LAUSR

Abstract The paper provides a diagnostic of service quality in electricity distribution in Brazil. Considering indicators related to customer complaints, average time for solving problems, and consumer satisfaction, there are some signs of deterioration over time. To address the multidimensional character of service quality, various well-established multivariate statistical analysis procedures are applied. A cluster analysis suggests that firms cluster into a small number of groups that are not directly related to the geographic region or firm size. A dynamic factor analysis for each firm provides mixed evidence, but in nearly 43% of the cases, important common patterns of indicators are found within firms. A canonical correlation analysis indicates a weak association between direct service quality and overall perceived satisfaction. The nature of firm heterogeneity can potentially provide guidance on how to incorporate service quality adjustments in the price-cap rule.