If annual observations are used for the sample of plants, an average figure for costs (e.g., per labor hour or ton of very severe inflation-hyperinflation-would the averaging of annual costs become a significant problem, but any studies undertaken under such conditions would surely be of dubious quality in any case.