Banca de DEFESA: WELLINGTON ALVES DE OLIVEIRA

RESULTS MANAGEMENT IN BRAZILIAN MUNICIPALITIES: PUBLIC SECTOR PERSPECTIVE.

Results management; Accruals in the public sector; Agency Theory; Brazilian municipalities

The study aimed to analyze whether there is earnings management in the Brazilian public sector under the accrual basis through municipal accounts. To this end, a systematic review of the literature was carried out in order to verify the possibility of EM in the public sector, since some authors disagree on this topic. Afterwards, the literature on earnings management in the private sector was analyzed to identify the main proxies used and how to associate them with the public sector. And finally, the analysis of earnings management in the Brazilian public sector was carried out through the proposed models. Thus, it was possible to identify that the international literature has already analyzed this possibility in a broader and more concrete way, with several studies on the subject. There are several forms of agency conflict that define the opportunistic behavior of managers beyond the simple contractual relationship between principal and agent. With the analysis of the literature on earnings management in the private sector, it was possible to identify errors that have been made by researches when not adequately selecting the method of choosing discretionary accruals and that the concept of discretionary accruals is the same for the public and private sectors. It was also identified in the literature that the most appropriate model for management testing is comprehensive accruals according to Richardson et al (2005), Larson et al (2018) and Christensen et al (2022). Some improvements were formulated to test the hypothesis of Richardson et al (2005) with the inclusion of the suggestion of the Dechow and Dichev (2002) model of past and future cash flows as explanatory variables. To this end, information was collected from the financial statements of municipalities in Brazil in the SICONFI system for the period from 2015 to 2022, in the 5,565 municipalities in Brazil registered in 2022. The results allow us to highlight that the data presented by the federated entities are not of great quality, such as interrupted series of permanent accounts, missing data, among other problems. The accruals selection model tests also showed that the inclusion of lagged and future cash flows did not present statistical significance with the other variables, therefore, the autocorrelation between them and the period flow cannot be treated by the POLS method. Immediately after their exclusion, the model proved to be significant, confirming that the change suggested by Christensen et al (2022) can partially solve the problem of omitted variables. This result allows us to infer that the use of comprehensive accruals solves part of the problem of the accruals selection method, which was the weak significance found. However, the problem related to the way this variable is selected remains, which is the indirect way in which it is generated, without direct observation of the data. The findings of the EM verification model suggest the presence of earnings management in Brazilian municipal administrations. The size variable informs how much municipalities manipulate accounts regardless of their size. In addition, management occurs through debt, in the search for better performance and through the recognition of uncollectible tax revenues. Thus, it is expected that this study will be a milestone in the discussion of results management models in the Brazilian public sector.
Enviar feedback
Painéis laterais
Histórico
Salvas