Can European Productivity Make Progress?

© The Author(s) 2018. Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 1 European Commission: Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Europe 2020 Flagship Initiative Innovation Union, SEC(2010) 1161, 2010; European Commission: Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, An Integrated Industrial Policy for the Globalisation Era, Putting Competitiveness and Sustainability at Centre Stage, SEC(2010) 1272, SEC(2010) 1276, 2010. Davide Castellani, Mariacristina Piva, Torben Schubert and Marco Vivarelli

Forum ades, the productivity gap between European countries and the US has not been signifi cantly reduced. By 2016, the average worker across the 28 EU member states generated $47.70 (constant prices, 2010 PPPs) of GDP for every hour worked; in the US the equivalent fi gure was $63.30; for comparison, in 2000 these fi gures were $40. 10 and $50.40, respectively. 2 In addition to the EU's widening productivity gap with the US, the entire world economy has entered a phase of low overall productivity growth in the aftermath of the Great Recession. Gordon was one of many prominent economists who wrote about this so-called "secular stagnation", and he largely attributed it to fading technological opportunities. 3 We posit that narrowing the EU's productivity gap with respect to the US may be even more problematic in times when declining technological opportunities reduce the overall potential for productivity gains. In addition, the EU industrial structure is much less high-tech intensive than that of the US, and this long-term structural effect can impact productivity. Moreover, the potential capacity of the EU to effectively increase productivity within each industry, compared to the US, is a critical issue. Finally, within the EU, country heterogeneity might be pivotal in determining differences in terms of the productivity consequences of economic cohesion policies. 4

Secular stagnation: Are there technological origins?
The past 20 years have seen a remarkable drop in the productivity growth rates for many Western economies. While the empirical observation of low productivity growth rates is well documented, some authors have raised the question of whether simple measurement issues may be the main drivers of these declining growth rates. In particular, one concern is that customary productivity measures do not appropriately refl ect the actual productivity gains due to new products. 5 However, in this respect, Syverson presents evidence indicating that the potential distorting effects of the measurement biases are not large enough to explain the slowdown in productivity growth. 6 Thus, most researchers today 2 OECD: Productivity statistics database, available at http://www. oecd.org/sdd/productivity-stats/. agree that the declining productivity growth rates are not illusionary and refl ect a real phenomenon.
What are the causes of this declining productivity growth? A widespread view among macroeconomists attributes the causes of stagnation to monetary policies 7 or abundant social welfare systems 8 . However, a more technology-centred view claims that the slowdown of productivity growth is largely the result of diminishing technological opportunities. 9 Indeed, this evidence is corroborated by Schubert and Neuhäusler's econometric evidence. 10 For a panel of 11 OECD countries, they show that the ability to sustain productivity growth through capital investment declined over the period 1993-2011. However, they also provide evidence that the effect of research and development (R&D) on productivity growth was almost unchanged throughout these years. All in all, given that the effect of embodied technological change has largely come to hold, there is little hope that productivity growth rates might return to "normal" after the economic crisis. Conversely, there is relevant evidence that the productivity-enhancing effects of genuine research activities are not declining.

EU versus US: What are the causes of the transatlantic productivity gap?
The literature has pointed to many different causes to explain the origin of the productivity gap between Europe and the US. 11 Researchers have assigned blame to, among other things, the quality of human capital, the rigidity of European labour markets, the role and diffusion of ICTs, 12 the importance of new managerial practices and organisational investments, 13  Forum be related to a revealed technological disadvantage in the EU.
In this regard, both at the aggregate and the microeconomic level, R&D expenditures can be considered a good proxy of technological investment. Therefore, the gap in corporate R&D investment can be seen as one of the main culprits of the European lag in terms of productivity growth in comparison with the US. 14 Indeed, in the last decade, corporate business expenditure on R&D in the EU has remained almost constant at approximately 1.2% of GDP. Meanwhile, in the US, this fi gure has fl oated between 1.7% and 2.0%.
Some scholars have argued that the lower European R&D expenditure is mainly due to a structural composition effect, which appears because the R&D-intensive manufacturing and R&D-  gains, receiving a smaller return on their investments than US businesses.
We use microdata on the top 1 112 R&D-spending companies in the EU (504) and the US (608), in both manufacturing and services, between 2004 and 2012. 19 Descriptive statistics show that, on average, R&D intensity and productivity are lower in EU companies than in US fi rms. Moreover, as expected, they are larger in hightech industries than in middle-and low-tech ones. In our panel data analysis, we estimate each company's productivity as a function of investments in R&D, physical capital, company size and industrial characteristics. Our results demonstrate that, for a US company, a ten per cent rise in R&D intensity yields a 2.7% increase in productivity. For EU fi rms, the same investment increases productivity by only approximately one per cent. The gap becomes even more interesting when we disentangle high-tech and middle-and low-tech sectors: for high-tech US companies, a ten per cent rise in R&D intensity yields a 3.3% increase in productivity, while for high-tech EU fi rms, the same investment increases productivity by just 1.2%. When we turn to middle-and low-tech companies, the gap shrinks, although US companies still remain better able to translate R&D investments into productivity.
We also split the sample into two periods, before and after the global fi nancial crisis. In 2004-08 the return on R&D spending for EU fi rms was not quite two-thirds that of their US counterparts, while in 2009-12 it fell to less than half. Therefore, since the economic crisis, it seems EU companies are even less able to translate R&D investments into productivity gains compared to their US counterparts. However, it is also interesting to underline that EU businesses gain more from capital spending than those in the US: a ten per cent increase in physical capital investments yields a 1.3% productivity increase in the EU, versus 1.0% in the US. This result seems to be mainly driven by middle-and low-tech industries, where EU fi rms gain a 2.2% productivity gain compared to 1.6% for US companies. In a sense, productivity growth in European non-high-tech fi rms is still heavily dependent on investments in physical capital (embodied technological change).
In general, R&D has potentially a much higher impact on productivity. Across all companies in our sample, a ten per cent increase in R&D spending leads to a 1.8% Forum rise in productivity. In comparison, the same increase in physical capital spending boosts productivity by only 1.1%.
On the whole, our results suggest that the transatlantic productivity divide can be explained by: • a lower level of R&D investments by EU fi rms compared to their US counterparts • a structural composition effect, which seems to be signifi cant both in aggregate and within the high-tech sectors • the presence of an intrinsic effect, i.e. the generally lower capacity of European fi rms to translate R&D investments into productivity gains.
Furthermore, our results show that EU companies have been more affected by the economic crisis with regard to their capacity to translate R&D investments into productivity. Indeed, the global economic crisis has worsened the productivity gap between the EU and the US.

Challenges for European policy makers
Turning Europe into the most competitive knowledgebased economy in the world is an ambitious goal. Key economic fi gures suggest that the EU continues to fall further behind the US -currently the leading economyrather than narrowing the gap. In this paper, we highlight that the EU faces the challenge of reducing its productivity gap in a period characterised by overall low potential for productivity growth. This makes the catching-up process even more diffi cult.
Focusing on the role played by R&D in the EU compared to its role in US, our empirical exercise provides robust evidence of both a "quantity effect" (relatively lower R&D spending by European fi rms) and a "quality effect" (lower ability of European fi rms to turn R&D spending into productivity gains). However, most policy attention has been devoted to the lower levels of R&D spending rather than the lower capabilities to make effi cient use of it; for example, the EU set a target of increasing R&D investment to three per cent of GDP. Instead of primarily focusing on the symptom (i.e. R&D investments that are perceived as too low), effective policies should take into account the reasons why EU fi rms obtain less productivity gains from their R&D investments compared to their US counterparts. To get the most out of R&D, EU fi rms need to address the structural problems that reduce their ability to translate investment into productiv-ity gains. The main cause might lie in some combination of available human capital and the organisational/managerial practices adopted (i.e. fi rms with better-managed R&D departments might provide relevant incentives and conducive work environments to their researchers, who in turn can be more productive).
In a sense, identifying best practices -embodied by those EU fi rms most successful at leveraging their R&D investments -would allow policy to be designed to foster knowledge transfer and learning between fi rms. Moreover, this would enable EU funding programmes to more effectively promote excellence by favouring such best practices. Indeed, several authors have acknowledged that research and innovation support may become more effective in promoting growth and productivity by taking a more selective approach. 20 Resources should be concentrated in promising high-tech fi elds that could potentially initiate and sustain the next phase of high productivity growth.
Although compelling, policies redirecting resources to the most effective uses will unavoidably result in a tendency to concentrate resources in scientifi c organisations, fi rms, sectors and, eventually, countries that already perform well in research and innovation. If the major concern to be remedied is the poor average/aggregate performance of the EU in comparison to the US, greater selectiveness in the allocation of support for innovation circumvents the problem of low effectiveness in some parts of Europe rather than addressing it. However, while such a strategy may be effective in increasing the average returns per euro spent, it may come at the high cost of lower European cohesion.