain intellectual property rights (IPRs) lead to reduced collaboration between firms and can, hence, hinder knowledge production.

Among firms with patent applications the average probability to collaborate with a competitor amounts to 13%. The probability of collaborating with a competitor decreases by 3% points for these firms if the share of pending patents in the patent application stock increases by one standard deviation at the mean. Thus, the average probability of collaborating with firms in the same industry is reduced by about 23% (=3/13), which is a sizeable impact.

Take that, Charlotte!

With property rights, there are plenty more fish in the sea

We at the Adam Smith Institute need little further evidence that property rights are the best way to an efficient allocation of resources. Even so, more literature on how property rights can work in different industries and regulatory environments is always welcome. A new National Bureau of Economic Research paper looks at how the strength of property rights can affect regulators' willingness to allow the exploitation of natural resources. They focus on the most common system of regulation, which sees a limited number of firms given the right to extract to the level of a cap set by a regulator. They attribute this, at least partly, to a benign form of regulatory capture.

Commonly, it is seen as an unwelcome anticompetitive force, leading to the overexploitation of resources by monopolistic producers in industries with clearly defined property rights. However, because of the temporary, weak, and ill-defined nature of rights in the natural resources sector, the authors suggest that this analysis is not applicable. Instead, they find that

when property rights to the resource are strong, the regulator’s choice (which is the product of resource harvesters’ influence) coincides with the public interest. However, when property rights to the resource are weak, the regulator’s choice leads to overexploitation. This suggests that the resulting extraction level is closer to the socially-optimal extraction level when rights to the resource are strong.

The authors distinguish between 'weak' and 'strong' property rights using the probability that such rights will be revoked – the more likely, the weaker the rights. They propose that, when rights are strong, firms influence regulators (either formally by voting in regulatory councils, or by informal means) to choose a lower extraction rate than they would in a situation with less secure property rights, because they are less concerned about those rights being revoked in the future. In addition, regulators discount utility from future harvests less when there is less risk of rights being revoked, causing them to favour less current extraction.

The paper tests this thesis empirically against novel panel data from 178 of the largest commercial fisheries, and finds that regulators are "significantly more conservative" in their management of resources when property rights are most secure. In those cases where poorly managed fisheries switch to a 'Catch Share' system, with more secure property rights, there is a significant fall in exploitation, supporting their thesis (the fall prior to the switch is attributed to a gradual policy change in the face of overexploitation):

If in practice the Coasean idea that the assignation and enforcement of property rights – through their effects on the decisions of regulators – lead to more efficient outcomes, this has important implications for policy. It gives us an even greater incentive (as if we need it) to promote the institution of secure property rights, especially in those resource-rich low-income countries which could be subject to a swift depletion of natural resources due not only to tragedies of the commons, but also to the insecurity of extractive firms property rights.

Uber: helping drivers, helping customers

The first comprehensive analysis of Uber 'partners' (i.e. drivers) has come out, written by Dr. Jonathan Hall, head of policy research at Uber, and Prof. Alan Krueger, of Princeton, and formerly Barack Obama's top economist. The results in short: Uber provides flexible employment at higher per-hour wages than traditional taxi driving, while building up reputational capital that traditional taxi systems cannot offer. It does not undermine traditional employment more general, or enhance inequality, but we all know how cheap the fares can be, and how useful the service is (this previously led me to believe that its stratospheric valuation might be justified).

This paper provides the first comprehensive analysis of Uber’s driver-partners, based on both survey data and anonymized, aggregated administrative data. Uber has grown at an exponential rate over the last few years, and drivers who partner with Uber appear to be attracted to the platform in large part because of the flexibility it offers, the level of compensation, and the fact that earnings per hour do not vary much with hours worked, which facilitates part-time and variable hours. Uber’s driver-partners are more similar in terms of their age and education to the general workforce than to taxi drivers and chauffeurs.

Uber may serve as a bridge for many seeking other employment opportunities, and it may attract well-qualified individuals because, with Uber’s star rating system, driver-partners’ reputations are explicitly shared with potential customers. Most of Uber’s driver-partners had full- or part-time employment prior to joining Uber, and many continued in those positions after starting to drive with the Uber platform, which makes the flexibility to set their own hours all the more valuable. Uber’s driver-partners also often cited the desire to smooth fluctuations in their income as a reason for partnering with Uber.

As we see above, Uber drivers really like their jobs, and that's probably why so many of them are still there a year later. I actually feel quite sympathetic towards existing taxi drivers both in the UK and US. They were forced by existing rules to invest heavily in getting their privileged spot in the market place, and Uber is effectively circumventing this process altogether.

This suggests we should compensate taxi drivers so that in the future people are not so worried that tech changes will force transformational rule changes that will ruin them. But this progress promises improvements on practically every margin of taxi driving; I can imagine a future where no traditional taxi driving exists—indeed with self-driving cars I can imagine a future where only an Uber-style rental-taxi system exists. So, compensation aside, it must go on.

Single sex schooling: three recent papers

Nearly all state schools are co-educational, but most independent schools are single sex. Three academic papers I came across in the last few months suggest that the education authorities might have something to learn from the private sector—all three find that randomly switching people to single-sex education leads to substantial improvements in their outcomes. The first, Lee et al. (2014) looks at random assignment of Korean youths to middle schools, comparing single sex classes in coed schools with coed classes in coed schools and single-sex classes in single-sex schools.

Male students attending single-sex schools outperform their counterparts in mixed-gender classes by 0.15 of a standard deviation. The significant impact of single-sex schools on male students’ achievement are not driven by classroom gender composition, but largely accounted for by increases in student effort and study-time. We find little evidence that classroom or school gender composition affect the outcomes of female students.

The second, Lu & Anderson (2015) looks more closely, at the effect in Chinese middle schools of being randomly assigned to sit next to others of the same gender. Contrary to the Korean paper, they find that girls do better if they sit near girls, but there is no effect on boys from who they sit near.

We exploit random seat assignment in a Chinese middle school to estimate how the gender of neighboring students affects a student’s academic achievement. We find that being surrounded by five females rather than five males increases a female’s test scores by 0.2–0.3 standard deviations but has no significant effects on a male’s test scores.

The third, Booth et al. (2013) looks at universities in the US, and does a true random experiment, again finding an effect on women but none on men.

We examine the effect of single‐sex classes on the pass rates, grades, and  course choices of students in a coeducational university. We randomly  assign students to all‐female, all‐male, and coed classes and, therefore, get  around the selection issues present in other studies on single‐sex  education. We find that one hour a week of single‐sex education benefits females: females are 7.5% more likely to pass their first year courses and score 10% higher in their required second year classes than their peers attending coeducational classes. We find no effect of single‐sex education on the subsequent probability that a female will take technical classes and there is no effect of single‐sex education for males.

Now these three studies are hardly conclusive—there's a big literature out there. And one of the things we really know well is that improving education is hard. Most of the quality of private schools is in their students, teachers, facilities and parents. But perhaps it's no coincidence that private schools are mostly single sex, and perhaps a system with more parental choice would tend toward a system with different genders in different classes or different schools.

Fiscal austerity might not have hurt growth in the Great Recession

I say 'might', but of course I don't think there's any evidence it did. Anyway, a new NBER paper (gated up to date version, full working paper pdf) from Alberto Alesina, Omar Barbiero, Carlo Favero, Francesco Giavazzi and Matteo Paradisi finds that it did not.

The conventional wisdom is (i) that fiscal austerity was the main culprit for the recessions experienced by many countries, especially in Europe, since 2010 and (ii) that this round of fiscal consolidation was much more costly than past ones.

The contribution of this paper is a clarification of the first point and, if not a clear rejection, at least it raises doubts on the second. In order to obtain these results we construct a new detailed "narrative" data set which documents the actual size and composition of the fiscal plans implemented by several countries in the period 2009-2013. Out of sample simulations, that project output growth conditional only upon the fiscal plans implemented since 2009 do reasonably well in predicting the total output fluctuations of the countries in our sample over the years 2010-13 and are also capable of explaining some of the cross-country heterogeneity in this variable.

Fiscal adjustments based upon cuts in spending appear to have been much less costly, in terms of output losses, than those based upon tax increases. The difference between the two types of adjustment is very large. Our results, however, are mute on the question whether the countries we have studied did the right thing implementing fiscal austerity at the time they did, that is 2009-13.

Finally we examine whether this round of fiscal adjustments, which occurred after a financial and banking crisis, has had different effects on the economy compared to earlier fiscal consolidations carried out in "normal" times. When we test this hypothesis we do not reject the null, although in some cases failure to reject is marginal. In other words, we don't find sufficient evidence to claim that the recent rounds of fiscal adjustment, when compared with those occurred before the crisis, have been especially costly for the economy.

The paper comes with a whole load of interesting charts, showing how much newspapers started talking about austerity in 2010, the evolution of Eurozone fiscal policy, the correspondence between different measures of governments' fiscal policy stances, and how much better cutting spending is as a means of belt tightening than raising taxes (contrary to what Keynesian intermediate macro textbooks tell you).

Interestingly, they don't make much mention of monetary policy whether conventional or unconventional. Of course, I believe that fiscal austerity would hurt growth if monetary policymakers weren't willing and able to steady aggregate demand. But so would a particular large firm, for example, reducing investment if this was also coupled with the central bank deciding to cut its nominal target for some reason.

Interrogating the evidence: women in academia

Lots of women have experienced individual instances of discrimination in academic settings. Women are underrepresented, relative to their half of the population, in some academic fields. Most people naturally conclude that one reason women are underrepresented is either (a) direct discrimination or (b) women being dissuaded from entering due to perceptions of an unwelcoming 'culture of discrimination'. A new paper argues that neither is a plausible explanation in philosophy, one of the fields most heavily criticised for its relative dearth of women. Authors Neven Sesardic and Rafael de Clercq, in the paper "Women in Philosophy: Problems with the Discrimination Hypothesis" present a strong case that there may well be bias in favour of women in the academic philosophy hiring process, with institutions going out of their way to try and find qualified women for positions if they can.

They point to a raft of previous work:

[At the University of Western Ontario 1991-1999] On average: 5.4% of female applicants were appointed compared to 2.9% of male applicants; 21.7% of female applicants were interviewed compared to 15% of male applicants; and 24.9% of female applicants who were interviewed were hired whereas 19.2% of men who were interviewed were appointed. Again, the results in each of the years are remarkably consistent. Women had almost twice the chance of being hired as did men.[9]

Similar results were obtained in a recent comprehensive study commissioned by the U.S. Congress to assess gender differences in the careers of science, engineering, and mathematics faculty—the area with the highest underrepresentation of women.[10] Conducted under the auspices of the National Research Council, Gender Differences at Critical Transitions in the Careers of Science, Engineering, and Mathematics Faculty included two surveys of major research universities, focusing on almost five hundred departments and more than eighteen hundred faculty members.

The authors reported that among those interviewed for tenure-track or tenured positions, the percentage of women interviewed was higher than the percentage of women who applied for those positions, and that tenure-track women in all disciplines received a percentage of first offers that was greater than their overall percentage in the interview pool.[11] The situation was the same with tenured positions in all disciplines except biology.

They attack the anecdotal evidence of discrimination presented in popular debate around the issue, and suggest that rather than discrimination either at the university level or perceptions of discrimination, biology and culture are to blame. Women end up, for whatever reason, with skill sets and preferences that don't favour the hard sciences and philosophy.

They document that existing studies showing bias have been overturned:

After re-examining the analyses, Nature has concluded that "[a study finding gender bias in journal acceptance of article submissions] can no longer be said to offer compelling evidence of a role for gender bias in single-blind peer review. In addition, upon closer examination of the papers listed in PubMed on gender bias and peer review, we cannot findother strong studies that support this claim. Thus, we no longer stand by the statement in the fourth paragraph of the Editorial, that double-blind peer review reduces bias against authors with female first names."[32]

It looks at evidence on research grant applications in the UK:

The authors looked at 1,741 grant applications to the Wellcome Trust and 1,126 grant applications to the Medical Research Council (in the UK). They concluded that “this study has shown no evidence of discrimination against women.”[35]

And across the developed world:

More recently, Ulf Sandström and Martin Hällsten investigated 280 grant applications submitted to the Swedish Medical Research Council in 2004.[36] Their conclusion is that “female principal investigators receive a 10% bonus on scores.”[37] More generally, Ceci and Williams report that “the weight of the evidence overwhelmingly points to a gender-fair review process” in grant funding.[38] Their conclusion is based on a number of smaller studies from different countries (including the abovementioned study by Grant et al.) as well as on six large-scale studies, including one by Herbert W. Marsh et al. that “found no significant gender differences in peer reviews of grant applications.”[39]

There is a huge amount more by way of evidence for their conclusion that neither (a) nor (b) is substantially true. Men and women sometimes want different things, and that's OK.

Ideas can mean the difference between wealth and poverty

Adam Smith never said that “The real tragedy of the poor is the poverty of their aspirations”, as some people who have never read him think. It is hard to think of a less Smithian view – he was the opposite of that quote's patrician and patronising voice, and had a deep compassion for people who had been unlucky in life. But there is some evidence that disadvantaged people underinvest their savings at a huge cost to themselves. This seems to be true even when there are no social constraints or market failures that might cause this to happen.

One reason for this may simply be that poor people do not realise that the investment opportunities exist, or do not really consider that they might benefit from them. Consider those bright young students from deprived backgrounds who have never even considered applying to university, just because nobody in their families ever has either. Your experience of the world shapes how you react to various opportunities that you get.

To test this hypothesis, a group of researchers at Oxford performed a controlled trial in remote Ethiopian villages, where they showed one of several one-hour documentaries about poor Ethiopian farmers who had expanded a business, improved their farming practices or broken cultural norms by, say, marrying for love. “Individuals succeeded largely through their own efforts and by drawing on assistance from community members and available resources, not through outside government or NGO intervention.”

The trial involved a placebo group (shown a comedy movie) and a control group (shown nothing at all) and it seems to have been a success. Six months after the screenings, the documentary group’s savings rate had risen significantly above the control group’s and had also begun to access credit at a higher rate. (These are some of the poorest people in the world, so the absolute amounts – a few pounds – may seem very small to our eyes.)

School enrolment was up by 15 percent in the documentary group, although it was also up by 10 percent in the placebo group so the effect is unclear, and spending on school expenses was up by 17% (compared to no change in the placebo group).

Overall, the results seem to show that showing extremely poor people examples of people like them who had made something of themselves inspired them to invest in themselves and their families.

It’s just one study, but it hints at something bigger. Incentives matter, of course, but you have to be aware of the existence of an incentive for it to work on you. Even if you’re aware of it, you might discount (or exaggerate) its significance according to your experiences. In a complex world, each of us uses a different pair of glasses to focus on what matters and filter out what doesn't. And no pair is perfect.

There is no obvious public policy lesson from any of this, except perhaps that people don’t always react predictably to incentives. Incentives matter – but so do ideas.

Men are not 'over', women are not discriminated against

In what seems to me slightly contradictory, two popular modern memes hold that, firstly, we are experiencing 'the end of men', who are steadily being eclipsed by women in many levels of academia, areas of the economy and so on; and secondly that women are discriminated against in the labour market, which is why they only earn around three quarters as much as men on average per hour. A 2013 paper by Kingsley Browne in the Boston University Law Review challenges both of these claims, arguing that the dominance women enjoy in many areas of society refutes the idea they are discriminated against, but their relative scarcity in other areas shows how men are still not 'over'. He explains this discrepancy between sectors to differences in preferences and characteristics between the sexes, differences he believes are biologically caused.

Common examples of perceived workplace inequality – the “glass ceiling,” the “gender gap” in compensation, and occupational segregation, among others – cannot be well understood if the explanation proffered for their existence is limited exclusively to social causes such as discrimination and sexist socialization.

Males and females have, on average, different sets of talents, tastes, and interests, which cause them to select somewhat different occupations and exhibit somewhat different workplace behaviors. Some of these sex differences have biological roots. Temperamental sex differences are found in competitiveness, dominance seeking, risk taking, and nurturance, with females tending to be more “person oriented” and males more “thing oriented.”

The sexes also differ in a variety of cognitive traits, including various spatial, verbal, mathematical, and mechanical abilities. Although social influences can be important, these social influences operate on (and were in fact created by) sexually dimorphic minds.

As I have written before, even if the very substantial work-related differences between men and women are socially constructed, satisfying their preferences as they are, rather than as an egalitarian might want them to be, makes men and women best off. I have also written how the gender wage gap isn't evidence for firm discrimination between the genders, because it is entirely explained by women's decisions (to take safer, more satisfying jobs, to work lower hours and to take substantial time out of the workforce).

He points out that women have recently come to dominate many high status fields and that most of the gender wage gap is between, not within, professions. Taken together, he argues there is no general 'glass ceiling', although of course there are many individual instances of discrimination.

In many respects, however, women have made breathtaking advances in the past several decades. Professions such as law and medicine are reaching parity among new entrants, and women represent over 60% of newly enrolled pharmacy students and over 75% of new veterinarians.

Even within fields in which men predominate generally, such as science, technology and engineering, there are interesting variations: women are underrepresented in metallurgical and mechanical engineering but much closer to parity in biomedicine and bioengineering.

And he gives strong evidence for differences between men and women in: competitiveness, which are not easily explained by 'stereotype threat' given that they start very early and only appear in specific sorts of high pressure competition; risk-taking (boys and men take more risk); interest in children (girls and women show more of it); spatial reasoning ability (men excel); verbal ability (women excel); and occupational interests.

Whether these are biological or social they massively affect the fields that women want to enter and the ones they can do well in. And this is OK! It makes people happier to do jobs (including work in the home) they want to do and jobs they are good at. It's OK for our labour markets to reflect this—it makes us better off overall.

People respond to incentives

That people respond to incentives is an obvious point but I feel like every reiteration is worth it. One of the clearest examples of where people respond strongly to incentives is retirement. If you raise the retirement age, many people who'd otherwise be eligible continue to work. Since retirement probably increases life satisfaction/happiness and perhaps even health we obviously want it to happen at some point, but since it's also very costly in terms of benefits paid and productive activity not done, we want to be mindful of both costs and benefits.

A new paper in The Review of Economics and Statistics by Kadir Atalay and Garry F. Barrett at the University of Sydney adds to a large literature:

Governments around the world are reforming their social security systems in light of the challenges posed by population aging. We study the 1993 Australian Age Pension reform, which progressively increased the eligibility age for women from 60 to 65 years. We find economically significant responses to the reform. An increase in the eligibility age of one year induced a decline in the probability of retirement by 12 to 19 percentage points. In addition, the reform induced significant program substitution, with increases in enrollment in other social insurance programs, particularly the disability support pension, which effectively functioned as an alternative source of retirement income.

Raising the retirement age for women led to lots more of them working, but also more of them claiming other benefits.

Every single paper I've ever seen on the topic has found a similar result. For example in the UK raising the pension age from 60 to 61 led to 7.3pp more women in employment at age 60 (separate paper with more evidence). In Spain, people with worse health were more responsive to financial incentives. Less generous pension payouts in France (normal retirement rather than disability insurance retirement) meant 14% higher total work hours, on average, between the ages of 55 and 64. Another paper found that pensioners respond to incentives in a different way: if they stand to gain more by waiting before they claim then they are more likely to wait.

The point of all this is not to say that we should pack the elderly off to the workhouse until they're 90, but more to note that incentives matter, against the common claims that the homo economicus model is rarely or never a good approximation for real humans.

Maybe we should like patents

For a long time I was very sceptical of the benefits of patents. For one thing, they seemed to interfere with other types of property—Apple's patents over certain shapes for phones mean Google cannot use its factories, materials, etc. in certain ways. For another, I coincidentally had come across work suggesting their benefits are overstated, including Against Intellectual Monopoly (appropriately available in full online) by Michele Boldrin and David K. Levine.

But three recent papers exploiting a novel source of data have made me reconsider, since they throw cold water on one of the popular alternatives to patents: innovation prizes. All three are by B. Zorina Khan, an economics Professor at Bowdoin College and fellow at Stanford University's august Hoover Institution.

The first looks at annual industrial fairs in 19th century Massachusetts and finds that (relative to patents awarded over the same period) prizes were mainly used for advertising purposes, were awarded unsystematically and unpredictably, and did not vary in line with how useful or popular an invention or innovation ended up being. What's more, prizewinners were typically from a more privileged class than patentees. This all implies that patents are more market driven and better at incentivising creative innovation, Khan says.

The second looks at similar data (American Institute of New York annual fairs) from a different angle. One argument against patents is that they limit what others can do on top of a given innovation, or how much they can be inspired by a particular breakthrough, because they might have to license the patent or risk infringing it. One argument in the other direction is that patents allow people to bring their information out into the open because others will not use it to jump ahead, so it encourages openness. What's more, all of their info surrounding the innovation is written down and easy to find.

Khan finds that the second effect predominates; patents encourage 'spillover' innovation more than prizes:

In keeping with the contract theory of patents, the procedure identifies high and statistically significant spatial autocorrelation in the sample of inventions that were patented, indicating the prevalence of geographical spillovers. By contrast, prize innovations were much less likely to be spatially dependent. The second part of the paper investigates whether unpatented innovations in a county were affected by patenting in contiguous or adjacent counties, and the analysis indicates that such spatial effects were large and significant. These results are consistent with the argument that patents enhance the diffusion of information for both patented and unpatented innovations, whereas prizes are less effective in generating external benefits from knowledge spillovers. I hypothesize that the difference partly owes to the design of patent institutions, which explicitly incorporate mechanisms for systematic recording, access, and dispersion of technical information.

Finally, her 2013 paper “Trolls and Other Patent Inventions: Economic History and the Patent Controversy in the Twenty-First Century” (the argument is given in less length in a Cato Unbound essay) argues that if you take a long historical view, current patent controversies around non-practising entities, patent thickets, litigation and so on are not new. She says they are part of a well-functioning and successful intellectual property system.

So maybe we should like patents. After all, we support regular property rights because the institution has been proven to lead to a wealthy, successful society, even if messed with substantially. If patents are the best tool we have for generating innovation—a key ingredient of continued social progress—then we should support them too.