ACCT90013 · Financial Accounting Theory And Practice
Earnings Management
This is the stewardship / contracting half of the subject in full flight — and it is where Positive Accounting Theory (PAT) lands. PAT is a positive theory: it does not say what accounting ought to be; it explains and predicts the accounting policies self-interested, wealth-maximising managers will actually choose, given their contracts. Hand it a manager's incentive setting and it predicts the direction of the reported-income choice via three hypotheses — the bonus-plan, debt-covenant and political-cost hypotheses. Earnings management is PAT made concrete: the manager pulling the accrual levers PAT says they will pull. You learn the debtholder–shareholder conflict that makes lenders write accounting-based covenants and demand conservatism; the patterns and methods of earnings management (income smoothing, big bath, accrual vs real); the “good” (signalling) vs “bad” (opportunistic) debate; and the detection tools (Jones model, Beneish, Benford) — all governed by the iron law: accruals reverse.
What this chapter covers
- 01Positive vs normative — what PAT is (and is not)
- 02PAT's three hypotheses: bonus-plan, debt-covenant, political-cost
- 03The debtholder–shareholder conflict; accounting-based covenants
- 04Accounting conservatism and why lenders demand it
- 05Earnings management: patterns (smoothing, big bath) and methods (accrual vs real)
- 06Good (signalling) vs bad (opportunistic); detection (Jones, Beneish, Benford) & the accruals iron law
Worked example: fire the right PAT hypothesis
- +1(a) Bonus-plan hypothesis: managers on a bonus plan choose accounting that maximises their bonus. Above the cap, extra earnings earn no bonus → the manager defers income (a ‘take a bath now, bank it for next year’ / income-decreasing choice) to shift earnings into a future bonus period.
- +1(a) Predicted choice: income-decreasing accruals this year (e.g. accelerate expenses, raise provisions) — PAT predicts the direction, here downward.
- +1(b) Debt-covenant hypothesis: firms close to breaching covenants choose income-increasing accounting to avoid the costly breach — the opposite direction.
- +1Resolve: the two incentives conflict; the prediction goes to whichever is more binding / costlier. A covenant breach (technical default, repricing, control rights) is usually far costlier than a forgone bonus, so the income-increasing choice likely dominates — state the dominant cost explicitly.
Key terms
- Positive Accounting Theory (PAT)
- Watts & Zimmerman's positive theory that explains and predicts the accounting choices self-interested managers make given their contracts — via the bonus-plan, debt-covenant and political-cost hypotheses. It predicts behaviour; it does not prescribe.
- Debt-covenant hypothesis
- PAT's prediction that, all else equal, the closer a firm is to breaching an accounting-based debt covenant, the more likely managers are to choose income-increasing accounting to avoid the costly breach.
- Accounting conservatism
- The asymmetric treatment that recognises losses earlier than gains and requires a higher verification standard for good news. Lenders demand it because, facing downside risk, they want the bad news fast — it disciplines managers and protects covenants.
- Earnings management
- Managers' use of judgement in reporting (accrual choices, real transaction timing) to alter reported earnings — e.g. income smoothing or a ‘big bath’. It can be ‘good’ (credibly signalling private information) or ‘bad’ (opportunistically misleading).
- The accruals ‘iron law’
- Because total accruals over a firm's life sum to zero, an income-increasing accrual today reverses later — earnings management borrows from the future. This reversal is what detection models (e.g. the Jones model's discretionary accruals) exploit.
Earnings Management FAQ
Is PAT a normative theory that says managers should manage earnings?
No — this is the headline trap. PAT is strictly positive: it explains and predicts what self-interested managers will do given their contracts; it makes no claim about what they ought to do. Saying “PAT recommends income smoothing” is wrong. Saying “PAT predicts a manager near a covenant will choose income-increasing accounting” is right. Keep the positive/normative line crisp.
How do I pick which of the three hypotheses to fire?
Read the manager's contracts and exposure: a bonus plan → bonus-plan hypothesis (choices that maximise the bonus, watch for floors and caps); leverage/covenants → debt-covenant hypothesis (income-increasing to avoid breach); large, visible, politically exposed firms → political-cost hypothesis (income-decreasing to dodge scrutiny, taxes or regulation). When two fire at once, adjudicate by relative cost — the costlier constraint wins.
When is earnings management ‘good’ rather than fraud?
On the efficiency (signalling) view, smoothing can credibly convey management's private information about sustainable earnings, lowering the cost of capital — within the rules. On the opportunistic view it misleads stakeholders for private gain. The line is intent and disclosure: judgement that communicates true persistence vs judgement that hides it. The exam wants you to argue both views and take a position, not to brand all of it fraud.
What's the ‘iron law’ and why do detection models rely on it?
Accruals must reverse — over the firm's life they net to zero, so income borrowed via accruals today is repaid later. Detection models like the Jones model estimate ‘normal’ accruals from firm fundamentals and flag the abnormal discretionary residual; Beneish (M-score) and Benford's-law digit tests work similarly. The reversal is the fingerprint: unusual accruals that later unwind signal management, not economics.
Exam move
Drill the PAT firing drill until it is automatic: read the contract/exposure, name the hypothesis (bonus-plan / debt-covenant / political-cost), and predict the income direction — up or down. The highest-value skill is adjudicating conflicts: when two hypotheses fire opposite directions, decide by which constraint is costlier (a covenant breach usually beats a forgone bonus) and say so explicitly. Keep the positive/normative line crisp — PAT predicts, it never prescribes. For earnings management, hold a three-way template: classify the pattern (smoothing, big bath), the method (accrual vs real), and the verdict (signalling vs opportunistic), then pick a detection tool and invoke the iron law (accruals reverse). As always, the marks are in the applied sentence that maps the theory to the named manager and contract.
