To mitigate the growing possibility of fraud and corruption in Public Administration, considering the rapid evolution of information technology currently used, tools that enable effective and continuous control of public accounts must be found. Therefore, the objective of this work was to empirically verify how Newcomb-Benford's Law (NB Law) could contribute to the analysis and control of public expenditure. To this end, a quantitative and exploratory analysis was conducted using an econometric model, based on the relationship between the aforementioned Law and the Chi-square and Z-tests, evaluating 93,825 purchase orders from a public institution in the state of São Paulo, from 2017 to 2019, a period in which legislative changes influenced expenditure behavior. Overall, a discrepancy was observed between the collected data and the law, with considerable statistical variations for the first digit 7 in the 2017 and 2018 fiscal years, evidencing possible evasion of bidding processes; and smaller differences for the digits extracted from the 2019 expenses, the year with the updated legislation, confirming that the change in limit values for the acquisition of products and services had a profound impact on the distribution of the digits related to public spending by the entity. The law makes important contributions to the evaluation and monitoring of public spending, serving as an auxiliary instrument for control work, especially with regard to planning and defining prominent data to be verified in the identification of relevant distortions.
