Inventors’ Two Cents

Even in societies in which markets were relatively free and developed, there was rarely any proportionality between the contribution of an innovator and the rewards he or she reaped.  At least in that sense, the situation was not different from what it is today: Nordhaus (2004) has estimated that in modern America only 2.2 percent of the surplus of an invention is captured by the inventor him/herself.  Things surely looked no better in the eighteenth century. … If ever there was a divergence between social and private net benefits, the Industrial Revolution was it.  The impact of the technological elite on the rest of the economy was thus vastly larger than proportional to their size.

That is from Joel Mokyr’s The Enlightened Economy: An Economic History of Britain 1700-1850 (p.88).  Not sure I believe the fraction is as low as 2%, though I expect it is well below 1.  We need better ways to encourage innovation!

Mokyr gave a talk at GMU yesterday, which I found disappointing.  His book convinces me that Britain’s exceptionally skilled workforce and high farm productivity were important enablers of the industrial revolution, but his claim that this was due to “enlightenment ideology” seems to me too vague to evaluate.

GD Star Rating
Tagged as: ,
Trackback URL:
  • In his interview with Kling & Shulz, in From Poverty to Prosperity, Mokyr mentions that he didn’t trust his original back-of-the-envelope calculations, so he suggested a much more conservative number, like 20 percent, but eventually the evidence became overwhelming, and he dialed it back down even lower than his original estimate.

  • Josh

    Did he factor in re-invention? If twenty people independently invent the same thing and each one gets 2% of the surplus, collectively the inventors get 40%.

    Also, 2% is not necessarily a poor reward. If an invention improves each human’s wealth by 1/1000 of 1 penny, multiply that by 6 billion and that’s still a very nice reward for a very slight improvement.

    • Exactly — multiple inventors throw kinks into the calculations. For example, you need to subtract off all of the social benefit that occurs when the next-brightest person would have otherwise invented it, when determining the inventor’s contribution to social wealth. (Of course, this conflicts with the intuition that, in cases of multiple simultaneous discovery, you divide the social benefit by the number of inventors to get the implicit per-inventor value. Then again, simultaneity implies obviousness…)

      Of course, I’m just an armchair amateur here who didn’t read the papers, and am obviously arrogant for assuming they wouldn’t account for such an obvious consideration. Has anyone read them and verified they thought of this? Robin?

  • MPS

    What should be the percentage?

    When someone has an “original” idea, that idea is based on the assimilation of many ideas that are already well known, and combines with many other existing ideas to form the actual innovation. I don’t know how to do a quantitative assessment of the relative sizes of the contributions, but I doubt anyone’s “original” idea composes more than 2% of the sum of knowledge needed to produce the final innovation.

    Also, it’s not obvious to me that the right way to give incentive for innovation is to give huge rewards to the innovators. Discovery is a risky process: one can toil for decades and not discover anything useful, or can get lucky and make a useful discovery very quickly. This by itself might indicate that the rewards of innovation should be huge, to motivate people to take that risk. But there’s another way it could play out. Institutions might recognize this feature, and employ many potential innovators, sharing the rewards of innovation and thereby spreading the risk of financing it. This might be the more “efficient” system.

    And it seems to me, this is largely how our economy works. That is, most people devoting the majority of their time to trying to develop innovations do so for some large institution, which pays them according to measures of productivity that relate more toward their potential to develop innovations, and not whether they succeeded or not. As extra incentive there are bonuses for success, but mostly the financial rewards are reaped by the funding institution, and pay, in part, for funding all the potential innovators who failed. People who recognize the high degree of “luck” in the whole process are happy with this arrangement, and I suspect in the end it generates more innovation.

    • Jess Riedel

      Nobody is suggesting that we stop spreading research risk with institutions. The problem is that if the owner of a patent (institution or not) is only capturing 2% of the value of the invention, there will be an underinvestment in research

      • Manufacturers and distributors also deserve to get paid. Is 2% an unreasonable cut? What *should* the figure be?

  • Arguing that the first person to come up with an idea deserves 100% of the surplus seems the same as arguing that people who improve on it further deserve 0%. I suspect that such a world would feature much less innovation than ours.

    • Yes, and if inventors always captured 100% of the surplus, then why exactly would we *care* if stuff gets invented? (Beyond our usual concern for any given person, of course.)

      • Nec

        Perhaps you care if you want that innovation to exist, if your life would be extremely impoverished without that innovation, say, a computer connected to a global network (if it didn’t already exist).

  • mgravity

    What about encouraging people who are particularly skilled innovators to become serial inventors? You can do very well at 2% surplus capture per invention with a lot of small inventions or improvements. If the payout was substantially larger, you might find too many one and done inventors, who make a fortune on one invention and then stop.

  • Jess Riedel

    How would one determine the optimal percentage?

  • josh

    How does the author’s definition of innovation correspond to the actual events that increase our PPF or social welfare or whatever? Perhaps we just misidentify which innovations matter.

  • I am not so sure of this:

    We need better ways to encourage innovation!

    Because I am not sure that more motivation will produce more innovation. Just bragging rights are pretty motivating.

  • Mokyr also offered the difference between the British guild system and guilds everywhere else. Since so much of the talk focused on the failures of enlightenment ideology in agriculture, I had trouble interpreting the central claim.

  • Grant

    Most innovation seems to be limited by other supporting technologies. With the exception of very simple things like sticky notes or nylon tie-wraps, I can’t think of many inventions which were mostly dependent on the inventor. I’m sure their contribution is greater than 2% but it still often seems pretty low.

  • Popeye

    We need better ways to encourage innovation!

    The problem is that if the owner of a patent (institution or not) is only capturing 2% of the value of the invention, there will be an underinvestment in research

    If innovation is driven by investment in research, and if investment in research is based on the % of value captured by inventors, and if we live in a nice world with seemingly lots of innovation… the conclusion may simply be that one of our assumptions is wrong.

  • Pingback: Tweets that mention Overcoming Bias : Inventors’ Two Cents --

  • Kip

    Why do we need better ways to encourage innovation?

    The entire IP system in the US is based on the idea that there isn’t enough innovation. But I question that premise. I am not sure what motivates it. All of the methods of encouraging more innovation, like IP rights, come with their own costs. It’s not clear that the alleged benefits outweigh the costs either.

    In short, it’s not clear that a free market, without IP rights, wouldn’t provide a near-optimal amount of innovation.

    I think the real motivation behind IP rights is that people have a gut-level disgust reaction whenever someone uses an invention that someone else invented. The worry is that the second person is a freerider.

    Of course, he is a free rider. Information is hard to keep contained. That’s not the freerider’s fault. We should be skeptical of attempts to build elaborate structures and systems to make information artificially contained. And we should be skeptical of those who say, because of that disgust reaction, that we need to artificially induce more innovation.

  • Microbiologist

    > It’s not clear that the alleged benefits [of IP] outweigh the costs either.

    How would we possibly measure? Cross-national or cross-temporal comparisons would assume, explicitly or not, that other variables didn’t confound our comparison. I would be sufficiently worried about confounds that I would actually consider a deductive approach to provide better evidence.

    • Kip

      There’s no real way to measure the benefits.

      On a related note, patents are often valued as “?” by economists.

  • Greg Clark says pretty much the same thing in “A Farwell to Alms”. His point is that we can’t attribute the Industrial Revolution to some great shift in recognition of property rights that allowed investors in technology to get rich, since inventors reaped so little gains. Rather, unskilled laborers did.

  • Negentropian

    Humanity gets the benefits of the invention, the inventor gets the money, every last trillionth of a penny of profit. It’s a fair win-win deal.

  • Nec

    The less physical matter is involved, the bigger percentage the inventor gets. Selling your software or music directly via web can yield close to 100%.