Lessons Learned

Lessons Learned from U.S. Technology and Innovation Policies

Although the United States has never had a coherent set of technology policies, government actions have profoundly influenced the rate and direction of technological change. Federal policies affecting technological change began with the codification of the patent system in the U.S. Constitution. Federal land grants supported the U.S. system of publicly financed colleges and universities, which became major players in R&D and innovation. In addition, government procurement during World War I transformed an infant aircraft industry that had produced only a few hundred planes; by the war’s end, U.S. firms had manufactured some 14,000 planes, learning a great deal in the process. Government-spurred innovation accelerated in the post-World War II period. Despite the heterogeneity in federal policies—or perhaps because of it, given the high levels of uncertainty that characterize innovation—government actions have been remarkably effective. Lessons learned from this rich experience are supported by a large body of literature in economics and other fields concerning innovation, and include the following:

• Technological change is a complex process involving invention, innovation, adoption, learning, and diffusion of technology into the marketplace. The process is highly iterative, and different policies influence outcomes at different stages. For example, the U.S. government spurred diffusion of know-how in microelectronics through policies including antitrust and defense procurement. In response to a federal government antitrust suit, AT&T released technical information about the transistor (which it invented), licensed the relevant patents at nominal rates to all comers, and refrained from producing transistors for outside sale. Texas Instruments then introduced the first commercially successful transistor, and the Department of Defense (DoD) and its contractors began to design the new devices into radar, sonar, missile guidance, and communications systems, stimulating further learning and cost reductions. In addition, DoD procurement contracts stipulating that the chips be available from at least two suppliers led to the sharing of design and process know-how, which encouraged new market entrants and accelerated inter-firm technology flows.
  
  
• Gains from new technologies are realized only with widespread adoption, a process that takes considerable time and resources and typically depends on a lengthy sequence of incremental improvements that enhance performance and reduce costs. For example, several decades of significant government and private sector R&D investments occurred before gas turbines derived from military jet engines improved in efficiency and reliability to the point that they were cost-effective for electric power generation. Today, gas turbines are the leading technology for new, high-efficiency power plants with low GHG emissions.
  
  
• Technological learning is the essential step that paces adoption and diffusion. “Learning-by-doing” contributes to reductions in production costs, and adopters of new technology contribute to ongoing innovation through “learning-by-using.” Widespread adoption, in turn, accelerates the incremental improvements from learning by users and producers, further fueling adoption and diffusion. For example, an entirely new class of products emerged as Intel (and soon, other firms) designed successive families of microprocessors, based in large part on feedback from users. When Intel began work on its 386 processor family, the lead technical and marketing specialist spent six months simply visiting customers to understand the features they valued most highly.
  
  
• Technological innovation is a highly uncertain process. Because pathways of development cannot be predicted, government policies should support a portfolio of options, rather than a particular technology or design. The unforeseen explosive diffusion of the Internet during the 1990s is illustrative. Both the Internet’s technologies and many of the formal and informal governance mechanisms that evolved to coordinate its standards and infrastructure sprang from DoD-sponsored networking research and trials.
  

In addition to these insights gained regarding the innovation process, lessons learned from U.S. experience with technology policies over the past several decades include the following:

• Federal investments contribute to innovation not only through R&D but also through “downstream” adoption and learning. For example, in the early years of computing, defense agencies made indispensable contributions to a technological infrastructure that propelled the industry’s rise to global dominance.
  
  
• Public-private R&D partnerships have become politically popular because they leverage government funds and promote inter-firm collaboration. Partnerships may have particular advantages in fostering vertical collaborations, such as those between suppliers and consumers of energy.
  
  
• Adoption of innovations that originate outside a firm or industry often requires substantial internal investments in R&D and human resources. Smaller firms may be less able to absorb innovations without government assistance.
  
  
• Just as competition in markets helps resolve uncertainties and improves economic performance, competition within government can improve performance in fostering innovation. The messy and often duplicative structure of U.S. R&D support and related policies creates diversity and pluralism, fostering innovation by encouraging the exploration of many technological alternatives.
  
  
• Because processes of innovation and adoption are lengthy and convoluted, effective policies and programs require sustained political support. Reliable political constituencies have been essential for the development of new technologies in defense and for research in the biomedical sciences. By contrast, technology policies for addressing climate change face a discordant political environment.

Previous | Next