Search results

Large digital platforms face intense scrutiny over self-preferencing, which involves a platform provider favoring its own offers over those of competitors. In online marketplaces, also called retail or e-commerce platforms, much of the academic and regulatory debate focuses on determining whether the marketplace provider gives preference to its own private labels, such as “Amazon Basics” or Walmart’s “Great Value” products. However, we outline, both conceptually and empirically, that self-preferencing can also occur through other dimensions of vertical integration – namely, retailing and fulfillment.

This article contributes by conceptualizing three dimensions of vertical integration in online marketplaces – private labels, retailing and fulfillment. Then, two studies empirically assess (1) which of the 20 most-visited global online marketplaces vertically integrates which dimension and (2) which share of 600 m available offers is vertically integrated to which degree in eleven international Amazon marketplaces.

The majority of the leading marketplaces vertically integrate all three dimensions, implying ample opportunities for self-preferencing. Across international Amazon marketplaces, only 0.02% of available offers consist of an Amazon private-label product. However, Amazon is a retailer for around 31% and fulfills around 38% of all available offers in its marketplaces. Hence, self-preferencing on Amazon can occur most frequently through retailing and fulfillment but comparatively infrequently through private-label offers. Still, these shares differ substantially by country – every second offer is vertically integrated in the USA, but only one in ten in India.

Most of the self-preferencing debate often focuses on private-label products. Instead, we present large-scale empirical results showing that self-preferencing on Amazon could occur most often through retailing and fulfillment because these channels affect much larger shares of offers. We also measure the variation of these shares across countries and relate them to regulatory environments.

Talent is a key resource for urban development, and building and disseminating urban brands have an important impact on attracting talent. This paper explores what kind of urban brand ecology (UBE) can effectively enhance urban talent attraction (UTA). We explore this question using a novel grey quantitative configuration analysis (GQCA) model.

To develop the GQCA model, grey clustering is combined with qualitative configuration analysis (QCA). We conducted comparative configuration analysis of UTA using fuzzy set QCA (fsQCA) and the proposed GQCA.

We find that the empirical results of fsQCA may contradict the facts, and that the proposed GQCA effectively solves this problem.

Based on the theory of UBE, we identify bottleneck factors for improving UTA at different stages. Seven configuration paths are described for cities to enhance UTA. Theoretically, this study expands the application boundaries of UBE.

The proposed GQCA effectively solves the problem of inconsistent analysis and facts caused by the use of a binary threshold by the fsQCA. In practical case studies, the GQCA significantly improves the reliability of configuration comparisons and the sensitivity of QCA to cases, demonstrating excellent research performance.