GB791 Unit 2 Research Article Critique Example

Course	GB791 Advanced Topics in Accounting
Unit	Unit 2
Paper type	Research article critique
Length	About 1,140 words, 7 pages double-spaced
Format	APA 7 student paper
School	Purdue University Global
Program	MBA
Updated	October 2026

Free sample paper for GB791 Unit 2

The Assumption Inside Every Accrual Model: A Unit 2 Critique of Jones (1991)

[Student Name]

Purdue University Global

GB791: Advanced Topics in Accounting

Unit 2 Research Article Critique

[Instructor Name]

[Date]

This critique concerns a real published article; its account of the paper's design is the candidate's reading.

What this part is doingPromising to expose an inherited assumption tells the reader the critique will reach beyond one article.

The Article

Jones, J. J. (1991). Earnings management during import relief investigations. Journal of Accounting Research, 29(2), 193-228.

Part One: The Question and the Incentive

The article begins from a provision of United States trade law. When a domestic industry petitions the International Trade Commission for relief from imports, the Commission considers, among other things, whether the industry has been injured, and reported profitability is one of the measures of injury. A firm in a petitioning industry therefore has an incentive to make its earnings look worse during the investigation, since lower reported profits strengthen the case for protection. Jones (1991) asked whether managers acted on that incentive by choosing accounting accruals that reduced income in the year the Commission examined.

The question is well chosen. Many earnings management studies struggle to say which way managers should push earnings, because incentives often point both ways. Here the incentive points one way, downward, and it applies at a known time. That combination gives the study a sharp prediction and a clear event window.

Part Two: The Design

The difficulty in any study of earnings management is that accruals are not all discretionary. A firm whose sales are rising will report growing receivables, and a firm with more equipment will report more depreciation, without any manager making a choice. The study needed a way to separate the accruals that the firm's economic circumstances would produce from the ones managers chose.

Jones built a model of normal accruals for each firm. Total accruals, scaled by the prior year's total assets, were regressed on the change in revenue and on the level of gross property, plant and equipment, both scaled the same way, plus a scaled intercept. The regression was estimated separately for each firm, using years before the investigation. The fitted model was then used to predict what accruals should have been in the investigation year. The difference between actual and predicted accruals, the prediction error, was read as the discretionary component.

The result was that prediction errors in the investigation year were negative on average: firms reported lower accruals, and so lower earnings, than their own history predicted. The study read this as evidence of income-decreasing earnings management during import relief investigations.

What this part is doingDescribing the design in plain steps before critiquing it lets the reader see exactly where the assumption enters.

Part Three: The Identifying Assumption

The design rests on one assumption: that the relation between accruals and their economic drivers in the estimation years is the same relation that holds in the investigation year. If the relation is stable, any departure from it in the investigation year must reflect something other than the drivers in the model, and the study attributes that something to managers' choices.

There are two reasons to doubt stability here, and both are specific to this setting.

The first is that industries petition for import relief when they are in trouble. Sales may be falling, inventories accumulating, plants underused. Under those conditions, accruals can behave differently for reasons unrelated to any managerial choice. Write-downs of obsolete inventory, allowances for doubtful customers and impairment of idle equipment all lower accruals when a business deteriorates. A model estimated on healthier years would not predict those accruals and would classify them as discretionary. The result could then reflect the industry's distress rather than its managers' choices.

The second reason is that the model treats revenue as given. If managers can influence reported revenue, for example by delaying shipments or tightening credit terms during the investigation, then the change in revenue is itself partly discretionary, and the model's predicted accruals already contain some of the management it is trying to measure. Dechow et al. (1995) proposed a modification that subtracts the change in receivables from the change in revenue, on the reasoning that credit sales are easier to manage than cash sales.

A prediction error is evidence of management only if nothing else changed between the years the model learned from and the year it judged.

Testing the Assumption

The assumption can be partly tested. One approach is to compare petitioning firms with firms that faced similar economic conditions but had no investigation, since distress without an incentive should produce the distress-related accruals without the management. Kothari et al. (2005) later proposed matching each firm with another of similar performance and measuring discretionary accruals relative to that match, which addresses the first concern directly. A second approach is to check whether the prediction errors reverse after the investigation, since managed accruals tend to unwind while distress-related write-downs usually do not.

Jones (1991) acknowledged the difficulty and used the firms' own pre-investigation history as the benchmark, which controls for firm-specific accounting practices. But a firm's own history cannot control for a change in the firm's circumstances, and the import relief setting guarantees such a change.

A Second Weakness: Precision

A further problem is statistical rather than conceptual. Each firm's model was estimated from its own annual history, so the number of observations behind each regression was small, and coefficients estimated from a short series are imprecise. Predicted accruals inherit that imprecision, and so do the prediction errors read as management. A negative error in the investigation year may therefore be partly noise, and a test that averages errors across a small sample of firms has limited power to tell noise from a real effect. Later research moved to estimating the model across all firms in an industry and year, which gains observations at the cost of assuming that firms in the same industry share one accrual process. Neither choice removes the stability assumption; each relocates it.

Part Four: What May Be Concluded

Within its setting, the study provides evidence consistent with income-decreasing earnings management, and the clarity of the incentive makes the interpretation plausible. The claim is stronger than the design alone can support, though, because the setting that supplies the incentive also supplies a rival explanation. A cautious reading is that the evidence fits earnings management and is not easily explained by distress alone, given the direction and timing of the effect, but that distress cannot be ruled out without a comparison group.

The study's larger legacy is its model. The Jones model, with the modification of Dechow et al. (1995) and the performance matching of Kothari et al. (2005), became the standard instrument for measuring discretionary accruals across hundreds of later studies. Each of those studies inherits the stability assumption. In settings where firms' circumstances change sharply around the event of interest, such as mergers, equity offerings or financial distress, the assumption is under the same pressure it faced in 1991, and the same tests apply.

What this part is doingEnding on the model's later use shows why a critique of one 1991 paper still matters to current research.

References

Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management. The Accounting Review, 70(2), 193-225.

Jones, J. J. (1991). Earnings management during import relief investigations. Journal of Accounting Research, 29(2), 193-228. https://doi.org/10.2307/2491047

Kothari, S. P., Leone, A. J., & Wasley, C. E. (2005). Performance matched discretionary accrual measures. Journal of Accounting and Economics, 39(1), 163-197. https://doi.org/10.1016/j.jacceco.2004.11.002

What the GB791 Unit 2 assignment asks for

Unit 2 in GB791 asks you to critique one published empirical article in depth. Give the full citation first. Restate the research question and explain why the setting makes it answerable. Describe the design plainly enough that a reader could follow each step. Then state the identifying assumption, the condition that must hold for the design to measure what the authors claim, and test it against the facts of the setting. Explain any rival explanation the setting itself creates. Discuss statistical concerns such as precision where they matter. Say how the assumption could be tested, citing later work that did so. Close with what may be concluded from the study and how far the conclusion extends beyond its sample. Six pages in four parts, with the article and two or three later studies cited, suits most sections.

How this GB791 Unit 2 example is built

The critique gives the full citation, then explains the trade law incentive that makes lower reported earnings useful during an International Trade Commission investigation. The design section walks through the firm-specific accrual regression and the use of prediction errors. Part three states the stability assumption and offers two reasons to doubt it: distress during petitions and revenue that managers can influence, with Dechow and colleagues' receivables adjustment. The highlighted sentence states the assumption as a test. A section on testing introduces Kothari and colleagues' performance matching and reversal checks, and a further section explains why short firm histories make estimates imprecise. The conclusion separates what the evidence fits from what it proves and follows the model into later research.

Where the points sit in the GB791 Unit 2 rubric

The GB791 critique earns its points through the identifying assumption. Instructors look for an assumption stated precisely and tested against the specific setting, not a generic list of limitations. A clear account of the design, step by step, is weighted because a critique cannot locate a weakness it has not described. Rival explanations count most when they arise from the setting itself, as distress does for import relief petitions. Suggesting how the assumption could be tested, with reference to later research, shows command of the literature. Statistical concerns earn credit when tied to the design. The closing section is graded on calibration, meaning claims sized to what the import relief evidence can bear and no larger. Citing the original and later studies accurately completes the paper.

Common GB791 Unit 2 mistakes, and how to avoid them

Many GB791 critiques summarize the article and then list limitations that would apply to almost any study, such as sample size or generalizability. Others criticize the authors for not using methods developed years later, which misses the point of an appraisal. The identifying assumption is often never stated, so the critique has no center. Students frequently describe the Jones model's equation without explaining why the prediction error is read as management. Later research is cited as a list rather than as answers to specific problems. State the question and incentive, describe the design, name the assumption, test it against the setting, show how it could be checked and calibrate the conclusion. Name the article on your syllabus, and a writer can build the critique around its own identifying assumption.

Get a GB791 Unit 2 example written to your instructions

Name the article your section assigned, or the research area you want one chosen from, and include the Unit 2 prompt and rubric. The critique drafted for you restates the question, rebuilds the design, names the identifying assumption and tests each threat against it. There is no fee for a first custom sample, typically ready within 24-48h. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.

More GB791 papers and MBA samples

GB791 Unit 2 questions, answered

Why critique a paper from 1991 rather than a recent study?

Because the Jones model still sits inside a large share of earnings management research, so its assumption is inherited by work published this year. Critiquing the original shows where that inheritance comes from. If your section requires a recent article, the same four-part structure applies, and a recent paper using a Jones-type model makes an especially good candidate.

What is an identifying assumption, in plain terms?

It is the condition that must hold for a study's comparison to measure what the authors say it measures. In Jones (1991), prediction errors count as manipulation only if normal accruals behave in the investigation year as they did before. If something else changed accruals that year, the model would label ordinary business as earnings management. Every empirical design has at least one such condition.

Should the critique propose a better design?

Briefly, where it clarifies a threat. This sample notes that a performance-matched control, of the kind Kothari, Leone and Wasley later proposed, would address the distressed-industry problem. A full alternative design belongs in later units. The critique's main task is judging whether this design answers this question, not writing the paper the reviewer would have preferred.

Where can I find a free GB791 Unit 2 sample paper?

A complete GB791 Unit 2 research article critique is on this page, examining Jones's 1991 earnings management study and the stability assumption every later accrual model inherits.

What does the modified Jones model change?

Dechow, Sloan and Sweeney subtracted the change in receivables from the change in revenue, on the reasoning that credit sales are easier for managers to influence than cash sales.