Abstract Modern macroeconomic theory and central bank practices have made expectations management a key ingredient in successful monetary policy. However, the challenge is greater in emerging economies, which are prone to instability and crises. Given that Brazil has frequently experienced major economic fluctuations, we study the behavior and forecasting performance of inflation expectations in a way that is robust to the presence of instabilities. The proposed tests provide more evidence against the rationality of the forecast for survey-based private forecasts. Furthermore, they uncover an informational or methodological advantage behind the medians of inflation expectations reported in the last five business days, and in forecasts in which the distance from the publication date of the observed IPCA prevails in relation to the projection time in the survey.

Abstract This paper uses Benford’s laws, BLs, as an analytical framework to explore the financial data of FTSE 100 companies, explicitly examining tax-related information to identify potential accounting irregularities that could indicate fraud or data manipulation. We utilise data from Refinitiv EIKON, focusing on revenue, profit before taxes, income taxes, payable income taxes, and deferred taxes from 2014 to 2023. Our analysis employs various testing tools, including the chi-square, the mean absolute deviation, called MAD, and the z -test, examining three data sets: the first-digit, the second-digit, and the first-and-second-digit data set across five specified variables, in order to test the various BLs, i.e. BL1, BL2, and BL12, respectively. The results from the MAD test reveal that revenue and profit before taxes exhibit the most notable irregularities, leading us to reject the null hypothesis for BL1 and BL12. Importantly, findings from BL12 indicate that none of the analysed variables meet conformity levels. This implies that when adjustments are made to meet benchmark requirements in order to avoid scrutiny, compliance with BL12 regulations becomes more challenging. Consequently, irregularities are more apparent in the first-and-second-digit data set compared to using only the first- or second-digit data set. However, conclusions on conformity (or not) to BLs should take into account chronological correlations between digits. Yet, failure to adhere to BLs may indicate that the financial statements of FTSE 100 companies are susceptible to inconsistencies or irregularities, which could result in potential tax discrepancies and financial issues.