This paper looks, using patents at how scientific resesarch is used in firms. Here are some points I took away
1. R versus D
Do universities do all the "basic" research? No. "
In 2017, the business sector funded about $85 billion in basic and applied research, accounting for about 22% of R&D funded by business, and 43% of all research in the United States".
Figure 1 gives a sense of this
" Figure 1, which is based on National Science Foundation data, shows that the share of basic and applied research in the total domestic R&D funded and performed by corporations, the \R" of R&D, in the United States has declined from over 31 percent in 1985 to about 20 percent in 2008 and has remained at that level thereafter. The share of research in total R&D performed by business shows a similar, albeit less dramatic, decline, from a peak of around 30 percent in 1991 to about 20 percent in 2008, and rising modestly thereafter. These trends suggest that the composition of corporate R&D is shifting away from \R" and towards \D".
And figure 2 shows some more:
"Many corporate labs were shut down or were oriented towards more applied activities. Bell Labs was separated from its parent company AT&T and placed under Lucent in 1996; Xerox PARC was spun o into a separate company in 2002, and Du Pont closed its Central Research and Developmet Laboratory in 2016.33 These trends are also re
ected in Figure 2, which presents trends in the annual number of publications (\R") and patent (\D") divided by sales, for our sample firms. The corporate publication rate fell by about 60% over the sample period, whereas patenting rates do not show any clear trend. The pattern suggests that the composition of corporate R&D is changing over time, with less \R" and more \D""
2. What about spillovers? Well, corporate scientists write papers and so disclose results. Isn't that a receipe for disaster, firms are giving away their own discoveries? maybe not. Perhaps the performing of basic scientific research enables firms to absorb research better, or provide careers for top scientists etc
And it looks like more knowledge is spilling over:
3. At the same time, another paper by the authors has documented that firms are doing less research than they were. So:
In this paper, we argue that private returns to corporate research depend on the balance between two opposing forces: the benefits from the use of science in own inventions, and costs of spillovers to rivals. ...increases in spillovers relative to internal use may be a proximate cause of the decline in corporate research that the literature has documented"
4. They match together US companies who spend on R&D and had and at least one patent over the period 1980-2015 (an unbalanced panel of 3,807 rms). (This is a major task since patents and companies can change ownership). They
We measure the use
of internal research in invention by citations made by the firm's patents to its own scientifc publications. Spillovers are measured by citations from the patents of rivals to the focal firm's publications.
5. what do they find?
First, we show that there is a positive relationship between the market value of a firm and its stock of scientic output. This relationship is stronger when the firm's patents use the science that the firm's
scientists produce.
Second, and consistent with this, we find that a firm produces
more scientific publications if it is more likely to use the science in its patents, but produces fewer publications if the science is more likely to be used by rivals' inventions.
6. This raises a series of questions.
a. other work finds that "ideas are getting harder to find". Perhaps this is because firms are doing less R, and even if universities are doing it, they might not be as good.
b. but then the question is why are firms doing less R? this paper would say that doing less will hurt them since they cannot benefit from their own research, but doing less helps them if more of it is spilling over to others.
The paper offers some reasons:
1. product market competition has become more intensive? No, if any thing the other way?
2. IP has got tighter? Maybe in the 1980s, but not recently. And anyway, life sciences, where patents are the tightest have not declined as much as other sectors.
