Trade Secrets

Rockwell v. DEV Industries: Posner Makes Secrecy a Cost-Benefit Problem

The Seventh Circuit reversed summary judgment to hold that whether a trade-secret owner took 'reasonable' precautions is almost always a jury question turning on the balance of costs and benefits.

March 17, 2025
Engineering drawings rolled in a secure storage vault
The case asked how leaky a drawing vault can be before the law withdraws its protection. Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

Rockwell Graphic Systems, Inc. v. DEV Industries, Inc., 925 F.2d 174 (7th Cir. 1991), No. 90-1499, decided February 11, 1991, is the opinion most responsible for the modern, economic way American courts think about the secrecy-measures element of a trade-secret claim. Writing for a panel that also included Judges Cummings and Flaum, Judge Richard Posner reversed a grant of summary judgment that the U.S. District Court for the Northern District of Illinois had entered for the defendants, and in doing so converted a seemingly binary question — were the owner’s precautions adequate, yes or no — into a contextual balancing of costs and benefits that ordinarily belongs to a jury.

The underlying dispute was unglamorous, which is part of its enduring teaching value. Rockwell manufactured printing presses and the replacement “piece parts” that keep them running. To have those parts machined, Rockwell sent independent vendors detailed “piece part drawings” specifying dimensions, tolerances, materials, and methods. When several former Rockwell employees turned up at DEV Industries, a competitor, discovery revealed that DEV possessed roughly one hundred of Rockwell’s confidential drawings. The claim proceeded under the Illinois Trade Secrets Act. The defense was not that DEV had developed the drawings independently but that Rockwell had been so careless with them that the drawings were no longer secret at all.

At a glance

  • Case: Rockwell Graphic Systems, Inc. v. DEV Industries, Inc., 925 F.2d 174 (7th Cir. 1991), No. 90-1499
  • Decided: February 11, 1991 (Posner, J.; Cummings and Flaum, JJ.)
  • Court below: N.D. Ill., which granted summary judgment for the defendants
  • Disposition: Reversed and remanded for trial
  • Law applied: Illinois Trade Secrets Act
  • Core holding: Whether an owner took reasonable precautions to keep information secret is a fact-intensive balancing of costs and benefits that “only in an extreme case” can be resolved on summary judgment; imperfect security does not automatically forfeit protection

The two functions of secrecy precautions

The intellectual core of Rockwell is Posner’s explanation of why the law conditions trade-secret protection on reasonable precautions at all. He identified two distinct functions, and the analysis still organizes the doctrine.

First, precautions are evidence that the information has real value worth protecting. “If Rockwell expended only paltry resources on preventing its piece part drawings from falling into the hands of competitors such as DEV,” Posner asked, “why should the law … bother to provide Rockwell with a remedy?” An owner’s own behavior is the best available signal of whether it genuinely believed the information conferred a competitive edge. A company that spends nothing guarding information is telling the court that the information is not worth much.

Second, precautions are circumstantial evidence of misappropriation. “The greater the precautions that Rockwell took to maintain the secrecy of the piece part drawings,” the court reasoned, “the lower the probability that DEV obtained them properly and the higher the probability that it obtained them through a wrongful act.” If material is locked down and a competitor nonetheless ends up with a hundred copies, the security itself narrows the universe of innocent explanations. Secrecy measures, in other words, do double duty: they prove the asset and they help prove the theft.

Why reasonableness is rarely a summary-judgment question

The district court had treated Rockwell’s lapses as dispositive. Rockwell did keep drawings in a vault, restricted building access, required confidentiality agreements from employees and vendors, used sign-out procedures, and stamped vendor copies with confidentiality legends. But it also let engineers make copies freely, did not always reclaim drawings from vendors, and allowed even unsuccessful bidders to keep them — so that tens of thousands of copies circulated outside the vault. To the trial court, that looseness defeated secrecy as a matter of law.

Posner rejected the inference that imperfect equals unreasonable. The right question is not whether an owner achieved airtight secrecy — “absolute secrecy” is never required — but whether it struck a sensible balance. More security always means higher cost, and not only in dollars. Posner stressed the indirect costs: forbidding engineers to copy drawings, for example, would force them to share single physical originals in ways that could cripple day-to-day productivity. A rule that demanded maximal precautions would force firms to spend more on secrecy than the secrets are worth, which serves no one.

Because that balance “depends on a balancing of costs and benefits that will vary from case to case,” the court held, “only in an extreme case can what is a ‘reasonable’ precaution be determined on a motion for summary judgment.” The reasonableness of Rockwell’s choices — how much leakage was tolerable given the cost of plugging it, and whether the outstanding copies actually reflected negligence or an acceptable trade-off — was a quintessential jury question. Summary judgment was reversed and the case sent back for trial.

A framework, not a checklist

Rockwell is frequently cited for the proposition that trade-secret owners need not be perfect, but its deeper contribution is methodological. It refused to reduce the secrecy element to a checklist of measures that a plaintiff either has or lacks. Two firms can adopt identical measures, and the same set of precautions can be reasonable for one and unreasonable for the other, because the calculus depends on the value of the information, the cost of additional security, and the realistic threat environment. That is why later courts treat the inventory of measures — vaults, legends, agreements, access controls — as inputs to a balancing test rather than as a scorecard.

The opinion also quietly reframed the relationship between secrecy and value. By treating precautions as a proxy for value, Posner gave defendants a powerful but double-edged argument. A defendant can point to leakage to argue the owner did not really value the information; but if the owner did invest heavily, that same investment becomes affirmative proof both that the secret mattered and that the defendant’s possession is suspicious. The framework rewards owners who calibrate their security to the asset, and it tolerates the rational decision not to gold-plate protection for information of modest worth.

Open questions

Rockwell announced a balancing test without supplying its boundaries, and the edges remain contested. How much leakage is too much before a court should take the question from the jury as the “extreme case” the opinion reserved? The decision does not say where that threshold sits, and trial courts still disagree about when outstanding copies, lax enforcement, or unreturned documents become fatal as a matter of law. The opinion also predates the digital era; applying a cost-benefit calculus to cloud storage, sprawling vendor ecosystems, and instantaneous mass copying raises questions Posner’s printing-press hypotheticals did not confront. Finally, the framework’s reliance on precautions as a proxy for value can strain in industries where security is cheap and ubiquitous, so that nearly every owner clears the bar and the element does little screening work.

Implications

  • Imperfect security is not forfeiture. Owners who took meaningful but incomplete measures can still reach a jury; defendants should not expect leakage alone to win on summary judgment.
  • Document the cost-benefit logic. Because reasonableness is a balance, owners benefit from a record explaining why they chose the measures they did and why more would have been disproportionate.
  • Precautions prove two things at once. Robust security supports both the value of the secret and an inference that the defendant obtained it wrongfully — a reason to invest visibly, not just adequately.
  • Reasonableness is usually a jury question. Counsel on both sides should plan for the secrecy element to survive to trial in all but extreme cases.
  • Calibrate to the asset. Gold-plating low-value information wastes resources; under-protecting high-value information undercuts both the value inference and the misappropriation inference.

Frequently asked questions

Did Rockwell win the case? Not outright. The Seventh Circuit reversed summary judgment for the defendants and remanded for trial. The holding was that Rockwell’s precautions could not be deemed unreasonable as a matter of law, not that they were conclusively reasonable.

Does a trade-secret owner have to keep information perfectly secret? No. The court was explicit that absolute secrecy is not required and that excessive precautions can be economically irrational. The owner must take precautions that are reasonable under the circumstances, judged by weighing their costs against their benefits.

Why does it matter how much a company spent protecting information? Posner identified two reasons. The level of precaution signals how valuable the information really is, and strong precautions make it less likely a competitor acquired the information innocently — both of which bear directly on whether a remedy is warranted.

Authorities and sources