Modelling the co-dependent diffusion of innovation in two-sided markets : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Marketing at Massey University, Albany Campus, New Zealand. EMBARGOED to 18 September 2024.
Loading...
Date
2023
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The Author
Abstract
With the advancement of technology, many innovations, like Electric Vehicles (EV) and contactless payment, are co-dependent. The diffusion of co-dependent innovation requires joint usage from more than one adopting group to enable functionality. For instance, EV owners will not drive their EVs unless they know that there are charging stations along their trips. Contactless payers will not pay with a “waving” or “tapping” of their contactless-enabled cards at the checkout unless they know that merchants accept this payment method.
Prior literature terms innovations that are compatible and can be used together as complementary innovations, and those adopted in sequence as contingent innovations (Peterson & Mahajan, 1978). Researchers build models of such innovations based on multi-product growth models or the hardware-software paradigm relying on the operation of the network effects (Bayus, 1987; Bucklin & Sengupta, 1993b; Stremersch et al., 2007). However, these terminologies fail to accurately describe co-dependent innovations, which require uptake by more than one adopting group and will only function with simultaneous use. When there are two distinct adopting groups, the market in which the innovation diffuses is a two-sided market. There is co-dependency between the adopting groups, and thus, between the diffusion path of each innovation. As the diffusion of these co-dependent innovations is yet to be modelled, the current study aims to fill this gap.
Using eight years of transaction-based data on a novel payment innovation in a developed western economy, we conceptualise co-dependent diffusion of innovation and examine its properties with three empirical studies. Results from Study 1 (presented in Chapter 2) demonstrate that prior models, including the multi-product Bass model, the model of indirect network effects, and the influx-outflow model proposed for a competitive two-sided market, fail to adequately depict the co-dependent diffusion of innovation. Building on findings from Study 1, Study 2 (presented in Chapter 3) shows that the Bass model with churn rates could be a promising candidate for modelling the co-dependent diffusion of innovation. In the payment innovation context, the churn rate represents the user dropout as a percentage of the current user base. Results reveal that merchants exhibit a higher churn rate than consumers, and the churn rates vary by industry. Simulated churn rates show opposite impacts on the innovation effect and the imitation effect in the diffusion process, where managerial implications can be drawn on tailoring strategies to different adopting groups based on the churn rates aiming to fuel the diffusion. Study 2 also highlights the potential of using churn rates as the proxy for the feedback effects between the adopting groups.
As the interaction effect between the adopting groups is established in Study 2, Study 3 (presented in Chapter 4) applies the Vector Error Correction Model (VECM) to account for consumer and merchant usage simultaneously. In the short term, consumer usage increases as a result of the variation in merchants’ usage. The positive response of consumers remains significant in the long run. On the contrary, merchants exhibit decreasing response to the variation in consumers’ usage; thus, only the immediate response is strong and significant in the short run. It is worth noting that, unlike the impact of marketing mix factors on sales that will die down over time, the variation of usage in one adopting group at the early stage of the diffusion could permanently lift the usage of the other group. This provides the first robust insights into the empirical patterns of co-dependency during the diffusion of innovation in two-sided markets and demonstrates how other such markets can be studied in the future. Managers can stimulate usage on one side of the market in the early stage of the innovation growth to leverage the interaction effect between the two sides. As emerged from the current work, an early push of usage on the merchant side may drive the co-dependent diffusion of the innovation in the long run.
Description
Embargoed until 18 September 2024
Keywords
Technological innovations, Economic aspects, Diffusion of innovations, Mathematical models, Multi-sided platform businesses