[Return to SSGM Home] [Return to SSGM Working Papers Page]

State, Society and Governance in Melanesia Project
Australian National University


 

ISOLATION, VULNERABILITY, AND GOVERNANCE: REFLECTIONS ON POVERTY ASSESSMENTS IN PAPUA NEW GUINEA, FIJI AND VANUATU

by

Maev O'Collins
Professor Emeritus, University of Papua New Guinea

Visiting Fellow, Political and Social Change
RSPAS, Australian National University.*



ABSTRACT
This paper will consider some of the issues involved in poverty assessment in Papua New Guinea, Fiji, Vanuatu and other island nations in the South Pacific. In particular, it will consider how isolation and vulnerability, and efficiency and effectiveness of government services, are related to the increase and persistence of poverty among vulnerable groups. These issues provide a focus for reflections on the approaches to poverty assessment used in the World Bank sponsored 1996/97 Poverty Assessment for Papua New Guinea, the UNDP/Government of Fiji 1997 Fiji Poverty Report, and the United Nations/Republic of Vanuatu 1996 report on Sustainable Human Development in Vanuatu.
 

Introduction: Poverty - a multi-faceted concept
 

Poverty has many faces. It is much more than low income. It also reflects poor health and education, deprivation in knowledge and communication, inability to exercise human and political rights and the absence of dignity, confidence and self-respect. ... Behind these faces of poverty lies the grim reality of desperate lives without choices and, often, governments that lack the capacity to govern (Human Development Report, UNDP 1997:iii). Poverty indeed has many faces. Many of these faces remain unrecognised or out of focus as they are linked to other aspects of deprivation, such as isolation and vulnerability. The assessment of poverty and the development of programmes to alleviate poverty require a multidimensional approach in which access to cash income or equivalent economic benefits is only part of the solution. Political and social factors may negatively impact on individuals, households or whole communities and limit their access to productive resources and services. Physical and social isolation may, in turn, negatively impact on health and educational status and make it increasingly difficult for poor people to take advantage of poverty alleviation strategies.
 

Interlocking and interdependent aspects of disadvantage create a 'deprivation trap', which serves to create and maintain poverty. As Chambers (1983:112) states:
 

Poverty contributes to physical weakness through lack of food, small bodies, malnutrition leading to low immune responses to infections, and inability to reach or pay for health services; to isolation because of the inability to pay the cost of schooling, to buy a radio or a bicycle, to afford to travel to look for work, or to live near the village centre or a main road; to vulnerability through lack of assets to pay large expenses or to meet contingencies; and to powerlessness because lack of wealth goes with low status: the poor have no voice. Describing himself (ibid. viii), as 'a puzzled and uneasy middle class Englishman who has lived and worked in Africa and Asia', Chambers was not specifically considering poverty in the South Pacific. Understandably, observers of development in the South Pacific during the early 1980s often viewed this description of poverty as not representative of the 'pockets of poverty' to be found in Papua New Guinea, Fiji or Vanuatu. In part, this reluctance to accept 'poverty' as a problem was because international definitions of poverty tended to emphasise cash income levels and poverty lines. As Bryant (1993:13) pointed out:
  Obviously there are problems with these definitions, particularly in establishing poverty levels in societies, which may still be partially or wholly subsistence. Poverty lines are also used where levels of income are measured and below a certain figure people may not be considered poor. Where a cash income is the only possible income such a line may be a useful assessment but in countries where there are an increasing number of people involved in informal sector employment, as well as multiple employment and subsistence farming for home consumption, poverty is difficult to assess. This difficulty in measuring the extent and seriousness of poverty does not mean that it is not a reality for many individuals, families and communities. During the 1980s and early 1990s, many 'puzzled and uneasy' observers noted the increasing problem of pockets of poverty in otherwise non-poor communities, and a marked increase in the number of households living in absolute poverty.1 They shared Bryant's conclusion that poverty was becoming a significant feature of life in many villages and towns.
  Both absolute and relative poverty exist in the Pacific. There are households which are absolutely poor and which do not have enough food or clean water, or access to a decent house or to basic education. There are others who live in relative poverty in that they do not have access to the basic needs, which a society may consider to be 'normal'. In the smaller countries of the South Pacific, increased poverty has often accompanied the shift to a more Western-style cash economy. Rural-urban migration has resulted in unplanned settlements where those with little or no cash income have joined the ranks of the urban poor. The situation does not equal the scale or depth of absolute destitution, which exists in many countries in Africa and Asia. Nonetheless, an increasing proportion of urban and rural households, and individuals within households, have very limited access to basic needs and services.
 

One explanation for the continuing reluctance to address poverty as a serious and pervasive problem, is the strongly held belief that families and communities are willing and able to look after their own. Yet, traditions of ceremonial gift-giving and reciprocal exchange imply the existence of a similar resource base. Where some members of the group have moved further into dependence on the cash economy than others, or where the cultural safety net has broken, this situation may no longer exist. After a review of the literature on safety nets in Papua New Guinea (Gustafsson 1997:32), noted that, on the basis of the research into poverty which had been conducted:
 

It is obvious that poverty does exist, even though it has not taken on such proportions as in Africa and Asia. Poverty in Papua New Guinea does not necessarily mean that people lack food, [but] a person who has no network of social relationships, preferably kin, might feel poor. The conclusion of this review was that:
  For the majority, there is no immediate risk of poverty, yet there are certain groups that are more exposed than others, such as the old, the sick, the handicapped, and the unemployed with no direct access to land in the home village (pp. 33-34). In addition to the 'at risk' groups identified in this review, there are other disadvantaged communities who have been persistently by-passed and are, in a very real sense, out-of-sight. For some, their geographic remoteness and minimal scattered populations make the delivery of services or economic development programmes difficult and unsustainable. For other communities, or households, the problem may be that of political or social isolation. This is because patterns of clan loyalties and reciprocal obligations are now operating most effectively for those who have greater access to the wider economy, and/or to political and administrative patronage. And, for urban employed relatives, the burden of providing a safety net for impoverished kin may itself be a factor, which places them 'at risk'. 2
 

It is clear that earlier conclusions as to the adequacy of safety nets are now out-of-date. Over the past two years Papua New Guinea has experienced an economic downturn, exacerbated by political instability, natural disasters, and a general deterioration in administrative capacity. These problems have seriously affected marginal households and communities already on the edge of poverty. The emergence of poverty in hitherto self-sufficient communities is related to the general deterioration of government services and national transport and communication infrastructure, and the inequitable distribution of development benefits. The lack of capacity of governments to govern means that assistance is not available to enable households and communities to improve their participation in the cash economy. The trap closes even more tightly as health and education services deteriorate. Law and order problems, particularly (but not exclusively) a problem with unemployed urban youth, may also threaten national and community stability.
 

In a speech during a conference on 'Australia-Papua New Guinea: Crime and the Bilateral Relationship', the Australian Minister for Foreign Affairs pointed to the links between effective governance, social order and poverty alleviation (Downer 1998:6). He stated that this meant that the need to strengthen institutional capacity has become an important focus of Australian assistance to Papua New Guinea.
 

While the media tends to focus on the more dramatic aspects of lawlessness, such as robbery, rape and other forms of violent crime, we also need to examine the more subtle problems which can arise when institutions do not function as well as they might.
 

Good governance fundamentally underpins poverty reduction and economic growth. Good governance means the effective management of a country's resources in ways that are open, transparent, accountable, equitable and responsive to people's needs. The rule of law is an essential component of this and the Australian Government has made assistance in this area a priority. In recognition of this key role, the programme to improve governance and civil society in Papua New Guinea has been expanding in recent years.

 The inescapable conclusion is that, unless and until the capacity of governments to govern is improved, any positive development obtained through government, bilateral, NGO or United Nations poverty alleviation programmes will be unsustainable. This paper will focus on the research and poverty assessments undertaken in Papua New Guinea, Fiji and Vanuatu. However, it should be stressed that the findings have much wider relevance, and reflect the international, as well as regional and national, faces of poverty.
 

The United Nations and Poverty Assessment
Reflecting a growing concern about the increase and persistence of poverty in Third World countries, the World Bank's 1990 World Development Report ushered in what has been termed the Decade of Poverty Assessment. By the mid-1990s, more than sixty national poverty assessments had been conducted in Third World countries. The persistence of poverty became a matter of increasing attention as the United Nations General Assembly proclaimed 1996 as the International Year for the Eradication of Poverty.
 

At the same time, other UN agencies were concerned with the identification of vulnerable groups at risk of becoming poor. Particular emphasis was placed on the problems experienced by women and children, as they were often the most disadvantaged members of low-income households (see UNICEF/Government of Vanuatu 1991 and UNICEF/Government of Papua New Guinea 1996). In addition to the poverty line/cash income approach, it was clear that a broader poverty assessment index was needed to identify vulnerable groups and individuals within households and communities. The United Nations Development Programme (UNDP) developed and refined a human development index. This index combined life expectancy, adult literacy, and real per capita income and aimed to measure relative poverty and vulnerability. Noting that: 'The quality of people's lives can be poor even in the midst of plenty' (Human Development Report 1996:43), UNDP broadened its development focus to include gender and participation measures, with emphasis on sustainability and assistance to vulnerable groups.
 

One might well question why the World Bank, UNICEF and UNDP studies and surveys of poverty and sustainable development proceeded in such splendid isolation from each other. After all, the problem had been recognised by all three agencies as being of utmost importance, and many of their findings were of a similar nature. In some situations, the same group of researchers and consultants were involved in surveys and reports sponsored by different UN agencies.
 

Examples of converging, but still divergent, attitudes are reflected in the different approaches to non-government organisations (NGOs) by the World Bank, UNICEF and UNDP. The information publication on partnership between the World Bank and NGOs, stated (World Bank 1996:1) that:
 

NGOs and community-based organizations (CBOs), often have closest contact with the poor, are best able to help them directly, and are well suited to helping them identify their most pressing concerns and needs. Their local knowledge and expertise and their ability to foster and promote people's participation thus give them strong comparative advantages and can make them valuable and experienced allies. Working with NGOs in Bank-financed operations can increase project reach and sustainability, provide the Bank with alternative perspectives, and facilitate consultation with local people. The foreword to this policy paper noted that 'building relationships is a long and complex process'. This is certainly true, but for UNDP, UNICEF and other United Nations agencies with somewhat different development orientations, the journey to closer partnership had begun earlier, and had been easier and less tension-filled.
 

A major source of continuing tensions has been the debate on the relative merits of quantitative economic measurements of poverty levels versus more qualitative and anecdotal indicators. At the heart of all these debates are differences in the way in which poverty is defined and in the appropriate unit of analysis. There are also problems in reconciling and integrating 'hard' data such as cash income and access to productive resources, and 'soft' data such as access to the psychological support of kin and community and the value of a sense of personal security. As one study published by the World Bank points out (Carvalho and White 1997:5):
 

The definition of poverty typically adopted under the qualitative approach involves a broader conception of poverty and deprivation than does the definition typically adopted under the quantitative approach. The qualitative approach defines poverty so as to capture the processes and interactions between social, cultural, political, and economic factors. It includes a wider range of factors such as vulnerability, isolation, powerlessness, survival, personal dignity, security, self-respect, basic needs, and ownership of assets than does the definition of poverty under the quantitative approach. This study ends on a very positive note, pointing to changes in attitudes to the 'correct' methods to be used in poverty assessment. However, separate approaches have not been completely eliminated. Often, qualitative data is seen only in terms of 'enriching' quantitative material and, even where statistical measures may be based on dubious assumptions, the fact that they are quantifiable appears to give them greater validity. Income and consumption data inevitably appear more reliable, when compared to qualitative assessments of 'vulnerability' or 'gender empowerment'.
 

Problems inherent in relying upon quantitative data obtained in earlier surveys are highlighted in the poverty assessments under discussion. In some cases, sampling inadequacies or logistic problems faced by interviewers tended to exclude remote marginal communities. In other situations household surveys were unable to identify poor individuals in non-poor households. Different assumptions regarding sampling or non-sampling errors, and reliance on official attendance records or individual recall, may also lead to widely differing results. 3
 

It has taken some time for the elements of the 'deprivation trap' to become mainstream issues in the poverty assessment debate. This delay may have benefited smaller South Pacific nations as it is now more acceptable to assess poverty by taking into account the different realities facing individuals or groups. And, while some analysts may still consider it necessary to warn that isolation and vulnerability should not be equated with poverty, few would deny that they are closely inter-related. 4
 

Another major question, which has been of concern to United Nations agencies from the beginning of the poverty debate, is the relationship between poverty and government capabilities. The theme of the 1997 World Development Report was 'The State in a Changing World'. The focus of this report was on the lack of government capabilities, ineffective service delivery, corruption, and government mismanagement. Pointing to the negative consequences on human lives and social and economic development, it stressed the need to reinvigorate state institutions. An implicit assumption was that the institutions of nation states had been vigorous and effective at some stage in the past. Yet, the very fragility and lack of vigour of many of these institutions may be related to the inappropriate and unwieldy nature of their colonial beginnings. The process of adaptation and evolution to more appropriate structures, which will better serve these new nation states, is a continuing process. While few would argue with the call for 'regulatory simplification and public enterprise reform', there must be some concern regarding the current emphases on privatisation and liberalisation, and the need for a contraction of the role of the state. For countries such as Papua New Guinea and Vanuatu, where economies of scale and private enterprise are inadequate responses to the needs of small, scattered, and remote communities, these may be premature and defeatist responses.
 

Despite statements in other World Bank publications, calling for a more sensitive and integrated economic and social approach to development, there are hardline rational undertones in the assertion (p. 75) that strategic options must focus on what is workable.
 

The policies which rely on these approaches may not be first-best policies in a textbook sense. But as state capability grows, countries can switch to more flexible tools, capable of squeezing out further efficiency gains. Throughout, states must maintain the confidence of firms and citizens that flexibility will not be accompanied by arbitrary behaviour - else the foundation for development crumbles. While the need to focus on the workable is understandable, some serious questions remain as to whether the apparent persistence of poverty will lead, not to alleviation programmes, but to an acceptance of increasing poverty as an intractable, if unfortunate, reality.
 

Poverty Assessment: Politics, Process and Preliminary Findings
It is important to stress the preliminary nature of even the most definitive of poverty assessments. This is an ongoing situation so the findings cannot be seen as absolute. Rather, this is part of the overall process of situating the poor within a particular society during a particular time period. Not surprisingly, the poverty and related human development assessments carried out in Papua New Guinea, Fiji, and Vanuatu reflect their different political circumstances, and the relationships between national governments and particular international funding agencies. The following summaries illustrate these differences and note some of the findings.
 

Papua New Guinea

The World Bank sponsored Papua New Guinea Poverty Assessment Project was initiated in partnership with the Government of Papua New Guinea in 1996. A number of national and international institutions and organisations were involved in the process. There were six areas of enquiry with a national household survey as the major component. Both quantitative and qualitative data were obtained from a number of additional surveys and analyses of secondary material. The interrelationships between poverty, health, education and geographic remoteness, and the role of NGOs in the identification and alleviation of poverty, complemented the major surveys.
 

The source of the following outline of the poverty assessment process is the status report presented at a seminar at the Australian National University on 17th February 1997 (World Bank 1997c).
 

1. Household Survey. This included analyses of data collected in six urban areas of Papua New Guinea for the 1985-89 Urban Household Expenditure Survey as well as data from the 1996 Household Survey, which surveyed 1,400 rural and urban households (Gibson 1996c and Gibson and Rozelle 1997). Additional working papers on 'boy-girl discrimination' and 'women's education and child growth' were based on data available from the Urban Household Expenditure Survey (Gibson 1996a and 1996b).
 

2. Poverty, Nutrition and Health Care. A rapid assessment of four types of poor communities included urban settlements, rural cash-cropping settlers and rural subsistence communities (Jenkins 1996).
 

3. Poverty and Education. A rapid assessment was undertaken of five different types of poor communities (Guy, Tawaiyole, Khambu and Avei 1996).
 

4. Poverty and Agriculture. This study was based on a long-term study of six of the poorest agricultural systems in the country (Allen and Bourke 1997a).
 

5. NGO Involvement in Rural Service Delivery. The report on NGO activities was collaboratively produced by twenty-five NGOs, and Interagency groups. It included findings from two forums, and case studies which illustrated the most promising NGO strategies (Education Development Center 1996).
 

6. Social Safety Net. This report involved case studies and a literature review of formal and informal support systems (Guy 1997, Gustafsson 1997).
 

The February 1997 status report noted that five seminars had been held in Papua New Guinea involving government officials, particularly from the Department of Finance and the National Planning Office, other departmental and provincial administrators and NGOs. Seminars were also held in Washington and Canberra. A National Seminar was held in Port Moresby (27-28 February 1997). The final World Bank report was scheduled for completion later in 1997.
 

However, events did not proceed as planned, and there have been continuing delays in the release of a final poverty assessment report. As a distant observer it is impossible to be certain why these delays have occurred. One explanation might be that the World Bank was not satisfied with the adequacy of the final report or that other material needed to be obtained to ensure its completeness. Another explanation relates to the already noted problem of integrating quantitative and qualitative data into a coherent assessment.
 

During 1997, the Government of Papua New Guinea and the World Bank were involved in continuing negotiations over the basis for further loans, and the conditions to be met if loan moneys were to be released. At a distance, it is not clear how far these problems had a spin-off consequence for the poverty assessment programme. By late 1998, the final report had still not been released and, although many of the papers and reports are now public documents, others remain in the limbo world of 'not for citation'.
 

Nonetheless, despite these political dimensions, it is still possible to draw conclusions from the quantitative and qualitative data available as to the degree and severity of poverty in Papua New Guinea. A survey of the Papua New Guinea economy (AusAID 1997:13-14) noted an increase in both poverty and inequality.
 

Preliminary findings from the recent assessment of poverty in Papua New Guinea (World Bank 1997[d]) indicate that the distribution of consumption is more unequal than in neighbouring countries - the richest 10 per cent of the population accounts for 36 per cent of consumption; the poorest 50 per cent for only 19 per cent of the total. ...Approximately 20 per cent of the population have a consumption level that is less than the threshold cost of achieving a basic standard of living. In addition, consumption levels have fallen in most cases; '... the poor in Port Moresby have become worse off than they were 10 years ago' (World Bank 1997[d], p. 9). A study of the relationship between poverty and agriculture found that many households in rural areas had very low cash incomes. some less than K50 a year. Despite the uncertain status and incompleteness of the Papua New Guinea Poverty Assessment Program, many of the findings contained in its different components are now accessible to government and non-government planners. What is really lacking is the political will to continue the process, as well as to ensure that any proposed poverty alleviation programmes are appropriately targeted and sustainable.
 

Fiji

In 1991, the Government of Fiji established a National Poverty Task-Force The 1996 Fiji Poverty Study also reflects the collaborative action of international and national agencies and institutions, and a number of individuals working in the field of poverty alleviation.
 

The study was commissioned by the Government of Fiji, with financial and technical assistance from the United Nations Development Programme (UNDP) through its Pacific Regional Equitable and Sustainable Development Programme. This was part of UNDP's Poverty Strategies Initiative, launched in conjunction with the 1996 International Year of Poverty Eradication. In addition to material obtained from existing secondary sources, the poverty assessment process included the following components.
 

1. Household and Expenditure Survey. A comparative analysis of the 1990/91 Bureau of Statistics Household and Expenditure Survey (See Ahlburg 1996).
 

2. Expert research papers and case studies. These were prepared by number of consultants with long-term experience in Fiji.
 

3. Case studies of people in poverty. These were obtained from welfare officers from the Department of Social Welfare and some of the 68 NGOs, listed as involved with poor and disadvantaged groups (Annex 3:130-147). Other input was provided by members of the Government/NGO Fiji Poverty Study Technical Working Group.
 

4. Participatory Needs assessment and subjective poverty line survey. These surveys and interviews by teams of research workers aimed at enhancing the participation of the poor and ensuring that their perspectives were included in the report.
 

5. Poverty Eradication Strategies. The analysis of different strategies to alleviate poverty involved a survey of 68 NGOs involved with poor and disadvantaged groups (Annex 3:130-147) and output from a workshop 'Brainstorming on Poverty Eradication Strategies' jointly organised by the Central Planning Office and UNDP/ESHDP in July 1996.
 

One of the main findings (Report pp. 38-39 and Ahlburg 14-15) was that between 23-25 per cent of the population reported income, which was below the poverty line. At the lower income level, between 10-12 per cent of households reported that their income in 1990-91 was insufficient to cover even basic food costs. Ahlburg points out that:
 

This was a significant increase over estimates for poverty in the late 1970s. However, a sizeable part of the increase in poverty was due to an upgrading of the minimum nutritionally adequate diet. Slow economic growth and rising food prices also contributed to a rise in poverty. (See Table 1) The composition of minimum nutritious diets took into account cultural differences between Fijian and Indo-Fijian households (Annex 2:127-129). An important part of the assessment was to identify ethnic patterns of poverty, which might otherwise be masked when using national or urban/rural averages. National averages suggest that Indo-Fijians have a higher average household income than Fijians. However, when income distribution is broken down into ten percentile categories (Report p. 26) a different result is obtained.
  It shows that, relative to Indo-Fijians, there are fewer very poor or very rich Fijian households, and that Fijians predominate in the middle income groups. Thus the higher average incomes for Indo-Fijian households reflect the high income of the highest income groups. Among the poor, Indo-Fijian households predominate.
 

However, other findings also point to the linked and widening gap between urban and rural areas, noting that: 'These geographic differences outscale ethnic differences and contribute to them (Ibid. p. 27). (See Table 2)

 The survey was undertaken at a time of relative peace and harmony in Fiji. The acceptance of a new Constitution and strategies to develop greater cooperation in national economic development also made it easier for material to be presented which reflected the major ethnic divisions in Fiji society. It draws attention to the vulnerable in society and the process of impoverishment. In doing so, it also attempts to link qualitative profiles of poverty with the quantitative material from the national Household Income and Expenditure Surveys.
 

The strength of this report is in this integrated focus, which provides data to support the conclusion that poverty is indeed a problem, and profiles of vulnerable 'at risk' individuals and households. Its description of the process of impoverishment and some of the strategies which could be undertaken to alleviate poverty also have particular relevance for other countries in the Region.
 

Vanuatu

The report on 'Sustainable Human Development in Vanuatu' was prepared as a joint undertaking between a number of United Nations agencies, the National Planning Office and other government and non-government agencies. It was financed and technically supported by the UNDP/UNOPS Pacific and Regional Equitable and Sustainable Human Development Programme. It relied heavily upon existing statistical source material, secondary analysis, and some case studies undertaken by researchers and NGOs.
 

Quantitative and qualitative data, which formed the basis of the report, included the following components:
 

1. Analysis of existing data. This included the 1988 National Population Census, the 1993 Demographic and Migration Analysis, the 1994 National Agriculture Census, preliminary data from the 1995 National Employment Survey, and other National Planning Office, Ministry of Education and Health Department surveys.
 

2. Literature survey. This included information in research reports, surveys, and policy documents prepared by and government departments, UN agencies, other aid donors, NGOs and individual researchers (for example, the 1991 UNICEF/Government of Vanuatu Situation Analysis of Children and Women in Vanuatu, low-income urban settlement data in the 1994 ADB/Ministry of Transport Report on Urban Infrastructure, and the 1994 AIDAB sponsored study of the economy by J. Fallon.).
 

3. Case studies. These included a study of an urban settlement (Bong 1995), a report on smallscale development programmes run by the Foundation for the Peoples of the South Pacific, and other examples provided by the National Council of Women, and members of VANGO (Vanuatu Non-Government Organisations).
 

4. Meetings and Workshops. Insights were provided by the members of the national core team, described as representative of government, NGO and donor interests.
 

The Report notes that the national core team met regularly over a twelve month period, and team members were involved in the collection and consideration of data. The final report was the qualitative of the three national assessments, with only a partial focus on poverty and inequality. This approach may reflect a need to 'tread softly' due to the reluctance of some ni-Vanuatu decision-makers to accept that 'poverty' is a growing problem. However, agreement was reached (p.99) that there were two types of poverty: those who are poor for structural reasons, and those who are 'at risk' of becoming poor in times of crisis. In Vanuatu, as in Papua New Guinea, geographic remoteness and isolation also means that smaller communities are finding it increasing difficult to maintain any form of economic and social access to the wider society. It was interesting to find that this inability to mobilise social and economic resources was noted (p.100), but not really explored in detail.
 

All three studies concluded that there was some degree of poverty, which needed to be addressed structurally and socially. In Papua New Guinea, the persistence of disadvantage and lack of access to services and resources were highlighted, and will be discussed in more detail later. The conclusion in the Fiji Poverty Report was definite and clear-cut and accepted that poverty was indeed a problem:
 

The safety nets available to many low-income people in Fiji do not fully bridge the gap between an adequate lifestyle and destitution. In Fiji there is a strong sense of community and kinship. Yet the poorest households tend to be small and isolated; family systems do not encompass everyone, nor does their support necessarily last as long as the need for it does. Financial support systems exist but they often do not protect the most vulnerable. Government provides many forms of assistance to low-income earners but direct welfare is available only to the poorest of the poor and provides even them little relief from severe hardship. Non-government organisation run many programmes to assist the poor, yet acknowledge that poor coordination and their often limited out-reach prevent them from fully meeting the needs of the poor in Fiji (Report, p.99). The threat of increased poverty in Vanuatu seems to echo the situation in Papua New Guinea and Fiji a decade or so ago and this may explain the ambivalent reaction to poverty and vulnerability. Yet, while preferring to focus on the more acceptable (and perhaps more manageable) questions of 'sustainable development', there was a recognition (p.19) that many households were experiencing hardship.
  Many families are experiencing great difficulty fulfilling their customary expectations today and low-income families in urban areas for example have very little to spare. ... Anecdotal evidence shows that for reasons of choice and necessity, some families are starting to minimise their contributions to kinsfolk, and others are already opting out of the family system altogether. It is clear that the subsistence support systems which have customarily linked the urban and rural families are under threat and some are being undermined. Despite the differences in focus and emphasis, the interlocking variables of isolation, vulnerability and governance and, explicitly or implicitly, their relationship to poverty were major themes for all three countries.
 

Isolation and vulnerability
In their study of poor rural communities in Papua New Guinea, Allen and Bourke (1997a) found that many development projects initiated in these areas had left almost no trace. The lack of cash incomes and deterioration in government services had further compounded the situation of disadvantage experienced by these communities. They reported (p.11) that:
 

...relative isolation from places where cash can be earned, primarily large and intermediate urban centres and mining enclaves, is associated with low incomes, and hence with low growth rates in children. (See Table 3) Isolation and lack of access to services and markets may be related to physical remoteness in inland areas or to climatic conditions which hamper transport to and from smaller islands. The costs, and in some cases the security risks, involved in travel to and from main centres, mean that already infrequent government services may cease altogether and the marketing of local produce is no longer commercially viable. For isolated communities, the presence, or absence, of regular radio communication is another important factor in the level of their connectedness with the wider community.
 

In some areas, the deterioration of existing infrastructure (roads, bridges, wharves etc) leads to a general impoverishment of whole communities. Lack of reliable access to markets mean that it is a waste of energy to prepare produce, just in the hope that transportation will become available. The problem is compounded when the occasional ship does call and commercial shipowners complain that there is not enough produce to make regular scheduling of visits worthwhile. 5
 

Scattered communities living in remote inland regions, or on small islands far from administrative centres, may never, or only very occasionally, be visited by health personnel. There are similar difficulties with regard to educational services. Even where primary schools have been established, there are shortages of teachers, school supplies and supervision from educational authorities.
 

The Vanuatu study pointed out (United Nations/GOV 1996:100) that: 'Provinces furthest from the capital of Port Vila record lower education, health and income levels than the more urban areas'. Evidence of the problems associated with maintaining adequate education services in remote communities was reflected in an analysis of the Radio Vanuatu Message Service (ADB 1992 Annex C.3-1). Several messages were sent to teachers who had not returned to their schools after the long Christmas vacation. The problems of holiday travel were two-edged. Teachers from remote areas often experienced delays and lack of sea or air transport, both when returning home, and when it was time to return to their assigned schools. The remoteness and difficulties associated with some of these postings also meant that staff were less willing to continue at a particular school, even if it meant facing some form of penalty, including loss of employment. These same problems of isolation and transport difficulties also help to explain why many health, education and other government workers are unwilling to conduct regular patrols or to be posted to more remote areas. 6
 

On the other hand, regular and reliable communication can be a positive factor, which mitigates against the tyranny of distance. The economic and social benefits to marginal families and communities provide the impetus for self-help and family or community economic independence.
 

The content of messages broadcast over Radio Vanuatu in 1991/92 reflected a wide variety of individual, family, community and national concerns and activities. Information regarding illness, death, and family and community conflicts were followed by items regarding business activities (special sales, requests for certain produce to be shipped to Vila or Luganville, etc), church celebrations, or visits from government field officers. Health supplies, schoolbooks, water tanks and livestock were to be delivered to particular health centres or schools. Passengers were to travel to Vila or be collected at their destinations. Tardy employees were threatened with 'strong action' if they do not return from their holidays. Teachers and other government workers were informed of their new postings.
 

Similar message services have been provided in Papua New Guinea. As part of a system which can reach out to isolated communities and improve their general standard of living, this service depends for its success on the availability of road, sea, or air transport for people, goods and services. The system also depends upon the maintenance of radio communications, the quality of reception and the availability of radio receivers in remote communities. In Vanuatu and Papua New Guinea communication and transport services to remote communities have deteriorated, while at the same time the costs of any remaining services have increased. This means that poorer communities are increasingly disadvantaged and that formal and informal support systems are less effective.
 

At the time of my visit in 1992, the lack of even minimum shipping services, and the costs of air travel and small boat transport, impacted on every aspect of life in the remote Torres Islands.7 The high cost of air freight made ordering of basic commodities difficult and some items could only be transported by ship. Government workers and their families experienced even greater hardship and social isolation. Trade stores were empty so rice, sugar, soap, batteries for radios or other items were often not available. Those from outside the community did not have easy access to garden produce or bush materials. Follow-up visits and regular contacts regarding rural water supply or other projects were also difficult to arrange. This meant that the few government workers stationed on the Torres had only their own strengths and initiatives to assist them, as they were often without external encouragement for long periods of time.
 

These problems of economic and social isolation and lack of government services are illustrative of the situations, which exist in many small communities throughout the Region. In these circumstances, stated national goals of sustainable human development may sound very hollow indeed.
 

In addition to spatial isolation, the Fiji Poverty Report points out that poor individuals, households and whole communities may be marginalised and isolated from the developments occurring within the wider community. For example, the Solomon Islands settlement in Suva contains some of the poorest households in Fiji but is subsumed under 'Other' in most statistical summaries. In these situations, small ethnic minorities may find that they are overlooked or are included with other more financially and socially advantaged groups. This social and psychological isolation is related, but not always the same as, economic isolation. Nevertheless, these problems have economic, as well as social, consequences and serve to create and maintain poverty.
 

In some circumstances, isolation may be a conscious decision of individuals or groups who are seeking to avoid negative official attention. Small unplanned settlements are established in out-of-sight locations so that government officials will be less likely to order their removal. However, this also means that in these generally very low income households, services such as clean water or sanitation will not be available and that women and children will be less likely to have access to adequate health and education services.
 

Social isolation may also be related to the fear or reality of domestic violence, abuse and neglect. Low-income migrant families may experience high levels of stress and may not have access to supportive kin and community networks. In these situations, women, children, the elderly and the handicapped may be most at risk of violence or neglect. The fear of violent reactions may also deter women from participating in community activities A study of the situation of women and children in Papua New Guinea reports (UNICEF/GOPNG 1996: 143) that:
 

Anecdotal evidence suggests that some women are reluctant to become actively involved in programmes and activities, and are even prevented from attending women's meetings, because of their husband's objections. Personal security fears also serve to limit the participation of women in activities outside the home. In some areas of Papua New Guinea, there has been a marked increase in reports of sexual attacks, including gang rapes. Newspaper reports of these crimes are often accompanied by police advice to women to be more circumspect, and not to move around the community unless accompanied by male escorts. These problematic situations seriously constrain women's economic and social participation and are further evidence that isolation and vulnerability are linked to the persistence and creation of poverty. 7
 

Vulnerability
Just as there are debates and divisions over how best to conceptualise 'poverty' the notion of 'vulnerability' means many different things, depending upon the focus of the discussion. The report on Sustainable Human Development in Vanuatu (p.99), defines vulnerability in these terms:
 

Vulnerability is defined as those who are poor in a sense of not having enough to fall back on in times of crisis, during a period of crop failure for example. Although poverty or poorness may be part of vulnerability, income is not the only factor. Poor individuals, households and communities are often more vulnerable to changes within their economic, social or physical environment. Their savings may be limited or non-existent, and their network of kin and community equally economically disadvantaged. Their low health, nutritional and educational status makes it difficult for them to cope with economic, social, or natural disasters. In these situations: 'Vulnerability includes a lack of ability to mobilise social resources, which could be used to address common problems' (cited from the Vanuatu National Workshop on Poverty Alleviation and Women in Development Report 1995, ibid.). Poor urban households may be particularly vulnerable, as they often have less access to food gardens, or other subsistence sector resources.
 

The Fiji Poverty Report (p. 99) notes that, in addition to those who are living in absolute poverty, many individuals and households are at risk of becoming poor and lacks the social and financial safety nets which might provide assistance in times of sudden need.
 

The safety nets available to many low-income people in Fiji do not fully bridge the gap between an adequate lifestyle and destitution. In Fiji there is a strong sense of community and kinship. Yet the poorest households tend to be small and isolated; family systems do not encompass everyone, nor does their support necessarily last as long as the need for it does. Financial support systems exist but they often do not protect the most vulnerable. Government provides many forms of assistance to low-income earners but direct welfare is available only to the poorest of the poor and provides even them little relief from severe hardship. Non-government organisations run many programmes to assist the poor, yet acknowledge that poor coordination and their often limited out-reach prevent them from fully meeting the needs of the poor in Fiji. Vulnerability may also be related to government regulations, which constrain or actively prevent marginal people from entering the informal sector. In a discussion of street vending in Port Moresby, Kopel (1996:162) cites an incident when police smashed the eggs of a street vendor who did not have the correct permit. She concluded that:
 
For the poor, unemployed and uneducated urban dweller who has nothing more than his own labour and wisdom, sustainable development in terms of all this great talk is meaningless.
The 1990 World Development Report noted (p.2) that the burden of poverty is spread unevenly and that some individuals or groups are often more vulnerable and disadvantaged than others. It pointed out that: 'Women in general are disadvantaged. In poor households they often shoulder more of the workload than men, are less educated, and have less access to remunerative activities'.
 

In these circumstances, the informal sector often represents the only opportunity for poor women to obtain some cash income, often in order to provide for their children's basic needs. Yet, it is when women move out from the home to seek economic benefits, that they may be vulnerable, not only from attacks from those labelled as 'criminals', but from the forces of formal authority. A report in the PNG National newspaper (29 October 1998) described an incident in which three Highlands women, accompanied by young children, were selling betel nut on Waigani Drive. They were violently arrested by a group of 'City Rangers', formerly unemployed youth but now working with the National Capital District Commission to enforce law and order, including NCDC by-laws against street selling. In a public letter of complaint to the Officer-in-Charge of the City Rangers, it was stated that:
 

These irresponsible youths, unemployable in any other context, have repaid the confidence of the community by abusing its members. No alleged lawlessness on the part of this small group of women and their toddlers can possibly justify the violence of yesterday's attack. These women are selling betel nut to augment the pitiful money they have to support their families. Not for them the gilded public service handouts from the Prime Minister. They must live by their wits.
 
And on the larger issue - the insidious growth of poverty in our midst - let us have no more government humbug about rising exchange rates and the improving value of the kina. Tell that to the women who were victimised yesterday, Prime Minister, as they clutched their second hand clothing around themselves and tried to protect their malnourished children ('NCD must control City Rangers' Thursday, 29 October 1998, National).
If we reflect further on the whole concept of vulnerability, it is clear that the 'City Rangers' are themselves marginal and vulnerable individuals, also living 'by their wits' to improve their chances of economic survival. It may also be true that the challenge posed by the need to demonstrate good governance, while at the same time responding to the needs of the poor, presents an insoluble dilemma for most administrators.8
 

Natural Disasters
Many countries in the South Pacific have always been subject to natural climatic hazards such as cyclones, tidal waves, volcanic eruptions and landslips. Traditional safety precautions have been developed to minimise the devastating impact of these events. Safety measures against loss of life include the building of emergency low-level structures or the designation of caves or other places where communities find refuge. 9
 

If local food gardens are destroyed, families. and even entire communities, may be taken in by other less affected communities. This is part of the normal cultural cycle of reciprocal assistance and mutual help. At some time in the future, they will be called upon to return the favour, and provide assistance to others within their safety network. In recent years, family members working in towns, or in rural areas away from the home community, may respond to disasters by sending food supplies, building materials or additional cash.
 

This customary disaster relief may, however, be unable to accommodate a major catastrophic event (or a succession of less serious events) which exhaust locally available resources, or for which emergency medical and other assistance is urgently needed. Those affected by these events are then dependent on government or international aid. In countries with areas prone to natural disasters, or facing severe economic downturns, families or whole communities may experience poverty for the first time. For many, this is only a transitional experience, but for others, particularly where they were living at the edge of poverty, external and sustained support may be required to enable them to once again become self-reliant and independent. This is where the links between vulnerability and governance become important elements in the process of immediate support and longer term economic and social rehabilitation.
 

Over the past few years Papua New Guinea has suffered several major natural disasters. In 1994, a volcanic eruption devastated Rabaul in East New Britain, and had serious economic consequences, not only for the local communities, but at a national level. In 1997, a severe frost, which affected highland areas, was accompanied by one of the most widespread and severe droughts in living memory. And, in July 1998, a catastrophic tidal wave devastated several villages in the Sissano Lagoon in the West Sepik, with an estimated loss of 3000 lives. All of these events required short term government and international emergency assistance, and longer term economic support and the provision of health, education and other social services.
 

In October 1997, a survey of drought affected areas reported (Allen and Bourke 1997b:1) that 'about 540,000 rural Papua New Guineans now have insufficient food supplies or the cash income with which to purchase food'. Local self-help responses and the mobilisation of family and community networks had been able to maintain some households, particularly in the less severely affected areas. But, for many communities there was no safety net and food supplies were almost exhausted. Schools had closed as children joined their parents in the search for 'famine foods' and water. The situation was exacerbated in more disadvantaged areas as most government services had deteriorated:
 

The assessment teams found that services to rural people have seriously run down over the last decade to the point where, in some places, they do not exist any more. Health centres, aid posts and schools are often closed more often than they are open. A variety of reasons exist, from a lack of staff to acute shortage of equipment and supplies. In a number of remote areas, the only effective services are being offered by foreign mining, gas and oil companies ... The majority of small centres ran out of tank water in about August. As a result, many public servants were either forced to carry water from streams and rivers, or chose to abandon their posts. On the other hand, many have stayed and are serving local people under extremely difficult conditions (ibid. p.12). It was clear that the ability of the worst affected communities to cope with this disaster was severely restricted because they had already been marginalised, and had very limited communication facilities, transport and other infrastructure. As far as many of these communities were concerned, government benefits and development had little or no meaning. In the short term, aid from within and outside Papua New Guinea provided immediate support. But, once the crisis was over, the capacity of the government to deliver ongoing services usually had not improved.10
 

It should be recognised by observers quick to criticise the current Papua New Guineas government that this situation of neglect and disadvantage has persisted since colonial times. Allen and Bourke (1997:16) provide an informative comparison between four surveys carried out by Wilson (1974), de Albuquerque and D'Sa (1986), the National Nutrition Survey (Heywood et. al, 1988) and Allen and Bourke 1997). Changes in district and sub-district boundaries, and some sampling differences, limit absolute comparisons. Nevertheless, it is noteworthy that seven of the low income areas included in the most recent survey (Ambunti, Angoram, Kabwum, Kandrian, Pomio, Ramu and Telefomin) were among those included in the 1974 listing of districts with the lowest socio-economic indicators. This is clearly not only a current problem of governance, but also of harsh terrain and physical isolation.
 

For these vulnerable and isolated communities, the concept of 'sustainable human development' is a distant dream. They remain, in a very real sense, among the poorest of the poor in Papua New Guinea.
 

In this litany of neglect, one apparent success story is that of the Menyamya area, which was listed in the first three assessments as one of the least developed districts in the country. In the early-1970s, it had also been subject to an official enquiry, after local leaders complained of constant ill-treatment by government officers. Under the heading 'Menyamya praised for its success', the National reported (2 November 1998) the launching of 15 projects, using funds from the rural action program (RAP), Morobe Provincial Government funds and aid from other donors. Projects included education and health facilities, a town water supply and a fish breeding centre. At the official launching, Prime Minister Skate urged the Menyamya people to take the lead in development and stated that the 1999 Budget would target rural development. Why Menyamya has suddenly received such favoured treatment is not altogether clear, but it will be interesting to see whether these developments are sustained over time.
 

Final Reflections on Poverty and Governance
As noted in the introduction, a major question is: How far have the poverty assessments and discussions in Papua New Guinea, Fiji and Vanuatu taken into account the governance factor?
 

The United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP) has sponsored research into the effectiveness and adequacy of poverty alleviation approaches. The study of projects initiated in Papua New Guinea (D. Cox Volume 2, 1996:60) found that, while they were clearly very worthwhile programmes, the results were transitory and locally based:

 
As approaches to poverty alleviation, they are ad hoc development underpinned by no national plan; they are development built on personal endeavour with no significant commitment of resources; they are household-based developments in the main which will not result immediately in any significant changes in the country's institutional base; and they are local initiatives which, commendable as they are, desperately need to be complemented by macro-level, centrally-funded and nationally-planned development strategies. The study concluded (p. 125), that Papua New Guinea did not have the necessary prerequisites for a cohesive rural poverty alleviation programme.
  First, Papua New Guinea lacked the political cohesion needed to plan and introduce a national strategy. Members of Parliament and others retained too strong a sense of loyalty to their own group to be able to decide on a distribution of resources for the good of the nation. Second, Papua New Guinea lacked the personnel and other resources to put into operation a resource-intensive plan, but also the ability for intersectoral cooperation. Third, the poor regions lacked much of the infrastructure needed to permit a range of income-generation activities. The overall conclusion was that the current lack of government capability meant that it would be impossible to develop and sustain effective poverty alleviation programmes.
 

As noted in the introduction, current political processes have increased social fragmentation and have undermined administrative planning capacities. In part, this is a reaction against the centralised control which existed at the time of Papua New Guinea's independence in 1975. To be effective, national (or even provincial) poverty alleviation strategies will have to be able to accommodate the many socio-political systems which make up the nation state of Papua New Guinea. In varying degrees, Papua New Guinea, Fiji and Vanuatu, have all found that the inherited, centralised patterns of government service provisions are cumbersome and unworkable. Regional divisions in Papua New Guinea, racial divisions in Fiji, and the condominium divisions between Anglophone and Francophone in Vanuatu, remain recognisable features on the development landscape.
 

In Papua New Guinea, the conflict in Bougainville also had its genesis in colonial times, and the severe social, economic and political consequences have negatively affected the overall capacity of the government to govern. While progress has been made in resolving this problem, scarce resources, which could be directed towards economic and social development, have been dissipated. In Fiji, progress has been made towards equity and political accord between ethnic Fijians and Indo-Fijians. But problems still remain, and the determination of the extent and causes of poverty, and the allocation of resources to alleviate poverty, will continue to be a sensitive political issue. In Vanuatu, problems inherent in the condominium legacy of different legal and administrative systems are still being resolved. Many of the dual services provided under the condominium were clearly unsustainable on economic grounds and solutions must now be found to assist remote disadvantaged areas.
 

Papers presented at the Accountability and Corruption in Melanesia Workshop (State, Society and Governance Project, November 1998, RSPAS, ANU, Canberra) pointed to the interlocking nature of poverty and governance issues. It is not that there is a simple cause and effect relationship but that, as Lamour (1997:10) concludes, corruption 'may be embedded in a wider, but non-corrupt, framework of inept governance'. Political instability, and the preferential allocation of resource, also make it inevitable that some areas or groups receive unequal development assistance. Politicians seek to reward those who have voted for them and to balance central coordination and financial control against regional or local demands.
 

Problems of poverty, marginalisation, vulnerability, and corrupt and ineffective governance are not new or unique to the South Pacific. For some observers, these have been seen as inevitable consequences of political and social change, and economic development. Often the response has been to develop grand plans and national programmes of poverty alleviation, with the hope that these will be funded by international aid donors. But, national planning must also take into account local needs and be appropriate to particular cultural contexts. These alternative strategies of poverty alleviation and sustainable human development are just as relevant for the 1990s as they were in the 1970s, when Schumacher (1973) presented his vision of the value of small-scale economic development, or in the 1980s when the first years of post-colonial governance were evaluated in Papua New Guinea.
 

Small is still beautiful and may also be a safer and more rational way of planning in uncertain times. The alternative is the acceptance of violent disruption as a pre-condition for economic growth. It is paradoxical that, in the long run, this may lead to rough distributive justice as the dispossessed shake the foundations of the homes of the wealthy and powerful in order to seek the equity denied them (O' Collins 1985:371). Considerations of distributive and restorative justice provide another perspective on ways to lessen the increasing alienation of marginalised and unemployed youth. These young people pose a threat to civil order, and are among the most familiar (if often unrecognisable) faces of poverty encountered by the non-poor. In considering reforms to the justice system, Dinnen (1998:265) points to the need to take contextual factors into account when considering strategies to lessen problems of raskolism.
  Where, for example, lack of economic or development opportunities are accepted as underlying particular kinds of crime, then an appropriate restorative outcome should seek to improve such opportunities by linking rehabilitation strategies with educational opportunities, vocational training, micro-credit facilities or project assistance. The proportion of those living below or just above the poverty line may be debated for different countries or groups within countries. It may be even easier to argue over whether specific 'baskets' of food or other items represent basic or generous survival needs. But, the inescapable conclusion is that in Papua New Guinea and Fiji, and, although to a lesser degree, in Vanuatu, there are many different examples of poverty and marginalisation. And, it is equally certain that, as the 1997 Human Development Report points out:
  Behind these faces of poverty lies the grim reality of desperate lives without choices and, often, governments that lack the capacity to govern. The assessments of poverty in Papua New Guinea, Fiji and Vanuatu outline a number of strategies for providing real choices for those living in, or on the edge of, poverty. They also provide very real and compelling profiles of vulnerable people trying to cope with social, economic and environmental disasters, which have affected themselves, their families, and their communities.
 

Papua New Guinea is often seen as the most fragile and embattled of these three nation states. Yet, as May (1998:70) concludes it 'has maintained a robust democracy, [and] there are strong redistributive elements in local cultures'. These positive elements, and the strength and resilience of many government and non-government institutions and individuals, may well defy the odds and provide the basis for strategies to achieve more effective governance, and sustainable human development.
 

Endnotes
*This paper was developed after a Political and Social Change work-in-progress seminar on 13 July 1998. I greatly appreciated suggestions from Sinclair Dinnen, and resource material provided by Bryant Allen, Heather Booth, Margaret Chung, and Gerry Ward.
 

1 For example, Barr (Poverty in Fiji 1990) drew attention to the increasing number of poor and vulnerable individuals and families in Fiji. Morauta (1984) described the growth of poverty and inequality in rural areas of Papua New Guinea, and E. Cox (1994) pointed to the gender aspects of poverty.
 

2 See Morauta, 1984, for a discussion of the relationship between redistribution and poverty in rural villages. Monsell-Davis 1993 considers the related question of the economic burden which traditional safety net obligations place on urban workers. A number of reports also note that rural-urban migration has tended to weaken support systems (UNICEF/ GOPNG 1996:139; UNICEF/GOV 1991:76, UNDP/GOF 1996:89-91). Recently, the 1999 PNG National Government Budget (handed down on 16th November 1998) contains provisions to reduce funding for government services and increase discretionary funds for use by politicians in their electorates. Despite official statements that the focus will be on rural development, the likely impact will be to increase inequalities between politically popular and unpopular groups, and further weaken existing support systems.
 

3 Findings from the 1996 Papua New Guinea Demographic and Health Survey (National Statistical Office 1997) highlighted the difficulties of estimates, which relied on interviewee recall and official records. For example, maternal mortality rates in PNG have been estimated at 700 to 900 per 100,000 live births. This more recent survey calculated national maternal death rates at around 370, with lower rates in coastal areas, and a high of 625 in the Highlands (Jorari and Marckwardt 1997:91). Marckwardt (Appendix B) and Pochapon (Appendix C) discuss problems of sampling variability, interviewer reliability, and recall problems (such as the misreporting of age), which need to be taken into account.
 

4 The point that poverty should not be equated with vulnerability was made quite strongly in the ODA guide to the social analysis of developing country projects (ODA 1995:37). It noted that poverty. 'is usually measured by flows of income or consumption'. However, this definition excludes social and relationship poverty, which may be significant when cultural and family ties have been weakened by distance or disaster.
 

5 For example, in 1992 a commercial ship owner made an unscheduled stop in the Torres Islands. He reported that it was useless to schedule more visits, as the villagers did not have any produce ready for sale. (Personal communication 1992).
 

6 The problems of distance, and the high costs of air transport, have also been noted for other Pacific Island countries. See, for example, the discussion of problems related to air transport, and remote runways in Tonga (Ward 1998).
 

7 A number of NGOs have developed projects to assist women's participation. These include the East Sepik Council of Women, and the Foundation for Law, Order and Justice (Education Development Center 1996). The Individual and Community Rights Advocacy Forum (ICRAF 1997) has also been active in this area.
 

8 In fairness to the NCDC management, it should also be noted that more than twenty years ago the informal sector in Port Moresby had already been described as 'an Administrator's nightmare' (Fitzpatrick and Blaxter 1975).
 

9 Personal observations, during visits to Tanna in 1989 and to Hiu and Mota (Torres Islands) in 1992.
 

10 In a sober assessment of the current situation in Papua New Guinea, Yala and Levantis (1998) refer to the 'continuing drought in development' which has had a serious impact on all aspects of social and economic development. They conclude that improved living standards will not occur unless revenues from resource industries are managed more effectively to improve infrastructure and the delivery of services, particularly in the areas of health and education.
 

References

Ahlburg, D.A., 1996. 'Summary of a report on income distribution and poverty in Fiji', UNDP, Suva.

Asian Development Bank (ADB), and Republic of Vanuatu, Ministry of Transport, 1992. Vanuatu Inter-island Shipping Study, Interim Report, Port Vila.

Australian Agency for International Development (AusAID), J. Fallon and T. King, Economic Insights Pty Ltd, 1995. The Economy of Fiji: Supporting Private Investment, AusAID, Canberra.

Australian Agency for International Development (AusAID), G. McColl, T. Owen, J. Zerby (Unisearch Ltd), and A. Cook (Keiraville Konsultants), 1997. Economic Survey of Papua New Guinea, AusAID, Canberra.

Allen, B.J., and Bourke, R. M., 1997a. 'Poverty and Agriculture in Papua New Guinea: a in-depth description of six agricultural systems', Department of Human Geography, RSPAS, Australian National University, Canberra.

------, 1997b. 'Report of an assessment of the impacts of frost and drought in Papua New Guinea, Australian Agency for International Development (AusAID), Port Moresby.

Barr, K. J., 1990. Poverty in Fiji, Fiji Forum for Justice, Peace and the Integrity of Creation, Suva.

-----, 1996. 'Overview of development in Fiji since independence, in The Reality of Aid 1996, edited by J. Randel and T. German for ICVA and EUROSTEP, Earthscan, London: 53-60.

Bryant, J.J., 1993. Urban Poverty and the Environment in the South Pacific, Department of Geography and Planning, University of New England, Armidale.

Carvalho, S., and White, H., 1997. Combining the Quantitative and Qualitative Approaches to Poverty Measurement and Analysis, Technical Paper No 36, World Bank, Washington.

Chambers, R., 1983. Rural Development: Putting the Last First, Longman, London.

Cox, D., 1996. Showing the Way: Methodologies for Successful Rural Poverty Alleviation Projects, Promoting HRD Services for the Poor, No 1, and Making an Impact: Innovative HRD Approaches to Poverty, Promoting HRD Services for the Poor No 2, United Nations Economic and Social Commission for Asia and the Pacific.

Cox, E., 1994. 'Gender sensitive approaches to the identification and alleviation of poverty in Papua New Guinea', Gender, Economic Growth and Poverty, edited by N. Heyzer and G. Sen, International Books: Utrecht, and Asian and Pacific Development Centre, Kuala Lumpur: 352-391.

Dinnen, S. 1998. 'Criminal justice reform in Papua New Guinea', in Governance and Reform in the South Pacific, edited by P. Larmour, National Centre for Development Studies, ANU, Canberra: 253-272.

Downer, A., 1998. 'Australia and Papua New Guinea: Crime and the bilateral relationship', Speech by the Minister for Foreign Affairs at the dinner for the Conference on Australia/Papua New Guinea: Crime and the Bilateral Relationship, 11 November 1998. Canberra.

Fallon, J., 1994. The Vanuatu Economy: Creating Conditions for Sustained and Broad Based Development, AIDAB (now AusAID): Canberra.

Fitzpatrick, P., and Blaxter, L., 1975. 'The informal sector - an administrator's nightmare', Administration for Development 4:29-35.

Gibson, J., 1996a. 'Testing for boy-girl discrimination with household expenditure data: results for urban households', Working Paper No.1, Institute of National Affairs: Port Moresby.

-----, 1996b. 'Women's education and child growth in urban Papua New Guinea', Working Paper No.2, Institute of National Affairs, Port Moresby.

-----, 1996c. 'Baseline poverty estimates for urban areas of Papua New Guinea', Working Paper No. 3, Institute of National Affairs, Port Moresby.

Gibson, J., and Rozelle, S., 1997. 'Results of the Household Survey Component of the 1996 Poverty Assessment for Papua New Guinea', mimeo.

Gustafasson, B. 1997. 'Poverty and informal social safety nets in Papua New Guinea: a literature survey', in 'Formal and informal social safety nets in Papua New Guinea', edited by R. Guy, National Research Institute, Port Moresby.

Guy, R. (ed.), 1997. 'Formal and informal social safety nets in Papua New Guinea', National Research Institute, Port Moresby.

Guy, R., Tawaiyole, P., Khambu, J., and Avei, D., 1996. 'Education and the quality of life in Papua New Guinea: Five village case studies', National Research Institute, Port Moresby.

Individual and Community Rights Advocacy Forum (ICRAF), 1997. Justice and Freedom, an NGO program for human rights, gender equality, land rights and natural resources for the 21st Century, ICRAF, Port Moresby.

Jenkins, C. (and research team), 1996. 'Poverty, nutrition and health care in Papua New Guinea: a case study in four communities', Papua New Guinea Institute of Medical Research, Goroka.

Jorari, A., and Marckwardt, A., 1997. 'Infant and child mortality, and maternal mortality', in Papua New Guinea Demographic and Health Survey 1996, National Statistical Office, Port Moresby: 75-94.

Kopel, E., 1996. 'Street vending in Port Moresby: a positive step towards self-help or an eyesore for the city', in From Rio to Rai: Environment and Development in Papua New Guinea up to 2000 and Beyond, edited by D. Gladman, D. Mowbray and J. Duguman, papers from the 20th Waigani Seminar (1993), UPNG Press, Port Moresby:152-163.

Lamour, P., 1997. 'Corruption and Governance in the South Pacific', Discussion Paper No. 97/5, State, Society and Governance in Melanesia Project, Australian National University, Canberra.

Marckwardt, A., 1997. 'Sampling Variability' in Papua New Guinea Demographic and Health Survey 1996, National Statistical Office, Port Moresby, Appendix B.

May, R., 1998. 'From promise to crisis: a political economy of Papua New Guinea', in Governance and Reform in the South Pacific, edited by P. Larmour, National Centre for Development Studies, ANU, Canberra: 54-73.

Monsell-Davis, M., 1993. 'Safety net or disincentive?: wantoks and relatives in the South Pacific', Discussion Paper No. 73, National Research Institute, Boroko, Papua New Guinea.

Morauta, L., 1984. 'Redistribution in the village: Can we rely on it?', in Economic Development and Trade in Papua New Guinea, edited by A. Sawyerr, Proceedings of the Fourteenth Waigani Seminar, 1981, University of Papua New Guinea, UPNG Press, Waigani:174-186.

National, PNG. 1998. 'NCD must control City Rangers', Thursday, 29 October 1998.

------, 1998. 'Menyamya praised for its success', Monday, 2 November 1998.

O'Collins, M., 1985. 'Small is still beautiful; social development and the Eight Aims', in From Rhetoric to Reality?, Papua New Guinea's Eight Point Plan and National Goals after a Decade, edited by P. King, W. Lee and V. Warakai, University of Papua New Guinea Press, Port Moresby:367-372.

Overseas Development Administration (ODA), 1995. A Guide to Social Analysis for Projects in Developing Countries, HMSO, London.

Papua New Guinea, National Statistical Office, 1996. Papua New Guinea Demographic and Health Survey 1996, National Statistical Office, Port Moresby.

Pochapon. A., 1997. 'Quality of the data: nonsampling error', in Papua New Guinea Demographic and Health Survey 1996, National Statistical Office, Port Moresby, Appendix C.

UNICEF and the Government of Vanuatu, 1991. A Situation Analysis of Children and Women in Vanuatu, UNICEF, Port Vila.

UNICEF and the Government of Papua New Guinea, 1996. Children, Women and Families in Papua New Guinea: A Situation Analysis, UNICEF, Port Moresby.

United Nations and the Republic of Vanuatu, 1996. Sustainable Human Development in Vanuatu, United Nations, Suva

United Nations Development Programme, 1994, 1996, 1997. Human Development Report, Oxford University Press/UNDP, Oxford.

United Nations Development Programme (UNDP), and the Government of Fiji, 1997. Fiji Poverty Report, Government of Fiji, Suva.

World Bank, 1990. World Development Report: Poverty, Oxford University Press, Oxford.

-----, 1996. The World Bank's Partnership with Nongovernmental Organizations, Participation and NGO Group, Poverty and Social Policy Department,The World Bank, Washington.

-----, 1997. World Development Report: the State in a Changing World, Oxford University Press, Oxford.

-----, 1997b Advancing Sustainable Development: The World Bank and Agenda 21. Development Studies and Monographs Series No. 19, World Bank, Washington.

-----, 1997c. 'Papua New Guinea Poverty Assessment: Status Report as of February 17, 1997', Canberra.

-----, 1997d. 'Poverty assessment in Papua New Guinea, Preliminary Report', Port Moresby.

Ward, R. G., 1998. 'Remote runways: air transport and distance in Tonga', in Australian Geographic Studies 36(2):177-186.

Yala, C., and Levantis, T., 1998. 'Recent economic events in Papua New Guinea: a continuing drought in development', Pacific Economic Bulletin 13(2), pp.16.