How is a but-for world different from a business forecast?
A forecast predicts a future that has not happened yet; a but-for world reconstructs a past that did not happen, and it is tested against what actually did. The two look alike on a spreadsheet, and that resemblance is what gets damages opinions excluded. A forecast answers to the business that wrote it. A but-for world answers to a record, to an opposing expert paid to find every departure from the evidence, and to a court applying a certainty standard.
The Federal Circuit named the risk directly in Grain Processing Corp. v. American Maize-Products Co., 185 F.3d 1341 (Fed. Cir., decided 4 August 1999), a patent damages case: “Reconstructing the market, by definition a hypothetical enterprise, requires the patentee to project economic results that did not occur. To prevent the hypothetical from lapsing into pure speculation, this court requires sound economic proof of the nature of the market and likely outcomes with infringement factored out of the economic picture.” The same opinion states the causal frame that gives the construct its name: “To recover lost profits, the patent owner must show ‘causation in fact,’ establishing that ‘but for’ the infringement, he would have made additional profits.”
The second difference is the defendant’s presence in it. A forecast assumes the world as the business expected it; a but-for world has to say what the other side would have done. Grain Processing turned on that: “A rational would-be infringer is likely to offer an acceptable noninfringing alternative, if available, to compete with the patent owner rather than leave the market altogether.” A counterfactual in which the defendant simply disappears, leaving every diverted sale to the plaintiff, is the easiest part of a damages case to attack.
How is a but-for world built and tested?
It is built from evidence about worlds that existed, and tested by whether independent methods converge. The strongest inputs are contemporaneous: budgets presented to a board, forecasts given to a lender, pipelines reviewed at a sales meeting, written before the dispute for reasons other than litigation. Their force comes from their timing rather than their optimism, and their absence is informative too, because a but-for world that outruns everything the business ever put on paper about itself needs an explanation.
Testing means running more than one construction and seeing whether they agree. A direct projection corroborated by a comparison to unaffected firms lands harder than either alone, because the two fail in different ways, so agreement is unlikely to be an artifact of either. Divergence is worth finding early and in private. The mechanics are set out at Constructing the counterfactual.
The expert doing the building works under a professional standard, whatever method is chosen. AICPA Statement on Standards for Forensic Services No. 1, issued by the AICPA’s Forensic and Valuation Services Executive Committee and effective for engagements accepted on or after 1 January 2020, carries at paragraph 6 the general standard to “[o]btain sufficient relevant data to afford a reasonable basis for conclusions or recommendations in relation to any professional services performed,” and provides at paragraph 7 that “[a] member performing forensic services should not subordinate his or her opinion to that of any other party.” The counterfactual is where that second requirement is tested, because it is the part of a damages model most responsive to pressure.
What is the before-and-after method, and when does it break?
Before-and-after builds the but-for world from the plaintiff’s own results in a period the conduct did not touch, usually the years before it began and sometimes the period after it ended. Its appeal is that the comparator is the business itself, so differences in market, scale and customer mix are controlled by construction rather than by argument. The Supreme Court accepted evidence of this kind in Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251 (1946), where the plaintiffs compared the Jackson Park Theatre’s own receipts across periods and also compared them with a competing theater’s.
It breaks when the clean period is not clean. A before period contaminated by a management change, a product transition, a financing event or a macroeconomic swing transmits every one of those effects into the damages number as though the defendant had caused them. It also breaks when there is no before period at all: a young business, a business transformed since the comparison years, or a venture that never traded. A counterfactual for a business without a trading history has to be built some other way.
Separating the conduct’s effect from everything else that moved in the same period decides more cases than the arithmetic does, and disaggregation is covered at Causation against damages. The comparison methods are set out at Before-and-after and yardstick.
What is the yardstick method, and what makes a comparator hold up?
The yardstick builds the but-for world from something else that was not affected: a comparable firm, a portfolio of them, an industry index, or the same company’s operations in another region. Courts have treated the availability of such a benchmark as part of the proof rather than as presentation. The Florida Second District Court of Appeal stated the requirement in Montage Group, Ltd. v. Athle-Tech Computer Systems, Inc., 889 So. 2d 180 (Fla. 2d DCA 2004): to recover lost prospective profits a business “must prove not only that the defendant’s action caused the damage but also that there is some standard or ‘yardstick’ by which the amount of damages may be adequately determined,” citing the Supreme Court of Florida’s decision in W.W. Gay Mechanical Contractor, Inc. v. Wharfside Two, Ltd., 545 So. 2d 1348, 1351 (Fla. 1989). That is Florida’s statement of the requirement; the phrasing differs elsewhere.
A comparator holds up when it resembles the plaintiff on the dimensions that drive the outcome being measured, a far stricter test than sharing an industry code. Market, scale, cost structure, customer mix and growth stage all matter, and the comparator must have been exposed to the same general conditions during the damages period while being untouched by the conduct. Portfolios often beat single firms, because idiosyncratic noise averages out and the selection looks less like choosing the one company that grew.
What kills yardsticks is asymmetry: a comparator chosen after the fact because it performed well, differences acknowledged only where they help, or a benchmark whose own results were affected by the same conduct, which quietly subtracts the damages it was meant to reveal. The discipline is to state the selection criteria before running the comparison, and to state what would have disqualified a candidate, because the first question on cross-examination is how many firms were looked at and why these survived.
What is the market-share approach to the but-for world?
The market-share approach claims a proportion of the diverted business rather than all of it, on the theory that in the but-for world the plaintiff would have captured its historical share of the sales the defendant took. It answers the standard objection to a lost-sales counterfactual, that some customers would have gone elsewhere. The Federal Circuit approved it in State Industries, Inc. v. Mor-Flo Industries, Inc., 883 F.2d 1573, 1580 (Fed. Cir. 1989), described in Grain Processing Corp. v. American Maize-Products Co., 185 F.3d 1341 (Fed. Cir. 1999) as “upholding award of lost profits in proportion to patentee’s market share of the relevant market including acceptable noninfringing substitutes.”
It sits inside the older four-part framework for lost profits in patent cases. In the earlier appeal in the same litigation, Grain Processing Corp. v. American Maize-Products Co., 108 F.3d 1392 (Fed. Cir. 1997), the Federal Circuit restated the test as requiring “(1) demand for the patented product, (2) the absence of acceptable noninfringing substitutes, (3) manufacturing and marketing capability to exploit the demand, and (4) the amount of profit it would have made,” citing Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152, 1156 (6th Cir. 1978). Market share is how a plaintiff who cannot satisfy the second factor across the whole market may still recover on the part of it that was its own.
Two cautions belong with it. The share must be measured in a market defined to include the acceptable substitutes, and the plaintiff must have had the capacity to serve the additional volume, a factual question about plants, people and working capital rather than an assumption. Whether infringement occurred is not an economic question and belongs to our sibling Intellectual Property Institute; what it is worth, once liability is assumed, is measured in this Institute’s intellectual property area.
How speculative may a but-for world be?
It may be uncertain, but not speculative, and the line between those two words is the whole of the reasonable-certainty requirement. Section 352 of the Restatement (Second) of Contracts (1981) states the limit in a single sentence, confining a contract plaintiff to the loss the evidence establishes with reasonable certainty. The Supreme Court of Texas gave the same limit a working form in Signature Industrial Services, LLC v. International Paper Co., 638 S.W.3d 179 (Tex. 2022), quoting Phillips v. Carlton Energy Group, LLC, 475 S.W.3d 265, 278 (Tex. 2015): “Proof need not be exact, but neither can it be speculative.” Texas asks parties to supply “objective facts, figures, or data.”
The requirement bites on the amount rather than on the fact of harm, and the Supreme Court separated the two in Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555 (1931): “It is true that there was uncertainty as to the extent of the damage, but there was none as to the fact of damage; and there is a clear distinction between the measure of proof necessary to establish the fact that petitioner had sustained some damage and the measure of proof necessary to enable the jury to fix the amount.” As to the amount, the Court held that “while the damages may not be determined by mere speculation or guess, it will be enough if the evidence show the extent of the damages as a matter of just and reasonable inference, although the result be only approximate.”
Where the defendant’s own conduct created the uncertainty, courts have declined to let it profit from that. Story Parchment: “The wrongdoer is not entitled to complain that they cannot be measured with the exactness and precision that would be possible if the case, which he alone is responsible for making, were otherwise.” Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251 (1946) put it as a principle: “The most elementary conceptions of justice and public policy require that the wrongdoer shall bear the risk of the uncertainty which his own wrong has created,” and added that although the jury “may not render a verdict based on speculation or guesswork,” it “may make a just and reasonable estimate of the damage based on relevant data, and render its verdict accordingly.”
None of that dissolves the requirement, and reading it as permission is how a counterfactual becomes an excluded opinion. Whether a cause of action permits a given measure is a question of law for counsel, the certainty standard is stated by the forum rather than by the expert, and the Institute supplies no projection, rate or figure. How these constructions are attacked on admissibility is covered at Daubert and damages.