En prensa
Capítulo «Proactive algorithmic transparency in government: The case of the Colombian repositories of public algorithms», con Sarah Muñoz-Cadena
Abstract: The adoption of artificial intelligence (AI) and other algorithmic technologies in the public sector is expanding around the globe, and transparency is one of the most common principles in AI guidelines. However, only a few national and sub-national governments proactively inform citizens about using automated decision-making (ADM) systems. This chapter aims to analyze how repositories of public algorithms can contribute to proactive algorithmic transparency. For that purpose, we study the case of the Colombian national government’s repositories, a unique case in Latin America, regarding the levels of transparency regarding the acquisition, development, piloting, adoption, and use of AI systems and robotic process automation (RPA). The chapter also presents the results of a novel database of 113 ADM systems in the Colombian public sector. We conclude that while the government publishes significant amounts of data about its ADM systems, the publicly available information is insufficient to make the algorithmic processes truly accountable in Colombia. This research follows a case study approach in which quantitative and qualitative data about algorithmic transparency in Colombia were collected and processed. The primary data was drawn from pre-existing public databases and repositories and primary information published by public entities in annual management reports, institutional web pages, social networks, and press reports. Secondary sources included information published by tech companies, news reports, grey literature –published by multilateral entities and civil society organizations– and academic literature.
Próximamente: Gutiérrez, J. D. y Muñoz-Cadena, S. (forthcoming), «Proactive algorithmic transparency in government: The case of the Colombian repositories of public algorithms». In Giest, S., Klievink, B., Ingrams, B., and Young, M. (eds.), «Handbook of Governance and Data Science», Edward Elgar.
Capítulo: “Instituciones para la competencia. Superintendencias y Comisiones de Regulación”, con María Carolina Corcione.
Resumen: El principal objetivo de este capítulo es describir el funcionamiento de las instancias, espacios e instrumentos del sistema de competencia que afectan la formulación, implementación y evaluación de las políticas públicas en Colombia. Actualmente, el sistema de competencia está compuesto por organizaciones estatales de las tres ramas del poder público y por actores no estatales. Todos los integrantes del sistema participan —en diferentes grados y momentos— en los procesos de formulación, implementación y evaluación de políticas públicas. Este capítulo expone y analiza cómo los diversos actores que integran el sistema de competencia contribuyen en dichos procesos de políticas públicas.
El texto profundiza en dos tipos de organizaciones que hacen parte del sistema de competencia y que están encargadas de promover y proteger la competencia en los mercados: las agencias de competencia y los reguladores económicos. Por tanto, el capítulo se enfoca en estudiar el caso de la Superintendencia de Industria y Comercio (SIC), que es la “Autoridad Nacional de Protección de la Competencia” y las tres principales comisiones de regulación: Comisión de Regulación de Comunicaciones (CRC), Comisión de Regulación de Agua Potable y Saneamiento Básico (CRA) y la Comisión de Regulación de Energía y Gas (CREG).
Próximamente: Gutiérrez, J. D. y M C. Corcione (próximamente). “Instituciones para la competencia. Superintendencias y Comisiones de Regulación”. En Cuervo y Sanabria (eds.) “Instituciones y Organizaciones para las Políticas Públicas”, U. Externado de Colombia y U. de los Andes.
En pre-print
Artículo: «The Gradual Impact of Sanctioning Cartels on Market Competition: Evidence from the Colombian Manufacturing Sector» con Sergio Pelaez
Abstract: This paper examines the market effects of fines imposed on hard-core cartels by the Colombian competition agency. We find that the fines had a positive and gradual impact on market competition in the manufacturing sector from 2012 to 2020. This suggests that the fines deterred cartel behavior but also that some implicit agreements take time to fade away. To measure market competition, we use the complement of a Lerner index, while an indicator variable is employed to capture the differential timing and sectors of fines imposed by the agency across various industries and years. Our data sources are a manufacturing census from the national department of statistics and public records of cartel cases from the agency. Our panel data consists of 67,671 observations for 10,316 firms over nine years. We employ a difference-in-differences (DID) method with multiple periods following Callaway and Sant’Anna (2021) and conduct robustness checks using Gardner (2021) and allowing for anticipation effects.
JEL codes: K21 – Antitrust Law, L41 – Horizontal Anticompetitive Practices, L60 – Industry Studies: Manufacturing, C23 – Panel Data Models; Spatio-temporal Models, D22 – Firm Behavior: Empirical Analysis
Citación sugerida: Sergio Pelaez, Juan David Gutiérrez. The Gradual Impact of Sanctioning Cartels on Market Competition: Evidence from the Colombian Manufacturing Sector, 16 May 2023, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-2928151/v1]
En progreso
Artículo: «Metering Antitrust Authorities: Performance Measurement in the Americas«
Abstract: This paper discusses how performance measurement of competition authorities is implemented and whether these assessments influence enforcement, advocacy and managerial decisions. The paper studies 36 competition authorities from 25 national jurisdictions and two supranational jurisdictions in the Americas. The data was collected through questionnaires answered by officials from competition authorities, interviews to directors, advisors or justices of the agencies or specialised tribunals, and from reports and statistics published in the organisations’ websites. The paper reports that the majority of the competition authorities are subject to some form of external evaluation and most of them also carryout internal performance measurements. However, over 22% of the competition authorities do not disclose online data of their self-assessments. Activity-based indicators are used more frequently than other types of indicators, such as managerial or impact-based indicators. Only four competition authorities regularly calculate the overall consumer benefits generated by their interventions and few authorities measure the rate of judicial reversal of their decisions. The majority of the surveyed officials answered that performance measurements influenced the strategic decision-making of the authority, but there were officials from 20% of these organisations who acknowledged that the criteria used for self-assessment were unclear. External assessments appear to influence the competition authorities’ agendas and incentive legal/institutional reforms.
Keywords: antitrust, enforcement, competition authorities, performance measurement, public management
Juan David Gutiérrez 