Strategizing in agency reform: a longitudinal case study from The Netherlands

ABSTRACT

This paper presents a longitudinal case study of a series of reforms affecting Dutch government agencies. It contributes to the developing literature on strategizing in processes of agency reform by identifying and explaining several tactics that agencies use to influence these reforms. Several previous findings on agency strategizing were replicated in the research, but the paper also presents new observations—most notably about the effect of the procedural legitimacy of a reform process and processes of learning. Furthermore, it shows that a parent ministry will also employ its own tactics to affect reforms imposed by a supranational body and will sometimes partner with agencies in doing so.

IMPACT

Insights and explanations are provided for the tactics that are used by agencies and parent ministries involved in agency reforms. Agencies are rarely passive recipients of reforms; instead, they will usually actively try to influence reform initiatives. The authors explain the conditions that make the use of a particular tactic more likely and the effect that parent ministries and agencies can have on a reform. The paper presents a repertoire of tactics that have been used. These insights will be helpful in terms of knowing how agency reforms will be influenced but will also help to limit the, often overlooked, transaction costs involved in a reform.

KEYWORDS:

• Agencies
• agency strategizing
• European Commission (EC)
• parent ministry
• reform
• tactics

Introduction

Governments across the globe have created large numbers of different types of agencies operating at arm’s length from their ministries and have charged them with policy implementation, regulation and service delivery in all policy fields. Examples include the Next Steps agencies and non-departmental public bodies (NDPBs) in the UK and Crown entities in New Zealand (Verhoest et al., 2012). Although many agencies have been subject to reforms (for example Boin et al., 2010; MacCarthaigh, 2014; Rolland & Ronnes, 2011; Flinders & Skelcher, 2012), these reforms have not always been realized. One factor affecting the success or failure of agency reforms is the strategies and tactics used by the agencies themselves to influence, resist or adapt reform initiatives. Agencies are not passive recipient of reforms; instead, they actively try to influence reform initiatives (Dommett & Skelcher, 2014; Overman et al., 2014). While still in its initial stage, a research theme is developing around the strategies and tactics that agencies use during reform processes, the factors influencing these and the conditions under which they will be successful.

We analysed various reforms of the Dutch plant health and quality inspection (PHQI) agencies implemented by the Ministry of Agriculture (MoA). This paper adds a new dimension to agency research by showing that not only agencies enact strategizing and resistance against agency reforms but that the parent ministries, too, can display similar behaviours, particularly when reforms are superimposed by a higher authority—in our case: the European Commission (EC).

The aim of this paper is to describe the different tactics used in the reform of the PHQI system by agencies and the parent ministry, and to explain them.

Several typologies have been developed to describe agencies’ strategic responses in agency reforms. Studies show that a range of strategies and more specific tactics are used by agencies to exert pressure on reforms. Some patterns have also identified. Dommett and Skelcher (2014), for example, show that agencies in the UK facing reforms typically start with a resisting response, before turning to a more conforming response. Our case study also shows this pattern.

In our study we follow Dommett and Skelcher (2014) in adopting a strategic-relational approach showing that agencies select their tactics based on the perceived opportunities and threats in the setting in which they operate. Whereas Dommett and Skelcher found that this setting also limits the legitimate range of agency’s tactics, our study highlights that the impact of procedural legitimacy on actors’ tactics also needs to be considered. Like Dommett and Skelcher, we show that agencies’ prior knowledge of a reform has an effect on the tactics they pursue. A lack of prior knowledge not only limits the time an agency has to influence a reform, it can also lead to a limited understanding of reforms and more challenging and attacking tactics.

Theoretical framework

Dependent variable: tactics

To describe and explain the strategies of parent ministries and agencies, we used Oliver’s (1991) typology that discerns five general strategies, each consisting of three more specific tactics. Oliver’s typology has been applied by Overman et al. (2014) and by Boon and Verhoest (2018) to describe organizations’ responses to reform initiatives and has proved useful in terms of understanding the assertiveness of agencies.

Not all of Oliver’s tactics were useful in our case, so we excluded five tactics. The remaining tactics are presented in Table 1 and discussed below.

Table 1. Tactics and strategic responses to institutional pressures (after Oliver, 1991, p. 152).

Strategy	Tactics	Indicated by
Acquiescence	Conform	Obeying rules and accepting norms
Defiance	Dismiss	Ignoring explicit norms and values
	Challenge	Contesting rules and requirements
	Attack	Assaulting the sources of institutional pressures
Compromise	Pacify	Placating and accommodating institutional elements
	Bargain	Negotiating with institutional stakeholders
Avoidance	Conceal	Disguising non-conformity
	Buffer	Loosening institutional attachments
Manipulate	Influence	Shaping values and criteria
	Control	Dominating institutional constituents and processes

Conformity shows an organization’s acceptance of a particular demand or expectation. In the case of dismissal, by contrast, the organization ignores expectations. The response of challenging is an active form of defiance to a demand. In this case, the organization resists an expectation in an open manner—agencies, for example, will publicly deny the validity of a particular reform. In the case of attacking, organizations even start assaulting the constituent of a demand. Organizations, however, may also try to conceal their activity through window-dressing. A variation on this tactic is when an organization attempts to create a barrier between its activities and a supervisory body: the tactic of buffering. The tactic of pacifying is where an organization shows a minor level of resistance, while devoting most of its energy to appease the constituent of a demand. Another tactic that can be used to find conciliation is the tactic of bargaining; an organization more (pro)actively aims to find a compromise by looking for concessions. Finally, organizations may try to manipulate the pressures facing them. The tactic of influencing is where an organization attempts to manage the interpretation of its activities. Organizations, alternatively, may also try to change constituents’ interpretation of the rationality or appropriateness of their own actions. The tactic of control describes the response of organizations that adopt a public and active role in shaping expectations.

Independent variables: features of the setting

To explain the use of these tactics, we follow Dommett and Skelcher (2014) in adopting a strategic-relational approach, which assumes that actors are strategic and pursue their goals in a setting that favours certain forms of strategic behaviour over others (see also Hay & Wincott, 1998). Based on the work of Oliver (1991), Overman et al. (2014) and Dommett and Skelcher (2014) it is possible to identify several features of settings that—depending on the actual threat or opportunity that actors attribute to them—have an effect on the use of certain tactics.

The occurrence of the different tactics is not predicted in detail in this study by these features. Oliver (1991) arranged the different tactics from more passive to active responses and only formulated general hypotheses about the effect of the different individual features of the setting on adopting a more or less active tactic. While we believe that the different tactics could in theory be predicted by combining the different features, we think it is more useful to use the different features of the setting as a framework for analysis. The framework we devised provided us with sufficient focus in the analysis, but was also broad enough to explain the ‘variance and complexity of our cases’ (see Scharpf, 1997, p. 29). Consequently, our explanations in this paper are partly deductive and partly inductive.

Content: An important feature is the content of a specific demand. This factor may, for example, refer to the extent to which particular reform initiatives are consistent with the goals of an agency. When an agency’s goals conflict with the reform, this will likely decrease its willingness to comply (Overman et al., 2014; Dommett & Skelcher, 2014). We assume that the same will hold for parent ministries.

Cause: An organization’s response to a particular demand is also found to be affected by the perceived legitimacy of a reform. For example if a reform is perceived as irrational or inappropriate, organizations are likely to be sceptical about the need to conform (Oliver, 1991, p. 162). Overman et al. (2014) showed that a high degree of legitimacy, on the other hand, facilitates compliance.

Constituents: An organization’s response can be affected by the number of constituents exerting expectations. Agencies, for example, have to deal with the expectations of their parent ministry but also those of different stakeholders. A parent ministry, in turn, also faces different expectations: besides expectations of agencies, it faces expectations of parliament, or, in our case, the EC’s expectations. Oliver points out that it can be difficult (or even impossible) for an organization to conform to different demands. When constituents’ expectations conflict, tactics will probably be used that grant the organization some leeway to act (Oliver, 1991). As Overman et al. (2014) point out, this will likely result in a tactic of compromise. Dommett and Skelcher (2014) show that agencies can also make use of particular constituents when there are different expectations: for example an agency may mobilize some of its stakeholders to publicly challenge a reform.

Control: Organizations’ responses will also depend on the control that other organizations have over their compliance. Constituents may actively steer the behaviour of organizations by monitoring their behaviour or using particular incentives or sanctions. Strong controls are likely to result, though not always directly, in a more conforming response.

Context: A final feature of the setting discussed by Oliver is the broader context. Based on Dommett and Skelcher (2014), we considered two dimensions of context:

• The prior knowledge that organizations have of a planned reform. Unexpected reforms were perceived by UK agencies as providing few possibilities to change an initiative and led to a more conforming response.

• The commitment of the organization that exerts a particular expectation. A strong commitment will limit the possibilities to change a reform and, thereby, the non-complying tactics that subordinate organizations can, or will, use.

The various features of the setting that affect the use of a particular tactic say little about the success of the tactic (Oliver, 1991, p. 160). Organizations may misinterpret the opportunities or threats of their setting due to inaccurate information. In some cases, this could lead to unintended—but not necessarily unwelcome—changes that alter the strategic selectivity of the setting.

Methods

Our longitudinal case study deals with the reform of the Dutch PHQI agencies over a period of almost 20 years—from 2002 to 2020. The analysis focuses specifically on the country’s initiative to delegate plant health inspection tasks (concerned with the control of plant pests and plant diseases) to four agencies that had previously only carried out plant quality inspections (concerned with the quality of plant products and production). ZBOs have legal independence and a high degree of managerial independence and are similar to NDPBs in the UK (Van Thiel, 2012).

Selecting a longitudinal case study allowed us to understand how the reform initiatives developed over time and to take into account the role of agencies and the parent ministry during different phases and in different settings (Pollitt, 2008). Outcome-oriented process tracing was used in our study (Beach & Pedersen, 2013), in which the different tactics that were employed by the ministry and its agencies were used as intermediate outcomes (our dependent variables) to disaggregate the case in smaller parts. To guide the reader through our case study, we have divided the case into five periods that describe different phases in the reform process.

We reconstructed the reform process from document analysis (although we were only given partial access to the MoA’s internal documents) and semi-structured interviews. Interviews were held with key respondents involved in the reforms to fill in gaps and collect background information. Interviews were held with four MoA officials, one official of the Dutch Permanent Representation in Brussels, an official of the Plant Protection Service (PD), a representative of one of the four ZBOs (NAK), an outside expert, a member of the Dutch parliament and a member of the European parliament from 2007 to 2009 (Zwaan, 2012). Representatives of the EC were not interviewed because the MoA officials made their participation in the research conditional on not contacting the EC about the case. The MoA feared that contacting the EC about the case would jeopardise the reform process.

Additional interviews were conducted in 2015 with three representatives of three of the four ZBOs and an MoA official by two of the authors, in the context of a commissioned research by the ZBOs and companies’ representatives (see also below).

To identify organizations’ tactics, we relied to some extent on our own observations: several tactics could be identified in the correspondence between different organizations or statements in the media. Most tactics were identified based on the interviews, by asking respondents to describe their own and others’ behaviour. The interviews also played an important role in reconstructing how the agencies and parent ministry perceived the strategic selectivity of the setting in which they operated. During the interviews, questions were asked about how respondents perceived the setting and the necessity, appropriateness and opportunities to act in a specific way. We largely focused on the content and cause of particular demands, the constituents exerting these demands and their controlling powers and the broader context of the setting in which they operated, as well as how this changed over time. Although we cannot fully rule out that some strategies were rationalized post hoc by our respondents, we tried to limit this by triangulating, as much as possible, different data sources. In addition, we gave the respondents anonymity, conducted the interviews face to face and used a semi-structured interview format that allowed the respondents to tell their own story and to ask (critical) follow-up questions. By doing so, we tried to create a reflexive environment. Coding techniques were used for the data analysis, for which the definitions of the different tactics and features of the setting were used.

Setting the scene: the Dutch plant health and quality inspection system

Traditionally, plant health (‘phytosanitary’) inspections were carried out by the plant protection service (PD). Under the authority of the MoA, the PD was responsible for implementing the EU’s plant health policy and the direct control of harmful organisms (EC, 2008). The PD operated as ‘type 1’ agency (‘agentschap’). As importing harmful organisms could close down the intensive plant sector in The Netherlands and exporting them would damage the reputation of The Netherlands, it was believed that the phytosanitary status could be best guaranteed by having inspections carried out by a semi-autonomous agency that would operate in close liaison with the MoA (MoA, 2005; see also De Vries et al., 2002).

Plant quality inspections in The Netherlands were delegated to four type 2 agencies: ZBOs (type 2 agencies have more freedom than type 1 agencies). The plant quality inspection tasks were carried out under the external control of the MoA ‘without an immediate hierarchical relationship’ (EC, 2008; Van Thiel, 2004). The agencies and their tasks are presented in table 2.

Table 2. Agricultural inspection agencies in The Netherlands.

Naktuinbouw Netherlands Inspection Services for Horticulture	Promotes and monitors the quality of products, processes and production chains of propagating related to floricultural, arboricultural and vegetable crops. Also performs research into varieties, pests and diseases. Plant health tasks include imports and plant passport checks. See https://www.naktuinbouw.com/board-structure
BKD Flower Bulb Inspection Service	Checks the quality of all flower bulb crops in The Netherlands on both quality defects and quarantine pathogens. It also performs import and export inspections and laboratory analyses. Plant health tasks include imports and plant passport checks. See https://www.bkd.eu/en/about-the-bkd/
NAK Dutch General Inspection Services	Controls the quality requirements for seed potatoes and arable crop seeds. Plant health tasks include plant passport checks and issuing phytosanitary certificates.
KCB Quality Control Bureau	Conducts import and export inspections of consumer products, such as fresh fruit and vegetables, potted plants and cut flowers. Plant health tasks include imports and plant passport checks and issuing phytosanitary certificates. See https://kcb.nl/en/over/organisatie-en-taken

The agricultural inspection agencies had been established in the 1930s by the agricultural companies to guarantee the quality of agricultural plant products that were being sold to other countries. This is still reflected in the funding of the ZBOs (fees from agricultural companies) and the composition of their boards, which include representatives of branch organizations that represent the agricultural sector. In the 1960s, the inspection bodies were granted the legal status of ZBO becoming part of the public domain when the EC started to set plant identity and quality standards for seeds and plants. It was considered a logical step to extend the tasks of the existing inspection bodies and give them the responsibility for monitoring and certifying seeds and planting material (EC, 2004; Tweede Kamer, 2009).

In practice, the division of tasks and responsibility of the different inspection agencies had become blurred. Most plants and plant products require plant quality and health controls and several health inspections and laboratory tests had been taken over by the different agricultural inspection agencies from the PD (MoA, 2005). In the seed potato sector, for example, the ZBO ‘NAK’ (Dutch General Inspection Services) was involved in phytosanitary inspections by testing for a number of diseases of seed potatoes (EC, 2006; NAK representative, personal communication, 3 April 2009). In addition, various tasks for issuing ‘plant passports’—which declare plants and plant products to be free of harmful organisms and which are needed for trade within the EU—were officially delegated from 1993 onwards to the agricultural inspection agencies by the PD (see below).

Reform process

Phase I (2002–2005): Operation Plantkeur (control)

In 2002, the PQHI system was critically assessed in a report commissioned by the MoA (De Vries et al., 2002). The report concluded that the PQHI system had developed into a ‘patchwork’; tasks and responsibilities were unclear. Moreover, the system was not very practical for companies because they often had to deal with different agricultural inspection agencies. In response, the MoA announced they would institute a more transparent inspection system in which the ‘companies should take their fullest responsibility to meet the standards’ (MoA et al., 2004). Project Plantkeur was set up to explore how the agricultural inspection agencies could collaborate better and inspection costs be reduced.

The result was another report—presented in September 2005—that suggested transferring all phytosanitary tasks to the four ZBOs under the authority, responsibility and supervision of the MoA and the PD (MoA, 2005). In both projects, the ZBOs as well as branch organizations were involved; the agricultural inspection agencies were given the opportunity to take control of the reform initiative.

Phase II (2005–2006): Maintaining international acceptance

An important concern during the reform was the need to maintain a high phytosanitary status in The Netherlands. The MoA was concerned that the use of ZBOs would be criticized (MoA, 2005; PD official, personal communication, 7 April 2009). Notwithstanding these anticipated difficulties, the MoA assessed that it could demonstrate that sufficient guarantees would be in place to uphold the good phytosanitary status of The Netherlands.

As part of the communication strategy about the reform, a meeting was set up with the EC in November 2005 to inform them about Plantkeur (MoA official, personal communication, 8 April 2009). Convincing the EC turned out more complex than expected, however, due to the results of a recent inspection mission conducted by the EC’s Food and Veterinary Office (FVO) in 2005. The FVO had visited NAK and, during that visit, it became clear that there was a close relationship between NAK and its commercial subsidiary: NAK Agro. Staff working for NAK Agro had also been involved in issuing plant passports. FVO found that (EC, 2006, p. 33):

The fact that NAK shares its staff, equipment and has financial connection with its commercial subsidiary, NAK Agro, is not fully in line with Article 2(1)(g) of Directive 2000/29, which stipulates that the ‘responsible official bodies in a member state may delegate tasks to be accomplished under their authority and supervision to any legal person, whether governed by public or private law, which under its approved constitution is charged exclusively with specific public functions, provided that such person, or its members, has no personal interest in the outcome of the measures it takes’. In this case NAK employees may have a personal interest in the outcome of the measures taken when charged with public functions to the advantage of NAK Agro.

However, the existing arrangement had been in place since the mid 1990s and had never before prompted discussions (MoA official, personal communication, 28 July 2009). In its reaction to the conclusions of the inspection report, the PD (acting on behalf of the MoA) therefore stated that ‘the sharing of staff, equipment and financial connection with its commercial subsidiary, NAK Agro is in itself not a violation of … article 2.1(g), which does not preclude such a construction’ (EC, 2006, p. 40).

Initially, the MoA dismissed the concerns by the EC: it ignored the concerns of the EC about Plantkeur, realizing that conformity to the concerns of the EC would imply an adjustment of the planned reform of the PQHI system. The content of the expectations of the EC clearly conflicted with the MoA’s goals. While the MoA realized that it was pushing the boundaries of the phytosanitary directive, it also believed that it could introduce the reform without making substantial adjustments to the NAK’s structure (MoA official, personal communication, 17 April 2007). From our interviews, moreover, it also became clear that the MoA showed little interest in the formal reasoning of the EC (PD official, personal communication, 7 April 2009). The cause for conformity with the EC demands was not accepted.

At the same time, the MoA was aware that it could not fully ignore the EC’s concerns because the EC would continue to monitor the MoA’s actions and could exert control by taking action in the case of non-conformity. The MoA, however, strongly believed in the effectiveness and feasibility of the reform and that the EC could change its view about the reform (PD official, personal communication, 7 April 2009). Put differently, the MoA believed that the EC would not remain committed to its concerns. Because of this, the MoA decided to influence the EC: during various informal face-to-face contacts it attempted to persuade the EC of the fact that Plantkeur would not cause a conflict of interests and did not breach EU legislation (MoA official, personal communication, 17 April 2007).

The EC, however, requested additional information in an official letter on the position of the NAK and NAK Agro in August 2006 (quoted in CBb, 2008). This letter made clear that the EC was taking the independence of inspections seriously and would not be easily influenced to adopt a different interpretation about the NAK’s (in)dependence (MoA official, personal communication, 28 July 2009). The MoA was thus made aware that the EC was committed to its interpretation and that it should approach the EC more cautiously on the reform (MoA official, personal communication, 17 April 2007).

In the meantime, preparations for Plantkeur continued; an agreement was signed on 15 May 2005 between the MoA, the PD, the four agricultural inspection agencies and the branch organizations, which committed the MoA to implementing the reform on 1 January 2007. Therefore conforming to the EC’s requirement to separate the tasks of NAK and NAK Agro became a difficult option for the MoA. At the same time, the MoA also realized that it would be risky to simply move ahead without informing the EC about Plantkeur and to undo the entire operation if the EC took control by starting legal action (representative NAK, personal communication, 3 April 2009; PD official, personal communication, 7 April 2009).

In a final attempt to address the EC’s concerns, while escaping full conformity, the MoA announced that it would amend the legislation concerning NAK. The amendment would indicate more explicitly that NAK was exclusively responsible for undertaking legal tasks and related specific public tasks. In addition, the MoA stressed that the inspection activities would be accredited under ISO standards (MoA, 2006) to further ensure the inspections’ independence. This attempt can be interpreted as a way of pacifying the EC’s concerns on the independence of the inspections. Some of the EC’s concerns were accommodated, without completely conforming with the EC’s expectations.

In further discussions with the EC, however, it became increasingly clear that a more formal segregation of NAK/NAK Agro personnel had to be made (MoA, 2006). The EC demanded a change in NAK’s constitution, as well as a stricter separation of the equipment and finances of NAK and NAK Agro—a point that had been stressed in the FVO report. Without doing so, the EC made very clear, it could not accept the reform and would take action.

The failed attempts to convince the EC that the Plantkeur was sound eventually made the MoA conform with the EC. The MoA realized that the EC was committed to its view and that the controlling power of the EC meant that non-compliance was not an option. To allay the concerns of the EC, NAK and NAK Agro had to be clearly split into a public body and commercial subsidiary with separate and transparent accounting for public tasks and commercial activities.

Phase III (2006–2007): Unexpected changes (challenging and attacking)

The MoA’s commitment to change NAK’s structure was not readily accepted by the four agricultural inspection agencies and caused tensions with NAK. The necessity to disentangle the public and private tasks of NAK in order to implement Plantkeur had not become clear to NAK until late 2006 (NAK representative, personal communication, 3 April 2009). The content of the changes suggested by the MoA conflicted with the interests of the agencies. Although the MoA had been taking NAK’s interests about Plantkeur into account in the discussion with the EC, NAK had not been directly involved in the discussion and was unaware of the EC’s concerns. As a result, NAK showed little understanding about the cause of these sudden changes and the requirement to make an adjustment to the statutory legislation.

The content of the changes and the limited understanding of the reasons behind them resulted in an intense discussion—behind closed doors—between NAK and the MoA (MoA official, personal communication, 17 April 2007; NAK representative, personal communication, 3 April 2009) in which the changes were challenged.

At the same time, representatives of agricultural branch organizations openly criticised (challenged and attacked) the separation, fearing that this would lead to higher inspection costs. They also questioned in a letter to the MoA distributed to the media whether the expected cost reduction resulting from Plantkeur would be realized (Anon., 2007). The pressure from NAK and the companies, however, did not change the MoA’s decision to separate NAK and NAK Agro; it would be too risky to implement operation Plantkeur without meeting the EC’s demands.

Eventually, NAK also realized it had to agree with the demand; without doing so, the MoA would simply not delegate the plant health inspection tasks (NAK representative, personal communication, 3 April 2009). The controlling powers of the MoA required the NAK to restructure in June 2007. From then on, NAK could only carry out public phytosanitary inspections and had to commit to avoiding any form (or appearance) of conflict of interest with NAK Agro (MoA, 2007). After implementing these changes, it became possible to implement Plantkeur and mandate the inspectors of NAK to carry out phytosanitary inspections as of 1 September 2007.

Phase IV (2007–2012): Changing EU legislation

While a conforming stance was initially adopted by the MoA, it remained dedicated to reforming the European PHQI system. A review of the EU plant protection system announced in the second half of 2008 (Council of the European Union, 2008) was seen as an opportunity to push for a change in the regulations, in particular with regard to a greater involvement of private sector companies. The MoA perceived this as an opportunity to take control over EU legislation. In the context of the evaluation, the MoA worked on a renewed vision for phytosanitary policy (MoA, 2009, internal document) and stressed that changes in EU legislation would have to fit and benefit the Dutch situation (Tweede Kamer, 2009).

A number of studies was commissioned by the MoA about the organization of plant inspections to provide input to the EU review study: in the review’s ‘terms of reference’ it was stated that it should be considered ‘whether or not more responsibility should be given to stakeholders’ (EC, 2009). The results of the review, however, were not fully supportive of the Dutch position. In the final report, it was concluded that ‘the majority view [among member states] has been that there is limited need or opportunity for further delegation of tasks to other bodies or legal persons’ (FCEC, 2009, p. 183). As a result, no changes were proposed by the EC with regard to the fact that delegated inspection bodies are ‘impartial and free from any conflict of interest as regards the exercise of the specific official control tasks delegated to it’ (EC, 2013, art. 26(b)(iii)).

In the face of these developments, the MoA adopted a stricter conforming stance about the sharing of responsibilities of official inspection tasks by the ZBOs. The discussion about Plantkeur with the EC, the limited willingness of other member states to change existing legislation, but also questions about the independence of the ZBOs from third countries, resulted in a more cautious position. In addition to the earlier changes that stressed the independence of NAK inspections, the MoA, together with branch organizations, decided in 2012 that the agricultural inspection agencies-inspectors carrying out phytosanitary inspections would from then on be designated as so-called ‘public officers’ mandated by the successor of the PD: the Food and Consumer Product Safety Authority (in Dutch: NVWA). Assigning the title of public officer to the inspectors of the different ZBO’s—a title used in international agreements—would further underline their independence (Tweede Kamer, 2012).

Phase V (2013–2020): Hiving in plant health inspection tasks?

The use of the title of ‘public officers’ was accepted by the four agricultural inspection agencies. In 2013, the MoA, however, took a new position and considered it necessary to give all phytosanitary inspection tasks to the NVWA (De Leeuw, 2013). This was part of a broader discussion on the role of arm’s length agencies, including wanting to reduce their numbers and make divisions between public and private tasks much clearer.

The MoA feared that concerns about the independence of NAK—despite the public officer title—would continue to fire up. According to the MoA, both the EC and other EU member countries kept pointing out that the existing PHQI system left doubts on the independence of the inspections and the mixing of public and private tasks (MoA official, personal communication, 7 March 2014; Tweede Kamer, 2015).

The MoA’s plans were not accepted by the branch organizations nor by the four agricultural inspection agencies. They argued that there had been no discussion about their independence since the designation of public officers. In their view, this change had appeased the concerns about the independence of the inspections and there was therefore no legal imperative to separate the different inspection competences and task (meeting ZBOs, branch organizations and two of the authors, 4 February 2014). The cause for this change was not accepted.

In combination with the conflicting content of these changes, this caused the agricultural inspection agencies to challenge and influence the MoA’s plans and to commission, together with the branch organizations, a study on the possible governance structures of the agricultural inspection agencies (carried out by two of the authors). The report described the pros and cons of several governance structures and was used by the branch organizations and the agricultural inspection agencies in discussions with the MoA about the reform plans (De Kruijf et al., 2014). In addition, the branch organizations lobbied members of the Dutch parliament to maintain the existing PHQI system. At the end of September 2014, a motion was proposed and approved by parliament to maintain the existing PHQI structure, while ensuring the public interest (Tweede Kamer, 2014).

The motion by parliament presented the MoA with a difficult task: accommodating the expectations of the EC and other countries with the Dutch parliament’s motion. In a response to parliament, the MoA stated that ‘the cabinet faces a number of dilemmas to develop an efficient and accepted model’ (Tweede Kamer, 2015). The MoA noted that it remained to be seen how it could balance the content of the different expectations. It would need time to figure this out, together with the branch organizations and the four agricultural inspection agencies (Tweede Kamer, 2015). In a further letter to parliament, in 2017. the MoA, however, postponed making any changes, making clear that the agencies performed well and that it would wait for the legal implementation of the EU plant health regulation and control regulation (Regulation 2016/2031) in December 2019 (Tweede Kamer, 2017).

Since then no specific reform steps have been undertaken. The reform of the PHQI system has become part of a larger transformation of the plant and veterinary inspection system, and reforming the veterinary inspection system is the priority. The tactics of the agricultural inspection agencies and branch organizations in this respect seem to be successful so far (early 2021).

Analysis and conclusion

This case analysis was aimed at improving our understanding of the role of strategizing in agency reform processes. As has been shown in other studies (Dommett & Skelcher, 2014; Overman et al., 2014), a variety of tactics can be pursued during this process. Based on the work of Oliver (1991), we distinguished several tactics that were used by the Dutch agricultural inspection agencies to resist reforms. In phase I, the four agricultural inspection agencies had some control over the reform plans. In phase III, however, they pursued the tactics of challenging and attacking before they moved to conformity when the MoA changed the statutory legislation that affected one of the agricultural inspection agencies. The tactics in this phase thereby formed a pattern that was also observed by Dommett and Skelcher (2014) in the UK: from a more resisting stance, the agencies moved towards a more conforming one.

When the MoA, in phase V, announced that the legal and organizational structure of the four agricultural inspection agencies would be changed, the agricultural inspection agencies, however, adopted a more pro-active stance, by commissioning a study into the pros and cons of various organizational structures and lobbying the parliament. So far (early 2021), these tactics of challenging and influencing seem to have been successful.

Our study shows that the threats attributed to the content of the MoA’s reform plans resulted in the agencies not passively complying with the proposed reforms. At the same time, the controlling position of the MoA made it difficult not to comply. In combination with a belief (or hope) that the MoA would not be fully committed to the content of the reforms (a context factor), this made the agencies adopt the tactic of influencing.

The belief that the MoA was able to change its position, however, turned out not to be the case during at least one of the reform phases—leading to the belief that the MoA was strongly committed to changing the reform. This resulted in a response of conformity.

At the same time, we also saw that the perceived limited procedural legitimacy of the changes caused the agencies to first pursue a number of more assertive responses: the cause for the demands of the MoA was not accepted, and combined with the threat attributed to the content of the change in the reform, this resulted in the tactics of challenging and attacking.

So far, very little attention has been paid to this procedural aspect in the literature on agency reforms and agency strategizing. We believe that doing so is important in order to understand processes of agency reform, but also to understand the relationship between a parent ministry and an agency after a reform has been implemented.

Our study also shows that the commitment of the MoA to the reform, in turn, resulted from its relationship with the EC and its conformity to the expectations of the EC.

The fact that government departments are confronted with different expectations, which limits their room for manoeuvre, has not been addressed explicitly in the literature, but it is crucial to being able to understand when particular agency tactics will be effective. Our study, however, also shows that a ministry can act strategically and pursue different tactics to address these expectations and that they will sometimes partner with their agencies. This strategizing deserves more attention.

The tactic of conformity by the MoA was not immediately used. Initially, the MoA dismissed the EC’s concerns and expectations. After (slowly) realizing that this was not an option, it attempted to actively influence and pacify the EC. In phase IV, when the EC announced a review and reform of the EU legislation, we saw that the MoA tried to take control by shaping the legislative review process.

The explanation of the tactics of influencing and eventually conforming by the MoA is similar to the explanation of these tactics by the agencies. We also see that the threats attributed to the EC’s demands and controlling position led to a situation in which neither simple conformity nor non-conformity was an option. And here we also see that the belief (or hope) that the EC would not be too strongly committed to the content of its demands was a reason to adopt the tactic of influencing. Based on these two observations, we can conclude that a conflict over the content of a reform, combined with strong control but assumed low commitment by the actor initiating the reform, are a sufficient condition for pursuing a tactic of influencing.

The range of tactics used by the parent ministry also shows that actors can misinterpret the commitment of others and that this may lead to failure. The MoA’s failed attempt to influence the EC showed the MoA that the EC was strongly committed and that conformity to the EC’s demands was the only option. In this case, the actors’ tactics resulted from a process of trial and error. This aspect of trial and error and updating of beliefs needs to be taken into account in the study of agency strategizing if we want to explain why actors adopt certain tactics and adjust these over time.

While more (comparative) research is needed to find out to what extent our findings can be generalized to other settings, our study shows that Oliver’s (1991) framework, in combination with a strategic-relational approach, has been helpful. We were able to identify and explain the range of tactics used in the face of reforms.

Further research could reveal whether there are typical sequences of tactics being used and investigate more systematically how different features of the setting combine and result in particular tactics. The tactic of influencing, which resulted from a combination of content, control and commitment, is one configuration of conditions that deserves further testing. Doing so could lead to the extension or refinement of the theory of different tactics, although there is a risk, as we have already pointed out, that we end up with a too broad typological theory of tactics that may difficult to be apply in an analysis.

Besides looking into the different factors that explain actor’s tactics, future research should, of course, focus on the effect of different tactics on the reforms taking place, but also on what impact different tactics have on the relationship between agencies and ministries after a reform has been implemented.

With this study, to conclude, we aimed to further open the black box of agency reform and the role of strategizing in this process. Several previous findings on agency strategizing were replicated in our case study, but we also made new observations, most notably about the effect of the procedural legitimacy of a reform process, the strategizing behaviour of a parent ministry and the importance of strategic learning. Our analysis also shows the need to adopt a more processual and relational understanding of how agency reforms take place. In such a perspective, it is not only the strategic behaviour of agencies during reforms that must be taken into account, but also the behaviour of other actors that have to strategically manoeuvre between different expectations.

Disclosure statement

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