President Donald J. Trump's concerns about the US’s security and competitiveness with other major economic players in the world culminated in tariffs on certain imports from China announced in March 2018. On 2 April, only three weeks later, China retaliated with tariffs on various US products. Trade tensions have also intensified with Canada, which announced retaliatory tariffs, and with the EU.
Similar concerns were raised in the 1970s and early 1980s, following the US exposure to a remarkable convergence by advanced countries, including Japan, Germany and France, in terms of technology and productivity (see Figure 1). In contrast to the protectionist rhetoric nowadays, the Reagan administration introduced, among other policies, an R&D tax credit scheme in 1981 for the first time in US history, followed by additional support from individual states. Figure 2 shows that, subsequently, R&D and innovation activity picked up in the US.
The recurring debate about global competition and the optimal industrial policy response calls for a deeper understanding of the diverse mechanisms through which these policies affect firms, with the implications likely to vary with the time horizon under consideration.
Figure 1 Convergence between the US and its peers, 1976–80
Notes: The figure shows the relationship between growth of average labour productivity in the manufacturing sector and growth in the number of patent applications for the United States and its major trading partners between 1976 and 1980.
Source: Taken from Akcigit et al. (2018), based on data on patent applications in the US from the USPTO and on international productivity comparisons from Capdevielle and Alvarez (1981).
Figure 2 R&D and innovation intensity of US firms, 1975–95
Notes: The figure shows the evolution of aggregate R&D intensity (defined as the ratio of total R&D spending to total sales) of the public US firms listed in the Compustat database, and the share of patents registered by US residents in total patents registered in the USPTO database from 1975 to 1995. The ratios are calculated annually. The bars show the total number of US states providing R&D tax credits, along with their names, for every year since the first adoption of such a measure in 1982.
Source: Taken from Akcigit et al. (2018).
A long series of studies have examined the effect of trade openness and trade policies on firm behaviour and aggregate productivity (e.g. Eaton and Grossman 1986, Grossman and Helpman 1991, Melitz 2003, Demidova and Rodríguez-Clare 2009, Costinot and Rodríguez-Clare 2014). Despite the early seminal work by Grossman and Helpman (1991), however, the dynamic link between trade and firms’ innovation decisions – the engine of factor productivity growth – has only recently gained traction (Burstein and Melitz 2013). Moreover, given the dynamic links between trade and innovation, policy horizons and transitional dynamics are of utmost importance in trade policy analysis. Taking these considerations seriously, in this column we examine the optimal mix of tariff policies and R&D subsidies to support domestic firms in global technological competition and to enhance aggregate welfare.
In a recent paper, we build a dynamic model of international trade and innovation where firms compete strategically in production and innovation (Akcigit et al. 2018). A central feature of our framework is that firms’ production and innovation decisions depend on their technological position relative to their foreign competitors, reflecting the strategic interactions between large firms. Such firms dominate global R&D races and international trade, and their choices affect market aggregates and give rise to strategic market power. For example, Airbus and Boeing dominate the technology-intensive aircraft industry (Baldwin and Krugman 1988, Irwin and Pavcnik 2004), and the top 1% of US trading firms account for about 80% of total US trade. Our framework captures the dynamic relationship between trade and firm innovation – and thus aggregate growth and welfare. In our quantitative exercises, we focus on the heterogeneous implications of industrial policies for aggregate growth and welfare across different policy horizons, accounting for the different transitional dynamics these policies generate.
We build a new two-country dynamic endogenous growth model where innovation determines the dynamics of technology and global market leadership. In both countries, final good firms produce outputs with a set of intermediate inputs, sourced from domestic and foreign producers. In each intermediate sector, a home firm and a foreign firm compete for global market share and invest in R&D to improve the quality of their product. Based on step-by-step innovation models of Schumpeterian creative destruction, this model structure allows for heterogeneous between-firm technology differences across intermediate sectors. Endogenous entry by a fringe of domestic and foreign firms creates additional competitive pressure for both leaders and followers in each sector. International markets are characterised by trade costs and the international diffusion of ideas through knowledge spillovers.
A key mechanism is that trade openness affects the economies' dynamics by changing firms' motives for innovation. With positive trade costs, firms intensify their innovation efforts when they risk losing their domestic market to a technologically superior foreign firm (defensive R&D) or when their technological advantage is large but not enough to cover trade costs and export (expansionary R&D). These novel implications generate a double-peaked distribution of R&D effort over the relative quality space of firms that, remarkably, is also supported by USPTO patent data. From a policy point of view, the distinction between defensive and expansionary R&D is crucial, because they generate different responses to alternative industrial policies.
We parameterise the model to match key trade, innovation, and growth facts in the late 1970s and reproduce the evolution of global leadership in those years, with the US initially representing the technological frontier in most sectors. The transitional dynamics of the model reproduce the convergence in technological leadership observed in patent data in the 1970s and early 1980s. The model’s mechanism is fairly successful in matching empirical patterns for the link between innovative activity and technological leadership that are not targeted and the elasticity of firm R&D spending to policy changes.
Raising import tariffs unilaterally
In the first policy exercise, we analyse the welfare implications of protectionism – i.e. raising import tariffs unilaterally. The welfare implications of this policy change depend on the time horizon over which the policy is evaluated. A rise in tariffs generates short-run gains, because it tames international business-stealing caused by foreign technological catch-up, keeping business profits at home (see Figure 3A). These gains more than compensate for the negative effect on aggregate productivity of replacing better-quality imported goods with inferior domestic counterparts. But protective measures reduce the incentives for domestic firms to invest in defensive innovation, because they weaken domestic firms' exposure to foreign competitive pressures (see Figure 3B). As time goes by, this force dominates, leading to a substantial welfare decline in the long run. Weaker foreign competition caused by protection and the ensuing reduction in defensive innovative activity also shape the optimal trade policy in our model: Lower tariffs are preferred when the welfare impact is evaluated over a longer time horizon.
Figure 3 Welfare effects of protectionism: Unilateralincrease in tariffs
a. Consumption equivalent welfare
b. Innovation response to protectionism
Notes: The figure illustrates the effects of a unilateral 50% increase in US tariffs (protectionist US policy without retaliatory response). Panel A shows the welfare change in consumption-equivalent terms over different time horizons. Panel B shows the shift in the innovation-effort profile of US incumbent firms over technology gaps
Increasing R&D subsidies
As an alternative policy option to protectionism, we fed the model with an increase in R&D subsidies, replicating the US move in the early 1980s. The subsidy increase generates non-negligible welfare gains in both the short and long run (except perhaps in the initial years, when higher R&D spending reduces the resources for consumption). By reducing the cost of innovation, subsidies stimulate both entrant and incumbent firms’ R&D in the US, thus accelerating productivity growth and allowing US firms to attain market leadership. Our findings suggest that the observed increase in subsidies was an optimal response only for a very short time horizon, while longer policy horizons call for much higher subsidies, as the growth-stimulating impact of subsidies becomes stronger over time.
Optimal policy mix
A key result for the optimal mix of tariffs and subsidies is that the direction of the trade policy component depends crucially on the assumption about the response of the trade partners. Protectionist trade measures become optimal only when unilateral changes are possible, combined with aggressive R&D subsidies. The reason is that protectionist policies protect domestic profits but reduce innovation incentives. So aggressive R&D subsidies are needed to make up for the reduced innovation efforts. However, if the trade partners retaliate, the optimal trade policy reverses and calls for a regime as liberal as possible. The risk of losing the export market is key in this reversal.
Figure 4 Optimal joint policy in unilateral and bilateral tariff changes
Notes: The figure compares horizon-dependent optimal joint policy in the case of (trade-policy) retaliation to that in the baseline.
Effect of falling trade costs
Last but not least, our analysis shows that less policy intervention is needed as the world becomes more globalised through reduced trade costs. This interesting result arises because lower trade costs intensify competition in the global marketplace. More competitive markets induce more innovation, both defensive and expansionary. In other words, with globalisation, markets take care of the innovation incentives and eliminate the need for policy intervention.
Authors’ note: The views expressed are solely the responsibility of the author and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or of any other person associated with the Federal Reserve System.
Akcigit, U, S T Ates and G Impullitti (2018), “Innovation and Trade Policy in a Globalized World,” NBER Working Paper 24543.
Baldwin, R and P Krugman (1988), “Industrial Policy and International Competition in Wide-bodied Jet Aircraft,” in R Baldwin (ed.), Trade Policy Issues and Empirical Analysis, University of Chicago Press, pp. 45–78.
Burstein, A and M J Melitz (2013), “Trade Liberalization and Firm Dynamics,” in D Acemoglu, E Dekel, and M Arellano (eds), Advances in Economics and Econometrics Tenth World Congress. Applied Economics, Vol. 2, pp. 283–328.
Capdevielle, P and D Alvarez (1981), “International Comparisons of Trends in Productivity and Labor Costs,” Monthly Labor Review 104: 14–20.
Costinot, A and A Rodríguez-Clare (2014), “Trade Theory with Numbers: Quantifying the Consequences of Globalization,” in E Helpman, K Rogoff and G Gopinath (eds), Handbook of International Economics, Vol. 4, pp. 197–261.
Demidova, S and A Rodríguez-Clare (2009), “Trade Policy Under Firm-level Heterogeneity in a Small Economy,” Journal of International Economics 78(1): 100–112.
Eaton, J and G M. Grossman (1986), “Optimal Trade and Industrial Policy under Oligopoly,” Quarterly Journal of Economics 101(2): 383–406.
Grossman, G M and Elhanan Helpman (1991), International Trade and Trade Policy, MIT Press, Chapter 8, pp. 141–163.
Irwin, D A and N Pavcnik (2004), “Airbus versus Boeing Revisited: International Competition in the Aircraft Market,” Journal of International Economics 64(2): 223–245.
Melitz, M (2003), “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity,” Econometrica 71: 1695–1725.