Innovation and productivity in the food vs. the high-tech manufacturing sector

ABSTRACT

The food sector is considered a mature industry characterized by low research and development (R&D) intensity. Nevertheless, food companies face numerous challenges and cannot do without innovation activity if they want to keep their competitiveness. In this study, we examine the impact of innovation on labor productivity in European food companies and compare it to results for firms operating in high-tech sectors. The central motivation of our study is that the low R&D intensity observed in the food sector should be mirrored in different productivity effects of innovation when compared to the high-tech sector. We use microdata from the European Union's ‘Community Innovation Survey’ (CIS) and apply an endogeneity-robust multi-stage model that has been applied by various recent studies. Our results point out major differences between the examined subsectors. While we find strong positive effects of innovation on labor productivity for food firms, we find insignificant effects in the high-tech sector. This might suggest that the returns to innovation might be best evaluated separately by sector rather than for the manufacturing sector as a whole.

KEYWORDS:

• Innovation
• productivity
• food industry
• high-tech vs. low-tech

JEL CODES:

• D24
• L6
• O3

Acknowledgments

We thank Geneviève Villette for her assistance in the Eurostat Safe Center.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 In terms of production value, the food sector was the largest subsector in EU manufacturing in 2015 (Eurostat 2018).

2 To our knowledge, panel identifiers are implemented starting from CIS 2014 so that first panel data estimations will be possible with the consecutive CIS wave.

3 The likely reason for this is that the same instrumental variables are used for both innovation indicators and the relationships between instrumental variables and the innovation indicators are similar, leading to high correlation between the results. High correlation between the innovation indicators is also observed by Raffo, Lhuillery, and Miotti (2008), Acosta, Coronado, and Romero (2015), Tevdovski, Tosevska-Trpcevska, and Disoska (2017).

4 Hashi and Stojčić (2013), as well as Tevdovski, Tosevska-Trpcevska, and Disoska (2017) find marginal effects of innovation on labor productivity higher than 100% in some cases.