Download The bottom billion: why the poorest countries are failing and more Study notes Global Economics in PDF only on Docsity!
The Bottom Billion: Why the poorest
countries are failing and what can be
done about it: Some insights for the
Pacific?
Terry O’Brien^1
A noted scholar of fragile states and of African economies, Paul Collier, argues that the
appropriate focus for today’s development effort is those countries whose residents have
experienced little, if any, income growth over the 1980s and 1990s. On his reckoning, there are
just under 60 such economies, home to almost 1 billion people.
Collier argues the plight of the ‘bottom billion’ is that they are caught in one (or often several) of
four traps; (i) conflict; (ii) mismanaged dependency on natural resources; (iii) weak governance
in a small country; and (iv) economic isolation among other very poor economies, with access to
big markets available only at high cost. Or as he puts it in the African context, ‘landlocked with
bad neighbours’.
Countries such as East Timor, Papua New Guinea and Solomon Islands suffer several of the
four traps Collier identifies. The growth performance over the last quarter-century of the six
Pacific economies in the bottom billion has been significantly weaker than the average of the
other states in the bottom billion. Effectively aiding the Pacific’s attempts to improve decades of
very weak per capita income growth may benefit from the insights into novel and ‘whole of
government’ forms of development assistance that Collier identifies for the ‘bottom billion’.
1 For the last four years, the author has been the Alternate Executive Director on the Boards of the World Bank Group for a constituency of 13 shareholder countries, including Australia. He has recently returned to the Macroeconomic Group. The views are those of the author, and not necessarily those of the Australian Treasury. He wishes to thank Paul Collier and World Bank colleagues Alan Gelb, Ginger Turner and Bruno Vincent for initial comments and assistance, and Elena Chanchu for extensive assistance with the data. The paper has also benefited from the comments and suggestions of Treasury and AusAID colleagues Simon Duggan and Julia Newton-Howes.
Introduction
A noted scholar of African economies and fragile states, Paul Collier, has recently taken stock of the pattern of progress in reducing global poverty in a short, non-technical book, The Bottom Billion: Why the poorest countries are failing and what can be done about it. The book rests heavily on Collierís and othersí earlier and more technical research, especially his work with Anke Hoeffler on the economics of conflict, aid, resources and fragile states. It proceeds mainly by practical and insightful narratives of the lessons learnt from the occasional success, and many failures, of work in some of the worldís most stagnant economies.
Collierís book was aimed at a G-8 audience of concerned citizens and policy makers, and received its European launch at a May 2007 event on improving African economic performance, under the aegis of Germanyís G-8 chairmanship. With that audience in mind, the book tends to be somewhat Africa-centred, but its insights are of much broader applicability. This article conveys some of Collierís insights, and points to their applicability to those Pacific states whose citizens are among the ëbottom billioní.
The need to focus development effort on the ‘hard cases’
Collier argues that the world is no longer best pictured as 1 billion rich and 5 billion poor. Rather, it is 1 billion rich, 4 billion in countries rapidly developing and converging in living standards on the rich (even if still home to the majority of the worldís extremely poor people), and 1 billion in states ëfalling behind, and often falling apartí.^2
This 1-4-1 division of the worldís 6Ω billion population is a useful approximation: there are about 1.1 billion people in the 30 developed countries of the Organization for Economic Cooperation and Development; and by various approaches to listing, there are almost 1 billion people in very poor countries that are not delivering much, if any, growth in per capita incomes. This leaves about 4 billion in the middle. But as we shall see, the definition of the ëbottom billioní cannot be wholly objective, and the category remains (at its margins) subject to debate and judgment.
2 Collier (2007a), p 3.
This second approach is the one chosen by Collier, for practical reasons of policy relevance: whether a community can grow richer is only partly a function of individualsí capabilities and efforts, but substantially a function of the national policy environment. Is the country at war, or peace? Can the government supply basic infrastructure, health and education, or not? Do governance structures secure property rights, or give rein to corruption? Is the national market growing, or contracting?
Identifying the countries of the ‘bottom billion’: relation to
‘fragile states’
Using this second approach, Collier speaks of a group of some 58 countries with a current population of 980 million who constitute the ëbottom billioní. While he does not list those countries, he argues they are distinguished by:
- low per capita income levels and weak per capita GDP growth; and
- being judged subject to one or more of the four traps.^5
These countries tend to be small: the average population is 17 million.
The problem of the ëbottom billioní is closely related to the difficulties confronting ëfragile statesí, which are also broadly defined in terms of low incomes, and poor policy, institutional and governance capacities.^6 The World Bank, the African Development Bank, the Asian Development Bank and the OECD donor countries that are members of the Development Assistance Committee have all discussed the special development challenges of these weakly performing states, and offered slightly different ways of identifying them. Box 1, Figure 1 and Appendix A offer some comparison of the countries these institutions have grouped into the ëfragile stateí category, and Box 2 provides a brief overview of the policy approaches suggested for aid donors to best help fragile states.
5 Collier (2007a), pp 5-12. 6 See, for example, OECD/DAC, (2007). En route to the current usage of ëfragile statesí, this topic has had a rapid succession of ëpolitically correctí names: ëfailed statesí, ëLow Income Countries Under Stressí (LICUS), and ëdifficult development partnersí.
Box 1: Listing ‘fragile states’: different approaches towards the ‘bottom
billion’?
The World Bank now identifies globally some 35 ëfragileí and ëborderline fragileí states and territories, home to almost half a billion people and characterised by weak institutions, poverty and vulnerability to conflict. Operationally, these characteristics are defined by: being an International Development Association (IDA) borrower, plus suffering weak policy and institutional capacity (measured by a score below 3. (roughly the bottom two quintiles) on the 0-5 scale of the Bankís Country Policy and Institutional Assessment (CPIA) index. (This listing includes Timor Leste, Papua New Guinea, Solomon Islands, Tonga and Vanuatu, plus Cambodia, the Lao PRD and Myanmar in the East Asia and Pacific region.)
Other agencies that have sought to enumerate the fragile state concept have generally arrived at similar or broader lists:
- The African Development Bank, focusing just on its membership in sub-Saharan Africa, and using both the CPIA index and its own ëcountry vulnerability indexí, has added six African countries to the 21 African countries identified in the World Bank listing.
- The Asian Development Bank, focusing just on its Asia-Pacific membership, proposes an identification of ëweakly performing countriesí based on its own Country Performance Assessment (CPA) system, and involving either a CPA score in the fourth or fifth quintile in two of the three most recent years, or vulnerability based on conflict or post-conflict status. The CPA for 2006 has been published, but the list of weakly performing countries is not published. We understand the Asian Development Bankís CPA and the World Bankís CPIA ratings are highly correlated, so probably much the same group of Asia-Pacific countries are identified as ëweakly performingí by the Asian Development Bankís approach as are classified as ëfragileí by the World Bank approach.
- The United Kingdom Department for International Development (DFID) (2005) includes a ëproxy list of fragile statesí, which it defined as countries that had appeared at least once in the fourth and fifth quintiles of the World Bankís CPIA between 1999 and 2003. That produced a list of 46 countries, home to some 920 million people, approaching in aggregate Collierís ëbottom billioní of 58 countries and 980 million. The criterion led it also to include Indonesia in its list.
Box 2: Suggested policy approaches for donors to help fragile states
OECD donor countries that are members of the Development Assistance Committee have participated in a dialogue with the World Bank, the European Community (EC) and the United Nations Development Programme (UNDP) to shape a consensus on how best to engage in fragile states. Resultant broad principles include:
- Take context as a starting point: understand the local limits to capacity, political will and legitimacy.
- Do no harm: for example, donít inadvertently worsen corruption or disrupt budget processes by large, uncoordinated swings in aid.
- Establish state building as the central objective: strengthen the capabilities of states to perform their core functions and address democratic governance issues.
- Align with local policies and/or systems: donors should use local systems for disbursements where they are satisfactory, and where not, disburse aid in a way that helps build local capacities.
- Recognize the political-security-development nexus: fragile states need help in all dimensions, and a ëwhole of governmentí approach among donors is likely to work best.
- Promote coherence between donor government agencies: where possible, use joint assessments and shared strategies among donors and with host governments.
- Act fast: take rapid action where the risk of instability is highest, but stay engaged long enough to give success a chance in what is inevitably a slow state-building process that may encounter setbacks.
- Avoid ëaid orphansí: areas characterised by low engagement and field presence.
OECD/DAC (2007).
Figure 1: Overlapping definitions of ‘fragility’, relative to the World Bank definition
13 extra countries 425 million extra people
6 extra countries
45 million extra people 'Weakly performing countries'
Fragile and marginal fragile states
35 countries or entities
490 million people
Total: 54 countries, 960 million people
Collier Bottom Billion : 58 countries, 980 million people
African Development Bank
World Bank
Asian Development Bank
UK Department For International Development
One would expect Collierís country list of 58 to include all 35 of the World Bankís fragile states, plus most of the countries added in the listings of the African Development Bank, the Asian Development Bank and DFID. He mentions examples such as Timor Leste, Fiji and PNG to illustrate his arguments, but says of the bottom billion that,
When I want to use a geographic label for them I describe them as ëAfrica +í, with the + being places such as Haiti, Bolivia, the Central Asian countries, Laos, Cambodia, Yemen, Burma and North Korea.^7
7 Collier (2007a), p 7.
in 2005), there is a beginning of convergence (with the middle groupís per capita income rising from one-eighth of the top billionís in 1980 to one-fifth in 2005). Collier notes that the current rate of sustained strong per capita income growth in the middle 4 billion is unprecedented in human history.
- For the bottom billion (as approximated by the 960 million people of the composite listing of 54 countries of Appendix A, including six Pacific states), incomes have grown at little more than half the rate in the top billion. Their living standards are still diverging from those in rich countries.^10
- Within this illustrative ëbottom billioní set of 54 countries, for the 35 ëfragile statesí identified by the World Bank (home to almost half a billion people, and including five Pacific states), real income levels have fallen by more than 5 per cent in a quarter century, compared to almost 60 per cent growth for the top billion. The citizens of the World Bankís listing of fragile states could now perhaps fairly be called ëthe very bottom half-billioní.^11
- Although not shown in Table 1, the contraction of GDP per capita in the bottom billion over the last 25 years is concentrated in sub-Saharan Africa. As a group, the other bottom billion statesí real per capita incomes grew a little, but either still slower than the OECD countries (if Indonesia is excluded from the bottom billion group) or somewhat faster (if Indonesia is included in the group). But in either case, incomes in the ënon-African, bottom billioní states grew much less than in the countries of the ëmiddle 4 billioní people.
Collier stresses that the problem of the ëbottom billioní is not the ëquality of growthí (for example, the inequity with which the fruits of growth are distributed), it is no growth.^12 Available data probably understate the problem: we lack internationally comparable GDP data for some of the poorest or most turbulent countries (such as North Korea and Iraq). Moreover, weak growth in the measures of national income that are available has parallels in social indicators: progress towards the Millennium Development Goals is disproportionately weak for the fragile states, compared to other developing countries (Table 2).
10 Even this modest result is heavily influenced by the inclusion of Indonesia in the illustrative bottom billion listing of 54 countries. If Indonesia were instead classified in the middle 4 billion group, that groupís growth rate would barely rise, but the new ëbottom three-quarter billioní, minus Indonesiaís 220 million people, would see their average annual per capita GDP growth fall to about zero (-0.4 per cent, to be spuriously precise). 11 The numbers are best regarded as approximate, because national accounts for many poor countries were of poorer quality, or non-existent, in the 1970s and 1980s, and the quality and coverage of Purchasing Power Parity estimates in that era were also limited. As a generalisation, coverage has been weakest for the poorest countries. 12 Collier (2007a), p 11.
Table 2: Fragile states face the largest deficit in most Millennium Development Goals Total in developing Total in fragile Fragile states Indicator countries (m illion) states (m illion) per cent share Total population (2004) 5,427 485 9% MDG1 - Poverty (2004) Extreme poverty 985 261 26% Malnourished children 143 22.7 16% MDG2 - Universal education Children of relevant age that did not complete primary school in 2005 13.8 4 29% MDG4 - Under five m ortality Children born in 2005 and expected not to survive to age five 10.5 3.3 31% MDG5 - Maternal health Unattended births 48.7 8.9 18% MDG6 - Diseases TB deaths 1.7 0.34 20% HIV+ 29.8 7.2 24% MDG7 - Environm ental sustainabiliy Lack of access to improved w ater 1,083 209 19% Lack of access to improved sanitation 2,626 286 11% Source: Global Monitoring Report 2007 , p 13.
Traditional aid: already past the point of diminishing returns in
the bottom billion?
Collier believes traditional ODA is not well suited to strengthening growth in the ëbottom billioní. Returning to his previous writings that tease out the similarities and differences between aid and oil (both being basically a source of ësovereign rentsí to governments in weak governance environments), he argues that traditional aid to African countries in the ëbottom billioní may already be at levels that produce diminishing returns. He uses the recent surge in oil prices as a natural experiment to test whether the Gleneagles G-8 promises to double aid to Africa might be beneficial. The oil price surge has already given African oil producers more resources than additional aid is likely to, while at the same time African non-oil producing countries have been hindered by high oil prices. Yet so far, there has been no difference in the economic performance of the two groups, suggesting that the four traps are sufficiently powerful to overwhelm the impact on oil exporters of additional resources.^13
Nonetheless, Collier remains a defender of traditional ODA. Aid is better than oil, because donorsí conditions on ODA expenditures (for example, auditing) and on the form of assistance (including, for example, the often-criticised use of international
13 Collier (2007a), pp 101-2.
Trap 2: Mismanaged dependency on natural resources
Almost one-third of the ëbottom billioní live in countries that are resource-dependent. Collier marshals evidence that resource rent dependency in developing country conditions tends to cause ëDutch diseaseí^18 , corruption, malfunctions of democracy through patronage politics, and often, autocracy.
Collier notes that moves to develop natural resources in the bottom billion, to reduce consuming countriesí dependency on autocratic and politically unstable producers in the Middle East, underestimate the role of resource dependency in producing autocracy and political instability: ëBecoming reliant upon the bottom billion for natural resources sounds to me like Middle East 2í.^19
However he affirms that notwithstanding the challenges, resource-rich countries have to exploit those resources to grow. Resources are their comparative advantage, so the policy question is how best to help them reduce the risks of the resources curse.
He judges aid as ëfairly impotentí to address the resources curse directly, though it can be indirectly powerful through addressing weak governance.^20
Trap 3: Weak governance in small countries
The World Bank defines public sector governance as ëthe way the state acquires and exercises its authority to provide and manage public goods and services, including regulatory servicesí.^21 Weak governance capacities have bedevilled the countries of the bottom billion, and Collier argues that small, poor states heavily reliant on aid and often resource-rich, face particular governance challenges: ëThe minimal state is not a viable model in the context of oil and aid; the government must transform its money into public servicesí.^22
Collier argues that in such cases an important contribution to strengthening governance can come from appropriate international standards and codes, including:
- a more comprehensive Extractive Industries Transparency Initiative;
18 ëDutch diseaseí is The Economistís (26 November 1977) memorable nickname for the tendency for large foreign currency inflows (such as from oil exports or ODA) to raise the real effective exchange rate, disadvantaging other exporters and import-competing industries. Collier argues the reason resource wealth has not had adverse political effects in, say, Norway, Canada or Australia, is that those countries had solid institutional and democratic inheritances before their largest surges of natural resource wealth; Collier (2007a), pp 42-4. 19 Collier (2007a), p 52. 20 Collier (2007a), p 107. 21 World Bank (2006), p 10. 22 Collier (2007a), p 66.
- a comparable initiative for the construction industry (particularly important given the vulnerability of this sector to corruption, and the very strong growth of ODA for infrastructure investment);
- continued roll-out of fiscal transparency standards and a charter for budget scrutiny;
- an international investment charter, working around limitations in Multilateral Investment Guarantee Agency and national risk insurance institutions and delivering on the promise of the failed OECD attempt to negotiate a Multilateral Agreement on Investment; and
- a ëpost-conflictí charter for donors and the international security system.
Such developments need to be driven by civil society and governments in poor and rich countries, in Collierís view, and he argues that developed countriesí non-government organisations would get better results stressing these areas than environmental and employment policies, which matter less to the ëbottom billioní.
He also argues that regional peer review arrangements to bolster applications of such codes would be valuable.^23 Fiscal codes are particularly important for resource-rich ëbottom billioní states, to help them perform the awkward reconciliation of high resource rents, low need for personal or consumption taxes, and the need for high government accountability. (The difficulty of this task is captured in Thomas Friedmanís (2006) aphorism, ëno representation without taxationí, also summarised in his ëfirst law of petropoliticsí: when oil prices rise, democratisation falls.)
Trap 4: ‘Landlocked with bad neighbours’ (or high-cost transport to large markets?)
Collier argues from his African research that a prominent trap of the ëbottom billioní is that they are landlocked with bad neighbours. Of course, being landlocked did not hurt Switzerland, but that is because Switzerlandís neighbours are big, rich markets and Switzerland invested heavily in transport infrastructure. Landlocked countries in Africa are hemmed in by very poor, small markets, often in conflict, and very poorly invested in the infrastructure that would permit trade at lower costs. He argues the only landlocked African states that have overcome this geographic penalty have been those well endowed with very high value natural resources and that have enjoyed good governance. But with that exception (illustrated by Botswana), Collier argues ëIf
23 A current example of developing country peer review occurs in the New Partnership for Africaís Development (NEPAD); http://www.nepad.org/2005/files/inbrief.php.
- encouraging migration and remittances. (That is, since it is costly to export goods, it may be more feasible to increase the mobility of people, in enlightened and mutually beneficial migration or work arrangements. Maximising the development benefits from such mobility requires the economic and financial policies that encourage remittances and business start-ups in the sending country);
- paying attention to rural development, since under the best of policies, most people in the bottom billion are going to remain rural. In their countriesí special cases, the technologies for rural productivity increase may require heavy investment in local knowledge;
- trying to attract responsible foreign investment in resource development through transparency and the assistance good host-country governance can provide to maintaining a good corporate image for the resource investor;
- trying to attract aid. Collier argues that ODA ought become more heavily concentrated in higher risk, bottom billion environments, which will require higher ëgovernance conditionalityí, and higher tolerance in donor communities of the supervision costs and risks of failure of operating in fragile states. Technical assistance, often reviled as ëaid to consultantsí rather than to the bottom billion, is very pertinent, given their countriesí skill gaps. Recipients who understand these dilemmas and work with donors can win ODA that might otherwise go elsewhere. At present, Collier argues (particularly from his UK and EU perspective) that donor politics and NGO support for debt forgiveness, ëvertical fundsí and so on are generally naive and likely to produce either avoidance of donor commitment to ëbottom billioní countries, or perverse results.^26
Trade policies for helping the bottom billion
Collier devotes a chapter to trade policy which includes insights that:
- rich country agricultural protectionism is the worst form of ëpolicy incoherenceí for donors who want to help the ëbottom billioní. Viewed from the objective of true economic development, it undoes the benefits of aid;
- protectionism in the ëbottom billioní occurs mainly because it is a key instrument for corruption (import licences, bribes to customs officials and so on);
26 Collier (2007a), pp 99-123.
- regional trade liberalisation among poor countries is very damaging, leading to accelerating divergence in living standards from global trends, and trade diversion. (There are more regional trade agreements than countries, and the typical African economy is in four agreements, usually incompatible with each other;)^27
- ëfair tradeí consumer movements in developed countries are not merely a harmless form of self-imposed consumption tax; they transfer resources to poor countries only in forms less useful than ODA (for example, higher coffee prices), because they are conditional on producers continuing traditional exports.
Such thoughts lead Collier to argue that:
- it is now getting harder for the bottom billion to benefit by trade, as China and India have driven competition from the field of labour-intensive, low-cost, simple manufacturing. (This claim may only be true for a limited time ó historically, countries have always moved up the ëvalue chainí as they became more productive, and it seems likely China and India will too);
- the ëbottom billioní need temporary protection from low-income Asia, before rich countriesí protection falls so low that there is no means to achieve this. Collier favours US-style African Growth and Opportunity Act initiatives (which in the area of apparel, have increased African exports to the US by 50 per cent, because of simple and unrestrictive rules of origin). In contrast, Collier judges the EUís ëanything but armsí initiative to have been ineffective;
- given the low utility of the WTO as a bargaining vehicle to the ëbottom billioní, Collier favours an unreciprocated reduction of the trade barriers against them, which he argues would be analogous to the emergence of IDA as a (near) gift to eligible countries after the recognition that the International Bank for Reconstruction and Developmentís post WWII design for European reconstruction would not benefit broader economic development^28 ;
- to the extent that he has any support for a ëbig pushí in ODA, he favours it being used in infrastructure (both physical and social) that would lower the barriers to the bottom billionís trade.
In his conclusion, Collier positions his views between the left (such as Jeffrey Sachsí The End of Poverty ) and the right (such as William Easterlyís The White Manís Burden ).
27 Collier (2007a), p 164. 28 Australia has made such an unreciprocated reduction in its barriers to 48 Least Developed Countries (see Lippoldt 2006).
Table 3: Pacific and the bottom billion (GDP per capita, PPP constant 2000 US$)
annual annual annual annual $ grow th $ grow th $ grow th $ grow th 1980 16,587 2,190 1,423 2, 1990 20,062 1.9% 3,086 3.5% 1,572 1.0% 2,027 -0.5% 2000 24,143 1.9% 4,025 2.7% 1,605 0.2% 2,440 1.9% 2005 25,975 1.5% 5,190 5.2% 1,834 2.7% 2,367 -0.6% 2005 as m ultiple of 1980 1.57 2.37 1.29 1. Annual % grow th, 25 years 1.8%^ 3.5%^ 1.0%^ 0.4%
m inus Pacific six 48 countries Top billion Middle 4 billion com posite list*
Bottom Billion
Pacific six*
- The Pacific six are Papua New Guinea, Solomon Islands, Timor Leste, Tonga, Vanuatu and Kiribati.
Table 4: The Pacific and World Bank fragile states (GDP per capita, PPP constant 2000 US$)
annual annual annual annual $ grow th $ grow th $ grow th $ grow th 1980 16,587 2,130 1,398 2, 1990 20,062 1.9% 2,963 3.4% 1,290 -0.8% 2,004 -0.5% 2000 24,143 1.9% 3,824 2.6% 1,153 -1.1% 2,408 1.9% 2005 25,975 1.5% 4,896 5.1% 1,294 2.3% 2,341 -0.6% 2005 as m ultiple of 1980 1.57^ 2.30^ 0.93^ 1. Annual % grow th, 25 years 1.8% 3.4% -0.3% 0.4%
Top billion
Pacific five Middle 4 billion (30 countries)*
states m inus
WB fragile
Pacific five*
- The Pacific five are Papua New Guinea, Solomon Islands, Timor Leste, Tonga and Vanuatu.
Throughout the Pacific, countries face the pervasive challenges of separation by high transport costs from large markets, weak governance, sometimes high dependency on natural resources, and in some cases, periods of conflict or coups.
Collierís diagnoses of the policies required to help the bottom billion are
thought-provoking for the Pacific:
- challenging the Pacific countries to develop policies to minimise the adverse effects of geographic barriers to growth, and manage the governance challenges of resource revenues and aid flows;
- challenging donors to sustain patient ëwhole of governmentí commitment to helping strengthen governance and repair breakdowns in security; and
- encouraging greater development of international codes to help strengthen countriesí own performance in mining and construction industry transparency and fiscal transparency.
