Innovation left to the market means existing path dependence will be reinforced – locking in inferior and dangerous tech

//DH

With the innovation spillovers that innovation policy scholars are used to considering, innovators benefit one another because each individual innovator is unable to reap the full rewards from her innovations. This feature captures an important part of the innovation process, but not the whole story. Another feature of innovation is path dependence, in which the innovation choices of past innovators influence the types of innovation pursued by future innovators. Path dependence does not necessarily directly benefit innovators. Rather, it redirects innovators toward the types of technologies with the larger stock of knowledge by changing the relative profitability of different types of technologies. Redirecting technological development toward cleantech may not increase the profitability of innovating or producing in the energy sector (and thus may not lead to innovation spillovers), but it does reduce pollution externalities. Thus, with innovation snowballing the population at large— not the innovators themselves—gain from technological development with social benefits like lower pollution externalities. And unlike spillovers, innovation snowballing does justify encouraging cleantech innovation even in the presence of an optimal carbon tax and optimal general innovation policy. In particular, new economic research suggests that there are substantial social welfare gains to encouraging clean technology innovation specifically, along with a carbon tax.57 Innovation snowballing resuscitates early environmental law’s goals (though not its means) of using policy to encourage the development of certain pollution-reducing technologies.

Today’s innovators “stand on the shoulders” of yesterday’s innovators.58 Pick almost any technology, and it will be the combination of a myriad of innovations developed across many years. There would be no Uber, for example, without the invention of GPS. There would be no electric car without the invention of the battery. Innovation is a cumulative process whereby past innovations create new opportunities for future innovations. All else equal, innovating in technologies with small knowledge stocks (e.g., cleantech) will tend to be less productive and less profitable than innovating in technologies with large knowledge stocks (e.g., dirtytech). As a result, profit-seeking innovators naturally gravitate towards technologies with the most productive knowledge base—and therefore the greatest expected returns to research and development investments.59 Innovation therefore begets more innovation, creating path dependencies and inertia in the direction of technological change. A corollary to this insight is that society will sometimes continue to innovate in certain technologies, even after it becomes clear that superior alternatives exist. Economists have been working with the “standing-on-shoulders” idea for a few decades now.60 What’s new is working out the implications of the path dependence where the dominant existing technology produces big negative externalities, like greenhouse gas emissions. The result of innovation snowballing is that the innovation system has multiple possible technology equilibria—and that major interventions can shift the system to a different equilibrium by altering the relative productivity of different lines of research.61