In a recent paper in this Journal, Gaumnitz et al. [1982] provided evidence on the consensus of auditors' judgments which seemed to reconcile conflicting results reported in Ashton [1974] and Joyce [1976]. In their study Gaumnitz et al. [1982] combined judgment tasks of evaluating internal control and recommending the number of audit hours needed to assess the propriety and collectibility of accounts receivable. They reported mean correlations of 0.704 and 0.617 for internal control and audit planning judgments, respectively. The former was consistent with Ashton [1974] and the latter was higher than observed by Joyce [1976], who reported a mean correlation of only .373 for audit hour of an accounts receivable subsystem. Gaumnitz et al. suggested that by linking the two judgment tasks they reinforced the link between internal control and planned number of audit hours. Specifically, they stated: link is provided by the auditor's knowledge that an inverse relationship should exist between the strength of internal control and the number of audit hours planned. Thus, the requirement to provide an explicit judgment on the strength of internal control enabled our subjects to operationalize this inverse relationship, resulting in a high correlation (relative to Joyce's findings) among audit hour estimates [1982, pp. 753]. This explanation appears plausible given the results of previous audit judgment research. That is, high levels of consensus have been found for

We investigate whether management's decision regarding the recognition of the valuation allowance (VA) for deferred tax assets provides incremental information about the persistence of accounting losses. We introduce a classification scheme that assigns loss firm‐years into three categories based on whether management appears to have recognized a material change in the VA, and whether or not the firm has positive taxable income (e.g., a net operating loss). The results of our study show that our tax categories contain information about the persistence of accounting losses over the following three years beyond variables previously identified to predict loss persistence. This incremental information is consistent with management using private information about the firm's future prospects in setting the VA. Finally, we find that investors’ pricing of the VA varies with the saliency of the tax signal and the information environment of the firm.