Innovation spillovers and market structure
Innovation spillovers and market structure: Evidence from the interwar aircraft industry
The slowing pace of economic growth in the US and Europe have rekindled fears of reduced innovation and prompted calls for institutional changes meant to increase returns on spending for research and development. This column uses the case of the US interwar aircraft industry to suggest some unforeseen hazards of such change. It recommends considering the design of innovation and antitrust policy in tandem, especially where attempts to incentivise innovation may alter the extent of competition and endogenously reconfigure market structure.
In economic models of endogenous technological change, economic agents take purposeful actions to allocate resources across investments in areas such as research and development or human capital (Romer 1990, Grossman and Helpman 1991, Aghion and Howitt 1992). The success and ultimate payoff from these investments is closely linked to the institutional environment that incentivises risk-taking and protects intellectual property (e.g. Williams 2017). The application of these models and their offshoots have been used to understand the forces underlying the most important change in the world economy in the last several hundred years: the onset of modern economic growth (Mokyr 2005). These models have also been useful for thinking about the redesign of current institutions to protect intellectual property (IP).
More recently, the slower pace of economic growth in the US and Europe has led some observers to highlight the increasing difficulty of ‘finding new ideas’ (Bloom et al. forthcoming). The potential ‘stagnation of science’ (Collison and Nielsen 2018) is linked to diminishing returns of time or money spent on inventive activity, and the need for either better funding for R&D or institutional changes that will help to increase the returns from current spending.
Evidence from historical experience, combined with economic theory, can shed light on the innovation policy gains associated with institutional change, as well as the challenges that attend large-scale reforms that touch many sectors of the economy. In Hanlon and Jaworski (2019), we consider the case of the aircraft industry in the US between WWI and WWII. In particular, we focus on the potential for changes in IP protection to generate spillovers – which may increase or decrease the incentive to innovate in connected markets – and the subsequent effects on market structure when firms choose to merge in response to changes in the IP regime.
Following WWI, allegations of war profiteering led the US Congress to limit IP protections available to airframe producers of military designs, which represented the vast majority of the aircraft market during this period. As a result, companies that developed a new airframe design did not have an exclusive right to their own design. Instead, the larger and more profitable production contracts were put up for competitive bidding, which led to perverse outcomes that diminished the incentive to innovate. In the case of the Glenn L. Martin Company’s MB-3 bomber, the War Department purchased 200 versions of the plane from another aircraft manufacturer, Curtiss, based on a competitive bid that did not include the costs of research and development.
Both the military and the airframe producers were aware that competitive bidding had the potential to impede technological progress. Eventually, pressure exerted by both parties led to changes in procurement procedures under the 1926 Air Corps Act, which included language that Air Corps and Navy procurement officers exploited to avoid competitive bidding in favour of contracts negotiated with individual manufacturers. This led to a significant shift in which firms developing new airframe designs also received exclusive rights to the production contract. Importantly, this change in the IP regime only applied directly to airframe producers; aircraft engine producers had access to IP protection throughout the interwar period.
Figure 1 illustrates the first key finding to emerge from our empirical analysis. In each panel, the vertical axis reflects technological progress of new designs as measured by wing-load for airframes (left panel) and horsepower per unit of piston displacement for engines (right panel). For airframes, technological progress was slow before 1926 and increased dramatically afterward; the reverse pattern characterised engines. In the absence of a change in IP regime for engine technology, and insofar as it is possible to rule out the overwhelming importance of other factors, we label the effects for airframes direct (i.e. IP protection increases the incentive to innovate in areas where IP is granted) and the effects for engines indirect (i.e. IP protection decreases innovation in areas where technology is a complement)2 to highlight the way that changes in innovation policy may have spillover effects beyond the sectors narrowly targeted for reform.
Figure 1 Airframe and engine innovation in the US
The insight from this set of results is that the total amount of innovation reflected in the final good will depend on the incentives to innovate in the sectors of all intermediate goods. In turn, these incentives will be shaped by the market structure that intermediate producers face in the short run. In the long run, producers may seek to alter the market structure in terms of the number and type of firms.
Indeed, our second key empirical finding highlights how the new IP regime led to changes in the incentive for airframe and engine producers to merge during this period. Theoretically, a change in IP protection would allow airframe producers to extract rents for new designs and give rise to the classic double-marginalisation problem. Empirically, two main producers of military aircraft engines operated in the US before 1926: Wright and Pratt & Whitney. By 1929, both had merged with major airframe producers: Wright combined with Curtiss to form the Curtiss-Wright Corporation, while Pratt & Whitney merged with Boeing and several smaller airframe producers to form United Aircraft. We call these ‘vertical-complement’ mergers.
Although our findings are related to a specific time and industry, the mechanisms we highlight are likely to be at work in other periods and sectors. One broader implication of our work is the need to consider the design of innovation and antitrust policy together, especially when attempts to incentivise innovation may alter the extent of competition and endogenously reconfigure market structure. In addition, following work by Schmookler (1966), we emphasise the interaction between market size and innovation and the consequences of spillovers across markets where products are linked as substitutes or complements.
Overall, our results suggest the challenges that lie ahead for the design of institutions to incentivise innovation and highlight that this will be more so in economies where the number of connections across firms – or individuals and regions – is growing rapidly. Of course, history also provides evidence for the benefits that flow from getting these institutions right.
Aghion, P and P Howitt (1992), “A Model of Growth Through Creative Destruction”, Econometrica 60 (2): 323–351.
Bloom, N, J Van Reenen, C I Jones and M Webb (forthcoming), “Are Ideas Getting Harder to Find?”, American Economic Review.
Collison, P and M Nielsen (2018), “Is Science Stagnant?”, The Atlantic, 18 November.
Grossman, G and E Helpman (1991), Innovation and Growth in the Global Economy, The MIT Press.
Hanlon, W and T Jaworski (2019), “Spillover Effects of IP Protection in the Inter-war Aircraft Industry”, NBER Working Paper 26490.
Mokyr, J (2005 ) “Long-Term Economic Growth and the History of Technology”, Handbook of Economic Growth 1: 1113-1180.
Romer, P M (1990), “Endogenous Technological Change”, Journal of Political Economy 98(5, Part 2): S71–S102.
Schmookler, J (1966), Invention and Economic Growth, Harvard University Press.
Williams, (2017), “How Do Patents Affects Research Investments?”, Annual Review of Economics 9: 441-469.
1 Alternatively, in the case where technology is a substitute, theory predicts that the indirect effects of a similar change in IP protection would be to increase the incentive to innovate.