2017年-世界发展银行全球_Can_Business_Input_Improve_the_Effectiveness_of_Worker_Training____Evidence_from_Brazils_Pronatec-MDIC_63页_1mb

报告摘要

Can Business Input Improve the Effectiveness of Worker Training? Evidence from Brazil's Pronatec-MDIC

Core Content

This paper evaluates the effectiveness of a demand-driven vocational training program in Brazil, Pronatec-MDIC, which incorporates direct input from businesses in determining the location, scale, and skill content of training courses. The study compares the outcomes of this program to those of a traditional, non-demand-driven program (Pronatec), both administered under the same institutional and logistical context.

Main Findings

Employment Effect: Trainees who completed the Pronatec-MDIC program had an 8.6% increase in employment in the year following course completion compared to those who did not receive a course offer.
Source of Impact: The employment gains primarily came from previously unemployed trainees who found jobs at non-requesting firms, suggesting that the program's demand-driven nature helped align training with broader labor market needs.
Skill Demand Alignment: Courses in occupations with higher growth had larger employment effects, indicating that business input captured meaningful information about future skill demand.
Comparative Effectiveness: The Pronatec-MDIC program was more effective than the traditional Pronatec program, which showed negligible employment effects.
External Validity: The study provides broad external validity by evaluating a national program, not limited to youth, with a large sample size (over 300,000 trainees).

Key Information

Program Overview

Pronatec-MDIC is a demand-driven program where businesses directly request specific skills training.
The program was launched in 2011 as part of Brazil's broader Pronatec initiative, aimed at improving earnings and employability for lower-income workers.
Course requests were submitted by firms, industry associations, and worker associations, with 97% of firm requests matched to administrative employment records.

Program Characteristics

In 2014, over 16,000 courses were requested by more than 2,000 firms.
Approximately 40% of registrants completed the courses.
Course size averaged 13 seats, with 200 hours of instruction, 8 hours per week, and 5–6 months duration.
MDIC forwards course requests to the Ministry of Education, which screens and approves them based on technical criteria, budget availability, and course viability.

Data and Methodology

The study uses administrative panel data from the RAIS (Relação Anual de Informações Sociais) system, which provides monthly employment and wage data for all formal sector workers.
The empirical strategy includes a difference-in-differences IV approach to estimate the causal effect of training, using exogenous course capacity restrictions to address endogeneity.
An event study framework is also used to examine the evolution of employment effects over time relative to course start.

Subgroup Analysis

Firm-supplied trainees had the highest employment rates (75%) and higher wages (R$1,492/month) compared to unemployment beneficiaries (60%, R$1,135/month) and others (53%, R$1,129/month).
The employment effect was statistically distinguishable between the demand-driven and non-demand-driven programs, with the former showing greater effectiveness across most subgroups.

Conclusion

The study concludes that business input enhances the effectiveness of vocational training programs by aligning them more closely with actual labor market demands. The demand-driven model of Pronatec-MDIC outperformed the traditional Pronatec program, particularly in terms of employment outcomes and wage gains, and provides empirical support for the importance of incorporating private sector needs in public training programs.

Future Work

The paper suggests that further research should explore the long-term impacts of such programs.
It also highlights the need for more rigorous evaluations of demand-driven training initiatives in other countries.

References

Institutional Context: The study was conducted in partnership with IPEA and funded by the World Bank's Productivity Programmatic Approach (P152871).
Data Sources: The study uses RAIS and administrative records from MDIC and Ministry of Education.
Methodological Innovations: The use of administrative data and IV strategy allows for credible estimation of program effects, avoiding issues of sample attrition.