Targeting versus Categorical Approaches in Social Safety Nets: The Case of Saudi Arabia's Regular Assistance Program Reform

This paper examines the reform of Saudi Arabia's flagship social safety net: the Regular Assistance (RA) Program. The program transitioned from a categorical-based design to a means-based design. The paper first identifies the pre-reform issues associated with the program using survey data from 2017 and 2019, including weak coverage, high exclusion and inclusion errors, low activation rates, and lack of eligibility verification mechanisms. The program's design caused regressive outcomes, with leakage of benefits to higher-income quintiles, which reduced the effectiveness and efficiency of the program. The reform, which was implemented in 2020, aimed to address these challenges and improve the program's effectiveness and efficiency. The reform introduced household means testing, a guaranteed minimum income, mark-ups for dependents and special needs categories, earnings disregards, and links to labor market programs. The paper analyzes the post-reform RA program using administrative data collected in August 2023, focusing on changes in beneficiary characteristics and targeting accuracy. The data suggests that the shift to a means-based approach may have reduced exclusion errors and improved the program's targeting accuracy. The benefits were better directed towards the intended vulnerable groups, leading to a reduction in leakage and an increase in adequacy for those most in need. The paper situates findings within the broader theoretical discourse concerning targeting mechanisms in social safety nets. Understanding the nuances of targeting methods is crucial for designing efficient and inclusive social transfer programs to alleviate poverty and promote socio-economic development.