Disclosure do guidance e excesso de confiança : análise da relação entre projeções confiantes e o valor de mercado das empresas

Abstract

This research investigates the relationship between managerial overconfidence, the practice of Guidance Disclosure (voluntary forecasts), and the market value of companies listed on B3, covering the period from 2018 to 2024, resulting in a sample of 2,058 observations. The study's primary contribution lies in its integrative methodological approach for measuring the behavioral phenomenon of overconfidence. It proposes a dual metric, novel in the Brazilian context, which combines an internal index—constructed through Principal Component Analysis (PCA) from characteristics of the CEO, CFO, and Investor Relations Director, including a proxy based on social media exposure and activity—and an external measure, based on corporate decisions (investment, acquisitions, and indebtedness). Guidance was measured using a multidimensional checklist applied to the companies' Reference Forms, overcoming traditional binary approaches. To address the censored nature of Disclosure data and potential endogeneity issues, robust econometric models were employed, including Tobit regressions and the Generalized Method of Moments (GMM). The empirical results indicate a negative and statistically significant relationship between overconfidence and the Disclosure of Guidance, leading to the rejection of hypothesis H1, which predicted a positive association. This result suggests that more confident managers tend to disclose fewer formal projections, possibly because they believe the market can already infer their future performance without the need for additional signals. This negative relationship was more pronounced for financial projections (H2), indicating that the behavioral bias can exert a differentiated influence on Disclosure components more closely associated with performance and market expectations. The association with investment projections (H3), while present, showed less statistical robustness. Finally, hypothesis H4 was confirmed, pointing to a negative association between projections influenced by overconfidence and the market value of the companies, which indicates that the market may negatively adjust pricing in response to communications perceived as overly optimistic. In summary, this research advances the integration between Behavioral Finance and the voluntary Disclosure literature by showing that the absence of Disclosure can also constitute a relevant behavioral signal. Methodologically, it presents an original and replicable framework. From a practical standpoint, the results reinforce the importance of considering managers' behavioral characteristics in the analysis of voluntary corporate communication and its impact on firm market value.