VoxEU Column Productivity and Innovation

Are the Nordic countries really less innovative than the US?

Do the ‘cuddly’ Nordic countries free ride on the ‘cut-throat’ incentives for innovation in US-style economies? Don’t PCs, the internet, Google, Windows, iPhones and the Big Mac speak for themselves? This column argues that, despite a higher overall tax burden and more generous safety nets, the Nordics have generated at least as much – if not more – innovation than the US. So far, ‘cut-throat’ capitalism has not been the only road to an innovative economy.

The cut-throat versus cuddly capitalism distinction (Acemoglu et al. 2012) resonates with widely-held stereotypes. The US is a ‘mean streets’ sort of place to live but the law of the jungle approach to capitalism produces breakthrough innovations. European nations – especially Nordic nations – are just, social democratic societies, but comfort saps the life out of invention. Acemoglu et al. (2012) argue that advancing the technology frontier fast requires US style cut-throat capitalism and that cuddly Nordic economies are free-riding on US innovation.

Stereotypes

Do the facts resonate as well as the stereotype? One cannot deny the importance of high-powered pecuniary incentives in advancing innovation and economic activity. We, however, argue, that – as a simple empirical matter – the claim that the Nordic countries are less innovative than the US is, to put it kindly, not unequivocally supported by data. Several measures of innovation activity and success suggest that the Nordic countries are doing equally well as, or even better than the US. There are also plausible explanations for these outcomes.

Nordics are dynamic

In their related working paper, Acemoglu et al. (2012a) use two measures to argue that the US is more innovative than the Nordic countries: the number of US patents and GDP per capita. Neither of these is a very good comparative measure of innovation.

It is quite understandable that US companies dominate patent filings in the US. For an international comparison, the so-called triadic patents – i.e. patents filed for the same invention in the US, EU and Japan – are a more suitable measure. By this indicator, US innovation activity lags behind that of Sweden, Denmark and Finland. The same result obtains when comparing some of the most frequently used indicators of innovation inputs, such as business expenditure on research and development, the share of researchers in total employment, and even the stock of venture capital as a share of GDP (Table 1). In Table 1, we don’t present statistics for Norway and Iceland because, arguably, they are special cases due to their large natural resources.

Table 1 Indicators of innovation activity, 2008

   USA  SWE  DEN  FIN
Triadic patents per million of population 48.7 88.3 60.5 63.9
Business expenditure on R&D, % of GDP 2.01 2.78 1.91 2.77
Researchers per 1000 of employed 9.5 10.6 10.5 16.2
Venture Capital, % of GDP 0.12 0.21 0.16 0.24
Worker reallocation, 2000–2007, % 43.3 32.0 45.5 39.8

Sources: Worker reallocation from Bassanini and Garnero (2012), other statistics from OECD (2010).

As Acemoglu et al. (2012a, 2012b) stress, innovation requires risk-taking. In a very innovative economy, one would therefore expect to find intensive job creation and job destruction, as firms that are successful in innovative activities expand rapidly while others are forced to exit the market.

The available data do not suggest that the US economy is unambiguously more dynamic than the Nordic economies (Bassanini and Marianna 2009, OECD 2004). In Denmark, worker reallocation is more intensive, and in Finland almost as intensive as in the US (Table 1). Moreover, time series from the US indicate a marked decline in job and worker flows since the late 1990s (Davis et al. 2012), whereas – at least in Finland – both flows have stayed intensive (Ilmakunnas and Maliranta 2011).

Nordics pushing productivity frontiers

Of course, innovation inputs, patents, or reallocation of labour are not ideal measures of a society’s success in pushing the technological frontier forward. They do not measure the outcome in terms of output or value-added.

But neither is GDP per capita a good measure. First, GDP per capita is affected by the degree to which labour resources are utilised, which does not add to innovation. Secondly, GDP includes the value-added by the public sector, which is hard to measure adequately. In an international comparison, a particular problem is that the size of the public sector varies a great deal across countries and even the data compilation practices differ.

These problems can be sidestepped by using productivity in the market sector as an outcome measure. It is also useful to look at productivity in different sub-sectors. Table 2 presents some labour productivity comparisons based on Inklaar and Timmer (2008).

Table 2 Labour productivity in 2007, % of the US figure

  SWE DEN FIN
Total economy 83 76 80
Market sector 89 75 86
Manufacturing excl. ICT industries 93 62 113
       
ICT 235 35 102
Trade 100 129 135
Transport and storage 50 73 82
Finance and business 74 70 38
Personnel services 48 92 58
Non-market services 67 79 63

Sources: Authors’ calculations based on Inklaar and Timmer (2008) and EU KLEMS database.
Note: ICT refers to electrical and optical equipment, and post and telecommunications.

The US advantage in labour productivity (value-added/hours) is somewhat reduced if we consider the market economy rather than the total economy (rows 1 and 2). Interestingly, relative productivity levels vary a great deal between sectors. The US seems to have a comparative advantage, especially in finance and business services.

In our view, there are good reasons to focus on the manufacturing sector. For one thing, measurement of productivity is the most reliable in manufacturing industries. Manufacturing also features important international technology spillovers.

Since the study by Inklaar and Timmer (2008) does not include productivity comparisons for the entire manufacturing sector, we need to exclude ICT when considering manufacturing. As Table 2 shows, in manufacturing that excludes ICT, US labour productivity is lower than in Finland and only marginally higher than in Sweden.

Figure 1 provides a comparison of productivity performance that is broader (with more countries), longer (with time-series), and more comprehensive (also including total factor productivity). This comparison of productivity performance takes into account the use of capital input as well as the quality of labour input. The figures give further evidence that the Finnish advantage in manufacturing productivity is long-standing, and not related to high capital input.

Figure 1 Relative productivity levels in manufacturing excluding ICT industries

Sources: Inklaar and Timmer (2008) and computations with EU KLEMS database.

Taken together, these observations are rather strong evidence against the claim that an American incentive system is necessary for being at the technology frontier.

The Nordics promote equal opportunity education and innovation

One explanation for Nordic good performance might be that they are better in mobilising human resources. While hours per capita are higher in the US, a larger share of the working age population is employed in the Nordics owing to more inclusive educational, social and employment policies.

This may imply that talents are harvested better for gainful economic activity. A second explanation could be the rather determined public policies to promote innovation.

Incentives are perhaps not miserable after all

A third explanation might be that the economic incentives for innovation in the Nordics, while weaker than in the US, are not miserable after all, at least not across the board. For instance, all Nordic countries have introduced dual income taxation, according to which capital incomes are taxed at a flat rate. This helps in motivating entrepreneurs, despite quite progressive taxes on earned income. Sweden has recently encouraged wealth accumulation by abolishing wealth and inheritance taxes altogether.

A well-designed safety net may also work to promote risk-taking. In particular, unemployment insurance may help risk-taking entrepreneurs by making it is easier for them to hire workers (see Acemoglu and Shimer 2000).

The Nordic model needs to be adjusted

The Nordic model has its problems. High labour and consumption taxes, together with social security benefits, reduce the labour supply. The increasing mobility of a well-educated new generation may lead to cherry-picking in different phases of life. Population ageing raises concerns about the sustainability of the generous public pension and health insurance systems.

Keeping the cuddly Nordic capitalism competitive under these pressures requires adjustments in policy. The generous ‘welfare promise’ requires an efficient public sector and high employment to be financially sustainable.

Conclusions

The US economy is undoubtedly very creative, flexible and at the productivity frontier in many fields. After all, PCs, the internet, Google, Windows, Apple Mac’s iPhone and iPad, and the Big Mac are all American innovations. Financial incentives do play an important role in this. Nevertheless, despite a higher overall tax burden and more generous safety nets, the Nordics have generated at least as much innovation activities as the US, and reached matching levels of productivity in important parts of the economy. So far, ‘cut-throat’ capitalism has not been the only road to an innovative economy.

References

Acemoglu, D, Robinson, C, and Verdier, T (2012a), “Can't we all be more like scandinavians? Asymmetric growth and institutions in an interdependent world”. NBER, working paper, 18441.

Acemoglu, D, Robinson, J, and verdier, T (2012b), “Choosing your own capitalism in a globalised world?”, VoxEU.org, 21 November.

Acemoglu, D, and Shimer, R (2000), “Productivity gains from unemployment insurance”, European Economic Review, 44, 1195-1224.

Bassanini, A, and Garnero, A (2012), “Dismissal protection and worker flows in oecd countries: evidence from cross-country/cross-industry data”, IZA/ DP, 6535.

Bassanini, A, and P Marianna (2009), “Looking inside the perpetual-motion machine: job and worker flows in oecd countries”. IZA/ DP,4452.

Davis, S J, R J Faberman and J Haltiwanger (2012), “Labor market flows in the cross section and over time”, Journal of Monetary Economics, 59, 1-18.

Ilmakunnas, P, and M. Maliranta (2011), ”Suomen työpaikka- ja työntekijävirtojen käänteitä: toimialojen elinkaaret ja finanssikriisi”, Työpoliittinen Aikakauskirja, 54, 6-23.

Inklaar, R, and Timmer, M (2008), “GGDC productivity level database: international comparison of output, inputs and productivity at the industry level”, EU KLEMS, working paper, 40.

OECD (2004), Understanding Economic Growth, Paris.

OECD (2010), OECD Science, Technology and Industry Outlook, Paris.

12,394 Reads