Industrial policy has become a bipartisan issue, though a sizable part of the electorate favors government ownership only when its favored party is in power. This year, a CNBC poll found that 66 percent of Democrats think it is not appropriate for the government to own companies, while just 34 percent of Republicans said the same.
The National Science Foundation (NSF) Regional Innovation Engines, established by the 2022 CHIPS Act, aimed to attract innovation investment to different parts of the country. Most Republicans opposed these types of measures when Joe Biden was pushing them. No policy created the world’s largest tech hub. The agglomeration of human capital is a process that happens serendipitously. Workers move where employers are, and employers move where skilled workers are. The process that creates agglomeration economies makes it hard to build them elsewhere once they are established, and governments in these places thus find it easier to tax and regulate companies and workers once they are stuck there.
Finally, a reason to check your email.
Sign up for our free newsletter today.
If innovation hubs are created by a few brilliant first movers who attract others to follow them, they cannot simply be built from scratch. This can happen organically, in which case a few hubs will have an innovative environment, while others won’t. Or it can happen through a war of taxes, infrastructure, and subsidies, where a locale tries to poach enough innovators from another place to improve its chances of becoming an innovation hub. A locale can choose this second path, and indeed it makes sense for Miami, Dallas, and Houston to do so. In the longer term, this competition may even create a better economic environment for everyone—at least until redistributive pressures arise in the new hub. The federal government should not be involved in these disputes.
Since skilled workers, entrepreneurs, and capital are scarce resources, the government should avoid subsidizing regional innovation without considering the general equilibrium consequences, which the original Biden NSF Regional Innovation Engines largely overlooked. It’s often possible to attribute increased investment in a region to a subsidy, but one must ask how much of the new economic activity is actually new, and how much would have existed anyway in another region of the same country but was simply poached, at a higher cost for the nation. The agglomeration process has many parts, and public R&D cannot be the only source of innovation.
The causal evidence on public R&D in the economics literature demonstrates these points. In a 2023 paper, economist Michael J. Andrews finds that only 5 percent of the increase in patents after a new college opens comes from that college’s alumni. More than 75 percent of patents come from individuals unaffiliated with their local college who lived outside the college county at the time the college was established. Andrews cites “consolation prize” cases—counties that came close to winning the state college but instead received another type of state institution, such as a prison or an insane asylum. The increase in patenting between the college and consolation-prize counties is statistically indistinguishable. Andrews’s conclusion? Population growth is ultimately what spurs additional innovation and patenting.
Andrews’s findings support the idea of a “complex system,” where adding parts creates something greater than the sum of its parts. If more people, rather than research investment, is the main driver of innovative activity, then subsidizing regional universities is not necessary to create more innovation. Increasing population and making areas denser are more likely to succeed in boosting innovation and economic growth.
A new paper in the Journal of Economic Perspectives reminds us why the government cannot create a new Silicon Valley by decree. Economists E. Jason Baron, Shawn Kantor, and Alexander Whalley note that even the best estimates of public funding’s positive effects on private R&D—0.8 euros in private funds for every public euro invested in private research—are no greater than those accruing from the agglomeration benefits of population growth. Reviewing evidence from the Diamond Light Source particle accelerator facility in the United Kingdom, they show that proximity to a major scientific facility increased related research output, primarily by attracting additional research inputs rather than increasing scientists’ productivity.
Though public research is necessary for the development of a healthy private sector, government must take a limited role. Some amount of public research is helpful not just for scientific discoveries but also as a focal point for positive collaborations, even if the research doesn’t need to be located in any particular place. In this way, the government creates both scientific findings for the private sector to apply and a place for capital, entrepreneurs, and skilled workers to meet and innovate.
Healthy tax competition for the development of these regions is positive, but it concerns only the two parties who are competing, not the federal government, which gains nothing overall by favoring one region over another. It is not the role of government to pick winners.