Loading...
Publication
Variáveis que influenciam o endividamento dos municípios: o caso dos municípios do distrito de Setúbal
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 2.56 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O objetivo desta dissertação de mestrado, consiste em analisar as variáveis, que podem influenciar o endividamento dos municípios do distrito de Setúbal.
Procurou-se identificar algumas variáveis, nomeadamente número de habitantes, despesas com pessoal, ativo corrente, ativo fixo, receitas de capital e receitas correntes, formalizou-se algumas questões, com o objetivo de perceber como é que as variáveis selecionadas influenciam o endividamento dos Municípios do distrito de Setúbal, num horizonte temporal de oito anos, de 2010 a 2017.
Os dados foram obtidos através do Anuário Financeiro dos Municipios Portugueses, base de dados do Portal Autárquico, e dos relatórios e contas das Autarquias em estudo.
Com o objetivo de responder às questões formuladas, numa primeira abordagem, optou-se por calcular o coeficiente de correlação linear de Pearson, entre as variáveis independentes e a variável dependente, o endividamento. Esta medida estatística, permite quantificar a relação linear entre duas variáveis, e perceber se a relação é positiva ou negativa, primeiro a análise foi feita em termos globais, e posteriormente por dimensão de município. Concluiu-se, que no global, todas as variáveis independentes selecionadas, tem uma correlação positiva, com a variável endividamento.
Numa segunda abordagem, optou-se por utilizar um modelo de regressão linear múltipla, com o objetivo de perceber, de que forma, as variáveis explicativas influenciam o endividamento, e obter um modelo, que permita estimar o endividamento, em termos globais e por dimensão do município. Concluiu-se que no modelo global, as variáveis “número de habitantes”, “despesas com pessoal”, “ativo corrente” e “receitas de capital” influenciam positivamente o endividamento, sendo negativo o impacto da variável “receitas correntes” e “ativo fixo”. No estudo por dimensão, algumas variáveis não são consideradas significativas, e nem sempre seguem o mesmo sinal do modelo global.
The main goal of this dissertation is to analyse, the variables that may influence, the indebtedness of the municipalities in Setúbal’s district. Firstly, some relevant variables were identified, namely the number of inhabitants, personnel expenses, current assets, fixed assets, capital income and current income, real estate, and capital and current expenses. Secondly, a few questions were formalized in order to help explain the dependent variable, the indebtedness of the municipalities in Setúbal’s district, in a timeline of eight years, from 2010 to 2017. The dataset used was gathered from different sources namely the Financial Annuary of Portuguese Municipalities, Municipalities Portal database and the financial statements from the municipalities in study. In an initial approach, it was decided to calculate Pearson’s linear correlation coefficient between the independent and dependent variables. This coefficient enables us to quantify the linear relation between the dependent and the independent variables, allowing us to understand if the relation is positive or negative. This analysis was performed both globally, across all municipalities, and individually, for each municipality. It was concluded that globally all the independent variables in study have a positive correlation with the dependent variable. In a subsequent approach, a multiple linear regression model was used to understand how the dependent variables impact the municipalities indebtedness both at a global level, across all the municipalities, as well as at individual level, in each municipality. In the global model it was concluded that the variables number of inhabitants, personnel expenses, current assets and capital income positively impact the indebtedness, on the other hand, the variables current income and fixed assets have a negative impact on the dependent variable. In the individual model, for each municipality, it was concluded that some variables are considered statistically non-significant and do not always follow the same order of impact as in the global model (e.g. a variable with positive impact in the global model may have negative impact in some individual models).
The main goal of this dissertation is to analyse, the variables that may influence, the indebtedness of the municipalities in Setúbal’s district. Firstly, some relevant variables were identified, namely the number of inhabitants, personnel expenses, current assets, fixed assets, capital income and current income, real estate, and capital and current expenses. Secondly, a few questions were formalized in order to help explain the dependent variable, the indebtedness of the municipalities in Setúbal’s district, in a timeline of eight years, from 2010 to 2017. The dataset used was gathered from different sources namely the Financial Annuary of Portuguese Municipalities, Municipalities Portal database and the financial statements from the municipalities in study. In an initial approach, it was decided to calculate Pearson’s linear correlation coefficient between the independent and dependent variables. This coefficient enables us to quantify the linear relation between the dependent and the independent variables, allowing us to understand if the relation is positive or negative. This analysis was performed both globally, across all municipalities, and individually, for each municipality. It was concluded that globally all the independent variables in study have a positive correlation with the dependent variable. In a subsequent approach, a multiple linear regression model was used to understand how the dependent variables impact the municipalities indebtedness both at a global level, across all the municipalities, as well as at individual level, in each municipality. In the global model it was concluded that the variables number of inhabitants, personnel expenses, current assets and capital income positively impact the indebtedness, on the other hand, the variables current income and fixed assets have a negative impact on the dependent variable. In the individual model, for each municipality, it was concluded that some variables are considered statistically non-significant and do not always follow the same order of impact as in the global model (e.g. a variable with positive impact in the global model may have negative impact in some individual models).