BIRTH OF A NATION

To understand the subject of this paper fully, we must begin with a brief look at the political economy of late colonial Angola. The start of the anti-colonial war in 1961 was characterised by extreme brutality, and elicited a ferocious response by the large metropolitan expeditionary corps sent to quell it. Yet contrary to Guinea-Bissau, and to a lesser extent Mozambique, where the insurgencies that erupted soon afterwards eventually took a heavy toll on local economies, Angola settled into a costly but low-intensity war that did not affect the cities or even the more populated and richer agricultural areas. The nationalist camp, divided into two and then three rival groups, was feeble and spent much of the time in internecine conflict. These were the coastal and urban Movimento Popular de Libertação de Angola (MPLA), with a leftist outlook and soi-disant pan-Angolan discourse, attracting additional support from mixed-race and a small number of young white Angolans; and the more ethnically based Frente Nacional de Libertação Angola (FNLA) and União Nacional para a Independência Total de Angola (UNITA), drawing the majority of their supporters from Bakongo and Ovimbundu communities respectively and making only limited efforts to enlarge these constituencies.Footnote ² Their long-standing and mutually exclusive claims to supremacy (rather than the greed-driven explanations provided by analysts in the 1990s) would furnish the leitmotif for four decades of intra-Angolan conflict which only came to an end with UNITA leader Jonas Savimbi's death in 2002.

Despite its military weakness, the nationalist challenge played a vital role in Angolan late colonial politics, pushing the Portuguese ‘substantially to alter the nature of their presence’ through an equivocal ‘hearts and minds’ campaign and, more importantly for our purposes here, major economic reforms (Clarence-Smith Reference Clarence-Smith1985: 194). This included opening up the previously protected economy, major metropolitan infrastructure investments, and the doubling of the European population to 350–400,000 (the second in sub-Saharan Africa after South Africa). As a consequence, the economy of the late colonial period was vibrant and diversified, posting annual real growth rates of 4·7% between 1961 and 1974,Footnote ³ despite the war raging in the remote provinces. Sisal, coffee, cotton, diamonds and iron ore were amongst the main exports, while oil, the extraction of which had started in the mid-1950s, became the major export in 1973 at 150,000 bpd. Angola also possessed a fast-growing manufacturing sector centred on consumer goods and light industry.Footnote ⁴

This process came to a halt with the 25 April 1974 democratic revolution in Lisbon. The Portuguese armed forces' decision not to go on fighting in Africa quickly led to talks with the nationalists and a commitment to decolonisation. However, soon after the Alvor peace agreement brokered between the Portuguese government and the three competing nationalist factions in early 1975, it became clear that peace would not hold. This resulted in a three-way war for control of the state, the mass exodus of an estimated 90% of the settler population, and the invasion of Angolan territory by South African, Zairian and Cuban troops, each in support of one of the competing parties. Barricaded in its Luanda stronghold on the day of independence, 11 November 1975, the MPLA and its Cuban allies prevailed to acquire control of the postcolonial state (Heimer Reference Heimer1979). The FNLA was routed and eclipsed as a major nationalist force, and UNITA retreated to its hinterland hideouts.

MPLA's President Agostinho Neto, however, soon realised that his country was in tatters. Angola remained under the threat of South African aggression and UNITA rebels; a great deal of infrastructure was destroyed. Most importantly, the vast majority of educated and technically able personnel, without whom a modern administration and economy cannot run, had fled abroad. Because of this vacuum, but primarily because it fitted the ideological assumptions of many MPLA cadres and their Soviet bloc patrons, the decision was made to expropriate Portuguese property without compensation and introduce central planning to the economy. This option was deepened by the official adoption by the MPLA of Marxist-Leninism in 1976, and its transformation into a purged ‘vanguard party’ in 1977 following the failed Nito Alves coup.Footnote ⁵ The practices and consequences of socialist management – heightened by the patrimonial practices recurrent in the running of postcolonial African administrations and parastatal companies (see Tangri Reference Tangri1999) – were catastrophic, if unsurprising, and no colonial-era economic sector of significance was ever successfully nursed back to life. Nationalised companies were inefficiently run and plundered for cash by unqualified political appointees, collectivised farms mismanaged and then abandoned, and most productive sectors cast off one by one (see Ennes Ferreira Reference Ennes1999). By the late 1970s, the chimera of ‘1973 output levels’ had become the mantra of civil servants, yet economic policy would not vary significantly until the late 1980s, when expropriation of the little Portuguese property left was still occurring.

There was, however, one significant exception to this, which is all the more surprising in view of the suicidal character of Angolan economic management during this period. Angola's oil sector would be spared from Marxist and postcolonial African economics alike, and allowed to flourish according to ‘modern’, ‘rational’ and ‘pragmatic’ criteria. The instrument for this policy was to be Sonangol, Angola's national oil company.

CREATION OF SONANGOL

As the country geared up for civil war in mid-1975, the MPLA put together a National Commission for the Restructuring of the Petroleum Sector (henceforth, ‘the Commission’) under the responsibility of Percy Freudenthal, a white Angolan businessman whose family had been close to MPLA leader Agostinho Neto since the 1950s. Together with Morais Guerra, a lawyer with banking sector expertise, Desidério Costa, an engineer, and a handful of other experts, Freudenthal's mission was to manage the transition in the oil sector in as smooth a manner as possible. Officially, the MPLA's approach to oil companies was still textbook 1970s resource nationalism: a statement of the period brazenly claims that ‘with the inevitable independence of Angola all those [oil] companies which operate offshore or inland will be chased from our national territory and all their equipment and assets seized’ (quoted in Africa Confidential 22.11.1974). But the exodus of white settlers and the impending civil war and foreign invasions made it obvious to the party leadership that oil production would soon be the pre-eminent, indeed the only, revenue-generating sector of the Angolan economy. For that reason, it had to be carefully sheltered from the coming political storms.

The earliest and most pressing task for the Commission was to arrange the return of the main oil operator, Gulf Oil, whose subsidiary the Cabinda Gulf Oil Company (CABGOC) was responsible for the bulk of Angola's oil production.Footnote ⁶ Gulf had exited Angola in November 1975 as the war raged. Through a mix of pressure from General Obasanjo's Nigeria, which hinted at retaliation against Gulf interests in the Niger Delta should Gulf boycott the new state,Footnote ⁷ and reassurance of pragmatic, business-minded intentions on the Angolan side, the Commission managed to bring the American company back into the new sworn enemy of the USA. By March 1976, Gulf Oil had paid an outstanding share of royalties to the MPLA, thus recognising it as the legitimate government of the country despite Henry Kissinger's vehement opposition.Footnote ⁸ Other companies active in Angola such as Petrofina and Texaco were also encouraged to return by their apparently reliable, professional Angolan interlocutors. Nationalisation of the oil sector was explicitly rejected.

The exception to this was the oil company ANGOL, a subsidiary of Portugal's SACOR which was just then in the process of nationalisation by the leftist government in Lisbon (SACOR was absorbed by the recently established Portuguese NOC, Petrogal). In accordance with the decision to expropriate all Portuguese property (Ennes Ferreira Reference Ennes2002), the MPLA took control of the company.Footnote ⁹ But behind this apparently radical measure lurks a transition more typical of conservative Francophone countries such as Gabon or Côte d'Ivoire than avowedly radical Angola, which testifies to the sang-froid of the Commission. At a time when the Portuguese government had not even recognised the MPLA as the legitimate government of Angola (the Portuguese cautiously waited till late February 1976, by which time the other two armed parties had been repulsed), members of the Commission had gone to Lisbon and signed a memorandum of understanding with SACOR/Petrogal to the effect that the Angolan NOC they were about to create would keep many of ANGOL's (mostly Portuguese) employees. According to this agreement, they would staff the new company and their years there would be counted as service to Petrogal for the purposes of promotion and retirement benefits when and if they decided to return to Lisbon.Footnote ¹⁰ Many simply stayed on and became Sonangol employees. Once this was settled, the founding team moved into the ANGOL building in downtown Luanda and proceeded to craft the new Angolan NOC.Footnote ¹¹ ‘In no other area of the Angolan economy was there such a degree of continuity, with both structures and people kept in place’, one Sonangol executive who started out in pre-1974 ANGOL put it. ‘You see, through colonialism, foreign invasion, Marxist-Leninism and capitalism, I have not left the same building.’Footnote ¹²

Sonangol itself was created in June 1976 as the Angolan oil concessionaire, sector regulator, and tax gathering agent, and although a Petroleum Ministry was later established, it was always a political lightweight and never rivalled the NOC's influence. According to a member of the founding group, the vision that animated the new company can be described as ‘flexibility, reliability, and pragmatism’.Footnote ¹³ Sonangol would not embrace the central planning and state-led economic policies that shaped Angola until the end of the Cold War: its ‘compass was the international oil economy, not domestic policy fads’.Footnote ¹⁴ In particular, it was to be insulated from the political race for a ‘share of the spoils’, the anti-corporate mentality, and the disregard for the rule of law that would soon become prevalent in all other sectors of the economy.Footnote ¹⁵ Sonangol executives also understood that with a Soviet and Cuban-supported government and the overt hostility of the US, they would have to go out of their way to appear trustworthy to Western oil investors. Finally, Sonangol did not define its mission in an overambitious manner, despite occasional references to the goal of creating a ‘fully integrated’ oil company. Acutely aware of the complexity of the oil sector and the paucity of human resources in Angola, Sonangol's priority became contractual negotiations with oil firms, while the latter would remain responsible for exploration and production. In order to improve its negotiating performance, Sonangol was committed to close partnerships with the best oil experts, lawyers and consulting firms available, an approach soon extended to any technical area the company was not self-confident about.Footnote ¹⁶

An important such partnership was with the Algerian NOC SONATRACH.Footnote ¹⁷ In May 1976, a high-level delegation arrived in Luanda with a view to helping the new government create Sonangol. Algerian experts became all-purpose advisors to Sonangol in ‘coming to terms with the practical side of things’,Footnote ¹⁸ and helped train a considerable number of first-generation Angolan personnel. Italy's ENI also played a role by providing technical education to many Sonangol workers at its Milan training institute. In addition to SONATRACH, Sonangol's rather surprising other major source of apprenticeship was Arthur D. Little, the reputable Cambridge, Massachusetts consulting firm. Arthur D. Little had been a much-appreciated oil sector auditor to the colonial government of Angola,Footnote ¹⁹ and was keen to remain involved in the country despite the US non-recognition of its new Marxist government. President Neto reportedly quizzed the Commission about the maintenance of such ‘imperialist’ advisors. But he was reassured to know that the Algerians (who also worked closely with Arthur D. Little) spoke positively of them and thus allowed the arrangement to stand.Footnote ²⁰

How could these men get away with creating a company that not only flouted most of the received economic wisdom of the period, at least in the socialist camp, but also cavorted with American oil and consulting firms, all the while exerting the pivotal role in the otherwise stillborn postcolonial Angolan economy? Several factors account for this. To start with, it had to do with the individuals in question. The key figures in Sonangol's core team were well-networked MPLA party members who enjoyed the confidence of President Neto. They shared the common social background of the upper ranks of the party, the exceedingly small, late colonial world of Luanda-based, mostly mixed-race educated Angolans. In this confined social circle, many influential actors in the government, the party and Sonangol were related by blood ties or by friendship. No matter how unorthodox their views may have sounded, their loyalty was never in doubt, a key feature of Sonangol officials up to the present time.Footnote ²¹ They were also respected as technocrats at a time when the party possessed very few university-educated cadres. More importantly, however, they enjoyed strong political support for their project. There was a general perception that the oil sector was a matter of life-or-death, ensuring the viability of the MPLA state and paying for its Cuban protectors. Whatever the management style, if it permitted the steady flow of resources it would be accepted. The Angolan economist Manuel Ennes Ferreira (2005: 5) sees this as evidence of the ‘realeconomik’ thinking of the MPLA leadership, whereby the oil sector was explicitly excluded from the domain of socialist policy-making and indeed made to finance its follies as well as the war effort.

The Marxist period of the MPLA that would last until the end of the Cold War was not one of mere salvaging of the oil sector from the nadir of the 1975–76 conflict. In addition to creating the legal framework for oil investment in AngolaFootnote ²² – including the state's ownership of all hydrocarbon resources – Sonangol and its foreign partners managed to re-establish and, by 1983, surpass late colonial oil production levels. In order to eliminate ‘possible agents and intermediaries in the marketing process’, Sonangol created Sonangol Limited, a London office responsible for the direct trading of Angola's share of the oil (an estimated 40% of the total) and the first of several Sonangol subsidiaries to span the globe.Footnote ²³ In its first decade of operations, Sonangol also managed to diversify the number of foreign investors in Angola's upstream (primarily through Elf-Aquitaine's acquisition of bloc 3 in 1979), although the largest share of production remained with CABGOC, which was bought out by Chevron in 1984. Not that this dependence on one company constituted a burden: Gulf and then Chevron (their operations protected from US-backed UNITA rebels by Cuban troops and Soviet weaponry) proved to be reliable allies for the duration of eighteen years of US enmity. As early as 1979, Gulf's impression of Angola's government as able ‘to understand the difference between a multinational and its home government’ was being publicly conveyed, and the hostility of the Reagan Administration did not change the thrust of its Angola engagement.Footnote ²⁴

Freudenthal's tenure ended a mere three years after the creation of Sonangol, partly, MPLA insiders have it, due to the fact that he was white, race being at that time an important fault-line in a party where mixed-race and white cadres had a disproportionate amount of power.Footnote ²⁵ It was not immediately obvious that the company would continue to be run in the same way: the second CEO of Sonangol, Hermínio Escórcio, was perceived as brash and ostentatious, and owed the position to his Politburo membership and good rapport with new President Eduardo dos Santos, rather than any particular managerial or oil sector expertise.Footnote ²⁶ But the fine technical team stayed on, enlarged by new legal, engineering, geological and economic expertise coming out of Sonangol's many educational partnerships, and Escórcio was always respectful towards them, never tampering with their responsibilities. Most importantly, there was never any competing attempt by Escórcio or anyone else to carve out a constituency and distribute patronage on the basis of oil money: Sonangol was under the political control of the Angolan presidency throughout this period as a trustworthy instrument of its interests. The trend was reinforced during the long tenure of Joaquim David (1989–99), a respected US- and UK-trained petroleum engineer.

By the late 1980s, an evidently amazed World Bank team reported ‘the shortage of skilled and trained manpower in Angola’, but went on to say that the petroleum sector was the least affected by the country's dearth of managerial and technical expertise. ‘So far Sonangol's high- and medium-level management positions have been staffed with comparatively experienced and competent personnel … Sonangol has [also] effectively and advantageously used external consultants to supplement its capabilities in all aspects of its operations.’ The report concludes that ‘general government policies in the oil sector have been enlightened and thus deservedly successful’ (World Bank 1989: viii–x, 16).

THE SONANGOL ‘UNIVERSO’

When the MPLA dropped its Marxist garb at the beginning of the 1990s and the ruling elite enthusiastically converted to crony capitalism (see Ennes Ferreira Reference Ennes1995), Sonangol was the leading domestic entity in Angola's political economy. This role, and the oil sector that underpins it, progressed on an essentially separate, upward path unaffected by the failed democratic experience of 1992, and UNITA's return to a civil war that would kill hundreds of thousands in only two years and continue intermittently for another decade. The rebel occupation of an estimated 80% of Angolan territory in 1993, for instance, did not damage Sonangol or the predominantly offshore oil production. (That said, Sonangol did play an absolutely vital role in the government's prosecution of the war effort through the acquisition of weaponry.) Instead, the 1990s were characterised by growing investor diversification and rising interest in the country's deep and ultra-deep waters, which new technology was then making accessible for the first time. In 1996 Elf made Angola's first ultra-deep water discovery, the Girassol field, and many followed soon afterwards. Angola quickly became the hotspot for companies from all over the world, including British Petroleum, Royal Dutch/Shell, ExxonMobil, Statoil, Norsk Hydro, Petrobrás, Marathon, the Chinese companies CNOOC and Sinopec, and a myriad of other firms of all shapes and sizes purposely drawn into the Angolan ‘chessboard’ by Sonangol.Footnote ²⁷ According to the International Energy Agency (2006: 84), the resulting level of investment is unheard of outside a few top oil producers worldwide: between 2003 and 2008 alone, it estimates that between US$17 and US$23 billion in FDI will have been brought into the Angolan oil sector. Oil production has grown throughout, and will continue to do so for the foreseeable future (see table 1).

Table 1

Angola's oil production

As the pivotal domestic institution in the sector that earns the Angolan government more than 90% of its revenues, Sonangol has at its disposal financial means and human resources that no other branch of the administration can rival. Sonangol's role in Angolan public life was further consolidated by its 1991 restructuring into a holding company, Sonangol EP or the Sonangol Group, containing various subsidiaries. It thus acquired its present form: that of a constellation of worldwide business interests that could be termed the Sonangol ‘Universo’.Footnote ²⁸ What follows is a non-exhaustive list of its activities.

The Sonangol Group directly employs an estimated 5,000 peopleFootnote ²⁹ and contains several major subsidiaries,Footnote ³⁰ all of which are subject to ‘preferential treatment … in the procurement of goods and services to oil companies operating in Angola’ (Alvesson et al. Reference Alvesson, Bhattarai and Pastor2003: 82). These are Sonangol Pesquisa e Produção (upstream activities); Sonangol Distribuidora (downstream activities, with affiliated businesses in the Democratic Republic of Congo, Portugal, Cape Verde and São Tomé & Príncipe); Sonangol Logística; Essa, a provider of professional training for the oil industry; SonAir, an air transport company catering to the oil industry that includes flights from Houston, Texas to Luanda; Mercury, a telecommunications group; Sonaship and Sonangol Shipping, two providers of maritime transportation; AAA, an insurance company described as providing ‘risk management for the oil industry, insurance brokerage and pension fund management’; Sonangol USA, Sonangol Limited (UK), and Sonasia (Singapore), marketing and trading units; and the Hong Kong-based China Sonangol International Holding, which in November 2004 joined a Chinese-Argentine offer to invest up to US$5 billion in Argentine oil exploration (Miami Herald 23.11.2004).

As part of an Angolanisation campaign, Sonangol has also established many joint ventures with foreign companies (some estimates are as high as sixty but no accurate figure is available).Footnote ³¹ High-profile joint ventures include Sonangol-Sinopec International (SSI), which recently paid a record US$2.2 billion for a 40% controlling stake in parts of offshore oil blocks 17 and 18 (Africa Confidential 7.7.2006);Footnote ³² Sonangol SGPS, a drilling services company; Sonasing, a joint-venture service company for the packaging and storage of crude; Wapo Angola, a services provider for the oil industry; Technip Angola, Petromar, Sonamet, Sonansurf and Sonamer, oil services companies; AngloFlex, a manufacturer of umbilicals and pipelines for underwater production systems in the oil and gas industry; the Banco Africano de Investimento, a bank; and many others. More recently, Sonangol's investment drive has reached even further away from its core business, and now includes hotel ownership and management, and the building of luxury housing. These activities are compounded by a host of other roles that Sonangol plays. In addition to a well-endowed bi-yearly literary prize and hundreds of scholarships for the children of insiders (including high-school education in Portugal and English-language training at Indiana University),Footnote ³³ the company has been a major patron of most cultural goings-on in Luanda and the lead sponsor of two football teams, Petro-Atlético de Luanda and Atlético de Namibe. In 2007, Sonangol was a lead sponsor of the Africa Pavilion (in reality, the collection of a Luanda-based private collector) at the Venice Biennale contemporary art event.

Sonangol's name is highly respected amongst its African counterparts: in the past five years alone, Sonangol advisory teams have been sent to Equatorial Guinea, Nigeria, Côte d'Ivoire, São Tomé & Principe and Gabon, while Sonangol assistance and the Sonangol example more generally have been crucial in the setting up of Congo-Brazzaville's SNPC.Footnote ³⁴ Despite international civil society criticism on the transparency issue (on which more in the next section), most Western banks and oil companies speak very well of Sonangol and report hassle-free interactions with the company. From their perspective, competence, predictability and a measure of mutual trust have long ago filled in for Sonangol's opaqueness. A director at Standard Chartered Bank, for instance, extolled the ‘excellent reputation and track record of Sonangol’ on the occasion of the largest-ever oil-backed syndicated loan in 2005.Footnote ³⁵ An oil company executive referred to the ‘unique amount of trust created by Sonangol's use of virtually the same team for twenty years’.Footnote ³⁶ A top oil executive of a major European firm with decades of involvement in Angola did not hesitate to say that ‘Sonangol is the Angolan miracle.’Footnote ³⁷ Problems that do show up between the oil companies and Sonangol concern the Group's assertiveness. Recent contentious issues included revisions of the Angolan Petroleum Law to increase local content, Sonangol's stated goal of slowing project developments,Footnote ³⁸ accusations that some companies have gone for ‘unduly costly technical options’, and demands for ever-more substantial signature bonuses. Ironically, these occasional conflicts are also a mark of the hard bargaining and relative technical competence that puts Sonangol in a more equitable position vis-à-vis foreign operators than any other NOC in the Gulf of Guinea region.

Riding high on the company's good reputation, the current CEO, Manuel Vicente, has even talked of creating a publicly traded company for flotation in the South African and New York stock exchanges by 2010.Footnote ³⁹ The Group certainly makes strenuous efforts to peddle its ‘modern’ image, as can be gauged from its many glossy publications, carefully manicured road-shows, lavish expenditure on lobbyists in Western capitals,Footnote ⁴⁰ and soundbite references to corporate social responsibility and other business-speak.Footnote ⁴¹ Sonangol's tasks, interests and footprint in Angola by the turn of the century were thus already exceedingly ambitious. But with the end of the conflict in 2002, the gradual shifting of Angola's war economy to a peacetime footing, high oil prices and increased production, the mission envisaged for Sonangol has been extended further. No longer merely an able negotiator of the best oil contracts its expertise and Angolan leeway can afford, the Group is now seen as the ‘engine’ and ‘promoter of national development’ (Guerra Reference Guerra2003: 23): the senior partner in even more joint ventures with foreign companies in which, for the first time, a new Angolan business class is meant to participate. Already Angolan-owned companies such as PRODOIL, ESAP, Poliedro, Majova and SOMOIL have been allowed into the oil sector, in hopes that within a decade, Angolan companies will be able to act as full partners in oil production.Footnote ⁴²

THE OTHER SIDE OF SONANGOL

What, then, is the other side of the coin to this seemingly exemplary instance of a well-governed, successful developing-world state corporation? In one word, it is that Sonangol's undeniable competence and sophistication are not, and have never been, put at the service of Angolan development, however defined. Instead, the company is the pivotal tool for the interests of the presidential clique known as the Futungo de Belas.Footnote ⁴³ The Futungo, a nebulous group of unelected officials and businessmen around President Eduardo dos Santos, became the key structure of power in the 1980s, in tandem with the relative sidelining of MPLA party organs and formal state structures. Sonangol essentially exists to harness and further their agenda. According to Global Witness, a campaigning NGO, Futungo mechanisms of misappropriation have included ‘offshore money laundering, over-priced military procurement and a deliberately opaque method of running up debts against future oil production’ (Global Witness 2004: 36). At the centre of all of them lies Sonangol. ‘If Futungo … was the brain of the system’, writes Nicholas Shaxson, ‘then Sonangol was its heart’ (Shaxson Reference Shaxson2007: 4).

From the mid-1980s, when a dip in oil prices led Angola to borrow from Western banks against future oil production for the first time, Sonangol has been the centrepiece of the presidency's vast system of parallel finances that has included up to half of Angola's yearly oil revenues. Oil-backed loans were initially used to pay for Sonangol's cash calls (and, from 1992, to acquire weaponry), but were soon ‘deployed for general government use’ (Vines et al. Reference Vines, Shaxson, Rimli and Heymans2005: 16–17). All of these operations were run through an extensive network of offshore accounts from which commissions were then paid to regime notables and foreign middlemen. Because such deals stipulate that loans should be ‘serviced through the direct delivery of oil’, Sonangol is ‘inevitably’ the Angolan party to these loans (Alvesson et al. Reference Alvesson, Bhattarai and Pastor2003: 80), and Sonangol's credibility as a business partner plays an important role here. Loans mostly take place through Special Purpose Vehicles, which are offshore financial tools that allow lenders to benefit from safe repayment structures through oil price protection, debt service reserve accounts and an accelerated repayment mechanism. Oil-backed loans are also syndicated (i.e. shared by a large number of banks), which further decreases risk for the creditor, and their short maturity of three to five years as well as inordinately high interest rates mean that there is never a lack of lending interest. While extremely onerous for the lender (the Angolan state), oil-backed loans provide the state elite with an easy source of money outside any sort of control.

This Sonangol and Futungo-centred ‘parallel state’ is vastly more efficient than Angola's dilapidated state structures, and can take on complex operations involving foreign partners in a way that the Finance Ministry or the Bank of Angola could not. However, this has contributed towards both the marginalisation of formal institutions in Angola and the increase of Futungo power,Footnote ⁴⁴ with the latter essentially misappropriating a large percentage of the sums involved. The resulting black hole in Angolan state finances, into which, according to the IMF, an extraordinary US$4.22 billion disappeared between 1997 and 2002 (IMF 2002: 47), has been dubbed the ‘Bermuda Triangle’ (Hodges Reference Hodges2004). The Angolan government has claimed that the major discrepancies found are the product of poor accounting and lack of capacity, not theft. But, Global Witness (2004: 50) quips, ‘a government and state oil company that handle billions of dollars through complex offshore arrangements, including the use of Special Purpose Vehicles and tax havens, can certainly manage a simple balance sheet’.

Sonangol also played a prominent role in the wartime acquisition of arms paid for by future oil production, as revealed by the string of scandals known as ‘Angolagate’ (because of their complexity, they cannot be described here in great detail). As the country re-entered civil war in 1992, the MPLA struggled to rearm itself. This took the form of an oil-for-arms deal which allowed the Futungo and its foreign middlemen, especially the businessmen Pierre Falcone and Arkadi Gaydamak, mainly through over-invoicing, to collect astonishing ‘fees’. When the scandal broke in France, which had been the centre of operations for the arms deals, Falcone was arrested. His release was eventually secured by ‘the largest bail in French legal history’, courtesy of Sonangol (Shaxson Reference Shaxson2007: 11). Finally, especially since the early 1990s, Sonangol has asked oil companies for the payment of a ‘signature bonus’ when bidding for oil blocks. These signature bonuses have reached worldwide records in Angola but disappeared into the company vaults. Indeed, when auditors KPMG were finally allowed (limited) access to Sonangol's books, they found that the primary source of oil revenue discrepancies was not foreign companies but Sonangol itself (HRW 2004: 23–6).

Africa Confidential (19.4.2002) is therefore right on the mark when stating that Sonangol ‘is not a normal company in any sense of the word’. In addition to its key role in the Bermuda triangle, the conflicting roles of concessionaire, equity partner, and operator would by themselves be the source of plenty of conflicts of interest (Vines et al. Reference Vines, Shaxson, Rimli and Heymans2005: 10; see also IEA 2006: 81–110). But Sonangol's noxious influence goes further. The company's tendency to invest directly, or allow insiders to invest, in all branches of Angolan economy however unrelated to its core business leads it to ‘squeeze out competitors which cannot match it in terms of access to capital, skilled personnel or political connections’ (Vines et al. Reference Vines, Shaxson, Rimli and Heymans2005: 10). A look at many characters benefiting from this sort of uncompetitive behaviour is a veritable who's who of the Futungo and the upper ranks of the bureaucracy, the military, the party, and their families. Unsurprisingly, it also includes most current or past senior technocrats of Sonangol itself. Indeed, while interviews with many of these officials are replete with formulaic expressions of a desire to roll back the patrimonial character of Angola's political economy, they do not actually envisage having to compete in a genuinely open market.

In a process reminiscent of 1970s Nigeria, when the laudable desire to build a domestic business class led to the creation of essentially parasitical positions (Nigerian entrepreneurs became silent partners in joint ventures with able foreigners, collected their share, but did not acquire much technical or managerial expertise),Footnote ⁴⁵ Sonangol is busy creating opportunities for well-connected but non-entrepreneurial individuals with little value added for Angola, while genuine entrepreneurs face formidable hurdles in all lines of business.Footnote ⁴⁶ This strategy essentially aims to secure continuing support for the presidency by sharing rentier opportunities and some of the spoils of the oil boom across the elite. It also detracts from the openness and competitiveness of the Angolan economy. By creating potentially hundreds of joint-venture-related opaque money flows instead of a centralised one, it sponsors a veritable ‘downstreaming of corruption’ (Vines et al. Reference Vines, Shaxson, Rimli and Heymans2005: 15) that will make fiscal accountability ever more challenging. In short, the Sonangol Group has been not only the manager of the Futungo's phantom revenue flows, but also the enabler of elite rentierism across important non-oil dimensions of the domestic economy, simultaneously guaranteeing that all profitable business opportunities are arbitrated by state rulers via the company (see Hibou Reference Hibou1998: 107).

This explains why, outside the carefully sheltered oil enclave, Angola is perceived as one of the riskiest economies worldwide in which to invest.Footnote ⁴⁷ The major oil firms, service and construction companies, banks and lawyers that form the mainstay of Sonangol's foreign business partners oblige partly because Angola is too valuable an oil province to pass over, and partly because they are spared the real burden of its monopolistic drive. When foreign partners do mind Sonangol's methods, they have to face up to the company's fierce nationalism and defence of ‘secrecy’. In 2001, Sonangol responded to BP's unprecedented decision to publish the value of its signature bonus by threatening to expel it from Angola.Footnote ⁴⁸ In 2004, Sonangol did not renew Total's concession for oil block 3/80 in apparent retaliation against French ‘Angolagate’ investigations, one of several aggressive stances towards French business interests.Footnote ⁴⁹ In sum, business in Angola happens according to Sonangol standards, which are in turn those of its political masters: otherwise, it does not happen at all. While the seeds of capitalist prosperity are not always pretty to look at, it seems unlikely that Sonangol's methods are conducive to a genuinely open, productive market economy.

GOVERNING A SUCCESSFUL FAILED STATE

Contemporary Angola presents two trends that seem contradictory and not to intersect. On the one hand, despite being sub-Saharan Africa's second largest oil producer, Angola is one of its worse-governed states, with negligible levels of social expenditure, epidemics of mortal diseases galore, famously sticky-fingered elites, and a public administration unable to perform basic sovereign tasks or cover much of the nominal Angolan territory. In the UN's Human Development Index, Angola ranks 160th out of 177 states worldwide: it is, in many ways, a very weak or even ‘failed’ state, at least from the viewpoint of the vast majority of its citizens. On the other hand, Angola's enormously wealthy rulers have successfully withstood major external and internal challenges, created the armed forces of a regional superpower, and developed strong and diversified networks of international support that are likely to guarantee incumbency for years to come. Yet the contradiction is only apparent, both segments amounting to one half of the political and economic makeup of postcolonial Angola. In other words, Angola is what one could describe as a ‘successful failed state’.Footnote ⁵⁰

The empirical element of failure is undeniable. A veritable basket case of dysfunctional state experiences, Angola brings together the pathologies of the colonial and postcolonial African state with the ailments of petro-statehood and socialist management. Over the past fifteen years in particular, the decay of state institutions, the privatisation of power and the abandonment by state elites of tasks they no longer believe are the state's own, especially the provision of public goods, has had an impact on the lives of Angolans that is difficult to overstate. However, the presence of oil deposits changes the calculus of state survival by preserving the Angolan state and other oil states in the region from Somalia-type demise. Oil guarantees it considerable freedom from the international financial institutions,Footnote ⁵¹ and a solidly long-term legal engagement (through the sale of the fuel that powers industrial civilisation) with the international economy. Oil ensures that whatever the domestic political conditions, there will be an interest by multiple external and internal actors in maintaining a notional central structure, and that enough resources will be available to prop up incumbents, guarantee their enrichment, and coerce or co-opt enemies. This allows them to build a political order that is violent, arbitrary, exploitative but fairly reliable. The resulting political process is and will be unstable and fragmentary, but the structure of politics itself will be stable and viable – while oil lasts. This political and material success of the elite does not cancel out the decay around itself but, as Prunier and Gisselquist (Reference Prunier, Gisselquist and Rotberg2003: 103) note, an analysis of failure must allow for the fact that certain forms of governance can be ‘successful as measured against its own parameters and judged by the standards of its political program’.

Provided two conditions are in place, the relationship between the Angolan state and the oil companies will work in a mutually beneficial way, regardless of the surrounding chaos. The first is a prerequisite for foreign involvement in the oil sector: the creation of a parallel economic system that insulates oil companies from the unreliability of local conditions, with its own acceptable legal framework and logistical efficiency. In such enclave contexts, companies can operate freely and do not face the rent-seeking, contractual uncertainty or threat of expropriation that are widespread outside the oil sector. The second condition – essential for elite survival and enjoyment of revenues – is the creation and maintenance of two sorts of state organisation that are spared the decline evident elsewhere. The first pertains to the instruments of coercion, in the form of one of Africa's largest armed forces, numerous police and paramilitary outfits, and an efficient and well-resourced intelligence apparatus active in Angola and abroad. The second is an entity that can articulate state interests in the oil sector with comparative prowess, bringing together its scarce human resources and enabling success in negotiations and joint ventures as well as access to oil-backed loans: in short, Sonangol. Insulated from the rest of the state apparatus and the incompetence of the civil service, conversant in the languages of high finance, business contracts and oil technology, Sonangol is a strange marriage of the latest expertise and market savvy with the narrow enrichment goals of a failed state leadership, and it works. Furthermore, its smooth, transnational sophistication is loyalist vis-à-vis the presidency:Footnote ⁵² the company's ethos is a mix of strong Angolan nationalism and corporate pride (similar to those of other NOCs such as PDVSA, Pemex, Petronas or Petrobrás) which, despite its key political role in Angola, does not phrase itself in a politically intrusive way.

Two questions seemed to be on the minds of the current and former Sonangol officials interviewed, for all tackled them without prompting. The first pertains to Sonangol's chances of survival (or the threat of ‘angolanisation of Sonangol’, as one expert put it), and the second to the potential for expansion of its business model to the rest of the economy (‘the sonangolanisation of Angola’). Regarding the first question, one executive noted the ‘constant struggle’ needed to ‘keep the credible nucleus of the company intact in an African (sic) business environment’.Footnote ⁵³ Another feared the proliferation of joint ventures and the development of ‘fiefdoms inside the Group’.Footnote ⁵⁴ Yet another expressed concern that ‘overly ambitious young executives would disregard negotiations expertise, which has served us well over the years, for the mirage of an integrated company’.Footnote ⁵⁵

To this author at least, the experience of the last thirty years shows that the company's approach can be sustained. This is not simply because it would be unreasonable to kill the proverbial goose with the golden eggs that has so capably served regime interests: after all, that has happened countless times in postcolonial Africa. Rather there seems to be a measure of consensus amongst the Luanda elite that Sonangol is a good thing and that it should be allowed to thrive. After the defeat of UNITA and its openly uncompromising challenge to the MPLA-built state, hypothetical regime changes would occur from within the current state elite itself. While the latter is divided as to the sharing of spoils, there is remarkable agreement on political economy issues and no major changes would occur (due to the paucity of human resources, it is unlikely that senior technical personnel at Sonangol would even be replaced).

The second question on the impact of some positive Sonangol features on the rest of the Angolan economy is more complicated, even assuming that its technical means can be separated from the political ends they serve. Early experiences with building a Sonangol-like state company in the diamond sector, ENDIAMA, failed because of corruption, the portable nature of the commodity, lack of human resources, and the decentralised character of the sector (involving hundreds of legal and illegal entities of many shapes, rather than a small group of world-class foreign companies). Attempts to manage other state companies such as Angola's air carrier TAAG in a capable manner met with only partial success. ‘Entrepreneurial’ efforts by Sonangol-trained executives are overwhelmingly instances of crony capitalism, rather than the beginnings of a business class willing and able to work in (and lobby for) an open business environment. In short, limitations of human resources and of predictability, transparency, and the rule of law in the non-enclave Angolan economy mean that it is impossible to engage in it without encountering its rentier, unproductive dominant logic. Sonangol may well be ‘a strategic island of competence ring-fenced against chaos elsewhere’ (Shaxson Reference Shaxson2007), but it is also very much part of that chaos and ultimately unfit to rise above it.

Does this mean the Sonangol experience cannot be put to good use? This is, of course, a function of Angolan politics: if it takes a more benign turn, then the tightly run and politically obedient Sonangol will doubtless be an asset other African states (including oil states) do not possess, the sort of self-standing and able institution many Western ‘capacity-builders’ can only dream about. This is the way some observers hope it will go. They point to substantial improvements at Sonangol over the past three years, in tandem with Angola's careful engagement with the anti-corruption reform agenda: Angola, for instance, has declared itself an ‘observer’ to the Extractive Industries Transparency Initiative (EITI) process, and speeches by would-be reformists in government are now peppered with talk of transparency.Footnote ⁵⁶ After submitting itself to partial audits by the consulting firm KPMG, Sonangol now has a website and is reportedly seeking to come into line with international accountancy standards, if only for Sonangol EP and not the sprawling empire of joint ventures and subsidiaries. For the first time, the 2003 revised budget included expenditures planned to be executed through Sonangol, and government is now publishing details of oil payments received (World Bank 2006: 44–5). More recently, Finance Minister Pedro de Morais has even claimed that ‘there were no longer any links between creditors and Sonangol’ (Angola News April 2006).

However, as the Economist Intelligence Unit (2005b) notes, it seems that this reformist turn is less ‘an effort to improve openness, although it has done so, and more an attempt by the presidency to make the system of political patronage (often described by outsiders as ‘corruption’) more acceptable and sophisticated rather than to abolish it’. As visions of Luanda as the ‘African Dubai’ take hold and a public works-driven attempt to rebuild the country proceeds apace,Footnote ⁵⁷ there is little doubt that Angola's political economy is in the process of substantial reconfiguration away from the wartime tenets described in this paper (Billon 2001). Yet there is no evidence that the company and those who define its course have fundamentally changed their successful, if non-developmental, vision for Angola and for Sonangol's role in it. In fact, their interest in engaging with the transparency rhetoric may already have peaked amidst all-time high oil prices and a new business partner, China, which has essentially replaced conditionality-ridden OECD donors and Bretton Woods institutions as the source of credit for Angola's ‘reconstruction’, and is unruffled by fashionable Western good governance agendas.Footnote ⁵⁸

For the time being, therefore, there is not much space for Sonangol to play a constructive role. The Group – a bedrock of expertise and income to a regime distinctly uninterested in providing for the majority – may in fact be a liability from the viewpoint of progressive politics in Angola. In a way, this substantiates a point often forgotten by western donors and IFIs in their attempts to depoliticise the undeniably political issue of state capacity: that ‘capacity’ has no intrinsic normative content, and cannot be separated from the political agenda it is made to serve.

*  *  *

Amidst the mostly colonial-era, dirt-covered buildings of decrepit downtown Luanda, one shiny new addition stands out: the twenty-three floors-high, US$131 million Sonangol Tower, complete with a heliport, two gyms, 91 underground parking spaces, an amphitheatre for 300 people, a restaurant for 450, and a fire-proof documentation room, all couched in the ‘latest engineering technologies’ made available by Portuguese, South African and South Korean contractors.Footnote ⁵⁹ The Sonangol Tower symbolises both the current buoyancy of the Sonangol Group and its political masters, and the extent to which its divergent development and methods are a structural feature of Angola's political economy rather than a freak occurrence. Sonangol's continuing rise, though principally the fruit of domestic trends described in this paper, is part and parcel of the general ascendancy of the national oil companies of oil-producing states, which control a large majority of the world's energy resources (Angola itself became a member of OPEC in January 2007) (OPEC 2007). With foreign investors and lending sources keen on deepening their presence in the ‘darling of the global oil industry’,Footnote ⁶⁰ IMF (2006) predictions of 31·4% real GDP growth for 2007, and a generally flaccid international push for transparency in the extractive industries, no external incentive structures are in place to force through meaningful changes in Angola.

This paper is a preliminary attempt to understand the role of Sonangol in Angola's politics, and its insertion in the international economy. Its conclusion is twofold. First, in a country where political and economic power has essentially been privatised, a competent and sophisticated NOC (perhaps the ablest state entity in sub-Saharan Africa excluding South Africa) plays the key role in allowing the Angolan elite's rewarding engagement with the international political economy of oil. This is achieved by separating the bureaucratic management of the company from the dominant rentierist logic of the Angolan economy, and the patrimonial last stop of the revenues it garners. While Sonangol and the agenda of its decision-makers can be understood in the generic terms of what Jean-François Bayart (Reference Bayart1989) called the strategies of ‘extraversion’ of African elites, there is little doubt that it occupies a very special place at the high, technically able, and cosmopolitan end of the spectrum of such strategies. In addition to holding sway over the most important external resource flows, Sonangol ensures that domestic economic opportunities are firmly controlled by those in power, and that independent businesses that could nurture alternative poles of influence do not develop. Sonangol is an exception within the weak, poorly paid and incompetent state apparatus, together with the paramilitary and armed forces that protect the regime. Most other state institutions, including those that should provide for public goods, barely work and are not able to fulfil their functions.

The second point to note is that the story of Sonangol's success is played out against a canvas of state dereliction, civil conflict and utmost impoverishment. Its tools and professional expertise that so many African states lack are put at the service not of broad-based prosperity, but the enrichment of the few. The abundant resources at the disposal of the Angolan government, together with the wise management of Sonangol for thirty years, show that the implosion of state institutions and the lack of provision for the basic needs of average Angolans are also political choices – a conscious divestment of responsibilities towards the population. They are not simply the result of lack of means, even if it is undeniable that Angola suffers from serious shortcomings at all levels which oil wealth alone cannot address. The Sonangol trajectory also shows that, provided that the basic instruments for elite empowerment are in place – tools such as the NOC to manage the Angolan side of the oil relationship and the armed forces to dissuade or crush would-be challengers – the elite of a state that is often deemed ‘failed’ or weak can be ‘successful’ amidst apparently inauspicious circumstances.