Evaluating What Works for Whom in Active Labour Market Policies
| Published date | 01 December 2015 |
| Date | 01 December 2015 |
| DOI | 10.1177/138826271501700403 |
| Author | Thomas Bredgaard |
| Subject Matter | Article |
436 Intersentia
EVALUATING WHAT WORKS FOR WHOM
INACTIVE LABOUR MARKET POLICIES
T B *
Abstract
In order to make informed and legitimate deci sions in labour market policies, European
and national policy makers need better knowledge of what type of interventions works
for whom. e European Commission and many Member States have high hopes that
‘experimental evaluation’ techniques (such as randomised controlled experiments,
systematic meta-analysis and econometric outcome evaluations) will deliver solid and
clear evidence to inform the de velopment of more rational decision-making processes.
is article reviews the evaluation literature on Active Labour Market Policy (ALMP)
and examines what works for whom, und er what circumstances. It assesses the st rengths
and weaknesses of the current dr ive towards ‘experimental evaluation’ and proposes an
integrated framework for ALMP evaluation that combines ‘experimental evaluation’
with ‘programme theory evaluations’ and quantitative with qualitative data collection.
Keywords: active labour market policy; employment policy; impact evaluation;
programme theor y; unemployment
1. INTRODUCTION
Active labour market policies (AL MPs) have been high on t he European agenda since
the Employ ment Guideli nes and Europe an Employment St rategy were lau nched a er
the Amsterdam sum mit in 1997. Since then, the European Commission has urged
Member States to make greater use of ALM Ps and pushed for cross-national learning
and evaluation (Europea n Commission 2015).
Eu rop ea n a nd n ati ona l d ec isi on m ak er s ha ve t ur ned to the re sea rc h a nd e va lua ti on
community to extr act ndings and seek inspiration for e ective policy-making.
However, the messages on ALMP from the researchers and evaluators are neither
clear-cut nor easily tra nsferred to political decision-making. First, t here is still a lack
* Associate Profes sor, Centre for Labour Marke t Research and Resea rch Centre for Evaluation ,
Department of Pol itical Science, Aalbor g University; Address: Fibigerst raede 1, DK-9220 Aalborg,
Denmark; phone: +4 5 9940 2605 / +45 2724 2128; e-mai l: thomas@dps.aau .dk.
Evaluating W hat Works for Whom inActive Labou r Market Policies
European Jour nal of Social Sec urity, Volume 17 (2015), No. 4 437
of systematic evaluation in Europe.
1 Second, the evaluation studies o en arrive at
contradictory results, making it d i cult to combine them in a coherent way. ird,
ALMPs are complex programmes that work under di erent contextual conditions,
and this makes t hem di cult to evaluate.
To overcome these problems, ALMP researchers and evaluators increasingly use
simplify ing techniques to synthesise and genera lise ndings. In ALMP eva luation, as
well as many other polic y interventions, empirical ndings are ra nked in a hierarchy,
in which randomised controlled trials (RCTs) and econometric impact evaluations
are considered the ‘gold standard ’ for evaluating what works (Pawson 2006 ; 2013).
rough systematic rev iews and meta-ana lysis, the evidence is su mmarised , quanti ed
and simpli ed to assess whether or not general classes or categories of interventions
work (cf. Card et al.2010; Kluve 2006; 2010).
is type of simpli cation process is problematic. First, the causal mechanisms
linking i nterventions and outcomes tend to become a ‘black box’. is makes it di cult,
or sometimes impossible, to ident ify why and how ALMPs work. We need contextual
and qualitative dat a and information to answer these ty pes of question. Second, most
ALMP interventions are complex, which make them di cult to evaluate. O en, the
target groups face major barriers to (re)integration in the labour market; techniques
for improving participa nts’ employability and employment opportunities are d i cult
to standardise; the interventions do not operate in isolation but in combination
with other policies a nd programmes and under di erent structural and economic
conditions; and outcomes are di cult to measure in the short-term. We, therefore,
need more sophisticated eva luation techniques for these type s of situation and public
problem.
e a rt ic le p ro ce ed s a s fo ll ow s. Fir st , A LM P i s d e ned in terms of its most common
components. Second, the programme eva luation literature on the impacts of ALMP
is reviewed and its shortcomings described. ird, suggestions for overcoming the
shortcomings of the current progra mme evaluation literature by combini ng impact
evaluations with prog ramme theory evaluations are put forwa rd.
2. DEFINING ALMPs
Historically, labour market policies evolved to address ‘market failures’ as well a s
socially and politically unacceptable outcomes of the labour market (see Bredgaard
et al. 2011a, Schmid et al. 1997, uy et al.2001). ere is tremendous national,
political and historical variation in what is perceived as a problematic situation in
the labour market that requires public intervention. Some of the most common
labour market problems include, for instance, a mismatch between supply and
demand for labour, long-term unemployment, lack of income support, skil ls
1 Surprising ly, the USA which spends much less on A LMPs has a stronger eva luation culture –
especial ly outcome evaluations – th an most European count ries (Kluve 2006, 2010).
To continue reading
Request your trial