Scientific article Andragoška spoznanja/Studies in Adult Education and Learning, 2024, 30(2), 63-80 DOI: https://doi.org/10.4312/as/16289 Shalini Singh, PhD, Director, Center for Lifelong Learning, University of Science and Technology Meghalaya & Visiting Researcher, Helmut Schmidt University of the Federal Armed Forces, contactingshalinisingh@gmail.com Søren Ehlers, Dr. Paed., Chairman, Center for Lifelong Learning, University of Science and Technology Meghalaya, soerenaugustehlers@gmail.com Shalini Singh, Søren Ehlers CHALLENGING THE MATTHEW EFFECT THROUGH INDIVIDUAL LEARNING ACCOUNTS? CASE STUDIES FROM DENMARK AND FRANCE ABSTRACT This paper discusses the role of Individual Learning Accounts (ILAs) as policy instruments in challenging the Matthew Effect and promoting self-directed learning among the non-participant but potential adult learners. The paper builds upon the Integrated Implementation Model of Winter (2012) explaining policy implementa- tion and the typology for classifying systems of lifelong learning according to their governance by Green (2000). The paper banks upon data from Denmark, France and the EU, and proposes a learner-centric model for the implementation of education policies along with revisions in Green’s typology. Keywords: low-qualif ied adults, non-participants, personal autonomy, personalised learning, social investments PREPREČEVANJE MATEJEVEGA UČINKA Z INDIVIDUALNIMI UČNIMI RAČUNI? ŠTUDIJE PRIMEROV IZ DANSKE IN FRANCIJE – POVZETEK Članek obravnava individualne učne račune kot političen instrument za preprečevanje Matejevega učinka in promocijo samostojnega učenja med neudeleženci izobraževanj, ki pa so še vedno potencialno učeči se odrasli. Prispevek na podlagi Winterjevega (2012) modela integrirane implementacije razlaga izvajanje politik in upo- rabi Greenovo (2000) tipologijo za klasif ikacijske sisteme vseživljenjskega učenja glede na sistem upravljanja. Na podlagi podatkov iz Danske, Francije in Evropske unije predlaga na učečega usmerjen model za implemen- tacijo izobraževalnih politik in revidirano verzijo Greenove tipologije. Ključne besede: odrasli z nizko stopnjo izobrazbe, neudeleženci, osebna avtonomija, posamezniku prilagojeno učenje, socialne naložbe 64 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 INTRODUCTION The EU has decided to establish a European Education Area by 2025 (European Com- mission [EC], 2020). In its agenda, Individual Learning Accounts (hereafter ILAs) have been prioritised since 2019 (EC, 2019c). The EU considers ILAs to be relevant policy instruments for achieving lifelong learning for all by ensuring that all adults participate in learning activities, irrespective of their socio-economic background. The EU suggests that its member states should ensure to provide diversified learning opportunities for address- ing the individual learning needs of all adults because of the long-term gains involved (EC, 2022; Fernandes & Kerneïs, 2020, 2021). To achieve this, the EU argues in favour of ILAs in the form of individual grants to finance learners rather than providers and is therefore demand-led rather than supply-led (Cedefop, 2023). Even though ILAs are an evidence-informed, well-elaborated and tested policy instrument to increase participation in learning among adults (and thereby the indicator reflecting resil- ient education systems), their effective implementation is a major challenge, even for Nordic countries like Denmark which invest abundant resources in education (EC, 2019a; Organ- isation for Economic Co-operation and Development [OECD], 2019a). This highlights the wide gulf between policy formulation (which is rather generalised) and implementation (more concrete and adapted version which might take decades to fill up; Singh & Ehlers, 2019). Meanwhile, the ineffective or incomplete implementation of policies might lead to several unintended consequences, which in turn may intensify the challenges a particular policy is supposed to address. The Matthew Effect is one such challenge which might get aggravated if ILAs are not implemented effectively. The Matthew Effect refers to the sit- uation where the rich become richer and the poor become poorer when social investments are made without equity considerations (Bonoli et al., 2017). Since the Nordic societies give high weightage to equity1 at least in their policy considerations compared to the rest of the world, equity is a significant parameter for policy implementation in the Nordic context. In other words, the relevance of the EU policy recommendations for the Nordic countries largely lies in their ability to challenge the Matthew Effect, i.e. the social phenomenon where- in the advantageous keep gaining more advantages and the disadvantaged keep becoming more disadvantaged. For instance, the differences between children who are poor readers and children who are excellent readers keeps increasing during their lifetime (Heckman, 2008). This article reflects on the possibility of implementing ILAs as a policy instrument to chal- lenge the Matthew Effect in Danish society by analysing their implementation in France, the context where they have been most successfully implemented. Furthermore, the French case is well-researched with data in alignment with the EU norms and thus can be easily compared with the Danish data. It focusses on giving low-qualified adults new possibilities which are individual-centric (rather than employability- or job-centric), and thus relevant for reflecting upon the Danish context. The research question for the paper is: 1 See Tuijnman & Hellström (2001). Nordic countries rank among the highest in terms of equality (meas- ured by the OECD using the Gini Coefficient). 65Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? Are Individual Learning Accounts (ILAs) a relevant policy instrument to challenge the Matthew Effect in the Danish context? The sources of research include policy documents, especially evaluation reports from the OECD, the EU, France, and Denmark. The article is thereby using the methods of case study, policy analysis (to focus on policies in the two cases and analyse the content of the policy documents based on how they describe and evaluate ILAs), and the comparative approach, taking the French case as the reference case and comparing the Danish case with it. The study covers the period from the 1970s, when the education and training reforms were introduced in France and Denmark after the post-WWII reconstruction period ended and a specific sector for the education of adults started taking shape in European countries with a view to achieve growth, prosperity and thereby development (Singh et al., 2022; Singh, 2024), and up to the year 2022, when a report was published rejecting the idea to implement ILAs in Denmark (Reformkommissionen, 2022). The point of departure for the paper is the European Commission’s recommendation from 2022 to implement ILAs in all EU member states based on the French experience (EC, 2022). The recommendation is a major shift from the EU’s previous standpoint. In its previous recommendations from 1995, the European Commission argued for invest- ment in human capital to ensure competitiveness, but in 2022 the EU recommendations argue for investments in social capital (i.e. social inclusion and equity) to ensure overall development (EC, 1995, 2022). Apart from other socio-economic factors, this recom- mendation was made especially in response to the continuously decreasing participation rate in adult learning amongst most EU member states during 2015–20 (see Figure 1) Figure 1 Adult participation in learning in Europe compared by country and over time Note. From Adult participation in learning, by country, 2015 and 2020 (% of population aged 25 to 64), by Eurostat (2020). 66 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 and the EU strategy for the European Education Area by 2025 which was overdue already (EC, 2020; Eurostat, 2020). The spatial context for the article is primarily Europe, even though the generalisations are relevant for comparison and reflection in other contexts. In 2016, about 15% of the participants in adult learning in the EU were low-qualified (Fernandes & Kerneïs, 2020). Even though the overall participation rate has been con- stantly increasing due to the competition among the providers of higher education, the numbers reflect the increasing participation of high-qualified adult learners while the participation of low-qualified learners remains almost the same. This implies that as the participation figures increase, the gap between the participation of the high- and the low-qualified increases. The same has been confirmed by statistics from Denmark high- lighting a declining number of low-qualified participants from 25% in 2009 to 14% in 2023 (Arbejderbevaegelsens Erhvervsraad, 2023). CONCEPTUAL FRAMEWORK The conceptual framework of the paper includes Winter’s (2012) Integrated Implementa- tion Model as it provides a comprehensive framework for understanding the implemen- tation process in a given context. Winter (2012) argues that the implementation process comprises of severalof such stakeholders and their behaviour in the Danish context re- garding a Matthew Effect among adult learners. Figure 2 Integrated Implementation Model by Winter Note. From “Implementation,” by S. Winter, in B. Peters & J. Pierre (Eds.), The Sage Handbook of Public Administration (p. 258), 2012, Sage. 67Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? Since the implementation guidelines are included in the policy formulation and design, at least two more conceptual formulations need to be considered: Green’s (2000) typology about the governance of lifelong learning systems and Vedung’s typology about types of policy instruments (Bemelmans-Videc et  al., 1998). Green (2000) categorises lifelong learning systems into three ideal types based on how they are governed: State-led, where the State is the primary decision maker; Partner-led, where the Social Partners and the State engage in tripartite negotiations for decision-making and the Social Partners are the most influential voice in decision-making; and Market-led, where policy decisions are primarily guided by Market forces of demand and supply. It is notable that in reality, life- long learning systems are governed by mixed forms or hybrids of these ideal types (Singh et al., 2022). In these hybrids, the proportion of influence of each ideal type may change when policies are revised. Thus, an ideal type may gain or lose influence as compared to other ideal types when policy revisions take place (Singh et al., 2022). Vedung classifies policy instruments into three categories: Carrots (benefits), Sticks (punishments) and Ser- mons (information) (Bemelmans-Videc et al., 1998). Both generalisations are used in the paper to analyse policy revisions and their implementation in Denmark and France. OPERATIONAL DEFINITION OF INDIVIDUAL LEARNING ACCOUNTS (ILAS) Because of the availability of several versions and descriptions of ILAs (Schuetze, 2007), the operational definition of ILAs for this article is: ILAs are personal accounts of individual adult learners wherein paid hours of work or money is provided by relevant authorities to cover their learning costs. The learning costs might be calculated in a narrow (including fee, travel costs, etc.) or broad (including Paid Education Leave, career guidance and validation, etc.) manner, de- pending upon how the decision makers are willing and able to determine the costs. While paid working hours2 were generally used in the past (termed the First Generation of ILAs in this paper), the transfer of money is the contemporary preferred approach (termed the Second Generation of the French ILAs in this paper) because the money provides more per- sonal autonomy to the learners for choosing what they consider as relevant. Furthermore, money includes the possibility of covering the broader learning costs and thus has more potential than paid working hours to increase participation in learning. Implementing ILAs is a complicated process and if designed appropriately, ILAs pro- mote inclusion, but the opposite might lead to exclusion as well (OECD, 2019c, 2019b). With a weak policy design, implementation might not give the desired results irrespective of low implementation f idelity (digressing from the planned implementation guidelines during the implementation process). This is so because the implementation process is guided by how policies are designed (Carroll et al., 2007; Winter, 2012). Special focus on 2 The 1974 convention about paid educational leave formulated by the International Labour Organization (1974) focusses upon paid hours for learning. 68 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 adequately informing (using Sermons) potential learners about the relevance of ILAs as Carrots is a prerequisite in the policy design for implementing ILAs effectively (OECD, 2019b). According to the OECD, ILAs are different from individual savings accounts and train- ing accounts (similar provisions) and only France has implemented them effectively (OECD, 2019c). While the individual savings accounts and training accounts are like perks connected to a specific employment and not portable (cannot be taken along while moving from one job to another), the ILAs are individual rights, available to all adult cit- izens, irrespective of their employment (OECD, 2019c). Moreover, ILAs reduce the role of collective actors like social partners and promote the effect of the market forces (com- petition among providers of individual learning opportunities; Perez & Vourc’h, 2020). This makes ILAs liberating instruments for adult learners, at least in principle, because to avail themselves of learning opportunities, the learners are not bound by their employ- ment status, restricted by the supply of offers and the conditionalities of social partners and providers and have the ability to choose what they want to learn like consumers, whom the providers try to entice. ILAs are therefore designed as Carrots (policy instru- ments) to ensure that working age learners have adequate resources and the freedom to buy the career guidance, validation, and education they consider to be relevant. Notably, career guidance and validation are included in the broader calculation of learning costs. The European Commission (2022) recommends enlisting all offers covered under ILAs in national registers for quality offers. Thus, while the learners are free in principle, they are restricted by national choices in the name of quality considerations in reality. REVIEW OF LITERATURE A vast body of research is available (especially from the 1980s) about the difference be- tween the learning behaviour of adults and children, and the way that the education of adults should therefore be designed. Since the 1970s, the literature reflects concepts like self-directedness (Knowles, 1980), self-determination (Deci et al., 2017), and per- sonalised learning (Wozniak, 2021) to highlight the development of an individual from a dependent to a self-directed learner. The same resonated in Nordic research as well (Tøsse, 2011; Wahlgren, 2010). Furthermore, the idea to develop education among equals ( Jarvis, 1985) through bottom-up approaches became wide-spread in the Nordic coun- tries (Tøsse, 2011; Wahlgren, 2010). Research done by Tough in the 1970s showed that more than 90% of adults in the US have their own personalised (could be informal and thereby not officially measured) learning projects irrespective of their employment status (Tough, 1979, as cited in Merriam et al., 2007). This is pertinent for understanding why adults might abstain from participation in organised (formal or non-formal) learning. In fact, adults have their own learning projects that do not necessarily count or are not meas- ured as organised and documented learning (Tough, 1971). The German adult education survey tries to capture these projects and map informal learning too, but it is not a widely accepted way of measuring learning activities yet (Singh et al., 2022). 69Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? Andragogical research emphasises that apart from employability and wage-related issues, adults (high- and low-qualified alike) may resist education from above, but they are inter- ested in learning if they perceive it to be relevant. This means that if the offers provide au- tonomy, enhance competence, and have relatedness, adults are motivated for participation. Thus, the challenge of non-participation might be rooted in the provisions rather than in the learners (Crowther, 2000; Deci & Ryan, 2017, as cited in Kalenda & Kočvarová, 2022; Jarvis, 1985; Tough, 1971, as cited in Merriam et al., 2007). One such challenge could be the lack of personalised learning because of which the potential learners might consider the offers as irrelevant (Wozniak, 2021). Personalised learning is apparently increasing in France due to new approaches such as Jobflix. This digital tool is designed to enable users to explore different career options and access personalised recommendations based on their skills and interests (Mrvcic & O’Doherty, 2022). A differential treatment of low-qualified and high-qualified adults as learners has charac- terised the EU policies (Brine, 2006). In the 1990s, the EU approach was influenced by the Vocational Training models applied in France and not the broader Adult Learning and Ed- ucation models which included non-vocational provisions (Field, 1997). A Norwegian case study showed how low-skilled adults were made to participate using Sermons (advice) and Sticks (punishment) (Dæhlen & Ure, 2009). A Czech study analysed the differences among Czech non-participants and recommended individualised offers (Kalenda & Kočvarová, 2021). It argued that giving autonomy and self-determination, building up the competencies required by the learners, and relatedness between learning outcomes and real-life needs could attract all learners irrespective of their qualification levels (Deci & Ryan, 2017, as cited in Kalenda & Kočvarová, 2022). The EU’s current approach recommends member states make social investments, that is, invest not only in human but also social capital (Hemerijck, 2018). Unlike in most parts of the globe, considerations about equity and thereby the low-qual- ified and the marginalised have predominated the Nordic policy discourse regarding the education of adults (Kosyakova & Bills, 2021; Rubenson, 2006; Tuijnman & Hellström, 2001). Between the 1970s and the 1990s, the policies in Denmark promoted the devel- opment of differentiated vocational and general (non-vocational) provisions under the adult education sector but post-2000s, this trend declined (Singh et al., 2022). The dif- ferentiation between vocational and non-vocational diminished gradually along with the relevance of the settings (formal, non-formal and informal) in which learning occurred (Singh et al., 2022; Singh, forthcoming). After 2000, the Danish policies were re-ori- ented from a focus on equity to a focus on achieving high productivity due to a highly competitive global economy (Ehlers, 2013; Singh et al., 2022). Consequently, the pub- lic spending on Adult Learning and Education and the participation rates of adults in learning became limited (Ehlers et al., 2011). By 2011, all Nordic countries revised their strategies, giving more voice to social partners as compared to the state or the market in policy design as well as implementation (Ehlers et al., 2011). Even though the Nordic countries report some of the highest participation rates for sever- al reasons (Desjardins, 2017), there is lack of research about whether such a participation 70 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 nowadays reflects high participation among the high-qualified adults only or among all adults irrespective of their qualification levels. Mapping the current target group behav- iour in relation to non-participation contributes towards filling this gap. Does the current status of participation indicate an increase or decrease in the Matthew Effect? THE DANISH CASE The Danish education system has been one of the high performing systems in relation to the participation of adults in learning (Andriescu et al., 2019; EC, 2019). It is flexible for adapting to changes and has been therefore restructured twice since 1960, in 2001 and 2018 (Singh et al., 2022). The Danish labour market is built upon consensus among the state and the social partners and about 68% of the labour force is organised (Munck, 2020). Since the mapping started, the participation rate in Denmark has remained above the EU average, but in recent years the participation rate has declined from 31.4% in 2009 to 25.3% in 2019 (EC, 2019a). The educational offers for adults are characterised by two types: vocational and general3 (EC, 2019a). These two types of offers have been independent of each other and spread across a large number of public and semi-public providers at least until the 1990s (EC, 2019a; Singh et al., 2022). The general education offers for adults existed since the mid-1800s mainly in the form of part-time evening classes (Ehlers, 2009). During the industrialisation period, most adults in the labour force only had 7 years of mandatory schooling, which was extended to 9 years in 1972 (Ehlers, 2009). The state, which funded all education, started introduc- ing non-formal vocational offers for the labour force gradually, including different target groups step by step (Pedersen et al., 2012). Until such offers became available, most adults relied on non-formal and informal (self-directed) learning (Ehlers, 2009). As a part of a labour reform in 1960, vocational training schools were established all over the country for unskilled workers of the private sector with a provision of paid education leave (paid working hours; Pedersen et  al., 2012). The state provided the resources until the 1990s (V. H. Pedersen, personal communication over email, 10th August 2022). The commission that drafted the labour market reform of 1960 included social partner representatives (an equal number of representatives from trade unions and employers’ associations) and provided influence to the social partners over the pro- visions and the content of offers (V. H. Pedersen, personal communication over email, 10th August 2022). The close cooperation between the state and the social partners led, for instance, to an agreement between the Ministry of Defence and the trade union for professional soldiers in the Danish Army. Haerens Konstabel- og Korporal Forening (hereafter HKKF) and 3 General offers included non-formal education in leisure time called folkeoplysning. For details about the several revisions of the 1990 law on Folkeoplysning, see Bache (2021). 71Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? the Ministry decided to offer professional soldiers paid educational leave to make sure that they could get a civil job after serving in the army.4 After a general revision in 1971, reskilling came back to the policy agenda for vocational education and training and the implementation of the revised law was once again shared between the state and the social partners. Access to public vocational schools (still under the Ministry of Labour) was extended to the skilled employees of the private sector in 1986 and to the skilled employees in the public sector in 1994 respectively (Pedersen et al., 2012). Unskilled workers employed in the private sector no longer remained the primary target group. The 10 Point Programme about the future education of adults, adopted by the Parlia- ment in 1984, maintained the difference between vocational (under the Ministry of La- bour) and the non-vocational (under the Ministry of Education) offers (Ehlers, 2009, 2013). However, the two types of offers began slowly to melt together due to reforms implemented in the 1990s (Ehlers, 2009, 2013). This process peaked in 2001 when the vocational offers were placed under the Ministry of Education (Ehlers, 2009, 2013). The Ministry of Labour was renamed as the Ministry of Employment (signifying precarious jobs) and the Ministry of Children and Education got the responsibility to monitor the initiatives for low-qualified adults (indicating that the education of low-qualified adults was to be treated in a manner similar to the education of children and youth Ehlers, 2009, 2013). The provisions regarding high-qualified adults were moved under the Ministry of Higher Education and Research (indicating that they were considered as the core of the knowledge economy; Ehlers, 2009, 2013). Since then, the provisions for the high-qual- ified have occupied a predominant position in the Danish policy on the education of adults (Ehlers, 2009, 2013). A large number of reforms that followed the suit thereafter were based upon the recom- mendations of committees comprising civil servants and experts. This meant a deviation from the labour market reform of 1960 where the social partners had direct influence in higher education. The state gradually withdrew from governance and took up monitoring after 2001. The non-formal folkeoplysning experienced funding cuts and was moved to the Ministry of Culture in 2011 (Bache, 2021). This administrative change signified that non-formal education had limited learning value compared to formal education. The policy choices after 2000 reduced the role of the state and gave influence to the so- cial partners owing to cost-sharing. The funding from the social partners was increased after tripartite negotiations. Resources were mobilised through collective competence funds administered by the social partners, while the market became the primary provider of offers for the education of adults (Rasmussen et al., 2019). It was evident how the social 4 For every month spent in the army, the soldiers were offered one week of paid education after the age of 35, just like ILAs (P. Soerensen, personal communications over email, 20th and 21st August 2022). The agreement existed from 1971 to 2013 and is in the process of being renewed due to the war in Ukraine (Holst Bach & Godtberg, 2023). 72 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 partners showed limited interest in non-vocational offers, resulting in an imbalance in the availability of offers on the market due to the imbalanced regulation of market offers (Singh et al., 2022). In 2016, a tripartite agreement was finalised between the state and the social partners to appoint an independent group of experts to review the existing provisions and mobilise resources for the education of adults. In 2017, apart from other recommendations, the expert group recommended introducing ILAs to increase participation (Ekspertgruppen for voksen-, efter og videreuddannelse, 2017). This was followed by a further tripartite agreement in 2018 that gave more influence to the social partners over all other stake- holders (Rasmussen et al., 2019). Based on Green’s typology, the Danish scenario can be depicted as shown in Figure 3: Figure 3 Hybrids of education of adults in Denmark Components of ideal types 1960 2001 2018 Predominant State-led Market-led Social Partner-led Medium influence Social Partner-led Social Partner-led State-led Least influence Market-led State-led Market-led Note. Created based on Green (2000) and the references used in the paper. After a change in government, a group of civil servants from the Ministry of Employ- ment, the Ministry of Education, the Ministry of Higher Education and Research, and the Ministry of Finance further analysed the possibility of introducing ILAs and con- cluded that the proposal had more negative than positive consequences (Beskæftigelse- sministeriet et  al., 2021). The proposal to implement ILAs was declared irrelevant by a commission of senior economists the following year, arguing that about 70% of the low-qualified adults do not want to participate and that most of the offers are utilised by the high-qualified adults (Reformkommissionen, 2022). The recommendations of the expert committee from 2017 were based on evidence from France and Singapore. The next section comprises of the pertinent aspects of the French case. THE FRENCH CASE An above-EU average (10.8%) and an increasing participation rate from 5.7% in 2009 to 19.5% in 2019 indicates that France has been successful according to the EU meas- urement standards (EC, 2019b). The system is characterised by Vocational Education & Training (and not Adult Learning & Education) provisions, influential social partners and very careful state governance. Since the labour market reform of 1971, the labour 73Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? force training in France has been mandatorily funded by the social partners and provided by the market. The implementation of the 1971 reform was both a success and a failure (Colardyn, 2004). The political aims were to correct, complement and compensate the human resources. The correction of the labour force worked 100%, the completion of the skills worked partly (mainly used by high-skilled adults), and the compensation (second chance education) did not work at all (Colardyn, 2004). Since industrial production was of the highest priority during the 1980s and 1990s, the participation of low-qualified workers remained poor despite several policy initiatives (Staugaard, 1997). It is difficult to make fundamental changes in the French system be- cause of the resistance from social partners (Munck, 2020). Only 8% of the French work- force is organised, strikes and other conflicts are frequent and the solutions are always decided by the state as the trade unions are funded either by the state or by the regions (Munck, 2020). The ILAs in France can be categorised into two generations because of the fundamental differences in their consequences despite continuity in their legal frameworks. In 2013, a Socialist government introduced reforms (Munck, 2020) which can be termed as the First Generation of ILAs. These ILAs counted paid hours for education in working time (Perez & Vourc’h, 2020). The funding of these hours was provided through the collective training funds pooled and managed by the social partners (Perez & Vourc’h, 2020). The vocational offers were realised through the middlemen hired and paid by the social partners (Perez & Vourc’h, 2020). In 2018, the Second Generation of ILAs, proposed by a Centrum government, replaced the hours with money. This revision faced much resistance from the social partners be- cause it gave the learners freedom to buy the offers of their choice from the market with- out the interference of the social partners in the form of career guidance and validation (Perez & Vourc`h, 2020). This revision which could be termed as the Second Generation of ILAs, removed the middlemen and thereby the control of social partners, while giving personal autonomy to individual learners. Furthermore, the low-qualified adult learners got access to more funds compared to the high-qualified adult learners (Cedefop, 2023; Perez & Vourc`h, 2020). For instance, the entitlement for low-qualified learners ranged be- tween 800€ to 8000€ per year, whereas the entitlement for high-qualified learners ranged between 500€ to 5000€ per year (Cedefop, 2023). Similarly an increase in vocational courses from 517,000 in 2019 to 984,000 was recorded in France in 2020 (Mrvcic & O’Doherty, 2022). Based on Green’s typology, the ILAs in France reflect the influence of stakeholders as shown in Figure 4: 74 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 Figure 4 The governance of ILAs in France Components of ideal types First generation (2013) Second generation (2018) Predominant State-led State-led Medium influence Social Partner-led Market-led Least influence Market-led Social Partner-led Note. Created based on Green (2000) and the references included in this article. DISCUSSION Since the 1960s, Denmark and France have applied different strategies. The Danish ap- proach was bottom-up and involved social partners (including NGOs). In the beginning, the state bore all costs but later, after tripartite negotiations, the costs were shared among the state and the social partners. The French approach was top-down, characterised by careful state governance. The funding came from social partners only. Paid Education Leave was introduced as an individual right in 1971 for every French citizen but inade- quately implemented (Staugaard, 1997). The Danish strategies favoured the low-qualified workers before Denmark started build- ing up a knowledge economy. Since the introduction of the Further Education System for Adults in 2001 owing to the needs of the knowledge economy, the policies changed in favour of the high-qualified. While France followed the EU policy of investment in human capital (economically-ori- ented) in the beginning, recently followed by both human and social capital (social invest- ments), Denmark has been doing the opposite. It continued to make social investments until 2000 but has switched to investments in human capital ever since. The French experience with the First Generation of ILAs shows that the predominance of social partners is a barrier in relation to implementation (Perez & Vourc’h, 2020) be- cause they may not prioritise the participation of low-qualified adults over the industri- al production (economic growth and profit). Further, their cost and benefits analysis in relation to the society may be different as compared to that of the state which aims at equity and social welfare, at least in the Nordic countries including Denmark. Thus, for a successful implementation of ILAs strategies and challenging the Matthew Effect, state governance (rather than monitoring) appears to be an effective option. The Second Generation of the French ILAs have been successful in two ways: 1) dealing with the current needs of the labour market and the economy for upskilling and reskilling the workforce, and 2) providing extra benefits to low-qualified adults, implying affirm- ative action or social investment. The policy change meant extra Carrots (in the form of money) provided to personal accounts owned by the low-qualified employees combined 75Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? with Sermons (career guidance and validation), chosen and paid for by the low-qualified employees themselves. The difference between the 2013 generation and the 2018 genera- tion of ILAs is profound. The Second Generation of ILAs carries more potential for con- structive change because it addresses the Matthew Effect by giving personal autonomy to individuals. Thus, the Second Generation of the French ILAs is more future-oriented, aimed at a fluid, flexible and suitable work force to address the needs of a labour market characterised by precarious jobs. The OECD and the EU have appreciated the French 2018 model reforms for the same reasons (EC, 2022; OECD, 2019). However, it shall be noted that the mentioned French model is not necessarily devoid of the Matthew Effect (Mrvcic & O’Doherty, 2022). In Denmark, the recommendations regarding whether ILAs should be implemented or not have changed according to the composition of authors. While the independent ex- perts authoring the 2017 report have favoured ILAs, the civil servants authoring the 2021 report and the senior economists authoring the 2022 report have rejected them, arguing that low-qualified adults are not interested in new learning opportunities after the age of 40 and that ILAs have more negative than positive consequences. The two latter reports which reject the implementation of ILAs in Denmark have not only ignored the findings of andragogical research existing for several decades but also the empirical evidence from France. Further, the two reports do not take into account the gradual shift of focus in education all over Europe from teacher-centric (education from above) to learner-centric (bottom-up) modes. In fact, the reports appear to adhere to a top-down approach, discrim- inating between the high and low-qualified adults, in alignment with the Danish system which has kept the low-qualified adults at the receiving edge over the last two decades. Despite the fact that low-qualified adults are the ones who lag behind in participation, the two latter reports label them as uninterested in learning anything and exclude this target group from further political debate. Thus, these reports do not address the core of the challenge for the decreasing participation in Denmark. The different organisation of provisions for the high- and the low-qualified might be justified for several reasons but it certainly increases the Matthew Effect, thus giving more personalised offers to the high-qualified and pushing the low-qualified to further disadvantage. The whole idea seems against the ideals of equity for which the Nordic countries are often lauded. Even though the overall Danish participation rates are higher than the French, the Danish rates are decreasing while the French rates are increasing. The Danish offers are personalised for the high-qualified and may therefore be perceived as relevant for the high-qualified adults, but they are not necessarily personalised for the low-qualified adults and thereby not necessarily perceived as relevant for the low-qualified. This appears to be one of the reasons for the low participation rates among the low-qualified adults in Denmark. After the optimum engagement of the high-qualified, the participation rates among the high-qualified may not increase. Thus, personalised offers for the low-qual- ified seem to offer the possibility of positive change. ILAs have the potential to provide liberty to low-qualified individuals for choosing the career guidance, the validation, and 76 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 the kind of learning they perceive as relevant and thereby increase their participation. For low-qualified adults, the implementation of ILAs in Denmark would mean a shift from supply (collective offers) to demand (individual choices). It can be argued that the learning outcomes of such personalisation might be difficult to measure but adaptive tests (personalised testing based on the knowledge of the learner) offer a relevant solution. A relevant model for the implementation of ILAs is therefore learner-centric. Building upon the model of Winter (2012), the following model (Figure 5) with learners as the core of the implementation process in education policies is proposed. Figure 5 Learner-centric Integrated Implementation Model for the education of adults Note. From “Challenging the Matthew Effect through Individual Learning Accounts. A comparative study from Den- mark and France” by S. Singh and S. Ehlers, paper presented at the 9th Nordic Conference in Adult Education and Learning on 18th May 2022, Tønsberg, Norway; adapted (modified) from Winter (2012). In Figure 5, the target-group behaviour is kept at the core of the implementation process to ensure that learning offers remain relevant for them. Similarly, Green’s typology needs to be adapted to include the individual as a relevant player when the education of adults is discussed. The French experience shows that the state, the social partners, and the individuals may share influence while the regulated mar- ket is available to fill in the gaps wherever possible. It must be noted that the market in EU member states is regulated in different ways. 77Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? CONCLUSION Despite a high participation rate and the high performance delivered by the Danish Further Education System for Adults (according to the EU measurement standards), Denmark is unable to challenge the Matthew Effect. A major cause of this is the post-2000 policy choice to aim at production rather than equity, withdraw state control and extend more influence to the social partners who favour production (economic growth) over social in- vestment. With the existing provisions, the participation of low-qualified adults is lagging in the process of overall development. This can have severe consequences in the future for equity in Danish society by increasing the Matthew Effect in the sense that despite access, participation levels are low. The decreasing participation rate indicates the need to revise the Danish approach. As indicated from the French experience, ILAs have the potential to challenge the Matthew Effect in Denmark and translate access into participation. REFERENCES Andriescu, M., Broek, S., Dohmen, D., Mulyneux, J., & Ravenhall, M. (2019). Adult learning policy and provision in the Member States of the EU: A synthesis of reports by country experts. Publications Office of the European Union. https://data.europa.eu/doi/10.2767/82188 Arbejderbevaegelsens Erhvervsraad. (2023). Det gaar den forkerte vej med efteruddannelsen i Danmark. Arbejderbevaegelsens Erhvervsraad. https://www.ae.dk/analyse/2023-01-det-gaar-den-forkerte- vej-med-efteruddannelsen-i-danmark Bache, P. (2021). Finkultur og mangfoldighed: 60 års dansk kulturpolitik. Gads Forlag. Bemelmans-Videc, M., Rist, R., & Vedung, E. (Eds.). (1998). Carrots, sticks & sermons: Policy instruments & their evaluation (1st ed.). Routledge. https://doi.org/10.4324/9781315081748 Beskæftigelsesministeriet, Børne og Undervisningsministeriet, & Uddannelses-og Forskningsministeri- et samt Finansministeriet. (2021). Rapport om sporskifte or uddannelse i hele arbejdslivet. https://www. star.dk/media/18212/rapport-om-sporskifte-og-uddannelse-i-hele-arbejdslivet.pdf Bonoli, G., Cantillion, B., & Lancker, W. (2017). Social investment and the Matthew effect: Limits to a strategy. In A. Hemerijck (Ed.), Uses of Social Investments (pp. 66–76). Oxford University Press. Brine, J. (2006). Lifelong learning and the knowledge economy: Those that know and those that do not – the discourse of the European Union. British Educational Journal, 32(5), 649–665. https://doi. org/10.1080/01411920600895676 Carroll, C., Patterson, M., Wood, S., Booth, A., Rick, J., & Balain, S. (2007). A conceptual framework for implementation fidelity. Implementation Science, 2(40), 1–45. https://doi.org/10.1186/1748-5908-2-40 Cedefop. (2023). Financing adult learning database: France. https://www.cedefop.europa.eu/en/tools/fi- nancing-adult-learning-db/overview/france Colardyn, D. (2004). Lifelong learning policies in France. International Journal of Lifelong Education, 23(6), 545–558. https://doi.org/10.1080/026037042000311460 Crowther, J. (2000). Participation in adult and community education – a discourse of dimin- ishing returns. International Journal of Lifelong Education, 19(6), 479–492. https://doi. org/10.1080/02601370050209023 Dæhlen, M., & Ure, O. (2009). Low-skilled adults in formal continuing education: Does their partici- pation differ from other learners? International Journal of Lifelong Learning, 28(5), 661–674. https:// doi.org/10.1080/02601370903189948 78 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 Deci, E., Olafsen, A., & Ryan, R. (2017). Self-determination theory in work organizations: The state of a science. Annual Review of Organizational Psychology and Organizational Behavior, 4, 19–43. https:// doi.org/10.1146/annurev-orgpsych-032516-113108 Desjardins, R. (2017). Political economy of adult learning systems: Comparative study of strategies, policies and constraints. Bloomsbury. Ehlers, S. (2009). Livslang læring som politisk strategi i 1900-tallets Danmark: Samspillet mellem civil- samfund, stat og marked. Uddannnelseshistorie, 44, 28–54. https://uddannelseshistorie.dk/wp-con- tent/uploads/2020/08/a2009-sren-ehlers.pdf Ehlers, S. (2013). Globalisering, videnskonomi og konkurrenceevne: Om restruktureringen af det dan- ske uddannesessystem. Uddanneseshistorie, 47, 21–35. https://uddannelseshistorie.dk/wp-content/ uploads/2020/08/a-2013soeren-ehlers.pdf Ehlers, S., Warvik, G., & Larson, A. (2011). Effektive strategier for livslang læring i de nordiske lande. Nordisk Ministerråd. https://nck.au.dk/fileadmin/nck/BNP/Effektive_strategier_for_livslang_laer- ing_i_de_nordiske_lande.pdf Ekspertgruppen for voksen-, efter- og videreuddannelse. (2017). Nye kompetencer hele livet: Frem- tidens voksen-, efter- og videreuddannelse. https://www.regeringen.dk/media/3545/nye-kompetenc- er-hele-livet_web.pdf European Commission. (1995). Teaching and learning: Towards the learning society [White paper on ed- ucation and training]. Office for the Official Publications of the European Communities. https:// files.eric.ed.gov/fulltext/ED461011.pdf European Commission. (2019a). Education and Training monitor 2019: Denmark. https://education. ec.europa.eu/sites/default/files/document-library-docs/et-monitor-report-2019-denmark_en.pdf European Commission. (2019b). Education and Training monitor 2019: France. https://education.ec.eu- ropa.eu/sites/default/files/document-library-docs/et-monitor-report-2019-france_en.pdf European Commission. (2019c, December 1). Mission letter to the Commissioner for Jobs and Social Rights issued by the President for the European Commission. https://commissioners.ec.europa.eu/system/ files/2022-11/mission-letter-nicolas-schmit_en.pdf European Commission. (2020, September 30). Communication from the Commission to the European Par- liament, the Council, the European Economic and Social Committee and the Committee of the Regions on achieving the European Education Area by 2025. https://education.ec.europa.eu/sites/default/files/ document-library-docs/communication-european-education-area.pdf European Commission. (2022). Council recommendation on individual learning accounts. https://data.con- silium.europa.eu/doc/document/ST-8944-2022-INIT/en/pdf Eurostat. (2020). Adult participation in learning, by country, 2015 and 2020 (% of population aged 25 to 64). https://ec.europa.eu/eurostat/statistics-explained/index.php?title=File:Adult_participation_in_ learning,_by_country,_2015_and_2020_(%25_of_population_aged_25_to_64).png&oldid=529938 Fernandes, S., & Kerneïs, K. (2020). Towards an individual right to adult learning for all Europeans [Report]. Foundation for European Progressive Studies. https://feps-europe.eu/wp-content/up- loads/downloads/publications/lifelong%20learning%20-%20a4%20-%20def%201%20-%2088%20 pages%20%204%20cover%20pp.pdf Fernandes, S., & Kerneïs, K. (2021). Towards a European individual learning account [Policy Brief ]. Foun- dation for European Progressive Studies. https://feps-europe.eu/wp-content/uploads/downloads/ publications/pb_210503_ila_fernandeskerneis_en.pdf Field, J. (1997). The learning society and the European Union: A critical assessment of supranational education policy formation. Journal of Studies in International Education, 1(2), 73–92. https://doi.org /10.1177%2F102831539700100205 79Shalini Singh, Søren Ehlers: Challenging the Matthew Effect Through Individual Learning Accounts? Green, A. (2000). Lifelong learning and the learning society: Different European models of organiza- tion. In A. Hodgson (Ed.), Policies, politics and the future of lifelong learning (pp. 35–48). Kogan Page Limited. Heckman, J. (2008). Schools, skills, and synapses. Economic Inquiry, 46(3), 289–324. https://jenni.uchi- cago.edu/papers/Heckman_2008_EI_v46_n3.pdf Hemerijck, A. (2018). Social investments as a policy paradigm. Journal of European Public Policy, 25(6), 810–827. https://doi.org/10.1080/13501763.2017.1401111 Holst Bach, A., & Godtberg, B. (2023, December 10). Uddannelse kan vaere noeglen til at standse me- darbejderflugten i Forsvaret, mener fagforening. DR Nyheder. https://www.dr.dk/nyheder/indland/ uddannelse-kan-vaere-noeglen-til-standse-medarbejderflugten-i-forsvaret-mener International Labour Organization. (1974). C140 – Paid Educational Leave Convention, 1974. http://ilo. org/dyn/normlex/en/f ?p=NORMLEXPUB:12100:0::NO::P12100_INSTRUMENT_ID:312285 Jarvis, P. (1985). The sociology of adult and continuing education. Croom Helm. Kalenda, J., & Kočvarová, I. (2022). “Why don’t they participate?” Reasons for nonparticipation in adult learning and education from the viewpoint of self-determination theory. European Journal for Re- search on the Education and Learning of Adults, 13(2), 119–123. https://doi.org/10.3384/rela.2000- 7426.3535 Knowles, M. (1980). The modern practice of adult education: From pedagogy to andragogy. Cambridge Adult Education. Kosyakova, Y., & Bills, D. (2021). Formal adult education and socioeconomic inequality: Second chance or Matthew Effects. Sociology Compass, 15(9), e12920. https://doi.org/10.1111/soc4.12920 Merriam, S., Caffarella, R., & Baumgartner, L. (2007). Learning in adulthood: A comprehensive guide (3rd ed.). Jossey-Bass. Mrvcic, V., & O’Doherty, A. (2022). Digital lifelong learning in France: Lessons from Jobflix. https://www. thersa.org/globalassets/_foundation/new-site-blocks-and-images/reports/2022/12/digital-life- long-learning-in-france-report-final.pdf Munck, A. (2020). Den rystede republik: Hvad i alverden er der galt med Frankrig? Forlaget 28B. Organisation for Economic Co-operation and Development. (1973). Recurrent education: A strategy for lifelong learning. Center for Educational Research and Innovation, OECD. https://files.eric.ed.gov/ fulltext/ED083365.pdf Organisation for Economic Co-operation and Development. (1996). Lifelong learning for all: Meeting of the Education Committee at Ministerial Level, 16–17 January 1996. https://www.voced.edu.au/con- tent/ngv%3A25305 Organisation for Economic Co-operation and Development. (2001). Economics and f inance of lifelong learning. OECD. https://doi.org/10.1787/9789264195868-en Organisation for Economic Co-operation and Development. (2019a). Getting skills right – Engaging low-skilled adults in learning. https://www.oecd.org/els/emp/engaging-low-skilled-adults-2019. pdf Organisation for Economic Co-operation and Development. (2019b). Individual learning accounts: De- sign is the key to success [Policy Brief on the Future of Work]. https://www.oecd.org/els/emp/individ- ual-learning-accounts.pdf Organisation for Economic Co-operation and Development. (2019c). Individual learning accounts: Pan- acea or Pandora’s box? https://www.oecd-ilibrary.org/sites/203b21a8-en/index.html?itemId=/con- tent/publication/203b21a8-en Pedersen, V., Andersen, S., & Lassen, M. (2012). En Fortaelling om AMU: Arbejdsmarkedsuddannelserne I en Verden af Forandring. Aalborg Universitetsforlag. 80 ANDRAGOŠKA SPOZNANJA/STUDIES IN ADULT EDUCATION AND LEARNING 2/2024 Perez, C., & Vourc’h, A. (2020). Individualizing training access schemes: France – the Comte Personnel de Formation (Personal Training Account – CPF) [OECD Social, Employment and Migration Working Papers No. 245]. https://dx.doi.org/10.1787/301041f1-en Rasmussen, P., Cort, P., & Larson, A. (2019). The vocational turn of adult education in Denmark – an analysis of adult education policy from the late 1900s. International Journal of Lifelong Education, 38(3), 254–267. https://doi.org/10.1080/02601370.2019.1586778 Reformkommissionen. (2022). Nye Reformveje 1. https://reformkommissionen.dk/media/25523/nye-re- formveje-1_web.pdf Rubenson, K. (2006). The Nordic model of lifelong learning. Compare, 36(3), 327–341. http://dx.doi. org/10.1080/03057920600872472 Schuetze, H. (2007). Individual learning accounts and other models of financing lifelong learning. In- ternational Journal of Lifelong Learning, 26(1), 5–23. https://doi.org/10.1080/02601370601151349 Singh, S. (2024). Beyond education: Mapping policy changes from the sectors of adult education to sys- tems of lifelong learning (1972–2022). Andragoška spoznanja/Studies in Adult Education and Learn- ing, 30(2). https://doi.org/10.4312/as/16287 Singh, S., & Ehlers, S. (2019). Recognition of prior learning: Policy analysis from Denmark and In- dia. Andragoška spoznanja/Studies in Adult Education and Learning, 25(1), 69–87. https://doi. org/10.4312/as.25.1.69-87 Singh, S., & Ehlers, S. (2022, May 18). Challenging the Matthew effect through individual learning ac- counts: A comparative study from Denmark and France [Paper presentation]. 9th Nordic Conference on Adult Education and Learning, Tønsberg, Norway. Singh, S., Schmidt-Lauff, S., & Ehlers, S. (2022). Transitions from adult education sectors to life- long learning systems: A Comparative Study from Denmark, Switzerland, Germany and the Eu- ropean Education Area. In. A. Grotlüschen, B. Käpplinger, G. Molzberger,  & S. Schmidt-Lauff (Eds.), Erwachsenenbildung in internationalen Perspektiven: Grenzen und Chancen (pp. 97–112). Ver- lag Barbara Budrich. https://doi.org/10.2307/j.ctv2whxcn7.10 Staugaard, H. (1997). Hovedtendenser inden for voksenuddannelserne I Frankrig. Aalborg Universitet. Tøsse, S. (2011). Historie, praksis, teori og politikk: Om kunnskapsgrunnlaget for voksnes læring. Tapier Akademisk Forlag. http://hdl.voced.edu.au/10707/114597 Tough, A. (1971). The adult’s learning projects: A fresh approach to theory and practice in adult learning. OISE. Tuijnman, A., & Hellström, Z. (2001). Curious minds: Nordic adult education compared. Nordic Council of Ministers. Wahlgren, B. (2010). Voksnes læreprocesser: Kompetenceudvikling i uddannelse of arbejde. Akademisk Forlag. Winter, S. (2012). Implementation. In B. G. Peters & J. Pierre (Eds.), The Sage Handbook of Public Ad- ministration (pp. 255–278, 2nd ed.). Sage. Wozniak, K. (2021). Personalized learning for adults: An emerging andragogy. In S. Yu, M. Ally, & A. Tsinakos (Eds.), Emerging Technologies and Pedagogies in the Curriculum: Bridging Human and Machine: Future Education with Intelligence (pp. 185–198). Springer. https://doi.org/10.1007/978- 981-15-0618-5_11