Profiling the Unemployed and Second Chance Education Schemes

Profiling the Unemployed and Second Chance Education Schemes Seamus Mc. Guinness The Economic and Social Research Institute Policy Learning Forum Cedefop 16 June 2017

Discussion Outline l Provide a description of statistical profiling and its potential use for the purposes of labour market activation. l Outline the data requirements for the development of a profiling tool and discuss its implementation in Ireland. l Outline the connection between profiling and other labour market activation measures. l Discuss the results of an evaluation of a “second chance education” activation programme.

What is Statistical Profiling? - Formal method for early identification of individuals with a high risk of long-term unemployment - We want an indication on the day an individual makes a claim for unemployment assistance how successful are they likely to be at finding a job - Profiling allows the PES to refer high risk clients to appropriate interventions in an objective way. l Advantages - More systematic, rigorous, and accurate identification of those with high risk - Allows for effective targeting of labour market activation measures and reduces dead-weight.

Data Requirements for Profiling and the Irish Approach § To develop a profiling model we need data that captures information on individuals when they make a claim and follow up (or longitudinal) information on their status 12 months after the claim was made § The Irish approach involved a specially devised questionnaire issued to all job seekers making a claim between Sept and Dec 2006 § This questionnaire collected information on a range of factors including: educational attainment, literacy/numeracy levels; labour market history, etc. Additional information was also drawn from administrative sources § We then tracked where the individuals were after 12 months, and again after 18 months.

Summary of Results from Irish Profiling Model § Factors Increasing the Likelihood of Remaining Unemployed § § Factors Increasing the Likelihood of Exiting to Employment § § Being older, having dependant children, having a history of long term unemployment, having literacy or numeracy problems, being in poor health and living outside Dublin. Being recently employed, higher levels of educational attainment, having access to transport, being willing to move for a job, being married and located in Dublin. In the Irish system the score a person receives in the model ranges from 0 to 1 and is termed a PEX score for probability of exit (to the labour market)

Exits to Employment by PEX Score 70. 0 60. 0 50. 0 40. 0 30. 0 20. 0 10. 0 Employment (%) Low PEX Medium PEX High PEX

How is Profiling Used in Practise in Ireland One-Stop Shop Integrated Reception Team Quicker Jobseeker claim processing Assessment / Decisions Team Info provided / direction given Ascertain appointment type Client provided with information pack, doc checklist, profile form, Record of Mutual Commitments, etc. Appointments set up All Claims Client selected according to profile Not “one-size” fits All Activation Case Management Activation Team Low PEX (20%) Group Engagement ii. Then every 2 months to review progress Client attends appointment Public Services Card Medium PEX (60%) Group Engagement i. Meet caseworker within 1 week of group engagement. ii. Then every 3 months to review progress Claim Decision and Profiling: PEX calculated Client signs RMC and is informed of GIS i. Meet caseworker within 1 week of group engagement. NEW High PEX (80%) Group Engagement i. Directed towards self-help tools during initial engagement session. ii. After 4 months, invited to attend caseworker to develop a personal progression plan New Claims (Profiled) JA/JB *

What are the limitations of profiling? l It is merely an allocation instrument and will not perfectly predict the outcomes of each individual. l The tool itself is designed to allow policy makers to allocate individuals to unemployment assistance programmes. However, the net benefit of the tool will be zero if the unemployment programmes that individuals are directed to are ineffective.

Evaluation of the Back to Education Allowance (BTEA) Scheme

l The BTEA is a second-chance education opportunities scheme: l l Eligibility criteria: l l l l Second-level (SLO) and Third-level (TLO) Full-time courses Education institutions under the remit of DES, while DSP administer the payment Qualifying benefit payment (e. g. , jobseeker’s, one parent family, etc. ) Duration of this payment (3 /9 months) Age (21/24) Commencing first year of a course that will lead to a QQI accreditation Received acceptance onto a qualifying course Progressing in educational qualifications Weekly payment: l Rate varies according to when course commenced and a person’s means.

Between 2007 and 2012, spending on the BTEA scheme more than trebled - from € 64. 1 m to € 199. 5 m; while the number of recipients grew from approximately 6, 000 to 25, 000 250 Millions (€) 200 150 100 50 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 BTEA Expenditure 30, 000 Number 25, 000 20, 000 15, 000 10, 000 5, 000 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 BTEA Recipients Source: Statistical Information on Social Welfare Services (Department Publications, 1998 -2013

Courses pursued under the BTEA l In 2008, the number of available places were split equally between the SLO and TLO options. Over 90% of individuals in the TLO stream were undertaking degrees while 90% of those in the SLO stream were undertaking the Post Leaving Certificate (PLC) Qualification l The PLC represents the cornerstone of Further Education and Training (FET) in Ireland, accounting for 50 per cent of total FET enrolments, and over 80 per cent of full-time enrolments l PLC programmes provide specific vocational skills to learners, with programmes directly related to areas of labour demand. Most popular course are in the areas of Childcare, Nursing, Business Studies, Sports / Recreation and Community / Health Services.

Evaluation approach l Used longitudinal administrative data to examine the labour market outcomes of 2008 entrants to TLO and SLO in 2012 and 2014 l These outcomes were compared to control groups of individuals also unemployed in September 2008, with similar characteristics, who while eligible for BTEA assistance did not avail of it l The Impact of the BTEA was assessed by comparing the proportions of “treated” individuals in employment 4 – 6 years following commencement of the programme with those of the control groups.

Results l The programme was found to be highly ineffective. Individuals who entered the SLO programme were 30. 5% less likely to be in employment in 2012, with the disadvantage falling to 25. 4% in 2014 l TLO individuals who entered the programme were 19. 9% less likely to be in employment in 2012, with the disadvantage falling to 14% in 2014 l There was evidence of administrative failures particularly about the effectiveness of monitoring arrangements that were in place for the BTEA scheme at that time and the degree to which eligibility rules regarding progression were being enforced.

Was the Quality of Education Delivered under the BTEA to Blame? l A forthcoming evaluation of the PLC using a completely different dataset, looking at a different time period and using a different control group will show that PLC participants are much more likely to be in employment 5 years following entry to the programme l When the BTEA participants are removed from the PLC population the positive impact of PLC participation on employment rises relative to the control group l Relative to the control group the BTEA (SLO) participants were again over 20% less likely to be in employment relative to the control group.

Summary 1 l Statistical profiling is a potentially useful tool that will allow PES officers to filter suitable individuals onto various activation paths l Profiling in its own right will have no impact on outcomes and will only be effective if the activation pathways positively influence individual outcomes l Evidence from a major second chance programme in Ireland show that this form of activation actually damaged employment prospects l Educational quality did not appear to explain this result. There were certainly administrative failings and it remains unknown if outcomes could have been better with more stringent conditionality and effective monitoring.

Summary 2 l It is not sufficient to assume that educational programmes that have been demonstrated to be beneficial to school leavers will automatically improve the employment outcomes of job seekers. l We could find very limited evidence of evaluations supporting the use of second chance education programmes as large scale labour market activation measures. l There is more evidence supporting the benefits of short-term skill intensive training, which also tend to be much cheaper allowing for greater numbers of claimants to benefit. 17