AUD · Assessing Risk and Developing a Planned Response · 10 practice questions
COSO Inherent Limitations: Management Override and Collusion
COSO provides reasonable, not absolute, assurance because collusion and management override can defeat well-designed controls. Below, the same vendor payment shows how the cause of an approval bypass changes the conclusion.
Try one first
Hint
Focus on what COSO says internal control can and cannot realistically achieve, even when the framework is well designed.
Answer C. COSO-based internal control is intended to provide reasonable, not absolute, assurance. Inherent limitations such as human error, collusion, and management override mean that even well-designed and implemented controls may not prevent or detect every material misstatement.
Why not A: This overstates control effectiveness; even when duties are properly assigned and followed, fraud can still occur through collusion or management override, and human error can persist.
Why not B: This choice is tempting because the five COSO components form a comprehensive framework, but COSO explicitly recognizes inherent limitations and does not guarantee absolute assurance.
Why not D: This confuses practicality with COSO requirements. COSO does not require senior management to review every transaction; effective control relies on segregation of duties, monitoring, and other procedures, and the key issue is inherent limitations, not lack of per-transaction senior approval.
Same scenario, one fact changes
Base case
Lake Co. segregates purchasing and accounts payable duties. Its system requires independent vendor approval, and monthly monitoring reviews payments. Without management involvement, clerks from the two functions secretly cooperate to bypass approval and cause a $20,000 fictitious-vendor payment. Monthly monitoring detects the payment one month later.
Answer: Collusion, an inherent limitation; COSO provides reasonable, not absolute, assurance.
The clerks jointly defeated the approval control. Their cooperation does not automatically prove poor design, a missing control environment, or ineffective monitoring.
Before you open each one, predict the answer.
Change 1The CEO directs the approval bypass instead of the clerks secretly arranging it.
Answer: Management override, an inherent limitation; reassess the control environment and fraud risk.
The CEO intentionally circumvents approval. This raises management-integrity concerns and calls for reassessing control reliance and the nature and extent of substantive procedures.
Change 2A software-patch failure permits the payment instead of the clerks colluding to bypass approval.
Answer: A correctable control-operation failure, not employee collusion or management override.
The required approval workflow failed to operate after a patch. Correcting and testing the workflow addresses this failure.
Change 3Monthly monitoring detects the payment the same day rather than one month later.
Answer: Collusion remains an inherent limitation despite prompt detection.
Detection timing does not change the cause. Prompt detection also shows why the fraud does not, by itself, prove monitoring was ineffective.
Key points
- Remember: Ask what defeated approval: employee cooperation, management direction, or a failed workflow.
- Employee collusion does not automatically mean the control environment is missing.
- Senior management override raises concerns about management integrity, fraud risk, and control reliance.
- Detection timing does not change whether the payment resulted from collusion.
How the exam traps you
- Any successful fraud proves the controls were poorly designed. Identify the bypass mechanism; collusion can defeat otherwise well-designed approvals.
- CEO approval makes the fictitious-vendor payment acceptable. A CEO's direction to bypass approval is management override, not a valid control.
- Monitoring must prevent every improper payment. Monitoring can detect the scheme without eliminating the underlying collusion risk.
Now the same facts as questions
Each question changes one fact from the one before. Watch which change flips the answer.
Question 1
Answer C. Right. Employee cooperation can bypass controls, so COSO provides reasonable rather than absolute assurance.
Why not A: Wrong. Collusion can defeat properly designed segregation and approval controls.
Why not B: Wrong. Management was not involved; the clerks secretly cooperated.
Why not D: Wrong. Monitoring need not prevent every fraud, and the review detected this payment.
Question 2
Answer A. Right. Senior management intentionally circumvented approval, raising control-environment and fraud-risk concerns.
Why not B: Wrong. CEO authorization does not make a fictitious-vendor payment valid.
Why not C: Wrong. The defining fact is the CEO's direction to bypass controls.
Why not D: Wrong. Later detection does not erase management override.
Question 3
Answer D. Right. The patch caused an existing control to fail to operate.
Why not A: Wrong. This workflow failure can be corrected; not every control failure is an inherent limitation.
Why not B: Wrong. No manager intentionally directed the bypass.
Why not C: Wrong. Independent vendor approval was required, but the workflow failed to enforce it.
Question 4
Answer B. Right. The clerks still colluded; faster detection does not change the bypass mechanism.
Why not A: Wrong. Detection does not establish management involvement.
Why not C: Wrong. Detecting this payment does not guarantee prevention or detection of every material misstatement.
Why not D: Wrong. Collusion does not automatically prove the control environment was absent.
Question 2
Hint
Ask whether the condition could be fully corrected by redesign, staffing, training, or monitoring, or whether it reflects an unavoidable human behavior that can only be mitigated.
Answer D. Employee collusion to falsify records and conceal theft is a classic COSO inherent limitation: collusion among people can defeat otherwise effective control activities and cannot be completely prevented by redesign or monitoring, only mitigated and detected. The other scenarios describe control design, operation, or monitoring deficiencies that management can correct directly (reassign duties, update procedures, or restore monitoring).
Why not A: This is tempting because lack of monitoring reduces assurance and lets weaknesses persist, making problems seem unavoidable. It is wrong because understaffing or monitoring lapses are resource/monitoring deficiencies that management or the audit committee can address (hire staff, outsource, change oversight), not an inherent control limitation.
Why not B: This is tempting because concentrated duties greatly increase the opportunity for error or fraud and are a serious control weakness. It is wrong because it describes a remediable design/operation deficiency, management can reassign duties, introduce independent reviews, or implement compensating system controls, so it is not an inherent limitation that cannot be eliminated.
Why not C: This option tempts by showing recurring control failures and inconsistent behavior that could lead to misstatements. It fails as the inherent-limitation choice because the problem is primarily documentation, training, and procedural, management can update procedures, clarify limits, and retrain staff to correct it.
Question 3
Hint
Recall COSO's listed inherent limitations (e.g., management override, collusion, human error); choose the observation that fits that category and would most directly prompt increased substantive testing.
Answer A. COSO explicitly identifies management override as an inherent limitation because persons in positions of authority can circumvent controls; this risk cannot be eliminated by control design. Because override can materially affect financial reporting regardless of other controls, auditors typically respond with heightened professional skepticism and expanded substantive procedures. The other items describe mitigable control deficiencies, third‑party risks, or management choices that can be addressed with compensating controls, oversight, or vendor controls.
Why not B: Tempting because inadequate segregation increases the likelihood of errors or fraud, particularly in small entities. However, it is a control design/implementation deficiency that can often be mitigated (for example, via independent reconciliations, supervisory review, or compensating controls) and is not the unavoidable, universal limitation COSO describes for management override.
Why not C: Tempting because reliance on a vendor with access to sensitive data raises the risk of error or unauthorized changes. Yet third‑party risks are addressable through vendor due diligence, SOC reports, contract terms, and entity monitoring, so they are not an inherent, unpreventable limitation.
Why not D: Tempting because cost-benefit tradeoffs affect which controls management implements, and such choices influence residual risk. Nevertheless, this reflects a deliberate management decision about acceptable risk and control investment rather than an inherent limitation that cannot be mitigated; auditors can respond with targeted substantive procedures or oversight.
Question 4
Hint
Think about the level of assurance internal control is intended to provide and the kinds of human behavior that can still defeat controls.
Answer D. Under COSO, internal control is designed to provide reasonable assurance, not absolute assurance. One of the inherent limitations of internal control is that collusion among employees can circumvent controls that would work if individuals acted independently. That is why a control failure can still occur even in a generally well-designed system.
Why not A: This distractor appeals to the incorrect idea that monitoring eliminates all failures. COSO supports reasonable assurance, not perfect prevention; regular monitoring improves effectiveness but does not remove inherent limitations such as collusion.
Why not B: This is tempting because candidates often treat any control failure as proof that a COSO component was missing. However, collusion is an inherent limitation of internal control and does not automatically mean the control environment was absent; a sound control environment can exist even though employees collude to bypass specific procedures.
Why not C: This may seem plausible because collusion involves intentional misconduct that is harder to detect than errors. But COSO explicitly contemplates fraud risk and recognizes limitations of controls, including the possibility that collusion can override or defeat procedures, so collusion is not outside its scope.
Question 5
Hint
Ask whether the issue is unavoidable even in a well-designed system, or whether it reflects a preventable weakness in the controls.
Answer B. COSO recognizes that internal control provides reasonable, not absolute, assurance. An inherent limitation is that collusion between individuals can circumvent even well-designed controls. Thus collusion is an unavoidable risk in practice, unlike issues caused by missing or ineffective controls.
Why not A: This is tempting because lack of an independent review increases the chance that errors or fraud persist. However, it describes a missing or ineffective control (a control deficiency) that management can design and enforce, not an inherent limitation that would exist even in an effective system.
Why not C: This may seem like a broad problem since tone at the top affects control effectiveness. Still, failing to communicate ethical expectations is a weakness in the control environment (a design or operating deficiency) that management can correct, not an unavoidable limitation like collusion or management override.
Why not D: This choice is plausible because lingering access creates risk of misuse. Yet it reflects a preventable operating failure in IT and access controls, not an inherent limitation that exists despite well-designed controls.
Question 6
Hint
Think about the level of assurance COSO says internal control can provide, even when controls are well designed and duties are segregated.
Answer C. Under COSO, internal control provides reasonable assurance, not absolute assurance. One inherent limitation is collusion: two or more employees can work together to bypass segregation of duties and approval controls that might stop a single employee acting alone. Because the facts state management was not involved and controls were operating as designed, collusion is the most consistent COSO explanation.
Why not A: This is tempting because a successful fraud often feels like proof that the system failed. However, COSO acknowledges inherent limitations and requires reasonable, not absolute, assurance; a single fraud occurrence does not by itself prove controls were ineffective.
Why not B: Management override is another known inherent limitation of internal control, but the stem explicitly states management was not involved, so override does not explain this scheme of employee collusion.
Why not D: Monitoring is important, but it does not guarantee prevention of all improper transactions and often finds issues after they begin. Given the controls were described as designed and operating, the better COSO explanation is collusion as an inherent limitation rather than a complete failure of monitoring.
Common questions
Why does COSO provide reasonable rather than absolute assurance?
People can defeat controls through collusion or management override. Lake's independent approval requirement cannot guarantee that every fictitious-vendor payment will be prevented.
How do I distinguish collusion from management override?
Identify who causes the bypass. Secret cooperation between Lake's clerks is collusion; the CEO directing the bypass is management override.
Is a software approval failure an inherent limitation under COSO?
In Lake's software-patch version, the approval control fails to operate. That is a correctable control-operation failure, unlike the risk that people deliberately defeat otherwise functioning controls.
Practice AUD like the real exam
The free ChatCPA simulator: real exam layout, timed testlets, starting with a question on this topic. No account needed to start.
Open the free simulator →More on COSO
Questions from the ChatCPA bank of 17,658 CPA exam questions, each with a written reason for every wrong answer. ChatCPA is built by Nicholas Miller, CPA (Oregon #14907). How these pages are made. Spot an error? Tell us.