home  /  the but-for world  /  before-and-after & yardstick
the counterfactual · economic damages

Before-and-after, and the yardstick.

Compare the plaintiff to its own past, or to someone else’s present. Either way, the comparison is the case.

begin here

What is the economic harm?

Start a conversation with Cournot, the Institute’s damages concierge, already scoped to before-and-after & yardstick. Pick a starting point, or describe the dispute directly.

Cournotbefore-and-after & yardstick · a diagnostic, not a damages opinion
Tell me about the business, its history before the conduct, and any comparable firms or markets you have in mind. I'll help you think about which comparison the evidence supports and where each would be attacked.

The comparison methods answer the counterfactual question with evidence about worlds that actually existed. Before-and-after compares the business to its own performance in an unaffected period: the years before the conduct, sometimes the period after it ended. The yardstick compares it to something else that was never affected: a similar firm, a portfolio of them, an industry index, the same company's operations in another region. Modern practice frequently runs these through regression, using the clean period or the clean comparator to fit a model of what drives the plaintiff's results, then projecting that model through the damages period, so that the forecast inherits its discipline from data rather than from judgment alone. Difference-in-differences combines both ideas, measuring how the gap between plaintiff and comparator changed when the conduct began. Every version lives or dies on one question: is the comparison clean? A before period contaminated by other changes, or a yardstick firm that differs from the plaintiff in ways that matter, transmits its flaws directly into the damages number, and opposing experts are paid to find exactly that.

mechanisms

The comparison toolkit.

Different machinery, one shared vulnerability: the quality of the comparison.

Before-and-after

The business against its own unaffected period. Strongest with a long, stable history and a clean break at the conduct.

The yardstick

A comparable firm, group of firms or index that was not affected. The choice of comparator becomes the battleground.

Regression forecasting

A model fitted on clean data and projected through the damages period, so the counterfactual is driven by measured relationships.

Difference-in-differences

How the plaintiff-versus-comparator gap moved when the conduct began. Controls for shocks that hit both.

Period selection

Which months are "before", "during" and "after". Small changes here can move the number materially, and everyone knows it.

Contamination

Anything else that changed at the same time: management, products, the market. The standard attack on both methods.

methodology

What the evidence shows — and what we examine.

How the Institute approaches a comparison-based claim.

Comparator auditWhether the proposed benchmark is comparable on the dimensions that drive the outcome, not merely the same industry code.
Period logicWhy the clean period starts and ends where it does, defensible independent of the number it produces.
Model scrutinyWhere regression is used: specification, fit in the clean period, and sensitivity to reasonable alternatives.
Contamination sweepA deliberate hunt for everything else that changed in the damages period, done before the opposing expert does it.
what's at stake

What the comparison decides.

The method is only as strong as the comparison inside it.

the credibility of the counterfactual admissibility of the forecast what comparator data must be obtained the size of the claim exposure to the contamination attack whether econometric expertise is needed

A yardstick is an argument, not a lookup.

Naming a comparable is the beginning of the work. The expert has to show comparability on the dimensions that drive the result, size, market, cost structure, trajectory, and address the differences rather than hope they go unnoticed. The other side will build its case on the differences.

common questions

The comparison methods: practical questions.

When is before-and-after the wrong method?

When there is no clean before, which happens more often than the method's popularity suggests. A young business has no before at all. A business that changed materially, new products, new management, an acquisition, a different scale, has a before that measures a different company. And a before period that ends in an unusual peak or trough builds a bias directly into the baseline. The method also assumes the before conditions would have continued, so an industry in structural decline makes the plaintiff's own past an overstatement of its but-for future. Each of these is fixable in principle, with adjustments or a hybrid method, but each fix is a new assumption to defend.

What makes a yardstick genuinely comparable?

Similarity on the dimensions that drive the outcome being measured, which is a far stricter test than sharing an industry. A useful yardstick resembles the plaintiff in market, scale, cost structure, customer mix and growth stage, and was exposed to the same general conditions during the damages period while being untouched by the conduct. Portfolios of comparators often beat single firms, because idiosyncratic noise averages out and the selection looks less like cherry-picking. What kills yardsticks is asymmetry: a comparator chosen because it grew, differences acknowledged only where they help, or a benchmark whose own results were affected by the same conduct, which quietly erases the damages it was meant to reveal.

Does regression make these methods stronger or just more complicated?

Stronger, when the underlying comparison is sound, because the model makes the reasoning explicit and testable. A regression fitted on the clean period shows which factors actually drove the plaintiff's results and by how much, and the projection through the damages period inherits those measured relationships instead of an expert's eyeballed trend. It also enables difference-in-differences, which absorbs shocks that hit plaintiff and comparator alike. What regression cannot do is rescue a bad comparison: a model fitted to a contaminated period or benchmarked to a wrong comparator is the same flawed argument with more decimal places, and the added complexity gives a skilled opposing expert more specification choices to attack. The machinery raises the ceiling and the stakes together.

How is the choice between the plaintiff’s past and a benchmark made?

By the evidence, and ideally by both. The practical sequence is to ask first whether a clean, representative before period exists; if it does, before-and-after is usually the anchor because a fact-finder trusts a company's own record more than a constructed comparison. A yardstick then serves as corroboration, showing that unaffected peers continued on the path the plaintiff's history implies. Where the two methods point the same way the claim is much harder to dismiss, and where they diverge you have found, in private, the argument the other side would otherwise have found in public. Where neither comparison is available the analysis moves to direct construction, which is its own subject.

related

Related specialization areas & resources.

Find the flaw in the comparison first.

Describe the business and the candidate comparisons. The Institute will help you test them the way the other side will.

Cournotdiagnostic · not a damages opinion
Tell me about the business, its history before the conduct, and any comparable firms or markets you have in mind. I'll help you think about which comparison the evidence supports and where each would be attacked.