Angola presents a graphic example of how developing countries with large natural resources – in particular oil and other minerals – are among those most prone to poor governance, armed conflict and poor performance in economic and social development. This hypothesis, backed by evidence from many parts of the developing world, postulates that countries with weak institutions (a characteristic of most developing countries) are poorly able to withstand the destructive forces of corruption and conflict unleashed by attempts to control and appropriate the rents accruing to the state from the taxation of oil and other mining operations. Mineral exports drive up the value of the local currency, undermining the competitiveness of the rest of the economy (such as agriculture and manufacturing) because imports are cheap. Often, these factors feed on and exacerbate other social cleavages, such as ethnic, regional or religious divisions.

Angola does indeed have an exceptional resource endowment. It is Sub-Saharan Africa's second largest oil producer (after Nigeria), with output currently running at over 900,000 barrels a day (b/d), almost entirely from offshore oil fields along the northern coast. It is also the world's fourth largest producer of diamonds (by value), accounting in 2000 for about 15 per cent of the world's production of rough diamonds. The diamond resources, both alluvial and kimberlite, are concentrated mainly in the north-east of the country.

Angola also has deposits of numerous other, yet-to-be exploited minerals and, in the colonial period, it was a medium-sized producer of iron ore. Substantial gas deposits have been found, but are yet to be productively utilised: most of the gas associated with producing oil fields is still being flared.

If its resources were properly harnessed, Angola would be one of Africa's most prosperous countries. With a relatively small population of only 14.5 million, it should be able to boast indicators of human well-being among the very best in Sub-Saharan Africa. Instead, war and mismanagement have led to the massive squandering of oil resources, while severely hindering the effective use of most other resources (notably in agriculture).

Although various non-economic factors were important in generating and fanning the conflict in Angola, from the early 1990s, the conflict became a raw struggle for power between rival domestic elites, in which minerals provided both the prize of victory and the means for achieving it. Due to the increase in oil production, government oil revenue, which had been less than $1 billion a year for most of the 1980s, was running at an average of just over $2.5 billion a year by 1995-2001. As for the National Union for the Total Independence of Angola (UNITA), its success in controlling the most valuable diamond mining areas in the north-east of the country enabled the rebels to market up to $3.7 billion worth of diamonds between 1992 and 1998, generating an estimated $2 billion in net revenue with which to bankroll its war.

Oil has undoubtedly also been at the heart of the Cabinda conflict. This small province accounts for more than 55 per cent of Angola's oil production. As a result, the cause of separation, championed over many years by various factions of the Front for the Liberation of the Enclave of Cabinda (FLEC), is bolstered by the prospect of financial viability. For the same reason, however, no Angolan government could ever accept Cabindan secession.

There is nothing inevitable about the 'resource curse'. Indeed, Angola now has an unprecedented opportunity to overcome the legacy of war. First, the peace agreement of April 2002 consecrated what amounted to a military victory by the government over UNITA, leaving little scope for the losers to return to war in the short to medium term.

Second, the rapid rise in oil production and government revenues in the next few years will provide exceptionally favourable conditions for reconstruction, development and poverty reduction – if these resources are managed well and put to better use than in the past. Following major new oil discoveries in deep water off the Angolan coast since the mid-1990s, investments currently underway will more than double oil production in the next few years, to 2.2 billion b/d in 2008. This will take Angola's oil production to about the same level as Nigeria's and it is quite possible that Angola will eventually overtake Nigeria as Sub-Saharan Africa's main oil producer. The potential benefits are all too evident given that Angola has only about one tenth of Nigeria's population.

The direct benefits of this industry to Angolans are actually quite limited. Capital-intensive, it employs only about 15,000 Angolans. Domestic backward and forward linkages to oil supply industries and refining within the country are also weak. The economic importance of oil therefore lies almost entirely in its role in generating fiscal revenue for the state. Indeed, 80 per cent of government revenue came from the oil sector in 2001 and this revenue is set to rise spectacularly in the next few years, as shown in Figure 1, which is based on projections made by the International Monetary Fund (IMF). Assuming oil prices between $20 and $24 a barrel, the IMF projects that fiscal revenue from oil will rise above $4 billion in 2004 and climb steadily thereafter, reaching about $8 billion by 2008.

Ultimately, the quality of resource management is likely to be the decisive factor in whether Angola succeeds in achieving long-term sustainable peace or succumbs once again to new forms of conflict, generated by factors such as a failure to develop the non-mineral sectors of the economy (and thus provide jobs and sources of income), frustration and anger at continuing high levels of poverty and social inequality and/or rivalries within the elite over access to the oil rent and other economic opportunities. There are two major issues here. The first is the question of where the government's expenditure priorities lie. The second is transparency in the management of public finances.

In the past, security imperatives have resulted in extremely high shares of government expenditure going to defence and security. As a percentage of GDP, defence and security expenditure peaked at 26 per cent (by far the highest level in the world) in 1999, but then fell back to 7 per cent in 2001 and 2002, although this is still very high by international standards. It must be stressed that the Angolan figures are likely to be large under-estimates, since they exclude substantial expenditure on defence and security that is off-budget (taking place outside the government's established rules and procedures for payments and not properly recorded in the government accounts). It is also worrying that the government is continuing to give high priority to defence and security, which were allocated 32 per cent of the budget for 2004.

Over the years, high defence and security expenditure has crowded out expenditure on the social sectors, notably education and health, and on basic infrastructure, such as roads and water supply. These are all critical for economic growth, job creation, poverty reduction and improved human well-being. There has been some recent improvement, with the social sectors' share of total classified expenditure rising from 9 per cent in 1999 to 22 per cent in 2002 – although these figures would be much lower if unclassified expenditure was included, because very little, if any, of that expenditure goes to health, education or other social services. It is also important to note that, within the social sectors, resource allocation has been biased by elite interests, such as funding for overseas scholarships and overseas medical evacuation, at the expense of the most elementary needs of the population (primary health care and basic education).

The second major issue is the inadequate transparency in public finances, which means that it is not totally clear how much revenue the government receives and how it is spent. According to IMF estimates, in 2002, 31 per cent of government expenditure was off-budget. This was the case for 36 per cent of government expenditure between 1998 and 2002. Furthermore, 11 per cent of expenditure in 1998-2002 could not be accounted for at all. This was the discrepancy (calculated by the IMF) between the government's known expenditure (recorded and unrecorded) and known revenue, including financing. There are other closely related transparency issues concerning government revenue and the external debt. With respect to revenue, there are large discrepancies in the data on oil taxes. This problem results partly from the fact that oil companies' taxes (and the 'signature bonuses' for new oil blocks) are sometimes not paid directly to the Treasury but pass through off-shore accounts held by the state oil company, Sonangol. Sonangol's delays in passing on tax payments to the Treasury and the failure to index them (against the dollar), in a context of high domestic inflation, have resulted in large losses to the state. Furthermore, the fact that Sonangol's accounts have never been independently audited means that it is impossible to verify that all taxes paid by the oil companies eventually reach the Treasury. Regarding the external debt, over the past decade the Angolan government has resorted to oil-guaranteed loans to cover its continuing large deficits. By the end of 2000, there was an outstanding stock of about $5 billion of these loans (accounting for roughly one half of Angola's total external debt), which are repaid with earmarked oil shipments. Often these loans, raised by Sonangol on behalf of the state, have not been recorded in the government's debt database and have been used for off-budget expenditures. The lack of transparency makes it difficult for the Angolan government to rally international support for reconstruction. There is a feeling among donors that, given the scale of its oil revenue, the government should be doing more for its own people and, as a minimum, should be able to account properly for the use of its own resources. By way of comparison, it should be noted that the unexplained discrepancy between known revenue and financing on the one hand and recorded expenditure on the other, which averaged $490 million a year in 1998-2002, is more than net overseas development assistance ($359 million a year in 1997-99). The lack of transparency also means that the government cannot provide the credible data required for negotiations in the Paris Club on the restructuring of Angola's large external debt, which, as of December 2002, included $5.3 billion in arrears.

Along with weak government procurement and auditing systems, the lack of transparency about revenue, expenditure and debt creates conditions in which corruption can thrive. Suspicions and accusations abound. In December 2000, the French authorities arrested a Franco-Brazilian businessman, Pierre Falcone, and Jean-Christophe Mitterrand, the son of the former French president, on charges of arms trafficking related to an arms-for-oil deal with Angola in 1993. The charges were later dismissed on legal technicalities, but 'Angolagate', as the affair came to be known, helped to create an aura of suspicion about the opaque world of Angolan oil-guaranteed loans, trade finance and arms contracts. A new scandal broke out in 2003 when the Angolan government appointed Falcone, still under investigation and banned from France and the United States, as its representative at UNESCO. The investigations continue.

A further controversy arose early in 2002 about the use of oil-guaranteed loans in the restructuring of Angola's bilateral debt to Russia. Suspicions regarding these transactions, through bank accounts in Geneva, prompted the Swiss judicial authorities to open an inquiry, which revealed that hundreds of millions of dollars had been paid into the private bank accounts of senior Angolan officials (including the head of state) and foreign businessmen, at the expense of the Russian and Angolan states. Several accounts were frozen.

It is hardly surprising, given such high profile incidents, that Angola is widely regarded as one of the world's most corrupt countries: it ranked 124th out of 133 countries in Transparency International's corruption perceptions index in 2003.

In a broader sense, going beyond outright fraud, corruption encompasses a wide range of practices which, in a system characterised by the absence or non-enforcement of clear rules and procedures, enable well-connected individuals to obtain privileged access to resources and benefits, at the expense of wider societal interests. In Angola, where the dispensation of patronage has been one of the key strategies for conserving political power, such practices have been commonplace, particularly with respect to diamonds and commercial farmland. Corruption has also 'trickled down' and is common in many everyday encounters, such as those between teachers and students, police and citizens. It is common practice in many transactions to ask for a gazosa (literally a 'soft drink'), meaning a bribe, although this so-called petty corruption is not just a case of lower level officials copying their superiors, but is a survival strategy adopted by extremely poorly paid Angolan officials.

Of course, corruption is not a one-way street. It involves both the 'corrupter' and the 'corrupted' and, as Angolagate and other scandals show, foreigners as well as Angolans are deeply involved. For this reason, international action is needed to prevent international criminal elements colluding with corrupt government officials to divert resources illicitly from countries where the institutional framework to prevent, detect and punish corruption is weak. Ultimately, however, Angola needs to defend itself by building just such an institutional framework, including rigorous Treasury, public accounting, procurement and auditing systems, effective scrutiny of public finances by the National Assembly and an independent judicial system.

Quite apart from their role in sustaining UNITA during the 1990s, diamonds have been one of the main sources of enrichment of senior government officials and military officers over the past decade. The regime has awarded diamond concessions as a means of rewarding loyalty, creating classic rent-seeking situations in which the Angolan concessionaires act as 'sleeping partners' with foreign companies, which bear all the prospecting, development and operating costs (and thus all the risks), while sharing the profits. Meanwhile, smuggling of diamonds has remained a major problem: the UN Monitoring Mechanism on Sanctions against UNITA estimated that, in 2000, about one third of the estimated $1 billion worth of diamonds exported from Angola were smuggled, including about $100 million worth of diamonds originating from UNITA and $250 million worth of other illegal diamonds. Despite rising somewhat in recent years, government revenue from the diamond industry remains very low. The 'scramble' for diamond wealth, which has been conducted mainly by outsiders, has fostered resentment among the local Lunda-Chokwe population in the diamond-rich provinces of Lunda Norte and Lunda Sul. This has been reflected in the rise of a regional party, the Social Renewal Party (PRS), which came second in both provinces (behind the MPLA but ahead of UNITA) in the 1992 parliamentary elections.

Due to the destructive nature of past external involvement in Angola, foreigners are rather poorly placed to influence resource management practices for the better. Generally, they have little credibility within the country: most Angolans assume that foreign involvement is motivated entirely by the desire to profit from the country's oil, diamonds and other resources. This is of course absolutely true for the multinational corporations engaged in Angola. By the very nature of their commercial interests (and their fear of losing mineral concessions or business permits), they are in a weak position to question government policies or practices. Most would argue that in any case this is not a legitimate role for them to perform. However, large international companies have increasingly come under pressure in the developed world to act within the norms of corporate social responsibility (CSR). Most of the major international companies operating in Angola have tried to cultivate an image of CSR by engaging in small-scale philanthropy. At its most misguided, this has taken the form of support to the Eduardo dos Santos Foundation (FESA), a body with a purportedly charitable purpose whose main function is to promote the beneficent image of the head of state. Most of the oil companies channel some of their philanthropic assistance through a 'social fund' managed by Sonangol (a so-called 'social bonus' has to be paid to this fund alongside the signature bonuses paid to the state for new oil blocks), while some also finance projects sponsored by NGOs and UN agencies, for activities ranging from demining to localised community development projects. Worthy though many of these activities are, in financial terms they represent a minuscule fraction of the profits these companies make (or hope to make) from their investments in Angola and they also do not directly address the resource management issues that ultimately are far more important to the Angolan people's well-being. More far-reaching are the initiatives that have been taken, at an international level, to curb the trade in 'conflict diamonds' and to promote transparency in the oil industry. Although worldwide in scope, both initiatives have been strongly influenced by the situation in Angola. In the first instance, the work of the British-based NGO Global Witness helped to strengthen implementation of the UN sanctions against UNITA imposed in 1998 (a ban on the purchase of unofficial Angolan diamonds and the freezing of UNITA bank accounts). Although not fully effective, the efforts to give teeth to the sanctions by setting up a monitoring regime did contribute to raising UNITA's transaction costs and thus diminishing its resources for war. Significantly, this was accompanied by a major shift in the role played by De Beers, which controls about 65 per cent of the world trade in rough diamonds. During the 1990s, De Beers had been systematically buying up smuggled diamonds from African conflict zones, including Angola, in accordance with its policy of acting as buyer of last resort, a role it had played since the 1930s to stabilise the world diamond market. Fearful that it would become the object of an international consumer backlash, De Beers decided in 2000 to take a strong stand against conflict diamonds and joined the Fatal Transactions campaign in efforts to curb the smuggling of these diamonds, through the introduction of the Kimberley Process Certification Scheme. The scheme was finally launched in January 2003, after three years of negotiations among governments, the diamond industry and NGOs, although both De Beers and NGOs criticised the failure to set up an effective, independent monitoring mechanism, an omission which could fatally undermine the scheme's credibility. Little, if any attention, has been given meanwhile to the resource management issues affecting the Angolan diamond industry, such as the patrimonial nature of diamond concessions and the potential for conflict between outside interests (Angolan concessionaires, foreign mining companies and traders) and local communities in the diamond rich areas. Apart from the issue of conflict diamonds, the main focus of international attention has been on the need for full and open disclosure of the tax and royalty payments made by oil and mining companies in the developing world. Internationally, this has been championed by the Publish What You Pay (PWYP) coalition of NGOs and by the Extractive Industries Transparency Initiative (EITI) promoted by the British prime minister, Tony Blair. However, consultations among governments and oil and mining companies have resulted in the rejection of a compulsory international framework, such as the one demanded by PWYP, which would require companies to disclose all their payments. A voluntary scheme, advocated by EITI, is unlikely to have any practical effect, as individual companies will not risk disclosing their payments unless all their rivals are obliged to do the same. Indeed, the potential risks of individual voluntary disclosure were brought home starkly to one oil company, British Petroleum (BP), in Angola in 2001. When BP decided unilaterally to publish the value of taxes paid to the Angolan government, Sonangol accused the company of breaking confidentiality clauses in its agreements and threatened to terminate its contracts. Full disclosure of tax payments would go only part of the way to improve transparency. While it would help clarify how much revenue is received by the Angolan state, it would not necessarily result in transparent management of those resources. On this broader issue, the IMF has been trying to bring about reforms in the management of public finances, notably through two 'staff monitored programmes', in 1995 and 2000-01. Both were unsuccessful, due in large part to the failure to bring all expenditure on-budget. Over the years, large and rising oil revenues have enabled the Angolan government, unlike the governments of poorer African countries, to keep IMF conditionality at bay and avoid fundamental reforms in public finance management, despite serious macroeconomic imbalances and the large external debt. This already weak external leverage will become even weaker as oil revenues soar in the next few years. Ultimately, fundamental change in resource management in Angola will come not from outside, but from within, as Angolans assert their right to benefit from the exploitation of their country's natural resources. However, by providing access to information and greater awareness of the nature of the problems facing resource-rich countries, the international pro-transparency campaign can assist those within the country (whether in parliament, in the press, in the churches, in professional associations, trade unions and NGOs) who are beginning to press for full transparency and better use of the rapidly rising resources available for reconstruction and poverty reduction.