Vehicle finance is the largest consumer credit category by rand value in South Africa. It is also the category most dependent on a single data source, the credit bureau score, for origination decisions. That reliance is understandable. Bureau scores are available in seconds, they correlate with long-term default risk, and they satisfy the minimum compliance signal that most F&I managers treat as the threshold for approval.
But a bureau score answers a specific, backward-looking question: has this consumer serviced registered credit reliably in the past? It does not answer the question that vehicle finance requires: can this consumer service a five-year, R250,000 instalment obligation starting today, given their current income and obligations? The gap between those two questions is where defaults originate.
The bureau score problem for vehicle finance
Credit bureaus in South Africa aggregate data submitted by registered credit providers on monthly reporting cycles. When a payment event occurs (whether a payment made, a payment missed, or a new credit account opened) that event typically takes 4–6 weeks to appear in the consumer's bureau record and affect their score.
The practical consequence: a consumer who was retrenched six weeks ago still shows their employment-period credit score today. A consumer who opened three short-term credit accounts in the past month, borrowing R15,000 in aggregate, may show no new obligations on bureau. A consumer who started missing their bond payment five weeks ago, a major financial stress signal, may not yet show any adverse listing.
For vehicle finance, where the loan term is 60–72 months and the rand value is significant, the 4–6 week visibility gap is material. An origination decision made on stale data carries the risk of the consumer's current state, not the state the bureau reflects. The bureau tells you who they were. It cannot tell you who they are.
Regulation 23A does not treat a bureau score as an affordability confirmation. The regulation requires verified gross income from payslips or bank statements, not from bureau data (which contains no income information at all). It requires enumeration of all current monthly obligations, including those not yet on bureau. It requires application of minimum expense norms keyed to income band. And it requires a documented calculation, not a score threshold, showing that discretionary income supports the proposed new obligation.
How stale bureau data creates origination risk
The scenarios where bureau staleness creates the most risk are predictable:
Recent retrenchment. A consumer with a stable R45,000 salary history and a 740 bureau score who was retrenched five weeks ago still shows that score. Their income is now zero or uncertain, perhaps a retrenchment package running down. The bureau provides no income signal at all; the bank statement shows the last salary credit was six weeks ago and deposits since then have been sporadic.
Rapid new debt accumulation. A consumer experiencing cash flow pressure will often take short-term credit from multiple providers in rapid succession (Boodle, Lime24, FairMoney, and store accounts) to bridge the gap. These new obligations take 4–6 weeks to appear on bureau. The bank statement shows the income credits from disbursements and the subsequent debit orders within days of origination. A consumer who appears to have no new debt on bureau today may have accumulated R8,000 in new monthly obligations in the past three weeks.
Informal debit load. Obligations to lenders who do not report to credit bureaus (unlicensed lenders, stokvel repayments, private individual loans) never appear on bureau regardless of how long they have been running. A consumer with R3,000 per month in informal loan repayments made as EFTs to private individuals has those obligations recorded nowhere except their own bank statement.
Income erosion. A consumer on commission income whose earnings have fallen 30% in the past two months shows nothing different on bureau. Bureau has no income data. The payslip from last month may show reduced earnings; the bank statement shows the trend.
What Reg 23A requires that bureau scores don't provide
Regulation 23A under the NCA specifies four distinct requirements for an affordability assessment. A bureau score satisfies none of them.
Verified gross income. Reg 23A(4) lists the acceptable verification documents: latest three payslips, three months of bank statements, or income statements. Bureau data contains no income information; it records credit account behaviour, not salary deposits. A credit provider who approves vehicle finance based solely on bureau data has not, on a plain reading of the regulation, verified income.
All current monthly obligations. The regulation requires "all monthly debt repayment obligations of the consumer," not "all obligations currently appearing on bureau." The distinction is not pedantic. Obligations from lenders who do not yet report (newly originated accounts), from lenders who report late, and from informal lenders who do not report at all are legally obligations under Reg 23A even though they are invisible on bureau.
Minimum living expense norms. The August 2025 draft amendments to Reg 23A introduce a mandatory minimum expense norms table. But even under the current regulation, applying no expense norms or applying figures below a reasonable minimum is a compliance failure. A vehicle finance originator who uses a flat R1,500 living expense estimate for all consumers regardless of income band is building a reckless lending risk into every assessment.
Documented discretionary income calculation. The credit file must contain a retrievable record showing the calculation: gross income, less tax and UIF, less obligations, less necessary expenses, equals disposable income. A bureau score and a dealership credit application form do not constitute this record. When the NCR inspects a vehicle finance provider's files, as it has done and will do again, the documentation question is concrete: show me the affordability assessment for this specific origination.
The NCT November 2024 vehicle finance case
Klusmann v FFS Finance South Africa (Absa VAF) [2024] ZANCT 50 is the recent NCT decision most directly relevant to vehicle finance affordability obligations. The case confirmed that the affordability assessment requirements under the NCA apply to vehicle finance originators, a clarification that matters because some F&I operations treat bureau-plus-income-declaration as a sufficient compliance methodology.
The case established that the substantive assessment obligation (not just a process compliance check) applies at origination. This aligns with the broader judicial direction on affordability assessment: courts are examining whether the credit provider actually assessed the consumer's ability to repay, not whether the provider went through the documentation motions. A vehicle finance originator who checked the bureau, received a signed income declaration, and approved the application has not necessarily conducted a substantive assessment of the kind the NCT is examining.
The NCR Consumer Credit Market Report tracks vehicle finance delinquency trends as part of the broader credit market picture. Vehicle finance default rates correlate with the origination assessment quality, specifically with the extent to which current-state income and obligation data was used rather than backward-looking bureau scores.
The bank statement advantage for vehicle finance
A three-month bank statement addresses the limitations of the bureau score directly, for vehicle finance specifically.
Income stability over time. Vehicle finance is a long-term commitment. A consumer whose salary has been consistent to within 5% for three consecutive months is a materially lower income-volatility risk than a consumer whose deposits have varied between R22,000 and R38,000. The statement makes this visible; no single payslip or bureau entry does.
Recent new obligations. New debit orders appearing in the most recent month (particularly to short-term lender reference patterns) flag recent debt accumulation before it reaches the bureau. A consumer with three new instalment debit orders in the past 30 days has a materially different obligation profile than the bureau reflects.
Informal debit orders. Regular EFTs to private individuals or cash withdrawals following a consistent pattern may represent informal loan repayments. These obligations exist regardless of their bureau visibility, and Reg 23A requires them to be included in the total monthly obligations figure.
Financial stress signals. High-frequency gambling transactions, chronic near-zero month-end balances, and repeated dishonoured debit order fees are behavioural signals that indicate financial stress not captured anywhere in a bureau score. Absa has publicly identified gambling spend as a "huge predictor of delinquency." That predictor is visible in the statement and invisible in the bureau.
The statement does not replace the bureau. It complements it. The bureau answers: has this consumer historically serviced registered credit? The statement answers: can this consumer afford a new R4,500 monthly instalment starting this month? Both questions matter for a vehicle finance decision. Neither answers the other.
Integrating statement analysis without slowing the deal
The practical objection to bank statement analysis in vehicle finance deal flow is speed. A dealership transaction is time-pressured. F&I managers work with consumers who have made an emotional purchase decision and are motivated to complete. Adding document uploads and processing time can kill deals.
That concern is legitimate for manual statement review, which takes 20–45 minutes per application and requires a trained analyst. It is not legitimate for automated statement analysis, where a digital PDF upload returns a structured result in under 30 seconds.
The practical workflow: the consumer's bank statement is requested as a digital PDF download from internet banking, the same request the F&I manager already makes for income verification. That PDF is uploaded to the analysis API. Within 30 seconds, the result returns: verified income, obligation schedule, Reg 23A-compliant affordability calculation, and a decision summary. The F&I manager adds this to the credit file alongside the bureau report and the credit application. Total additional time: under two minutes. Compliance achieved. Origination risk reduced.
AffyScore layers current-state behavioural data on top of bureau history for vehicle finance originators: recent income stability, new obligations not yet on bureau, and financial stress signals the bureau structurally cannot carry. The decision pack satisfies the Reg 23A documentation requirement at origination. For the comparison between what bureau scores provide and what statement analysis adds, the detailed breakdown is in our piece on bureau score versus statement score, currently being prepared and available soon.
