2026 Modernising New Zealand’s retail payment system (RBNZ)

Reserve Bank of New Zealand (RBNZ) Future of Money Work Programme

Issues paper: Modernising New Zealand’s retail payment system (PDF, 811KB) Opened 18 Aug 2026. Closes 27 Oct 2026

TEXT AS SUBMITTED TO THE RBNZ.

PSGRNZ has deliberately not selected responses to the substantive tick-box questions in this consultation. Many of the propositions are superficially desirable, but selecting them would risk implying agreement with the assumptions, problem definition or direction of reform within which they have been framed. Our concern is precisely that the consultation does not provide respondents with sufficiently developed alternative institutional and infrastructure models against which those propositions can meaningfully be assessed. Our responses are therefore provided in narrative form. No inference of agreement, disagreement or consent should be drawn from an unmarked box.

At a time when modern research tools make broad international and interdisciplinary enquiry easier than ever, there is diminishing justification for consulting the public on nationally significant infrastructure before the institutional alternatives and contrary evidence have been seriously examined.

Q1. Do you agree with our description of the performance and capability gaps in New Zealand’s payment system?

1. Which of the following statements do you agree with? X none of the above.

Are there any other issues with how New Zealand's retail payment platform works that we should consider?

The RBNZ issues paper did not analyse the extent to which older EFTPOS systems might be built upon. Instead, the consultation asks respondents to agree or disagree with propositions such as:

some payments take hours or days rather than seconds;
people cannot easily pay using a phone number or email address;
limited information can accompany payments; and
sending money overseas is comparatively difficult.

These may be genuine limitations, but the Issues Paper does not provide the data necessary to understand their scale, frequency, economic significance or causes. Nor does it distinguish clearly between observable limitations in existing functionality and more consequential claims about restricted innovation, deteriorating resilience or risks to economic sovereignty.

The Issues Paper states that other countries are modernising their payment systems to combat fraud and scams. Yet it later acknowledges that a significant share of losses arises from authorised push payment (APP) scams and other forms of social engineering, which exploit behavioural vulnerabilities and information asymmetries rather than technical weaknesses in core payment infrastructure. Modern infrastructure may support improved safeguards, but this weakens any simple inference that infrastructure modernisation itself will meaningfully reduce fraud and scams. Wider data sharing and third-party connectivity may also introduce additional privacy, cyber-security, operational and governance risks that should be assessed alongside the proposed benefits.

RBNZ states that previous consultations, engagements and studies have ‘consistently identified common themes’, including the Council of Financial Regulators (CoFR) Payments Vision work. Yet the Issues Paper does not set out or critically assess that evidence before asking the public to endorse its diagnosis. Its international feature comparison is principally descriptive: it shows that selected countries possess capabilities New Zealand lacks, rather than demonstrating the consequences of their absence or whether adopting them would resolve the problems identified here.

The headline estimate of $0.7–$1.3 billion in annual benefits is labelled ‘Indicative RBNZ analysis’ and is based on an assumption that New Zealand could achieve performance gains comparable to estimates for UK account-to-account payment infrastructure. The Issues Paper does not provide sufficient methodology for respondents to assess whether those assumptions are transferable to New Zealand.

OPEN BANKING

Open Banking provides an important example of the missing analysis. RBNZ acknowledges that Open Banking has only recently been implemented, yet includes it among recent initiatives it characterises as having delivered improvements but limited structural or overall economic impact. It then concludes that Open Banking does not remove the need for system-wide back-end overhaul. Given how recently the regime has been implemented, this assessment warrants considerably more scrutiny.

In September 2025, the Commerce Commission described New Zealand as still ‘on the cusp of realising the benefits of open banking’.

It identified incumbent control, delayed implementation, API functionality, pricing, partnering arrangements and unequal access to payment channels as constraints on competition, while continuing to envisage Open Banking payment providers developing as domestic competitors to international scheme networks.

https://www.comcom.govt.nz/assets/Documents/retail-payment-system/Retail-Payment-System-Update-on-open-banking-progress-2-September-2025.pdf

The regulatory architecture itself is also still developing. Alade's 2026 legal analysis of New Zealand's new Consumer Data Right describes the CPD Act as providing a robust foundation for Open Banking, but identifies unresolved questions concerning informed consent, intermediary governance and accountability, allocation of liability between banks and Accredited Requestors, and the sustainability of API access arrangements.

These concerns cannot simply be attributed to technical limitations in the underlying payment rails. They concern the regulatory and institutional design of an emerging market.

Alade, Isa, The Customer and Product Data (CPD) Act 2025 and the Legal Limits of Open Banking in New Zealand (May 01, 2026). (2026) Victoria University of Wellington Law Review (forthcoming), Available at SSRN: https://ssrn.com/abstract=6923519

Indeed, as recently as March 2026 the Commerce Commission described the Open Banking market as being ‘in its infancy’. It continued to identify barriers to adoption and incomplete API coverage, and expected Kiwibank's entry during 2026 to expand coverage to approximately 90% of New Zealand bank accounts and substantially reduce competitive asymmetry. The Commission simultaneously supported consideration of modern open-access payment infrastructure sitting alongside existing rails, while continuing to pursue the development of Open Banking and address barriers to its adoption.

https://www.comcom.govt.nz/assets/Documents/retail-payment-system/Retail-Payment-System-Open-banking-update-letter-5-March-2026.pdf

This distinction is particularly important because RBNZ itself identifies incumbent control, access arrangements and misaligned incentives as significant constraints on competition. Those findings make it even more important to distinguish limitations caused by the technical infrastructure from those produced by its ownership, institutional governance and access arrangements.

It is therefore concerning that RBNZ does not adequately establish whether the present limitations of Open Banking arise from the underlying payment infrastructure, incumbent control, regulatory design, API access and pricing, incomplete implementation, or simply the immaturity of the regime. Before those limitations are relied upon in support of system-wide back-end overhaul, these competing explanations should be independently evaluated.

Otherwise there is a risk that the presently limited effects of a major statutory and competitive reform are used to support substantially more far-reaching infrastructure reform before New Zealand has established what Open Banking can achieve, why particular limitations remain, and whether they could be addressed through less far-reaching interventions.

The missing analysis is central to the public’s capacity to respond to this consultation: How often do these problems occur? Who is affected? How significant are they? What causes them? Which can be addressed through improvements already underway, and which genuinely require new infrastructure? Without this evidence, respondents cannot meaningfully assess either the diagnosis or the proportionality of the proposed response.

Nor does Ministerial endorsement of RBNZ leading this work resolve the separate institutional question of whether RBNZ is the appropriate agency to lead decisions extending beyond payment-system functionality into the governance of critical national infrastructure.

Finally, the repeated assertion that ‘the time to act is now’ should itself be supported by evidence. Urgency is a policy conclusion. If the existing system continues to perform its core function while substantial reforms, including Open Banking, SBI365, wider ESAS access and the Consumer Data Right, are still being implemented or maturing, RBNZ should demonstrate what material harm would result from allowing sufficient time to evaluate their effects, establish the remaining problems and independently scrutinise the alternatives.

A question that therefore warrants an answer is:

What analysis did RBNZ undertake of the Commerce Commission's 2023–2026 Open Banking work, including its findings on API pricing, incumbent behaviour, partnering, governance and access, in reaching its assessment of Open Banking's structural and economic impact and its conclusion that Open Banking did not remove the need for system-wide back-end overhaul?

Q2. Do you agree with our description of the strategic leadership gap and misaligned incentives in New Zealand’s payment system?

2. Which of the following statements do you agree with? X None of the above

Are there any other issues with leadership, decision-making, or incentives that we should consider?

QUESTION 2: THE ‘STRATEGIC LEADERSHIP GAP’

We do not agree that RBNZ has established a ‘strategic leadership gap’. The Issues Paper treats responsibilities distributed across different institutions principally as a problem of fragmentation, unclear leadership and difficult coordination. Yet it also acknowledges that this distribution reflects distinct public-policy objectives, including prudential soundness, market conduct, competition, infrastructure oversight, anti-money laundering and data sharing, and that separation of conduct and prudential regulation between FMA and RBNZ is consistent with New Zealand's ‘twin peaks’ model.

What is largely missing is an equivalent analysis of why these functions are separated, what checks and balances that separation provides, and what risks could arise from concentrating strategic leadership. New Zealand may need better coordination between appropriately separated institutions, rather than greater concentration of authority. The paper moves too quickly from the existence of multiple mandates to the proposition that stronger system-wide leadership is required, without adequately communicating the importance of a separation of powers.

RBNZ is well aware that the institutional landscape is already changing, but does not adequately address the governance and constitutional implications. It identifies the Consumer Data Right and open banking, the Digital Identity Services Trust Framework, wider ESAS access and Payments NZ's Next Generation Payments programme as recent reforms. CoFR developed its Payments Vision in 2023 and, RBNZ acknowledges, could provide cross-agency governance through targeted taskforces. Yet the consultation does not ask what these reforms have achieved, what has yet to mature, or whether better coordination, priority and resourcing could address the identified problems without further concentrating authority.

The international comparisons fail to address these broader issues. RBNZ concludes that modernisation is generally led or strongly steered by governments or central banks, but its comparison principally records who leads, builds, funds, settles and oversees. It does not systematically examine the safeguards surrounding those functions: separation of ownership and regulation, operator governance, incumbent conflicts, access arrangements, independent oversight, or the division of responsibility between central banks and other regulators. The comparison therefore demonstrates that public institutions participate in payment modernisation. It does not establish that concentrating strategic leadership in a central bank is the appropriate institutional model for New Zealand.

That distinction should have been central to Question 2. The public is being asked whether New Zealand has a leadership gap before being shown whether the underlying problem is leadership, coordination, resourcing, incumbent control or institutional design. Nor are respondents shown mature international alternatives demonstrating how ownership, operation, settlement, regulation, strategic direction and independent oversight can be deliberately separated.

Without that analysis, the consultation risks treating institutional plurality as a defect when it may also be an important public-law safeguard.

These issues are much ‘bigger’ than central banking. Questions of long-term governance and public power warrant an independent inquiry, not a central-bank-led process.

Q3. What benefits do you think payment modernisation can deliver? Are we missing any key benefits?

Which benefits are the most important for New Zealand’s retail payment system? Select all that apply. X None of the above.

Are there any other benefits that we should consider?

What's working well today and why? We want a retail payment system that is reliable, efficient and meets the needs of all New Zealanders. Please tell us which parts of the existing retail payment system are working well.

QUESTION 3: BENEFITS ARE PRESENTED WITHOUT THEIR CORRESPONDING RISKS

Unfortunately, the RBNZ’s framing is almost entirely couched in the affirmative. Respondents are told that modernisation could make payments ‘easier, safer and more efficient’ and generate around $1 billion in economic benefits each year. They are then invited to select benefits including competition, innovation, lower costs, speed, fraud protection, reliability and better international payments.

Therefore if we agree with the claims – we are not equally granted boxes to tick where we explain that there are costs and risks that must be considered. The RBNZ framing suggests that such an analysis has been conducted when this has not, to our understanding, taken place.

There is no corresponding question asking respondents to identify the costs, risks or adverse consequences of modernisation, even though RBNZ acknowledges elsewhere that the net outcome will depend upon choices about platform design and governance. The consultation therefore asks the public to consider the benefits of modernisation before the architecture capable of producing those benefits, and its corresponding risks, has been determined.

PSGRNZ are concerned Issues Paper places payment modernisation within a much wider technological trajectory and the RBNZ’s September proposal similarly takes this position. They anticipates infrastructure capable of supporting tokenisation, CBDCs, stablecoins and AI-enabled ‘agentic’ payments. Modern infrastructure can alter who operates essential infrastructure, who determines access, what information is generated and visible, how readily different systems can be connected, and how dependent New Zealand becomes upon particular technologies, standards and institutions. These are consequences of architecture, not merely technical details.

A credible consultation should therefore examine benefits and risks together. Faster payments may improve efficiency while common infrastructure increases dependency. Greater interoperability may improve convenience while increasing informational linkability. Stronger fraud controls may improve security while expanding monitoring capabilities. Common platforms may reduce costs while creating new infrastructural incumbents. The relevant question is not simply what modernisation can deliver, but what capabilities and dependencies different forms of modernisation create.

INTERCONNECTED DIGITAL INFRASTRUCTURE CREATES PUBLIC RISKS

A growing legal literature demonstrates that interconnected digital infrastructure raises governance questions extending well beyond efficiency. Boundaries between personal and non-personal data become porous; linkage and biometric identification can erode individual autonomy; third-party data flows strip away practical control; and meaningful avenues for redress may be limited or absent.

Pandey examines privacy, transparency, accountability and statutory authority as India's DPI expands; Sasi (2025) considers Aadhaar, UPI and data architecture as distinct but interoperable layers; Mishra and Agarwal (2026) examine governance of personal and non-personal data at scale; Lie, Austin, Sun and Qiu (2021) identify the ‘third-party problem’ created by increasingly complex information flows; and recent Malaysian scholarship (Musa, Shuaib & Zulhuda 2026). examines the tension between more efficient inter-agency data sharing and informational autonomy and state accountability.

Mishra N, Agarwal B, (2026) The emerging framework for non-personal data protection in India: perils, promises, and lessons for the developing world, International Journal of Law and Information Technology, 34:eaaf025, https://doi.org/10.1093/ijlit/eaaf025
Pandey, V. (2026) Governing the Stack: Constitutional Limits on India's Digital Public (March 12, 2026). Available at SSRN: https://ssrn.com/abstract=6595278
Lie D, Austin LM, Sun PYP and Qiu W, ‘Automating Accountability? Privacy Policies, Data Transparency, and the Third Party Problem’, In University of Toronto Law Journal, pp. e20200136, 2021.
Musa SS, Shuaib FS, & Zulhuda S. (2026). Data Sharing Act 2025 and the Constitutional Safeguards on Informational Autonomy in Malaysia. Journal of Information Systems and Digital Technologies, 8(1), 81–95. Retrieved from https://journals.iium.edu.my/kict/index.php/jisdt/article/view/693
Sasi A. (2025) Decoding the Indian data governance model: Relooking at Aadhaar. Social Sciences & Humanities Open, 11:101408, https://doi.org/10.1016/j.ssaho.2025.101408

These studies do not establish that New Zealand payment modernisation will produce these harms. They establish that these are foreseeable classes of risk that a serious options analysis should investigate. The question is not whether real-time payments inevitably produce surveillance, informational concentration or institutional centralisation. It is whether particular architectures for payments, identity, authentication and data interoperability create those capabilities, and what safeguards prevent their misuse.

Instead, this question could have asked:

What benefits, costs and risks could payment modernisation create for New Zealand, and how might these differ between alternative infrastructure and governance models? What new dependencies, concentrations of power or informational capabilities could each model create, and what safeguards would be required?

Q4. Which parts of the current payment system are functioning well, and what factors do you see as contributing to this performance?

Which parts of the current retail payment system are working well? Select all that apply. X None of the above.

Are there any other parts of the retail payment system that are working well and why?

QUESTION 4: WHAT IS ACTUALLY NOT WORKING?

RBNZ has established that the existing system has limitations; it has not established that it is performing badly. Nor has it demonstrated that replacing or substantially redesigning the existing infrastructure would produce greater net benefits than improving what already exists.

The current retail payment system appears to perform its fundamental function reasonably well. Payments are routinely made and received, the system is generally reliable, and RBNZ does not identify widespread failure in the core function of transferring money. The problems identified are principally limitations relative to newer systems: batch rather than real-time processing, restricted rich-data capability, limited cross-border interoperability, dependence on existing bank infrastructure, and declining competitiveness of domestic EFTPOS. RBNZ itself acknowledges that SBI365 has increased the speed and availability of the existing system, while open banking can provide significant end-user benefits. These limitations justify investigation and improvement, but do not establish the need for wholesale infrastructural replacement or major institutional redesign.

More importantly, the Issues Paper does not establish the counterfactual. It identifies costs and constraints associated with legacy infrastructure, but provides no comparable analysis demonstrating that a replacement platform would produce greater net benefits once construction, migration, cybersecurity, governance, compliance and ongoing operational costs are included.

Nor would a new common platform necessarily produce greater competition. RBNZ recognises that payment markets exhibit network effects, high fixed costs and economies of scale capable of producing concentrated market structures. Yet it does not carry that insight forward and ask whether new common infrastructure could itself become the next powerful incumbent. Modernisation may change the location of concentration rather than remove it.

NOR HAS RBNZ ESTABLISHED THE NEED TO HURRY

The Issues Paper repeatedly and emotively conveys a sense that New Zealand as falling behind its peers and insists that it is ‘time to act’. But technological lag is not evidence of urgency, and urgency should not be manufactured through rhetoric where the underlying evidence does not establish it. If the existing system continues to perform its basic functions while SBI365 and open banking are already delivering improvements, there is time to distinguish what genuinely needs replacing, what can be incrementally improved, and what should not be integrated at all.

Indeed, being behind presents an important opportunity. New Zealand can examine not only the benefits achieved elsewhere, but the problems that emerged afterwards: infrastructure consolidation, incumbent advantage, access restrictions, data linkage, informational autonomy and the adequacy of legal safeguards. Countries such as India and Malaysia are already generating legal and governance scholarship from which New Zealand can learn. RBNZ's international comparison makes remarkably little use of this advantage.

Decisions about common payment infrastructure can create long-lived dependencies, incumbent advantages and institutional powers that become difficult and expensive to reverse once participants, businesses and government systems depend upon them. Network effects can also entrench a small number of infrastructure providers, potentially reproducing at a new layer the concentration RBNZ is seeking to address. The appropriate response to technological change is therefore not necessarily speed.

Where payment infrastructure may become critical (also known as systemically important) national infrastructure, careful institutional design before implementation may be considerably cheaper than attempting to restore competition, separation or accountability afterwards. Being late gives New Zealand the opportunity not to repeat other countries' mistakes. We should use it.

We should not rush headlong into a system that might confine us and ultimately, mirror the dominance by a small range of institutions that we currently have.

The relevant comparison is not between an ‘old’ system and an imagined modern one. It is between the existing system, including improvements already underway, and credible alternative architectures whose costs, benefits, risks, institutional consequences and reversibility have actually been examined.

Q5. Given emerging technologies, what are the key considerations for a modern payment system in New Zealand to ensure it can remain adaptable into the future?

Which of the following will help New Zealand's retail payment system adapt to future needs? Select all that apply.

How the retail payment platform works / How the retail payment platform is run. X None of the above.

Are there any other considerations for the future of the retail payment system that we should take into account?

QUESTION 5: FUTURE-PROOFING WITHOUT ASKING WHAT SHOULD REMAIN SEPARATE

RBNZ asks whether the future payment platform should be interoperable, adaptable and information-rich, but does not ask what it should be interoperable with, what information should remain unlinked, which infrastructure layers should remain separate, or what limits should constrain the capabilities created by the wider digital architecture.

The options offered are all broadly desirable: capable operation, public-interest governance, openness, accountability, ongoing investment, upgradeability, speed, finality, richer information and interoperability. The problem is not these objectives themselves. It is that RBNZ has defined the future primarily as a better-performing payment platform, rather than asking what kind of wider digital architecture New Zealand may be constructing and what should remain deliberately separate from it.

That omission is particularly striking because the Issues Paper expressly introduces Digital Public Infrastructure. RBNZ cites the World Bank model of payments, data sharing, and identity and e-signature as ‘separable but interconnected’ building blocks, and notes the ‘multiplier effects’ from combining them. It then places New Zealand's open-banking and Consumer Data Right arrangements within the data-sharing layer and the Digital Identity Services Trust Framework within the identity layer. RBNZ therefore plainly recognises that payment infrastructure may sit within a much larger interoperable architecture. Yet Question 5 does not ask whether some of those layers should remain technically, informationally or institutionally separate.

India makes the omission particularly important. RBNZ presents UPI principally as a successful fast-payment system with public-private ownership and public strategic leadership. But UPI exists within a much wider digital-public-infrastructure ecosystem in which payment, Aadhaar-based identity and data-exchange capabilities have developed alongside one another. India should therefore be examined not simply as evidence that modern payment rails can work, but as an important institutional case study of what happens as payments become one layer of a broader digital stack.

Emerging legal scholarship raises precisely these wider questions. Pandey's 2026 analysis of India's expanding DPI is particularly instructive. He argues that the development of Aadhaar, UPI, DigiLocker and related infrastructure has outpaced the legal and institutional safeguards governing the stack, producing gaps in purpose limitation, mandatory linkage, third-party data sharing and system-wide accountability. Critically, responsibility remains distributed between institutions responsible for individual layers, without equivalent accountability for the cumulative architecture they create. Pandey argues that these gaps are constitutional in nature and cannot simply be resolved through technical fixes.

Sasi similarly examines the relationship between Aadhaar, UPI and India's wider digital architecture; Mishra and Agarwal examine the increasingly difficult boundary between personal and non-personal data and governance at scale; Lie et al. identify the ‘third-party problem’ that arises when interconnected digital services make meaningful transparency and accountability increasingly difficult; and Musa, Shuaib and Zulhuda examine informational autonomy and state accountability as government data-sharing expands.

These studies do not establish that payment modernisation necessarily produces harm. They demonstrate why the architecture matters: individually governed systems can become collectively powerful when identity, payments, authentication and data-sharing become interoperable. The resulting governance problem may exist between the layers, precisely where no single institution is responsible for the system as a whole.

Mishra N, Agarwal B, (2026) The emerging framework for non-personal data protection in India: perils, promises, and lessons for the developing world, International Journal of Law and Information Technology, 34:eaaf025, https://doi.org/10.1093/ijlit/eaaf025
Pandey, V. (2026) Governing the Stack: Constitutional Limits on India's Digital Public (March 12, 2026). Available at SSRN: https://ssrn.com/abstract=6595278
Musa SS, Shuaib FS, & Zulhuda S. (2026). Data Sharing Act 2025 and the Constitutional Safeguards on Informational Autonomy in Malaysia. Journal of Information Systems and Digital Technologies, 8(1), 81–95. Retrieved from https://journals.iium.edu.my/kict/index.php/jisdt/article/view/693
Sasi A. (2025) Decoding the Indian data governance model: Relooking at Aadhaar. Social Sciences & Humanities Open, 11:101408, https://doi.org/10.1016/j.ssaho.2025.101408

‘Interoperability’ cannot therefore be treated simply as a desirable capability. Interoperability with what, for what purpose, through which identifiers, and subject to what limits? Interoperability between banks is one thing; interoperability between payments, digital identity, government credentials and administrative data is another. Persistent identifiers, common authentication systems, APIs and data-sharing rules can increase the capacity to connect previously separate domains without any single institution formally owning the entire architecture.

This is not simply a privacy issue but extends to the architecture and distribution of power: who establishes identity, authenticates individuals, observes transactions, connects information across domains, determines access and authorises third parties, and what prevents capabilities created for one purpose being repurposed for another.

Lie et al. identify the resulting ‘third-party problem’: as information passes through increasingly complex networks of third parties, meaningful transparency and accountability deteriorate, leaving individuals with little practical ability to know where their data travels or how it is subsequently used.

Lie D, Austin LM, Sun PYP and Qiu W, ‘Automating Accountability? Privacy Policies, Data Transparency, and the Third Party Problem’, In University of Toronto Law Journal, pp. e20200136, 2021.

PSGRNZ highlight the troubling gap in RBNZ's framing. RBNZ describes DPI components as ‘separable but interconnected’, but Question 5 asks almost exclusively about the benefits of making them interconnected. It does not ask what should remain separable.

A more meaningful question would have been:

As payments become increasingly interoperable with digital identity, data-sharing and other public and private digital infrastructure, what capabilities should New Zealand enable, what functions and information should remain deliberately separate, and what legal, institutional and technical safeguards are required to prevent concentration of power, purpose creep and inappropriate cross-domain linkage?

Payment interoperability need not entail universal informational interoperability.

Q6. Do you agree with our principles for strategic leadership and reform? Are there any additional principles we should consider?

We would like your views on the principles that should guide any future improvements.

Which of the following statements do you agree with? Select all that apply. X None of the Above

System-wide leadership

One or more lead organisations should be able to:

Set the long-term direction for New Zealand’s retail payment system.

Make decisions that balance the needs of the retail payment system as a whole.

Work with government, regulators and industry to coordinate change.

Support goals such as resilience, competition, innovation, inclusion and economic sovereignty.

None of the above.

System-wide rules and oversight

A modern retail payment system should have:

Rules that are easy to understand and are inclusive, especially for new organisations.

Regulators working together to avoid gaps, overlaps and conflicting requirements.

A way of monitoring how the retail payment system is performing and how it can be improved.

X None of the above.

Platform level governance and decision-making

Governance should ensure that:

Consumer, business and public interests are represented.
Critical payment infrastructure is managed by organisations that can support long-term modernisation.
Decisions about who can access the platform, how it operates and where investment is made support competition, innovation and resilience.
X None of the above.

Are there any other principles we should consider when improving New Zealand's retail payment system?

RBNZ's principles are management principles designed primarily to make the future system governable. They are inadequate for defining the legal boundaries of infrastructure that may become part of a wider interoperable digital public architecture. They do not address what the future system should not become capable of doing. ‘Accountability’ cannot simply identify who runs the system; it must also address responsibility for the cumulative capabilities created between systems, institutions and third parties.

THE QUESTIONS DO NOT TEST THE PRINCIPLES

The survey questions compound the problem because they do not meaningfully test RBNZ's proposed principles. Respondents are asked whether lead organisations should set long-term direction, coordinate change and support resilience, competition and innovation; whether rules should be understandable and regulators should avoid gaps and conflicts; and whether consumer, business and public interests should be represented. These are overwhelmingly desirable propositions, with ‘none of the above’ effectively the only alternative.

Agreement therefore tells policymakers remarkably little. A respondent can support resilience, coordination and public-interest representation while strongly opposing concentration of strategic leadership, platform governance and regulatory influence. Agreeing that somebody should ‘set the long-term direction’ says nothing about who, what powers they should possess, what should remain outside their control, who independently oversees them, or which decisions properly belong to Ministers and Parliament.

The questions also blur important distinctions between leadership, ownership, operation, regulation, settlement and independent oversight. These are different functions, yet system-wide leadership is presented as an essentially desirable outcome rather than asking whether distributing these functions may itself provide important checks and balances.

The ‘system-wide rules and oversight’ questions are similarly weak. Avoiding regulatory ‘gaps, overlaps and conflicting requirements’ sounds unobjectionable, but some overlap may be intentional and protective. Privacy, competition, financial conduct, prudential stability and public accountability legitimately impose different constraints on the same infrastructure. The objective should not be regulatory neatness, but ensuring that no important public interest disappears between mandates and no institution can exercise significant infrastructural power while effectively supervising itself.

The platform-governance questions likewise ask respondents to endorse representation of consumer, business and public interests, capable infrastructure management, competition, innovation and resilience. They do not ask whether the infrastructure operator should be separate from the regulator; strategic direction from operational control; central-bank settlement from retail-platform ownership; or whether Parliament should establish boundaries that agencies and operators cannot subsequently alter.

Most importantly, the questions do not address the wider architecture RBNZ itself introduces. Its DPI discussion describes payments, data sharing, and identity and e-signature as ‘separable but interconnected’ building blocks. Yet respondents are not meaningfully asked what should remain separate. There is no equivalent consideration of limits on payment-identity linkage, persistent identifiers, third-party data flows, purpose creep, citizen visibility or circumstances in which interoperability should be constrained.

The consultation is therefore unusually one-directional. It asks whether respondents support desirable outcomes, rather than asking them to make the difficult institutional choices necessary to achieve those outcomes without creating unacceptable concentrations of power.

Once again, PSGRNZ cannot responsibly tick these boxes. Doing so risks implying prima facie agreement with propositions whose underlying risks have neither been identified nor properly parsed out.

A DIGITAL PUBLIC INFRASTRUCTURE ACT?

Principles of leadership, coherent regulation and public-interest governance cannot substitute for an overarching statutory framework if payments may increasingly interoperate with identity, credentials, data-sharing and other digital infrastructure. The prior question is whether New Zealand requires a Digital Public Infrastructure Act establishing the purposes, limits, institutional responsibilities and safeguards governing the architecture as a whole.

Such an Act could distribute responsibilities across RBNZ, the Commerce Commission, FMA, Privacy Commissioner and digital-government authorities; preserve separation between operation, regulation, settlement, identity and oversight; constrain persistent identifiers, cross-domain linkage, third-party access and function creep; provide for independent audit and citizen visibility; and require parliamentary authority for significant expansions in infrastructural capability.

The Act need not dictate a particular technical model. It would establish the constitutional and public-interest boundaries within which technical systems could evolve. This is only one possible approach. Yet the consultation provides no meaningful space to propose alternatives of this kind because its questions already embed significant presumptions about the problem, the direction of reform and the institutional response.

A meaningful consultation could instead have asked:

How should ownership, operation, settlement, strategic direction, regulation and independent oversight be distributed between institutions? Which functions should be legally required to remain separate? What limits should apply to interoperability between payments, identity and data infrastructure? Should these boundaries be established by Parliament through an overarching Digital Public Infrastructure Act rather than determined incrementally by the agencies and infrastructure operators developing the systems?

Q7. What do you see as the key trade-offs to consider for the future of the payment system?

QUESTION 7: THE TRADE-OFFS HAVE BEEN TOO NARROWLY FRAMED

This should be one of the most important parts of the consultation. Decisions about critical payment infrastructure inevitably involve trade-offs. Yet RBNZ frames these surprisingly narrowly and, in doing so, appears to build assumptions favouring centralisation into the analysis before the institutional alternatives have been properly examined.

RBNZ suggests trade-offs between competition and reliability, and between competition and efficiency, warning that ‘a fragmented settlement system with too many players defeats its purpose’. But this conflates different layers of the payment system. Competition in retail services, clearing, access arrangements or payment provision is not the same as fragmentation of final settlement. Central-bank settlement can remain concentrated to provide finality in central-bank money while ownership, operation, clearing, access, service provision and oversight remain plural or institutionally separate.

The relevant question is therefore which functions genuinely require centralisation, and which should remain competitive, redundant, independently governed or institutionally separate. The need for common settlement infrastructure does not establish a corresponding need to concentrate platform ownership, operation, strategic direction, technical standards, access decisions or regulatory oversight.

RBNZ also gives much greater attention to the risks of insufficient centralisation than to those of excessive centralisation. Common infrastructure can improve efficiency and reach, but it can also become indispensable infrastructure through network effects, switching costs and economies of scale. Competition at the consumer-facing layer can therefore coexist with increasing concentration underneath. Modernisation does not solve the incumbent problem if the modernised infrastructure itself becomes the new incumbent.

The treatment of resilience is similarly incomplete. Centralisation can strengthen settlement finality and common standards, but diversity, redundancy and independently operable alternatives are also sources of resilience. A highly integrated system may be efficient under normal conditions while increasing common-mode dependencies during cyberattack, technical failure or disruption to a critical third party. The relevant trade-offs therefore include efficiency versus redundancy, integration versus diversity, and common infrastructure versus independent alternatives.

The analysis becomes still narrower when payments are considered as part of wider digital infrastructure. RBNZ itself introduces payment services, data sharing, and identity and e-signature as ‘separable but interconnected’ building blocks. Greater interoperability can improve functionality, but it can also increase linkability, dependency and institutional capability. Relevant trade-offs therefore include interoperability versus informational separation, convenience versus privacy and autonomy, fraud detection versus surveillance capability, and integration versus institutional separation.

These are not simply additional benefits and costs to be optimised. Privacy, independent oversight, institutional separation and limits on infrastructural power may need to operate as boundaries and safeguards, rather than values that can simply be traded against efficiency or convenience.

RBNZ concludes that these trade-offs underscore the need for ‘strong leadership’. But difficult trade-offs establish the need for legitimate decision-making, expertise, transparency and accountability. They do not establish that greater concentration of leadership is the appropriate institutional response. In some cases, the appropriate response may instead be institutional separation, independent oversight and statutory limits established by Parliament.

The everyday functioning of democracy depends upon powers being visible, contestable and constrained by independent institutions capable of preventing overreach and abuse. In an age of interconnected digital infrastructure, where agreements, operations, information flows and decision-making can sit largely out of public sight across the public-private interface, these protections become more important, not less. Getting the institutional architecture right is therefore not simply another trade-off between competing policy objectives; it is fundamental to democratic governance.

A more meaningful question would have been:

Which functions within the future retail payment system genuinely require common or centralised infrastructure, and which should remain competitive, independently governed, redundant or institutionally separate? What trade-offs arise between efficiency and redundancy, interoperability and informational separation, convenience and privacy, common infrastructure and competition, international integration and economic sovereignty, and central coordination and independent oversight?

The central issue is not whether New Zealand should choose competition or centralisation. It is where centralisation is genuinely necessary, where it creates new risks, and where multiple providers, redundancy and institutional separation are themselves essential safeguards.

Q8. Do you agree with our key questions for considering potential options? Are there any additional questions at the system or platform level that you think are important?

Questions about the retail payment system / Questions about the retail payment platform. X None of the above.

QUESTION 8: THE OPTIONS ARE ALREADY NARROWLY FRAMED

Question 8 divides the future into the ‘retail payment system’ and the ‘retail payment platform’, but in both cases the questions are framed around identifying who should lead, regulate, control, operate and make decisions. This positions the problem as one of finding the appropriate leader or controlling organisation, rather than first asking how powers and responsibilities should be distributed and which functions should remain separate.

The framing gives little visibility to models based on multi-agency oversight, independent checks and balances, or deliberate separation of functions. The prior questions should be: which functions require coordination, which should remain separate, and who should independently oversee each? Only then does it make sense to ask who should operate the platform or exercise particular powers.

This is deeply consequential because payments are being considered alongside increasingly interoperable identity, data-sharing and other digital infrastructure. There is no obvious reason why strategic direction, infrastructure operation, access, competition, privacy, financial stability, identity governance and independent oversight should sit within a single institutional hierarchy. The architecture should determine the allocation of responsibilities; the search for a leader should not determine the architecture.

MULTI-AGENCY OVERSIGHT IS NOT A DEFECT

RBNZ repeatedly characterises distributed responsibility as fragmentation. Yet critical payment infrastructure intersects with financial stability, competition, consumer protection, privacy, digital identity, data governance and public accountability. These are distinct public interests, and there are good reasons for different institutions to protect them.

Different regulators may legitimately reach different conclusions because they are charged with protecting different statutory interests. That tension can be a safeguard rather than a failure. RBNZ itself acknowledges that existing responsibilities reflect distinct policy objectives and are broadly consistent with New Zealand's twin-peaks model.

Coordination may need improvement. That does not establish that authority should be concentrated. For infrastructure that could become essential to New Zealand, keeping responsibility spread across independent bodies may itself be an important safeguard. One statutory framework does not require one regulator; coordination does not require consolidation; and central-bank settlement does not require the same institution to lead, operate, regulate and oversee the wider architecture.

An overarching statutory framework could instead establish common duties and clear lines of accountability while distributing responsibilities across RBNZ, the Commerce Commission, FMA, Privacy Commissioner, digital-government authorities and Parliament.

THE INTERNATIONAL COMPARISON REINFORCES THE SAME FRAMING

RBNZ draws from Australia, the UK, Singapore and Sweden a set of ‘common themes’, including strong public-sector strategic leadership, an overarching legislative framework with a ‘clear lead regulator’, shared operational responsibilities between industry and central banks, and platform governance shared between the operator and central bank.

But this is precisely where a much richer institutional comparison was required. Who independently oversees the central bank? Who protects competition? Who supervises privacy and data use? What happens if infrastructure consolidates after implementation? What safeguards govern access? Which decisions remain with government or Parliament? How are payment, identity and data functions kept separate? What constrains third parties and function creep?

Instead, the comparison is largely confined to who leads, regulates, operates and funds. The RBNZ have elected to retain its enquiry to a narrow spectrum of institutional possibilities under consideration and risks circular reasoning: jurisdictions are examined through a framework centred on leadership and platform delivery, and the resulting ‘common themes’ are then used to support stronger leadership and more consolidated governance.

RBNZ's comparison also gives remarkably little attention to jurisdictions where central-bank functions are more deliberately bounded. Japan, for example, distinguishes the Bank of Japan's provision of final settlement and systemic oversight from privately operated retail clearing infrastructure. The Philippines similarly separates central-bank regulation and oversight from operation of its principal retail payment systems. These models do not remove the central bank from payments, nor should they. But they demonstrate the important institutional point that responsibility for settlement and systemic stability does not require the central bank also to assume strategic leadership, platform operation or control of the wider retail architecture.

The relevant international question is therefore not simply ‘who leads?’, but which powers have other democracies deliberately chosen not to give their central banks, and why.

It is troubling that the RBNZ says it is defining the problems, not yet presenting policy options. But problem definition, as we understand, is not neutral. By framing distributed authority as fragmentation and repeatedly directing respondents towards questions of leadership, control and consolidation, the consultation is already shaping the range of solutions the public is invited to consider.

The options analysis should go further:

Should New Zealand establish an overarching Digital Public Infrastructure Act defining the purposes, limits and safeguards applying across payment, identity and data infrastructure, while deliberately distributing operational, regulatory and oversight functions among independent institutions?

Q9 Do you have any feedback on criteria that we should use to assess future options to modernise the payment system?

What criteria should we use to assess future options for improving New Zealand's retail payment system?

Examples include:

value for money
competition and innovation
reliability and resilience
inclusion and accessibility
future readiness
how well it works with other payment systems in New Zealand and overseas.

QUESTION 9: THE EVALUATION HAS ALREADY BEGUN

RBNZ states that it is ‘not evaluating potential options at this stage’. While no detailed preferred model may yet have been formally assessed, substantial evaluative choices have already been made.

RBNZ has diagnosed an outdated system, a leadership gap and regulatory ‘fragmentation’; argued for stronger leadership; identified governance and leadership reform as a ‘minimum requirement’; and structured its options around increasing public leadership, regulatory powers and control. The problem definition, direction of travel and boundaries of the options space are therefore already being shaped.

The proposed criteria predominantly revolve around system performance: efficiency, innovation, competition, resilience, inclusion, interoperability and modernisation. These are important, but the criteria neglect the institutional conditions required to preserve those benefits over time, including institutional separation, independent oversight, privacy and informational autonomy, limits on infrastructural concentration, democratic accountability and preservation of independent alternatives.

Even apparently straightforward criteria such as ‘value for money’ require a longer-term public-interest perspective. People reasonably want payments to remain inexpensive, but they also need confidence that access to essential payment infrastructure cannot later be restricted because of lawful political beliefs or other activities unrelated to financial risk. Value for money should therefore be assessed over decades, not simply through present transaction costs or implementation efficiencies. Its preservation ultimately depends upon governance arrangements that constrain the abuse of infrastructural power, whether exercised by government or private actors.

There is also a consultation problem. The public is being asked to agree or disagree with RBNZ's diagnosis and principles before mature competing options have been presented. Agreement could later appear to support the emerging direction, while disagreement reveals little unless RBNZ investigates why people disagree. Opposition to ‘strong leadership’, for example, could mean support for multi-agency oversight, opposition to RBNZ leadership specifically, preference for parliamentary control, or rejection of the underlying diagnosis.

RBNZ may not yet be formally evaluating options, but it is already defining the problem, selecting the objectives, structuring the option space and seeking public agreement with important premises upon which the later evaluation will rest.

A better Question 9 would ask:

What alternative institutional and infrastructure models should be developed before options are narrowed, and what criteria should be used to assess their performance, concentration of power, independent oversight, privacy, democratic accountability and capacity to preserve resilience, access and competition over the long term?

PSGRNZ is a New Zealand Charitable Trust. No. 9 of our Charitable Objectives states this obligation:

To encourage scientist and physicians, as those members of our society who carry specialised knowledge in science and technology, to take up their responsibility to engage in public debate to serve the public’s right to quality independent information.

Download

Information