Public Procurement in Brazil: Evidence of frauds using the Newcomb-Benford Law (NB-Law)

• Autor(a) principal: Sampaio, Adilson da Hora
• Data de Publicação: 2022
• Outros Autores: Figueiredo, Paulo Soares, Loiola, Elisabeth
• Tipo de documento: Artigo
• Idioma: por; eng
• Título da fonte: Cadernos Gestão Pública e Cidadania (Online)
• Texto Completo: https://periodicos.fgv.br/cgpc/article/view/82760
• topic: public procurement, electronic auction, corruption, Newcomb-Benford law, bidding anomalies.; contratación pública, negociación electrónica, corrupción, ley Newcomb-Benford, anomalías en licitaciones.; compras públicas, pregão eletrônico, corrupção, lei de Newcomb-Benford, anomalias em licitações.

This article uses the Newcomb-Benford(NB-Law), or Law of Anomalous Numbers, to analyze the values of Electronic Bidding Processes that occurred in the Purchasing Portal of the Brazilian Federal Government and contained in the newly created website "DadosAbertos.gov.br". In the analysis, all services contracted in the period from 2014 to 2018 were considered. The objective of the research was to analyze the conformity of the Electronic Auction referred to the NB-Law, aiming to verify anomalies, which represent signs of fraud. It can be said that there was a statistically significant anomaly in the analysis of the first digit of the values bid in the Electronic Bidding Tenders. It is also noted that the trading sessions with the first digit of numbers 4, 8 and 9 are those with the largest differences between expected and observed values, strengthening the hypothesis that these represent the trading sessions with the highest incidence / probability of deviations, to be tested in future studies. The study, pioneer in this type of analysis in the Brazilian data source, aims to contribute to the literature focused on the detection of accounting or financial fraud in the public sector. The results collected here can also contribute to the practice of inspection in public management. It is recommended to deepen the studies based on the Economics of Corruption, on issues that involve the detection of fraud, so that corruption becomes unviable or inopportune.