You file it every quarter. Has anyone read it?
Your PSD008 return tells the FCA, loan by loan, who you lent to, on what terms and through whom. Many lenders file it without ever analysing it. The FCA can analyse it without asking you a question. We analyse it first.
Why it matters
The FCA built these returns so it could “spot and stop harm faster” and “intervene more quickly and boldly where we have identified risk of harm in the consumer credit market”, relying less on ad hoc requests for information 62. After you submit, it runs its own data quality checks for consistency between your sales and performance returns, completeness and outliers, and may email you naming the transactions it has questions about 63. Any lender reporting £2m or more of consumer credit balances or new advances in its CCR003 returns is in scope 63.
Your return is built from your own systems, so it tells the truth about how you lend, whatever your policies say. Analysed properly, it also shows you where you spend money and where you lose it, the debt building up across your book, signs of fraud, customers with several loans at once, and errors in your systems. Better that you see it first.
What the return is
A consumer credit lender must send the FCA a sales data report for each reporting period, within 20 business days of its end (SUP 16.11.3R), electronically in the FCA's standard format (SUP 16.11.9R), containing all the information required (SUP 16.3.11R). It only counts as provided once every mandatory field is completed correctly and the FCA's system has accepted it (SUP 16.11.10G). Passing that check proves the file is well formed. It does not prove the lending it describes was sound.
Before you file
- Loans dated outside the reporting period, and loan numbers already sent in an earlier return
- Duplicate loan numbers, impossible or missing dates, and every loan against the high-cost short-term credit cost cap
- Customer references written the same way as in earlier returns, so returning customers still link up: the FCA's systems will not reject a change, but it breaks the link between your returns
- Codes against the FCA's data definitions, every field the analysis relies on filled in, and the file through RegData's test environment
- Whether last quarter's filing mistakes stayed fixed, and a signed pre-filing checklist for the senior manager who owns the return
What the return shows about your lending
- Repeat borrowing counted across quarters, not one quarter at a time: loans running at once, new loans given while earlier ones were still running, and chains of loans taken within weeks of each other (CONC 6.7.22G)
- Affordability on your own reported figures: repayment against spare income, whether income is recorded before or after tax, and whether existing loans ever reach the spending figure (CONC 5.2A.12R)
- Data that cannot be right: customer details that change on every repeat loan, two loans on the same day that disagree, identical loans paid out twice
- Brokers and direct lending: for each channel, which customers it brought, how many loans, the commission per loan, how often its customers borrowed again, how much spare income they had and how complete its data is
- Age, postcode area and loan size: where cover is thin and repeat borrowing concentrates, with care over protected characteristics
- Charges as a share of money lent, automated decisions, step-ups to bigger loans and rounded incomes
The fields we test
- Borrower reference (66A): the same reference for the same borrower across back-book, sales and performance reports, as the FCA's data definitions expect
- Net or gross income (72A), income used and declared (73A, 75A), expenditure used and declared (74A, 76A): if both net and gross income are held, net should be reported
- The expenditure breakdown (77A, 77B): whether existing credit commitments, including your own loans, reach the figure you lent on
- Whether the creditworthiness assessment included a manual underwriting check (67A), and what income and expenditure information you hold (68A)
- How the sale was made (21A), whether a credit broker introduced the borrower (23A), the broker's FRN and name (24A, 25A) and the commission paid (26A)
- Dates (31A, 33A), amount, charges and repayment (99A, 107A, 126A to 128A), agreement type (51A) and the APR against the representative APR in your promotions (16A, 103A)
What you receive
- A plain-English report: each finding, the loans behind it, the rule or definition it relates to, and whether it is new, better or worse than last quarter
- A fix list: the problem, the fix, how you will know it worked, and which past customers to look back at
- Every figure calculated twice, by our checking tool and by a separate calculation, so the numbers stand up to challenge
- A finding means “look at these loans”: only a person reviewing the loans can decide whether a rule was broken
Questions
Short answers, with the provision each one rests on.
What is PSD008?
PSD008 is the FCA's product sales data return for consumer credit lending. A lender reports each loan made in the period within 20 business days of its end (SUP 16.11.3R), in the FCA's standard electronic format (SUP 16.11.9R).
Our return passes the FCA's checks. Why analyse it?
Passing the FCA's format check shows the file is well formed, not that the lending is sound. A return can pass every check and still show customers with several loans running at once, income recorded on the wrong basis, or the same customer paid two identical loans on one day.
Can you analyse returns we have already filed?
Yes. Send the filed files under NDA. Several quarters together show what one quarter hides: loans still running from earlier quarters, chains of repeat loans and problems that recur in every return.
Will the report show each broker?
Yes. The report is yours, so it shows every channel: how many loans each broker sent, the commission per loan, how often its customers borrowed again, their spare income and the quality of their data. Nothing is published.
Talk to someone who has done your job.
A 30-minute call, no deck. Bring the problem; we will tell you what we would do.