The effectiveness of tax incentives in attracting investment: The case of Puerto Rico
The contribution of this dissertation is the empirical understanding of the effectiveness of Puerto Rico's investment incentive program. In 1978 the local government enacted a tax incentive law, in an effort to decentralize the location of firms. The goal is to encourage firms to locate in rural/less developed areas outside the San Juan/Metro area. The government divided the island into three industrial zones. In the high industrial zone of the San Juan area, tax exemptions are available for only 10 years, in the intermediate industrial zone for 15 years, and in the low industrial zone tax exemptions are available for 20 years.
The focus of the dissertation is to measure the impact of this program in four areas: (1) location of firms; (2) job expansion; (3) forgone revenues, and (4) a comparison of forgone revenues and job expansion benefits. Traditionally, Conditional Logit (CL) has been the methodology used for firm location analysis. However, CL confronts several limitations, and for that reason, I perform a Poisson Regression analysis. This methodology will give the same results as the CL model and, in certain cases related to location decisions, is a better approach since it handles more properly the limitations inherent in the CL methodology. Using Poisson Regression I find that firms tend to locate in a statistically significant fashion at both the intermediate and low zones. I analyze job expansion through Shift-Share (SS) analysis. One feature of SS analysis is its descriptive power when explaining the change in regional employment over time. Based on the Shift-Share analysis, I find that job expansion at both the intermediate and low zones is significantly higher than what would have occurred if these zone would have grown at the same rate of the high industrial zone. Finally, the program has a statistically significant negative impact on government revenues. In general, revenues naturally decline because firms are exempted from paying taxes through the program. This impact is greater within firms locating at both the intermediate and low zones. Nonetheless, forgone revenues are more than compensated, by salaries and wages earned in jobs created by firms.
Area planning & development;
Latin American history;
0999: Urban planning
0999: Area planning & development
0336: Latin American history