EIoP, Special Issue 1, Vol. 13 (2009), Art. 7

European Integration online Papers / ISSN 1027-5193
EIoP Logo    Special Issue 1, Vol. 13 (2009), Art. 7

How to cite?
Hartlapp, Miriam. 2009. Learning about policy learning. Reflections on the European Employment Strategy. In: Kröger, Sandra (ed.): What we have learnt: Advances, pitfalls and remaining questions in OMC research, European Integration online Papers (EIoP), Special Issue 1, Vol. 13, Art. 7, http://eiop.or.at/eiop/texte/2009-007a.htm.

DOI: 10.1695/2009007

Learning about policy learning.
Reflections on the European Employment Strategy*

Miriam Hartlapp
Social Science Research Center Berlin (WZB), Germany
E-Mail: hartlapp@wzb.eu

Abstract: The concept of policy learning has been attracting increasing attention in the political science literature, notably in the many publications on the Open Method of Coordination (OMC). However, much research faces fundamental methodological problems regarding the mode of functioning and extent of policy learning. This paper shows that these problems can be avoided if we take a closer look at the potential of the OMC instruments to stimulate learning. The paper differentiates between three types of learning: learning by own experience, learning from others and learning with others. Systematic links are established between them and the learning conditions created by the European Employment Strategy (EES) tools casting doubt on the steering capacity of the EES through learning. Moreover, there is a fundamental contradiction between actor orientation in deliberative policy learning and competitive policy bargaining which questions the learning potential of the EES.

Keywords: open coordination; employment policy; policy-learning; benchmarking; participation; social policy; European social model; governance; political science

Table of Contents:

1. Introduction to contents list

In recent years, both scholars and practitioners have shown increasing interest in policy learning and related concepts like policy transfer and policy diffusion. This can be attributed not least to European integration, which has seen the establishment of the Open Method of Coordination (OMC), a mode of governance based on the assumption of mutual policy learning. There are hopes that this instrument can stimulate policy learning, bringing positive innovation and expansion compared with the classical forms of regulation with all their limitations. Against this backdrop, I examine the OMC with an eye to what characteristics of the instruments it deploys can in fact stimulate and guide policy learning. I am thus concerned with the effectiveness of OMC in promoting policy learning, not with its factual effect in the member states.

The OMC finds application in a number of policy areas. Despite differences in process and modalities, OMCs in all policy areas share a conviction that member states should learn from one another through iterative reporting, benchmarking, and peer review in the pursuit of common policy goals. The toolbox provided by the European Employment Strategy (EES) is older and better developed than OMC instruments in other policy areas; it is therefore particularly useful to take a closer look at it (also Niechoj, Radulova, López-Santana and Weishaupt this issue). It is likely that the design and processual use of the instruments stimulate various forms of learning (vs. Pfister this issue). But they possibly also stimulate other transfer mechanisms. This is of practical and theoretical importance. On the one hand, differentiation into various forms of learning allows systematic comparison of the learning potential offered by EES instruments. On the other hand, it is important to investigate interaction between instruments and their compatibility. Corresponding instruments may reinforce each other and lead to more comprehensive learning effects. At the same time, even if certain instruments have the potential to stimulate learning, if they are used at cross purposes to other tools this could reduce or neutralise the learning potential of EES as a whole.

The aim of the article is to contribute to understanding learning as a causal mechanism and thus to a better appreciation of the learning potential of the OMC.

  • I begin by outlining different learning concepts discussed in the research literature (section 2).
  • I continue with a critical summary of research findings on policy learning in the context of the OMC and discuss the methodological problems involved (section 3).
  • Finally, conducting my own analysis of learning in the EES framework that departs from the instruments, I show that their steering potential consists only to a limited extent in stimulating or initiating policy learning. Moreover, the fundamental idea of the steering function of deliberative learning processes runs counter to increasing differentiation in the form of more concrete targets and the associated policy conflicts (section 4).
  • I finish by summarizing findings and discussing possible recommendations on expanding the potential for learning by modifying existing instruments (section 5).

2. Different forms of learningto contents list

This section provides a basis for empirical analysis by considering forms of learning and the underlying conditions for learning. With reference to the much quoted definition by Heclo (1974), I understand learning for my present purposes to be an alteration in the behaviour or preference of a rational actor (in our case a government) in reaction to a modified information pool (in this case the EES). The information pool is a necessary but not sufficient condition for change through learning. By making a selection of experience and data out of the information pool, by filtering and interpreting, the actor generates new knowledge. This knowledge forms the basis for her decisions.

As far as the depth of learning processes is concerned, the most important distinction is whether actors’ strategies for action and steering instruments alter because the information pool has changed with respect to the possibilities and contexts of problem solving, or whether actors also learn with regard to their preferences. If actors seek effectively to attain a constant goal the process is referred to as “single loop learning” (Argyris and Schön 1996: 20) or “simple learning” (Deutsch 1969: 146-147); adaptation of existing instruments is called “first order change”, and the introduction of an instrument designed to attain an existing goal “second order change” (Hall 1993: 279). In the case of “complex learning” (Deutsch 1969: 147), actors also consider whether the values on which their action is based are ‘wrong’ and change their preferences on the basis of new cost-benefit considerations (updating beliefs). Here Argyris und Schön (1996: 20-22) refer to “double-loop learning”; Hall (1993: 279) to “third order change” in policy goals or the policy paradigm.

A change in the information pool is necessary for both simple and complex policy learning. It can be generated in various contexts (see Visser and Hemerijck 2001) (1): First, the most everyday form of learning is expanding the information pool through one’s own experience. This form of learning can be fostered by focused inducement to learn in the sense of reflexive evaluation or support in the search for information on the context of problems. But decisions need not be based on an actor’s own experience. Second, if governments use information on the effects of policy decisions and on the policy instruments of other countries without experiencing them themselves, they can learn from others (“lesson-drawing”, Rose 1991). On the assumption that rational actors seek solutions to problems, this allows transaction costs to be reduced (“learning ahead of failure”, Visser and Hemerijck 2001). Who learns from whom depends on a number of factors such as the pull of successful or the push of powerful countries, geographical distance, similar problems or functional interdependencies (Trubek and Trubek 2005: 93; Rose 1991: 14-17). Third, deliberative, experimental processes seek to generate new, innovative and better solutions to problems on the basis of voluntary exchange of information and positions in decisional situations. The approach proceeds on the assumption that frequent interaction can alter the preferences of member states and thus enable joint decisions to be taken (2). The concept of deliberative learning was developed with the aim of preventing bottle-necks in production and optimising processes at the plant level. By exchanging information, actors are able to check and adjust their own decisions with regard to anticipated changes in the behaviour of other actors (Sabel 1994: 158). This school of thought argues that these considerations can be transferred from business practice to governmental action (“democratic experimentalism”, e.g. Dorf and Sabel 1998). Consulting, experimentation, the exchange and pooling of information enlarge the scope for action and enable policy learning in the sense of pursuing common goals.

Steering instruments that improve the general conditions for generating and exchanging information between governments can foster learning processes (e.g. Sabatier 1986: 323). In contrast, no learning from others is to be expected if actors are called upon to defend (competing) or not easily incompatible (national) interests. If the institutional context increases competition between actors, confrontation or bargaining are more likely than joint learning or problem solving. Scharpf (1997: 124) describes the difference between the two forms of interaction as follows:

“The successful joint search for better overall solutions requires creativity, effective communication, and mutual trust, whereas success in the distributive battle depends on the strategic and even opportunistic, communication and withholding of available information […] the ‘cooperative’ interaction orientation that is conducive to joint learning in the production dimension is psychologically incompatible with the “competitive” orientation that facilitates success in distribution […]”.

In principle competitive settings still allow for “reactive learning [where] governments learn individually by responding to the challenges of the competition” (Benz 2007: 512). Yet this form of positive competition in the market of political ideas is exigent institutionally and “there is only limited evidence that, in practice, policy competition does play a significant role” (Benz 2007: 513). The following analysis of instruments and their operational mechanisms imply that the learning potential of the EES is rather determined by the dichotomy of competitive bargaining versus deliberative exchange and learning.

3. EES performance monitoring and related conceptual problemsto contents list

Policy learning is not an end in itself, it serves to implement political programmes or in general terms to attain goals. EES performance review is therefore an important element in examining policy learning. However, many of the studies on the subject suffer from major conceptual problems, which make conclusions about policy learning in the EES context more difficult (see Kröger this issue). Strictly speaking, a critical consideration of the research literature on the OMC reveals no more than that changes have been taking place in member states. And we know that there is a mode of governance at the European level that allegedly prompts or facilitates learning processes. To associate these two observations under the heading “learning process” is unsatisfying for three reasons.

1. The problem of lacking empirical evidence: Many (particularly early) studies turn to official documents in evaluating learning in the EES context. They take no account of the fact that these documents are not neutral assessments but reports prepared or at least approved by the national governments in question (e.g. Biagi 2000: 156; Mosher and Trubek 2003: 75). No check is made on whether a selective presentation of national policies (deliberately) overestimates or underestimates the influence of the OMC. Nor does this approach ensure comparability between countries. More recent studies comparing a larger number of member states can better cope with this problem (Zeitlin et al. 2005: 27-28). But even a great deal of empirical effort cannot eliminate difficulties in adequately measuring learning processes.

2. The problem of alternative explanatory variables: New policies can be introduced without the EES playing a role (Scharpf 2002: 654). Over and beyond the empirical questions on how various explanatory factors are to be weighted, here my criticism is levelled at the analytical negligence of alternative explanatory factors in identifying learning processes.

Figure 1

Very few studies conceptualise external factors (e.g., international financial crises) and internal factors (e.g., demographic change) as counter-factual explanations. Policy learning in the EES cannot be distinguished from other causes of policy change. Where different countries find similar solutions to common external or internal problems, this can result in policy change or even in convergence. Moreover, there are international organisations other than the EU with an interest in steering employment policy, such as the ILO and OECD (Schäfer 2005), equally causing (potential) policy change.

3. The problem of functionally equivalent transfer mechanisms: The third point of criticism is the underspecification of the mechanism underlying the change to be observed. In most studies on the OMC, the relationship between EES and change in national policies resembles a black box.

Figure 2

Many studies on the OMC assume implicitly or explicitly that the transfer mechanism in the ‘black box’ is learning (see López-Santana this issue, Vanhercke this issue). “Its [OMC] real effectiveness and influence nationally is part of a collective learning process” (De La Porte and Pochet 2002a: 13). Or “it organises a learning process in order to promote the exchange of experience and best practice” (Regent 2003: 191). Although assessment of learning processes in the OMC has become more sceptical over time, there is still a lack of systematic in-depth analysis showing that observable policy changes are due to learning by actors or the overall political system.

What is more, other transfer mechanisms can be posited for the EES. One important such mechanism is pressure. The OMC is described as a soft steering instrument because actors have no possibility for imposing direct sanctions. Ashiagbor (2004: 314 and 327) claims that learning processes can be initiated by exerting pressure (naming and shaming). However, I argue that changes in national employment policies driven by this logic must be fundamentally distinguished from learning, since any change is not brought about by alteration of the information pool.

Figure 3

Change in employment policy through national government political strategies (typically motivated by economic or political interests) also differs from the transfer of European policies through learning. Soft European tools like the OMC may also be used by national actors for their “leverage effect” (Erhel et al. 2005) or as “selective amplifiers” (Visser 2005) for advancing specifically national interests. In such cases the OMC is a means to an end but not the factor that activates learning processes (see Weishaupt this issue).

A careful research design is needed to obtain an empirically clear distinction between political calculation, coercion, learning, and other transfer mechanisms such as imitation or copying. Of central importance is the precise analysis of government motives for action and the question whether changes in behaviour or preferences are subsequent to a change in the information pool. It comes as no surprise that the country chapters in Zeitlin et al.(2005) reach diverging conclusions on different transfer mechanisms. Since the individual articles do not systematically examine potential transfer mechanisms, it remains unclear whether this divergence points to country-specific differences or whether the applied research perspective is to blame.

Finally, another analytic-conceptual difficulty arises in the European multi-level system. The EES often reflects national policies, for what is taken to be Brussels policy is the result of negotiations between members states and European institutions. This is the case for the creation of the EES (Schäfer 2004) as well as for the elaboration of goals, guidelines, and indicators (De La Porte and Pochet 2002b; Jobelius 2003). The distinction between what is to be learned and what has already been learned tends to blur.

This paper is also unable to resolve many of the conceptual problems mentioned. But it points to a possibility for addressing the issue of policy learning in the EES in an alternative manner by examining the potential of steering instruments for stimulating learning.

4. The potential of EES instruments for stimulating learning to contents list

The EES consists of a number of instruments and processes that are repeated in regular cycles. The Council adopts Employment Guidelines and country-specific Recommendations (since 2005 part of the Integrated Guidelines for Growth and Jobs). On this basis member states draw up National Reform Programmes (NRPs; prior to 2005 National Action Plans – NAPs) which are assessed by the European Commission and the Council through the Joint Employment Report with a view to setting the next annual guidelines (3). Apart from the NRPs, additional indicators help in collecting information and comparing countries. Then the process starts over again. In parallel, peer reviews seek to identify exemplary procedures and to encourage their transfer to other countries through meetings and exchanges. The institutional design and processual deployment of these EES instruments determine what forms of learning are probable, or what other transfer mechanisms come into play. Does the given instrument satisfy the structural and procedural preconditions for stimulating policy learning to achieve the practical realisation of EES employment policy goals? If the nature of the instrument presents a major obstacle to learning, policy learning is unlikely to take place through the EES (4).

The following table assigns EES tools to one of the three analytically distinct forms of learning – for reasons of clarity, it shows the dominating form; further combinations and complementarities are discussed in the paper.

Table 1

It should be emphasised that the table makes no statement about actual learning processes or the relevance of learning as compared with other transfer mechanisms. After all, even in the best of cases, learning is not a sufficient condition for successful steering by the EES.

4.1. Guidelines and recommendations to contents list

Guidelines are proposed by the European Commission and adopted by the Council acting by a qualified majority. They set common priorities for European employment policy. Since 2005, they have applied for a period of three years (previously one year) and are integrated into European economic policy guidelines. The eight guidelines are subsumed under three priorities, which are concerned with both policies and outcomes:

  1. attract and retain more people in employment, increase labour supply and modernise social protection systems,
  2. improve adaptability of workers and enterprises and
  3. increase investment in human capital through better education and skills.

The guidelines are supplemented by country-specific recommendations (in 2008 between 5 and 6 per country).

The relatively great potential of guidelines and recommendations for promoting learning consists in encouraging the interlinkage of known problems, generating information, and reassessing existing policies. Policy learning would then take place through inducement to expand one’s own experience. Limits are that country-specific recommendations are in practice – with the exception of external studies or EES indicators (see 4.5) – largely based on reports prepared by national governments. Overall guidelines and recommendations are likely only in rare cases to cause alterations to the information pool.

At the same time recommendations contain no information about the policy of other member states from which a country could learn. First, deliberative learning with others presupposes a conviction that cooperation enables more successful problem solving. In view of the unrealistic goals of the Lisbon Strategy, the motivation for the individual to invest in an exchange of experience and common problem-solving should be limited (Zängle 2004). Second, guidelines and recommendations set goals and impose concrete targets. While deliberative exchanges between actors are expected to produce new and better solutions, the guidelines are based on compromise between political actors. In brief, the European Commission is interested in greater quantification and concretisation of goals and obligatory reporting (Bruno 2008: 105). However, it is in the interests of member states to avoid specific commitments as much as possible. Empirical studies show that consultations on formulating guidelines have increasingly taken on the character of “interest-driven bargaining“ rather than “result open deliberation” (Jobelius 2003: 25; Radulova 2007: 374-5). Third, the growing substantive differentiation of guidelines into (partially) quantified, group-specific goals is at odds with the prerequisites for open learning processes. Finally, withh their ideologically stamped plea for the promotion of employability and a shift from passive to activating social security benefits, the EES is in substance leaning towards liberal and social-democratic welfare state models (Salais 2006). Whether this policy orientation is itself based on a learning process (for example, owing to the relative poor performance of corporatist-conservative regimes) or is to be explained in terms of (ideological) interests of the given governments or the Commission cannot be conclusively judged in this context. What is certain is that the guidelines reflect a specific steering intention – in contrast to open-ended learning processes (5).

Overall, guidelines and recommendations appear not to be well suited to promoting learning processes. Relatively speaking, their greatest potential seems to be in inducing the expansion of actors’ own experience. But they do not make information about the policies of other member states available. Unrealistic goals, uniform and increasingly concrete targets are likely to leave little scope for learning with others.

Over and beyond learning, other transfer mechanisms offer explanations for change. In recent empirical studies changes become apparent where guidelines or recommendations are used to defend their interests against criticism at the national level (see Vanhercke this issue) or where financial support from the EU for social policy initiatives enhances the responsiveness of national actors to OMC guidelines and recommendations (Mabbett 2007: 88; Weishaupt this issue).

4.2. National Reform Programmes to contents list

On the basis of guidelines and recommendations, each member state draws up a National Reform Programme (NRP). Because NRPs are prepared at the national level, processes of learning with other national actors should be to the fore. The NRPs could prompt repetition of learning processes oriented on the European level between relevant actors. To this end they should be formulated in a wide circle of major stakeholders, national parliaments, taking account of sub-national interests and with „active participation of civil society” (Ferrera et al. 2002: 237).

In practice, the circle of actors involved is small (Kröger 2007; Kerber and Eckardt 2007: 236). Research criticises a failure to involve the local level (Jacobbson and Schmid 2003: 117) and social partners (De La Porte and Nanz 2004: 279; even where social partner involvement in national policy making is traditionally strong, e.g. Sweden, Vifell 2004: 17). Moreover, responsibility for preparing NRPs is often entrusted to departments specialised in international policies. Work proceeds in relative isolation from the decisive arenas of national employment policy and overall strengthens the executive(s) (e.g. for Sweden: Vifell 2004: 14; on other OMCs Kröger 2006; Radulova 2007: 373-374). At the European level, too, the circle of actors involved is limited and, in comparison with the early years of the EES, an “intergovernmental drift” is apparent (Goetschy 2003: 88; Salais 2006: 201). What is important is that the limits to learning described are grounded in the interests of national governments, in their endeavours to retain control of the process. Other than in the case of guidelines and recommendations, criticism is levelled not at the design of the instrument but at how it is used and deployed in most member states. For this very reason, the challenge to learning through NRPs is posed not only by their discussion of learning potential under ideal conditions but also by their concern with (power) political interests (see Flear this issue).

NRPs also have the potential to stimulate learning not only in the domestic arena but also with other member states at the European level. Here, too, assessment tends to be sceptical. One hope set in deliberative learning is that is can improve the basis for decision making through exchanges on future decisions and strategic planning. Although NRPs could in principle combine annual review and planning, they tend to be predominantly evaluative in nature, thus precluding their anticipation in their own policy decisions by member states.

Analysis of the instrument shows that NRPs primarily seek to promote learning with others at the national level. A major obstacle is the factual limitation to the actors involved in formulation. To this must be added the evaluative nature of NRPs, which reduces the potential for learning with others at the European level.

4.3. ‘Good practice’ benchmarking in the Annual Progress Reportto contents list

The Annual Progress Report (prior to 2005 Joint Employment Report) serves to evaluate labour market policies and their outcomes. In the report, the European Commission assesses NRPs, compares member states, emphasises good practice, and benchmarks member state labour market performance. The main object of the report is to exchange information and learn from others. Two forms of interaction are to be distinguished: Horizontal discussion among member states in the Council, where the European Commission is also present (Cambridge Process) and bilateral discussion among representatives of member states and the Commission.

Horizontal exchange at meetings of the Council was originally restricted to less than an hour per country (Mosher and Trubek, 2003: 48). With the expansion and enlargement of the Council, NRPs are now discussed in four smaller working groups and summary results are presented in the plenum (Zeitlin, 2005: 489). The text of the report as adopted by the Council is available on the EES website. In 2005 a scoreboard was introduced, assessing the policy response of member states to each country-specific recommendation on an A to D scale: A (completed), B (in progress), C (limited), or D (incomplete).

Bilateral exchange serves firstly to clarify issues and secondly, according to Casey and Gold (2005: 25) “to negotiate around any ‘recommendations’ that the report will contain.” This suggests that no learning processes are likely. Institutionalised horizontal exchange could facilitate learning from others. Although it seems unlikely that, in the brief period of formal exchange, decision makers can find “new solutions to problems often thought to be unsolvable” (Trubek and Trubek 2005: 93), the change in procedure in 2004 has continued to enable individual discussion of NRPs offering more potential not only for discussing reform outcomes but also impact chains and conditions for reform.

In the Progress Report, and more so in the scoreboards, bad performance is named (naming and shaming) and good performance praised (faming). The scoreboard allows member states’ policy successes to be directly compared and ranked. The scoring system recalls marks at school. A poor mark can amount to a communicative sanction “which can bring about real pressure on member states to change their policies” (Ashiagbor 2004: 315; Regent 2003: 210; Bruno 2008: 111).

This could have two consequences. The establishment of competition in the context of benchmarking could induce member states to take account of new strategies for solving problems. From this point of view, ‘sporting competition’ would provide a positive framework for reactive learning at an individual level (cf. Benz 2008: 37-38). However, it can be objected that even without the EES, there would in principle be competition in employment policy performance. On the other hand, the incentive for member states to engage in (potentially) costly reforms to do better in the Progress Report is likely to be small. Another interpretation is that benchmarking seeks to steer behaviour by discrediting member states. This sort of competition is likely to worsen conditions for learning from others. If so the assessment that EES instruments create “trust and cooperation” and thus stimulate learning processes (Ferrera et al. 2002: 227) would not hold true. Where member states change policies only to avoid being at the bottom of the class, the transfer mechanism involved is pressure (cf. Lodge 2007). Dissonance between conditions necessary for stimulating learning – trust and the conviction that cooperation is of mutual benefit – and the competition engendered by direct comparison are weighty obstacles to learning.

To judge conclusively the potential of Progress Report benchmarking for learning from others would require additional information on the selection criteria for the groups and on communication in the plenum. These factors influence whether horizontal exchange amounts to bargaining or deliberative problem-solving. Thus all that can be established at this point is that there is a basic incentive to exchange information.

Benchmarking in the Progress Report could serve to distinguish successful policies from those that are less successful, to select models (Regent 2003: 209) and to create coherent targets (mimesis, Trubek and Trubek 2005: 92). However, concentrating on single examples of good practice runs counter to the diversity of national policies (Scharpf 2002; Bruno 2008). This suggests, firstly, that certain models are suitable for all member states. Secondly, diversity is reduced and alternatives excluded. In complex fields like employment policy, various reform paths can lead to improved performance and attainment of EES goals even where problems are similar (equifinality). Benchmarking tends to suppress the necessary independent innovation and creativity that could provide responses to country-specific problems (Lundvall and Tomlinson 2001: 123). Thirdly, insufficient account is often taken of systemic contextual factors; incomplete or uninformed transfer can prove harmful and cause policies to fail (e.g. Kerber and Eckardt 2007: 234). Fourthly, best or good processual practices or strategies cannot as a matter of course be identified on the basis of outcomes (Arrowsmith et al. 2004: 320 and 328). Finally, not all policies that today appear to solve problems will be successful in the long term and with sustainable impact (“lemming effect”, Lundvall and Tomlinson 2001: 123). In all, the instrument of benchmarking appears primarily suitable for stimulating learning from others where the problems involved are technical – and not so much in complex matters like employment policy.

4.4. Peer review to contents list

Peer reviews were initiated in 1999 by the European Commission to stimulate learning from others through the exchange of good practices. Each country sends two representatives, a government representative and an independent expert; Commission officials also participate. The meeting lasts two days. Lectures and discussions on the policy example are mostly supplemented by a field visit, e.g. to a firm.

Each year six examples are selected (in earlier years between 8 and 10). At the same time, thematic seminars are held under the umbrella of the Mutual Learning Programme. Focal topics are dealt with in a semi-annual cycle. A start is made with a thematic seminar. Participation has been extended to include social partners and NGOs from the national and European levels. Finally, funding is available in connection with peer review for bilateral and multilateral follow-up activities, e.g. information events to pass on the knowledge gained at the EU level. With these reforms the instrument of peer review has on the whole improved the potential for learning from others as well as with others.

The limits within which peer review can stimulate learning are set by the general difficulty of transferring successful problem-solving in a complex economic, social, and ideological context. In empirical studies, Casey und Gold (2005) identify a number of institutional obstacles to the transfer of policies. Potential impediments include the lack of infrastructure for implementation, divergent legal systems (regulation by collective bargaining agreements or by legislation), the pattern of industrial relations (e.g. collective bargaining autonomy for the social partners or statutory minimum wage), social security systems (e.g. financing through taxes or contributions), and political structures. In a similar vein, Scharpf (2002) argues that learning processes can work only between countries with relatively similar, historically evolved institutional arrangements.

This fundamental criticism can be mitigated in two aspects. First, the intention of peer review is not necessarily one-to-one transfer. In positive terms, it can serve to breach patterns of thought and “to encourage poor performers to rethink their strategy” (Trubek and Trubek 2005: 94). Secondly, criticism does not primarily address the conception of the instrument but its organisation. Member states can choose countries and topics that display similarities to their own, institutional structures, and actor constellations. If peer review provides more information on these factors and if the selection process were more strongly structured in this regard, the preconditions for learning could improve. The difficulty of transferring policies in complex fields with historically evolved institutions remains – especially when the political will is lacking. In sum, peer review offers only limited potential for promoting learning.

4.5. Indicatorsto contents list

Since 1997, performance indicators have been used in the EES context to facilitate evaluation of member state employment policy. Like guidelines, indicators have become more precise and numerous in the course of time. Key indicators (monitoring indicators) are used to measure the performance of labour markets with respect to prevention, activation, and employment, and to substantiate NRPs, for example, employment rates and long-term unemployment rates, as well as compatibility of work and private life. In parallel, context indicators (analysis indicators) permit better classification and evaluation of the key indicators of the given country, for example GNP growth, employment rate in full time equivalents or spending on life-long learning.

Of the instruments under study, indicators seem to be best suited to stimulate learning processes through reflexive self-evaluation. The regular capture of key indicators gives member states an incentive to rethink the causal connections between the expected impact of policies and the results of action, and, where necessary, systematically to record existing imbalances (Zeitlin 2005: 479). It is assumed that, in response to the newly generated knowledge, governments adapt their policies to take account of the changed information pool in seeking to attain an employment goal (simple learning). The positive influence on learning processes is expected to go beyond single indicators. Empirical studies stress the effect of the instrument at the institutional administrative level, for example, where the obligation to capture data induces member states to introduce or improve national structures for supervising the labour market, or where some member states exploit the indicators for the purpose of “tracking national policy targets – not only European ones” (Radulova 2007: 376). Furthermore, it is to be assumed that comparable indicators in different member states structure information, and are thus a precondition for learning with others and from others in the context of NRPs, benchmarking, and peer review (sceptical, based on empirical analysis Mabbett 2007; more optimistic Atkinson et al. 2004: 68).

What indicators are recorded is decided each year by an expert group of the Employment Committee. De La Porte und Pochet (2004: 282) argue that an expert body is a better prerequisite for mutual learning than bodies dominated by political interests. However (final) decisions on indicators and their quality are made at the political level. Backdrop to the debate are both political positions and a conflict on competence between national governments and the European Commission (see Salais 2006: 205 on indicators „politically monitor[ing] the EES“). While these conflicts hamper learning processes that could be stimulated by the instrument (cf. bargaining versus problem-solving), it shows that the role of indicators is not limited to the generation of information. Quantitative indicators help consolidate the OMC. If it is true that member states have so far supported the OMC “because of its low degree of legalization and its limited potential for unintended consequences” (Schäfer 2004: 13), this development is likely to produce tensions. Whether this concern of member states is justified cannot be conclusively decided on the basis of the research literature. There are certain signs that the OMC is used by the Commission to obtain consensus for central decisions (Scharpf 2001). Similarly, the Commission could regard the OMC as an instrument for expanding competence. The OMC enables the Commission to act in policy areas where its scope for action has hitherto been restricted by the treaties or by persistent conflicts.

Indicators stimulate learning through reflexive evaluation and targeted inducement to generate information. In comparison with the other EES instruments, the stimulation of learning in this case is less affected by inherent problems like the non-transferability of complex solutions to problems or the selection of particular models but it is affected by the very choice of indicators. However, political conflicts are emerging about the finalité of the OMC, which could hinder the optimum exploitation of the learning potential offered by indicators.

5. Summary and recommendations for action to contents list

The point of departure for this paper has been the growing prominence of the learning concept in the research literature and the wide-spread hope that instruments that can promote policy learning constitute a positive innovation and expansion vis-à-vis classical forms of regulation with all their limitations. At the same time, research into ‘learning’ in the context of the OMC and the EES shows methodological problems in the empirical analysis of causal connections, in taking due account of alternative internal and external explanations, and in establishing an analytical distinction between various transfer mechanisms. I have argued that these problems do not arise if analysis of learning in the EES context starts with the instruments employed. Without permitting conclusions to be drawn on the actual extent of learning, this approach enables the potential of the tools for stimulating learning to be addressed.

The characteristics and processes of the steering tools used under the EES can foster various forms of learning and can vary in their potential for stimulating learning. If the design and processual application of an instrument is not suited to promoting learning, the steering effect of the EES through policy learning will be weak. Overall, guidelines and recommendations appear to be not well suited to stimulating learning processes. Relatively speaking, their greatest potential seems to be in inducing actors to enlarge their own experience. However, like other forms of learning, this is limited by the fact that guidelines and recommendations seldom alter the information pool, that goals are unrealistic and uniform, and that increasingly concrete targets leave little room for learning with others. National Reform Programmes are designed to foster learning with others in a deliberative process. However, the relevant sub-national actors are very little involved at the national level in formulation and exchange processes. Benchmarking in the Progress Report aims to stimulate learning from others, but focusing on only a few models and the decontextualised perspective call learning from others into question. The peer review instrument promotes potential learning from other member states. What limits learning is the difficulty of transferring policies in complex fields with historically evolved institutions. Finally, indicators stimulate learning through reflexive evaluation and targeted inducement to generate information. While indicators have the potential to promote learning through own experience, political conflicts about the consolidation of the OMC may in future hamper the optimum exploitation of this potential. Overall, the analysis on the basis of existing studies has shown that the potential of the EES instruments to stimulate learning is subject to considerable limitations.

In some contexts these limitations could be eliminated (at least to some extent) by altering processes and characteristics. Learning through own experience could be further improved if indicators were more strongly geared to criteria developed by expert groups and researchers – the downside being low political feasability and lack of democratic legitimacy. Nor is conflict about the consolidation of the method through the quantification of guidelines and indicators necessarily inevitable; it depends on the use the European Commission makes of the instruments as well as member state responses supported in turn by political majorities. The second form of learning, namely learning from others, could also gain potential from the modification of instruments. Greater process orientation and use of performance monitoring and case studies would enable loose but coherent steering (Arrowsmith et al. 2004: 325). Peer review should take place between groups of countries with similar institutional arrangements and structures. Policies in less problematic areas are likely to be more successfully transferred and changes more easily achieved through soft steering instruments, as well. Clusters, which could vary depending on the subject matter, could offer better conditions for learning from others. In general, the exchange of information can be promoted by greater transparency and the wider dissemination of the relevant documents among the actors affected. Finally, learning with others would be more likely if employment policy guidelines formulated realistic goals that could actually be attained through joint action (Zängle 2005). Deliberative learning processes would also benefit from changes in member state practices in preparing NRPs. The circle of actors involved should be extended and stabilised to include a broader group of sub-national actors. Zeitlin (2005: 486) proposes using benchmarking not only for policy results and outcomes but also for measuring the broad involvement and relative influence of relevant actors (determining the nominal participation and the points in the NRP that substantively concern sub-national interests).

However, this suggestion for stimulating deliberation through greater competition risks replacing learning processes by pressure to adopt policies. With respect to interaction between instruments and their compatibility, this constitutes the weightiest obstacle to stimulating learning processes through EES tools. The assumption was formulated at the outset that, where the steering intentions of different instruments contradict one another, this could reduce or neutralise the learning potential of EES as a whole. The underlying conditions for exchanges between countries in the sense of deliberative processes include similar or at least compatible interests, as well as trust, sustained relations between actors, and the conviction that cooperation in solving problems is to the benefit of all. These conditions are at odds with the interests of actors in situations that promote bargaining as an interaction mode. Benchmarking and (soft) sanctions seek to attain EES goals through competition between member states. The setting for learning with others is also negatively affected by the institutional context of increasing intervention by the highest political levels to influence operation of the OMC, the debates on consolidating the method, and political controversy about the quantification of goals. From a theoretical and abstract point of view, this tension arises from the contradiction between two interaction modes, namely problem-solving and bargaining (Scharpf 1997: 124). A positive aspect that should be noted is that learning through own experience stimulated by EES instruments appears to depend less on the interaction orientations of the actors involved (but see Horvath this issue). If the EES wishes to contribute to attaining employment policy goals through policy learning, this is the field that offers the greatest potential.

Referencesto contents list

Argyris, Chris and Donald A. Schön. 1996. Organizational Learning II. Theory, Method, and Practice. Reading MA: Addison-Wesley.

Arrowsmith, James, Keith Sisson and Paul Marginson. 2004. What can 'benchmarking' offer the open method of co-ordination? Journal of European Public Policy 11 (2): 311-28.

Ashiagbor, Diamond. 2004. Soft Harmonisation: The 'Open Method of Coordination' in the European Employment Strategy. European Public Law 10 (2): 305-32.

Atkinson, Tony, Eric Marlier and Brian Nolan. 2004. Indicators and tragets for social inclusion in the European Union. Journal of Common Market Studies 42 (1): 47-75.

Benz, Arthur. 2007. Accountable Multilevel Governance by the Open Method of Coordination? European Law Journal 13(4): 505-22.

Biagi, Marco. 2000. The Impact of European Employment Strategy on the Role of Labour Law and Industrial Relations. The International Journal of Comparative Labour Law and Industrial Relations 16 (2): 155-73.

Bruno, Isabelle. 2008. Y a-t-il un pilot dans l'Union? Tableaux de bord, indicateurs, cibles chiffres: les balises de la décision. Politix 21 (82): 95-118.

Casey, Bernard H. and Michael Gold. 2005. Peer review of labour market programmes in the European Union: what can countries really learn from one another? Journal of European Public Policy 12 (1): 23-43.

De La Porte, Caroline and Patricia Nanz. 2004. The OMC - a deliberative-democratic mode of governance? The cases of employment and pensions. Journal of European Public Policy 11 (2): 267-88.

De La Porte, Caroline and Philippe Pochet. 2002a. Introduction. In Building Social Europe through the Open Method of Co-Ordination, ed. Caroline De La Porte and Philippe Pochet,. 11-26. Brussels: P.I.E. Peter Lang.

De La Porte, Caroline and Philippe Pochet. 2002b. Supple Co-ordination at EU Level and the Key Actors' Involvement. In Building Social Europe through the Open Method of Co-Ordination, ed. Caroline De La Porte and Philippe Pochet, 27-68. Brussels: P.I.E. Peter Lang.

Deutsch, Karl W.. 1969. Politische Kybernetik. Modelle und Perspektiven. Freiburg: Rombach.

Dorf, Michael C. and Charles Sabel. 1998. A Constitution of Democratic Experimentalism. Columbia Law Review 98 (2): 267-73.

Eberlein, Burkard and Dieter Kerwer. 2004. New Governance in the European Union: A Theoretical Perspective. Journal of Common Market Studies 24 (1): 121-42.

Erhel, Christine, Lou Mandin and Bruno Palier. 2005. The Leverage Effect. The Open Method of Co-ordination in France. In The Open Method of Coordination in Action: The European Emplyoment and Social Inclusion Strategies, ed. Jonathan Zeitlin, Philippe Pochet and Lars Magnusson, 217-47. Brussels: Peter Lang.

Flear, Mark. 2009. The Open Method of Coordination on health care after the Lisbon Strategy II: Towards a neoliberal framing? In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (12), http://eiop.or.at/eiop/texte/2009-012a.htm.

Ferrera, Maurizio, Manos Matsaganis and Stefano Sacchi. 2002. Open co-ordination against poverty: the new EU 'social inclusion process'. Journal of European Social Policy 12 (3): 227-39.

Goetschy, Janine. 2003. The Emplyoment Strategy and European integration. In Five years' experience of the Luxembourg employment strategy, ed. David Foden and Lars Magnusson, 69-109. Brussels: Peter Lang.

Hall, Peter A..1993. Policy Paradigms, Social Learning, and the State. The Case of Economic Policymaking in Britain. Comparative Politics 25 (3): 275-96.

Heclo, Hugh. 1974. Modern Social Politics in Britain and Sweden. From Relief to Maintenance. New Haven: Yale University Press.

Horvath, Anna. 2009. What kind of consensus? Conflicting notions of effectiveness within the Social Protection Committee. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (17), http://eiop.or.at/eiop/texte/2009-017a.htm.

Jacobbson, Kerstin and Herman Schmid. 2003. The European Emplyoment Strategy at the crossroad: Contributions to the evaluation. In Five years' experience of the Luxembourg employment strategy, ed. David Foden and Lars Magnusson, 111-40. Brussels: Peter Lang.

Jobelius, Sebastian. 2003. Who formulates European Emplyoment Guidelines. Paper presented at the ESPAnet conference "Changing European Societies - The Role of Social Policy", November 13-15, in Copenhagen, Denmark.

Kerber, Wolfgang and Martina Eckardt. 2007. Policy learning in Europe: the open method of co-ordination and laboratory federalism. Journal of European Public Policy 14 (2): 227-47.

Kröger, Sandra. 2006. When learning hits politics or: Social policy coordination left to the administrations and the NGOs?, European Integration online Papers 10 (3), http://eiop.or.at/eiop/texte/2006-003a.htm (accessed on October 5, 2009).

Kröger, Sandra. 2007. The End of Democracy as We Know It? The Legitimacy Deficits of Bureaucratic Policy Governance. Journal of European Integration 29 (5): 565-82.

Kröger, Sandra. 2009. The Open Method of Coordination: Underconceptualisation, overdetermination, de-politicisation and beyond. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (5), http://eiop.or.at/eiop/texte/2009-005a.htm.

Lodge, Martin. 2007. Comparing Non-Hierarchical Governance in Action: the Open Method of Co-ordination in Pensions and Information Society. Journal of Common Market Studies 45 (2): 343-65.

López-Santana, Mariely. 2009. Having a Say and Acting: Assessing the effectiveness of the European Employment Strategy as an intra-governmental coordinative instrument. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (15), http://eiop.or.at/eiop/texte/2009-015a.htm.

Lundvall, Bengt-Ake and Mark Tomlinson. 2001. Learning by comparing: Reflections on the use and abuse of international benchmarking. In Innovation, Economic Progress and the Quality of Life, ed. Gerry Sweeney, 120-36. Cheltenham: Edward Elgar.

Mabbett, Deborah. 2007. Learning by numbers? The use of indicators in the co-ordination of social inclusion policies in Europe. Journal of European Public Policy 14 (1): 78-95.

Mosher, James S. and David M. Trubek. 2003. Alternative Approaches to Governance in the EU: EU Social Policy and the European Employment Strategy. Journal of Common Market Studies 41 (1): 63-88.

Niechoj, Thorsten. 2009. Does supranational coordination erode its national basis? The case of European labour market policy and German industrial relations. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (10), http://eiop.or.at/eiop/texte/2009-010a.htm.

Pfister, Thomas. 2009. Governing the knowledge society: Studying Lisbon as epistemic setting. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (6), http://eiop.or.at/eiop/texte/2009-006a.htm.

Radulova, Elissaveta. 2007. The OMC: An Opaque Method of Consideration or Deliberative Governance in Action? Journal of European Integration 29 (3): 363-80.

Radulova, Elissaveta. 2009. The construction of EU’s childcare policy through the Open Method of Coordination. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (13), http://eiop.or.at/eiop/texte/2009-013a.htm.

Regent, Sabrina. 2003. The Open-Method of Coordination: A New Supranational Form of Governance? European Law Journal 9 (2): 190-214.

Rose, Richard. 1991. What is lesson drawing? Journal of Public Policy 11 (1): 3-30.

Sabatier, Paul A.. 1986. What Can We Learn from Implementation Research? In Guidance, Control, and Evaluation in the Public Sector: The Bielefeld Interdisciplinary Project, ed. Franz-Xaver Kaufmann, Giandomenico Majone and Vincent Ostrom, 313-26. Berlin: De Gruyter.

Sabel, Charles. 1994. Learning by Monitoring: The Institutions of Economic Development. In The Handbook of Economic Sociology, ed. Neil G. Smelser and Richard Swedberg, 275-96. Princeton: Princeton University Press.

Salais, Robert. 2006. Reforming the European social model and the politics of indicators: from the unemployment rate to the employment rate in the European Employment Strategy. In Unwrapping the European Social Model, ed. Anna Serrano Pascual and Maria Jepsen, 189-212. Bristol: The Policy Press.

Schäfer, Armin. 2004. Beyond the Community Method: Why the Open Method of Coordination Was Introduced to EU Policy-making. European Integration online Papers 8 (13), http://eiop.or.at/eiop/texte/2004-013a.htm (accessed on October 5, 2009).

Schäfer, Armin. 2005. Die neue Unverbindlichkeit: Wirtschaftspolitische Koordinierung in Europa. Frankfurt/M.: Campus.

Scharpf, Fritz W.. 1997. Games Real Actors Play. Actor Centered Institutionalism in Policy Research. Boulder: Westview.

Scharpf, Fritz W.. 2001. European Governance: Common Concerns vs. the Challenge of Diversity. In Mountain or Molehill? A Critical Appraisal of the Commission White Paper on Governance, ed. Christian Joerges, Yves Mény and Joseph H. H. Weiler, 1-11. Florence: Robert Schuman Centre, European University Institute.

Scharpf, Fritz W.. 2002. The European Social Model: Coping with the Challenges of Diversity. Journal of Common Market Studies 40 (4): 645-70.

Trubek, David M. and Louise G. Trubek. 2005. The Open Method of Coordination and the Debate over "Hard" and "Soft" Law'. In The Open Method of Co-ordination in Action. The European Employment and Social Inclusion Strategies, ed. Jonathan Zeitlin, Philippe Pochet and Lars Magnusson, 83-103. Brussels: Peter Lang.

Vanhercke, Bart. 2009. Against the odds. The Open Method of Coordination as a selective amplifier for reforming Belgian pension policies. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (16), http://eiop.or.at/eiop/texte/2009-016a.htm.

Vifell, Asa. 2004. Speaking with forked tongue. Swedish employment policy and European guidelines: a case of Europeanization through soft-coordination. Paper presented at the Conference of Europeanists, March 11-13, Chicago; USA.

Visser, Jelle. 2005. The OMC as a Selective Amplifier for National Strategies of Reform. What the Netherlands Want to Learn from Europe. In The Open Method of Coordination in Action: The European Emplyoment and Social Inclusion Strategies, ed. Jonathan Zeitlin, Philippe Pochet and Lars Magnusson, 173-215. Brussels: Peter Lang.

Visser, Jelle and Anton Hemerijck. 2001. Learning Ahead of Failure: a Research Agenda. Paper presented at the Max Planck Institute Cologne, November 13, Cologne, Germany.

Weishaupt, J. Timo. 2009. Money, votes or ‘good’ ideas? Partisan politics and the effectiveness of the European Employment Strategy in Austria and Ireland. In What we have learnt: Advances, pitfalls and remaining questions in OMC research, ed. Kröger, Sandra. European Integration online Papers, Special Issue 1, 13 (14), http://eiop.or.at/eiop/texte/2009-014a.htm.

Zängle, Michael. 2004. The European Union benchmarking experience. From euphoria to fatigue? European Integration online Papers 8 (5), http://eiop.or.at/eiop/texte/2004-005a.htm (accessed on October 5, 2009).

Zeitlin, Jonathan. 2005. The Open Method of Coordination in Action: Theoretical Promise, Empirical Realities, Reform Strategy. In The Open Method of Co-ordination in Action. The European Employment and Social Inclusion Strategies, ed. Jonathan Zeitlin, Philippe Pochet and Lars Magnusson, 447-503. Brussels: Peter Lang.

Zeitlin, Jonathan, Philippe Pochet and Lars Magnusson (eds). 2005. The Open Method of Coordination in Action: The European Emplyoment and Social Inclusion Strategies. Brussels: Peter Lang.

Endnotesto contents list

(*) An earlier version of this paper was published as WZB Discussion Paper SP I 2006-114. Many thanks to Sandra Kröger, Armin Schäfer, Günther Schmid and an anonymous reviewer for valuable comments.

(1) In contrast to Visser und Hemerijck (2001), I do not treat the OMC as a homogeneous instrument that promotes only one form of learning but look at each tool used seperately.

(2) Taking this argument further, much of the literature on deliberative processes stresses the value of decisions taken under this approach as increasing democratic legitimacy for the OMC (e.g. Eberlein and Kerwer 2004).

(3) Since 2005 the Employment Report forms one chapter within the overarching EU Annual Progress Report on the Lisbon Strategy.

(4) Kröger (2006: 13) makes a similar argument about the (lack of) potential of OMC social inclusion NAPs and peer reviews to stimulate learning. Lodge (2007) stresses the high instrumental and contextual prerequisites of OMC pensions and information society to operate effectively.

(5) A further point of criticism concerns the legitimation of this steering intention, seeking both new goals and a change in underlying standards and values (cf. Schäfer 2005: 219, Flear this issue).

European Integration online Papers | ISSN 1027-5193
© 1997-2015 ECSA Austria | All Rights Reserved. Webmaster Editorial Assistant