There’s More than One Way to Skin a Cat: Construction Claims by Extrapolation
Trial by sample is a critical tool in the hands of the Court’s case managing judges, without it the Court would struggle. In the recent case of Standard Life Assurance Ltd v Gleeds (UK) & Others [2020] EWHC 3419 (TCC), the Plaintiff advanced a ￡38 million professional negligence claim against the defendants (the design team), which comprised of a ￡25 million claim extrapolated from 122 samples taken from 3,600 variations. The Plaintiff contended that due to the defendants’ negligence, it had paid the Contractor more than it would otherwise have had to pay. Ultimately, the Plaintiff wanted the Court to draw an inference that based on the 122 samples taken; other variations were caused by the negligence of the defendants. The defendants brought an application to strike out and dismiss the extrapolated part of the claim on the grounds that the extrapolated claim had no prospect of success. The plaintiff contended that extrapolation is permissible and that it would be disproportionate to plead and prove its case separately in respect of each and every component of the claim. The plaintiff claimed that there was no scientific right or wrong way for such a claim to be tried. Counsel advocated that there is ‘more than one way to skin a cat.’
Kerr J accepted on the facts that the defendants had demonstrated that inferences concerning certain matters in the pleaded claim could not be drawn to support parts of the extrapolated claim. Kerr J noted that a balance had to be struck between allowing the plaintiff redress in the Court, at a proportionate cost and within a reasonable time, and allowing the defendants know the case against them. Although, acknowledging that the directions given were not scientific, Kerr J directed that, inter alia, the plaintiff narrows down the variations it was pursuing and allowed the defendants nominate additional variations to be sampled.
In giving the directions Kerr J gave some useful guidance around using extrapolation to support a construction claim. Kerr J noted that:
‘114. A case where performance is chaotic or close to non-existent is much less well suited to extrapolation than one where the defendant is alleged to have mass-produced the same error, so that the effect of the error can reasonably be proved by sampling and statistical analysis leading to extrapolated losses …
115. … I reject … that extrapolation cases are confined as a matter of law to cases where the pleaded breach or breaches are "systemic." There is no magic in that adjective. It is a convenient label to describe cases where the defendant operates a system that reproduces an error many times over … Extrapolation may also be legitimate in other kinds of case, applying ordinary principles of evidence including the drawing of permissible inferences from primary fact.
116. For example, where the defendant operates no "system" at all but fails to perform properly in the same way repeatedly – for example, by failing to read documents properly or report properly – in relation to different aspects of the works, the court could, if appropriate, draw an inference from one part of the works to another ..
117. If the claimant proves that building, say, two or three of the ten houses has cost double what it should have cost, through negligent lack of reporting and lack of control over cost overruns and delay, the court may be willing to draw the same inference in the case of the other seven or eight houses, absent rebutting evidence from the defendant to differentiate the other houses …’