UC Berkeley

ABSTRACT We revisit evidence that conditional conservatism aggregates up from the firm level causing corporate profit estimates in the National Income and Product Accounts (NIPA) to be more sensitive to negative relative to positive aggregate return news. Our study delivers three main messages. First, annual estimates of NIPA corporate profits are designed to provide neutral estimates of income from current production regardless of whether times are good or bad. Second, we find spurious evidence of asymmetric sensitivity to aggregate return news in placebo test variables that are void of conditional conservatism, clearly indicating that an alternative explanation is in order. Third, we find that the variance of aggregate returns decreases with the level of economic activity. Although this return variance effect is unlikely to be attributed to accounting, it has the potential to generate spurious evidence of conditional conservatism at the macro level. Overall, our study highlights the importance of construct validity tests in archival studies and contributes to research in the properties of NIPA corporate profits. JEL Classifications: M41; G12; E43; E52. Data Availability: Data are available from the sources identified in the text.