| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 653.63 KB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Na literatura sobre a política de dividendos, não existe um modelo universalmente aceite para explicar o comportamento dos dividendos. O objetivo desta dissertação é identificar os fatores que a influenciam o pagamento de dividendos pelas instituições bancárias que operam em Portugal.
Para esse propósito, foi formulado um modelo de pagamento de dividendos e testado por meio de análise de regressão múltipla. O estudo utiliza os dados financeiros de 10 bancos no período entre 2005 e 2018. Os dados são ainda subdivididos em três subamostras, correspondentes aos períodos pré-crise (2005-2007), crise (2008-2012) e pós-crise (2013-2018). O modelo proposto é testado, para dados de painel, através do estimador panel least squares (PLS), com e sem efeitos fixos, por forma a avaliar o impacto de cada um dos determinantes específicos dos bancos sobre o indicador de dividendos pagos.
Conclui-se pela significância estatística global do modelo definido e de vários dos determinantes incluídos. Os resultados para a totalidade da amostra evidenciam que o pagamento de dividendos é impulsionado principalmente pela distribuição passada/histórica e pela influência positiva da performance em termos de rentabilidade (ROAt). Além disso, o tamanho do banco também se mostra significativo embora com sinal oposto ao esperado. O estudo evidencia ainda que os efeitos da maturidade e alavancagem têm menor importância estatística para explicar a política de pagamentos de dividendos. Os resultados são consistentes com a teoria de sinalização da política de dividendos e com o modelo de Lintner de estabilidade dos dividendos.
Outro resultado valioso está associado às alterações na política de distribuição de resultados antes, durante e no pós-crise financeira. O estudo suporta a hipótese de influência das crises na política de pagamento de dividendos. Os resultados fornecem evidências que os dividendos no período pós-crise foram condicionados pela política de retenção de resultados imposta pelas autoridades, e que os bancos de maior tamanho estão associados à retoma dos pagamentos. Já o indicador de alavancagem (ALAVt) apenas se apresenta estatisticamente significativo no período pré-crise, e com sinal positivo, contrariando a tese de que a pressão regulatória reduz os dividendos.
In the literature on the dividends’ policy, there is no universally accepted model to explain the dividends’ behaviour. The goal of this article is therefore to identify the factors that influence the payment of dividends by the banking institutions operating in Portugal. For this purpose, it was created a model for the payment of dividends, which was then tested with multiple regression analysis. The study uses the financial data of ten banks in the period from 2005 to 2019. The data is then subdivided in three subsamples, corresponding to the pre-crisis (2005-2007), crisis (2008-2012) and post-crisis (2013-2018) periods. The proposed model is tested for panel data using the Panel Least Squares (PLS) model, with and without fixed effects, in order to evaluate the impact of each of the specific bank determinants on the payed dividends indicator. It is then concluded that both the defined model and the included determinants are statistically significant. The results for the entire sample provide evidence that the payment of dividends is mainly explained by the historical payment rates as well as by the positive return performance (ROAt). Besides this, the size of the bank is also statistically significant. However, the correlation signal is the opposite of the one that could be expected. The study also provides evidence that the maturity and leveraging effects have less statistical significance in explaining the dividends’ payment policy. These results are consistent with the signalling theory of the dividends’ policy as well as with the Lintner’s model on the stability of dividends. Another insightful conclusion is associated with the changes in the policy of returns’ distribution on the periods before, during and after the financial crisis. The study supports the hypothesis of the crisis influence on the policy of dividends’ payment. In addition, the results provide evidence that the dividends on the post-crisis period were influenced by the policy of retaining returns imposed by the regulating authorities. The results further suggest that the largest banks are associated with the faster rate of payments normalization. On the other hand, the leveraging indicator (ALAVt) is only statistically significant in the pre-crisis period, with positive signal, which contraries the thesis that the regulatory pressure decreases the dividends’ payment.
In the literature on the dividends’ policy, there is no universally accepted model to explain the dividends’ behaviour. The goal of this article is therefore to identify the factors that influence the payment of dividends by the banking institutions operating in Portugal. For this purpose, it was created a model for the payment of dividends, which was then tested with multiple regression analysis. The study uses the financial data of ten banks in the period from 2005 to 2019. The data is then subdivided in three subsamples, corresponding to the pre-crisis (2005-2007), crisis (2008-2012) and post-crisis (2013-2018) periods. The proposed model is tested for panel data using the Panel Least Squares (PLS) model, with and without fixed effects, in order to evaluate the impact of each of the specific bank determinants on the payed dividends indicator. It is then concluded that both the defined model and the included determinants are statistically significant. The results for the entire sample provide evidence that the payment of dividends is mainly explained by the historical payment rates as well as by the positive return performance (ROAt). Besides this, the size of the bank is also statistically significant. However, the correlation signal is the opposite of the one that could be expected. The study also provides evidence that the maturity and leveraging effects have less statistical significance in explaining the dividends’ payment policy. These results are consistent with the signalling theory of the dividends’ policy as well as with the Lintner’s model on the stability of dividends. Another insightful conclusion is associated with the changes in the policy of returns’ distribution on the periods before, during and after the financial crisis. The study supports the hypothesis of the crisis influence on the policy of dividends’ payment. In addition, the results provide evidence that the dividends on the post-crisis period were influenced by the policy of retaining returns imposed by the regulating authorities. The results further suggest that the largest banks are associated with the faster rate of payments normalization. On the other hand, the leveraging indicator (ALAVt) is only statistically significant in the pre-crisis period, with positive signal, which contraries the thesis that the regulatory pressure decreases the dividends’ payment.
Description
Keywords
Dividendos Setor bancário português Crise Dados em painel Dividends Crisis Portuguese banking sector Panel data