Concept
Competing Anticipation Tests
Is the material protected work product?
Competing Anticipation Tests
Rule 26(b)(3)(A) bars discovery of documents and tangible things that a party or its representative prepared “in anticipation of litigation or for trial,” and Evidence Rule 502(g)(2) defines work-product protection with the same words.1 Neither says how likely a suit must be or how large a part litigation must play in the document’s creation. The gap matters because discovery is otherwise extremely broad, so the anticipation test decides how much of a party’s own paper stays out of its opponent’s hands.2
Courts have filled the gap with three tests that answer two different questions. The specific-claim approach asks how definite the prospect of litigation was when the document was made. The primary-purpose test and the because-of test ask how central litigation was to the document’s creation, the first narrowly and the second broadly. A majority of circuits use the because-of test, and the Fifth Circuit uses the primary-purpose test.3 The choice carries stakes in both directions. A looser test lets routine business paper hide behind the bare possibility of suit, while a stricter one denies protection to preparation that prudent parties begin before any claim arrives.4 Because the circuits differ, a gray-area answer should name the test it applies and, when the jurisdiction is unknown, report the result under more than one.
Specific-Claim Approach
The narrowest reading asks the party to identify the lawsuit it anticipated. Courts that take this view insist on a showing that, when the document was created, there was an identifiable prospect of particular litigation in the near future.5 The D.C. Circuit’s decision in Coastal States Gas Corp. v. Department of Energy states the demand in these words: the document must have been prepared “with a specific claim supported by concrete facts which would likely lead to litigation in mind.” Janicker v. George Washington University speaks of an “identifiable resolve to litigate.”6 Both formulations require a particular claim, and the facts behind it must make suit likely and not merely imaginable.
The demand answers a real problem. Almost every business activity carries some potential for litigation, and courts requiring proof of a particular prospect reason that routine reliance on counsel’s input about that potential should not cloak ordinary business documents in protection.7 A claim that has already been asserted gives the court something it can check. A party’s private expectation that it would someday be sued gives the court nothing to check.
The First Circuit reaches a similar result by looking at how remote the litigation was. A report that serves a business purpose is not work product unless litigation was clearly anticipated, and “done * * * in advance of litigation” means more than that the report takes into account that its “subject might conceivably be litigated.”8
Hickman itself fits the model. After an accident, a public hearing was held, and shortly afterward counsel for the adverse party privately interviewed the survivors “with an eye toward the anticipated litigation.” By then, representatives of two of the deceased crew members were already in communication with him.9 The claims had arisen, and the Court spoke of preparation for possible litigation “after a claim has arisen.”10
That phrase creates friction when the litigation is expected only if a party later chooses to act. Suppose a document is plainly motivated by the prospect of suit over a proposed business transaction, but the suit will follow only if the company decides to go ahead with the deal. The motive is real, yet the litigation depends on a business decision not yet made, and Hickman described a claim that had already arisen.11 A test built on identifiable claims struggles with that situation, and the because-of test below was built to handle it.
The gearbox report shows how the approach works. Nobody had demanded compensation or threatened suit, and the company’s expectation that a serious injury would draw a lawsuit is exactly the generalized potential for litigation these courts refuse to count. The answer changes if, before the engineer wrote, the worker’s lawyer sent a letter demanding payment and threatening to sue. The letter names a claimant, identifies an injury, and threatens litigation, so a court can confirm an identifiable prospect of particular litigation from the document itself, and a report written in response becomes a candidate for protection. It is only a candidate. Even with a claim pending, a report must still have been prepared for the dispute and not in the regular course of business, which is the question the next two tests take up.12
Primary-Purpose Test
The primary-purpose test moves the inquiry from how definite the claim was to why the document exists. The Fifth Circuit, in United States v. El Paso Co., asks whether the “primary motivating purpose” behind the document’s creation was to aid in future possible litigation, a narrower standard than the because-of test.14 Courts using it treat reports that businesses commonly create for business reasons as less likely to be work product.15
The working question is whether the same document most likely would have been prepared apart from any concern about potential litigation.16 Courts answer it from circumstances. They read the document’s content, asking whether it refers to prospective litigation, and they consider the usual role of counsel, particularly where only in-house counsel was involved.17 A document the business would have written anyway is a business record that happens to be useful in a lawsuit, and the test denies it protection on that ground.
Documents the law requires a party to prepare sit at the far end of the spectrum. When regulatory law requires a party to prepare a document, the document cannot be said to have been prepared primarily in anticipation of litigation, because the party made it to conduct business in compliance with that law, and absent some claim of privilege the party should produce it.18 The Fifth Circuit applies the same logic more broadly: if the possibility of a dispute is remote, or the information was prepared in the ordinary course of business or while preparing a client’s tax returns, the information is not entitled to work-product protection.19 Stamping a legally required report “prepared in anticipation of litigation” changes nothing. The test asks what the document was for, and the regulation, not the legend, explains why the report exists.20
The hard case is the document that serves two masters.
Under the primary-purpose test the railroad probably loses. Repairing crossings and disciplining engineers are important purposes the railroad would have had apart from any lawsuit, so litigation was probably not the primary motivating purpose.21 The lawyers’ design of the form and the copy to general counsel point toward a legal purpose, and the role of counsel is one thing courts weigh.22 But they show that counsel wanted the information, not that litigation was the main reason the report exists.
The railroad report also marks the line between this test and the next. A document that serves litigation and operations must show that litigation was the dominant aim under the primary-purpose test, and a document with an important operating function usually cannot. The because-of test asks the broader question whether the prospect of litigation brought the document into being, and a dual-purpose document can satisfy that question. The same report can therefore be unprotected in one circuit and arguable in another.
The Because-Of Test
Most circuits ask whether, in light of the nature of the document and the factual situation in the particular case, the document can fairly be said to have been prepared or obtained because of the prospect of litigation.24 The Second Circuit’s decision in Adlman uses the words “the prospect of litigation.”25 The test makes causation the question, and three consequences follow. A document can qualify before any claim exists. It can qualify even though it also serves a business function. It cannot qualify if the party would have produced it in similar form without any prospect of litigation.
The first consequence reflects how parties actually behave. Prudent parties anticipate litigation and begin preparing before an action formally starts, so a test tied to a pending claim would deny protection to exactly that early preparation.26 That is why a contingency does not defeat the claim. Under the because-of test, it is not critical that litigation depends on a later business decision if the document would not have been created but for the possibility of litigation.27
The second consequence is the holding of United States v. Adlman. In contemplation of a challenge by the Service, the taxpayer’s tax advisor prepared a memorandum evaluating the tax issues raised by the taxpayer’s proposed merger. The court held that “work-product protection should not be denied to a document that analyzes expected litigation merely because it is prepared to assist in a business decision.”28 A company worried about a tax challenge wants an assessment of that challenge in order to decide whether to go forward, and a rule that stripped protection whenever the analysis informed a business choice would remove protection from the most useful analyses. The court gave a second example: a memorandum by company attorneys estimating the likelihood of success in litigation and analyzing the company’s legal strategies, prepared at the auditors’ request to help estimate what to reserve for litigation losses.29
The third consequence marks the boundary of the doctrine. Even though litigation is already in prospect, there is no work-product immunity for documents prepared in the regular course of business rather than for purposes of the litigation.30 Courts apply the boundary by asking whether the same document most likely would have been prepared apart from any concern about litigation.31 They use the same question on internal investigations. In many industries certain investigations, such as accident reviews, are routine, and placing all of them under the work-product shield would go too far. The because-of test overcomes that business purpose only if the investigation was shaped by concern for litigation.32
The Roxworthy decision shows the test working on a taxpayer’s facts. The Service argued that the memoranda had been prepared for penalty protection and not in anticipation of litigation. The appellate court disagreed, holding that the taxpayer did anticipate litigation because a yearly Service audit was a certainty given the company’s size and the transaction at issue, and because the accountants had advised that the Service targeted that type of transaction.33 The same facts that made a dispute foreseeable also explain why the memoranda were created.
A contrasting tax dispute shows what an ordinary-course document looks like. The district court there noted that the tax memorandum did not specifically refer to litigation and that, given the size and complexity of the transactions, the same comprehensive analysis would have been provided in similar form regardless of any anticipated audit or litigation.34 Those two observations, silence about litigation and likely existence in the same form anyway, are the kind of evidence that points away from protection.
Tax accrual workpapers have divided the circuits. The First Circuit, sitting en banc in United States v. Textron Inc. in 2009, denied work-product protection. The next year the D.C. Circuit in Deloitte upheld protection on similar facts, and the Wells Fargo court protected tax accrual workpapers to the extent they reflected legal analysis for the “recognition step” of FASB ASC 740-10.35 In 2010 the Supreme Court denied certiorari in Textron, leaving a conflict among the circuits over the correct test.36 The state of the law on these workpapers is far from settled.
Both documents come from one company and touch overlapping subject matter, so the subject cannot be what separates them. A recurring reporting need dictates the schedule’s content, and that need continues whether or not anyone threatens a dispute. One specific inquiry dictates the memorandum’s content, and its second use does not change why it exists.
The because-of test is the broadest of the three, so a document can win under it and lose elsewhere. The Fifth Circuit’s standard is narrower, and under it reports that businesses commonly create for their own purposes are less likely to be protected.37 The First Circuit looks to the remoteness of litigation and treats a report with a business purpose as work product only if litigation was clearly anticipated.38 When the document’s purpose is plain, the tests converge. In Regions Financial Corp. v. United States, a district court held tax opinions analyzing two listed transactions protected under both the primary-purpose and because-of tests, because under either the critical issue is the purpose for which the documents were created.39