Charges on UPI Transactions- Why it makes sense?

– Shreyas Sinha

Introduction

In mid-August, 2022, the Reserve Bank of India (‘RBI’) released a discussion paper on the various kinds of charges and fees levied on digital transactions (e.g., transactions through debit cards, etc.) and asked for stakeholder comments by October, 2022. The Central Bank, amongst other things, mooted the idea of whether transaction charges/fees ought to be introduced for transactions conducted through the Unified Payments Interface (‘UPI’), which have so far, due to heavy subsidisation by the Government of India, been free of these costs.

The RBI’s proposal has drawn mixed reactions. While the payments industry has reacted positively, commentators have argued that the transaction charges might disincentivise increasing adoption of UPI-based digital payment platforms amongst consumers. As of time of writing, UPI remains a widely popular payment system, having clocked a whopping 6.28 billion transactions in July, 2022.

This paper undertakes a mixed Law and Economics and regulatory theory analysis of the RBI’s proposal to argue that the move is, all things considered, beneficial for the payments sector and for consumers alike as it is Kaldor-Hicks efficient and conforms to principles of smart and responsive regulation. The paper proceeds by – first, undertaking a Kaldor-Hicks efficiency analysis of the RBI’s proposal; second, analysing the proposal vis-à-vis principles of smart and responsive regulation; and third, concluding.

Why The Transaction Charge is Efficient

Kaldor-Hicks efficiency, as a concept, is a matter of social choice between multiple outcomes. For example, assume that there exist a particular status quo X with benefits allocated between Group A and Group B in a particular proportion. A legal regulation would be Kaldor-Hicks efficient if the change it brings about results in a situation (situation Y) wherein Group A benefits vis-à-vis Group B but retains the ability to compensate the latter for its loss, whilst also retaining surplus benefits that leave it in a better off state than it was in situation X. Thus, Kaldor-Hicks efficiency, essentially, undertakes a utilitarian calculus and determines whether a particular legal action results in a net welfare gain to society.

But before delving into a Kaldor-Hicks analysis of the RBI’s proposal, it is pertinent to flesh out why exactly this framework is appropriate to qualitatively assess why transaction charges are good in the present case. The Kaldor-Hicks framework is, due to its origins in utilitarianism, a far more realistic model for assessing legal rules than, say, the Pareto efficiency model, which refers to an economic framework where resources cannot be reallocated to make one individual better off without making at least one individual worse off and which has been criticised by scholars for its absolutism and general inapplicability. It is intuitively logical to assume that any legal regulation would result in a shift from one outcome to another, leaving certain groups better off vis-à-vis other groups. The model’s wide acceptance in regulatory studies and in the framing of public policy is testimony to its relative superiority vis-à-vis other models. This is not to say that the model has no flaws; it does. However, it is still the closest model that can explain regulatory effects vis-à-vis society. So, having demonstrated why a Kaldor-Hicks efficiency analysis is suitable, one can now critique the RBI’s proposal using this model.

At the outset, let us assume that there are three principal stakeholders in this scenario – the payments sector (P), the Government (G), and the consumers (C). P includes entities like the banks (i.e., payer’s bank and beneficiary’s bank) and UPI app providers (e.g., Google Pay, Paytm, etc.). G includes the Government of India, the National Payments Corporation of India (‘NPCI’), and other public authorities. C simply refers to the end-users of the UPI service.

In the current scenario, C does not incur any costs for making a UPI transaction. The other stakeholders, however, incur a cumulative cost of INR 2 per UPI transaction (assuming the value of the average UPI transaction to be INR 800, which is categorised as a “low-value transaction” by the Government of India). As the RBI acknowledges in its discussion paper, facilitating UPI transactions “requires the PSO and banks to put in place adequate systems and processes to address the settlement risk”, thereby leading to significant costs. So, this scenario is inherently detrimental to P from a costs perspective. P is able to continue operations because public authorities, or G, provides it with subsidies and “incentives” to the tune of INR 1,300 crores (as of the 2021-22 fiscal year) so that the requisite systems keep functioning.

One may argue that this status quo is normatively desirable as it has led to the adoption of UPI-based payment systems on a mass-scale and given impetus to the Government’s goal of digitising payments across the economy, which reduces economic leakages. Furthermore, this money, which is being spent by G so as to enable P to continue UPI operations, comes from the public exchequer and is now not being spent on other socio-economically productive activities, thereby precluding the application of the multiplier effect in the larger economy.

Additionally, given that the amount of the subsidies is dependent on G’s budget for a particular fiscal year and that further costs need to be incurred in order to better provision of service and innovate new payment systems (including ones based off of the extant UPI system), P is disincentivised from undertaking investment activities, thereby stifling innovation in the sector.

A charge imposed on UPI transactions (which is over-and-above the value of a particular transaction itself) addresses this issue. By imposing a standard charge, the RBI effects a benefit transfer from C to P, which, by implication, obviates the need for continued subsidisation from G. So, if the proposed charge is indeed levied, C will become worse vis-à-vis P and G. However, this is not the complete picture.

Kaldor-Hicks efficiency is achieved in this outcome because P and G can, in principle, compensate C for the loss it has suffered as a result of the charge and still retain enough benefits such that there is a net welfare gain. For the purposes of this paper, ‘compensation’ is not conceptualised as direct, monetary compensation but as the creation of value, i.e., P and G can now utilise the money raised from the charge to invest in further innovation and betterment of service, which would have direct (e.g., more innovative payment services with a wider array of features and safety measures) and indirect benefits (e.g., better accounting of transactions and formalisation of payments in general) for C. Thus, barring some discontent with the charge in the short run, C would be better off in the long run as it would be the principal beneficiary of innovations in the UPI ecosystem. Finally, a charge on UPI transactions would incentivise further investment into the online payments sector as prospective investors or innovators would no longer think themselves to be dependent on government largesse.

Therefore, all things considered, an imposition of a charge on UPI transactions would lead to a Kaldor-Hicks efficient outcome and there would be a net welfare gain to society.

Regulatory Theory Justification for a Transaction Charge

A charge on UPI transactions also adheres to responsive regulation and smart regulation metrics. Before undertaking that analysis, however, it is important to flesh out what these terms mean.

‘Responsive regulation’, as a concept, essentially suggests that the regulation of any particular sector must be responsive to new trends in the sector, the general regulatory environment, and to the reaction of regulated entities to different regulatory approaches. Responsive regulation is dynamic and places ‘adaptability to changing circumstances’ as its principal normative value.

‘Smart regulation’ refers to regulation based on normative principles like flexibility, creativity, and innovativeness. Smart regulation often incorporates multiple single regulatory instruments to bring about comprehensive regulation, as opposed to the traditional bipartite mode of regulation involving the regulator and the regulated entity. One of the design principles of smart regulation is maximisation of opportunities for mutually beneficial outcomes between the various stakeholders of a regulated sector to encourage regulated entities to pursue innovation.

How does a charge on UPI transactions satisfy this two-pronged metric? The charge is in response to changing trends in the online payments sector. The RBI, in its discussion paper, takes note of the increasing acceptance of and reliance upon UPI-based payment systems for day-to-day transactions and its importance for the greater governmental effort towards digitisation. However this increased volume of transactions necessarily implies increased costs on banks (and the Government of India) as they have the ultimate responsibility of maintaining UPI-based systems. Furthermore, if the cost burden on banks and the Government is not alleviated, it would preclude the investment of financial resources in bettering system efficiency and innovation. Thus, the RBI’s proposal to impose a charge on UPI transactions is a direct and appropriate response to changing market conditions.

With regards to smart regulation, as this paper has already argued in Part II, the outcome achieved upon the imposition of a charge on UPI transactions would be Kaldor-Hicks efficient in that it would result in a net welfare gain to society. Of course, there would be some pushback from consumers in the short run. Indeed, there may even be a temporary dip in the volume of UPI-enabled transactions once a charge is in place. However, over the long run, new innovations in the online payments sector are developed (with consumers as their principal beneficiaries), transaction volumes will go up. Furthermore, the economic situation itself would be better off as there would be positive externalities from the development of new fintech innovations in the payments sector and the Government’s saving of the money otherwise being spent on subsidising costs related to UPI.

Conclusion

In conclusion, this paper has demonstrated that the RBI’s proposal to place a charge on UPI transactions is efficient and desirable from a regulatory standpoint. It would benefit from spurring innovation and relieving stakeholders of the burden that zero-charge transactions currently impose. However, this paper has not considered the actual modalities of the charge itself, i.e., whether it is mandated by the RBI as a flat fee or a percentage of the transaction value, or if it is to be market-determined, etc. Moreover, the argument does rest on some primary assumptions which warrant their own interrogation, such as – (1) the current UPI model does not adequately sustain fintech innovation; (2) the value created by levying a charge on UPI transactions would be, under a Kaldor-Hicks model, adequately transferred to end consumers; and (3) a UPI model that is Kaldor-Hicks efficient is normatively more desirable than the current model which treats UPI as a public good. In any case, these aspects do not render the principle-based argument of this paper untenable and Kaldor-Hicks efficiency, as an analytical model, still stands as a useful framework to look at UPI and attendant issues such as a service charge.

Shreyas Sinha is a law student at the National Law School of India University (NLSIU).