NGOs as new Guardians of the Treaties? Analysing the effectiveness of NGOs as decentralised enforcers of EU law

ABSTRACT

The European Union is increasingly resorting to a decentralised enforcement strategy in which it relies less on the Commission-initiated infringement procedure, and more on domestic litigation by particularly NGOs. In order to better understand whether this is a suitable strategy to tackle the compliance deficit, we need more insight into NGOs as decentralised enforcers of EU law. What do NGOs do to stimulate domestic compliance, and is this a potentially effective alternative to the centralised enforcement strategies? Analysing the actions by the German NGO Deutsche Umwelthilfe in relation to the Ambient Air Quality Directive, reveals high potential for the role to be played by NGOs. This exploratory case illustrates that particularly the strategy of flooding the country with a wave of legal proceedings was effective in increasing the salience of the topic, thereby indirectly releasing those capacities that the authorities had still lacked to comply.

KEYWORDS:

• compliance
• decentralised enforcement
• environmental policy
• European Union
• NGOs

Introduction

Air pollution is Europe’s largest environmental health risk, responsible for around 400,000 deaths per year (EEA, 2020, p. 10). The European Union (EU) already started addressing air pollution in the 1980s, and since then air quality legislation has been continuously updated. The Ambient Air Quality Directive (AAQD) of 2008 has been legally transposed by all member states, however, significant portions of the EU population are still exposed to pollution levels that exceed the Directive’s limits (ECA, 2018). This is one of the many cases that illustrate that the EU suffers from a compliance deficit. Although the size of this deficit is debated, the wider body of academic scholarship on EU compliance agrees about the severity of the problem (see Treib, 2014 for a stocktaking overview of the literature).

The most pervasive tool available to the EU to address compliance problems is the infringement procedure, ultimately penalising non-complying member states before court. Over the last two decades, however, we witness a shift from the use of the centralised infringement procedure initiated by the European Commission, to relying on stakeholders such as NGOs to raise non-compliance issues before national courts (Mathieu et al., 2018). Kelemen (2012, p. 56) already captured this development under the label ‘Eurolegalism’, which he used to identify the trend of the EU increasingly relying on ‘formal, transparent legal norms backed by more aggressive public enforcement and expanded opportunities for private enforcement litigation’. More recently, Hofmann (2018, 2019) provided more insight into how the Commission is withdrawing from enforcement and is increasingly ‘outsourcing’ this to private actors at the national level. Particularly in the environmental sector this change can be observed. Prior to 2010 there was a strong reliance on the central infringement procedure, but since then there has been a sharp decline and the number of proceedings initiated by NGOs before national courts is increasing considerably (Hofmann, 2019). However, we still know little about how exactly NGOs act as decentralised enforcers. Therefore, to better understand the potential of the shift in strategy from centralised infringement procedures to decentralised enforcement, we need more insight into the role that NGOs play in this process.

While the body of literature on NGO influence on compliance with EU law is slowly expanding (e.g., Andanova & Tuta, 2014; Cent et al., 2013; Kaya, 2018), and we are gaining more and more insight into private litigation in the EU (e.g., Conant et al., 2018; Hofmann & Naurin, 2021), we still have limited empirical insight into whether NGOs as decentralised private enforcers can be perceived as an effective replacement of the centralised enforcement strategy. This exploratory article asks what strategies NGOs use in domestic litigation, and how such strategies are perceived by the relevant actors in the implementation process, in order to be able to better judge the effectiveness of decentralised enforcement. We do so by presenting a most likely case of NGO involvement. The Ambient Air Quality Directive represents the basis for a range of precedents in which NGOs and other claimants have contributed to the creation of a right to clean air. The German NGO Deutsche Umwelthilfe (DUH) was at the forefront of this development.

Trying to gain more insight into the role of NGOs as decentralised enforcers of EU law, and thus as potential additional Guardians of the Treaties, this article analyses the case of DUH and the AAQD implementation in Germany based on interviews with DUH representatives, German implementing authorities as well as Commission officials. These insights lead us to conclude that there is strong potential for the role to be played for NGOs in stimulating domestic compliance with EU law.

Contextualising and conceptualising the role of NGOs as decentralised enforcers of EU law

Traditionally, most national court cases and preliminary reference cases concerned free movement and the internal market, where companies have strategically used litigation to try to overrule unfavourable national legislation (Hofmann, 2018). More recently, however, those policy sectors that concern public interests (e.g., consumer protection, non-discrimination, free movement and posting of workers, data protection, and environmental protection) also see more private litigation. Kagan (2003, p. 3) already coined the term ‘adversarial legalism’ for such a style of ‘policymaking, policy implementation, and dispute resolution through lawyer-dominated litigation’ for the United States. While Kagan doubted this style to take root in an EU that relies more on consensus and cooperation, Kelemen (2011, 2012) observes a convergence towards Eurolegalism. The increased use of detailed, enforceable legal norms, combined with increased access to justice, paved the way for more private litigation, in turn resulting in more stringent public enforcement in the EU member states. While it remains contentious whether we witness such a convergence towards Eurolegalism, several studies suggest that an increasingly adversarial legalistic style is developing in the EU (e.g., Bastings et al., 2017; Rehder, 2009).

Before we turn to our increasing understanding of private litigation in the EU, what do we already know about NGOs as facilitators of compliance in general? Whereas plenty of research exists discussing interest group strategies and influence at the pre-adoption stage (e.g., Dür & Mateo, 2013; Junk, 2015; Klüver, 2013), research discussing the implementation stage is scarcer. We know that NGOs seek access but not whether they gain access during the implementation stage (Kaya, 2018). The several qualitative case studies that are available illustrate that NGOs mostly play an important role as watchdogs and data suppliers, i.e., by providing expertise and data that national ministries need to comply (e.g., Andanova & Tuta, 2014; Cent et al., 2013). This is labelled as the cognitive power of NGOs by Boström and Hallström (2010, p. 46), who state that NGOs often provide information, expertise, on-ground experience and framing that is perceived as more reliable than any information states and corporations can provide. This in turn can have a socialization effect, persuading member states into compliance (e.g., Parau, 2009). NGOs can also send complaints to the Commission pushing them to open the infringement procedure, thereby impacting domestic compliance (e.g., Andanova & Tuta, 2014).

In addition to our understanding of how NGOs influence EU law compliance, we also recently gained more insight into what encourages private litigation in the EU. Factors that favour the occurrence of private litigation are grouped under the label ‘legal opportunity structures’ (e.g., Conant et al., 2018; Hilson, 2018; Hofmann & Naurin, 2021). Important examples of such opportunity structures are the degree of judicial independence, access to courts (standing and costs), and the possibility for judicial review. Access to justice is not the same in all member states, ranging from high accessibility in countries such as Portugal (Eliantonio, 2018), to low accessibility in, for example, German-speaking countries that are traditionally hostile to public interest litigation and slow to adapt their national systems (Hofmann, 2019). Making use of existing national opportunity structures requires NGOs displaying certain characteristics. Apart from the fact that there must be a relevant NGO in the first place (Dimitrova & Buzogany, 2014), this NGO must then also have access to the necessary resources, expertise and legal consciousness to leverage the existing opportunities (Börzel, 2006; Hofmann, 2018; Hofmann & Naurin, 2021). In addition, the position of the group, i.e., whether they can be considered insiders or outsiders, matters (Hofmann & Naurin, 2021), as this position influences a group’s political opportunities for influence (Conant et al., 2018). The more political opportunities for NGOs to influence the domestic system, the less need there is for following the route of private litigation. This thus leads to the hypothesis that outsider groups are more likely to turn to litigation (Hofmann & Naurin, 2021). Using comparative survey data from five EU countries, Hofmann and Naurin (2021) are the first to provide empirical insight into the anticipated occurrence of private litigation. Their results illustrate that overall litigation is a strategy that is still used relatively infrequently by interest groups and is most likely to be used by politically active and resourceful groups. In addition, outsiders are more likely to start litigation, and those that receive public funding are less likely to do so.

This recent literature thus provides important clues as to what factors seem to drive or push the presence of private litigation, thus answering the when and why question of the occurrence of decentralised enforcement. What is still missing in this wider body of literature, is more insight into the how and with what effect question of private litigation. As already outlined by Conant et al. (2018, p. 1378), we need a better understanding of how different parties mobilize law and how national institutions respond to this activity before being able to judge the impact of this trend. We know that impact of NGO litigation can vary across member states, as illustrated by Slepcevic (2009) who analysed how environmental organisations have addressed their national courts in France, Germany and the Netherlands, in relation to the Natura 2000 Directives. He showed that only in the Netherlands court litigation was successful in remedying formal and practical implementation problems. In Germany, only formal transposition problems were effectively countered, and in France litigation proved unsuccessful. The French case illustrated how a negative public stance (i.e., public support for open hunting periods) worked against the interest group litigation attempts. We also know from the literature on NGO impact on compliance as discussed above, that NGOs can resort to different strategies and activities (e.g., Andanova & Tuta, 2014; Cent et al., 2013; Parau, 2009). In addition to starting private litigation, they can also mobilize the Commission by submitting a complaint to trigger the infringement procedure, they can influence the national framing and persuade governmental actors into compliance, and they can provide lacking information and expertise. This article contributes to this body of literature by providing concrete case study evidence of what an NGO can do as a decentralised enforcer, and how this activity is perceived by the core actors involved.

Case selection and research approach

Exploring how and with what effect NGOs act as decentralised enforcers requires a most likely case, in order to be able to explore what NGOs do and how this is perceived by the relevant actors involved. Based on the literature discussed above that outlines the factors that seem to drive or push the presence of private litigation, we can illustrate how the role played by DUH in relation to the AAQD in Germany is a suitable case.

In order to analyse the role of NGOs as decentralised enforcers, we need to be able to detect non-compliance and there need to be directly enforceable rights. The AAQD being adopted in 2008, we can safely argue that non-compliance after so long cannot be excused with start-up difficulties but hints at more substantial problems. Furthermore, non-compliance itself is relatively easily detectable as the Directive established upper pollution limits that must not be exceeded. Hilson (2018) argues that an explicit framing of rights often is not clearly visible in the EU environmental sector, however, the AAQD case is a notable exception. The ‘right to clean air’ is directly enforceable, which is the prerequisite for NGOs to be theoretically able to resort to adversarial legal enforcement strategies (Vanhala, 2018, p. 384).

Active NGO involvement, further, requires a favourable legal opportunity structure (e.g., Hofmann & Naurin, 2021). Most importantly, the member state needs to ensure judicial independence, a possibility for political review, as well as court access. The AAQD case in Germany adheres to these preconditions. In the landmark case Janecek v Freistaat Bayern (2008), DUH supported a citizen of Munich who sued his federal state because the pollution levels at his street exceeded the AAQD limit values. To determine whether citizens were entitled to file such lawsuits, the case was referred to the European Court of Justice. It held that persons directly affected by exceeded limit values must be able to bring action against the competent authorities before national courts requiring them to draft effective air quality plans. The court thereby effectively introduced a citizen right to clean air not only granting citizens locus standi¹ ‘but also a right to a plan (the right to a legal remedy) and the right to demand judicial scrutiny of that plan (the right to substantive review)’ (Andrews & Taddei, 2015, p. 10). In a later DUH-initiated case, a German court also granted these rights, contrary to normal German procedural requirements, to claimants without direct and individual concern (Töller, 2021). This includes DUH and opened henceforth the doors for legal enforcement of the AAQD by NGOs.

Finally, DUH itself exhibits all the characteristics commonly identified in the literature as prerequisites for an NGO capable of taking diverse forms of action (Hofmann & Naurin, 2021; Rodela et al., 2017). These include besides material power resources, also symbolic resources (issue ownership and recognition), cognitive resources (expertise and experience), and social resources (access to networks) (Boström & Hallström, 2010). DUHs considerable material resources with an annual budget of more than 11 million euro (DUH, 2021) and its locus standi, represent the two essential conditions for accessing the power resource of legal enforcement (Börzel, 2006). In the field of air pollution, it has become the German NGO recognized for its exposure of ‘Dieselgate’ and its involvement in landmark legal proceedings (Hehemann & Uhlmann, 2018). The personnel of about 100 people are composed of experienced professionals with technical and legal expertise. Moreover, DUH is well connected in society and politics to an extent that it sometimes negotiates measures with public authorities on an equal footing (Doll, 2018).

Analysing the role of DUH in enhancing compliance with the AAQD in Germany, asks for gaining insights into how it interacts with those actors who are responsible for compliance, i.e., the Implementing Authorities (IAs). With respect to these authorities, Kaya (2018, p. 41) has found ‘anecdotal evidence’ that state actors value and benefit from interest group involvement in policy implementation. However, he himself acknowledges that these findings ‘should be taken with a pinch of salt without first studying the perceptions of formal implementing actors’. This article contributes to the testing of this ‘anecdotal evidence’ by not only interviewing the NGO but also hearing the IAs. A total of nine interviews were conducted², including two representatives from DUH, two Commission officials and five national officials from the IAs. The five national officials work(ed) at the ministerial level for five different federal states, the governance level responsible for the implementation of the AAQD in Germany. DUH initiated cases in nine federal states, and we interviewed the core responsible officials in the five states that showed the most serious cases. Often the official interviewed was the sole responsible actor, or part of a very small team. The decision to interview the IAs rather than elected politicians is based on Lenschow et al.’s (2017, p. 530) findings that the AAQD with its focus on technical matters and planning empowers the administrations to the extent that they are taking ‘most crucial implementation decisions, with elected bodies mostly “rubber stamping” these decisions’.

In the analysis we explore, first, how DUH acted as a decentralised enforcer. Based on the interviews and other available documentation, we illustrate DUHs strategy. We know from the literature identified in the previous section that there are four categories of activities that can be expected when analysing the role of NGOs in compliance: 1) influencing the framing, national debate, and thereby persuading a government into compliance, 2) providing lacking information and expertise, and thereby helping governments to comply, 3) launching a complaint, and thus triggering the start of the centralised infringement procedure, and 4) initiating litigation itself (decentralised enforcement strategy).

To assess whether these strategies do in fact enhance compliance, the second step in the analysis asks whether they are perceived as effective. We aim to analyse whether resorting to NGOs as decentralised enforcers is an effective compliance strategy. Following Miles and Underdal (2001), a compliance strategy is defined as effective when it stimulates that the given policy is actually put into practice and helps in solving the political problem. Analysing effectiveness is no easy endeavour, but the longstanding tradition of assessing interest group influence can help in understanding how we can assess the effectiveness of NGOs as decentralised enforcers. The methodological challenges identified in this literature relate to the difficulty to attribute outcomes aligned with NGO preferences to its deployed strategies (Klüver, 2013, Dür, 2008a; b). There is a reliability problem deriving from the data sources available to make an effectiveness assessment, and a causality problem of identifying causal chains between NGO strategies and outcomes while excluding external factors. An often used strategy is the use of triangulated interview or survey-based self-assessments of involved stakeholders (e.g., Binderkrantz & Pedersen, 2019; Dür & De Bièvre, 2007). These actors, however, will either subconsciously or consciously overstate or understate interest group influence. Dür (2008b) argues that the reliability problem can be resolved by cross-checking the self-assessment with perspectives of other stakeholders. DUH claiming that a respective strategy was effective (self-attributed influence) will be thus counterchecked with the perception of the Commission and IAs. Interviewing Commission officials, who do not have an interest in overstating the NGOs effectiveness, will enhance the reliability and will help to check to what extent the NGO has interacted with the Commission (e.g., by providing information for the Commission to initiate the infringement procedure). Tackling the causality problem requires, according to Klüver (2013), that researchers should show an association between actor properties and outcomes. If authorities suddenly complied after the deployment of legal proceedings or if technical solutions proposed by DUH were adopted, an association between the actor property and the eventual outcome would be present, indicating effectiveness. Nevertheless, even then one must actively check for alternative explanations (Klüver, 2013). In analysing DUHs effectiveness as a decentralised enforcer, we thus check whether we can indeed observe air quality improvements in regions that have been sued, as this would indicate that the NGO stimulated the policy to be put into practice and thus reach its intentions (i.e., cleaner air). In addition, we countercheck DUHs self-assessment of effectiveness with the effectiveness assessment of other parties such as the IAs and the Commission.

The final step in the analysis evaluates what it was in DUHs role in decentralised enforcement that is perceived as effective (or not). How do the various actors involved in this process evaluate the opportunities for NGOs as alternative guardians of the treaties, based on the experiences in this case study?

The Deutsche Umwelthilfe as a new Guardian of the Treaties?

The Ambient Air Quality Directive (2008/50/EC) aims to reduce harmful effects on human health and the environment. It requires the member states to implement standardized air quality monitoring practices, to report their findings to the European Commission (EC), and to adopt Air Quality Plans wherever limit values are exceeded (Andrews & Taddei, 2015, p. 6). These plans must contain appropriate measures to keep the period of exceedance as short as possible (Taddei, 2020, p. 152). The AAQD is considered the essential piece of legislation regulating matters of air pollution in Europe which explains why one official even referred to it as ‘the Bible’ of clean air (IA #1). The following analysis firstly examines how DUH acted as a decentralised enforcer. Secondly, we assess how effective DUH has been in enhancing compliance with the AAQD in Germany. Finally, we explore how all actors involved in this process perceive the role of NGOs as decentralised enforcers. What seems to work or not work in this decentralised enforcement strategy?

DUHs role as a decentralised enforcer

While Hofmann and Naurin (2021) convincingly illustrate that litigation is still a relatively underused strategy for NGOs, the case explored in this article forms an exception. DUH mostly resorted to litigation. One implementing authority representative described this as a ‘wave of lawsuits sweeping over Germany’ (IA #5). DUH proudly refers to lawsuits in 40 German cities for exceeding limit values of which it has not lost one. Being bothersome and searching the confrontation via litigation is referred to as a unique selling point of the organization (DUH #1). Wherever DUH initiated legal proceedings, it usually remained intransigent even when the Implementing Authorities showed increased efforts. This is described negatively by several IAs, who remark that increased efforts on their side were not acknowledged and that it seemed that DUH simply wanted to push through processes on principle (IA #3, 5). DUH explains its intransigence, however, by remarking that non-judicial settlements could not be relied upon. Experience had shown that when legal demands were dropped because authorities promised to take effective measures, matters were stalled. This ultimately triggered a situation of mutual distrust that led DUH to only conclude settlements that are legally binding and recorded in writing (DUH #2).

Next to the use of litigation, DUH also resorted to information provision, in the form of measuring of pollution levels to detect non-compliance. NGOs are expected to document non-compliance and submit their findings to the responsible enforcement authority, in this case, the Commission (Kaya, 2018, p. 29). Yet, the Commission officials remark that they have not received any data on breached limit values from DUH (EC #1). However, this is, as Commission officials and DUH confirm, only because the official measuring network required by the AAQD already sufficiently proved Germany’s non-compliance (EC #1, 2; DUH #2). Nonetheless, DUH carried out measurements to exert political pressure and raise issue-salience; in other words, to influence the framing of the debate. For instance, a large-scale citizen science campaign in 2019, in which citizens were equipped to take samples at hundreds of locations in Germany, gained significant attention as it revealed that air pollution was a widespread problem not limited to the major cities (DUH #2).

Rhetorically, DUH states that it likes to use somewhat pointed wording to generate public attention (DUH #1), although it claims to always maintain a balance between clearly naming problems and proposing solutions (DUH #2). While some national officials recognise that DUHs rhetoric is not exclusively confrontational (IA #1, 5) they all agree that confrontational naming and shaming prevails (IA #1-5). ‘The rhetoric is not exclusively negative but more often negative’ (IA #5).

DUH, thus mostly relying on an adversarial strategy of litigation and influencing the frame via public naming and shaming, emphasises that ‘cooperation, [and] seeking consensus is part of DUHs DNA’ (DUH-#1). The NGO states that it is in constant contact with the authorities and supports them in finding solutions to stop limit violations, for instance, by organizing expert panels (DUH #1). Although the representatives of the NGO refer to the repeated reaching of settlements to demonstrate their willingness to compromise (DUH #1, 2), they too must admit that finding compromises in a situation characterised by mistrust is complicated and that its public appearance is not that of ‘Mr. Compromise himself’ (DUH #2). There is one example where a city developed a comprehensive and far-reaching plan that DUH praised publicly and considered as one of the best they had ever seen (DUH #2). Nevertheless, the interviewed official in whose federal state this occurred was eager to emphasise that this was a unique phenomenon that happened only due to the extraordinarily courageous actions of the municipal environmental director, a member of the Greens (IA #2).

Despite its core strategy of litigation and public naming and shaming, DUH thus also has a self-image of being cooperative and consensus-seeking. Interestingly, the IAs predominantly reject this characterization. Only one official stated that the relation ‘was never really confrontational’ (IA #1). He reports regular invitations to expert discussions on best practices for reducing pollutants (IA #1). The four others, however, report little to no cooperation (IA #2-5). One emphasises that the relationship with DUH was confrontational from the very beginning and stresses that solution-oriented cooperation with DUH is impossible. ‘They tried to destroy us […]. It is almost like war’ (IA #3). DUH is repeatedly accused of a lack of willingness to compromise (IA #3-5). ‘You cannot cooperate if DUH comes with the maximum demand and shows no mercy. This is how you destroy a willingness to cooperate’ (IA #3). Moreover, some officials have never experienced that DUH would have sought any type of cooperation (IA #2, 4). In legal matters, DUH was perceived as intransigent and advancements were only rarely rhetorically lauded (IA #2, 3).

In sum, DUH initiated that a wave of lawsuits swept over Germany and heavily influenced the framing via naming and shaming. While the NGOs self-image is that of an actor seeking cooperation and consensus, the parties on the ‘receiving end’ – the implementing authorities – describe the strategy of the NGO as a rather adversarial approach. Having described how DUH (is perceived to have) acted as a decentralised enforcer, the next section analyses to what extent this strategy was effective. In other words, did this strategy indeed lead to better compliance with the AAQD in Germany?

The effectiveness of DUH as a decentralised enforcer

The first indicator for effectiveness is to see whether there is a sudden air quality improvement in regions that have been sued. Here we witness some evidence. DUH proudly refers to the fact that the air pollution levels have declined much quicker in those cities where it has sued compared to those where it has not (DUH #1, 2), a development that has been empirically confirmed (Töller, 2020, p. 292). In addition, when in June 2021 the European Court of Justice ruled in the Commission-initiated procedure against Germany for violating the AAQD, DUH had already concluded most of its cases and the number of cities violating limit values had dropped from 90 in 2016 to only six in 2020 (Abnett, 2021). However, as an alternative explanation, this sharp decline could also be driven by the fact that in cities with extremely high pollution and thus also high pollution reduction potential, a faster decline is to be expected anyway, and thus one should remain sceptical to fully attribute the positive development in these cities to DUHs activities alone.

As suggested by Dür (2008a, b) and Klüver (2013), we can provide further evidence to analyse the potential impact an NGO had on a particular event, by relying on self-assessment of the NGO and counterchecking this with the perceptions of other stakeholders involved. Based on the interviews, we can conclude that all actors confirm that DUHs actions led to a state of improved compliance with the AAQD in Germany. Firstly, DUH identifies that its legal action was effective and responsible for improved compliance in Germany. ‘We see that the legal way is successful. Legally but also in de-facto-terms’ (DUH #1). Likewise, a Commission official surmises that DUHs lawsuits might indeed have expedited the compliance in Germany (EC #1). While the IAs argue that on a case-by-case basis, the legal proceedings were often counterproductive as they tied up capacities and occasionally triggered acts of defiance, in the broader picture they agree that the pressure of DUH ultimately helped them to comply. Particularly helpful has been that DUH indirectly influenced the financial resources available for IAs to comply with this directive. DUH observed that ‘the federal government took 1.5–2 billion euros and made it available, primarily to the cities where we had sued’ (DUH #1). DUH does not seem to be overstating its self-attributed influence, because the IAs also stress that this money is attributable to DUH pressure (IA #2, 4, 5). Eventually, it was this fund that enabled the IAs to implement previously too expensive measures like the modernization of public transport (IA #2, 4).

Evaluating NGOs as new Guardians of the Treaties

Having described what strategy DUH resorted to when acting as a decentralised enforcer, and having indicated how this strategy seemed to have led to improved compliance, we now turn to the evaluation hereof. What is it that is perceived as effective (or not) in this relatively underexplored role of NGOs as new Guardians of the Treaties? What general assumptions can this single case study provide about what NGOs should and should not do in order to be able to be perceived as reliable alternatives to the centralised infringement procedure?

A first observation is that NGOs litigation strategy is not perceived as effective in all situations of non-compliance. The implementing authorities interviewed for this case study hint at the importance of identifying the underlying reason for infringements. If non-compliance is the result of lacking capacities in the form of insufficient financial resources, litigation is not perceived as helpful (IA #2, 4). Some highly effective pollution-reducing measures like the retrofitting of old diesel busses, the states would have gladly implemented (IA #4). However, as DUH confirmed, although politically not contentious, measures like this were simply too expensive for the states (DUH #1; IA #4). Another lacking capacity concerns enforcement tools tackling the emissions. The main limit value the IAs struggled to comply with was nitrogen dioxide which is mainly emitted by diesel cars. Officials, however, lament that these limits could not be met with the car fleet on the road, which drives on historically developed infrastructure designed to concentrate traffic flows. ‘It’s like you’re trying to bring the moon down to earth’ (IA #3). Thus, the officials had to comply with upper limits without having the enforcement capacities to tackle the underlying reason for exceeded limit values, the emitters, diesel cars especially (IA #1-3, 5). Since DUH could neither provide lacking financial means nor equip authorities with enforcement tools to regulate the car fleet, it was unable to help in domains where help was needed and thus not perceived as helpful by the IAs. A notable exception and vivid example for effective implementation expediting through capacity provision was a case in which DUH re-measured the emissions of allegedly low-emitting diesel busses and found out that their emissions were much higher than contractually agreed with the respective state. ‘The transport company has ripped us off. That only came out thanks to DUH’s measurements. For that I am thankful’ (IA #1).

Another example in the same vein is when non-compliance takes place because of lacking technical knowledge. Air quality policy is highly complex and requires profound technical expertise (Lenschow et al., 2017, p. 526). NGOs have been identified as having cognitive power (e.g., Boström & Hallström, 2010), as by providing relevant expertise and information they can aid compliance. This component of NGOs potential in enhancing compliance did not seem to materialise in this specific case. Although DUH did propose technical solutions suitable to reduce pollutants (DUH#1, 2), these solutions were rarely or never implemented. Sometimes because they were deemed to be ‘cookie-cutter proposals’ (IA #4), maladjusted for the local situation. This accusation is seemingly not unfounded, as DUH exclusively refers to standardized factsheets sent to IAs when asked about technical measures that they proposed (DUH #1, 2). Other times because of technical errors in DUHs proposals: ‘We talk about NOx and then they propose soot particle filters. Sorry, I’m not in the fourth grade. We laughed ourselves to death’ (IA #3). This, and the fact that the Commission officials interviewed confirm that they are not aware of a systematic lack of capacity of the German authorities in technical matters (EC #1, 2), suggests that there was simply no need for technical NGO proposals. This has remarkable parallels with Versluis and Tarr’s (2013, p. 331) observation that EU agencies are having a low impact as facilitators of compliance in countries with strong institutions because they do ‘not seem to add much to the levels of knowledge or resources present’.

DUHs strategy as a decentralised enforcer was thus not perceived as helpful by the IAs because they experienced compliance difficulties that were ‘out of their control’, i.e., the difficulties they experienced were mostly described as missing the financial resources and enforcement tools to really tackle the infringements. Yet, as also the IAs acknowledge, litigation by DUH has been described as stimulating compliance, although more indirectly. As identified by the relevant actors, what turned out to be very effective in DUHs strategy as a decentralised enforcer, was the raised issue-salience. Together with the threat of more legal action, this resulted in pressure on all governance levels to finally comply.

The main positive side-effect of DUHs litigation, as identified by the IAs, was that it strengthened their internal and external leverage. As one official put it: ‘Of course, it’s a scolding, but you have to use it skilfully to achieve internal and external goals’ (IA #1). Internally, officials report that they were able to use the pressure of DUH to assert far-reaching measures within their state. For instance, three officials report that before they were sued, they had been unable to convince the transport authorities of measures that disturb the traffic flow but reduce air pollution (IA #1-3). These inter-administrative dissensions are also highlighted by one Commission official as a factor complicating the necessary decision-making to achieve compliance (EC #2) and have been described as compliance-hindering in previous literature (Bondarouk et al., 2020). DUH observed that through their pressure, however, the balance of power shifted in favour of the proponents of ambitious air quality measures within the states, which enabled them to implement previously inconceivable measures (DUH #1, 2). The authorities confirm this perception: ‘The pressure of the lawsuit allows me to put pressure on other authorities.’ (IA #3). This is in line with Lenschow et al. (2017) who found that ‘behind closed doors, these legal proceedings are regularly welcomed by the administration – more concretely by environmental departments that previously had lost the battle over most suitable policy measures against other departments’ (p. 528).

The enormous pressure exerted by DUH strengthened the IAs’ leverage not only internally but also influenced the framing externally. Most crucially, in contestation with the federal governance level which was pressured into providing the capacities that the IAs had needed to comply (IA #1, 4). Before the lawsuits, they were unable to make their demands heard, but when DUH sued half of the country, ‘Merkel realized that things were getting serious’ (IA #1). Precedents initiated by DUH established that far-reaching measures like diesel driving bans were not only proportionate but necessary in the case of persistent limit value violations. DUH stresses that this decision became a turning point (DUH #1, 2). ‘Majorities for far-reaching measures could only be found under the sword of Damocles of the diesel driving bans’ (DUH #1). This observation is shared by the IAs, who were now equipped with a drastic but extraordinarily effective enforcement tool. ‘The highest court rulings that DUH obtained, in which it was determined which measures are proportionate, were very helpful. Because until then, many politicians claimed measures like driving bans were disproportionate or illegal’ (IA #2). The lawsuits also eased the frustration for the authorities of lacking the financial resources to finally implement effective measures and achieve compliance. ‘The federal government had to react. The states called for help saying that they can’t be left out in the cold’ (IA #2). In addition, authorities describe that DUH has also been an important balancer in external relations with other stakeholders, especially with influential German car manufacturers. Bondarouk et al. (2020) have shown that stakeholder constellations influence how extensive adopted air quality measures become. This is confirmed here, where lawsuits allowed officials to push through far-reaching measures against the resistance of industry representatives (IA #1, 2, 5). ‘Without the lawsuits, there would have been more pressure from industry opponents’ (IA #5).

All in all, DUH had managed to influence the framing and spotlight the issue of air pollution in Germany in the prior years as especially the lawsuits attracted significant media-attention (DUH #1; IA #4, 5). As a result, one official stated, many measures had already been taken in his state before DUH finally sued there as well (IA #4). The increased attention had made air pollution control an electoral issue, which motivated politicians to act (DUH #1). To conclude, the interplay between the general public pressure caused by the increased issue salience of air quality and the legal pressure in the individual cases combined is perceived as having played a crucial role in triggering compliance.

While we observe that the adversarial litigation strategy initiated by DUH had its effect – all interviewees remark that DUHs actions led to increased compliance – this strategy is also perceived to have potentially dangerous side-effects. DUHs increasingly adversarial demeanour triggered an alienation that ultimately precluded cooperation. Almost all national officials emphasise that they would have been willing to cooperate in principle (IA #1-3) or have done so with other NGOs (IA #3). However, as the first contact with DUH was often a threat of legal action by a lawyer, the relationship was perceived as charged from the beginning (IA #3, 5). DUH was recurringly criticised for putting up unrealistic demands (IA #2, 3, 5). ‘Sometimes they overshot the mark with their demands from the outset, making it difficult to work together decently’ (IA #2). Due to DUHs ‘aggressiveness’ (IA #3, 5) and ‘impatience’ (IA #5), the officials working in the ministries were soon convinced that cooperation was impossible and developed a pronounced antipathy against DUH (IA #3, 5). ‘Everyone was sick of hearing DUHs name’ (IA #5). DUH confirmed this alienation, admitting that ‘sometimes one ends up not talking to each other’ (DUH #2). The process of alienation was described particularly vividly by one official who identified a shift from cooperation to confrontation with the adoption of the AAQD. Before the adoption, DUH brought together all kinds of stakeholders to jointly fight air pollution (IA #5). However, when the AAQD enabled DUH to legally enforce the right to clean air, the conflicts emerged. DUH implicitly confirms this shift by referring to the experience that they had tried cooperation for many years but realized that matters only began to change through confrontation (DUH #1). Since NGOs have been found to rationally choose their strategy based on perceived opportunities (Rodela et al., 2017, p. 246), it appears that the AAQD and the resulting availability of coercive legal tools caused DUH to rationally shift to a seemingly more effective adversarial strategy, thereby confirming Kelemen’s (2011, p. 2012) Eurolegalism hypothesis.

Conclusion

In a European Union where compliance problems are increasingly tackled via a decentralised enforcement strategy, it is crucial to gain more insight into the working and effectiveness of this strategy. This article has provided more detailed empirical insights into how NGOs play their role in this system of decentralised enforcement, and to what extent they can effectively expedite compliance with EU directives.

DUHs efforts to enhance compliance with the AAQD in Germany, has illustrated how this NGO mostly resorted to litigation as the core strategy to push infringing German states into compliance. All actors involved in this process confirm that this strategy has been effective. This seems to indicate that NGOs can be valuable additional Guardians of the Treaties, next to the Commission. The decentralised enforcement trend as observed the last decade (Hofmann, 2018, p. 2019) can thus indeed be an effective way to tackle non-compliance at the domestic level. It also illustrates how expanded opportunities for private litigation pave the way for increasing Eurolegalism (Kelemen, 2011, p. 2012).

This case provides first clues about what makes private litigation potentially effective. Litigation seemed to have influenced the national debate, where the media attention around the court cases impacted the framing. The officials concerned with air quality consequentially gained internal as well as external leverage to push through measures that would have been infeasible before. Next to the direct threat of more legal action, also indirectly the lawsuits seemed to have released those capacities that the authorities had still lacked to comply. The precedents initiated by DUH legitimised drastic but effective measures such as diesel bans, which on the one hand provided the authorities with new enforcement tools, but on the other hand – and this was much more decisive – motivated the federal government to release extensive financial resources to prevent driving bans in the ‘country of cars’.

The evaluation of DUHs adversarial strategy by the ‘receiving end’ – the implementing authorities – illustrates that such a harsh litigation strategy can also have potential downsides. It can lead to situations of mutual mistrust and thus go at the expense of needed cooperation. This can be particularly counterproductive when the underlying reasons for non-compliance are not lifted via the court cases. In this specific case, the NGO hardly resorted to another important role as outlined in the literature, namely providing information and expertise. This appears to be partly due to the fact that German authorities have experts available, and are less in need of assistance in technical knowledge. In addition, it appears to be partly due to the fact that the confrontational relationship caused alienation between the NGO and the IAs, and this thwarted potential cooperation. We hypothesize that NGOs are most likely to have a potential positive effect on compliance by providing information and expertise in member states and/or sectors where local implementing authorities lack this, and where there is less alienation between NGO and authorities. Future research should focus on cases with profoundly different situations and see how NGOs choose their strategies and how effectively they work. Are NGOs operating in states with weaker administrative capacity perhaps more likely to opt for cooperative capacity-building activities because they can add more to the levels of knowledge or resources present?

While this case positively illustrates the potential for NGOs as additional Guardians of the Treaties, we subscribe to Hofmann’s (2019, p. 342) conclusion that ‘[p]rivate enforcement is […] not a panacea for compliance problems’, and as such it can never be a complete substitute for the centralised infringement procedure. This would require favourable legal opportunity structures in all member states, as well as the presence of well-resourced NGOs in all relevant sectors, and this situation is currently not materialized within all parts of the EU.

Acknowledgements

We would like to thank the respondents for their willingness to talk to us and help us gather the information required for this article, and we would like to thank the reviewers for their very insightful and constructive feedback that helped improve the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Konstantin Reiners

Konstantin Reiners studied the BA European Studies at Maastricht University, and is now an MA student at the MA International Security at Science Po.

Esther Versluis

Esther Versluis is Professor of European Regulatory Governance at Maastricht University.

Notes

1 The right of a party to appear and be heard before a court.

2 The interviews were conducted between April-May 2021. Names are not listed in order to guarantee anonymity, and interviewees are referred to as IA (Implementing Authorities), DUH (Deutsche Umwelthilfe) or EC (European Commission). As not many officials are primarily responsible for air quality per state, we do not reveal which five states have been selected for this research. EC officials’ statements do not reflect the position of their institution.