Evaluation of the effects of the active labour measures on reducing unemployment in Romania

Transcription

1 National Scientific Research Institute for Labor and Social Protection Evaluation of the effects of the active labour measures on reducing unemployment in Romania Speranta PIRCIOG, PhD Senior Researcher 1 st degree, Scientific Director, NSRILSP Madalina Ecaterina POPESCU, PhD Senior Researcher 3 rd degree, NSRILSP EC-RWC 2015 Eastern-Central Regional Workshop on Counterfactual Impact Evaluation Methods 2015 Budapest, September 2015

2 Project: ASSESSMENT OF THE ACTIVE MEASURES EFFECTS ON REDUCING UNEMPLOYMENT AND INCREASING EMPLOYMENT Stage 1: Stage 2: Stage 3: Stage 4: Elaborating the assessment methodology of the active measures impact on reducing unemployment and increasing employment Ex-post impact assessment of active measures on employment in Romania Factor analysis on sustainable employment. Recommendations for increasing the efficiency and effectiveness of active employment measures. Elaborating an ex-ante methodology to assess the active measures impact on reducing unemployment and increasing employment

3 Main results Methodology for ex-post impact assessment of active measures upon employment Ex-ante methodology to assess the active measures impact upon employment The ex-post impact assessment report of active measures upon employment in Romania Recommendations for increasing the efficiency and effectiveness of active employment measures 4 3 Factor analysis on the active successful measures.

4 The approaches

5 Data collection According to the beneficiary s request, different survey subsamples were required, in order to be able to form distinct groups for both the participants and the non-participants of each of the 7 labour measures considered in the analysis. Based on the particularities of each measure, 7 different questionnaires were elaborated addressing the participants of each measure, while other 3 questionnaires were required to address the non-participants of the measures at different moments in time: in 2010, 2011 or The respondents were randomly selected from the database of the National Romanian Agency for Employment. Main problems occurring: some cases of incorrect or inexistent addresses in the database some cases of people unwilling to participate the survey some cases of unreachable persons

6 The micro model The propensity score matching method (PSM) was applied distinctly for each labour measure using STATA software, by: first estimating Probit models and generating the propensity coefficients checking the balancing property of the models applying a non-parametric method for matching the treated with the untreated units based on the propensity scores and determining the net impact of the intervention. In order to select the control and the treatment group using PSM, the following variables were used: the output variables: employment (in 2013 and at the time of the survey); duration of unemployment and net wage the covariates grouped into: socio-demographic variables (sex, age, marital status, education, residence and region); economic variables (sector of activity, years of work experience);

7 The micro model The general form of the Probit model was: Treatment = f (AGE, GENDER, MARITAL STATUS, REGION, EDUCATION LEVEL, RESIDENCE, SECTOR OF ACTIVITY, WORK EXPERIENCE) Once the dichotomous model was estimated and the propensity scores were determined, 4 matching algorithms were applied: The Nearest Neighbour Matching Method The Radius Matching Method The Kernel Matching Method The Stratification Method The model allowed us to assess the ex-post impact of the active measures upon employment in Romania, by quantifying the odds of a participant to the measure to find a job as compared to a non-participant.

8 Labour active measures Professional training measure Labour mobility stimulation measure "Salary income supplementation" active measure Structure of the sample: 295 treated and 346 non-participants Beneficiary s profile: young urban male, with medium level of education Structure of the sample: 155 treated and 346 non-participants Beneficiary s profile: urban male aging 25-45, with high education level Structure of the sample: 165 treated and 317 non-participants Beneficiary s profile: urban male over 45, with medium education level Measure Subsidies for employers who employ individuals aged over 45 Structure of the sample: 154 treated and 251 non-participants Beneficiary s profile: urban male over 45, with medium education level Measure Subsidies for the employers who employ individuals who have 3 years more to meet the requirements for early retirement or for age limit retirement, unless they meet the conditions to request partial early retirement Structure of the sample: 125 treated and 251 non-participants Beneficiary s profile: urban male over 45, with low education level Measure Subsidies at employers who employ graduates of education institutions Measure Employment premiums for graduates Structure of the sample: 183 treated and 247 non-participants Beneficiary s profile: young urban female, with high education level Structure of the sample: 84 treated and 81 non-participants Beneficiary s profile: young urban female, with medium education level

9 Net impact Subjective impact Net impact at the end of the application Net impact at the time of the survey Professional training measure 55.5% -7% -8% Labour mobility stimulation measure 22.6% 35% 28% "Salary income supplementation" active measure 27.5% 36.6% 27.1% Measure Subsidies for employers who employ individuals aged over % 44.7% 11.6% Measure Subsidies for the employers who employ individuals who have 3 years more to meet the requirements for early retirement or for age limit retirement, unless they meet the conditions to request partial early retirement 65.6% 56.2% -18.1% Measure Subsidies at employers who employ graduates of some education institutions 24.6% 33.5% 15.7% Measure Employment premiums for graduates 26.2% 35.4% -24.5%

10 Factor analysis Active policies are only some mechanisms among many others (unemployment insurance policies (passive policies), wage policies, policies concerning the safety of employment, policies on wage-setting mechanisms, policies concerning social dialogue, etc). The lack of a correlation of all the objectives of these policies may cancel or mitigate the partial positive effects obtained. Given the multitude of possible factors that may influence the results of implementation of some active measures, it is absolutely mandatory that any impact assessment try to contextualizeze the best possible results. Therefore we decided to customize the impact analysis at the level of sociodemographic characteristics (gender, residence area, level of education and age groups) in order to emphasize the effectiveness of each measure at the level of some specific subgroups.

11 Factor analysis Several Logit models were then estimated on each subgroup of individuals: Male versus female Urban versus rural Low, medium and high level of education Age groups: under 25; 25-45, over 45 years old The output variable was the occupational status of the respondents, while the explanatory variables were the covariates: sex, age groups, marital status, levels of education, residence area and the "treatment" binary variable. Valid econometric models were estimated in relation to LR test and pseudo R². and also with statistically significant coefficients. A particular attention was paid to the significance of the "treatment" variable, which indicates the chances of a beneficiary of a certain measure to find a remunerated job at the time of the survey, compared to those of a nonbeneficiary of the measure, with the same characteristics.

12 Factor analysis Professional training measure Has stronger effect on: young people under 25 years Labour mobility stimulation measure Has stronger effect on: young rural male with low education level "Salary income supplementation" active measure Has stronger effect on: young rural female with medium education level Measure Subsidies for employers who employ individuals aged over 45 Has stronger effect on: rural female over 45 with medium education level Measure Subsidies for the employers who employ individuals who have 3 years more to meet the requirements for early retirement or for age limit retirement, unless they meet the conditions to request partial early retirement Has stronger effect on male over 45, in general. Measure Subsidies at employers who employ graduates of education institutions Has stronger effect on: rural male with high education level Measure Employment premiums for graduates Has stronger effect on : young rural male

13 Conclusions Young people under 25 years of age proved to be more receptive to the active employment measures regarding Labour mobility stimulation, Wage income supplementation and Employment premium granted to graduates of educational institutions, showing a higher level of availability for changing jobs and home address for an additional financial gain. Young people aged up to 24 years were the only category that recorded a positive impact of the Training courses measure, showing the occupational inadequacy of them related to labour market requirements, which result in reporting the need for better information on the dynamics of the labour market. In terms of gender differences, we notice that the active employment measures relating to Labour mobility stimulation, Subsidizing of jobs for graduates, Employment premium granted to graduates of educational institutions, as well as Subsidizing of jobs for people who still have 3/5 years up to retirement proved to be more effective among men, while the effects of measures regarding the Wage income supplementation and Subsidizing of jobs for people over 45 years were felt more strongly among women.

14 Conclusions Effects with higher intensity amoung people from the urban area were recorded only for the measure regarding the Employment premium granted to graduates of educational institutions, while the remaining measures exercised a stronger impact upon individuals from the rural area. Considering the level of education as an influence factor of the impact of active employment measures, we emphasize the following: graduates with general education resonate more powerfully to the implementation of measures relating to Labour mobility stimulation, Wage income supplementation and Subsidizing of jobs for persons over 45 years of age, while the most pronounced effects of the measure on the Employment premium granted to graduates of educational institutions are felt especially among people with higher education.

15 Conclusions Policy recommendations aimed at increasing the efficiency and effectiveness of active employment measures are considering three levels: 1. At a conceptual, strategic level: Periodic assessment of the impact of active measures Designing a new and larger database of unemployed persons looking for a job Reinforcing the role of activation of labour market policies Periodic studies and forecasts on short and medium term of the labour demand Strengthening the principle for equal opportunities and access to labour market; Stimulating labour demand through balancing the ratio between the level of unemployment allowance and the minimum wage Reinforcing the role of informing for the diminishing of activities in the informal economy.

16 Conclusions Policy recommendations aimed at increasing the efficiency and effectiveness of active employment measures are considering three levels: 2. At measure level: A better profiling of the persons looking for a job before being recommended for a certain measure; Offering training vouchers within vocational training active measure Offering vouchers for apprenticeship to support employment in jobs that require long-term specialization (e.g. industry); Supporting the development of start-ups Development of customized mediation or based on a standardized mechanism of profiling the unemployed and persons looking for a job 3. At the level of packages of measures: Setting up packages of measures for groups with difficulties in accessing jobs (e.g. youth) and finding a better way to define the education graduate;

17 Questions?