The excellent Kevin A. Bryan on innovation

He has a forthcoming JET paper with Jorge Lemus, here is the abstract:

We construct a tractable general model of the direction of innovation. Competition leads firms to pursue inefficient research lines, because firms both race toward easy projects and do not fully appropriate the value of their inventions. This dual distortion will imply that any directionally efficient policy must condition on the properties of hypothetical inventions which are not discovered in equilibrium, hence common R&D policies like patents and prizes generate suboptimal innovation direction and may even generate lower welfare than laissez faire. We apply this theory to radical versus incremental innovation, patent pools, and the effect of trade on R&D.

Here is a slightly different version of the abstract, along with other research papers.  Here is Kevin on travel.  Kevin is at the University of Toronto, and also is author of the excellent blog A Fine Theorem.

Here is Kevin’s reading list on innovation, recommended.


How innovation happens and how to make it more effective is one of the most important questions for humanity not just economics. But I bet there will be a lot more comments on the Celebrity Misbehaviour post above.

Competition produces many mini-innovations, which all add up to a big innovation, while monopoly produces neither many mini-innovations nor another big innovation as the monopolist protects the big innovation that made it a monopolist. I believe it's Tim Taylor (following the research of Edmund Phelps) who has written on the role of dynamism in the workplace in producing relatively small innovations (such as Eli Whitney's improvements to the cotton gin) that add up to a big innovation and serve as real drivers of economic growth. Today's pursuit of the big innovation (which I sarcastically refer to as flying cars and spaceships to Mars) by billionaires are vanity projects that impress a certain class of people who are more familiar with comic books and science fiction than with actual production of goods.

Innovation is clustered in big cities and so is economic growth that flows from it. Look at the map at the link of how states rank in terms of disruption from globalization. Amazing.

I got part way through it and the implicit assumptions of true "ability to view the future" limit its value to trivial or obvious inventions. Real innovation is more like evolution, it can't plan ahead to prevent blocking and dead ends of non-optimal solutions. The future is opaque beyond the laws of physic and thermodynamics and that lack of future vision means non-optimal outcomes for both human development and evolution.

These models in this and similar papers imply that all problems can just be solved with more R&D, which is only true for trivial problems. However, internal combustion engines won the automotive power supply technology with high useful energy density and easier energy distribution. The "rocket scientists" of the time would put this in the trivial class where minimal future vision is obvious. It was also obvious that solid state physics would kill the vacuum tube industry.

For an example of this of how this thinking goes wrong, just look at how much money the DOD and DOE "wise men" have wasted on trying to get biofuel from algae. Lots of R&D money down the rat hole and that was trying again for something the same "wise men" tried in the 70's.

Started to read the paper but found some of the assumptions funny, Eg, assuming that B is a better innovation than A, but that once A is discovered the marginal value of discovering B is zero. I guess you could imagine such a situation but it does not seem common.

Comments for this post are closed