Key to this approach is a model of personal service, i.e. profiling clients according to their employability, based on statistical predictive model and focused counselling. Resonating with Employment Guideline 7 (Enhancing the functioning of labour markets), in particular in the area of tailoring services and interplay of active labour market policy measures (ALMPs) with “passive” measures, the overall emphasis was to move to an inverted pyramid model. The focus is put on self-help (through strengthening career management and life-long learning and career support) for the people closer to the labour market and to individual case management services focused on those hardest to employ and who are furthest away from the labour market. StAP is a tool to segment clients into specific groups who need to access different levels of services.

The situation on the Croatian labour market has continued to improve. According to the latest available Labour Force Survey data published by Eurostat, the employment rate of the population aged 20 to 64 increased from 62.9% in the third quarter of 2016 to 65.6% in the third quarter of 2017. Over the same period, the unemployment rate decreased from 10.9% to 9.0%, and the share of long-term  unemployed (12 months or more) in the total number of the unemployed fell from 49.7% to 42.9%. Based on those data, it can be calculated that the long-term unemployment rate, i.e. the share of the long-term unemployed in active population, declined from 5.4% to 3.9%.

It should be mentioned, for the sake of comparison, that in the EU-28 as a whole, the share of the long-term unemployed in the total number of the unemployed in the third quarter of 2017 stood at 44.7% and the long-term unemployment rate amounted to 3.3%. Thus, the long-term unemployment rate in Croatia was slightly higher than the EU-28 average, while the share of the long-term unemployed in the total number of the unemployed was lower than the EU-28 average.

Identifying clients’ risk of not becoming employed within a year by using an IT-tool, the practice aims to achieve the following:

• provide needs-based services to jobseekers;
• facilitate the work of employment counsellors in managing the inflow of cases; and
• efficiently allocate resources over a longer time period.

A statistical predictive model of employment within 12 months from entry to the unemployed register (for newly registered) was designed using 970,000 entries in 2012-14, tracked until the end of 2015. Dropout from the register within 12 months was also predicted and implemented into counsellors’ IT support application.

A number of job-seeker attributes are used in prediction of employment including indicators for human capital (education and experience), the history of employment and unemployment, affiliation to vulnerable groups, economic sector (occupation and industry of previous employment, area of education), and monthly labour supply-demand ratio (education-level specific).

Client segmentation is based on estimating the probability of employment within 12 months following the registration and has four risk group outcomes:

• high probability of employment;
• no specific indication (medium risk);
• reduced likelihood of employment; and
• very low probability of employment.

Low employability groups are recommended to be given more frequent and more in-depth counselling, as well as priority for other activities (including ALMMs), while highly employable group is recommended to be nudged towards online tools. Low employability groups, who tend to have above-average predicted drop-out probability as well should receive more information and advice on their rights and obligations.

Initially, 64 counsellors in the first two regional pilot offices used StAP. At the end of the piloting phase, a total of 84 counsellors used StAP (the final group of counsellors started in August 2017).

During the period December 2017 to March 2018 an additional 10 offices started using StAP. By February 2018, in total, 285 counsellors (around 60% of all counsellors) were trained and had started to apply StAP profiling. Another expansion will happen in May 2018 when an additional five regional offices will start implementing StAP profiling.

Outcomes that have been identified so far include the following:

• Counsellors’ focus is directed to clients needing more support.
• Better distribution of counsellors’ time.