The Deposit Takers Act 2023: RBNZ’s New Prudential Regime and Its Tech and Data Demands

The Deposit Takers Act itself is short on the things that will consume your technology function. The demands live in the standards underneath it, and those are still landing.

Most of what has been written about New Zealand’s Deposit Takers Act 2023 is a legal summary: it consolidates two old regimes, it introduces depositor protection, it hands the Reserve Bank of New Zealand a broader supervisory toolkit. All true, and all beside the point for the people who will actually have to deliver against it. The Act is a frame. The work sits in the prudential standards RBNZ is issuing beneath it, and in the reporting those standards assume you can already produce. This is a reading of the regime as a technology and data problem, for the deposit takers who have to build the evidence before a supervisor asks for it.

What actually changed

The DTA replaces the Banking (Prudential Supervision) Act 1989 and the Non-bank Deposit Takers Act 2013 with a single regime covering both registered banks and the former non-bank deposit takers. That consolidation matters technically because it removes the old distinction that let smaller institutions operate under lighter, separate machinery. Everyone is now inside one prudential perimeter, with proportionality applied by size rather than by category.

Free · 4 minutes

Do you know where AI is already being used in your business — and what it can see?

Fourteen questions on shadow AI, data exposure, oversight, and governance debt — the gap between how fast AI is arriving and how much control you have over it. Banded finding on screen, full sheet by email.

RBNZ’s proportionality framework sorts locally incorporated deposit takers into three groups by total assets: Group 1 for the largest, systemically important institutions, Group 2 for the mid-tier, and Group 3 for the smallest. The thresholds proposed sit at roughly NZ$100 billion and NZ$2 billion, but treat the exact boundaries as provisional and check them against the final standards, because the numbers have moved through consultation. The principle is stable even if the figures are not: a Group 3 institution carries a lighter documentation and stress-testing load, but it does not escape the substance. Where a large bank needs depth, a small one still needs the capability to exist.

The timeline that should set your sequencing

The dates are the part worth internalising, because they are what reward early preparation. RBNZ has been consulting on exposure drafts of the standards in tranches since late 2025. On the current published plan, all DTA standards are to be issued by 31 May 2027, the licensing window runs from 1 June 2027 to 30 November 2028, and the standards commence on 1 December 2028 — with crisis-related standards on a later track. The Depositor Compensation Scheme, by contrast, is already live: it commenced on 1 July 2025 and protects eligible depositors up to NZ$100,000 per depositor, per institution.

Read that as an engineering schedule rather than a legal one. The gap between standards being issued in 2027 and commencing at the end of 2028 is not slack — it is the window in which you are expected to have built and evidenced the capability. Firms that wait for the final text before starting will be building data plumbing under a licensing clock, which is the most expensive way to do it. The core standards, the ones RBNZ uses to assess eligibility for relicensing, are the ones to sequence first.

Operational resilience becomes a standard, not a slogan

The proposed Operational Resilience Standard is where the DTA stops being a capital conversation and becomes a technology one. As drafted, it reaches into operational risk management, oversight of material service providers, ICT governance, and business continuity, built around the idea of “critical operations” that must keep running through disruption. If that shape looks familiar to anyone who has read Europe’s DORA or the UK’s operational-resilience rules, it should. New Zealand is arriving at the same destination by its own route: the regulator wants to see that you have identified what is critical, mapped the systems and suppliers those operations depend on, set tolerances for disruption, and can show the arrangements actually hold.

The recurring weak point will be the supplier map. Cloud hosting is usually captured. The harder dependencies — payment switches, core-banking providers, identity services, the smaller integrations nobody classified as critical — are where the gaps sit. A resilience standard that turns on material service providers means your third-party inventory has to be complete and current, not a spreadsheet last touched at onboarding. The right to understand and, if needed, move away from a critical provider is now a resilience control, not merely a commercial preference, and the same exit and functional-equivalence questions raised by the EU Data Act apply directly here.

The reporting the DTA quietly demands

The obligation that will surprise the most institutions is not a headline standard at all. It is the depositor data the regime assumes you can produce on demand. Because the DCS pays out per depositor, RBNZ needs institutions to be able to resolve their deposit book to the level of the individual customer: direct insured balances, uninsured balances, look-through and trust accounts, and reconciliation against a Single Depositor View. Under Open Bank Resolution pre-positioning, that is not a report you run once a year — it is a capability the resolution machinery relies on being accurate at a moment’s notice.

For many deposit takers, particularly those built on older core systems or a patchwork of acquired products, this is genuinely hard. Customer identity is fragmented across product silos. The same person appears as three account holders with no reliable key linking them. Trust and nominee structures obscure the beneficial holder the scheme cares about. Producing a clean Single Depositor View is a data-lineage and entity-resolution problem before it is a reporting one, and it is exactly the kind of latent data debt that stays invisible until a regulator sets a deadline against it.

What to build first

Sequence the work against the phasing, not against the final text. Start with the depositor data model, because it underpins DCS reporting that is already live and OBR pre-positioning that follows, and because entity resolution is slow to get right. In parallel, complete and maintain the material-service-provider inventory the resilience standard will test, since you cannot evidence resilience over dependencies you have not catalogued. Then build the evidence trail — the artefacts a supervisor will ask to see — rather than the policy documents alone. As with any standards-based regime, what matters is not the framework on the shelf but what your technology function can actually demonstrate on the day of the visit.

None of this is unique to New Zealand, and that is the useful part: the same data foundations serve the country’s parallel push toward open banking under the CDR designation. A deposit taker that can resolve its customers cleanly, map its critical dependencies, and produce evidence on demand is prepared for more than the DTA. The institutions that struggle in 2028 will not be the ones that misread a standard. They will be the ones that could not tell the regulator who their depositors actually are.

Free interactive tool

Website compliance checklist

What your site has to do, based on what it actually does

Answer as much or as little as you like — the list builds as you go. Nothing is stored against your name and no email is required.

Free interactive tool

Interactive deadline calculator

Check which regulations apply to you and when

Regulation across the EU, UK, US and Asia-Pacific has moved considerably in the past eighteen months, and several headline dates have shifted more than once. Twelve questions, about three minutes.

Results are shown on screen — no email required. A dated summary is available to download, and can be sent on if that's more useful. What we do with your answers.

Most technology problems are not technology problems. They are control problems.

The systems exist. The investment has been made. The question is whether leadership can understand, direct, evidence, and sustain what those systems produce. Find out where control exists — and where it only appears to.

Full Governance by Sixteen Pillars

Govern your business. Prove your compliance.

A board assurance cockpit for EU-regulated financial firms — tamper-evident, hash-chained proof of governance across DORA, GDPR, NIS2, ISO 27001, the EU AI Act and MiCA. In development.

See what's coming