Financial projections are among the most important—and potentially most misleading—parts of a new bank licence application. A technically sophisticated model may show rapid growth, early profitability and comfortable capital and liquidity ratios. The central supervisory question, however, is not whether the spreadsheet balances. It is whether the proposed bank can plausibly generate the projected business, control the resulting risks and remain viable when actual outcomes differ from management’s plan.
Under the Basel Core Principles, the licensing process consists, at a minimum, of an assessment of the applicant’s ownership structure and governance, its strategic and operating plan, internal controls, risk management and projected financial condition, including its capital base (Basel Committee on Banking Supervision [BCBS], 2024, Principle 5). The assessment must therefore extend beyond numerical compliance to the credibility and sustainability of the institution as a whole.
A sound review can be organised around four connected areas:
- business-model risks and vulnerabilities;
- the basis of the assumptions, tested against independent benchmarks;
- prudential viability; and
- business-model-specific stress testing.
These elements should lead to a single integrated supervisory judgement rather than four disconnected technical reviews.
Business Model Review
A financial model cannot be assessed properly without first understanding how the proposed bank intends to make money. The reviewer should identify its target customers, products, distribution channels, revenue sources, funding structure, geographic reach and competitive proposition.
The business model should then be converted into measurable drivers. Interest income depends on asset volumes, product mix, pricing, repayment behaviour and origination timing. Fee income may depend on transactions, subscriptions, assets under management or trade volumes. Funding costs depend on deposit composition, customer sensitivity, market competition and pricing strategy.
Regulatory guidance for new banks emphasises that a business plan should explain how the institution will become profitable and self-sufficient, why this is plausible and what risks could prevent delivery of the plan (Bank of England, 2026a). The plan should also be internally coherent as a package: the financial projections, the ICAAP, the ILAAP and the recovery and wind-down material should rest on the same assumptions and the same growth path, and should have been subject to independent review and challenge before submission (Bank of England, 2026b).
Particular attention should be paid to vulnerabilities such as:
- exceptionally rapid customer or deposit acquisition;
- reliance on a few depositors, borrowers or commercial partners;
- dependence on shareholder or group-related business;
- uncommitted future capital injections;
- high fixed costs combined with uncertain revenue;
- aggressive lending before risk capabilities are established;
- dependence on critical technology, outsourcing or correspondent banks; and
- inadequate expenditure on compliance, cybersecurity and operational resilience.
The fourth of these deserves particular weight. Supervisory experience with new banks indicates that the loss or delay of expected external capital support is a frequent contributor to start-up business models becoming distressed and growth targets being missed (Prudential Regulation Authority [PRA], 2025, para. 4.3).
Projected growth should also be consistent with the operating model. A fivefold increase in customers or transactions without corresponding investment in staff, systems, controls and assurance functions is unlikely to be operationally credible.
Challenge the Assumptions and Benchmarks
Each material assumption should have a clear definition, an identifiable source, a documented rationale and a traceable relationship with the financial statements.
Customer and volume assumptions
Projected customer numbers, account balances, loan sizes, transaction volumes, conversion rates and retention should be reconciled with the target market and acquisition strategy. Claims that founders, shareholders or group companies can introduce customers should not automatically be treated as committed business.
The reviewer should distinguish among:
- broad market potential;
- expressions of interest;
- non-binding indications;
- contractual commitments; and
- business already capable of being transferred.
Pricing, margins and costs
Projected margins and fees should reflect customer quality, product risks and market competition. Net interest income is especially sensitive to the repricing assumptions embedded in the model. A common pattern is an asymmetric treatment of betas—assets assumed to reprice quickly and fully as rates rise, while deposit costs are assumed to lag—which flatters income in a rising-rate path and should be tested against a symmetric alternative and against a falling-rate path, where the same asymmetry works in reverse.
The cost base should cover not only technology and front-line staffing but also governance, risk management, compliance, internal audit, regulatory reporting, cybersecurity, insurance, professional services, premises and outsourced services. Costs required before commencement should not be deferred until after profitability has been achieved.
Credit losses and accounting integrity
Credit-loss assumptions should be consistent with the proposed underwriting standards, borrower segments, collateral, concentrations and economic environment. Where the applicant has no historical data, external evidence and conservative overlays may be necessary.
The reviewer should verify that the income statement, balance sheet and cash-flow forecast are fully integrated. Loan growth should generate funding requirements, interest income, expected credit losses, risk-weighted assets and liquidity effects. Profits should reconcile with retained earnings and capital.
Monthly projections are particularly important during the start-up phase. Annual figures may conceal temporary breaches arising from rapid growth, delayed capital injections or concentrated implementation expenditure. Supervisory guidance for applicants reflects this granularity: forecast start-up losses in the early years should be set out precisely, and the projections should run at least to the point of break-even (European Central Bank [ECB], 2018).
Independent benchmark
Management’s base case should not automatically become the supervisory base case. An independent benchmark should be developed from relevant peer data, market growth, product economics and realistic operating capacity. This mirrors the business-model analysis that supervisors apply to authorised institutions, where the assessment turns on the viability of the model over roughly a one-year horizon and its sustainability over a longer strategic horizon (European Banking Authority [EBA], 2026).
Benchmarking should cover customer growth, deposit acquisition, margins, credit-loss rates, control-function expenditure, productivity, cost-to-income ratios, profitability, funding composition and capital consumption.
The purpose is not to force an innovative applicant to replicate an existing bank. It is to identify where the application depends on exceptional performance. Material differences should be supported by demonstrable competitive advantages, committed customers, differentiated distribution or credible pricing power.
A useful comparison contains three paths:
- Applicant base case: management’s intended outcome.
- Supervisory benchmark case: independently assessed assumptions.
- Adverse case: material business-model and external risks.
A large difference between the applicant’s base case and the benchmark case is itself a significant finding, even where both remain above regulatory minima.
Assess Prudential Viability
Prudential viability is broader than compliance with minimum ratios. A proposed bank may remain above every numerical minimum in its base case and still have an unsustainable business model.
Profitability and capital
The reviewer should determine when the bank breaks even, the cumulative losses incurred before that point and whether initial capital can absorb those losses while preserving an appropriate buffer. A useful discipline is to ask whether the applicant would still meet its Pillar 1 requirement, its institution-specific Pillar 2 add-on and its buffers throughout the first year of trading on the supervisory benchmark case, not merely on management’s own case (PRA, 2025, para. 4.4).
The quality of profitability should also be examined. Recurring income should be distinguished from one-off fees, valuation gains, shareholder-related business and temporary interest-rate effects. Abrupt improvements in the final forecast years may indicate unexplained margin expansion, understated costs or unrealistically low credit losses.
Capital ratios should be independently recalculated. The review should cover capital eligibility, deductions, risk weights, off-balance-sheet exposures, provisioning effects and operational-risk requirements. Operational risk deserves specific attention in a start-up: because the standardised measure is driven by income and balance-sheet components the applicant has not yet generated, the requirement is itself a projection, and it grows automatically as the business plan is delivered.
The direction of the ratios is as important as their level. A bank may remain technically compliant while consuming its initial capital buffer more quickly than it can generate replacement capital. It is also worth noting that capital held to meet a start-up buffer is not calibrated on a stress test and may therefore be insufficient both to absorb a severe stress and to fund an orderly exit (PRA, 2025, para. 4.11).
The leverage ratio should provide a complementary view, because rapid growth in low-risk-weighted assets may appear acceptable under risk-based capital measures while materially expanding the bank’s total exposure. The Basel Committee has taken the view that a simple leverage ratio framework is critical and complementary to the risk-based capital framework, reinforcing risk-based requirements with a non-risk-based backstop (BCBS, 2014, paras. 2–3; see also BCBS, n.d.).
Liquidity and funding
Liquidity analysis should consider whether obligations can be met throughout the projection period—not merely at each year-end. It should assess depositor concentration, behavioural stability, maturity mismatches, collateral requirements, encumbrance, intraday needs and contingent outflows. Both the liquidity coverage ratio and the net stable funding ratio should be recalculated on a monthly basis during the growth phase, since a plan that builds long-dated assets faster than stable funding can satisfy the short-term measure while steadily eroding the structural one.
Deposits obtained through promotional pricing, intermediaries or digital channels may be more price-sensitive than established transactional deposits. Similarly, reliance on a small number of corporate or shareholder-related depositors may create simultaneous funding and concentration risks.
International liquidity principles emphasise the need for a liquid-asset cushion, severe stress scenarios, contingency funding arrangements and supervisory assessment tailored to the nature and complexity of the bank (BCBS, 2008).
Concentration and operational capacity
Concentration should be assessed by depositor, borrower, sector, geography, product, currency, correspondent bank, service provider and related party.
Correlations are particularly important. Depositors, borrowers and shareholders may be exposed to the same economic sector. One event could therefore cause deposit withdrawals, credit deterioration and reduced shareholder support simultaneously.
Financial viability also assumes that the proposed bank can safely deliver the projected scale. Inadequate governance, staffing, systems or controls can make growth itself a prudential vulnerability. Where a licensing regime provides for a restricted or mobilisation phase before unrestricted authorisation, the projections should distinguish clearly between the resources required to reach that gateway and those required to trade beyond it (ECB, 2019).
Business-Model-Specific Stress Tests
Stress testing should challenge the events that could cause the business plan to fail. A uniform percentage reduction in revenue is rarely sufficient.
A digital retail bank may need to test weak customer acquisition, deposit-rate competition, cyber disruption and accelerated withdrawals. A corporate or trade-finance bank may need to test borrower concentration, falling trade volumes, collateral deterioration and correspondent-bank disruption.
The Basel stress-testing principles provide that scenarios should cover a range of material risks and business areas, capture bank-specific vulnerabilities and be sufficiently severe, supported by appropriate governance, methodology, resources and documentation (BCBS, 2018).
The licensing stress test should contain four elements.
Base-case reconciliation
The reviewer should first reproduce the applicant’s base case. Failure to reproduce it may indicate errors, undocumented adjustments or weak model governance.
Sensitivity testing
Individual assumptions should be varied to identify the most influential drivers. These may include slower deposit growth, delayed lending, margin compression, higher staffing costs, increased credit losses or launch delays.
Integrated and reverse stress testing
Integrated scenarios should combine related shocks. For example, an economic downturn may reduce customer growth and fee income while increasing defaults, collateral haircuts and funding costs.
Reverse stress testing should begin with an outcome such as a capital breach, liquidity failure or exhaustion of available resources and identify the conditions that would produce it. This establishes whether failure requires an extreme event or only a modest departure from management’s assumptions.
Management actions
Stress results should be assessed before and after proposed management actions. Actions such as reducing growth, repricing products, cutting costs, raising capital or obtaining group support should be accepted only where they are specific, timely and feasible under stress.
Uncommitted capital, rapid asset sales at book value and immediate reductions in contractual lending should not be treated as reliable recovery actions.
Where recovery actions fail, the relevant question becomes whether the bank can leave the market in an orderly way while still solvent—repaying or transferring deposits and meeting the costs of doing so. Solvent exit is increasingly treated as a business-as-usual planning requirement rather than a late-stage contingency (PRA, 2024), and the projections should show that the resources needed to execute it would still exist at the point they were required.
Overall Assessment
The final assessment should answer five questions:
- Is the business model economically coherent?
- Are the material assumptions supported and internally consistent?
- Does independent benchmarking support the projected growth and profitability?
- Can the bank preserve adequate capital and liquidity while building the business?
- Can it survive plausible stresses, and exit in an orderly way, without relying on uncertain management actions?
The projection may then be classified as:
- Credible: assumptions are supported, resources match the planned scale, and vulnerabilities remain manageable.
- Credible subject to conditions: the model is viable but requires safeguards such as additional capital, growth limits, concentration limits or enhanced reporting.
- Materially deficient: assumptions are unsupported, documents are inconsistent or the model cannot be independently reproduced.
- Non-viable: realistic assumptions eliminate profitability, capital is insufficient, funding is unstable or modest stresses cause early failure.
In sum , Reviewing a new bank’s financial projections is not an attempt to predict its exact future size or profitability. It is an assessment of whether the applicant understands its economics, risks and resource needs and can withstand reasonable departures from its plan.
References
Bank of England. (2026a, February 2). Regulatory expectations. https://www.bankofengland.co.uk/prudential-regulation/new-bank-start-up-unit/regulatory-expectations
Bank of England. (2026b, March 13). Regulatory business plan. https://www.bankofengland.co.uk/prudential-regulation/new-bank-start-up-unit/regulatory-business-plan
Basel Committee on Banking Supervision. (n.d.). LEV10: Definitions and application. Bank for International Settlements. https://www.bis.org/basel_framework/chapter/LEV/10.htm
Basel Committee on Banking Supervision. (2008). Principles for sound liquidity risk management and supervision. Bank for International Settlements. https://www.bis.org/publ/bcbs144.htm
Basel Committee on Banking Supervision. (2014). Basel III leverage ratio framework and disclosure requirements. Bank for International Settlements. https://www.bis.org/publ/bcbs270.pdf
Basel Committee on Banking Supervision. (2018). Stress testing principles. Bank for International Settlements. https://www.bis.org/bcbs/publ/d450.htm
Basel Committee on Banking Supervision. (2024). Core principles for effective banking supervision. Bank for International Settlements. https://www.bis.org/bcbs/publ/d573.pdf
European Banking Authority. (2026, June 26). Guidelines on common procedures and methodologies for the supervisory review and evaluation process (SREP) and supervisory stress testing (EBA/GL/2026/06). https://www.eba.europa.eu/activities/single-rulebook/regulatory-activities/supervisory-review-and-evaluation-process-srep-and-pillar-2/guidelines-common-procedures-and-methodologies-supervisory-review-and-evaluation-process-srep-and
European Central Bank. (2018). Guide to assessments of fintech credit institution licence applications. https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.201803_guide_assessment_fintech_credit_inst_licensing.en.pdf
European Central Bank. (2019). Guide to assessments of licence applications: Licence applications in general (2nd rev. ed.). https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.201901_guide_assessment_credit_inst_licensing_appl.en.pdf
Prudential Regulation Authority. (2024, March 12). Solvent exit planning for non-systemic banks and building societies (Policy Statement PS5/24, incorporating Supervisory Statement SS2/24). Bank of England. https://www.bankofengland.co.uk/prudential-regulation/publication/2024/march/solvent-exit-planning-for-non-systemic-banks-and-building-societies-policy-statement
Prudential Regulation Authority. (2025, October). Non-systemic UK banks: The Prudential Regulation Authority’s approach to new and growing banks (Supervisory Statement SS3/21). Bank of England. https://www.bankofengland.co.uk/-/media/boe/files/prudential-regulation/supervisory-statement/2025/ss321-october-2025.pdf
Leave a Reply