A recent working paper from the Bank of England sheds light on the complex dynamics between traditional banks and non-bank lenders in the UK’s unsecured lending market for small and medium-sized enterprises (SMEs).
Authored by Negar Mohammadi Jazi and Felipe Netto, Staff Working Paper No. 1,191, published on 19 June 2026, uses detailed loan-level data and a sophisticated structural model to examine the interplay of risk-based capital requirements, information gaps, and market competition.
SMEs play a vital role in the economy but often struggle to secure financing due to limited transparent financial histories, which creates significant information asymmetries.
Lenders must invest considerable effort to evaluate creditworthiness accurately.
Banks have long dominated this space, but non-bank lenders have gained ground by offering alternative approaches to assessment and funding.
A key distinction lies in regulation: banks face risk-weighted capital requirements that tie lending costs to perceived borrower risk, while non-banks operate with greater flexibility but typically higher funding expenses.
The researchers analyzed confidential data from major UK lenders and identified several important patterns.
Lenders appear to rely heavily on nuanced, non-public information when setting rates, as evidenced by wide variations in pricing even after accounting for observable firm characteristics.
Higher rates, after controlling for known factors, reliably predict elevated default probabilitiesāa sign of effective risk differentiation.
Non-banks tend to serve riskier borrowers overall and apply higher interest rates, reflecting their different cost bases and capabilities.
A particularly striking finding concerns loan sizes.
Larger loans correlate with higher default risks, even after adjustments for visible traits.
Banks actively participate in smaller loan segments but pull back sharply on bigger unsecured facilities.
When banks do extend larger loans, defaults are notably lower than those seen in non-bank portfolios.
Banks also show greater sensitivity in pricing to hidden risk factors for substantial exposures, suggesting regulatory pressures encourage caution in higher-stakes lending.
To unpack these observations, the authors built and estimated a structural model incorporating borrower heterogeneity in observable and unobservable risk types, endogenous loan demand, and lender-specific screening technologies and costs.
Borrowers select from personalized rate offers based on each lender’s risk assessment, which combines public signals (like requested loan amount) with private screening results.
Banks‘ costs explicitly include balance-sheet constraints from capital rules, which become more burdensome for riskier assets.
Estimation results reveal meaningful differences.
Demand proves relatively elastic, with borrowers showing greater sensitivity to non-bank rate changes.
Non-banks incur higher baseline funding costs, but their superior screening precision helps them navigate information challenges effectively.
For banks, regulatory capital considerations represent a substantial portion of marginal costsāaround half for lower-risk casesādirectly influencing pricing and allocation decisions.
Overall, the study demonstrates that capital regulation does not operate in isolation; it interacts with lenders’ varying abilities to process information and manage costs.
Non-bank participation stems not solely from lighter oversight but also from technological edges in screening.
This framework offers policymakers a valuable tool for assessing how changes in rules might affect credit availability, pricing, and the mix of funding sources for SMEs.
As regulators continue refining frameworks for financial stability, insights like these underscore the need to consider competitive spillovers and information frictions in SME markets. The paper provides a quantitative lens for evaluating policies in mixed regulatory environments.