Denmark

X

X

X

X

X

X

X

X

X

(X)

X

X

X

X

A. Specific reference: (author(s), year, title, source, pages)
Andersen, S. H. (2017) Disconnected youth: Past, present & future. Rockwool Foundation Research Unit, Study Paper 116. Copenhagen.

B. Main topic covered by this specific study and how it relates to the barriers identified
Barriers to education and employment among young member of the NEET-group aged 19-29 years. The author argues that we should regard youth vulnerability as a continuous variable (in the paper such youth are called ‘disconnected youth’ and disconnectedness should be seen as a continuum) and be more aware of the fact that this group of young people is very heterogeneous. Hence, the report investigates how background characteristics, risk behavior and future prospects differ between different subgroups among young members of the Danish NEET-group. 

C. Target group(s), information on how the target group is identified
The target group is categorized as ‘disconnected youth’ in Denmark aged 18-29 years and identified as members of this age group who have neither been in education nor had a job for two consecutive years during the period 2000 to 2013. A usual definition of NEET is ”young people aged 16-24 who are neither in education, employment or training”. A Danish delimitation of NEETs is used since in a

Danish setting 18 and 30 years are crucial thresholds. Hence, eligibility for and level of welfare benefits as well as requirements of the unemployment system (e.g. with regards to participation in active labor market programs), depend on whether the claimant is above or below 30 years of age, and citizens below 18 years of age are not eligible for such benefits.

D. Location of the study: Country, region, city 
Denmark.

A. Data stems from Danish population registers and registers on treatment for drug addiction, psychiatric hospitalization, early parenthood and criminal convictions. Data covers: For men (18-24) 115.769 individuals, men (25-29) 133.587 individuals. For women (18-24) 137.794 individuals, women (25-29) 188.250 individuals.

 

Criminal behavior is defined as having spent (any) time in prison in either t or t-1 (this is anything from having spent a night in prison to getting a maximum sentence of as much as 16 years) or having gotten probation.

Mental health problems are defined as having been hospitalized for at least one day due to a mental health condition at any point in time, both before and after t and t-1 where we observe the individual’s status as disconnected youth

Drug addiction is defined as having received treatments against drug abuse at any point in time, having been convicted of one or more drug related crimes in t or t-1 (regardless of the sanction), and having been hospitalized for conditions related to drug abuse at any point in time.

 

Early pregnancy as becoming a parent before age 22. This might appear late by international

standards, but reflects the fact that actual teenage pregnancy (defined as conceiving a child before age 19) is very rare in a Danish context (as of 2015, age of first parity in Denmark is 29.1).

First, the study identifies distinct subgroups of disconnected youth through cluster

analysis. Second, it uses standard regression models to estimate the correlation between a) cluster

membership and background characteristics and b) cluster membership and subsequent membership of the group of disconnected youth. 

The analysis distinguishes between two subgroups of male ’disconnected youth’:

1. The first and largest group engages in risk behavior, such as crime, drug use and early parenthood, and at elevated levels compared to average male youth.

2. The second and smaller group engages in these activities at even higher levels. Moreover, they also suffer from mental health problems, and significantly more have grown up with mothers with mental health problems and in broken homes.

 

The analysis distinguishes between three subgroups of female ’disconnected youth’:

1. The first group suffer from mental health problems,

2. The second group struggles with all problems, but drug addictions in particular. In the second cluster there is a significant overrepresentation of persons who have grown up in a home with maternal mental health problems

3. The dominant problem in the last cluster is early parenthood but it also struggles with problems pertaining to mental health and drug use.

NB. The barriers themselves are not characterized in the article beyond the definitions provided above.

G. Brief description of the specific barrier(s) the target group faces as identified in this study
Barrier 1: Criminal behavior (see definition above). 

Barrier 3: Mental health problems (see definition above).

H. Quality of the study: High, medium, low + comments
We regard the quality of the study as high given a comprehensive dataset (full population of the target group in Denmark 2000-2013) and given well-conducted analyses. However, the impact from barriers on youth outcomes (education/ employment) is not a causal analysis.

A. Specific reference: (author(s), year, title, source, pages)
Børne- og Socialministeriet (2016) Socialt udsatte og kontanthjælpsreformen. Velfærdspolitisk Analyse nr. 9. December 2016.

B. Main topic covered by this specific study and how it relates to the barriers identified
The topic of this analysis produced by the Danish ministry of Children and Social Affairs is how a reform of the Danish cash benefits system taking effect January 1, 2014 affects vulnerable young people receiving such benefits. Cash benefits is a means-tested type of welfare benefit for unemployed persons in Denmark with no other sources of income. The reform implied that cash benefits were abolished for all young recipients (18-29 years) with no professional education (‘erhvervskompetencegivende Uddannelse’). Instead, this group of young unemployed became entitled to a new type of benefits (called ‘educational benefits’) with a lower amount of transfer income. The introduction of this lower benefits type intended to provide young people with an economic incentive to enroll in education or find a job. The aim of the study is to analyse the extent to which the reform also succeeded in lowering the number of vulnerable youth aged 25-29 years receiving educational benefits compared to the same group of persons previously receiving cash benefits – as well as the extent to which the target group increases its enrollment in education or takes up a job. Hence, the study sheds light on the extent to which economic incentives were prior to the reform a barrier to these young peoples’ path toward taking education or finding a job.

C. Target group(s), information on how the target group is identified
The target group is vulnerable young people aged 25-29 years January 1, 2014 who received educational assistance, approximately 5000 persons. Vulnerable youth is defined as persons who live up to one or more of the following three criteria during the last three years:

It is important to notice that far from all vulnerable young people aged 25-29 were affected by the cash benefits reform. The reform entailed legal provisions requiring the municipalities to categorize young benefits recipients into one of three categories:

Only the first two categories of recipients were affected by the reform. Hence, only the vulnerable young persons aged 25-29 years not categorized as activity ready were affected by the reform – and hence only those persons are in the treatment group investigated by the analysis.

D. Location of the study: Country, region, city 
Denmark.

E. Data (a.), specific measures (b.), and methods (c.), please also specify whether the study deploys cross-country data (d.)
Data stems from the Danish DREAM-register, registries under the National Board of Social Services and the National Social Appeals Board, the National Registry on Alcohol Treatment, and other registries available through Statistics Denmark.

Vulnerability is measured according the three criteria mentioned above. Employment and education is measured via the Danish DREAM-register.

Concerning methods, the study uses an econometric difference-in-difference approach. The authors use vulnerable cash benefit recipients aged 30-34 (vulnerable according to the same three criteria mentioned above) as the control group. Cash benefit recipients above the age of 30 years were not affected by the reform and given a small age difference compared to persons aged 25-29 years are likely to be relatively similar to the target group. 

F. Overall characteristic of the barriers identified by this study
The study investigates the extent to which economic incentive structures related to the Danish cash benefits system prior to the cash benefits reform constituted a barrier to the enrollment in education or the taking up of a job among vulnerable youth (i.e. young persons with the three mentioned characteristics).

Overall, the study finds that the increased economic incentives work – also in relation to vulnerable youth – in the sense that the share of vulnerable persons aged 25-29 years with no professional qualifications who receive cash benefits drop by 5 percentage points after the reform. The similar group of vulnerable youth aged 30-34 years, who are not affected by the same reduction in transfer income, sees no similar drop after the reform.

The report also contains an analysis of the extent to which vulnerable youth aged 25-29 leave the cash benefits system to a higher extent after the reform in favor of either enrollment in education or a job. Comparing this exit rate for this age group before and after the reform and using youth aged 30-34 as control group the exit rate doubles after the reform.

 

The last analysis pertains to all vulnerable youth aged 25-29 receiving cash benefit before and educational assistance after the reform - and not just those who are directly affected by the economic incentives ingrained in the reform (i.e. those receiving educational assistance categorized as educationally ready).

G. Brief description of the specific barrier(s) the target group faces as identified in this study
Barrier 1 Insufficient economic incentives to start education or find a job. See above.

Barrier 3 [Title, for example: Lack of proficiency in the host country language] 

H. Quality of the study: High, medium, low + comments
The quality of the study is medium – given that many unobserved differences may exist between the vulnerable 25-29 years old and the control group of vulnerable youth aged 30-34 years old.

I. Other comments: 

Bolvig, I. et al. (2019) Unge uden job og uddannelse - hvor mange, hvorfra, hvorhen og hvorfor? En kortlægning af de udsatte unge i NEET-gruppen. København: Det Nationale Forsknings- og Analysecenter for Velfærd. Available at: https://www.vive.dk/media/pure/14561/3641625 

B. Main topic covered by this specific study and how it relates to the barriers identified 

Barriers to education and employment among the most vulnerable part of the youth NEET-group in Denmark (that the authors name ‘vulnerable NEETs’). The report consists of three parts:

1. A literature review surveying research (mainly Danish) on the barriers and challenges that confront such young people when it comes to taking education and/ or finding employment.

2. A register-based study of the absolute number of young people in the target group, its size as a proportion of young people aged 19-24 years in Denmark, and its geographical distribution.

3. Barriers to education and employment among members of the target group as perceived from the Danish municipalities and the caseworkers responsible for helping them.

C. Target group(s), information on how the target group is identified 

The target group is defined as young people in Denmark aged 19-24 years who have not been enrolled in education or had any employment for a full year (2017). This group corresponds roughly to the most vulnerable young people receiving services in the Danish job centers. Furthermore, the quantitative register-based study (2.) defines the target group as not having tertiary education and as not having been awarded disability pension at the age of 18 years. 

D. Location of the study: Country, region, city 
Denmark.

Ad 1: Literature review: The literature review has focused on Danish studies on vulnerable youth and their characteristics and using the NEET-category published since 2008. The search found 85 such studies and selection process resulted in 12 studies being included.

 

Ad 2: We will not draw on the insights from the quantitative register-based study and therefore not specify types of data included.

 

Ad 3: Interviews in 6 Danish municipalities (Lolland, Randers, Varde, Svendborg, Horsens, Aalborg). A total of 10 focusgroup interviews were conducted with a total of 44 managers and caseworkers from: the youth departments of the jobcenters in these 6 municipalities, their youth guidance departments (ungdommens uddannelsesvejledning, UU), teachers from lower secondary schools, and a single manager from preparatory basic education and training (den forberedende grunduddannelse, FGU).

The study does not apply cross-country data.

F. Overall characteristic of the barriers identified by this study 

As it is evident from the outline of the three parts of the report above, the report covers both individual barriers and barriers relating to public employment services. The coding below focusses on barriers related to public employment services, but initially outlines (as barrier no. 0) the barriers identified in the report relating to the individual. 

 

Risk-factors that can be identified in primary/ lower secondary school:

The study finds that risk-factors among parents are less important than risk-factors among the young people themselves (i.e. their own risk-taking behavior).

Barrier 1: Lack of coherence and coordination. Overall, the study identifies problems in the municipal youth services relating to a lack of coherence and coordination among the many different institutions and actors involved in assisting vulnerable youth. This problem pertains to both minors (15-17 years) and adult youth (above 18 years). Many municipalities have during recent years chosen to co-locate the job centers and youth guidance departments in the same buildings. Nonetheless, several municipal employees find that a sufficiently coherent and integrated service targeting vulnerable youth is still lacking. This problem also relates to the fact that many different actors are involved in the interventions and the support targeting both minors (<18 years) and adults (> 18 years). A further complication is that there is a break and no smooth transition in the support provided when a minor turns 18 years. 

Barrier 2: Lack of common targets and approaches. There is a lack of common approaches and targets among caseworkers working for the youth guidance departments and those working for the job centers. Roughly speaking, according to the report, the jobcenter caseworkers aim at the fastest possible part toward education while the youth guidance caseworkers aim at the best and most sustainable part toward education (to minimize the risk for subsequent dropout). Moreover, lack of research-based evidence concerning the best types of support and interventions leaves much discretion to the individual caseworker. Therefore, caseworker practices and decision on the best type of support differ very much. 

Barrier 3: Lack of integrated IT-systems. Different IT-systems pertaining to different part of the municipal administration are insufficiently integrated implying that caseworkers find it difficult to achieve sufficient oversight of the life situation of vulnerable young citizens and to coordinate efficiently. 

Barrier 4: Lack of sufficiently early interventions. All interviewed caseworkers argue that early support and interventions are crucial if vulnerable youth are receive adequate support during their upbringing and youth. They find that vulnerable youth are challenged by complex (familial and social) problems at a very young age and may have different types of medical diagnose—also from an early age. Youth guidance caseworkers need early warnings from the schools if a pupil risks dropout. Nonetheless, the reality is that they are typically first advised when the young person has already dropped out. 

A. Specific reference: (author(s), year, title, source, pages)
Larsen, M. & Amilon, A (2019). Tilbagetrækningsalder og tilbagetrækningsårsager. Opgørelser på Ældredatabasen. VIVE Det Nationale Forsknings- og Analysecenter for Velfærd.                                                               

B. Main topic covered by this specific study and how it relates to the barriers identified
One of the main purposes of the report is to examine the influence of different factors on retirement age (actual retirement and planned retirement). 

C. Target group(s), information on how the target group is identified
In the analyses of actual retirement is the target group 67-year-olds. In the analyses of planned retirement is the target group 57-year-olds. 

D. Location of the study: Country, region, city 
Denmark. 

E. Data (a.), measures (b.), and methods (c. ), please also specify whether the study deploys cross-country data (d)

a. The report builds on data from The Danish Longitudinal Study of Ageing (Ældredatabasen), which is a questionnaire survey for a representative sample of the Danish population aged 52 and older. Data are collected in 1997, 2002, 2007, 2012 and 2017. The data are merged with register data from Statistics Denmark.

b. Measurement of actual retirement age is based on the question “how old were you when you stopped working?”. Measurement of planned retirement age is based on the question “When do you expect to stop [working]?”.

Reasons for retirement are divided into four groups related to:

1.  Economy

2.  Family/leisure time

3.  Work

4.  Health

Group 1 and 2 are categorized as pull factors and group 3 and 4 as push factors. “Push” means that a person are pushed out of the labor market, for example due to health problems or condition at the workplace (involuntary retirement). “Pull” means that there are conditions outside the labor market that make it more attractive to retire than to stay in the labor market.

c. The report reports findings as simple cross-tabulations and frequency tables.

d. The study use Danish survey data. 

F. Overall characteristic of the barriers identified by this study
Seniors retire to a greater extent due to pull factors than push factors. Among the 67-year-olds in 2017, 45 percent retired solely due to pull factors, while 31 percent retired solely due to push factors. 18 percent retired due to a combination of pull and push factors. The remaining 5 percent cannot be categorized.

 

The reasons for retirement do not differ much for men and women. However, there are differences between educational groups. Push factors are most important for seniors without a vocational or further education.

G. Brief description of the specific barrier(s) the target group faces as identified in this study

Push factors. Poor health is the single most important push factor. However, together are 9 work related factors a bit  more important than poor health. Some of the work related factors used to explain retirement were “was fired”, “to avoid poor health” and “did not thrives at the workplace”. 

Pull factors. Frequent used reasons for retirement are the possibility of receiving early retirement pension or pension and a desire to decide for yourself how your time should be used. But also a desire for spending more time with family and friends and on leisure activities are used a reasons for retirement. 

H. Quality of the study: High, medium, low + comments
Medium. Fine analyses to study the reason for actual retirement/planned retirement age. 

I. Other comments: 

A. Specific reference: (author(s), year, title, source, pages)
Bingley, P., Datta Gupta, N., Kallestrup-Lamb, M. & Pedersen, .J. (2021): Labour force exit in Denmark, 1980-2016: Impact from changes in incentives, in Börsch-Supan, A. & Coile, C.: Social security programs and retirement around the world reform and retirement incentives. University of Chigago Press, 109-131. 

B. Main topic covered by this specific study and how it relates to the barriers identified
The average age at exit from the labour force to the early retirement program (efterløn) or to old-age pension has decreased from 1980 to 1998, and increased from 1998 to 2014. The purpose of the chapter is to determine to which extent this observed trend reversal can be explained by changes in incentives.

C. Target group(s), information on how the target group is identified
55-69-year-olds. 

D. Location of the study: Country, region, city 
Denmark. 

a. Register data

b. Two central measures are:

c. The authors simulate SSW and ITAX using data including information on earnings, benefits for the largest programs supporting consumption in retirement and pathways to these benefits. Figures showing the development in SSW and ITAX with calendar years and age are discussed in relation to policy changes, and OLS regressions of the effects of ITAX on employment rates are estimated.

d. The study use Danish data. 

F. Overall characteristic of the barriers identified by this study
The study find a strong relationship between simulated incentives and employment in Denmark and show that incentives to some extent explain the trend reversal. 

H. Quality of the study: High, medium, low + comments
Medium.

I. Other comments: 

A. Specific reference: (author(s), year, title, source, pages)
Amilon, A. & Larsen, M. (2019): Forventet tilbagetrækningsalder og arbejdsrelaterede forhold. Opgørelser baseret på ældredatabasen. VIVE Det Nationale Forsknings- og Analysecenter for Velfærd.

B. Main topic covered by this specific study and how it relates to the barriers identified
The purpose of the article is to examine the influence of work related factors on the planned retirement age, and whether the influences of these factors differ for men and women, for unskilled/skilled workers and white-collar workers/civil servants, and for the private and public sector. 

The target groups are wage earners, who were 52, 57 or 62 years old. The status as wage earners are self-reported (based on the questionnaire survey). 

Denmark. 

a. The report builds on data from the 5^th wave of The Danish Longitudinal Study of Ageing (Ældredatabasen), which is a questionnaire survey for a representative sample of the Danish population aged 52 and older. Data are collected in 2017. This report use data for wage earners, who were 52, 57 or 62 years old in 2017. The data is merged with register data from Statistics Denmark.

b. Measurement of planned retirement age” is based on a question, where the respondents are asked about what age they expect to retire.

The work related  factors are measured by the following:

c. Ordinary least squares regression (OLS) are used to analyse the influence of work related factors on the planned retirement age. The models include control for age, education, health and economic factors.

d. The study use Danish survey data. 

F. Overall characteristic of the barriers identified by this study
The results show that many  of the work related factors has a significant influence om the planned retirement age, but also that there are some differences between the examined subgroups. 

G. Brief description of the specific barrier(s) the target group faces as identified in this study
For the unskilled/skilled workers are only two of the work related factors associated with the planed retirement age: physically demanding work and stressful work. For the white-collar workers/civil servants are most of the examined work related factors associated with the planed retirement age. Exceptions are: physically demanding work, satisfaction with cooperation with colleagues and option of less demanding job. However, the work related factors with the strongest association for the white-collar workers/civil servants are: custom with regard to retirement age, dissatisfaction with working hours, stressful work and influence on work situation.

 

More of the work related factors is associated with the planned retirement age for women than for men and for workers in the public sector than workers in the private sector. 

H. Quality of the study: High, medium, low + comments
Medium. Fine analyses to study relationship between work related factors and planned retirement age.

The authors argue for the advantages of focussing on planned retirement age: (1) previous research have shown strong correlation between planned retirement age and actual retirement age, and (2) it makes it possible to obtain reliable and detailed information about job related factors (it is more difficult to obtain reliable information on jobrelated factors before retirement after the person is retired). 

I. Other comments: 

A. Specific reference: (author(s), year, title, source, pages) 

Jakobsen V, Jeppesen T, Kleif HB, Mehlsen L, Thuesen F. Kortlægning af integration i kommunerne. Kommunernes indsatser for at bringe ikke-vestlige indvandrere i beskæftigelse. VIVE - Det Nationale Forsknings- og Analysecenter for Velfærd; 2021. 

B. Main topic covered by this specific study and how it relates to the barriers identified 

This reference is a report published by VIVE in 2021. The report was a survey- and interview-based mapping of what Danish municipalities do to bring unemployed non-western immigrants into work. As part of the mapping, survey respondents (managers and caseworkers working in the integration field in Danish municipalities) replied to a questionnaire on their work practices directed towards reaching this goal. Among other topics, the survey investigated which barriers the target group faces to obtaining employment according to the respondents. 

C. Target group(s), information on how the target group is identified 

Unemployed non-western immigrants in Danish municipalities. The target group was identified as those individuals with a non-western background who are registered as recipient of unemployment benefits, welfare benefits, integration benefits and the like in the Danish municipal job centers, and who by law are obliged to attend interviews and participate in programs to help them find a job.

 

The report distinguished between newly arrived immigrants (typically refugees and family reunited persons) and immigrants who have been living in Denmark for a longer period. The Integration Act regulates rights and obligations pertaining to the first group (‘integration law citizens’), while the Act on an Active Employment Effort regulates rights and obligations pertaining to the second group (‘employment law citizens’). 

D. Location of the study: Country, region, city  

Denmark as a whole and its 94 municipal job centers. 

a. The report builds on:

1. An internet-based survey to managers and employees in Danish municipal job centers who work professionally with this group of citizens (Response rate= 83% building on N=79 for caseworkers working under the Danish integration law, and response rate=49% building on N=47 for caseworkers working in the Danish law of employment)

2. Case studies in 10 Danish municipalities encompassing in total interviews with 58 interview persons. 

b. The survey encompasses a battery of 7 items on types of barriers that the target group faces when seeking work. These items are: insufficient Danish proficiency, lack of knowledge on the Danish labour market, lack of professional qualifications as compared to the demand in the labour market, lack of motivation to find a job, Ill health or insufficient health coping, insufficient coping with everyday life challenges, social control.

c. With respect to the survey items on barriers, the report reports findings as simple cross-tabulations and frequency tables. The findings are reported separately for integration

d. The study uses Danish survey data. 

The report distinguished between the 7 types of barriers mentioned above (see E,b). These barriers focus on knowledge of the Danish language, general knowledge of the labor market, professional qualifications, job motivation, health, everyday life challenges, and social control.

 

The study finds that the main barriers the two main target groups, both men and women, faces are insufficient Danish proficiency, lack of professional qualifications as compared to the demand in the labour market, and lack of knowledge on the Danish labour market. 

G. Brief description of the specific barrier(s) the target group faces as identified in this study 

Barrier 1. Lack of proficiency in the host country language: For unemployed integration law immigrants (i.e. newly arrived immigrants) proficiency in Danish is the main challenge. However, it is also a challenge to unemployed employment law citizens (i.e. immigrants who have been living in Denmark for a longer time). 

Barrier 2. Lack of qualifications relevant in relation to labour market demands: For unemployed employment law citizens this is the main challenge, but it is also a challenge to unemployed integration law citizens. 

Barrier 3. Lack of knowledge on the Danish labour market: This is mainly a barrier to integration law citizens, and mainly to the females in this group. It is especially is less of a barrier to male employment law citizens, but also to some females in this group. 

Barrier 4-7. The respondents also identify a number of the other barriers mentioned in the survey as important obstacles to the target groups employment. Especially among female newly arrived citizens (integration law citizens). They prioritize these barriers (after the three mentioned above) in the following order: Insufficient coping with everyday life challenges, lack of motivation to find a job, ill health or insufficient health coping, social control. 

H. Quality of the study: High, medium, low + comments
The quality of the study is medium. Barriers relating to the target group are not observed directly in relation to the target group itself (and there is no attempt to identify causal effects from barriers on their employment in an econometrics sense) but reported and ranked by caseworkers working professionally with this group. 

A. Specific reference: (author(s), year, title, source, pages)
Andersen, L.H., Dustmann, C. & Landersø, R. (2019): Lowering welfare benefits: intended and unintended consequences for migrants and their families. Study paper 138, The Rockwool Foundation. (https://www.rockwoolfonden.dk/publikationer/lowering-welfare-benefits-intended-and-unintended-consequences-for-migrants-and-their-familie )

B. Main topic covered by this specific study and how it relates to the barriers identified
The purpose is to examining the effects of the of Denmark’s Start Aid welfare reform: in 2002 the welfare benefits for refugees, who had been approved asylum after July 1^st was reduced to around 50% of the previous social assistance level. The new benefit was referred to as Start Aid benefits.

C. Target group(s), information on how the target group is identified
Refugees and family reunited to refugees. 

Denmark. 

a. The data used are register data from Statistics Denmark. The sample consist of refugees aged 18-55 when granted residency, and who were granted residency between January 1 2001 and December 31 2003 (refugees from Afghanistan and the former Yugoslavia are excluded from the data, because of contemporaneous conflict-induced changes in the inflow from the two countries, which could compromise the balancing of the sample around the reform).

b. In analysis of reform effects the following are used to measure the effects:

c. Focus on Refugees and family reunited to refugees: compare individuals who receives residency before the reform with individual who receives residency just after the reform using a regression discontinuity design.

d. The study useuses Danish data. 

H. Quality of the study: High, medium, low + comments
High.

I. Other comments: 

A. Specific reference: (author(s), year, title, source, pages)
Dahl, M. & Krog, N. (2018): Experimental evidence of discrimination in the labour market: intersections between ethnicity, gender, and socio-economic status. European Sociological Review, 34(4), 402-417. 

The main purpose of the article is to present evidence of ethnic discrimination in the recruitment process from a field experiment conducted in the Danish labour market. Three hypotheses are examined:

H1: Applicants with Middle Eastern-sounding names are less likely to receive a callback than applicants with traditional Danish-sounding names.

H2: The difference in callbacks between majority and minority applicants is larger among male applicants than among female applicants.

H3: The difference in callbacks between majority and minority applicants is larger when the majority applicant holds a popular name than when the majority applicant holds a low-SES name. 

C. Target group(s), information on how the target group is identified
People with Danish or Middle Eastern-sounding names. 

D. Location of the study: Country, region, city 
Denmark (all regions are included). 

a. Experimental data collected in 2015 and 2016. 800 applications were sent in response to 400 job (jobs from jobindex.dk). The jobs are in the following occupational categories: Administrative support, Construction & Extraction, Education, Health Care, Retail, Sales & Related.

b. Specific measures:

c. The article reportreports the results of a correspondence experiment, where applications are submitted to advertisements in the name of fictitious applicants. The authors randomized the assignment of three different treatments (ethnicity, gender and SES). Each job received two applications, one with a Danish sounding name and one with a Middle-Eastern-sounding name. Gender was randomly assigned pairwise. Thus, the applicants for any given job were either two men or two women. Finally, the Danish-sounding names were randomly assigned from to pools:  (1) the most popular names in Denmark and (2) the Danish low-SES names.

d. The study uses Danish data. 

The article finds evidence for ethnic discrimination in the hiring process

H1: Find a callback-ratio of 1.52 for job interviews between applicants with traditional Danish names and Middle Eastern-sounding names.

H2: Find a larger callback ratio for men than for women. Thus, minority men are more exposed to discrimination in the hiring process than minority women.

H3: Do not find that the difference in callbacks between majority and minority applicants is larger when the majority applicant holds a popular name than when the majority applicant holds a low-SES name.

Medium-High. The authors mention some limitations of the study. One limitation is that the sample of job openings not are representative for all job openings (which is very difficult to obtain)    . Another limitation is that the names used to signify low SES not necessarily match the employers’ perception of low-SES names.

I. Other comments: 

A. Specific reference: (author(s), year, title, source, pages)
Ali, S.M., Hansen, C.L., Tønder, A. & Graae, L.E. (2019): (Somaliske kvinders fortællinger om arbejde og ledighed. Væksthusets Forskningscenter (https://vaeksthusets-forskningscenter.dk/publikationer/somaliske-kvinders-fortaellinger-om-arbejde-og-ledighed/ )

B. Main topic covered by this specific study and how it relates to the barriers identified
The purpose of the report is to examine the reasons why women with origin in Somalia have a very low employment rate. This is done through a focus on the women’s narratives concerning their experiences in Denmark, including their experiences with the public employment services. Thus, a citizen's perspective is applied to examine the reasons for the low employment rate among the Somali women in Denmark.

C. Target group(s), information on how the target group is identified
Immigrant women with origin in Somalia. 

D. Location of the study: Country, region, city 
The interviewed women are resident in Copenhagen, Aarhus, Odense or Aalborg. 50% of the Somalian women in Denmark are resident in one of these four cities. 

E. Data (a.), specific measures (b.), and methods (c.), please also specify whether the study deploys cross-country data (d.)
a. Qualitative interviews with fifty 30-59-year-old immigrant women from Somali, and qualitative interviews with seven professionals. 

H. Quality of the study: High, medium, low + comments
The quality of the analyses in the report is good. However, the purpose of the report and our literature review differs. The report focus on the women’s narratives concerning their experiences in Denmark, which is important knowledge. However, experienced barriers is not necessarily actual barriers.

A. Specific reference: (author(s), year, title, source, pages) 

Bredgaard, T.  & Shamshiri-Pedersen, D.  (2018): Bedre Match Mellem Personer med Handicap og Arbejdsmarkedet. En litteraturgennemgang. Aalborg: Forskningscenter for Handicap og Beskæftigelse. 

B. Main topic covered by this specific study and how it relates to the barriers identified 

The study is a Danish literature review of research on different types of barriers affecting negatively the employment chances among persons with disabilities. The review covers mainly Danish but also, where relevant, international research on this topic. Moreover, the focus is mainly on empirical studies. The study by Bredgaard and Shamshiri-Pedersen distinguishes between barriers affecting the individual, the employers and the employment services. The coding in this scheme focusses on barriers relating to the individual.   

The target group ‘persons with disabilities’ is defined very broadly since the literature review includes all relevant studies whatever definition of disability they apply. Since, however, there is a strong focus in the study on Danish empirical research a definition of disability common in this research is relatively dominant in the literature review. In Denmark a relational concept of disability dominates (as opposed to the medical and the social concept) implying that ‘disability’ is to be understood as a joint product from health-related (bodily or mental) limitation and social barriers. As a practical social-scientific measurable phenomenon, this concept is operationalized as a self-reported condition, i.e. the target group is identified as those who self-report some sort of disability – irrespective of whether they have any sort of diagnosis or experience any particular barriers in the labour market or otherwise (see also Larsen, Jakobsen and Mikkelsen, 2022).       

D. Location of the study: Country, region, city  

Denmark, Nordic countries and to some extent other Western countries. 

a. Data stem from a literature search identifying 200 potentially relevant scientific publications. Although the search did not use a date interval criterion, almost all of these publications were published after 1990. 71 references focus on the supply-side of the labour market – i.e. persons with disabilities and their barriers.

b. Most Danish research included in the study deploys the relational concept of disability and the measure of disability describes above (see C).

c. Different search strings have been used to identify relevant research in Danish and international databases. Subsequently the researchers have scanned the literature to identify the most relevant

d. No cross-country data.  

The literature study identifies a number of barriers to employment among individuals with disabilities. Some of these barriers could be termed ‘restraining factors’ in the sense that they are not insurmountable barriers to employment but rather factors that in various ways decrease the employment chances among persons with disabilities. 

Barrier 1 Severity of disability. Several studies find that persons with minor (self-rated) disabilities have an employment rate that is almost on the same level as persons with no disabilities. However, persons with major (self-rated) disabilities experience much lower rates compared to persons with no disabilities. 

Barrier 2 Type of disability. Persons with mental illnesses have the lowest rate of labour market participation and the lowest employment rate among persons with disabilities. 

Barrier 3 Age. The prevalence of persons with disabilities in the population increases with age. Moreover, older persons with disabilities have lower work motivation and lower employment rates than younger persons with disabilities. 

Barrier 4 Education .Studies from Denmark show that persons with disabilities have a lower level of education compared to persons without a disability, and fewer are enrolled in education. This is unfortunate because a higher level of education among persons with disabilities decreases the employment gap to persons with no disabilities. Hence, although the level of education has gone up among persons with disabilities during the last 10 years in Denmark, persisting low levels of education among some persons with disabilities constitute a barrier to increasing their employment chances. 

Barrier 5 Work ability. Self-rated work ability among persons with disabilities correlates positively with their employment degree among persons already employed. Moreover, self-rated work ability correlates positively with intensity of job search among unemployed persons with disabilities. Self-confidence is also positively correlated with likelihood of employment. Hence, both low self-rated work ability and lack of self-confidence may constitute barriers to employment among persons with disabilities. 

H. Quality of the study: High, medium, low + comments 

Medium. Most of the studies included in the literature review are correlational. The quality of the literature review itself is also medium. There is an explicit and systematic search strategy but how the evaluation of the quality of the included studies was conducted is not specified.  

A. Specific reference: (author(s), year, title, source, pages) 

Holt, H., Larsen, M., Bach, H. B. , Jensen, S. (2015): Borgere i Fleksjob efter Reformen, København: SFI. 

The main topic covered by this report is the Danish reform of the governmental flexjob-scheme that took place in 2013. Flexjob is a Danish governmentally supported job for persons with a permanent and substantial reduction in their work ability.

 

Overall, the aim of the flexjob scheme is to help persons with either a disability or other health problem find a job – and at the same time avoid that persons with limited but some remaining work ability end up outside the labour market on disability pension. The Danish disability pension and flexjob reform taking effect in 2013 had the aim of increasing the possibilities for being employed in a flexjob with very few working hours. At the same time, the new legislation aimed at establishing an economic incentive for persons with a flexjob to increase the number of working hours.

 

Persons granted a flexjob have to find a job in the ordinary labour market, and they receive a salary from the employer according the number of hours they work and work intensity with which they work these hours in this job. For example, if they work 16 hours a week with an intensity of 50%, the employer has to pay them for 8 hours work. For the rest of the time up to 37 hours a week (standard working hours in the Danish labour market), they receive a supplementary subsidy from the state. Typically, the salary they receive for working ‘ordinary work hours’ exceeds the subsidy from the state. Hence, they should have an economic incentive to increase the number of hours they work. Providing persons in a flexjob with an economic incentive to increase their weekly number of work hours was a key element in the 2013 reform.

 

The report mainly investigates how persons who have been granted a flexjob according the new rules experience their employment situation. 

C. Target group(s), information on how the target group is identified 

The target group is persons who have been granted a flexjob. It focusses on those who have been granted a flexjob job, i.e. the subsidy, and found a job with a public or private employer. It also focusses on those persons who have been granted a flexjob, but who are still unable to find a job (and therefore receive the particular type of unemployment benefits for such persons, in Danish termed ‘ledighedsydelse’). 

Denmark. 

E. Data (a.), specific measures (b.), and methods (c.), please also specify whether the study deploys cross-country data (d.)
 

a. Data for the analysis stem from register data and a survey among persons in a flexjob (N= 1950) or who have been granted a flexjob but haven’t found one yet (receiving ‘ledighedsydelse’, N= 465). The sample of persons in a flexjob includes both persons who have found their flexjob prior to the 2013 reform and persons who have found their job after the reform. The response rate was 76%, which is high.

b. The specific measures concerning the effects from economic incentives are survey items for the respondents asking them whether they expect to increase their weekly work hours over the next three years (yes/ no/ don’t know).

c. The analysis uses standard methods for analyzing surveys (i.e. concerning statistical inference).

d. No cross-country data. 

F. Overall characteristic of the barriers identified by this study
Overall, the report finds that the 2013 reform is a success in the sense that the aim of bringing more people granted a flexjob with very few hours (up to 10 hours a week) in to work has been fulfilled. Prior to the reform this group constituted 9% of persons in a flexjob. After this reform, this group constitutes 38% of persons in a flexjob. In this sense the economic incentives ingrained in the reform appear to work according to the intention. Still, part of the explanation for the rise in the number of persons with a flexjob with few work hours must most likely also be ascribed to the fact that the reform tightened the rules providing access to disability pension. 

G. Brief description of the specific barrier(s) the target group faces as identified in this study
Barrier 1 Insufficient economic incentives. Concerning the aim of creating an economic incentive that would make persons awarded a flexjob increase their working hours, this aim is only partially fulfilled (p. 24, 71, 149). In the survey, all respondents have been asked whether they expect to increase the number of hours they work over the next three years, i.e. in the years after they reply to the survey. 17% of persons working in a flexjob with few hours (up to 10 hours a week) say that they expect to increase their working hours, while 6-9% of persons in a flexjob with more than 10 hours say so. Given that the reform aimed at that persons working few hours should increase their number of work hours, one may argue that these economic incentives have only fulfilled their aim to a very limited degree. 83% percent of those who work few hours (up to 10 hours a week) and 91-94% of those working more hours (above 10 hours a week) say that they do not expect to increase work hours. 

H. Quality of the study: High, medium, low + comments 

The quality of the study overall is medium given a good sample of persons in a flexjob for the survey and a high response rate. However, concerning the identified barrier, i.e. an insufficient economic incentive for increasing work hours, the quality is low. We merely have information on the self-reported expectation among persons in a flexjob concerning increasing their work hours over the next three years. Hence, the report does not provide evidence concerning potential causal effects (or lacking effects) in an econometric sense from an economic incentive on the number of hours worked. 

I. Other comments: 

We should remember, that the barrier identified in relation to increased work hours is not necessarily due to a lack of economic incentives. Persons in a flexjob (especially those with few hours) have a limited work capacity given disability or health problems. Therefore these health problem may be real underlying barrier to progress in work ability and hence an increase in work hours – and not an insufficient economic incentive.

A. Specific reference: (author(s), year, title, source, pages) 

Shamshiri-Petersen, D. and Krogh, C. (2020) ‘Disability disqualifies: A vignette experiment on danish employers’ intentions to hire applicants with physical disabilities’, Scandinavian Journal of Disability Research, 22(1), pp. 198–209. doi: 10.16993/sjdr.661 

B. Main topic covered by this specific study and how it relates to the barriers identified 

This article investigates whether Danish employers discriminate against persons with physical disabilities when hiring new employees. The analysis is based on a survey to a representative sample of employers in Denmark. The study aims at shedding light on the extent to which differential treatment negatively affects the employment chances of persons with physical disabilities in Denmark, i.e. the extent to which differential treatment is a barrier to their employment. More specifically the study examines the intentions of employers to hire a physically disabled applicant and whether financial compensation, public employment service recommendations, and the prospect of increased workload, influence such intentions. Results show that employers’ intentions regarding hiring drop significantly when they receive a description of an applicant who uses a wheelchair. Adding information regarding the opportunity to apply for financial compensation to the description increases such intentions slightly. This implies that disability in and of itself is a disqualifying factor and that the intended recruitment practise is discriminatory. 

C. Target group(s), information on how the target group is identified 

Persons with physical disabilities are not identified in a statistical sense but described to the respondents as the topic of the survey. Moreover, a vignette describes a fictitious job applicant to the respondents. 

D. Location of the study: Country, region, city  

Denmark. 

 

a. The study builds on a survey to a representative sample of Danish employers. 5000 workplaces were selected from the Danish company register, in which all Danish companies are required to be registered. The response rate was 41% (n = 1901).

b. The study contains five different vignettes. The survey also addressed questions about employer attitudes, knowledge, and experience regarding disabled employees. The survey also includes background information about company size, sector, industry, and geography.

c. The respondents were asked after reading the vignette: ‘On a scale from 1–5, 1 being very likely and 5 very unlikely, how likely is it that you would hire the applicant?’. The means of the vignette groups are compared on the 5-point scale using ANOVA.

d. The study uses Danish survey data. 

The study finds that if an applicant reports using a wheelchair, the intention of the employer to hire declines substantially. None of the factors thought to affect employers’ intentions either positively or negatively resulted in any notable changes. Information on financial compensation increased the number of employers who intended to employ the disabled applicant slightly, but neither information on workload nor a PES recommendation had any effect. 

G. Brief description of the specific barrier(s) the target group faces as identified in this study 

Barrier 1 Lack of employment opportunities due to negative differential treatment.

On a five point scale from 1 (very likely to hire the applicant) to 5 (very unlikely to hire the applicant) there is a drop of 1,71 when comparing the control group to the treatment group. The control group receives a description of the applicant with no disability. The treatment group receives a vignette text with added information that the applicant is a wheel chair user. For more description, see F.

 

The quality of the study is medium. The response rate among employers is relatively high and the vignette experiment well-designed. However, other studies question the external validity of survey experiments as compared to field experiment, see for instance Wulff, J. N. and Villadsen, A. R. (2020) ‘Are Survey Experiments as Valid as Field Experiments in Management Research? An Empirical Comparison Using the Case of Ethnic Employment Discrimination’, European Management Review, 17(1), pp. 347–356. 

I. Other comments: 

Amby, F. (2020): Jobcentrenes indsats, viden og holdninger, i: Bredgaard, T., Amby, F., Holt, H., Thuesen, F.: Handicap og Beskæftigelse – fra Barrierer til Broer. København: DJØF Forlag. 

B. Main topic covered by this specific study and how it relates to the barriers identified 

The main topic covered by this study is what Danish job centers do to help persons with disabilities find a job as well as knowledge on and attitudes toward this target group among employees and managers in the job centers. The author wrote his political science PhD. Dissertation on this topic with the title “Målgruppen der forsvandt: handicapområdets position i dansk beskæftigelsespolitik og hvordan ledige med handicap kan komme i ordinær beskæftigelse” (2015) arguing that a lack of focus and knowledge concering this target group characterizes Danish job centers. This article is a follow-up study of the empirical analysis presented in the PhD.-dissertation based on new data collected in the job centers.   

C. Target group(s), information on how the target group is identified 

The purpose of the analysis is to investigate how job centers handle unemployed persons with disabilities (the primary target group). In order to do so, the analysis focusses on case workers and managers who work with persons with disabilities. Since, however, persons with disabilities are not a category of their own in Danish job centers and can be found in different sections in these job centers focusing on different categories of unemployed (e.g. unemployment insurance benefits, sickness benefits, cash benefits), identifying respondents with knowledge on the primary target group is difficult. The analysis seeks to handle this problem by collecting survey data from three categories of employees in the job centers: 1. middle managers, 2. ordinary case workers (both 1. and 2. from different sections in the job center) and 3. Disability key persons (a position that should exist in all job centers implying that one employee is appointed in each job center whose function is to know disability and employment legislation in depth to be able to share this knowledge with other employees). 

D. Location of the study: Country, region, city  

Denmark/ Danish job centers. 

E. Data (a.), specific measures (b.), and methods (c.), please also specify whether the study deploys cross-country data (d.)
 

a. Data stem from three surveys conducted June-September 2019:

1. Survey to middle managers (N= 106, response rate 63%)

2. Survey to case workers (N= 175, response rate 61%)

3. Survey to key persons (N= 336, response rate 68%)

 

b. All three survey cover the same themes and use the same measures, survey questions on:

 

To analyse the extent to which the case worker believe the employment chances are good or bad compared to non-disabled unemployed persons, the survey targeting case workers also includes a survey experiment. Some respondent were presented with one vignette describing an highly employable accountant, others a vignette describing a cash benefits recipient with some years of experience working as a hotel receptionist. Two treatment groups of respondents received the vignette with added information that the unemployed persons were wheel chair users.  Respondents were asked to evaluate on a 5-point scale the extent to which they found it likely that this persons would on his own find a job.

 

c. Data are analysed using standard methods for statistical inference and correlational analysis (linear regressions, Pearsons’ R etc.). 

d. No.

F. Overall characteristic of the barriers identified by this study 

The study finds several different types of barriers to improving the employment chances of persons with disabilities in the administrative practices of the Danish job centers. 

Barrier 1 Lack of strategy and priority. Only a minority of Danish job centers have a strategy to improve the employment chances of persons with disabilities. In roughly half of Danish job centers, persons with disabilities are a low priority. 

Barrier 2 Lack of identification of unemployed with disabilities. Merely around 25% of job center systematically seek to screen unemployed for disabilities or practice other types of identification. A minority of job centers seek to evaluate whether unemployed belong to the target group for receiving compensatory measures. 

Barrier 3 Lack of counselling or courses targeting persons with disabilities. A minority of Danish job centers have special job counselling or job search courses for persons for disabilities. 

Barrier 4. Lack of belief in the employment chances of persons with disabilities. The survey experiment shows that case workers faith in the employment chances of persons with disabilities is much weaker that the similar faith concering non-disabled persons. Given that previous Danish research indicates that the case workers’ belief in the employment chances of unemployed persons in itself affect the likelyhood that unemployed persons find a job, this lack of belief is a barrier in itself . 

Medium quality – given high quality of information based on a triangulation of data from different surveys targeting different groups of employees in the job centers, but a strategy for identifying causal effects from the identified barriers is lacking.

I. Other comments: