Record Details

Explaining DEA Technical Efficiency Scores in an Outlier Corrected Environment: The Case of Public Services in Brazilian Municipalities

Brazilian Review of Econometrics

View Archive Info
 
 
Field Value
 
Title Explaining DEA Technical Efficiency Scores in an Outlier Corrected Environment: The Case of Public Services in Brazilian Municipalities
Explaining DEA Technical Efficiency Scores in an Outlier Corrected Environment: The Case of Public Services in Brazilian Municipalities
 
Creator de Sousa, Maria da Conceição Sampaio; Departamento de Economia, Universidade de Brasília, Campus Universitário Darcy Ribeiro s/n, Brasília-DF, 70910-900
Cribari-Neto, Francisco; IBMEC-RJ and BNDES.
Stosic, Borko D.; Departamento de Estatística e Informática, Universidade Federal Rural de Pernambuco, Rua D. Manoel de Medeiros s/n, Dois Irmãos, Recife – PE, 52171-900
 
Description We estimate DEA (Data Envelopment Analysis) technical efficiency scores for nearly five thousand Brazilian municipalities using a recently proposed method that combines bootstrap and jackknife resampling to eliminate the influence of outliers and possible measurement and recording errors in the data. We use both the constant and variable returns to scale variants of the DEA method. After computing the efficiency scores, we use econometric methods to investigate the determinants of those scores. The results reveal that: (a) the level of computer usage (a proxy for administrative capability) tends to have a positive and highly significant impact on efficiency; such gains seem to be greater for the most inefficient municipalities, thus indicating the existence of diminishing marginal benefits; (b) the urbanization rate is also shown to be positively correlated with the efficiency measures; (c) excess spending due to royalty revenues and economies of scale seem to explain why efficiency scores increase with the size of the municipality; (d) municipalities which take part in the Alvorada Program (a federal program for low income communities) tend to display higher efficiency scores than those they would have achieved had they not been included in the project. Additionally, the results show that the more power awarded to municipal councils, the better the effectiveness in resource utilization as measured by efficiency indices. This is because such councils tend to increase the transparency of the budgeting process, hence reducing corruption and the misuse of local funds.
We estimate DEA (Data Envelopment Analysis) technical efficiency scores for nearly five thousand Brazilian municipalities using a recently proposed method that combines bootstrap and jackknife resampling to eliminate the influence of outliers and possible measurement and recording errors in the data. We use both the constant and variable returns to scale variants of the DEA method. After computing the efficiency scores, we use econometric methods to investigate the determinants of those scores. The results reveal that: (a) the level of computer usage (a proxy for administrative capability) tends to have a positive and highly significant impact on efficiency; such gains seem to be greater for the most inefficient municipalities, thus indicating the existence of diminishing marginal benefits; (b) the urbanization rate is also shown to be positively correlated with the efficiency measures; (c) excess spending due to royalty revenues and economies of scale seem to explain why efficiency scores increase with the size of the municipality; (d) municipalities which take part in the Alvorada Program (a federal program for low income communities) tend to display higher efficiency scores than those they would have achieved had they not been included in the project. Additionally, the results show that the more power awarded to municipal councils, the better the effectiveness in resource utilization as measured by efficiency indices. This is because such councils tend to increase the transparency of the budgeting process, hence reducing corruption and the misuse of local funds.
 
Publisher Sociedade Brasileira de Econometria
 
Date 2005-11-01
 
Type info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion


 
Format application/pdf
 
Identifier http://bibliotecadigital.fgv.br/ojs/index.php/bre/article/view/2507
10.12660/bre.v25n22005.2507
 
Source Brazilian Review of Econometrics; Vol 25, No 2 (2005); 287-313
Brazilian Review of Econometrics; Vol 25, No 2 (2005); 287-313
1980-2447
 
Language eng
 
Relation http://bibliotecadigital.fgv.br/ojs/index.php/bre/article/view/2507/1490