Session 2: Why do companies patent?
Discussion leaders: Lisa Ouellette: Accounts indicate that non-incentive reasons drive much patenting, contrary to standard incentive story. What do sociologists etc. think about this divide in accounts? Sociology/psychology give us tools when economic rationality (or what the economists say rationality is) falls short. How do those fields feel about our cooptation, mostly from a law & econ perspective?
Economic rationality of individual firm choice is different from whether patenting is utility- or welfare-maximizing. Evidence of whether they’re needed for commercialization is shaky, but that’s different from the question of why firms patent.
Burk on sociology of patenting: he says it’s irrational b/c most patents are unenforced unlicensed and ignored. [Quoting Lemley saying that, Burk points out.] But that’s like saying buying options is irrational b/c most options don’t pan out. The total rents from patents are larger than the total costs; similarly true for university patent licensing income v. costs overall, even w/a lot of losers.
Dan Burk: most of our discussion is oriented to individuals b/c at the end of the day it’s natural people who invent things, as patent though not © recognizes. At the same time, we know this happens in organizational contexts. Interested in how often the phrase “tell a story” came up. There are certain themes, narratives, stories that lend cohesion/coherence to social activities—trying to expand the stories beyond economic ones that are common in our field.
Concepts of efficiency and welfare and preferences are socially constructed—economics doesn’t talk about how preferences are constructed or where they come from, and that is exactly what we are concerned with. Small level differences in reactions—Brownian motion—engineers and managers may have different reactions to changes. Network models are another approach. Theory of the firm appears outdated to many sociologists; networks instead. Formal legal structure of the firm may not be relevant. Some organizations are within, others outside the firm. We have surveys (which may be methodologically flawed by others’ standards), but we don’t have fieldwork where you embed over months/years. What people tell you they do and what they do may not be the same. How do we add that to the kit? Sociologist may spend 7 years studying chefs, but most of us don’t have the training to know what to do and tenure etc. norms mean that we can’t do that and hope to thrive in our careers.
Commentators: Tim Holbrook: Race/gender play big roles. W/in the institutions, patents are currency: example, wall of names at business showing who had patents—didn’t care what the patent was for or how much money it made to get on the wall. Externally it may not look rational but has internal productivity payoff. Norm-setting in institutions. Patents may also have more salience in helping develop brands, goodwill in today’s society. Global question: clearly empirical evidence suggests that patents don’t operate in the traditional ways to incentivize invention. Has that always been true?
Peter Lee: Economic reasons for patenting: can be socially rational to follow myths, and even economically rational: if you don’t follow these myths, market may punish you. Irrational exuberance of markets; shareholders have certain expectations for what companies do and reward/punish. Narrative of patenting as semantically equivalent to patenting is important—if used as innovation proxy, may be misleading. Smaller companies and larger may differ in reasons: need to accumulate capital/send signals may be more important for smaller. Smaller cos. often tend to be more innovative; patent or perish may be the dynamic. Stage of company also matters. 20 inventers at MakerFair: those who just started out, excited about idea, didn’t know/care much about patents/IP. More sophisticated entrepeneurs, those seeking financing, claimed savvy/ were seeking patent protection.
Different types of companies, here including universities. Sociological evolution of university from anti-commercial to remarkable shift to seeing patents as part of own institutional mission. Scientists encourage tech transfer offices to patent, seek out licensees, go to work for them.
Lemley: Patent wall is important—you see it for individuals, at companies, in marketing literature—people value it in some way. Do they value it in a way IP law ought to care about? Holbrook called it an incentive to generate ideas, and that might be true, but it’s also possible that it’s an incentive to turn uncodified knowledge you’d have anyway over to the patent office, with unclear social benefits/costs. Maybe it’s just triggering the segmenting of things that otherwise wouldn’t have been worth patenting or would otherwise have been put together into one large entity, which could be a bad thing for society. A lot of what we want here is a metric for innovation and we suck at measuring it. The number of economists who make careers by using patent counts as a proxy for innovation suggests that this is really a usable metric even by people who are fairly sophisticated and probably ought to know better. The desire for such a metric by scholars and companies may be driving us to valorize patents.
Patents are also a trading card. They are our tulips. Allow us to structure a market for something difficult, which is know how. The patent is carrying with it the knowledge/know-how that allows companies to work together. That could be extremely valuable, but he then wonders what the law ought to be doing.
Silbey: Patents as brands/signalling devices, and that’s absolutely one way patents function. Patents as cultural objects whose meaning has shifted over time; we need to map that shift in function as semiotic object. That will help us understand what rights we want to confer. Inevitably, you interview different people in companies w/different perspectives and goals: you find tensions and commonalities w/in the institution. Easier to think about how the companies use patents as a sign of why they might be collecting them. Andrew Currah, J. Econ. Geog., on how Hollywood collects © as a signal about what it is. A model for studying institutional motivations for IP.
Buccafusco: Ask why don’t firms patent, or what firms do instead of patenting, as well. Should we assume patentable inventions? There’s stuff that matters that isn’t patentable.
Jim Bessen: Economists model innovation, not inventions. This leads to problems talking to you guys. Value of patents, estimating: one method is looking at whether they’re willing to pay the renewal fee, assuming they’re rational about that. We don’t care about the source of that value. Value of patents is relatively small part of value of innovation. Knowledge development happens in lots of ways. Value of patents resides largely in small number of most valuable patents. Legal system is looking at the tip of the iceberg that is a tip of another iceberg, and that’s where the money is, but we have to think more broadly.
Dan Burk: We’ve been using incentive as shorthand for “incentive to invest in economically rational way,” but social science will always talk about motivation, not just return on investment but other motivations. Note that suffragettes used patenting as evidence that women should have the vote—they set up a booth at the World’s Fair. May not have anything to do w/rationality and incentives in the usual sense, but still a motivation.
Jeanne Fromer: Women and patenting: the way that work is assigned in companies may mean women get things that are harder to patent. It’s not the Q why companies patent, but how companies are organized and how we assign work to people. Women who can’t get patents are not going up on that wall of praise.
McKenna: note also the me-tooism. Universities patent b/c it’s the way to be a serious research university these days, although they are also delusional about how much it’s likely to pay off. But mostly it’s a signal about seriousness rather than a rational strategy to pay off through licensing. Also part of a larger shift about how we think about and fund universities, driven by trustees who think universities should be a business.
Rob MacCoun: Interesting shift from morning’s focus on images/accounts of human nature to domain in which economics is a variable.
[sorry, dazed from lack of sleep]
Silbey: myth of self-interest. Companies want to survive, but what does it mean to survive? Lots of startups, their goal is to be acquired. University’s definition: survival depends on reputational cachet. Self-interest variable needs to be better delineated to understand how patents function. EpiPen: learning from profound failures in innovations in the patent space—why we don’t have a competitive market in the EpiPen and the role of patents in that failure—is a rich space for study.
Sprigman: Signals and marketing tools: there are always other ways to send a signal. When we can’t compete on houses we compete on the schools we send our kids to. The question is the relevant costs and benefits of having these particular signals, if that’s what patents are doing.
Bessen: The value of a signal is related to the value of the cost of giving the signal, and literature on startups suggests that patents may not be that valuable.
Sprigman: law & econ: total social welfare criterion. In business school, they talk about producerist welfare. It might be startups are getting funded b/c of patents, but maybe they’re crappy and social welfare does down. Second-best for producers might be first-best for society, even if VCs have to do work and live with people and figure out what they’re doing, instead of giving Elizabeth Holmes a billion dollars for Theranos.
Burk: different kinds of signaling: firm value, managerial competence. At other times, we talk about signaling legitimacy, social players, American innovativeness, etc. These might overlap but don’t always. Women scientists patent at a lower rate than male counterparts; why? Ethnography: women are socialized not to think in these terms. When a male engineer comes up with something, he’s happy and thinks he might be able to make money. A woman in the same situation will be happy but is less likely to think that she should get out and pitch it; more likely to be excluded from advisory boards, committees, etc. where she might make contacts to use the patent to commercialize. Even if you think patents are traditional econ incentive, it’s swamped by other social structures/signals that are preventing what we want to happen from happening. There’s virtually no evidence on race. One piece of work: Af-Am inventors used to patent at about the same rate as the general population; dropped significantly in the 20th century due to a series of lynchings.
Nagaraj: New study tracking HS students in NY, and there are some differences by race, holding ability constant, in whether they end up in patent literature.
Sunder: design patent turn—it’s about your image as an innovator. It’s less about utility and more about creating new experiences and lifestyles and image. Shaky theoretical foundations for design patent; this conversation is making me more nostalgic for utility patents where we focused on the quid pro quo and whether the benefit outweighed the cost. Design patent calculus is much harder to cabin.
Stephanie Bair: Our ability as humans to rationalize: split-brain patients make up reasons for what they do that don’t track what actually happened. Same thing happens with our explanations for why we have patents: we have them, so there must be a good reasons. If we switched to another system, people would line up to explain it too.
Burk: Women tend to get the crummy jobs worldwide, so where scientists are ill-paid (like Spain) they may be relatively more likely to be scientists and get patents.
Lemley: VCs want patents is self-limiting: once people know that VCs want patents, it gets manipulated as a signal. But if it’s a bargaining chip that lets people participate in the market, then it may retain force.
Silbey: in my work, patent as “a chit” comes up a lot, VCs and startups alike. Tells us something about power dynamics, patents as power, how they can be manipulated.
Greg Mandel: Lay attitudes about IP; trying to get understandings of justifications—they tend to think it’s about anti-plagiarism, anti-copying. Are they searching for a moral justification to hang this on, as with the other instances of retroactive rationalization? Sometimes we give them a scenario and ask if it’s ok w/o mentioning IP; what comes out in those scenarios is a concern about attribution, so that seems to be prior to concepts of IP for laypeople. Strong internal concern about taking credit for other people’s works underlying that.
Burk: Culturally constructed?
Merges: you get the same results with little kids in all cultures.
Silbey: it’s not that sharing is bad, it’s that taking credit for others’ work is bad. [I expect the cultural construction comes in on the question of what counts as “work.” See, e.g., Joanna Russ’s How to Suppress Women’s Writing.]
Holbrook: Do we think the law should just be reflective of views?
Lemley: almost never a good idea.
Silbey: how law is involved in institutions incorporating IP: engineering curriculums have been changing dramatically to incorporate more joint appointments w/law profs and with computer science. They are IP/privacy/data security professors. Engineering is institutionally becoming infused w/the importance of some fields and not others. Other schools, trying to stand out, are trying to infuse ethics/humanities. MIT is on a huge curricular reform to get engineers to study more literature and social science. When law is taught in engineering, starts reaffirming the value of that law. Engineers start valuing particular signals of what it means to be excellent in the field; that’s a way that law as a system is functioning to affect innovation spaces.
Suchman: but engineers often get a black letter, these are the rules approach, which further reifies the law.
Lemley: disconnect b/t reasons companies patent and what happens afterwards may have policy implication. B/c reasons for patenting are largely disconnected from legal rights that result, we have flexibility in designing our legal rules in ways that won’t significantly impact corporate decisions. Insensitivity of patent applications/issuance to even dramatic looking changes in substantive law, both in 80s-2000s to greater protection, and to lesser protection in recent years. Hard to tell whether to be happy or sad that the system will merrily chug along regardless of what we do. If we thought w/in a certain range we won’t actually disrupt reasons for patenting or uses of patents as chits, we could save a lot of social cost on the back litigation end w/o damage to ecosystem as a whole.
Bessen: Is it the upper tail driving the litigation? If most patents are used defensively, that may be immune to substantive changes, but other areas, small number of patents we care a lot about, may not.
[redeye daze intensifies]
Merges: we know there are some negative effects in introducing patents into a patent-free space—dissemination goes down, but maybe net dissemination doesn’t decrease. There are other reasons to disseminate. Tends not to think in terms of oppositions, but about how IP fits into the larger scheme.