This review discusses the evolution and performance of active labour market policies in Latvia since 2012, as the Latvian labour market emerged from the particularly severe effects of the global financial crisis. In the immediate aftermath of the crisis, Latvia experienced one of the largest increases in the unemployment rate – 15 percentage points – of any OECD country and Latvian workers experienced a drop in both real and nominal wages. Yet the recovery from the crisis was relatively strong and rapid, and Latvia’s unemployment rate halved between 2010 and 2015. Nevertheless, in 2018, the unemployment rate was still at 8%, close to pre-crisis levels in Latvia, but above the OECD average. Moreover, long-term unemployment has remained a challenge in Latvia: in 2017, just seven other OECD countries had a higher proportion of the labour force unemployed for 12 months or more.

Latvia’s Inclusive Employment Strategy 2015-2020 places significant emphasis on developing and improving active labour market policies to help unemployed people access good jobs, with a particular focus on disadvantaged groups. While participation in active labour market policies remains relatively low in Latvia, the menu of such policies has expanded and diversified in recent years. There has been a shift away from public works towards providing employment incentives alongside rehabilitation for the long-term unemployed, while programmes have also been introduced to promote job seekers’ mobility across Latvia’s regions.

The bulk of this review is devoted to evaluating selected active labour market policies in Latvia. In particular, the review focusses on three types of activation measures: (1) training; (2) measures to support mobility and entrepreneurship for Latvia’s regions; and (3) wage subsidies for Latvia’s most vulnerable groups. The analysis uses detailed linked administrative data collected by several key agencies in Latvia between January 2012 and October 2017 and specialised econometric techniques to estimate the effectiveness of the selected measures.

Training for the unemployed has had positive effects on labour market outcomes, although the voucher system used to allocate training may be improved. Participants in both formal trainings – which seek to build specific, accredited skills – and non-formal trainings – which seek to build more general skills such as languages and information and communications technology – experienced an increase in their chances of finding a job and in their earnings. While these effects differed according to the gender, age, and social assistance receipts of training participants, virtually all types of participants benefited from taking part. In addition, combining training for the unemployed with other active labour market policy measures, especially measures to support regional mobility, appeared to boost effectiveness. Nevertheless, even though the voucher system used to disperse training carries many advantages, some disadvantaged groups may need additional support when using their vouchers, training providers are not distributed evenly across municipalities which sharpens the need for supporting regional mobility, and the voucher system currently in place in Latvia may compound so-called lock-in effects.

Active labour market policies have also been explicitly deployed to address, and even exploit, the sizeable differences in labour market outcomes between Latvia’s regions. The Riga region and surrounding Pieriga region have substantially lower unemployment rates and higher job vacancy rates than the other regions in Latvia and there has been a long-run trend of individuals migrating towards urban and suburban areas. Since 2013, the Latvian State Employment Agency has offered financial support to those accepting job offers more than 20km away, a programme that increased the probability of an unemployed person moving to take up a job by around one-half.

The Latvian State Employment Agency also offers a targeted programme of subsidised employment, which has helped many disadvantaged groups (re)connect with the labour market. The long-term unemployed, older workers, and young people were all more likely to hold jobs – even non-subsidised ones – after a spell of subsidised employment. Nevertheless, employment subsidies impose a heavy administrative burden on employers, which may discourage their participation. Moreover, after the end of any employment subsidies there are no clear positive effects on the labour market outcomes of persons with disabilities.

The analysis undertaken in this review suggests several key policy messages, which may improve the performance of active labour market policies in Latvia:

• Introduce possibilities for less severe sanctions when individuals turn down job offers, but require that unemployed persons without family commitments accept job offers from anywhere in Latvia.

• Simplify the tool used to profile unemployed people, better link it to different streams of activation measures, and improve its accuracy by profiling unemployed people as soon as they register with the State Employment Agency and making better use of existing statistical information.

• Extend activation measures to those who are not (yet) unemployed and provide online services to those unemployed people who are more likely to resume work quickly.

• Revise the voucher system used to disperse training for the unemployed, by reducing the time for which individuals must wait to receive a voucher and lengthening the time for which vouchers are actually valid, in order to limit lock-in effects.

• Monitor choice and competition in the training voucher system as the number of training providers is reduced and ensure that caseworkers are able to provide special support to those disadvantaged voucher recipients who need help in exercising effective choice.

• Enhance regional mobility support for training participants, extend regional mobility support for families by arranging for additional access to credit, and link access to regional mobility support to the profiling tool rather than requiring people to have been unemployed for at least two months before becoming eligible.

• Consider differentiating the programme of employment subsidies for persons with disabilities according to the degree of assessed disability or work capacity.

• Continue to invest in building and maintaining a well-functioning data infrastructure and develop mechanisms for conducting ongoing monitoring and evaluation of active labour market policies.