How Well Do Compustat NOL Data Identify Firms with U.S. Tax Return Loss Carryovers?
Although Compustat net operating loss (NOL) data are an underlying component of most proxies of corporate tax incentives, there is little empirical evidence regarding how Compustat NOLs relate to firms' tax-loss positions per their U.S. tax returns. We use confidential U.S. tax return data for a sample of large corporations over the period 1981–1995 to compute firms' U.S. tax-loss carryovers and construct a matched sample with Compustat data. We then evaluate how well Compustat NOL data works as an indicator of firms' U.S. tax-loss positions, identify sources of misclassification errors, and investigate the effectiveness of using additional Compustat data screens. We find that screening for U.S. current income tax or total pretax income successfully reduces certain types of misclassification errors. We conclude that care should be used in constructing U.S. tax-loss proxies, particularly when the research setting involves firms with foreign operations or corporate acquisitions activity.