Innovation Policy and the Economy, Volume 11

number of such articles, then the fraction of extant knowledge known by an individual would
decline at the same rate: -5.5% per year.²

Figure 1: Journal Article Publications per Year, Worldwide

Notes: This figure presents the number of publications in each year, worldwide, as recorded by the Institute of
Scientific Information’s Web of Science database, pooling articles across all fields of science and engineering and
social sciences. Growth rates in publications are similar looking only at authors with U.S. addresses. Over 90% of
the articles are, consistently, in science and engineering fields. See text for further discussion.

Below we will examine richer and more systematic evidence about the implications of
such expansions of knowledge. But it should be clear at this point that innovators face a shifting
landscape in which they become educated and produce new ideas. In fact, one may expect two
natural responses in innovators’ educational decisions as the volume of knowledge expands:

1. First, innovators may spend longer in education;

2. Second, innovators may seek narrower expertise.

² That is, let N be the total number of papers (or other codified ideas) in the world and let this number grow at rate
gN. Let Q be the fixed number of papers that an individual has time to learn. Then the share of extant knowledge
known by the individual is s = Q/N, and the growth rate of s is then gs = - gN.

<<   1   2   3   4   5   6   >>

More intriguing information