Historical cost measurement and the use of DuPont analysis by market participants

Abstract

We investigate whether historical cost measurement of assets lowers the usefulness of DuPont analysis for investors. Because firms report assets at modified historical cost under US GAAP, accounting ratios can be biased upward when assets have appreciated. Thus, variation in asset turnover, which is the DuPont ratio most affected by asset measurement, can be due to both economic forces and measurement effects. We assess the extent of measurement effects using the average age of a firm’s assets and find that asset turnover ratios are higher and more persistent for firms with older assets. Forecast errors of asset turnover are associated with the change in asset age, and these forecast errors are positively associated with contemporaneous and future returns. Our results are weaker in non-US samples, in part reflecting deflation and upward revaluations, consistent with our US results capturing biased asset turnover ratios due to historical cost measurement.