Practice the explanation as well as the calculation
These three exercises are original fictional teaching cases. They are not employer interview questions and are not representations of any firm's selection process. NERA notes that a case study may be requested; other employers and offices use different processes. Check the actual invitation rather than assuming every economic-consulting interview has the same format.
For each case, spend a few minutes defining the question before calculating. Write the formula, track units and make the arithmetic reproducible. Then explain one reason the numerical result could be misleading. A useful answer separates a fact supplied in the exercise, an assumption you introduced, and an inference that depends on that assumption.
Sources[1]
Case 1: higher prices, lower contribution
A fictional product sells 10,000 units at $20, with variable cost of $12 per unit. Next period it sells 9,000 units at $22, with variable cost of $15 per unit. Fixed costs are unchanged and not supplied. Calculate revenue and contribution, then explain why the changes differ.
| Period | Units | Price per unit | Variable cost per unit |
|---|---|---|---|
| Before | 10,000 | $20 | $12 |
| After | 9,000 | $22 | $15 |
Revenue falls from $200,000 to $198,000, or $2,000. Contribution falls from 10,000 × ($20 − $12) = $80,000 to 9,000 × ($22 − $15) = $63,000, or $17,000. Calling the $2,000 revenue decline the profit effect would ignore the cost increase.
One exact contribution bridge is volume: (9,000 − 10,000) × $8 = −$8,000; price: 9,000 × ($22 − $20) = +$18,000; unit cost: −9,000 × ($15 − $12) = −$27,000. The three changes sum to −$17,000. This is an accounting decomposition, not proof that price caused the volume decline.
Case 2: a comparison group changes the interpretation
A fictional treatment group has an outcome of 100 before an event and 120 afterward. A comparison group moves from 80 to 88. The treatment group's raw change is 20; the comparison group's change is 8. The difference in changes is 12 outcome units.
| Group | Before | After |
|---|---|---|
| Treatment | 100 | 120 |
| Comparison | 80 | 88 |
That result is not automatically a causal estimate. You would need a defensible reason the comparison group represents what would have happened to the treatment group without the event. Ask about prior trends, changing composition, exposure to other events, measurement consistency and spillovers. Two aggregate periods cannot establish those conditions.
Do not report 12% simply because the difference is 12. The result is in the supplied outcome units. A percentage requires a stated denominator and a reason for using it. As a sensitivity exercise, change the comparison group's final outcome to 95: its change is 15 and the difference becomes 5. Explain why the comparison choice matters.
Case 3: a lost-sales claim is not a lost-revenue shortcut
A fictional claimant says it would have sold 4,000 units at $50 without an event, but actually sold 3,000. Avoidable variable cost is assumed to be $30 per additional unit. Under those assumptions, lost contribution is (4,000 − 3,000) × ($50 − $30) = $20,000, not the $50,000 difference in revenue.
| Counterfactual units | Incremental units | Lost contribution |
|---|---|---|
| 3,500 | 500 | $10,000 |
| 4,000 | 1,000 | $20,000 |
| 4,200 | 1,200 | $24,000 |
The main uncertainty is the 4,000-unit counterfactual. If a supported range is 3,500–4,200 units, the same formula yields $10,000–$24,000. That is an assumption-driven range, not a statistical confidence interval. Ask whether capacity could support the additional sales, whether demand was present, which costs were avoidable, and whether any offsetting gains should be considered.
Do not add interest, taxes, mitigation or other adjustments without a defined scope and evidence. This simplified exercise teaches arithmetic and questioning; it is not a legal damages methodology or a professional valuation.
Check the answer before making it impressive
Use this review sequence: verify the formula; recalculate independently; label currency, units and time period; state the counterfactual; identify the assumption carrying the conclusion; and explain what evidence could overturn it. A chart should not hide missing data or suggest precision the inputs do not support.
Record an explanation of one case without reading a script. When you reach an unsupported inference, say what you would check rather than filling the gap with a confident story. Then ask someone to change an input and explain how your answer changes. That tests understanding better than memorizing the final number.
For a portfolio, label the work as an independent fictional exercise and include the input file and answer key. Do not put an employer logo on it or suggest that completing it certifies readiness for that employer's interviews.