25 May 2018

Bank of England: Competition for retail deposits between commercial banks and non-bank operators: a two-sided platform analysis

This paper assesses competition between a bank and a non-bank operator running two-sided platforms that allow payments between consumers and merchants (besides providing their own deposit-storage facilities) under three configurations: exogenous single-homing, endogenous multi-homing and interoperability.

Retail banks’ mainstream business model, which is reliant on a stable supply of retail deposits, might come under threat as a result of the emergence of a new substitute for commercial banks’ personal and saving accounts that provides a safer money storage option thanks to access to a central bank’s balance sheet.

In line with the extant literature, when platforms are perceived as undifferentiated by agents on both sides, pricing rivalry is intense and only one platform can be active. The ability of the incumbent bank to fend off the competition threat from the non-bank operator is lower when multi-homing is an option. This is particularly the case when the distribution of deposits skewed towards a majority of depositors with relatively low balances. When consumers value the ability to make payments particularly strongly (i.e., compared to the remuneration of deposits), the incumbent bank is better off under interoperability. Under undifferentiated competition, the prevailing pricing structure whereby consumer access to payments systems is subsidised by merchants is maintained.

Alternatively, when consumers have split preferences regarding the non-bank new entrant, they end up being the ones who subsidise merchants when the option of multi-homing is available, that is, thanks to the fact that the latter are in principle indifferent between the two platforms. Indeed, the incumbent bank can be totally crowded out on the merchant side, thus being left with only a minority of very loyal customers who strongly dislike the new platform (e.g., perhaps because they are concerned about cyber risk). Therefore, the incumbent greatly prefers to compete under interoperability where the extent of merchant subsidisation is capped to zero transaction fees so that they all opt for multi-homing. Therefore, the incumbent can hold on to a larger number of consumers and both platforms can charge a higher membership fee thus making higher profits, which tends to suggest that the non-bank new entrant would be willing to cooperate on allowing interoperability.

In conclusion, perhaps counterintuitively, the risk that banks could be exposed to a large deposit outflow as a result of the entry of a new platform relying on access to the central bank’s balance sheet can be mitigated by the presence of demand-side frictions due to opposing brand preferences. These must be sufficiently high, though, as to neutralise the incumbency advantage due to the presence of network effects on the consumer side, thus weakening the tendency towards a winner-takes-all outcome that can beset competition among two-sided platforms. Furthermore, platform coexistence is facilitated when both platforms are willing to allow seamless payments across them, or when public intervention imposes interoperability in the absence of cooperation.

Full working paper

© Bank of England