Non-price Effects and Market Definition in Digital Markets

Zhang, Junjun (2026) Non-price Effects and Market Definition in Digital Markets. Doctoral thesis, University of East Anglia.

Preview

PDF Download (7MB) | Preview

Abstract

This thesis investigates how mergers and acquisitions reshape outcomes of competition in digital markets, focusing on the non-price mechanisms. It combines basic theoretical modelling, structural estimation, and large-scale empirical causal analysis to study settings in which conventional price-based merger tools provide an incomplete account of competitive effects. Across three core chapters, the thesis shows that digital consolidation often operates through adjustments in perceived quality, innovation, and portfolio composition rather than through overt price movements. Since any assessment of merger and acquisition effects on competition depends on how the relevant market is defined, the thesis advances the literature by Natural Language Processing (NLP)-based approaches to digital market definition. The first core chapter analyses Amazon’s proposed acquisition of iRobot through a model of platform-driven quality allocation, in which a vertically integrated platform can selectively amplify perceived product quality via algorithmic rating and visibility contributions. Estimating a structural model from the United States (US) robotic vacuum cleaner market, it shows that even modest, algorithmically driven boosts to iRobot’s ratings post-merger could substantially increase both the acquired firm and platform profits, reallocate market shares, and alter firms’ incentives to invest in quality.

The second core chapter turns to the app economy and examines 1,387 Mergers and acquisitions (M&A) meticulously identified among US developers on the Google Play Store, using panel data on roughly 8 million apps and 1.8 million developers from 2015–2019. It documents that M&A reduce new app launches and external acquisitions while also limiting app removals and sales, leading to a net decline in product variety but an increase in maintenance updates and quality-enhancing activity on existing apps. The analysis further shows that acquisitions drive a sharper fall in new launches than mergers, and that horizontal and non-horizontal deals defined on product category exhibit distinct post-M&A strategies.

The third core chapter addresses two central questions in digital industrial organization: how to delineate relevant markets when user-side prices are often zero, and how mergers within those markets reshape non-price conduct. Using Google Play data for 2015–2019, it develops an NLP-based framework to construct ‘text-defined markets’ from app descriptions and update notes, computes concentration measures in these markets, and then studies 1,914 horizontal M&A events, identified by the presence of apps from both the acquirer and the target in the same market, using a staggered Difference-in-Differences (DiD) design with matched controls. The findings show that horizontal mergers that would be flagged by European Union (EU) or US screening thresholds tend to increase developer exit (‘pruning’) and modestly raise innovation, without clear price effects, while below-threshold horizontal and non-horizontal integrations mainly reallocate update effort and slightly raise prices.

Item Type:	Thesis (Doctoral)
Faculty \ School:	Faculty of Social Sciences > School of Economics
Depositing User:	Chris White
Date Deposited:	22 Jul 2026 10:09
Last Modified:	22 Jul 2026 10:09
URI:	https://ueaeprints.uea.ac.uk/id/eprint/103943
DOI:

Downloads

Actions (login required)

View Item