Download to read offline
Uploaded byVerónica Escudero
What Works: Active Labour Market Policies in Latin America and the Caribbean
The document discusses the labor market performance and social achievements in Latin America, highlighting improvements since 2000, including reduced unemployment and working poverty. It emphasizes the role of active labor market policies (ALMPs) and conditional cash transfer programs in improving employment outcomes, while noting challenges in job quality and high levels of informality. The report concludes by advocating for a comprehensive, evidence-based approach to leverage ALMPs effectively and enhance job matching and quality.
What Works: Active Labour Market Policies in Latin America and the Caribbean
- 1. Verónica Escudero ILO ResearchDepartment December 15, 2016
- 2. MOTIVATION Latin America: findingthe path to sustainable progress Weak labour market performance in the 1990s • In this decade unemployment increased from 8% to 11% Labour market and social achievements since 2000: • Drop of unemployment to 6.5% • Reduction in working poverty (from 18 to 8%) • Decrease in income inequality It means employment growth exceeded growth of working age population and social conditions improved considerably 2
- 3. MOTIVATION Figure: Share ofthe population covered by a CCT (%) Innovative Social Policies played a crucial role: Conditional cash transfer (CCT) programmes are comprehensive interventions aimed at raising social security coverage More than 20% of the population is covered by a CCT in the region This corresponds to an expenditure of 0.4% of GPD 3
- 4. MOTIVATION Figure: Average productivitygrowth rate (2000–2015) and productivity levels (2000) by region However, there was a limited impact on quality job creation: The region had low levels of productivity growth Moderate increase in employment in the service sector ( from 60.8 to 63.3) Only slight increase in employment in high skilled occupations (from 17.6 to 19.9) 4
- 5. MOTIVATION Figure: Informal employmentand GNI per capita, latest available year And high levels of informality: Represents almost half of total non- agricultural employment Heterogeneity across countries More common among women and youth and low educated individuals One reason for the prevalence of vulnerable employment 5
- 6. The potential roleof ALMPs: ALMPs could prevent the current economic slowdown from translating into a structural deceleration • Improve matching between jobseekers and employers • Keep individuals attached to the labour market Improve employability and support ongoing economic transformations • Guarantee continuous skills’ upgrade • Improve employment quality Even during crises when effects are weaker, ALMPs can be an efficient option • Unemployment, long-term inactivity and CCTs are expensive options • ALMPs can be self-financing in the medium-term MOTIVATION 6
- 7. • Interventions thathelp individuals finding more sustainable jobs WHAT ARE ALMPS? 7
- 8. Leveraging the potentialof ALMPs: Public expenditure in ALMPs has increased in virtually all countries in Latin America since 2000 Often higher than expenditure in passive policies However, in most countries expenditure in ALMPs is far from the levels in advanced countries Moreover, not enough is known with respect to the impact of ALMPs as well as the implementation and design characteristics needed to maximise their effectiveness MOTIVATION 8
- 9. MOTIVATION Figure: Public expendituresin active and passive labour market policies as a share of GDP 9
- 10. MOTIVATION Figure: Share ofexpenditure in ALMPs by type of programme 10
- 11. OBJECTIVES AND WAYFORWARD Address the knowledge gap: What works in terms of ALMPs and how can they best be leveraged? Fulfil the ILO mandate: Provide evidence based policy advice that is relevant and applicable Work rigorously: Involve a Council of experts to make sure the analysis is at the forefront of research in policy evaluation Involve the constituents: Ensure that the research accurately reflects the policy preoccupations of the regions and its constituent countries 11
- 12. PRODUCTS Compendium of labourmarket policies First Meta-analysis of ALMPs in the region Three impact evaluations: Argentina, Colombia and Peru Synthesis report: in English and Spanish Website: www.ilo.org/almp-americas 12
- 13. ILO COMPENDIUM oflabour market policies Information about all ALMPs and some PLMPs in: Argentina, Brazil, Colombia, Ecuador, Peru, Uruguay, and Mexico. 218 policies with detailed information: • Type of policies and components • Year of implementation • Policy characteristics and target group • Sources of information and impact evaluations Available in the project’s webpage 13
- 14. META-ANALYSIS Goal: Assess whathas been done, what are the results and what still needs to be researched Based upon 44 studies, 52 programmes and 152 impact estimates 18 studies added from Card et al (2015) and 26 using the ILO network Keep in mind: This approach relies heavily on the quality of studies included, the selection process is crucial Method: Linear probability model with different specifications 14
- 15. META-ANALYSIS There isa peak of studies between 2005 and 2009 75% of the studies are of quasi-experimental nature Patterns in the evaluation of ALMPs by type of intervention and country • 18 out of 52 programmes: Implemented in Argentina and Peru • 67% are training programmes • 70% of the evaluations are on youth These patterns are not strictly representative of ALMPs in the region (e.g. 44% are training programmes, 24% are for youth) Various findings in terms of effectiveness by type of policy 15
- 16. Share of ALMPsby type of intervention in the meta-analysis and the compendium Meta-Analysis Compendium 67 13 8 8 4 44 28 11 5 12 0 10 20 30 40 50 60 70 80 Training Self-employment Public works programmes Employment subsidy Labour market services 16
- 17. IMPACT EVALUATIONS: PERÚ- Background Motivation: Informal employment is high in Peru (64% in 2013) and affects mostly the poorest segments of the population Public works programmes are increasingly used in the region Often in form of workfare programmes (short-term jobs and training) Construyendo Perú (2007-11): Provided temporary employment to the unemployed in situations of poverty and extreme poverty; and also training in order to enhance their employability Implemented through the financing of public investment projects (max. 4 months) Targeting was an important component in the planning. It involved 3 stages: • Geographical (by district using an index – FAD) • Self-targeting • Individual Targeting (based on socioeconomic profiling) 17
- 18. IMPACT EVALUATIONS: PERÚ- Evaluation Data and methodology: Encuesta Nacional de Hogares and a special survey on programme participants (2007-10) Regression discontinuity design (RDD) based on the geographical targeting mechanism Public Works programmes: effective in the medium- to long-term? Positive effect on employability for women and low-skilled individuals However, higher probability of being informal or working poor Source: ILO based on Escudero (2016) All Women Lower educate d Higher educated Employment status Employed 2SLS 2.1 2.3* 4.7* 2.0 LLR 5.1 4.5* 39 4.3 Inactive 2SLS -2.3 -2.5* -4.7* -2.4 LLR -4.2 -4.9* -52 -2.9 Employed informally 2SLS 5.5** 3.9** 3.3 6.6** LLR 16*** 7.5** 26 15*** Own-account worker 2SLS 3.6** 2.8** 1.9 3.5** LLR 5.5** 4.5** 20.7 7.1** Waged worker 2SLS -0.03 0.12 0.7 -0.1 LLR 3.6** 2.1* 3.7 4.3** Waged employee 2SLS -2.8** -1.6** 0.2 -2.8* LLR -7.5*** -3.2* 0.6 -8.9** Job quality Working poor 2SLS 7.6*** 5.6*** 0.8 8.9*** LLR 14* 10* -21 7.1** Excessive working time 2SLS 1.7 1.1 1.3 2.6 LLR 14* 11 13 16* Observations Maximum 24 427 12 374 34 257 18
- 19. IMPACT EVALUATIONS: PERÚ- Conclusions In more detail: Consider heterogeneity in effects: Higher participation of and clearer effects on women Desired effects are not always fully achieved: deadweight loss possible , effects can fade over time Possible unintended results: detrimental effects on job quality (informality, working poverty, etc.) Policy Considerations: Nature of projects is key: • Training, duration, resources, needs of the local labour market Ensure that resources are allocated efficiently • In Peru financing did not go to the most efficient project Ensure that programmes are well resourced: • In Peru, spending was considerably lower than in other countries in the region 19
- 20. IMPACT EVALUATIONS: ARGENTINA- Background Motivation: Crisis in 2001-02: unemployment rate of 21.5% and more than 50% of the population under the poverty line In this situation CCTs (Plan Jefes) were a key tool for the improvement of social indicators Yet, CCTs by themselves are insufficient to reduce poverty in a sustainable manner From the Plan Jefes to the Seguro de Capacitación y Empleo (SCE): When the economic situation improved and the fiscal space was greater, the policy approach was reoriented: • Plan Familias was targeted • SCE and plan familias were targeted towards participants of Plan Jefes with high probability of finding a job Data and methodology: Encuesta Permanente de Hogares (2006-10) Need for comparison group: Transfer to new policy was gradual, hence in reference period there were eligible participants of Plan Jefes that had not yet been transferred to the new programmes Two sets of estimates: DID and DID combined with PSM 20
- 21. IMPACT EVALUATIONS: ARGENTINA- Evaluation Results: Are active interventions effective in reducing dependency from CCTs? Positive effect on earnings and employment quality (e.g. formal employment, working hours) Participants are also more likely to move to inactivity Souce: ILO based on López Mourelo and Escudero (2016). Full sample Matched sample Employment status Employed informally -0.02*** -0.06** (0.003) (0.013) Unemployed ns ns Inactive 0.03* 0.08** (0.009) (0.019) Job quality Real hourly wages 0.04*** 0.03** (0.005) (0.009) Low-paid job ns ns Number of hours worked 0.64* -1.55*** (0.226) (0.079) Excessive working time -0.02** -0.05** (0.004) (0.009) Underemployment -0.03*** -0.02*** (0.001) (0.001) 21
- 22. IMPACT EVALUATIONS: ARGENTINA- Conclusions Policy Considerations: Ensure equality of access: • High cost for potential beneficiaries in remote areas to reach the cities where services were offered • Design resulted in the discrimination of women Reinforcing incentives: • The intervention had a shorter duration (2 years) than Plan Jefes (indefinite), potentially creating a disincentive for participants to switch Strengthen the institutional capacity: • Years after the implementation, many Plan Jefes participants had not yet been given the option to join 22
- 23. IMPACT EVALUATIONS: COLOMBIA- Background Motivation: The share of informal employment (54.4% in 2013) is high, even for the region Public employment service has the potential to increase formal employment and improve labour market outcomes of jobseekers They represent 10,8% of spending in ALMPs in Colombia (second item in terms of public spending in ALMPs) The Agencia Pública de Empleo (APE): Constituted in 2013 as the public employment agency in charge of providing labour market services, as part of SENA Part of the SPE (a network of public and private providers of employment services) APE provides services to job seekers and to employers Data and methodology: Gran Encuesta Integrada de Hogares (2008-14) Propensity score matching, using information on the method of search 23
- 24. IMPACT EVALUATIONS: COLOMBIA- Evaluation All Women Men Low skilled High skilled Control: classified advertisements Employed formally 0.09*** 0.13*** 0.04*** 0.24*** 0.05*** Hourly wage -0.04* -0.06** -0.07** 0.13*** -0.10*** Control: private employment agencies Employed formally -0.01* -0.01 -0.03*** -0.03** -0.01** Hourly wage -0.02* -0.09*** 0.03 0.05*** -0.07*** Control: contact with employer Employed formally 0.05*** 0.07*** 0.04*** 0.1*** 0.03*** Hourly wage -0.06*** -0.11*** 0.0 0.06*** -0.12*** Control: family and friends Employed formally 0.31*** 0.33*** 0.28*** 0.46*** 0.25*** Hourly wage 0.06*** 0.01 0.07*** 0.22*** -0.04*** Results: Are APE services more effective than other job-search channels? Positive effects on formality, but mixed effects on wages More effective for low-skilled jobseekers and women, and when services are provided face-to-face Source: ILO based on Pignatti (2016). 24
- 25. IMPACT EVALUATIONS: COLOMBIA- Conclusions Policy Considerations: Institutional framework: • Limited scope: Only 1% of population reached • Multiple labour market services platforms are inefficient • Too onerous requirements on job seekers and employers can create disincentives Reaching out to jobseekers and employers: • Pool of jobseekers concentrates around low-skilled individuals • Registered jobseekers need to be encourgaed to continue with their job search • The reputation of the SENA should be leveraged for its provision of labour market services 25
- 26. THE REPORT INSHORT 1. Innovations from the past have been successful in raising living standards but have not necessarily improved employment outcomes 2. ALMPs are increasingly used in the region, but there is considerable scope to leverage them further 3. ALMPs need to be implemented as a policy package rather than isolated programmes to take advantage of their complementarities 4. Effectiveness hinges on particular design and implementation characteristics 5. Some of these characteristics are particular to each type of policy 6. There is the need to guarantee sufficient institutional capacity 7. Data collection and robust impact evaluations of ALMPs must be part of the comprehensive strategy 26
- 27.
- 28. Labour market outcomesimprove when ALMPs are combined with existing measures: • Income support and job-search assistance • Employment subsidies as a complement to CCTs ALMPs have to constitute a policy package: • Public works programmes with a training component • Assistance for funding with technical support for the self-employed WHAT WORKS: The importance of the design and implementation Leveraging complementarities: 27
- 29. WHAT WORKS: Theimportance of the design and implementation Characteristics: Duration is the key driver of programme effectiveness Particular measures are needed to provide incentives for the use of ALMPs Target groups: Women are more likely to succeed in the labour market thanks to ALMPs Youth and low-income people do not show better results Adapt the characteristics to increase effectiveness: 28
- 30. TRAINING: • Effectiveness doesnot depend on the number of components • On-the-job training components are the most effective (particularly when they deal with barriers to individual employability) • Need to take into account the local firms’ needs • Effectiveness increases when training providers are selected through a bidding process PUBLIC WORKS: • Ensure targeting is applied and enforced (e.g. avoid multiple participation) • Designed to fully benefit those targeted (e.g. projects tailored to the group) • More successful when they include employability-enhancing components (particularly when relevant for the local labour market) WHAT WORKS: Drivers of effectiveness by type of policy 29
- 31. WHAT WORKS: Driversof effectiveness by type of policy EMPLOYMENT SUBSIDIES: • Effective mainly for youth or when targeted specifically to vulnerable groups • Acts mainly through substitution and deadweight effects MICROENTERPRISE CREATION POLICIES: • Effectiveness hinges on the educational level of participants • And on the provision of technical assistance in addition to the financial support • Sustainability depends on the development of the local market LABOUR MARKET SERVICES: • More effective when provided face-to-face • The challenge is to reach a higher number of beneficiaries • PES can have an important role during periods of weak demand (e.g. to deploy other ALMPs) 30
- 32. WHAT WORKS: InstitutionalCapacity Provide sufficient financial means • Make sure that the programme reaches all individuals in need who comply with the eligibility criteria • Guarantee the stability of financing (also during periods of crisis) • Supervision of activities and tracking of the pre-established objectives Strengthen the link between policies • PES offices present in the whole country • Reach remote areas Systematic gathering of data and impact evaluation of policy efforts 31