Documentos de Trabajo sobre Economía Regional y Urbana - Municipal categories in Colombia: Moving towards an asymmetric decentralization

Documentos de Trabajo sobre Economía Regional y Urbana
Number: 
321
Published: 
Authors:
Yuri Carolina Reina-Aranza
Classification JEL: 
C45, R11, E62
Keywords: 
Decentralization, capabilities, results, Unsupervised learning, ICOT, Colombia.

The most recent

Julián Alonso Cárdenas-Cárdenas, Deicy Johana Cristiano-Botia, Eliana Rocío González-Molano, Carlos Alfonso Huertas-Campos
Luis E. Arango, Juan José Ospina-Tejeiro, Fernando Arias-Rodríguez, Oscar Iván Ávila-Montealegre, Jaime Andrés Collazos-Rodríguez, Diana M. Cortázar Gómez, Juan Pablo Cote-Barón, Julio Escobar-Potes, Aarón Levi Garavito-Acosta, Franky Juliano Galeano-Ramírez, Eliana Rocío González-Molano, Maria Camila Gomez Cardona, Anderson Grajales, David Camilo López-Valenzuela, Wilmer Martinez-Rivera, Nicolás Martínez-Cortés, Rocío Clara Alexandra Mora-Quiñones, Sara Naranjo-Saldarriaga, Antonio Orozco, Daniel Parra-Amado, Julián Pérez-Amaya, José Pulido, Karen L. Pulido-Mahecha, Carolina Ramírez-Rodríguez, Sergio Restrepo Ángel, José Vicente Romero-Chamorro, Nicol Valeria Rodríguez-Rodríguez, Norberto Rodríguez-Niño, Diego Hernán Rodríguez-Hernández, Carlos D. Rojas-Martínez, Johana Andrea Sanabria-Domínguez, Diego Vásquez-Escobar
Luis Armando Galvis-Aponte, Adriana Isabel Ortega-Arrieta, Adriana Marcela Rivera-Zárate

Abstract

Asymmetric decentralization has become relevant in developing countries like Colombia, where capabilities between subnational governments differ dramatically. Although these discussions have been present since the 1960s, the success of decentralization is still a relevant discussion in these contexts, where the classification of the territory is fundamental. This document discusses the possible obstacles of the territorial classes currently used in Colombia. Likewise, it proposes and discusses the usefulness of using recent algorithms from the unsupervised machine learning literature to classify subnational territories. Specifically, this document implements Clustering via Optimal Trees (ICOT), an algorithm that allows us to classify the territories and identify the rules underneath the defined classes. This study also proposes the existence of different territorial typologies according to their uses.