In ruling on a case management/protective order, the U.S. District Court for the Western District of Pennsylvania recently turned down the taxpayer's request that the U.S. government be limited to statistical sampling with respect to documentation of business components in litigation over the taxpayer's research credit claim.
In a case covering tax years 1990 through 2006, Bayer Corporation (Bayer) sought to limit the government's discovery request for documentation on qualifying business components to a statistical sample. Bayer also asked the court to limit the evidence evaluated during the trial to those sampled items. The government objected, claiming it could not evaluate whether a statistical sample would be appropriate and sufficient before Bayer provided evidence of the qualification of its business components for the research credit.
The court held for the government, concluding that while statistical sampling and other estimation methods long have been considered acceptable methods for determining the characteristics of a large universe, there is no precedent for sampling to be forced upon the government in the liability phase of a trial. The court further noted that the government did not believe Bayer was capable of meeting its burden of identifying the business components for which the credits were claimed and that Bayer first needed to satisfy this burden before quantifying the credits.