Symbolic Institutional Traps: Language Regimes, Legal Legacy, and Organizational Constraint in Postcolonial Economies

This study draws upon multiple original and secondary datasets to empirically test the relationship between language systems, institutional performance, and economic integration. The analysis is grounded in four core data sources that enable comparative institutional assessment across both national and subnational contexts.

Dataset 1: Doing Business North America (DBNA)This dataset utilizes the 2022 Doing Business North America (DBNA) data [DBNA-2022-All-Data], compiled by the Center for the Study of Economic Liberty at Arizona State University. The DBNA dataset offers standardized, city-level institutional metrics across 83 U.S. cities, including Puerto Rico. It evaluates six key dimensions: Starting a Business, Employing Workers, Getting Electricity, Land and Space Use, Paying Taxes, and Resolving Insolvency. Each dimension is further disaggregated into sub-indicators for detailed institutional analysis. This study uses DBNA data to investigate how language systems and legal origin shape institutional performance, using San Juan as a focal case of institutional divergence within a nominally common-law jurisdiction. All variables and derived metrics are made publicly available for transparency and reproducibility, with full documentation provided in the appendix.

Dataset 2: EF English Proficiency Index (EF EPI)English language proficiency data were obtained from the EF English Proficiency Index (EF EPI), published by EF Education First. The EF EPI ranks countries and territories based on results from the EF Standard English Test (EF SET), a standardized online assessment. Countries are grouped into five proficiency bands: Very High (600+), High (550–599), Moderate (500–549), Low (450–499), and Very Low (<450). While subject to self-selection bias, EF EPI remains one of the most widely used proxies for national English proficiency in global research. In this study, EF EPI scores are employed as indicators of language regime characteristics hypothesized to influence institutional performance, human capital development, and integration into global economic systems.

Dataset 3: U.S. Census Bureau – ACS Language Data (Table B16004 & B16001)Spanish language use and English proficiency data were obtained from the U.S. Census Bureau’s American Community Survey (ACS), specifically Table B16004 (Language Spoken at Home and English-Speaking Ability) and Table B16001 (Language Spoken at Home by Ability to Speak English for the Population 5 Years and Over). The primary variable is the proportion of individuals who speak Spanish at home, disaggregated by English proficiency levels. Puerto Rico is included via the ACS Puerto Rico Community Survey. These data serve as proxies for regional linguistic environments and are incorporated into our econometric models to examine whether English proficiency—interpreted as a commodified economic skill—affects business participation and inequality.

Dataset 4: World Bank Ease of Doing Business (EODB) 2020This study includes data from the World Bank’s 2020 Ease of Doing Business (EODB) rankings, which assess regulatory efficiency across more than 100 countries. Key indicators include Starting a Business, Getting Electricity, Registering Property, and Enforcing Contracts. Although the EODB project was discontinued in 2021 due to data integrity concerns, its inclusion remains critical, as it is the only global dataset that includes Puerto Rico, enabling consistent cross-national comparisons. To mitigate known limitations, EODB data are complemented by DBNA metrics and the World Bank’s Worldwide Governance Indicators (WGI), providing triangulation and robustness to the empirical design.

All datasets have been cleaned and standardized, with derived variables constructed to support cross-sectional econometric analysis. Full codebooks, documentation, and supplementary files are included in the accompanying appendix to facilitate replication and secondary research.