continue to see growing patenting efforts and commercial innovation in equilibrium even as
education duration rises and credit is diffused through teamwork.12
The “natural” compensation of market size is, however, much less clear for basic
research. Unlike patents, the commercial value of a basic research idea (which may typically be
zero) does not obviously scale with world GDP, even though the potential value from the idea’s
downstream spillovers does scale with world GDP. Hence, while the motive to encourage basic
research remains extremely strong - and grows -- patent law does not easily transmit this
commercial value into basic science. It remains for other institutional forms of support, through
agencies like the NIH, NSF, government labs and public universities, to confront the life-cycle
challenges and encourage entry into basic science.
B. Policy Mechanisms
To encourage entry into science, one may either (i) increase the value, V, of the scientific
career, or (ii) speed up training, to bring V earlier in the life-cycle. This section will consider
policy mechanisms that can influence both dimensions.
The value V to being a scientist likely has several important components, including
wages, status, and creative freedoms (see, e.g., Stern 2004). To increase V, one could therefore
consider several targets. Wages can be increased most obviously through public support of
researchers, either in public universities, government labs, or the salary components of research
12 In practice, productivity growth, resulting in per-capita income growth, will enhance the market size for ideas but
also increase wages in other careers, conveying no innate bias toward innovative careers. However, population
growth expands the market size for ideas without affecting wages in other careers directly (according to standard
neoclassical growth theory where the aggregate production function is constant returns to scale). So population
growth, unlike per-capita income growth, will tend to asymmetrically advantage commercial innovative careers over
other careers. We can therefore consider the following back-of-the-envelope calculation. World population growth
has averaged approximately 1.5% per year since 1975. Meanwhile, average team size in patenting has risen by
1.1% per year since 1975, suggesting that the individual share of the patents’ commercial value falls by about 1.1%
per year. The rise in training, at about .08 years annually and with a 10% discount rate, reduces the relative value of
an innovative career by 0.8% per year. Thus, assuming that the rate and size of ideas is fixed once the individual is
educated, the personal value of the innovative career would be increasing by 1.5% - 1.1% - 0.8% = -0.4% per year.
Thus population growth may compensate substantially for the educational challenges in commercially-oriented
innovative careers. If we consider a weighted population growth measure, which incorporates rapid increases in
relevant technology buyers in China, India, etc, then the compensation will be higher.
20