Weaker jobs, weaker innovation. Exploring the effects of temporary employment on new products

ABSTRACT

This work explores the relationship between temporary employment and product innovation focusing on five major European economies (France, Germany, Italy, Spain and the Netherlands) observed between 1998 and 2012. The analysis distinguishes sectors according to their technological characteristics and regimes finding that industries using temporary employment tend to have a weaker product innovation propensity. The negative correlation between temporary employment and innovation turns out to be stronger in those sectors where tacit firm’s specific knowledge is crucial to the development of innovations. These sectors are identified using both the ‘Cumulativeness’ proxy stemming from Peneder’s classification as well as distinguishing between different Schumpeterian regimes – Schumpeter Mark I vs. II – of knowledge accumulation.

KEYWORDS:

• Product innovation
• labour market flexibility
• industry-level analysis
• technological change

JEL CLASSIFICATION:

• O31
• J21

Acknowledgments

The authors wish to thank Alfred Kleinknecht, Mario Pianta, Federico Tamagni, Arianna Martinelli and Maria Enrica Virgillito for their comments and suggestions. This paper was produced as part of the ISIGrowth project on Innovation-fuelled, Sustainable, Inclusive Growth. All the usual disclaimers apply.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Assessments of the spread of labour market reforms after the crisis are available, for the OECD economies in Adascalitei et al. (2015). The European cross-country heterogeneity in terms of labour market reforms is analysed by Picot, and Tassinari (2017).

² In this study, we do not consider flexible contracts used as screening devices during trial periods due to the lack of information concerning contracts’ duration.

³ These kind of assumptions are peculiar of traditional contract theory models as those in (Milgrom, and Roberts 1992). We thank an anonymous referee for providing specific comments on this point.

⁴ Both countries implemented a general reduction in workers’ rights and introduced the recognition of the economic reason as a fair motivation for firing.

⁵ For a detailed study of European labour market reforms, consult the LABREF EU database, available online at https://webgate.ec.europa.eu/labref/public/.

⁶ In 2016, the share of young people (15–24 years old) employed with temporary contracts was 72,9% in Spain, 54.7% in Italy, 53.8% in Netherlands, 58.6% in France and 53% in Germany. Source Eurostat LFS.

⁷ The EPL measures the level of strictness of labour legislation across countries, in particular, in relation to the cost of dismissing an individual workers or a group of workers employed on various types of contracts. For further details see http://www.oecd.org/els/emp/oecdindicatorsofemploymentprotection.htm.

⁸ New goods with higher volatility.

⁹ Improvement of mature goods with lower levels of volatility and more stable demand.

¹⁰ This kind of comparison is proposed also in the studies based on the ’Varieties of capitalisms’ approach (Hall, and Soskice 2001).

¹¹ We need to distinguish between functional flexibility and numerical and wage flexibility. The former is based on the possibility for workers to assume different roles and responsibilities within the same firm (for a seminal distinction see Beatson 1995). This internal mobility can facilitate cooperation, increase workers’ motivation and engender trust. In a context of incomplete contracts and bounded rationality, employees’ participation can facilitate the solution of problems and promote a trial error learning process lorenz1999trust.

¹² On this point, see also Kleinknecht, van Schaik, and Zhou (2014).

¹³ To enable comparison, the innovation and employment variables taken, respectively, from CIS6 and LFS were converted into Nace Rev.1 using the conversion matrix proposed in Perani, and Cirillo (2015).

¹⁴ According to CIS metadata, ‘Product innovative enterprises’ are those who introduced new or significantly improved goods and/or services with respect to their capabilities, user friendliness, components or sub-systems. Changes of a solely aesthetic nature and the simple resale of new goods and services purchased from other enterprises are not considered as innovation.

¹⁵ Note that the Eurostat definition of temporary contracts refers to employees on temporary contracts ‘that declare themselves as having a fixed term employment contract or a job which will terminate if certain objective criteria are met, such as completion of an assignment or return of the employee who was temporarily replaced’. Lisi, and Malo (2017) emphasize that the Eurostat definition of temporary employment does not allow to distinguish between fixed-term contracts and temporary agency workers.

¹⁶ The Herfindahl–Hirschman index is built using AMADEUS firm-level data.

¹⁷ Descriptive statistics for the main variables are provided in Table A5.

¹⁸ The Peneder (2010)’s measure of cumulativeness is based on two CIS questions: the first regarding the relative importance of internal versus external sources of information; the second classifies firms depending on whether the firm appears to be a technological leader or follower. Therefore, high-cumulativeness sectors are those ones where firms are classified as ‘creative’ in terms of technology development and internal sources of knowledge are important as well as external sources. High-cumulativeness sectors are also those where ‘adaptive firms’ – namely firms adopting technology from outside – acquire knowledge from external sources. Conversely, sectors are classified as low-cumulativeness if creative firms relying on external sources of information prevail, or if adaptive firms relying equally on internal and external sources of information dominates. The main idea is that if knowledge is highly cumulative, creative firms closer to the technological frontier will more ‘heavily rely on their own sources of information. Conversely, adaptive firms – more distant from the technological frontier – will have to acquire knowledge for their innovation activities from external sources’ (Peneder 2010, 327).

¹⁹ Note that, in this paper, we cluster sectors according to their technological features provided by the Schumpeter and Peneder characterizations (cumulativeness criteria). Thus, our approach differs from Lisi, and Malo (2017) where sectors are ranked on the basis of technology-related employees’ characteristics – such as the share of workers with a tertiary education or the share of workers engaged in science and technology tasks.

²⁰ For instance, Spain presents the highest shares of temporary employment – almost double that in Germany, France, Italy and the Netherlands. This holds for all Spanish sectors and is probably due to the deep and generalized labour market ‘flexibilization’ put forth in that country starting from 1984.

²¹ It loses its significance in regression (5) and (6) where we introduce the interaction term with the Peneder technological dummy and the Herfindahl index, and in regression (7) and (8) where we further introduce a dummy for Southern Europe.

²² As explained, the Peneder (2010)’s is a categorical variable thus the coefficients associated to the high and low-cumulativeness modality are interpreted in relation to the omitted category (low-cumulativeness).

²³ Concentration of intangible assets is the variable used to discriminate between Schumpeterian regimes, see Section 3.2.

²⁴ The case of Spain is counter-intuitive since the country experienced the highest level of short-term contracts from eighties. With the Workers’ Charter reform in 1984, that essentially allowed the use of temporary contract for any kind of job and without specific constraints, the demand of temporary jobs significantly rose well before the other countries and remained high.

Additional information

Funding

This work was supported by the European Union’s Horizon 2020 research and innovation programme ISIGrowth [grant number 649186].