Innovation Policy and the Economy, Volume 11

is, using a discount rate of 10%, it is generally true that a person will value one dollar next year
at only 90 cents today. By analogy, a wage stream must rise by 10% to compensate the
individual for an extra year of schooling. That this reasoning can apply usefully to labor markets
is demonstrated by the fact that average wage returns to an extra year of schooling are about
10%, a relationship that holds in a fairly stable fashion across time and across countries (Mincer
1958, Psacharopoulos 2004).^9,10

In science, this problem can lead to two straightforward selection effects. First, there is
selection across types of scientific careers. While certain areas of science, such as
biotechnology, have become increasingly deep, with lengthening Ph.D.’s and the development of
post-doctoral phases, other areas of innovation have milder training commitments. Perhaps the
most spectacular example in recent years surrounds the “dot.com” boom. In many instances,
important innovative ideas, including retail concepts (e.g. book seller + internet), internet search,
and HTML software applications required relatively little technical training at first. The relative
ease of entry into such innovative careers, other things equal, will tend to draw entrants away
from sciences that feature long and extending training phases. Note, however, that this form of
selection is not necessarily a concern for scientific and economic progress. Indeed, diverting
talent and effort from a harder area of innovation to a less costly but possibly equally fruitful
area of innovation may well be efficient.¹¹

The second kind of selection effect occurs when talented individuals avoid science
entirely. For example, if careers in finance, management, or law require more static levels of
training, then scientific careers will be increasingly costly by comparison. The estimated 6-8

⁹ The 10% benchmark is true for richer countries. Returns to education tend to be somewhat higher in poorer
countries, which is consistent with higher discount rates in poorer countries, as reflected in higher interest rates in
poor countries.

¹⁰ The 10% discount rate may not apply perfectly to scientists, who may, for example, particularly enjoy learning or
may be especially attracted to non-pecuniary benefits (see, e.g., Stern 2004). Nevertheless, standard discounting
likely applies to wage aspects of the scientist’s career decision and presumably the individual would, other things
equal, rather not delay other benefits as well - whether social status or the joys of creativity and discovery.

¹¹ Some evidence for this selection effect appears in Table 1. When adding field fixed effects, the age trends rise.
This means that the increasing delay is actually higher within individual fields, but that scholars appear to be shifting
over time to those fields where great achievements can be had at younger ages.

18

<<   16   17   18   19   20   21   22   >>

More intriguing information