• No results found

The World Bank in Vietnam and Indonesia

Introduction: Case Study Selection

Part II of this thesis comprises two case studies of how the PWC has translated into Bank lending. This section introduces the case studies and outlines the methodology used in the selection of case study countries. The case studies examine the Bank’s overall relationship with the borrower but pay particular attention to Bank lending. This is a timely exercise because surprisingly, in the last decade or so — apart from the odd controversial dam project or the like — Bank lending has received limited critical focus.

Further, investment lending receives even less than adjustment lending. Attention has in the main been focused on the Bank’s development discourse. This focus is undoubtedly warranted and had been facilitated by the Bank’s increasing disclosure of documents from 1993. The new disclosure policy, though, also applies to loans thus making analysis of the Bank’s lending significantly easier. However, there is now sometimes so much information available that it is difficult to analyse it.

Bank lending is more than just a way of maintaining the relationship with a country in order to influence its policy directions, as has on occasion been suggested.1 Rather, as is shown in the case studies, the Bank’s lending program involves the pursuit of an often long-term agenda. Moreover, the form and nature of Bank lending — project lending, design and implementation according to Bank standards and procedures and international tendering — binds borrowing countries to market procedures and enmeshes them with international capitalism.

In terms of the timeframes studied, the Washington Consensus dominated the Bank from the early 1980s to the mid to late 1990s. Ideally, a case study of Bank lending in this period would have covered a period in the late 1980s to early 1990s. However, in

1 See for example, Kapur, Lewis and Webb, World Bank, p. 270.

1993 the Bank relaxed its official disclosure rules for projects approved after that date.2 The new rules provide for disclosure of: Country Assistance Strategies, Staff Appraisal Reports (later renamed Project Appraisal Documents) and (some) Implementation Completion Reports. These reports form the basis of the analysis undertaken below.

Given this change, the first case study period starts in 1993. It was considered that five year’s analysis of new project approvals would provide a fairly comprehensive overview of Bank lending and trends, so the first case study period runs from 1993 to 1997. For the post-Washington Consensus period, the latest period possible was chosen to allow its influence to have filtered through the Bank’s pipeline of projects — the period studied is 2000 to 2004.

In order to select the case study countries, a range of data was collected on eight developing countries across four regions with active Bank portfolios. In South-east Asia, Indonesia and Vietnam were explored, in South Asia, Pakistan and Bangladesh, in South America, Peru and Brazil and in Africa, Nigeria and Uganda. Given the focus of the project on changing lending practices, two factors were considered crucial for case study countries, first a fair sized portfolio with the Bank (approximately $1 billion in gross disbursements over a five year period) so that there are a range of loans to compare in the different periods. Second, borrowers should have at least a ‘medium-sized’ economy so that development policy and lending has not normally been seriously impacted by a single event such as a large fall in the price of a key export or a natural disaster. In addition, it was considered important that the countries have a strong relationship with the Bank and that it’s lending is likely to be of significant economic consequence. Thus, although it is beyond the scope of the current research project to determine the impact of Bank lending (indeed this is a virtually impossible task), the project may in addition be of some use in indicating where Bank programs have been most influential in case study countries and some of the possible implications of this.

Data collected on each potential case study country included:

ƒ Eligibility for IBRD and/or IDA;

ƒ Total number of Bank projects and the cumulative lending;

ƒ Bank Approvals in 2003;

2 The Bank adopted the more conservative of two proposals developed by a committee convened to study disclosure policy, Caufield, Masters of Illusion, p. 266.

ƒ Type of loans (investment, adjustment and other non-project);

ƒ Gross and net disbursements over the period 1997-2002;

ƒ Net Official Development Assistance (ODA) received 1997-2002;3

ƒ Net Bank disbursements as a percentage of net ODA 1997-2002;

ƒ ODA as a percentage of Gross National Income (GNI) or Gross National Product (GDP), 2001;

ƒ FDI flows 1997-2002; and

ƒ ODA as a percentage of FDI, 1997-2002.

Data was obtained from official World Bank publications and its website, the UNDP Human Development Index and the OECD Development Assistance Committee website.4

The South American countries were quickly excluded as in both cases, whilst Bank funds are significant in relation to net flows of ODA, ODA as a per cent of GNI overall is minute (for Peru 0.85 per cent and Brazil 0.07 per cent). In addition, both countries only have access to IBRD funding and both, as of July 2004, had only environment and investment loans with the Bank, that is, they had no adjustment loans. In Nigeria, ODA as a percentage of GNI was also small (0.47 per cent) and, in addition, Bank gross disbursements were quite low and net disbursements were negative for 1997-2002.

Uganda had the highest ODA as a percentage of GNI of the countries studied by far (14.28 per cent); the Bank was a significant contributor of funds overall and there was a good mix of loans. However, Uganda had the smallest population examined and the high levels of poverty mean the economy is quite small, add to this the high dependence

3 The OECD Development Assistance Committee defines ODA as:

Grants or Loans to… (developing countries) which are: (a) undertaken by the official sector; (b) with promotion of economic development and welfare as the main objective; (c) at concessional financial terms [if a loan, having a Grant Element of at least 25 per cent]. In addition to financial flows, Technical Co-operation is included in aid. Grants, Loans and credits for military purposes are excluded.

Developing countries (Part I) are defined by the DAC every three years, countries in transition (Part II) are classified as receiving Official Aid (OA) not ODA. Credits from the IDA are generally concessional enough to be regarded as ODA. Other funds from donor countries or multilateral agencies that do not meet this definition are classified as: Other Official Flows (OOF). IBRD loans and IMF loans are generally included in this category as their grant element is less than 25 per cent. Funds for commercial purposes or designed to facilitate exports are also regarded as OOF. Source OECD DAC:

www.oecd.org/dac/ accessed November 2005.

4 UNDP Human Development Index: www.undp.org, OECD DAC: www.oecd.org/dac/ both accessed February 2004.

on primary commodities and the ongoing unrest in the northern border regions and it was not attractive for a study of changing Bank lending.

For both Pakistan and Bangladesh, ODA is a small but significant contributor to GNI — 3.37 and 2.12 per cent respectively. The level of ODA to Pakistan increased notably from 2001— a direct result of its role in the so-called ‘war on terror.’ The instability in Pakistan and ongoing travel warnings essentially ruled it out as a case study country.

World Bank gross disbursements to Bangladesh were over $2 billion for 1997-2002, a significant amount both in relation to total ODA flows and total FDI flows. However, as of July 2004, the active Bank portfolio contained only one adjustment loan. It was a possible case study.

Official Development Assistance as a percentage of GNI in Vietnam was 4.42 per cent, a comparatively significant level. Between 1997 and 2002 gross Bank disbursements were over $1.3 billion, a significant amount in relation to both ODA and FDI flows. The Bank portfolio is made up predominately of investment loans, however, structural adjustment lending increased from almost nine per cent of lending in 1993-1997 to over 16 per cent in 2000-2004. Although this is well below the Bank’s current average of 33 per cent of loans in structural adjustment, it is a significant level for a country defining itself as socialist.

In Indonesia, ODA as a percentage of GNP in 2001 was approximately 2.5 per cent, which is a historically average level. Although a spike in 1998-1999, due the Asian Financial Crisis, took the percentage to over four.5 Even 2.5 per cent is a significant level for such a large economy. Gross Bank disbursements to Indonesia 1997-2002 were over $5.8 billion, although interestingly net disbursements were quite low, indicating a high repayment burden. This meant that net Bank loans as a percentage of net ODA were a low 3.7 per cent. However, the significance of the Bank to Indonesia is clear when the gross Bank disbursements (1997-2002) of over $5.8 billion are compared to FDI flows during the same period, which were minus $13.3 billion. As of July 2004, Indonesia had only one active adjustment loan out of a total of forty loans.

5 Anis Chowdhury and Iman Sugema, "How Significant and Effective Has Foreign Aid to Indonesia Been?," ASEAN Economic Bulletin, Vol. 22, No. 2 (2005), p. 189.

Prior to the Asian Financial Crisis, Indonesia had only ever accessed one adjustment loan, since then it had four bring the total to five out of its total of 317 loans.

There are a range of other reasons why Vietnam and Indonesia stand out as attractive options for the case studies, including the fact that there has been relatively little study of Bank activity in these countries. But there is also their accessibility, the interesting contrast between their forms of government and relationships with the Bank. Bank lending to Vietnam resumed in 1993, during the period when the Bank was shifting away from the Washington Consensus. It seems reasonable to assume that the assigned staff would have fewer preconceptions about the ‘right’ policy actions. It may mean that new directions in the Bank’s approach to development could be more apparent here.

Indonesia and Vietnam share a region and are now at similar levels of development — Indonesia’s 2006 Human Development Index ranking was 108 and Vietnam’s was 109.6 Indonesia’s Gross National Income per capita is still significantly higher than Vietnam’s (2005: $1280 versus $620) but Vietnam’s life expectancy is higher (2004:

70.3 years versus 67.4 years).7 Both countries have an adult literacy rate of just over 90 per cent and similar levels of reported income inequality (2002 GINI Index Indonesia 34.3, Vietnam 37.0).8 Of course, Indonesia has a much larger population than Vietnam, they are geographically very different and their development and coherence as ‘nation-states’ is quite distinct.9 They also have interesting and contrasting forms of government, levels of political stability and approaches to the role of the state in development. These contrasts should be useful for the study in terms of analysing the Bank’s changed approach to the role of the state in the economy and specificity to local conditions in two diverse settings.

The contrast continues with the impact of the Asian Financial Crisis — Indonesia was hit early and hard by the Crisis and is only just returning to earlier levels of GDP growth. Whereas the relatively insulated Vietnamese economy was only indirectly

6 United Nations Development Report 2006, "Statistics", available from:

http://hdr.undp.org/hdr2006/statistics/, accessed: April 2007.

7 World Bank Group Website, "Data and Research, Country Data.”

8 United Nations Development Report 2006, "Statistics."

9 In both countries too the military has played, and continues to play, an important role in both politics and the economy although in both cases this role has declined over the past five to ten years.

impacted and growth rates quickly returned to the equal the previous high levels.

Indonesia was an early favourite of the World Bank and their relationship through to the time of the Asian Financial Crisis, was if not always perfect, certainly strong. However, since then Indonesia has fallen out of favour with the Bank and the level of lending has decreased. Vietnam in contrast, after decades in the wilderness, has become a Bank favourite — one of it’s proclaimed ‘success stories.’ Since the Asian Financial Crisis though, there have been hints of strains in the relationship.

There is one identifiable weakness in the suggested case studies, which is that both countries have a low proportion of adjustment loans, compared to the Bank’s overall portfolio. In 2003, adjustment loans accounted for 33 per cent of new lending commitments and for the period 1997-2003 they averaged 38.2 per cent. However, adjustment loans appear to be disproportionately concentrated in a few countries. A brief survey of loans in 30 countries confirmed this with twenty countries (including India, China and Thailand) having none or only one adjustment loan, whilst a few countries had a very high proportion of adjustment loans. For example, over 48 per cent of Turkey’s loan portfolio by value was adjustment lending.10 Nevertheless, overall the majority of Bank loans are project loans and relatively less analysis has been done on this area, in comparison to structural adjustment, which has been the subject of much study.

Structure of the Case Studies

Each case study comprises two chapters. The first chapter focuses on the general background starting with some information on the political, economic and social structures and policies of the country. The country’s dealings with donors are outlined as are their relationship with the World Bank. The initial chapter also details the link between the policy framework of the country and the Bank’s program through an analysis of the Bank’s key country planning documents: the Country Assistance Strategies and Poverty Reduction Strategy Paper.

10 Survey of active portfolios in July 2004, undertaking using the World Bank’s projects database, available online at World Bank Group Website.

The second chapter provides a detailed analysis of Bank lending structured by type of lending: adjustment or investment. Adjustment loans were studied together but analysis of investment lending was separated into sectors, utilising the Bank’s own sectoral classifications of activities. Thus loans in say the transportation sector in the first case study period are compared with loans in the transportation sector in the second case study period. This generally worked well because, despite the broad nature of the sectoral classification, Bank projects in both case study periods tend to be similar. For example, in Vietnam energy and mining sector projects are made up only of activities in electricity generation, distribution and retailing. It did create some problems in sectors where Bank lending was small. For example, in Indonesia there is only one loan in each case study period in the information and communications sector and they are in quite different areas. This made any comparison impossible but such problems areas made up only a very small proportion of lending so their exclusion or incorporation into other sectors is unlikely to have any detrimental impacts on the overall outcomes of the loan analysis.