A damages expert proves but-for causation by building the “but-for world” — a credible picture of what the plaintiff’s financials would have looked like had the wrong never happened — and then disaggregating the actual decline so that only the portion the defendant’s conduct produced is claimed, with every other force that moves a business (the economy, the industry, new rivals, the plaintiff’s own decisions, normal customer turnover, seasonality) accounted for and stripped out.
I am a CPA, Accredited in Business Valuation (ABV), and forensic accountant who works as an economic-damages expert witness. In that work, the question I hear most often from attorneys is not “how big is the loss?” It is the harder one underneath it: “can you show the defendant actually caused it?” A business can lose money for a hundred reasons. The job of a damages analysis is to separate the loss the defendant is responsible for from everything else that was happening at the same time. This article explains how I approach that separation — the financial side of causation — and where the line sits between my lane and counsel’s.
Causation Is Its Own Element, Not a Byproduct of Liability
It helps to start by naming the two distinct things a plaintiff has to prove. One is liability — that the defendant did something wrongful. The other is that the wrongful act caused a quantifiable financial loss. Lawyers often call the second piece “causation of damages,” and it is genuinely separate from the first. A defendant can be plainly in the wrong and still owe little or nothing, if the plaintiff cannot connect the wrong to a measurable harm.
I keep those two ideas apart in my own work, because they belong to different people. Whether the defendant breached a contract, infringed a patent, interfered with a relationship, or made a misrepresentation is a legal and factual question for counsel, the fact witnesses, and ultimately the trier of fact. I do not opine on liability. What I opine on is the economics: assuming the conduct occurred as alleged, what financial consequence did it produce, and how do I know it was that conduct rather than something else?
That distinction matters because it disciplines the analysis. If I treat liability as a given and then assume every dollar of decline flows from it, I have skipped the part of the work that actually earns the opinion. Courts notice when an expert does that, and so does opposing counsel. The causation element is exactly where a damages opinion is won or lost.
For the broader framework these pieces fit into, see my overview of economic damages. The discussion here drills into one element of that framework.
The But-For World: What Would Have Happened Without the Wrong
Every lost-profits analysis rests on a counterfactual. I have to describe a world that never occurred — the one in which the defendant’s conduct never happened — and estimate what the plaintiff’s revenue and profit would have been in it. That hypothetical world is the “but-for world,” and the damages are the gap between it and what actually happened.
Two estimates make up that gap:
- The but-for figure. What the business would have earned absent the wrong. Because this scenario is hypothetical by definition, it is built on assumptions, and every assumption has to be supported by something concrete — the company’s own history, comparable operations, industry data, or contemporaneous projections.
- The actual figure. What the business in fact earned (or, for losses that run into the future, a reasonable forecast of what it will earn) with the conduct factored in.
The loss is what is left when I subtract the second from the first. And it must be a loss of net profit, not revenue. Revenue that never came in did not carry its variable costs with it, so the costs the business avoided by not earning that revenue come out of the figure. Claiming lost revenue without removing avoided cost overstates the harm and invites a well-deserved attack on cross-examination.
The two main techniques for building the but-for figure are the before-and-after approach and the yardstick approach. I have written about how I choose between them in before-and-after vs. yardstick method, so I won’t re-derive the mechanics here. What matters for causation is this: neither method, by itself, proves the defendant caused anything. They produce a number. Causation is the separate work of showing that the number reflects the defendant’s conduct and not the dozen other things moving the business.
Disaggregation: Separating the Defendant’s Effect From Everything Else
Disaggregation is the heart of a causation analysis, and it is the step weak opinions skip. A real business is pushed and pulled constantly. The economy expands and contracts. The industry grows or shrinks. New competitors open down the street. Customers age out, move, or simply drift away. Sales rise and fall with the season. The owner makes a good decision one quarter and a poor one the next. When revenue drops after a defendant’s wrongful act, the act is rarely the only thing that changed — and my task is to figure out how much of the change belongs to it.
Among the forces I work to isolate and account for:
- The macroeconomy. A recession, a spike in interest rates, a shift in commodity or input prices — these move whole sectors at once and have nothing to do with the defendant.
- The industry. Some industries are contracting while others grow. A declining tide lowers the plaintiff’s boat regardless of the conduct at issue.
- New or stronger competition. Legitimate competition that erodes the plaintiff’s sales is not compensable. If a capable rival entered the market, that effect has to be carved out.
- The plaintiff’s own choices. A price increase, a service-quality slip, the loss of a key employee, a strategic decision that backfired — these belong to the plaintiff, not the defendant.
- Ordinary customer attrition. Every business loses some customers each year as a baseline. Only attrition above that baseline can be attributed to the wrong.
- Seasonality. A quarter-over-quarter comparison that ignores the plaintiff’s normal seasonal pattern can manufacture a “decline” that is just the calendar.
- Capacity limits. If the plaintiff could not physically have produced or served the additional volume it claims it lost, those profits were never available to lose.
The disaggregation question I ask of every downward movement in the numbers is the same: would this have happened even if the defendant had done nothing? Whatever survives that question is the candidate for damages. Whatever does not survive it comes out.
A Hypothetical Illustration: Splitting a Revenue Decline
The following figures are hypothetical. I invented them to show the mechanics of disaggregation; they are not drawn from any actual case or client, and no real engagement is implied.
Suppose a regional distributor earned $5,000,000 in revenue the year before the conduct at issue. In the loss year, revenue fell to $3,800,000 — a drop of $1,200,000. A careless analysis would simply call the whole $1,200,000 the defendant’s doing. A disaggregated analysis asks where each piece of that decline actually came from.
| Driver of the decline | Revenue effect | Attributable to defendant? |
|---|---|---|
| Industry-wide contraction (sector down ~8%) | $400,000 | No |
| New competitor opened nearby | $250,000 | No |
| Plaintiff discontinued a product line (own choice) | $150,000 | No |
| Remaining decline tied to the wrongful conduct | $400,000 | Yes |
| Total revenue decline | $1,200,000 |
In this illustration, only $400,000 of the $1,200,000 drop survives the causation test. The other $800,000 reflects forces that would have moved the business whether or not the defendant ever acted. To turn that $400,000 of lost revenue into lost profit, I then remove the variable costs the plaintiff avoided by not earning it — so if those costs ran 60% of revenue, the lost profit on this slice would be roughly $160,000, not $400,000. A damages number built this way can withstand scrutiny precisely because it does not ask the defendant to pay for the economy, the competition, or the plaintiff’s own decisions.
Proximate Cause and Foreseeability, in Plain Terms
Two legal ideas sit close to the financial analysis, and while they are counsel’s to argue, I keep them in view because they shape what a defensible number looks like.
The first is proximate cause — the requirement that the loss flow directly and naturally from the wrong, not from some remote chain of events several steps removed. The law does not make a defendant pay for harms only distantly connected to its conduct. As a practical matter, this pushes me toward losses I can trace cleanly back to the act, and away from speculative ripple effects that depend on a long string of contingencies.
The second is foreseeability — the idea that the kind of loss claimed was a reasonably probable consequence of the conduct, and in contract cases, within the parties’ contemplation when they made the deal. I am not the one who establishes what the parties foresaw, but the concept reminds me to keep the damages tethered to consequences a reasonable party would have expected, rather than to every conceivable downstream effect.
These standards run alongside the requirement that the amount itself be provable to a reasonable degree of certainty. I treat that certainty standard in depth in reasonable certainty for lost profits; here it is enough to say that causation and certainty work together. A loss can be real and still fail for lack of certainty, or be precisely calculated and still fail for lack of causation. A credible opinion has to clear both bars.
The Post-Hoc Trap: A Decline After the Wrong Is Not Proof the Wrong Caused It
The most seductive error in this work is the oldest one in logic: assuming that because B followed A, A must have caused B. In Latin it is post hoc, ergo propter hoc — “after this, therefore because of this.” In a damages context it looks like this: the defendant did something, the plaintiff’s profits fell afterward, therefore the defendant caused the fall. Stated that baldly, the flaw is obvious. Dressed up in a spreadsheet, it can slip past an inattentive reader.
Sequence is not causation. A business’s profits might have fallen in the same period for reasons that had nothing to do with the defendant — the very forces disaggregation is designed to catch. The fact that the decline came after the conduct establishes timing, and timing is necessary, but it is nowhere near sufficient. I treat a post-event decline as a question to investigate, never as an answer.
This is also why correlation is not causation. Two things can move together — the conduct and the decline — without one driving the other, especially when a third factor is moving both or when the alignment is coincidental. Showing that the lines on a chart bend at the same moment is a starting observation, not a conclusion. The conclusion has to come from ruling out the alternative explanations, one at a time, until the defendant’s conduct is the explanation that remains standing.
Intervening and Superseding Causes
Even when the defendant’s conduct genuinely set a loss in motion, something else can break the chain. Lawyers call these intervening and superseding causes — events that occur after the wrong and that either contribute to the harm or, in the stronger case, become the real cause of it.
A recession that arrives mid-loss-period is a classic example. So is a later, independent business decision by the plaintiff, a regulatory change, the departure of a key person for unrelated reasons, or a separate market shock. When one of these enters the picture, part or all of the continuing loss may belong to it rather than to the defendant.
I watch especially for the moment when a loss that started with the defendant’s conduct stops being about that conduct. Picture a business that suffers a real, defendant-caused hit, begins to recover, and then loses a beloved long-time employee whose departure drives away a block of loyal customers. A plaintiff may want to keep blaming the original defendant for the revenue that never returned. An honest analysis recognizes that the later, unrelated event is doing the work by then, and the damages clock attributable to the defendant has effectively stopped. Drawing that line — identifying when the defendant’s effect ends and an intervening cause takes over — is part of defining the loss period correctly.
Why Courts Throw Out Opinions That Ignore Obvious Alternative Causes
Judges act as gatekeepers for expert testimony, and one of the surest ways for a damages opinion to be excluded or gutted is to ignore the plain-sight alternative explanations for a loss. I have seen the pattern many times in the case law that governs this work: an expert compares “before” to “after,” finds a gap, attributes the whole gap to the defendant, and never seriously asks what else was going on. Courts have rejected exactly that — opinions that made no effort to separate the defendant’s effect from unrelated business factors, that assumed continued growth with no support, or that failed to account for competition, capacity, or the plaintiff’s own role.
The lesson I take from those decisions is not defensive box-checking. It is that the alternative-cause analysis is the opinion. An expert who refuses to engage with the other forces moving the business has not produced an unreliable opinion at the margins; the opinion has no reliable foundation at all, because it never did the one thing that distinguishes the defendant’s harm from background noise. Reliability comes from confronting the competing explanations head-on and showing, with evidence, why they do not account for the loss I am attributing to the conduct.
How I Actually Demonstrate Causation
Stripped to its essentials, demonstrating causation is a process of elimination supported by affirmative proof. These are the techniques I lean on.
Control and Benchmark Comparisons
The strongest causal evidence often comes from a comparison that holds the defendant’s conduct as the main thing that differs. If I can find a benchmark — the plaintiff’s own unaffected location, a closely comparable peer business, a relevant industry index, or the same company in a period free of the conduct — and that benchmark did not suffer the decline the plaintiff did, the contrast points to the conduct. The benchmark functions like a control group: it absorbs the common forces (the economy, the industry) so that what remains is more plausibly the defendant’s signal. The comparison only carries weight if the benchmark is genuinely similar in size, market, customer base, and other relevant respects; a poorly matched benchmark proves nothing, and courts say so.
Eliminating Alternative Explanations One by One
Most of my causation work is methodical subtraction. I list every plausible non-defendant explanation for the decline — recession, competition, the plaintiff’s pricing, attrition, seasonality, a discontinued product, a lost contract — and I work through each, using the financial records, industry data, and management’s own contemporaneous documents to size it or rule it out. What I cannot attribute to any of those alternatives is what I attribute to the conduct. This is the disaggregation discussed above, applied as a disciplined checklist rather than a single sweeping assumption.
Regression and Statistical Methods, Where the Data Support Them
When there is enough clean data, a regression analysis can quantify how much of a movement in the plaintiff’s results is explained by measurable factors — say, overall market demand, price, or a seasonal pattern — and how much is left unexplained once those are accounted for. Used properly, regression is a powerful way to isolate the defendant’s contribution from the rest. Used carelessly, it manufactures false precision. I reach for it only when the data are adequate and the model is genuinely appropriate to the facts, and I am always prepared to explain, in plain terms, what the analysis does and does not show.
Grounding Everything in Real Evidence
None of these techniques work on assumption alone. Causation opinions live or die on the underlying record — the plaintiff’s historical statements and tax returns, contemporaneous projections prepared before anyone was thinking about litigation, customer and order data, comparable-business results, and credible industry and economic sources. I verify what I rely on rather than parroting what a client hands me, because an opinion connected to the data only by my say-so is exactly the kind courts exclude. The discipline of independent verification is also what lets the opinion survive cross-examination intact.
How Causation Connects to the Rest of the Damages Analysis
Causation does not stand alone. It shapes — and is shaped by — the other pieces of a lost-profits claim. The duty to mitigate, for instance, interacts directly with causation: losses the plaintiff could reasonably have avoided are not chargeable to the defendant, which is a causation-flavored limit on the claim. I cover that in mitigation of damages. And in matters where the proper remedy is the defendant’s gain rather than the plaintiff’s loss, the causal question shifts to what profit the defendant earned from the wrong, which I address in unjust enrichment and disgorgement of profits.
For the full methodology of building a lost-profits number from the ground up, see lost profits damages for Florida commercial litigation. And if you are evaluating whether a damages opinion will hold up — or need one built to withstand challenge — that is the heart of my expert witness and litigation support practice. Across all of it, the through-line is the same: a number is only as good as the causal story that supports it.
FAQ
What is but-for causation in a damages case?
But-for causation asks whether the claimed loss would have occurred but for the defendant’s wrongful conduct. The damages expert answers it by constructing a hypothetical “but-for world” — what the plaintiff’s finances would have looked like had the wrong never happened — and measuring the gap between that world and what actually occurred, after removing the effect of every other force that moved the business.
How is causation different from simply showing a loss?
Showing a loss only proves that the plaintiff’s profits fell. Causation proves why they fell. A business can lose money because of a recession, new competition, its own mistakes, or normal customer turnover, none of which a defendant is responsible for. The causation analysis separates the portion of the loss the defendant produced from the portion that would have happened anyway.
What is disaggregation and why does it matter?
Disaggregation is the process of breaking a decline into its causes and attributing each piece to its actual source — the economy, the industry, competitors, the plaintiff’s own decisions, seasonality, attrition, and finally the defendant’s conduct. It matters because only the portion tied to the conduct is recoverable. An analysis that skips disaggregation and blames the entire decline on the defendant tends to be excluded or sharply discounted by courts.
Doesn’t a drop in profits right after the wrong prove the wrong caused it?
No. That reasoning is the post-hoc fallacy — assuming that because the decline came after the conduct, the conduct caused the decline. Timing is necessary but not sufficient. The decline still has to be tested against every other explanation that was active in the same period. Correlation and sequence are starting points for investigation, never proof of causation.
What is an intervening or superseding cause?
It is an event occurring after the defendant’s conduct that contributes to the harm or becomes its real cause — a later recession, an independent decision by the plaintiff, a regulatory change, or an unrelated market shock. When such an event takes over, part or all of the continuing loss belongs to it rather than to the defendant, which often defines where the loss period attributable to the defendant ends.
How can I discuss a potential causation analysis with you?
You can reach me at 954-282-9615 to talk through the financial side of causation in your matter. This kind of work is typically billed hourly, at approximately $400 per hour, the Florida market average. I am glad to discuss how the loss in your case might be separated from the ordinary forces that move any business before you decide how to proceed.
About the Author
Joey Friedman is a CPA, Accredited in Business Valuation (ABV), and forensic accountant who holds a Master of Accounting and a Master of International Business and is a member of the AICPA and the Association of Certified Fraud Examiners. He also holds a Florida real estate license. Beyond those credentials, he has personally owned and operated more than a dozen of his own businesses across industries including marketing, printing, transportation, restaurants, hospitality and entertainment, and event planning — so he separates the loss a defendant actually caused from the ordinary noise of the market with both a forensic accountant’s discipline and an operator’s first-hand grasp of the many forces that move a real business’s results.