Thread by @mattsclancy

Where do new ideas and technologies come from? One school of thought says they are born from novel combinations of pre-existing ideas. It’s an appealing notion, but does this perspective get us anywhere useful? #ThursdayThreads

Let’s look at a few papers.

One nice thing about the combinatorial perspective on innovation, is that if you know all the possible technological building blocks you can exhaustively enumerate all possible inventions, including ones that are never actually invented.

For example, once components A, B, and C, exist, then the combination ABC is a possible invention. Then you can see what kinds of factors are correlated with possible inventions being realized.

In Clancy (2017) (yes, that’s me), I used the 13,000+ “mainline subclasses” in the US patent classification system as proxies for the underlying technological “components” of patented innovations. Examples of mainline subclasses: "bridge; truss", "AI (neural network)"

When a patent gets assigned more than one, I interpret it as an invention combining these technological components.

I show a patent with a given combination ABC is more likely to be filed in a year when there are more patents combining any of the pairs AB, AC, and BC, but less likely to be filed when there are more patents combining all three (ABC).

These figures illustrate these dynamics (the different lines correspond to different values for other explanatory variables). Left is expected patents as a function of how often pairs of components are used, right is the same for triple-combinations.

The implication is that innovation enables neighboring ideas that draw on similar but not exactly the same components.

A 2nd nice thing about the combinatorial perspective is it suggests a natural way to measure the novelty and creativity of innovations. A classic paper here is Fleming (2001), which also uses patent subclasses to proxy for combining technologies. funginstitute.berkeley.edu/wp-content/upl…

For a sample of patents granted in 1990 he calculates the number of prior patents assigned the exact same set of subclasses. He shows patents assigned combinations without much precedent tend to receive more citations, but with more variance.

This is sort of what we would expect: compared to incremental improvements, novel innovations have a higher chance of being hits or flops.

Others have applied a similar idea to academic papers. Wang, Veugelers, and Stephan (2017) define highly novel papers as those whose citations “combine” journals not previously connected (weighted by how distant the journals are). sciencedirect.com/science/articl…

As with Fleming, they find papers making unusual connections have a higher variance of citations received, but tend to be more highly cited in the long run.

However, they also find evidence of bias against novelty in science: papers making highly novel connections are less likely to appear in high-impact journals, receive a higher share of citations from outside their own field, and citations they do receive take longer to arrive.

Uzzi, Mukherjee, Stringer, and Jones (2013) also looks at papers that make atypical connections between cited journals. science.sciencemag.org/content/342/61…

They also find some ambivalent results about about novelty. The papers most likely to become “hits” (top 5% most highly cited) are those that make a small number of atypical connections but are otherwise highly conventional.

Papers that are novel across the board (“low median convention” and “high tail novelty” in the figure below) don’t fare nearly as well.

So the observation that innovation is about combining novel ideas isn’t “true-but-useless.” If you have ways of measuring the knowledge building blocks that patents and papers embody, a combinatorial perspective can tell us something interesting we didn’t already know.

PS - If you liked this thread, I write this kind of thing on Thursdays. You can also subscribe to a newsletter (mattsclancy.substack.com) or RSS version (mattsclancy.substack.com/feed/).

Today's thread delayed by travel. Hoping to meet some of you this weekend at the ASSA!