Value-oriented public management in vulnerable societies: assessing the impact of Public Value Management practices in three Caribbean countries

Abstract

Innovative public management practices can greatly benefit citizens, but often fail due to a mismatch with the societal context. Public Value Management was developed in North America and Western Europe, emphasising public leadership, stakeholder involvement, and information sharing. This article examines whether these practices advocated in resilient societies which are large, egalitarian, and high trust also work in vulnerable societies which are small, hierarchical, and low trust. The article compares management practices at 16 public utilities in 3 Caribbean societies, concluding that Public Value Management can improve outcomes when applied comprehensively. Partial implementation, however, leads to worse outcomes instead.

Keywords:

• small states
• public management
• governance
• Public Value Management
• public utilities
• state-owned enterprises
• Caribbean
• Aruba
• Curacao
• St Kitts

Introduction

Innovative public management practices can bring great benefits to citizens. Many new practices, however, fail to deliver on this promise due to a mismatch with the societal context of government. For example, governments replaced the old emphasis on legitimacy and control with a business-like focus on efficiency and effectiveness in the 1990s (Hood, 1991). These New Public Management practices were championed by the Organisation of Economic Co-operation and Development in North America and Europe, but also enthusiastically exported by the World Bank and International Monetary Fund across the world. New Public Management is now criticised for over-simplifying the societal expectations of government and indiscriminately promoting public management practices not suited to societies with limited state capacity and weak civic associations (Briguglio, Cordina, Farrugia, & Vella, 2009; Crook, 2010; Grindle, 2013; Manning, 2001; Stoker, 2006).

Alternative approaches to government now seem set to take over the world. The new generation of approaches address the innate ‘publicness’ of government and the multitude of stakeholders involved (Moore, 1995; Stoker, 2006). Public Value Management as conceptualised by Mark Moore, the focus of this article, urges public officials to seize the initiative in the creation of socially desirable outcomes, mobilise a multitude of external partners, and open up government to the community (Moore, 1995, 2013). Originating in North America, this alternative approach to public management is now spreading slowly but surely across the world (Alford, Douglas, Geuijen, & ‘t Hart, 2016; Williams & Shearer, 2011).

Public Value Management could repeat the mistakes of New Public Management in its spread across the world. First, although imaginatively conceptualised and thoroughly theorised, Public Value Management practices have not been extensively tested empirically (Hartley, Alford, Knies, & Douglas, 2016). Second, just as the ideas of New Public Management did not prove to be universally applicable, the practices of Public Value Management may not work in all types of societies either.

This article addresses both the need for empirical examination and cross-societal research by evaluating the impact of Public Value Management practices at 16 public utilities on the Caribbean islands of Aruba, Curacao, and St Kitts. Examining these public utilities – through stakeholder interviews and expert reviews of performance outcomes – allows us to empirically examine the impact of Public Value Management practices. Moreover, the societal setting of these utilities is radically different from the large, egalitarian, and high-trust settings such as Massachusetts (USA) or England (UK) where Public Value Management has been predominantly developed (Williams & Shearer, 2011). These three islands can be defined as vulnerable societies, as an adverse combination of small population size, hierarchism, and low trust dominates the relationship patterns within society (Briguglio et al., 2009; Commonwealth, 2007; Douglas, 2011; Oostindie & Sutton, 2006). These societies so provide a challenging setting for public management practices.

The findings suggest that Public Value Management practices can work in societies which are small, hierarchical, and low trust. However, the different practices associated with Public Value Management strongly rely on each other to create positive outcomes. The best performing utilities displayed all three management practices associated with Public Value Management, from entrepreneurial public officials and extensive stakeholder involvement to broad information sharing. The worst performers were those officials which acted as public entrepreneurs, but did not actually involve stakeholders or share information. They merely used Public Value Management as a cover for patronage and clientelism. These findings highlight both the potential use of Public Value Management beyond its initial setting and the potential dangers if Public Value Management is only applied selectively.

The promise of Public Value Management

Public Value Management was developed during the 1990s at the Harvard Kennedy School of Government by Mark Moore in reaction to the dominance of ‘business-like’ approaches to government (Benington & Moore, 2010). Public Value Management provides both a new definition for public management outcomes and a new framework for public management practices (Williams & Shearer, 2011). Moore evaluates public organisations on three dimensions. He argues that successful agencies combine (a) a strong public value proposition with (b) an operational capacity to deliver value beyond the confines of their own organisation, while (c) maintaining legitimacy amongst their stakeholders (Moore, 1995, 2013). For example, a waste management agency creates public value if the organisation is able to (a) keep the community clean and green by (b) optimising the waste management chain from the kitchen bin to the recycling plant, while (c) maintaining a high degree of support amongst politicians, citizens, and other partners. This combination of (a) public value delivery, (b) operational capacity, and (c) legitimacy provides the foundation for the evaluation of the public utilities studied here.

The main focus in this article is on the management practices making public value creation possible. There is no definitive list of Public Value Management practices. Moore urges managers to pursue a range of activities from managerial and technical interventions in their own organisation to philosophical and political work in their network environment (Moore, 2013). Given the focus on the interaction between public management and its societal setting, the three key practices highlighted here specifically address the relationships between the agency and society as envisioned by Public Value Management. The practices include the role of public leadership, the involvement of stakeholders, and information sharing. Each of these practices harbours potential pitfalls, necessitating careful empirical examinations.

Firstly, Public Value Management argues that public officials should act as public leaders with ‘restless value-seeking imaginations,’ continually labouring to define and deliver public value (Moore, 2013). They have a duty to reshape their environment to generate the greatest public value. Some scholars have countered that this gives a blank cheque to bureaucrats who want to side-step their political masters (Rhodes & Wanna, 2007). Public Value proponents respond that everyone in the public realm – whether civil servant, politician, or citizen – could act as value seekers and that even the highest public officials are subject to the restraints of their authorising environment (Alford, 2008; Benington & Moore, 2010). However, the practical effects of this license for leadership must be examined empirically, especially when looking at societies where the leadership dynamics are potentially intertwined with nepotistic relationship patterns (Baldacchino, 2001).

Secondly, Public Value Management advocates the engagement of external stakeholders (Moore & Hartley, 2008; Williams & Shearer, 2011). This is in line with a wider movement from government towards governance, where organisations are encouraged to engage with a broad regime of governing actors rather than focus on a single principal (Sorensen, 2006; Talbot, 2010). For example, a waste management agency cannot only interact with the Department of Public Works, but must also engage with a wide network of clients, environmental groups, industry experts, and regulators to define and deliver public value. Public officials should therefore play ‘an active role in steering networks of deliberation and delivery’ (Stoker, 2006). Again, this call for stakeholder involvement may have adverse effects in some societal settings, especially in communities where officials already struggle to eliminate nepotistic relationships (Piattoni, 2001).

Finally, Public Value Management advocates broad information sharing. Rather than invest in information monopolies, public officials are encouraged to initiate public debate and educate stakeholders (Moore, 1995). As Stoker (2006) observes; ‘to launch a waste recycling scheme [ … ] requires an intensive dialogue and high levels of trust between the public and authorities.’ ‘Democracy in this sense is about government by information exchange and consent’ (Hirst, 2000, p. 27). Yet transparency and openness have a mixed track-record for success and could be counterproductive in societies already cynical of their governments (Douglas & Meijer, 2016; Hood & Heald, 2006). Like the other two management practices advocated by Public Value Management, the relevance and validity of information sharing needs to be examined beyond the setting of large, egalitarian and high-trust societies.

The societal context of public management

Public management practices do not operate in a vacuum; their impact is shaped by their institutional, economic, and societal context (Putnam, Leonardi, & Nanetti, 1994). A range of classic studies have examined this interaction (Dahl, 2002; De Tocqueville, 2003; Lijphart, 1975; Putnam et al., 1994). The starting point of these studies is that the patterns of relationships dominant in a society replicate themselves into the government sphere (Field, 2003). If people are used to interacting on an equal footing, across communal divides, and in a transparent manner, they will conduct their government activities in a similar vein. If relationships are hierarchical, factional, and opaque, the public management interactions will mirror this pattern instead.

Following the work of Putnam et al. (1994) and Fukuyama (2013) on social capital and trust, three different dimensions of relationships can be said to engender effective government. First, the nature of relationships needs to be based on equality and fairness rather than hierarchy and patronage (Piattoni, 2001). Second, the structure of the relationships needs to be outwardly focused, providing bridges between different parts of the community rather than just bonds within segregated factions (Granovetter, 1973; Narayan, 1999). Third, the expectations of relationships need to be positive, with actors expecting a mutual advantage, rather than acting out of fear (Douglas, 2012; Przeworski, 1991).

With its roots in places such as Massachusetts (US) and England (UK), Public Value Management was developed in resilient societies generally displaying the nature, structure, and expectations of relationships patterns as described above, although these societies also have their particular societal challenges (Fukuyama, 2013). The assumed beneficial role of public leadership, stakeholder involvement, and information sharing in Public Value Management could be said to rely on the relationship patterns existing in these societies already. The question is whether Public Value Management works in vulnerable societies where the societal relationship patterns differ from the standard recipe for effective government.

This does not mean that we can simply select any small country as a contrast to the large societies of North America or Western Europe. The nature, structure, and expectations of relationships differ greatly amongst the global set of small countries. Societies such as Barbados, the Gambia, the Isle of Man, Malta, or Vanuatu all have very different relationship patterns. Where some small societies could be said to be fundamentally too vulnerable to create public value, others are well positioned through a strong tradition of egalitarianism and community connections (Anckar, 2002; Baldacchino, 2005; Briguglio et al., 2009; Dahl, 2002). The toughest test for Public Value Management will not be in societies which are only small, but in societies which are different on all fronts from the egalitarian, large and high-trust settings where Public Value Management was first developed.

Even then, success is not impossible. Although Putnam et al. (1994) argues that it takes decades if not generations to improve relationship patterns, the occurrence of effective agencies in otherwise weak societies suggests that value creation might still be possible (Crook, 2010; Fukuyama, 2013). Leonard (2010) argues that certain ‘pockets of effectiveness’ within vulnerable societies can overcome their disadvantageous context through factors such as technology, political economy, and helpful public management practices. Public Value Management practices might just be such a helpful public management practice.

Interactions of Public Value Management with relationship patterns

Based on the theoretical review of Public Value Management and the impact of relationship patterns within society, we can formulate several expectations about the potential impact of Public Value Management practices in small, hierarchical, and low-trust societies. The central question here is whether Public Value Management practices can be a factor in supporting the occurrence of ‘pockets of effectiveness’ or that these management practices fail to translate from the egalitarian, large, and high-trust societies where they are currently advocated.

Interaction 1: public leadership and the nature of relationships

The emphasis on active public leadership may prove to be a dangerous tactic in societies which already have hierarchical relationships, but could also provide the breakthrough required. Under the guise of public value leadership, public officials could turn the agency into their personal network of patronage, vindicating the critics of Public Value Management (Rhodes & Wanna, 2007). Alternatively, the only way to eliminate existing patronage and invigorate the public utilities might be the will and skill of entrepreneurial public officials (Crook, 2010).

Interaction 2: stakeholder involvement and the structure of relationships

The emphasis on stakeholder engagement could be a bane or a blessing as well. If relationship patterns do indeed replicate themselves across society, stakeholder engagement in an already nepotistic society could further intertwine agency governance and patronage (Piattoni, 2001). Alternatively, mobilising more stakeholders around a specific agency could prove to be an effective substitute for the civic associations lacking in these societies (Lowndes, Pratchett, & Stoker, 2006). Engaging stakeholders could so lend the agency more resources and support, aiding public officials as they steer the agency towards public value creation.

Interaction 3: information sharing and the expectations of relationships

Information sharing could also have a negative or positive impact on management outcomes. Especially in low-trust settings, openness may actually add to the confusion and cacophony of opinions. Transparency and information distribution could lead to more disorder and dissent within the governance arena (Hood & Heald, 2006). Alternatively, transparency and information may be the only way to improve the stakeholder expectations of cooperating with the government. Their expectations are probably poor already due to negative experiences in the past. Information about the potential pay-off could stimulate them to give cooperation another shot (Douglas & Meijer, 2016).

Research design and case selection

The interaction between management practices and society could be explored through various research designs. A ‘most different systems design’ comparing small versus large, egalitarian versus hierarchical, or independent versus dependent countries, could reveal the relative success of Public Value Management in these different circumstances (Gerring, 2006). This approach would require a large data set of comparable agencies in a diverse selection of countries and risks the loss of analytical depth without a qualitative focus on actual management practices. This article therefore opts for an ‘extreme or crucial case analysis,’ focusing on settings which are ‘most- or least- likely to exhibit a given outcome given their strongly deviating scores on an explanatory variable (Gerring, 2006, p. 89).

The three islands of Aruba, Curacao, and St Kitts differ greatly from the North American and Western European contexts where public value was first developed through the relationship patterns dominant in their societies. As will be discussed below, the nature, structure, expectations of relationships shaping the public sphere in these countries did not match the ‘ideal’ circumstances. A focus on these countries allows us to test whether Public Value Management is possible at all in these small, hierarchical, and low-trust settings. This research design does mean that the results are not directly applicable to other countries, but the mechanisms uncovered in these extreme circumstances are likely to occur in different contexts as well, albeit it harder to observe (Tables 1 and 2).

Table 1. Overview of selected countries and cases.

Aruba	Curacao	St Kitts
Population: 100,000 Status: Country within the Kingdom of the Netherlands	Population: 135,000 Status: Country within the Kingdom of the Netherlands	Population: 35,000 Status: Member Federation of St Kitts & Nevis
State-owned utilities:	State-owned utilities:	State-owned utilities:
Airport Seaport Bus transport Electricity and water Electricity distribution Waste management	Airport Seaport Bus transport Electricity and water Gasoline distribution Waste management	Air- and seaport Electricity Water Waste management

Table 2. Average public value score per public utility.

Case	(a) Value proposition Min. 1 point Max. 4 points	(b) Operational capacity Min. 1 point Max. 4 points	(c) Legitimacy Min. 1 point Max. 4 points	Total score Min. 3 points Max. 12 points
Waste A	2.50	4.00	3.25	9.75
Waste B	3.00	2.50	3.00	8.50
Energy A	3.00	2.25	3.00	8.25
Energy B	3.33	1.67	3.33	8.33
Seaport A	3.25	2.50	2.25	8.00
Transport A	3.00	1.50	3.00	7.50
Airport A	2.00	2.50	2.75	7.25
Energy C	2.25	2.50	2.50	7.25
Seaport B	2.00	2.25	3.00	7.25
Waste C	2.25	1.25	2.50	6.00
Transport B	2.25	1.50	2.00	5.75
Seaport C	2.50	1.25	1.75	5.50
Energy D	1.50	2.25	1.75	5.50
Energy E	2.00	2.25	1.00	5.25
Airport B	1.00	2.00	2.00	5.00
Energy F	1.25	1.25	1.50	4.00

Country selection

Although Aruba, Curacao, and St Kitts all faced substantial institutional and economic challenges, it is argued here that the chief challenges for public management stemmed from the shape of the relationship patterns dominant in their societies (Baker, 1992; Briguglio et al., 2009; Commonwealth Secretariat, 2007; Dahl, 2002; Douglas, 2012). Aruba, Curacao, and St Kitts are all small, democratic, multi-party countries, with populations ranging from 35,000 people on St Kitts to 135,000 on Curacao. Their economies are based on tourism and financial services, making them middle- to high-income countries. Aruba and Curacao are both autonomous countries within the Kingdom of the Netherlands, retaining autonomy over their internal affairs such as public utility policies. St Kitts is part of the fully sovereign federation of St Kitts and Nevis, and also fully controlled its utility policies (Oostindie & Sutton, 2006).

The vulnerability of these islands largely stemmed from the nature, structure, and expectations of relationships shaping the context of government. As already emphasised by Roe (1988), people may display very different relationship patterns when they are in a private setting compared to when they are in a public setting. In the high-risk environment of government interactions, relationships might show a more vicious edge, as seems to be the case of otherwise cordial and friendly relationships on these islands (Douglas, 2012; Marcha & Verweel, 2003).

The nature of the relationships in the public sphere on these islands was generally characterised by inequality and patronage. From a historical perspective, this inequality is the legacy of centuries of colonisation and slavery (Marcha & Verweel, 2003). From a demographic perspective, the small size of the population reinforces hierarchical relationship patterns. The elites on the islands were by necessity small, concentrating a lot of roles and power in a small number of hands (Baldacchino, 2001). One could argue that this means that a few individuals can already make a difference, but unchecked power can easily lead to excesses (Selwin, 1999).

The structure of the relationships on these islands was also problematic. The small size of the populations created an inward-looking collection of factions, rather than a network of outward-looking or bridging connections (Oostindie & Sutton, 2006). Even compared to other small nations such as the Bahamas or Barbados, the three islands had relatively few civic associations such as sport clubs or community action groups (Douglas, 2012). The small scale of the islands meant that the civic associations which were present would overlap with the personal, political, and professional spheres of government officials. Most people would therefore share strong ties of obligation with each other, rather than the loose web of weak ties associated with effective government (Granovetter, 1973; Putnam et al., 1994).

The expectations of relationships in the three societies were characterised by mistrust and zero-sum games rather than confidence and mutual gains. Taking the Corruption Perception Index as a proxy for trust, people on all three islands showed a high degree of mistrust towards government (Transparency International, 2008). These three islands scored worse than Italy, Trinidad, Tobago, and Oman. Surveys on Curacao reported that 92 per cent of people felt that the government and 91 per cent of society was affected by corruption (Oostindie & Verton, 1998). Although the societies displayed moments of collectivism, especially during national holidays such as Carnival, the assumption is that the antagonistic expectations of societal interactions will dominate interactions in the government sphere.

Case selection

This article examines the impact of Public Value Management practices in these challenging societal circumstances. The ambition is not to identify public officials who have read the work of Mark Moore; the goal is to compare utilities which do or do not apply these practices in action. A relevant comparison required a set of agencies which were similar in structure, but different in management practices. These utilities were similar in their organisational structure and faced similar technical, economic, ecological, financial, and personnel challenges. The relative autonomy of the agencies meant that the management team could choose to implement their own management practices, with some mirroring a Public Value Management approach where others opted for a different style.

The 16 public utilities were responsible for the management of the airport, sea port, bus transportation, drinking water, electricity, gasoline, or household waste. Like most places around the world, a wave of quasi-privatisation had reshuffled the relationship between the central government and the utilities, but all of them remained within the public domain (Goede, 2005). Two of the utilities were full government departments, while 13 utilities were private corporations in name but were wholly owned by the government. One utility had a private as a minority shareholder, but the government retained 80 per cent of the shares. The case selection left out the two remaining types of utilities on the islands, the telecom and postal providers, as they were unique in facing private competition or had an alternative ownership structure.

The performance of these utilities was of vital importance to the community. The utilities provided essential services, claimed a significant part of the budgets of local households, and were amongst the largest employers on the island (Martina, 2009). As part of the public sphere, and with so much at stake, it is reasonable to expect that the relationship patterns dominant in these societies would assert themselves in the management of these utilities (Goede, 2005; Martina, 2009). However, due to their semi-autonomous status, the officials of these utilities did have the liberty to shape their own management practices, constructing alternative relationship patterns around their specific agencies. Many officials had studied or worked in Western Europe and North America, bringing their experiences with them back to the islands (Douglas, 2012). As a consequence, a variety of styles could be observed, including management practices in line with Public Value Management or practices stemming from different approaches.

Assessing management outcomes and practices

In order to assess the impact of the public value practices, the management outcomes of the utilities were evaluated. This public value score was based on the three dimensions as detailed by Moore (1995): (a) value proposition, (b) operational capacity, and (c) legitimacy. This assessment was based on the judgement of experts, in line with Moore’s emphasis on evaluation through stakeholders (Moore, 2013). Local representatives of Court of Auditors and the Ombudsman were asked to rate the outcomes of the utilities on their islands, with the idea that the Auditors would rate more from a governance perspective and the Ombudsman more from a consumer perspective. Although these experts were relatively independent from the local politicians, it should be considered that their judgement might be coloured by personal connections, especially in such small communities. The researcher therefore conducted additional ratings himself per utility, based on annual reports and audit reports by external accountants. These three evaluations are admittedly still influenced by the local environment, but together they do assess the utilities from different angles and allow us to detect conflicting performance evaluations between coders.

Per utility, each expert could award between 1 and 4 points for (a) the public value proposition of the utility, (b) the operational capacity of the utility, based on their evaluation of the costs versus the quality of the service provided, and (c) the legitimacy of the utility, based on their assessment of the democratic control over the utility. Each public utility could so earn a total of 12 points. The disagreements between the local and overseas expert were limited, on average there was 0.68 confidence interval at 0.90. This means that we can expect 90 per cent of additional evaluations to return similar results.

The identification of the management practices was based on 107 semi-structured interviews with various stakeholders within and around each utility. Per utility, the respondent group included at least the senior manager in charge of the utility, one former senior manager, one supervisory board member, and the minister responsible. In addition, respondents included numerous auditors, consultants, politicians, trade union representatives, community leaders, and journalists. The interviews followed a semi-structured format, with a topic list covering the different aspects of the management practices and outcomes of each utility. The interview reports were typed out after each conversation.

The presence or absence of entrepreneurial officials, active stakeholder involvement, and information sharing was based on a qualitative assessment of the comments of respondents through open text coding. If the respondents consistently described the official in charge of the utility as entrepreneurial, able and visionary leader, than this official would classified as an active public leader. Similarly, if respondents consistently reported that a large number of stakeholders were involved in the governance of the utility – for example, frequent consultations with trade unions, the minister or community groups – a utility would be consider to actively engaging stakeholders. Finally, if the stakeholders reported that they received a lot of management information about the utility, surpassing mere consumer marketing, the utilities were deemed to actively share information.

Active public leaders

The fieldwork first reviewed which utilities had entrepreneurial public officials. Ten of the 16 utilities had public officials which possessed considerably more knowledge, skills, years of experience, and financial security than their counterparts in the regulators or political principal. These 10 officials would assert their extra resources to shape policy and the policy arena; frequently appearing on the radio and television to campaign for a policy change in their field, organising community events to celebrate their organisations, mobilising citizens to aid the utility in cleaning up the streets, pressuring politicians for better bus services, etc. Three of the utilities were not led by a strong administrative manager, but by a strong political leader dominating the agency. A final three utilities lacked an active public leader, with no individuals seizing the initiative.

The plethora of entrepreneurial leaders at 13 of the 16 cases could be explained by the small size of these communities. For example, there were very few people on the islands who could discuss electrical engineering at the same level as the public manager of the local electricity plant. With few peers, some public officials considered themselves strong leaders, even if they did not have all the relevant expertise or qualifications. One external observer commented wryly: ‘We have many dwarfs here who think they are policy giants.’

Although paying lip-service to public value or the general good, many of these public leaders did not act as the ‘rest-less value seekers’ envisioned by Moore. They adhered to the practice, but not the principle of public value, aiming to create value for themselves instead of the community. Respondents observed that several public utilities were run as private kingdoms, with managers personally dispensing favours such as jobs, contract tenders, or financial rewards. In a sign of the deep entrenchment of the hierarchical relationship patterns across society, the underlying party would often accept this behaviour. As one utility manager commented on his successor handing out jobs to allies: ‘Of course I don’t like it. But you know; it is their turn now.’ However, the interviewees also commented that some public officials would use their power for the good of the organisation and the community at large. In their view it took a strong leader to ‘sort out the mess.’

The ultimate question is whether the presence of active public leaders coincided with better public value outcomes. Interestingly, active public leaders were present in both the best and worst performing agencies. The top performers (Waste A, Waste B, etc.) and bottom performers (Energy F, Energy E, etc.) all were led by a public official who commanded significantly more skills, expertise, or finances than their stakeholders. The presence of strong individuals at all value-creating utilities underlines their necessity for creating value in small societies, but their presence in the worst performers also signals that they could destroy public value if left unchecked.

Active stakeholder engagement

The fieldwork then explored to what extent the 16 public utilities would actively engage a multitude of stakeholders. Due to the small size of the communities, many of the key stakeholders were already connected. For example, the minister and public manager would not only meet each other at the formal shareholders meeting, but might also be members of the same political party, sports club, or even family. This web of connections combining formal and informal relationships could lead to a messy policy field with a range of actors jousting for their own private interests.

Public officials responded to this challenge in one of two ways. In the first approach, they would seek to limit the amount of stakeholders involved. They could do this through a strict interpretation of the laws and regulations governing the utilities, using the nominally private status of these utilities as a tool to keep the public out. One public manager commented: ‘There is no such thing as a public utility on this island. We are a private company whose owner happens to be the government.’ Officials with these convictions would restrict their engagement of stakeholders to the bare minimum.

In a second approach, in line with the practices of Public Value Management, senior officials would actively engage a multitude of stakeholders. At Waste agency A, for example, the manager would frequently consult with his minister, but also with other ministries, the Chamber of Commerce, and community groups. He even sponsored the founding of an environmental group to get more attention for the ecological dimension of waste management. Importantly, stakeholder involvement was not a free-for-all. The officials would structure the stakeholder field, using company statutes to appoint experts to official advisory boards or formalising interactions with community groups by holding regular meetings on specified issues.

The question is again which type of practice produced the best results. Possibly, keeping out stakeholders is the only way to avoid nepotism. Alternatively, a connection with the wider society might be a necessity in this unstable environment. The analysis showed that the high performers engaged more stakeholders than the poor performers. For example, the top five utilities gave a major role not only to the minister and public manager, but also involved trade unions, investors, community action groups, and business groups. By contrast, the worst performers would often use legal barriers to keep out external stakeholders. The utilities so used formal guidelines to establish a helpful relationship pattern.

Active information sharing

The fieldwork finally explored how active the public utilities were in sharing information with the community. Respondents reported that a lot of different forms of information were mixed in with the public debate. For example, the public debate about a price hike of the electricity tariffs were as much about the rising oil prices as about the expensive new car of the wife of the manager of the electricity plant. The inward-looking associations and low-trust attitudes result in frequent contradictions and confusion. One official concluded that it was impossible to get a message across in this environment; ‘Even if you get it right yourself, there is always someone at your back who will completely twist your stories. People hear all these different versions and do not know what to believe anymore.’

Several public officials stopped sharing information with the public. ‘I tried explaining my policy to these people, but they just will not understand and get upset. Now I just keep a low profile.’ Yet these public officials continued to broadcast information: ‘The management of perception is the only lever they can really control. Those that have done this well were guaranteed a stable position.’ Several public officials were actively and constantly broadcasting information – going on radio shows, speaking to community groups, producing strategy papers – aiming to reach everyone from the minister to the people in the street.

The question is again which practices were observed amongst the high public value utilities. Without an exception, all the high performers had public officials who actively and consistently broadcasted large amounts of information. Where the underperforming public officials would not share information, the performing agencies would keep sharing information. Even though the extra information might result in a new round of gossip and disputes, the public officials at the successful utilities kept broadcasting information.

Analysis

The findings on separate practices can be brought together into one Venn-diagram to highlight the joint occurrence of different practices (Douglas, 2011, 2012). Looking at Figure 1, it becomes apparent that the top performers (Waste A, Waste B, Energy A, Seaport A, Energy B, Airport A) applied all three practices of Public Value Management. Each of these utilities demonstrated strong public leadership, stakeholder engagement, and information sharing. One middle performer (Seaport B) applied stakeholder engagement and information distribution, but lacked an entrepreneurial leader. Two middle performers (Waste C, Energy E) applied only stakeholder engagement. Three poor performers (Energy C, Transport A, Energy D) applied none of the public value practices. Interestingly, the worst performers (Seaport C, Transport B, Airport B, Energy F) also did not engage stakeholders or share information, but did have strong public leaders.

Figure 1. Management practices per case, incl. rank in value production in diamonds.

Before exploring how the public value practices may have helped the top performers escape the restraints of adverse relationship patterns, it is important to emphasise that seven agencies did not escape these limitations. The relationships around these utilities corresponded to the dominant pattern of patronage, factionism, and distrust. All of these utilities (Energy C, Transport A, Energy D, Seaport C, Transport B, Airport B, Energy F) had strong leaders who acted more like patrons dishing out private favours than public value seekers. It did not matter whether these leaders were elected politicians or appointed public officials; the governance dynamic reflected the relationship patterns prevalent in the public domain. The failure of this group of utilities would seem to support the criticism that Public Value Management only provides cover for power-hungry officials and that public management practices developed in North America or Western Europe lack global relevance. However, the fact that the most successful utilities also had strong leaders suggests that there is a more complicated mechanism at work and that this practice could still improve outcomes.

Closer analysis suggests that the adverse relationship patterns cannot be broken by applying a single practice of Public Value Management. The most successful utilities did not cherry-pick within the practices, but committed to the full set of strong active public leadership, active stakeholder engagement, and active information sharing. The public utilities which only had entrepreneurial leadership, and nothing else, scored the worst of the whole sample. Cases which only displayed stakeholder engagement or information sharing fared slightly better, but still lacked the momentum to truly excel. Only those agencies which combined stakeholder engagement and information distribution with active public leadership truly brought value to their community.

Conclusion

The findings provide a new perspective on the match between innovative public management practices and the challenges of small scale, hierarchical, and low-trust societies. Approaches such as Public Value Management can improve the government outcomes in these societies, although each of the practices involved requires an important qualifier in order to match the societal context. These qualifiers are detailed by revisiting the interactions between the three components of Public Value Management and the three dimensions of relationship patterns.

Interaction 1: public leadership and the nature of relationships

Both the public officials of the very best and very worst utilities overshadowed their surroundings in terms of initiative, experience, and expertise. It may seem paradoxical that agencies working within societies already plagued by hierarchical patronage structures need yet more strong leaders to create public value (Baldacchino, 2001, 2012). However, officials at the high-performing agencies combined this leadership with engaging stakeholders and spreading information, whereas the active public leaders at the worst-performing utilities preferred to work alone and in opaque manners. The successful officials were using their position of power to educate and mobilise other actors, creating a network of supporters and civic relationships otherwise lacking in these societies (Piattoni, 2001). Public officials should indeed not get a blank check, as argued by opponents of Public Value Management (Rhodes & Wanna, 2007), but should be charged with building a strong stakeholder network around the organisation to strengthen the partnerships and accountability structures (Talbot, 2010).

Interaction 2: stakeholder involvement and the structure of relationships

All successful utilities invested time in engaging stakeholders, confirming the importance of stakeholder engagement propagated by Public Value Management. However, without a tradition of constructive relationships within in society, the agencies needed to define anew what it means to interact with each other in an equal but professional manner (Putnam et al., 1994). The high-performing utilities used corporate governance rules to generate roles for the external stakeholders, creating several supervisory bodies or advisory networks, rather than using roles to keep people out. This formalisation of the relationship structures helped utilities to avoid the opaque patronage networks prevalent in society, while still bolstering their legitimacy (Douglas, 2012). This practice signals that the emphasis of Public Value Management on stakeholder engagement does also apply in these types of societies, but that it requires work to build the right structure of relationships.

Interaction 3: information sharing and the expectations of relationships

Information sharing is also confirmed as a prerequisite for creating public value. Value creation relies on the response of external stakeholders to a call to create value together, yet as citizens in these societies have negative expectations of government relationships, they are not likely to invest their time. For the agency to succeed, citizens need to be convinced anew that there is mutual gain in working together (Przeworski, 1991). The problem is that the officials cannot wash away centuries of bad experiences overnight. Instead of actual positive outcomes or good experiences, the only thing utilities can offer is information about how it might be different this time round. As the officials share information about their plans, processes, and decision-making, they convince citizens to give it another shot. Step by step, as the agency starts to improve, they validate this trust. Society may still be characterised by distrust of government on the whole, but the performing utilities may have convinced at least their circle of stakeholders to suspend their disbelief.

On the whole, the practices championed by Public Value Management are observed to work in small, hierarchical, and low-trust societies. However, each practice comes with a qualifier signalling its reliance on a comprehensively reviewed and implemented change of public management. Strong entrepreneurial leaders can work in these hierarchical societies, but only if they use their power to build strong accountability networks. Active stakeholder management can work in nepotistic settings, but only if formal rules and regulations are used to reset otherwise clientelist relationships. Active information sharing address mistrust, but only if combined with strong leadership and involvement actually delivering results. When the qualifying measures are absent, outcomes do not improve ore even deteriorate. These findings give hope to troubled societies seeking new approaches to government, but also caution that selective copy-pasting from overseas could lead to worse results.

Discussion

The results give a first indication of how Public Value Management practices play out in small, hierarchical, and low-trust societies. However, several reservations about the analysis and the outcomes are in order. First, as stated before, this article is not meant as a full-scale evaluation of the principles of public value, but as an assessment of the impact of some of the practices associated with this new approach to public management. The conclusions are therefore aimed at some techniques of Public Value Management, rather than the concept as a whole. Second, it only offers a limited insight in the direction of the mechanism; the performing agencies might be able to display Public Value Management practices only because they were successful to begin with, although the reports of the respondents suggest the reverse. Finally, this article takes a largely qualitative approach. The rating of the utilities is clearly structured and the 107 interviews provide a rich source of information, but the analysis ultimately relies on qualitative judgements. Further research efforts could test these results through quantitative analyses, involving a wider selection of countries.

The relevance of the findings can be specified for the three different groups of countries. In these three Caribbean islands, and possibly in countries very much like them, Public Value practices seem to work if they are applied comprehensively. It should be remembered that these islands had challenging relationship patterns, but that the relatively strong institutional and economic conditions may have supported the Public Value practices by providing the judicial framework or basic education of stakeholders. Speaking for small, hierarchical, and low-trust societies more generally, the outcomes do suggest that a joint intervention in leadership, stakeholder involvement, and information sharing could unlock value creation, although the precise mix and nature of interventions will depend on the specific societal context. For all societies struggling, this article does confirm that public agencies often fall victim to the adverse relationship patterns within society, but also that agencies and officials can nurture new types of relationships around them and create value for the community.

Acknowledgements

This article is based on the fieldwork conducted by Scott Douglas as part of his Ph.D. for the University of Oxford, supervised by Professor Christopher Hood. The author would like to thank the editors and the anonymous reviewers for their many incisive comments on earlier drafts of this article, greatly helping to enhance its quality and clarity.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Part of the fieldwork was financed through a grant by Alexander & Simon and United Trust.