Publication | U4 Issue

Reciprocity networks, service delivery, and corruption: The wantok system in Papua New Guinea

In all countries, informal systems of reciprocity (ISRs) play an important role in shaping societal norms and state functions. ISRs are informal social networks underpinned by reciprocal obligations linking families, friends, colleagues, and associates. During times of crisis these networks can provide critical support, ensuring that people have somewhere to sleep, food to eat, and access to other essential resources. However, ISRs can also present significant challenges for practitioners, particularly when public servants and other elites direct state resources to their own ISR networks and exclude people outside these cliques. In countries where the state is weak, ISRs can present an even greater dilemma, as they can undermine state laws and rules and lead to various forms of corruption that inhibit reliable and impartial service delivery. 

This U4 Issue addresses the important policy question of how practitioners can best engage with ISRs? It examines how practitioners can think about building on the strengths ISRs can bring while mitigating the threats they can pose to good governance and service delivery. Recognising the varied nature of ISRs and the contexts in which they operate is a first step in formulating integrity-strengthening interventions that can support better-governed service delivery at subnational levels. Meaningfully responding to the influence ISRs can have over public administration will require practitioners to embrace diverse policy and programmatic approaches tailored to different contexts, even within the same country.

Papua New Guinea’s (PNG) ISR is known as the wantok system. Wantok means ‘same language’ or ‘one talk’ in Tok Pisin (the country’s lingua franca) and refers to a reciprocal relationship of favours between kin and community members. As with ISRs in other contexts, the benefits of the wantok system are disputed. Some suggest that it provides important forms of social protection for citizens.988acd8d9c26 Others are concerned that it can lead to a misallocation of state resources and widespread corruption.7fca40947591 This paper provides a framework for practitioners to assess the challenges and benefits of the wantok system and, by extension, ISRs in other contexts. 

While the governance issues facing PNG have been driven by a variety of social, historical, and geographic factors,5b2504f7a92c the wantok system increasingly shapes service delivery in the country. Decentralisation reforms have placed greater resources and administrative power within subnational administrations, where the wantok system is particularly influential. We argue that practitioners need to understand and respond to the diverse ways in which the wantok system shapes public administrations. Drawing on original research by one of the authors (Walton) and on other studies conducted with citizens and public administrators in provinces across PNG, we examine the dilemmas that the wantok system has created for subnational administrations and discuss how these might be addressed.8d22b9e68499

The first section of this paper engages broader debates about the role of ISRs in shaping service delivery and good governance. It then provides a background on the wantok system in PNG and the dilemmas it presents for practitioners. Next we outline a model of analysis based on three interrelated elements. The first explains what variations in PNG’s subnational administrations mean for understanding and responding to the wantok system. The second highlights the importance of understanding how different leadership styles shape the wantok system, and the third focuses on how the wantok system impacts individuals. This three-tiered analysis helps explain how the wantok system varies between and within public administrations in PNG. This analysis does not lead to easy solutions, but it does encourage practitioners to be more strategic in deploying resources and interventions aimed at improving service delivery at the subnational level. The final section provides practical tips on how practitioners can gather meaningful information about the impact of ISRs, such as the wantok system, within public administrations. 

Informal systems of reciprocity and their potential for corruption

ISR obligations are an essential part of social interaction in all societies.b4b460757300 In his book Bribes, John Noonan3831da28b746 analysed the origins of bribery to demonstrate how, as far back as 3000 BC, not reciprocating a favour and thus violating norms around reciprocity would elicit punishment. However, the reciprocal obligations associated with ISRs are not in themselves necessarily corrupt; indeed, they often involve legally sanctioned and socially legitimate behaviour. And where the state is weak, ISRs can be an important source of resilience and social protection. In other words, by strengthening informal relationships of mutual assistance, ISRs provide important social benefits.

Nevertheless, for practitioners seeking to develop more accountable, impartial, and less corrupt public service delivery, ISRs represent a potential risk to service delivery and good governance. When ISRs dominate public administrations they can reinforce particularism, meaning that the treatment citizens receive is based on their connection to informal social networks rather than on their rights as citizens.264d2f30c1d1 This can result in rule bending, collusion, fraud, nepotism, and other practices that help redistribute resources to a public servant’s or politician’s supporters, friends, and kin.

While not all forms of corruption can be traced to ISRs, their influence on public administrations does pose a challenge. How can practitioners respond to the corruption risks ISRs can pose to the delivery of public services? As more governments around the world decentralise service delivery, answering this question becomes even more important. This is because ISR networks are often more intense at lower levels of government, where reciprocal relationships can be more important than at the central level because subnational government officials are more closely integrated within the community they serve.

Practitioners have generally responded to ISRs with rather standardised models that emphasise the role of formal rules, organisational procedures, and managerial codes in preventing corruption associated with these networks.e219f366860a These approaches, however, may have done little to curb the more pernicious behaviours associated with ISRs. Indeed, many standardised reforms have resulted in superficial policies that have failed to improve public institutions and have further undermined service delivery.099a44be2f93 In turn, many governments are still plagued by corruption and persistent institutional failure,8c7ac3a3dc62 and some analysts are concerned that decentralisation reforms have exacerbated these problems.6e31f266b44d

Understanding and responding to the challenges posed by ISRs therefore requires analysing the variety of ways they can impact different administrations. In the following analysis we focus on the wantok system – PNG’s ISR – to highlight how ISRs can shape public administration and service delivery in different ways. In doing so, we develop a framework that practitioners can draw on to understand and respond to the dilemmas associated with ISRs in PNG and other contexts. 

Background: Decentralisation in PNG

Located to the north of Australia and east of Indonesia, PNG is a nation of approximately 8.6 million citizens (Figure 1, Table 1). Part of the Pacific subregion known as Melanesia, the country gained independence from Australia’s colonial administration in 1975. PNG is a parliamentary democracy with a unicameral legislature consisting of 111 Members of Parliament (MPs) representing 89 districts (or ‘open’ electorates) and 22 provincial electorates. As in many developing countries, poor governance is a key concern for the PNG government, citizens and development partners. Transparency International’s Corruption Perceptions Index 2018 ranks PNG 138 out of 180 countries, with the country scoring 28 out of 100. PNG also rates poorly on other indices that attempt to measure levels of corruption, such as the World Bank’s Worldwide Governance Indicators.e0339aec4195077154fe1691

Figure 1: Map of Papua New Guinea’s provinces 

Map of Papua New Guinea’s provinces

Table 1: PNG by the numbers (all figures 2018 except as noted)

Population, total (millions)  8.61
Under-five mortality rate (per 1,000 live births) 53
Life expectancy at birth, total (years)  66
Primary completion rate, total (% of relevant age group)  79
Mobile cellular subscriptions (per 100 people) 49
Internet use (% of population)  11
GDP per capita, PPP (current international $)  4,299
Literacy rate, adult total (% of people ages 15 and above2010) 65

Source: World Bank (2019)

Although the country is rich in natural resources, successive governments have struggled to deliver basic services. PNG failed to achieve any of its eight Millennium Development Goals (MDGs). Indeed, while many developing countries made great strides between 2000 and 2015 (the time span covered by the MDGs), surveys have shown that, at least for some indicators of service delivery, PNG has gone backwards. In the health sector, for example, declining revenues, poor policies, and implementation failures have undermined health outcomes (Macintyre 2018). In a survey of eight of the nation’s twenty-two provinces, researchers found that the condition of health facilities declined significantly between 2002 and 2012. By 2012, there were fewer drugs, fewer patients, and fewer health workers in these facilities than there had been a decade earlier.f4c4de222452 

Recently, academics have painted a dire picture of PNG’s economic situation, noting ‘falling government revenue, large expenditure cuts to basic services, evidence of negative economic growth, and a fixed, overvalued exchange rate, supported by foreign exchange rationing’.1ecc182e8239 The public service has felt the effects of this fiscal tightening. Cash flow crises have meant that public servants are often paid late, and sometimes government departments do not receive promised funds.4d85478b4b87 

PNG’s fiscal situation has not been helped by problems associated with decentralisation2706fefe70cc reforms. Since the country’s independence in 1975, successive PNG governments have made efforts to devolve powers from the national level to subnational governments.f3e21ed3c0a7 However, many observers have criticised these reforms for inadequate funding, meagre capacity, poor oversight, and political manipulation.ab3aaa3386d9

Decentralisation has three dimensions: political, which aims to provide citizens or elected representatives with more power in decision-making; administrative, which redistributes authority, responsibility, and resources for service provision among different levels; and fiscal, which can include responsibility for revenue collection. Recent reform in PNG has focused on administrative decentralisation. The District Development Authority Act (DDA Act) of 2014 devolves administrative power to the country’s 89 districts, known as District Development Authorities or DDAs.c66442112c39 DDAs have significant resources allocated to them through the District Services Improvement Program (DSIP), an annual constituency grant introduced in 2007.f3f4c59534905c87c72954d7 According to government guidelines, DSIP funds should support projects in areas such as health, education, transport, water supply and sanitation, law and justice, and communication.e0b0ffd2debb

The effectiveness of PNG’s decentralisation reforms has been shaped by the pervasive influence of the wantok system, as we explain below. 

PNG’s wantok system

The wantok system is the most important informal system of reciprocity in the country and is a part of the everyday life of Papua New Guineans. Wantok networks vary in size and tend to overlay kin, community, and ethno-linguistic groupings. While informal, the wantok system encourages social obligations whereby those within a wantok network look after each other – for example, by providing funds, housing, or food in times of need. This mutual assistance helps build strong intra-group trust and collective identity. While it has changed over time, the system is a vestige of pre-state social organisation, when the territory now known as PNG was home to small-scale, egalitarian, and non-stratified societies that made decisions mainly by consensus.bf98e0482d31 Today the system resonates with these older forms of sociality, with wantok loyalties often stronger than loyalty to the nation-state. In practice, this means that individuals may give more weight to their wantok obligations than to state rules and laws.

The nature of the wantok system has changed over time and differs throughout space. For example, scholars note a distinction between the way the system operates in PNG’s urban and rural areas. Among urban dwellers, the wantok system has been adapted and broadened to include others from the same region of the country, even if not from the same clan. Urban residents can call on their wantok network for assistance, particularly financial help and housing.73b6bc113d01 In rural areas, where approximately 85% of the population lives, the wantok system is more often embedded in smaller familial and clan units and is rooted in communal land holdings. 

Given the weakness of the state, PNG’s wantok system provides its citizens with many benefits. For a start, it offers social protection when the state fails to supply basic social services. For example, a wantok – an individual who is a member of a specific wantok network – may call on a public official within the same network to use the office car to transport a sick relative to hospital. While the types of requests from wantoks to public servants vary, they are frequent, and refusal can fracture social ties. Insofar as the wantok system is essential for ensuring that PNG citizens have access to critical resources and support, some question whether accusations of corruption associated with this system are always appropriate.ee18faa5bd51 This is not to say that the wantok system should replace official mechanisms for distribution and use of public goods, but it is widely recognised that without it, life would be worse for many of PNG’s citizens.33e9cf921ba6 

Despite the benefits the wantok system provides, some scholars and practitioners are concerned that it is incompatible with notions of good governance.ba7fbf598e94 The reciprocity demanded by wantoks can mean a recurring dynamic of unofficial favours, where public resources are distributed unfairly and informally to the benefit of exclusive groups. So, the same official car that transports the sick relative to hospital may end up also ferrying around wantoks to parties and sporting events. This is a fairly innocuous example of how the wantok system can lead to abuses of public goods; more egregious incidents also take place. Gelu1966432ffd25 provides examples of such practices by noting that public servants often employ their wantoks outside of formal hiring processes and issue driver’s licences without tests. The system can also help perpetuate other corrupt practices such as election-related patronage and bribery.338934e1c8bf

Decentralisation reforms have put more money and power into subnational governments where – despite the weakness of the PNG state and the fiscal squeeze – relationships between citizens and the state can be more intense. These legislative changes have helped MPs solidify and expand patron-client networks that exclude those outside these cliques. In addition, they can increase the opportunities for nepotism and favouritism. Indeed, some observers suggest that the DSIP, by directing funds to district administrations often effectively controlled by MPs, has resulted in politicians increasingly channelling resources to their clan and supporters.6f65efb27289 The DDA Act8f087b992973 also provides districts with greater autonomy and resources with which to engage private contractors. This has increased the likelihood that MPs’ wantoks will benefit from contracts, to the detriment of quality service provision. Walton, Davda, and Kanaparo21469877624b find that school infrastructure built by contractors tied to politicians and officials working in DDAs was of poorer quality and more expensive than infrastructure built by community groups. 

The following sections shed further light on the challenges and benefits of the wantok system. Drawing on research conducted across PNG over the past decade, we present three levels of analysis that can help us understand the wantok system’s strengths and more problematic aspects; this approach provides a first step in developing programmes and interventions to improve governance and service delivery within subnational administrations.

Analysis level 1: The wantok system and subnational administrations

This section draws on in-depth provincial case studies to highlight the ways in which cultural, geographic, and historical factors shape the wantok system.d52fa51f028c We examine how the system operates in two provinces, East New Britain and Gulf, with very different development outcomes. We then discuss how practitioners might respond to the impact of the wantok system across subnational administrations.

Understanding the wantok system in context: Comparing ENB and Gulf

Many consider East New Britain (ENB), an island province off the northeast coast of the PNG mainland, to be one of the country’s best-performing provinces. ENB generally operates according to a formal framework of accountability and has a long history of relatively good service delivery. In contrast, Gulf province, in the country’s south, epitomises the challenges faced by most of PNG’s provinces (see Figure 2).

Comparing ENB and Gulf illustrates just how variable PNG’s subnational administrations can be. For a start, compared to Gulf, ENB has a better record of delivering government services to its constituents. A review of health and education outputs in eight provinces showed that ENB was the best performer across a number of indicators, while Gulf was one of the worst performers. In 2012, for instance, 90% of parents in ENB said that most children in the community went to school; in Gulf, only 37% said this was the case.8cf185dfb5da One study found that Gulf had the lowest level of functional literacy out of five provinces.9b422380a22e Nonetheless, Gulf and ENB implement national policies through similar administrative and political systems. Both have been executing the national government’s decentralisation reform through the DDA Act. And both are home to a variety of ethno-linguistic groups (Gulf has 24 such groups and ENB has 13), within which the wantok system is used to redistribute resources and provide social protection to fellow wantoks. These similarities beg the question: why do these two provinces experience such different development outcomes? 

Figure 2: Map of PNG showing Gulf and East New Britain provinces

Map of PNG showing Gulf and East New Britain provinces

Understanding these differences requires examining the ways state-society relations have developed over time in these two provinces, and as we shall see, the wantok system plays a central role. Historically, ENB province had, by PNG’s standards, a long engagement with modernity and the West, which helped lay the groundwork for today’s relatively well-functioning formal administration. In 1884 Germany proclaimed a protectorate over New Britain Archipelago, and plantations, particularly copra and coconut, were introduced.5fed2f8fc6d4 The province includes the town of Rabaul, which has been a centre of administration and commerce since the late 1800s, even though volcanic activity severely affected the town in 1937 and 1994. As a result of external engagement, ENB emerged from PNG’s independence in 1975 as one of the few provinces willing and mostly able to autonomously administer its own affairs. It was one of the first to have a fully elected provincial government, and later became one of only three provinces to assume full responsibility for its finances (in 2004). It has also had a level of formal education that is among the highest in PNG.a0b1324191b1 So while there is variation within the province,ff2739de0cb7 ENB has had a long familiarity with formal institutions of Western government, and many citizens living in the province have sufficient levels of education to understand how government should work. With a stronger framework of formal accountability in place, leaders are more likely to follow rules and laws in spite of pressures from their wantoks.ee46d74ad5f9 These factors set ENB apart from most other provinces in PNG.

In comparison to ENB, Gulf province faces challenges that are more significant. In much of the province, inferior land resources and remoteness have inhibited state penetration and undermined opportunities for income generation.8501b179d37e Widespread rivers and swamps hinder infrastructure development and transportation links.0cd913ced4c0 Gulf’s rich natural resources suggest a hidden potential, but the exploitation of these resources has resulted in few tangible benefits. The province was the site of the first recorded discovery of oil in the country,ed4d32d56438 and today it is traversed by both oil and gas pipelines, which provide royalties for some landowners. The gas pipeline is part of PNG’s large-scale liquified natural gas (LNG) project, which connects fields in adjoining Western province to processing plants in the bay of Gulf and the capital, Port Moresby. Another LNG project, the Elk-Antelope gas field, is set to be based in Gulf province and is in initial stages of development. Despite these projects, the region is significantly underdeveloped and suffers from poor service delivery.587fcf1139d3 In turn, formal state institutions have never been firmly consolidated in Gulf. As a result, as explained further below, the traditional influence of political and community leaders known as ‘big men’ remains strong.

ENB and Gulf’s historical and cultural differences also help explain why the two provinces have implemented decentralisation reform in very different ways. Gulf’s emphasis on ‘big man politics’ has undermined the principal-agent relations enshrined in the DDA Act and supporting policies.e58563fd373a According to the Act, citizens (as principals) are tasked with monitoring MPs (as agents). DDA regulations call for citizen participation in DDA meetings and require DDA committees to make funding decisions publicly available. Voters are – in theory at least – able to keep MPs accountable through elections every five years. Public administrators and elected local-level officials are also supposed to monitor MPs. However, research shows that in Gulf, political leaders avoided accountability and successful manipulated the citizens and administrators tasked with monitoring them. For example, in allocating DSIP funds through the DDA, MPs favoured communities that voted for them. One Gulf MP bullied and cajoled DDA members into accepting his budgetary allocations, overriding their concerns about how this funding was to be spent. Gulf MPs and district officials also failed to provide citizens with information about funding decisions and denied public access to DDA meetings where decisions about funding allocations were made.d4dc89e9722c Moreover, voters failed to punish such behaviour and sometimes actively encouraged it. This created a collective action problem, where the distinction between principal and agent is not apparent, big men hold sway, and few people trust that others will not engage in corruption and favouritism.

Findings from Gulf highlight the drawbacks of applying principal-agent frameworks when basic accountability mechanisms do not function. The DDA Act, which sets out the basic model of principal-agent accountability for PNG’s provinces, was undermined in Gulf because ties connecting MPs, administrators, and citizens through the wantok system impede the monitoring of public resources. Gulf is not the only province where this is the case. While DSIP funding is aimed at addressing poor rural service delivery, it is widely acknowledged that the distribution of these funds is vulnerable to patronage demands and has been poorly monitored.7f82e6a0a22e In Gulf and some other provinces, decentralisation reform has institutionalised, rather than undermined, patronage ties between politicians, administrators, and citizens, as MPs direct resources to their supporters. Thus, reforms designed to address PNG’s poor record of service delivery have often strengthened the most detrimental aspects of the country’s wantok system. 

In comparison, administrators and politicians in ENB mostly respected the principal-agent relationships outlined in the DDA regulations.fd2ff058773d Voters were made aware of when and where DDA meetings were to be held. Relationships between public administrators and politicians were more formalised and transparent than in Gulf.

In sum, taking a close look at the differences between ENB and Gulf shows the importance of understanding and accounting for the very different state-society relations that emanate from the fragmented processes of state formation that are apparent throughout the country. These relations – and the historical, cultural, and political factors shaping them – help determine the capacity of administrations to deliver services and implement national decentralisation policy.

Policy implications: Engaging with the wantok system in different administrational contexts

In each subnational administration, the wantok system presents different challenges that are shaped by cultural, historical and geographic factors. Responding to these will require flexibility and adaptability. Below we highlight some of the ways that practitioners might respond to the impact of the wantok system in Gulf and ENB, and places like them. 

Address collective action challenges

In places such as Gulf, citizens and officials typically do not trust those in other wantok networks will not engage in corruption. This lack of trust often results citizens being more willing to support corruption and less likely to report it. This situation represents a ‘collective action problem’, which is when corruption becomes systematic – that is all levels of society participate in corruption and principals cannot keep agents to account.

Addressing collective action problems is difficult; however, it could involve nurturing trust between different ethno-linguistic groups to help break down the bonds of patronage that can exacerbate corruption. Trust-building measures might include building roads, improving access to quality educational services and strengthening communication networks. In some cases, practitioners may need to proactively bring different ethnic groups together to shape government policy. This could mean organising participatory budgeting programmes that engage citizens in government decision-making.

This is where non-governmental and religious organisations can play a particularly important role. In PNG, more people trust the churches to address corruption than key state institutions.9c88d718b446 In turn, churches could play an important role in bringing together various communities with the aim of improving citizen engagement with the state – this might include helping communities demand accountable, impartial, and fair service delivery for all citizens. 

Demonstrate the impact of corruption on local communities

Recent research has also found that when citizens understand the impact that corruption has on their communities, they are more likely to overcome collective action problems and report corruption. In a social experiment conducted in Port Moresby with over 1,500 citizens, Peiffer and Walton915dfba0e416 found that respondents who were exposed to anti-corruption messages emphasising impacts on their own wantok group were more likely to see corruption as widespread and to have favourable attitudes about reporting corruption. This suggests that, when citizens believe corruption threatens their wantoks, they are more likely to act against corrupt practices. In turn, the wantok system could – in some situations – be drawn on in a positive way to help fight corruption. 

Strengthen formal accountability

In ENB and other districts where the administrative functions of government are relatively strong, practitioners should work to strengthen principal-agent relations. This could mean improving information to principals (for example, by publicising information about government spending) and improving mechanisms for monitoring (for example, by ensuring that communities are involved in deciding about government budgets and spending). It could also involve promoting and resourcing independent anti-corruption efforts at the subnational level – in PNG this could involve the police, the Ombudsman Commission, and the Phones Against Corruption initiative.9d89eb8c95a7

Analysis level 2: Variations of wantokism 

In this section, we look at the way PNG’s wantok system can vary by examining the difference between ‘big man’ and ‘big shot’ leadership styles. We highlight what these leadership styles can mean for service delivery and good governance. 

Big man versus big shot

The big man style of leadership has long been associated with politics in PNG, and many believe it underpins a form of patron-client relations between political elites and citizens that enables corruption.acdd396be3bf ‘Big man’ is a heuristic term for a model of leadership based on, among other things, the ‘ability to accumulate and distribute resources’.9032758bad84 Resources distributed to constituencies and wantoks often take the form of government funds that are distributed to gain big man status.fc6a757bc02c Kombako summarises the influence of localised kinship distribution on PNG’s modern state system:

When corruption permeates the State, political office becomes the most promising avenue for gaining access to resources for local ‘redistribution’. Political leaders build prestige and following through the allocation of grants, development projects, infrastructure services, and other resources to their constituencies. State ‘gifting’ becomes a means for constituting political ‘bigmanship.’3b7e8463abbf

Relations of patronage, strengthened by the wantok system, thus characterise what has come to be known as big man politics. 

Martincaffe9e71536 suggests that in recent times citizens have come to distinguish between two categories of leader: the big man and the big shot. Big man leaders are prevalent in many parts of PNG and defer to ‘traditional’ forms of governance and reciprocity through the wantok system. Big shots, by contrast, are personally successful but eschew the reciprocity associated with the wantok system in favour of ‘possessive individualism,’ which means they see themselves as owners of their own capacities and minimise the importance of social ties. While Martin’s research focuses on the rise of big shots in ENB, others have argued that this style of leader is found across the country.ee5ab78ab5e3 

Martin45344e2f172c suggests that citizens in ENB view the rise of big shots in negative terms. Many consider big shots distant from their kin in the villages and less likely to distribute resources to them than big men. In turn, Martin275f579208b7 finds that politicians and statesmen, public servants, and citizens fear being cut off from their wantoks and becoming a big shot. In other words, ignoring the connections enabled by the wantok system can result in separating oneself from the broader community, which is difficult for many who rely on informal networks for social and economic goods.

There are also no guarantees that those who become big shots will be any less corrupt than their big man counterparts. Ketan notes that, at the clan level, big men are careful with their transactions because their ‘leadership status demands transparency and accountability’261491afbb88 – although he notes that this does not necessarily translate into greater transparency within the public service. In comparison, some citizens suggest that higher levels of education – a trait often associated with big shots – can exacerbate corruption among the elite.49bc263d8f5a 

Policy implications: Responding to different types of leadership 

In responding to concerns about governance, practitioners tend to promote the very traits displayed by big shots; that is, by pushing for policies that stress impartiality, they can reward leaders who cut themselves off from their wantoks. Even in relatively strong administrations, the price paid for adhering to principles of impartiality can be steep, as Martin87cef97ebd34 has demonstrated with the case of ENB. Pushing for these traits within the public sector could lead communities to view their leaders as disconnected, unsympathetic, and uninterested in the plight of their kith and kin. This suggests that practitioners need to be careful in implementing even the most well-meaning reforms. Given this, the following suggestions provide guidelines on how to think about engaging in good governance reforms in different administrative contexts. 

Promote dialogue between leaders and citizens

In many contexts, practitioners will need to balance the advantages of pushing for impartial leadership qualities against the disadvantages of promoting leaders who are viewed as disconnected from their communities. Achieving such a balance won’t be easy; however, promoting dialogue between government and citizens could be a good place to start to break down the disconnect between them. Forums should be a site for explaining the role of government and service delivery and for receiving feedback on how government could work better. Forums should give citizens the chance to raise their own concerns. An example is PNG’s Consultative Implementation and Monitoring Council, which – notwithstanding its significant resource constraints – provides consultative mechanisms between the government, the private sector, and civil society. 

Demonstrate that impartiality leads to improved services

Citizens often support the patronage relations that underpin big man politics because it is the only way to ensure they benefit from state resources. In focus groups with citizens in remote parts of the country, many complained that resisting various types of corruption was futile because corruption often offered one of the few ways to secure benefits from the state.2ed1ab6bdf82 This suggests that the onus is on practitioners to prove that rejecting corruption – and accepting the big shot style of impartial leadership – will provide material and social benefits. Practitioners often assume that anti-corruption activities improve material outcomes, as there is much evidence, usually based on large country-scale analysis, that countries with lower levels of corruption have better resources and social outcomes.8617e48603c4 However, in PNG, particularly at the local level, this relationship still needs to be established. In turn, practitioners need to continually demonstrate that impartial behaviour actually results in improved service delivery. 

Where the big man style of leadership is entrenched, practitioners have, arguably, a more difficult job to do. They will need nuanced understandings of and responses to the strategies that individuals employ to maintain the status quo. International development agencies and practitioners increasingly understand the importance of engaging with political power. ‘Thinking and working politically,’ as practitioners have labelled this approach, will be particularly important where big man politics dominates. In practice, this means carefully and continually reviewing the changing political economy of subnational administrations to understand the ways in which leaders pander to or resist patronage networks. 

Support grassroots action

Even where big man politics dominates, it is important to seek out and engage with citizens who, in their own way, are addressing corruption and poor governance. In focus groups conducted with citizens across four provinces in PNG, respondents spoke of directly confronting and even assaulting those involved in what they thought was corrupt behaviour.e16c9979f9b1 This research shows that even where big man politics is most acute (for example, in the Highlands region of the country), some citizens are turning against corruption. This suggests that more could be done to tap into local anger about corruption that is seen to directly threaten community resources. Channelling this anger could involve non-government organisations and churches working together to build civil society coalitions for change. 

Analysis level 3: Individual responses to the wantok system

In this section we examine how public servants respond to requests from wantoks and how these relationships affect their work. Drawing on research conducted in 2018 with 136 public servants working across four provinces,4419350a48af we first examine the ways in which public servants respond to the wantok system. We then examine differences between women and men, between junior and more senior staff, and between respondents working in different provinces.

Public servants’ responses to the wantok system

Recent research with public servants shows that most find it very difficult to ignore demands from their wantoks.e13c68e3c637 Ninety percent of respondents said that they contributed to their wantoks, with half donating money (out of their own pocket), food, and other resources more than once a month. More than 70% of respondents said their wantok had asked for a favour related to their work. Yet less than one-third of all respondents believed that it is sometimes acceptable for government employees to use their positions to benefit their wantoks. So, while the pressure to contribute to wantoks is intense, most public servants acknowledged that providing unofficial favours is wrong. 

The research suggests that unofficial favours can take many different forms. Many requests involved help with administrative tasks or with obtaining employment. Typically, such help consisted of, in the words of a senior male public servant, ‘photocopying of papers, emails, typing, scanning, getting applications for jobs.’ 

Refusing such requests was very difficult, and many respondents expressed concern about how to manage their personal and professional obligations. A senior public servant from Milne Bay said: 

I am dealing with my flesh and blood but it is a challenge [because] I myself would want to maintain my leadership [position within the government]. If I were able to accept my wantoks’ [requests], everyone else would do the same. 

Indeed, many feared they would be ostracised by their wantoks if they refused requests. One junior public servant from Eastern Highlands said:

It is very difficult to turn down their favour because if I do, I will be seen as a stranger, or family or friends will keep away from me. Even in times of need, no one would want to show up. I just have to give in so nothing of this sort can happen to me. 

However, not everyone said they acquiesced to such requests; one-fifth of the 136 respondents said they refused, with some saying it was easier to refuse when one is an ‘outsider’ to the local community. A mid-level male public servant in Madang said, ‘All my family is in [another province]; it’s good, because there are no wantoks here to ask me for anything, but if [they were here], then I think it would affect my job.’ Those with experience of having worked outside of the provincial government also said they were in a better position to resist requests from wantoks. 

Variation across space, seniority, and gender

Out of the four provinces included in the survey, respondents from Eastern Highlands – a province in the Highland region of the country known for its big man politics – were most likely to receive requests for favours from their wantoks in their official capacity: 79% said their wantoks had asked for a favour. Respondents working in this province were also least likely to report corruption, given their fears about payback. However, it was in Madang, located on the northeastern side of the PNG mainland, where most respondents believed it acceptable to use public office to benefit wantoks: 46% agreed it is sometimes okay for government employees to use their position to benefit their wantok. This finding is likely explained by the province’s cultural diversity. Residents of Madang speak 175 local languages, far more than in any other province in PNG. It is likely that this ethnic diversity prompts a more intense competition for state resources, resulting in public servants being more willing to comply with these requests.4cc9c3ddc5e9

The research also found that the wantok system affected junior and senior public servants in different ways. For junior public servants, helping wantoks obtain jobs meant assisting them craft their résumés and providing advice on how to navigate the bureaucracy to secure short-term contracts and jobs. For those more senior, it meant favouring wantoks and friends for jobs and promotion to the exclusion of other, possibly better-qualified candidates. One female administrator from Eastern Highlands province said:

Favouritism and nepotism [are] becoming a very big threat. For instance, I am in the middle management position, and I would like to engage my own people, you see, without considering others… good governance in not evident in such practices.

Senior staff were better able to resist demands from wantoks; for those in junior positions, demands were sometimes relentless. A junior male employee in New Ireland said:

I am unable to complete my tasks when wantoks show up and are waiting outside the office… That leaves me in an unstable mental situation because I am caught between… serving them and my colleagues not being happy with me, and the fear of my wantoks not accepting me if I do not attend to their requests.

As a result, many junior staff found it difficult to say no to such requests.

When it came to pressure from and perceptions about their wantoks, differences between the genders were less stark than differences of seniority. Men and women were equally likely to be asked to provide favours in their official capacity, although male public servants were slightly more likely to believe it acceptable to provide favours to their wantoks. Thirty-one percent of women and 35% of men agreed, or agreed strongly, that it is sometimes acceptable for government employees to use their position to benefit their wantok. 

Differences between the genders were more apparent when it came to reporting corruption. Women were less likely to know how to report corruption (47% of women and 64% of men said they knew how to report). Even if women found out about a case of corruption, they were less likely to report it (36% of women who had discovered corruption said they reported it, compared to 51% of men). This finding resonates with a household survey conducted with over 1,800 citizens across nine provinces of the country, which also found females less likely to report corruption.9e8f463d01aa In part, the patriarchal nature of integrity organisations explains these differences. The Royal Papua New Guinea Constabulary, the country’s police force – often a first stop for those wanting to report corruption – is dominated by men, with male police officers making up around 87% of the force.76a420b70ab0 Senior public service roles are also dominated by men. Haley5351c8895e8d notes that in PNG’s public service, ‘women occupy 18% of all senior management appointments and 7% of all executive appointments.’ These factors likely dissuade women from reporting, as studies in other contexts have found women generally more likely to report corruption to other women than to men.16a590b04da6 

Policy implications: Responding to the impact of the wantok system on individual public servants

As we have stressed throughout this paper, the wantok system often plays an important role in the lives of Papua New Guineans, and that is certainly the case for the country’s public servants. Anyone working within the PNG government system will need to allow staff to respond to requests when they do not involve unofficial favours. This might mean taking time to fulfil cultural obligations, such as attending a haus krai (house of mourning), even if the time spent exceeds allocated compassionate leave. In some situations, encouraging public servants to work with their wantoks might even prove beneficial. Anecdotal evidence suggests that when people are working towards a common goal, such as nation building, the wantok system might help promote good governance because group affiliation can mean that others are held to account.50c3fc5db516 More research is required to test this claim. However, when the wantok system does result in corruption it is important to respond to the various impacts it can have on individuals. The following suggestions outline how practitioners might respond in such situations. 

Shift staff

Public servants who said they refused their wantoks’ requests said it was easier to do so when their wantok networks were located far from their job posting. Although wantok relations are fluid and can follow individuals throughout the country, moving staff between subnational governments could help free some civil servants from wantok pressures. Rotation is, however, costly and would require significant investment to be implemented effectively.747ba6ebedc6

Support resistors

In subnational administrations some individuals are better able than others to resist the pressures associated with the wantok system. Identifying and working with these particular individuals and groups will increase the chance that governance reform will be successful. In PNG, for example, senior staff are typically better able to resist wantok pressures than junior staff, which suggests that supporting and promoting these leaders could help set the tone for sections of the public service. Some have argued that supporting pro-integrity leaders can change organisational values and create environments where it is safer to challenge norms.2b4b7513053e 

Support the most vulnerable

Practitioners should also support those who find it most difficult to refuse requests for unofficial favours. In PNG, junior staff find it particularly difficult to say no to their wantoks’ pleas for assistance, which suggests that these staff need greater support from donors and the government. As suggested elsewhere, this should include targeted training about how to resist pressures, along with awareness programmes and more effective induction programmes.c8bacaccf535

Support female officials

When wantok pressures give rise to corruption, women are less likely to report it. To help women resist corruption, practitioners need to support female whistleblowers and provide more gender-sensitive pathways for reporting violations. PNG’s women are not alone in finding it difficult to report corruption. Studies have found that women in developing countries often lack the knowledge or means to report.e1565211a01a International experience suggests a host of ways that practitioners can develop more gender-sensitive responses. These include collecting and monitoring gender-disaggregated data, designing anti-corruption policies with women and men in mind, promoting women’s participation in political life, getting women involved in budgeting, and increasing the percentage of women in the public service.462eef013058 PNG would likely benefit from such approaches. Given the low levels of women in public service and in integrity organisations, there is a need to ensure that the PNG government employs more women in these sectors. While increasing the number of women might not lessen wantok pressures, it could increase the frequency with which women report corruption.7529fe96a384

Conclusions: Reflections and a toolkit for researching ISRs

All countries feature informal systems of reciprocity that can play a role in determining the distribution of state resources. In Papua New Guinea, given the fragmented and often weak nature of the state, the wantok system has significantly shaped public administration and resource allocation. This means that service delivery, particularly at the subnational level, can be particularistic, with resources funnelled to the family, friends, and supporters of bureaucrats and politicians. However, as we have shown, responding to these challenges with one-size-fits-all responses will likely cause more problems than they solve. Good governance reforms require understanding and responding to both the positive and negative aspects of the wantok system that manifest across and within subnational administrations. 

This U4 Issue has presented a tripartite model that practitioners can draw on to anticipate the problematic impacts of ISRs – in this case, the wantok system – and adopt policies accordingly. We argue that understanding the broader environment of accountability, how the wantok system is structured, and how individual public officials respond to pressures from wantok networks can help guide policy choices. While we urge caution in copy-pasting our policy recommendations into other contexts, there is scope for applying this investigatory approach elsewhere. That is, this tripartite model could be applied to understand how ISRs disrupt service delivery in a variety of settings. Of course, the results and implications will vary, but gathering information around these aspects could be important for designing meaningful responses to the challenges posed by ISRs.

We note that ISRs such as the wantok system should not be thought of in dichotomous terms: that is, they are not to be considered either good or bad. Rather, the role they can play in shaping state-society relations falls along a continuum, ranging from providing potentially lifesaving benefits, at one end, to encouraging corruption, at the other. In the former circumstances, finding ways to accommodate ISRs is wise, while the latter demands pro-active and targeted initiatives to disrupt the pernicious activities these relationships can facilitate.

Preventing the most egregious forms of corruption linked to ISRs requires responding to its individual and structural causes. In some cases this will mean creating and nurturing ‘impartiality-enhancing institutions’ that reduce the possibilities and incentives for favouritism, and encouraging officials to act in the public interest.85a640e1bf0a In other cases it will require interventions that respond to more difficult challenges, such as building trust between citizens to reduce corruption rooted in collective action problems. The broader implication is that practitioners should embrace diverse approaches even within the same country or region.

Addressing the problems associated with PNG’s wantok system and with ISRs in other contexts will not be straightforward. Indeed, as we have outlined, practitioners will undoubtedly face dilemmas and trade-offs that will need to be navigated. We hope that we have persuaded readers that the first step in addressing these dilemmas is to try to understand the role ISRs play in a given context. To this end, Table 2 provides tips on how to ask questions and collect data to reveal the challenges posed by the wantok system and other ISRs. 

Table 2: Researching ISRs: Key questions and data collection methods

Topic Key questions  Data collection methods, sources, and tips
Understanding broader state-society relations and accountability How do historical, cultural, and economic differences shape state-society accountability? How does this vary in different settings? How well are principal-agent relationships established? What undermines the formation of principal-agent relations? What role has decentralisation policy played in shaping these relations?
Understanding the social economic, and historical factors shaping state-society relations should start with a literature review. This can include literature from the academic and public policy sectors.
Again, key stakeholder interviews and focus groups can help practitioners understand the social dynamics that help shape the types of challenges associated with an ISR.  

Resources: Walton and Jones, Decentralisation and the potential for corruption in PNG (DevPolicy blog, 30 June 2017); Walton and Jones, The geographies of collective action, principal-agent theory, and potential corruption in Papua New Guinea (Development Policy Centre, 2017); Ketan, The name must not go down (Institute of Pacific Studies, 2004). 

Understanding the nature of an ISR  How do ISRs vary across time and space? To what extent do ‘big man’ or ‘big shot’ leadership systems (which may be known by various terms across societies) prevail? How do existing policies exacerbate problems associated with big man or big shot leadership? What mechanisms exist for bridging divides between the public sector and communities?  Focus groups and key stakeholder interviews are good ways to understand the leadership dynamics within different contexts. Focus groups and interviews can be conducted with public servants as well as other stakeholders, including community representatives. 

Resources: Schmeer, Stakeholder analysis guidelines (refer to this guide for information about how to identify key stakeholders).

Understanding individual and group pressures  Which public servants are most pressured by their ISRsand which are most likely to resist? What kind of requests do they receiveWhat strategies do they use to respond to these requests? How well do different individuals understand how to report corruption? Examining differences between women and men and between junior and senior staff in this respect is particularly important. Focus groups can help identify key challenges facing public servants. They can be used when the nature of the problem is unclear. 

Once the pressures on public servants have been clarifiedsurveys or key stakeholder interviews with individuals within a department can be conducted to understand the extent of the problem and who is most affected 

Observation is also an important method for understanding how individuals are pressured by their ISRs, and how they respondThis can help reveal how state resources are used and to whom they are directed. 

Resources: Springer, Haas, and PorowskiApplied policy research (Routledge, 2017); Young and ShaxsonThe research-policy connection (Overseas Development Institute, 2006); Robson, Real world research (Wiley, 2002). 

  1. E.g., Payani 2000.
  2. E.g., Dinnen 1996.
  3. The research includes interviews with administrators and politicians from provincial, district, and local-level government conducted in 2016 (Walton and Jones 2017; Walton 2019b); a 2018 survey of subnational public servants working in four provinces (Walton 2019a); and a nine-province survey on citizens’ perceptions of corruption, conducted in 2010 and 2011 (Walton and Dix 2013; Walton 2018). The research does not cover the role of wantokism in relation to administrative and political systems within the central government.
  4. Pieper 2004.
  5. 1988.
  6. Graycar and Jancsics 2017.
  7. Schweitzer 2004; Mungiu-Pippidi 2006.
  8. Clayton, Noveck, and Levi 2015; Ivanyna and Shah 2011.
  9. Pritchett, Woolcock, and Andrews 2013.
  10. Pritchett, Woolcock, and Andrews 2013.
  11. Johnson, Taxell, and Zaum 2012; Bliesemann de Guevara 2010.
  12. While there are serious questions about the ability of these national-scale indices to depict actual levels of corruption, they give a good indication of how outsiders perceive the levels of corruption within PNG.
  13. Kaufmann and Kraay 2019.
  14. Howes et al. 2014.
  15. Garrett 2016.
  16. Fox et al. 2017: 1.
  17. Kalinoe 2009.
  18. Howes et al. 2014.
  19. Decentralisation is defined as ‘the transfer of authority and responsibility for public functions from the central government to subordinate or quasi-independent government organisations or the private sector’ (Rondinelli 1999, 2).
  20. Auditor-General’s Office 2014.
  21. For the five-year period between 2007 and 2012, the government provided each of PNG’s 89 open (district) MPs a total of around 20 million kina (US$7.6 million) in DSIP funding. In the years leading up to the establishment of DDAs in 2014, these funds significantly increased. In 2013, DSIP funding reached 10 million kina (US$3.15 million) for each open MP for that year (Howes et al. 2014, p. 128).
  22. Howes et al. 2014, 128.
  23. Each DDA is overseen by a management board consisting of an MP from the district (also known as an open electorate), elected presidents of local-level government, and three other members appointed by the district MP (Department of Provincial and Local-Level Government Affairs 2015). A district administrator, also known as the chief executive officer (CEO), heads the management board. The CEO implements board decisions and is meant to provide a check on the MP. In reality, given that MPs often directly appoint them, CEOs often work as the MPs’ proxies rather than as independent administrators.
  24. While the pre-state societies of Melanesia were largely egalitarian in terms of social class, it is important to note that they were unequal in terms of gender, with women often treated as inferior to men (Gregory 2014).
  25. Levine 1999; Goddard 2005.
  26. Scholars have documented a variety of ways that the wantok system has made up for the state’s failures. Dinnen (1996) has argued that it enables communities to organise their own security in the midst of a deteriorating criminal justice system. Others point out that women can sometimes use the wantok system to their advantage. Although men are often advantaged by the system at the expense of women, McNae and Vali (2015) found that some female academics used the wantok system to gain leadership roles and to help with child rearing.
  27. Walton 2013; Rooney 2017.
  28. Dinnen 1997; Wood 2016.
  29. 2006.
  30. de Renzio 2000; Fukuyama 2008.
  31. 2017.
  32. 2014.
  33. Auditor-General’s Office 2014.
  34. Findings presented in this section are discussed at greater length by Walton (2019b) and Walton and Jones (2017).
  35. ASPBAE Australia 2011.
  36. Howes et al. 2014, ix.
  37. Walton and Jones 2017; Walton 2019b.
  38. Bray 1985; Errington and Gewertz 1993.
  39. Bray 1985.
  40. Connell 1997.
  41. Howes et al. 2014.
  42. Connell 1997.
  43. Howes et al. 2014.
  44. Connell 1997; Howes et al. 2014.
  45. Walton and Jones 2017; Walton 2019b.
  46. Principal-agent relations refers to an arrangement where one actor/group (the principal) appoints another (the agent) to act on their behalf.
  47. Wiltshire 2013; Auditor-General’s Office 2014.
  48. Walton and Jones 2017; Walton, 2019b.
  49. Walton and Dix 2013.
  50. 2019.
  51. An initiative to get public servants to report corruption over the phone; for more information, see Watson and Wiltshire (2016).
  52. Dinnen 1997, p. 191.
  53. Crocombe 2001, 412.
  54. The type of reciprocity associated with the wantok system is often related to traditional governance systems. As in other Melanesian societies (including East Timor and the Solomon Islands), clans in PNG are broadly organised around two types of governance systems: chieftain leadership, which is hereditary, and ‘big man’ leadership, which is based on ability to distribute resources. The big man style of leadership has permeated politics across the country.
  55. Kombako 2007, 24.
  56. Gregory 2014.
  57. 2007, 2010.
  58. 2010.
  59. 2010.
  60. Walton and Dix 2013.
  61. 2013, p. 5.
  62. 2010.
  63. E.g., Gupta, Davoodi, and Alonso-Terme 2002.
  64. Walton 2018.
  65. Walton 2018.
  66. Walton 2019a.
  67. Walton 2019a.
  68. Some of the literature supports this hypothesis, with Dincer (2008) finding a linear and positive relationship between ethnic fragmentation, along with religious fragmentation, and corruption.
  69. Hossain, Nyamu Musembi, and Hughes 2010.
  70. 2015.
  71. Putt et al. 2018: 66.
  72. Walton and Peiffer 2017.
  73. Brigg 2009.
  74. Fjeldstad 2003.
  75. Heywood et al. 2017.
  76. Walton 2019a.
  77. Hossain, Nyamu Musembi, and Hughes 2010.
  78. Rheinbay and Chêne 2012.
  79. Hossain, Nyamu Musembi, and Hughes 2010.
  80. Kolstad and Wiig 2009, 5321.