Green together? The effects of companies’ innovation collaboration with different partner types on ecological process and product innovation

ABSTRACT

This paper investigates the effect of companies’ innovation collaboration with different partner types on the emergence of different typologies of ecological innovation (EI), specifically process- and product-EI. Econometric analyses, based on a sample of 546 German manufacturing companies collected as part of the Community Innovation Survey, indicate a differential effect of collaboration with individual partner types. Specifically, we find that collaboration with consumers is associated positively with both process- and product-EI, whereas collaboration with universities and suppliers is associated positively only with process-EI. Collaboration with enterprise customers and competitors is neither associated with process-EI nor product-EI. Our results shed light on the mechanisms within the recently established open eco-innovation mode and emphasise the importance for theory and practice of distinguishing among collaboration partners, contingent on the underlying typology of EI. We discuss important implications for theory and practice.

KEYWORDS:

• Ecological innovation
• process ecological innovation
• product ecological innovation
• collaboration

JEL:

• M10 General Business
• O30 Innovation - General

Disclosure statement

Please note that the authors used CIS data in the following article, too, to investigate the effect of project-level collaboration breadth (i.e., the number of collaboration partner types) and collaboration depth (i.e., the intensity of the interactions with these partners) on the incremental and radical innovation performance of innovation projects: Kobarg, S., J. Stumpf-Wollersheim and I. M. Welpe. 2019. “More Is Not Always Better: Effects of Collaboration Breadth and Depth on Radical and Incremental Innovation Performance at the Project Level.” Research Policy 48(1): 1–10.

Notes

¹ Please note that we do not consider the role of service providers in our study, as this group is highly heterogeneous and thus neither allows for the formulation of a consistent set of hypotheses against the background of the literature nor for drawing meaningful findings regarding this group.

² Please note that given the high correlation between the dependent variables, we investigated potential dependencies of estimations for process-EI vs. product-EI. Simultaneous Poisson estimation indicated that our results are robust to the independent estimations, i.e., the inter-relationship between process-EI and product-EI does not influence the reported results, and reporting two different models appears justified.

³ Please note that our data would allow for a comparative analysis across additional sectors, e.g., (financial) services. However, comparative analyses would require further differentiated and cumbersome hypothesis development, which would not be possible against the background of the existing literature and would dilute the focus of our study.

⁴ We thank an anonymous reviewer for this valuable suggestion.