Journal entry testing with Benford's Law and a suspicious round-number cluster
You are auditing the procurement cycle of Taraz Construction JSC for 2025. The general ledger has 84,000 journal entries. You run data analytics: (a) Benford's Law first-digit test on amounts above KZT 500,000 shows a significant deviation (chi-square p < 0.01), driven by an excess of entries beginning with '9'; (b) 62 manual journal entries were posted on 31 December 2025 after 22:00 by a single user, all round amounts between KZT 900,000 and KZT 999,999, all debiting 'Construction materials' and crediting 'Accounts payable - other'; (c) the user's profile shows access to both vendor master data and journal posting. The CFO explains the entries as 'year-end accruals'.
1. Explain how you would investigate the 62 manual journal entries and what audit evidence you would obtain.
Read the worked answer
Why these entries are high-risk:
- Posted after hours on the last day of the year - classic fraud red flag.
- Round amounts just below KZT 1,000,000 may indicate an attempt to stay under an approval threshold.
- A single user with access to both vendor master data and journal posting is a segregation-of-duties failure.
- The credit to 'Accounts payable - other' rather than a named vendor is unusual.
Investigation steps:
- Extract the full population of the 62 entries with user ID, timestamp, amount, description, and approval trail.
- Trace each entry to supporting documentation (goods received notes, contracts, invoices, delivery notes).
- Confirm whether the 'accruals' relate to actual goods received before year end; if not, they are likely fictitious.
- Match the credits to vendor master records - check for new or altered vendors created by the same user.
- Review the user's access log and whether the entries were subsequently reversed in January 2026.
- Interview the CFO and the user; obtain written explanations.
- Extend testing to other periods and to the vendor master changes made by the same user.
Conclusion: If the entries are unsupported, they represent a material misstatement (overstated assets and liabilities) and possible fraud; the auditor should escalate to those charged with governance and consider the impact on the audit opinion.
2. Interpret the Benford's Law result and explain its limitations and how it should be used in the audit.
Read the worked answer
Interpretation:
- A chi-square p < 0.01 indicates the first-digit distribution deviates significantly from Benford's expected distribution. The excess of '9' entries is consistent with the 62 round-number entries (900,000–999,999).
- Benford's Law is a red-flag tool, not proof of fraud. Deviations can arise from legitimate causes: assigned thresholds, fixed-price contracts, or a genuinely skewed population.
Limitations:
- Benford's Law applies to naturally occurring data spanning several orders of magnitude; it is not valid for small populations, assigned numbers, or amounts constrained by policy.
- Round-number entries are expected to deviate and may be legitimate.
- The test does not identify which entries are wrong - it only prioritises testing.
Use in the audit:
- Use the result to focus substantive testing on the '9' cluster and on the 62 entries.
- Combine with other analytics (duplicate payments, weekend postings, vendor master changes).
- Document the rationale and conclusion; do not rely on Benford's Law alone as audit evidence.
Conclusion: The Benford result corroborates the need for detailed testing of the 62 entries but is not, by itself, evidence of misstatement.