From Institutional Performance to System Design

These implications are addressed to the system rather than to any single actor. They do not prescribe specific interventions, but identify the structural changes in thinking, design, and governance that the evidence demands. They extend beyond Nigeria’s immediate context and are relevant to any financial system undergoing rapid digital transformation.

Implication 1Trust must be treated as a system design objective, not an institutional outcome

The traditional assumption in financial governance is that systemic confidence is built through the performance and integrity of individual institutions: that if each actor is well governed, financially sound, and consumer-oriented, public trust will follow. The findings challenge this assumption directly. Platform-led systems have fundamentally changed where accountability sits. When a transaction spans multiple actors and something goes wrong, no single institution is clearly responsible – and no single institution can fully remedy the failure. Trust at system level depends on the design of the shared arrangements between actors, not on the performance of any one of them.

Institutional soundness alone is no longer sufficient to generate systemic confidence. Trust must be deliberately designed into the system architecture through clear allocation of responsibility across multi-party transaction chains, reliable shared infrastructure, and governance frameworks capable of managing interdependencies across actors. This is not a refinement of existing approaches. It is a reorientation of the policy frame: from supervising institutions to designing systems.

Implication 2Operational reliability and recourse are primary drivers of confidence

Prevailing assumptions about the drivers of financial confidence tend to focus on innovation, product quality, and access expansion. The Executive Table evidence tells a different story. Transaction failure and dispute resolution ranked at the top of every breakdown point assessment. Data privacy, cybersecurity, and AI decision-making ranked at the bottom.

This should pivot the focus of policy from enabling innovation to ensuring reliability. The credibility of digital financial systems rests heavily on how they respond when transactions fail, disputes arise, or consumers are exposed to loss, much more than on what they deliver when things go well. Effective recourse mechanisms, clear dispute resolution pathways, and transparent communication during system failures are therefore central, not peripheral, to financial system design.

Implication 3Identity infrastructure is foundational to system integrity

Identity infrastructure emerged from the discussion as a critical enabler – and in its current state, a persistent inhibitor – of trust across multiple dimensions simultaneously. Fragmented KYC processes, inconsistent NIN implementation, and the inability to trace accounts to verified individuals enable fraud, weaken consumer recourse, and diffuse institutional accountability at the same time. No other single factor identified in the discussion operates across that breadth of system failure simultaneously.

Without a unified and interoperable identity layer, financial systems cannot reliably link transactions to verified individuals, enforce responsibility across institutions, or provide effective recourse to consumers. Identity must therefore be treated as shared infrastructure – a public good that underpins the integrity of the entire system – rather than as a compliance function managed independently by individual institutions. The policy challenge is not the creation of identity assets, which Nigeria already has in the BVN and NIN systems. It is their integration into a seamless, interoperable verification environment that the system as a whole can rely upon.

Implication 4Risk management must be embedded within system architecture

The increasing speed and complexity of digital financial systems – particularly through real-time payments and AI-enabled decision-making – have fundamentally altered the nature of risk. Risks now emerge at the velocity and scale of the system, frequently outpacing the capacity of traditional institution-level controls. Once funds move beyond recall in a real-time system, losses are realised immediately. Once an AI model embeds bias into a credit-scoring process, it scales at the rate of the model’s deployment. The reactive risk management model in which institutions identify, assess, and respond to risks after they emerge, is structurally misaligned with the systems it is meant to govern.

This reality requires a shift from reactive risk management to embedded risk design. Fraud prevention, anomaly detection, and governance mechanisms must be integrated directly into financial infrastructure from inception, not layered on top of it after deployment. In fast-moving systems, the question is no longer whether risks can be eliminated, but whether systems are designed to detect, absorb, and resolve them without undermining confidence. Risk architecture is therefore a founding design requirement, not an operational afterthought.

Implication 5Coordination is a structural requirement, not a voluntary choice

The analysis reveals a persistent and structurally significant coordination failure: institutions operate in shared environments, face shared threats, and depend on shared infrastructure – but respond to those threats individually. The Executive Table polling confirmed that the primary barriers to collaboration are not regulatory or legal, but relational and competitive. Institutional mistrust and competitive incentives to hoard information ranked significantly above legal constraints, regulatory uncertainty, and technological limitations.

Effective defence against fraud, cyber threats, and systemic risks requires coordinated responses across institutions. That coordination cannot depend on voluntary participation, which the evidence shows has been systematically undermined by competitive dynamics. It must be supported by formal mechanisms: shared intelligence platforms with binding participation, neutral convening structures with mandates that span all institution types, and regulatory frameworks that enable and where necessary require collective action. Nigeria’s own track record – BVN, NIBSS, GSI, NeFF – demonstrates that this is achievable. Coordination failures are not inevitable, they are design choices that can be recalibrated.

Implication 6Regulatory models must evolve from institutional oversight to system governance

Regulatory frameworks designed to supervise discrete institutions were developed for a financial architecture in which risk was primarily located within individual actors. Platform-led financial systems have changed that architecture fundamentally. Risk now arises from interactions between institutions, from the gaps, interfaces, and coordination failures between actors, rather than from the behaviour of any single one. A regulator monitoring each institution’s compliance and soundness in isolation cannot observe, let alone govern, the systemic risks that emerge from how those institutions interact.

Supervisory approaches must therefore evolve from monitoring institutional compliance to shaping system governance. This includes developing frameworks for cross-institutional accountability that assign responsibility when failures span multiple actors; strengthening regulatory capacity in emerging domains such as artificial intelligence, where supervisory understanding has not kept pace with deployment; and creating structured channels for ongoing dialogue between regulators and market participants that are substantive rather than compliance-oriented. The role of regulation shifts from controlling institutional behaviour to governing system dynamics: from asking whether each institution is sound to asking whether the system as a whole can be trusted.

Implication 7Cultural confidence is a policy variable, not a behavioural by-product

The persistence of cash and the relative trust commanded by informal financial providers (despite their offering fewer legal protections and less sophisticated products) highlight that confidence in financial systems is not determined solely by technical performance or regulatory compliance. It is also shaped by perception, experience, and the relational signals that institutions provide to users. Consumers judge financial systems not only on what they can do, but on how they communicate, how they respond when things go wrong, and whether they feel known and valued as participants rather than processed as customers.

This implies that cultural confidence is a policy variable: one that requires deliberate attention rather than passive expectation. Consumer education, transparency, responsiveness, and the design of user experience all contribute to whether individuals choose to engage with formal financial systems and whether they remain engaged over time. Financial inclusion is therefore a function not only of access, but of active usage and the confidence that sustains it. Systems that neglect the behavioural and perceptual dimensions of trust will continue to struggle to convert access into durable participation – regardless of how technically capable or formally compliant they become.

Implication 8Trust architecture requires a feedback mechanism to function

The Trust Resilience Score of 5.4 produced at the Executive Table was generated by a single closed-door session of thirty senior practitioners. It is a significant data point, precisely because it represents the candid collective assessment of the people closest to the system’s performance. But it is not a monitoring system. There is currently no standing, recurring, publicly available instrument that tracks systemic confidence in Nigeria’s digital financial ecosystem over time, across consumer segments, geographies, institution types, and financial literacy levels. Governance interventions are therefore made without a reliable feedback mechanism, and course correction currently depends on crisis rather than data.

Trust infrastructure such as shared infrastructure, accountability frameworks, and coordination mechanisms – the substance of the implications above – cannot be assessed for effectiveness without a recurring measure of whether systemic confidence is improving, stagnating, or deteriorating. This places trust measurement benchmarks within the core functions of financial system governance rather than a one-time diagnostic exercise.

This gap has now been recognised at the level of national strategy: the Payments System Vision 2028 commits the CBN to launching a National Payments Trust Index. The value of any such index will depend on whether it measures trust as a system-level, demand-side property or merely aggregates the institutional performance data already collected. The measurement problem this implication identifies is precisely the problem a national index will have to solve.

These eight implications do not list separate problems. They describe one structural problem seen from eight angles: the absence of an architectural approach to confidence in Nigeria’s digital financial system. Individually, each implication identifies a specific gap in how the system is currently designed and governed. Together, they define what a system-level approach to trust would require – what the recommendations in Section VI are designed to address.