Theme: This ARI discusses some of the lessons to be learned from the failure of the model implemented in the Republic of Chad to manage oil revenue flowing into the country’s public coffers since 2003.
Summary: The Chadian model to reduce poverty was portrayed as an opportunity to overcome the ‘resource curse’. A complex institutional system led by the World Bank sought to ensure that oil revenue would be applied to growth and to cutting poverty. However, in a very short time the mechanism proved to be fragile and ended up succumbing to international pressure and the growing political instability in Chad. There are some lessons to be learned from this failure: the limitations of the discourse of good governance and transparency, the importance of more careful planning of the project schedule, the significance of the focus on obsolescing bargaining theory, the excessive complexity of the institutional design and the relevance of external factors, overlooked by the resource curse theory. Lastly, the case of Chad also highlights the emerging geo-energy rivalry between China and the US in Africa.
In October 2003, Chad joined the oil exporters club, on the heels of the largest private investment in a single project in the history of Sub-Saharan Africa. The approximately US$4.2 billion of investment were used to launch operations at the Doba oil fields (in southern Chad, very close to the border with Cameroon), and also to build a 1,076 km oil pipeline to the port of Kribi (Cameroon), on the shores of the Atlantic.
The project, led by a consortium comprising Esso, Chevron and Petronas (the ‘Consortium’), had raised many expectations both domestically and abroad. Previous negative experiences in developing countries warned of the possibility that the project might not turn oil revenue into development, in yet another example of what is known as the ‘resource curse’. Substantial external intervention spearheaded by the World Bank sought to overcome the ‘curse’ by designing a model institutional system, consisting in guaranteeing that a substantial part of oil revenue would be applied to policies to aid growth and help reduce poverty.
In a previous ARI (nr 73/2008) we detailed the model and the relationship between its failure and the intensification of the internal political conflict in Chad. In this ARI we intend to look at the model’s failure from a different standpoint, over a longer time-frame and drawing lessons from the institutional perspective and in terms of the resource curse theory.
In 1988, the first contract was signed between the Consortium and the government, but due to the structural political instability in Chad, until the end of the 1990s the possibility of exploiting the Doba reserves was not seriously considered. The World Bank played a crucial role, without which the project would never have gone ahead. Apart from limited financial support, the Bank acted as a ‘moral guarantor’ and in 1999 it forced approval in the Chad parliament of Loi 001/PR/99, establishing that 85% of direct revenue (royalties and dividends) should be applied to sectors seen as a priority in reducing poverty. In June 2000 the World Bank gave the project the green light and construction on the oil pipeline and the rest of infrastructure began. Finally, months ahead of schedule, in October 2003 operations came on stream. Another of the model’s specific characteristics is that the World Bank also forced through the creation of a expenditure supervision body called Collège de Contrôle et Surveillance des Ressources Pétrolieres, which would manage this 85% of revenue. Members of the Collège were appointed by the government, but also by civilian society, which via NGOs would have a say in the management of resources. This institutional device that aimed to avoid the ‘resource curse’ overlooked some aspects which later proved to be crucial in explaining the project’s failure, such as external factors, initial institutional weakness and structural political instability in the country, but it was nevertheless considered to be sufficient and appropriate.
Chronology of Failure
As early as January 2004, the first problems emerged. After barely three months exporting oil, the arrangement between Chad and the IMF (PRGF) collapsed, and in view of the government’s macroeconomic non-compliances, the IMF froze access to part of the oil revenue, triggering fiscal tensions in the first half of 2004 which ended with an attempted coup in May stemming from a failure to pay wages to the country’s army. Meanwhile, in October of the same year, a press release by the government indicated its frustration with the terms of the contract with the Consortium, urging a renegotiation in an attempt to obtain more access for discretionary use of the funds in tackling political instability.
In 2005, internal political tensions persisted, with the reform of the Constitution in May, enabling Déby to run for re-election; this did not go down well with his possible successors, who ended stirring up a rebellion. Meanwhile, in October Déby announced plans to reform Loi 001/PR/99, and despite opposition from the World Bank, in December Chad’s Parliament approved the reform, which afforded the government greater access to oil revenue for discretionary use, now including security as a priority sector (and thereby paving the way for an increase in military spending). Automatically, the World Bank suspended all payments and froze the oil revenue account.
The situation deteriorated when in April 2006 a rebel attack, which almost ousted Déby, could only be curbed with the aid of a military intervention by France. The prospect of a collapse of the state led the World Bank to accept most of the conditions proposed by the Chadian government, which were specified in July 2006 in a Memorandum of Understanding, effectively signalling the resumption of World Bank payments, and allowing the Chadian government to commit 70% of the 2007 budget to priority sectors. During the rest of 2006 and 2007, Déby’s bargaining position vis-à-vis the Consortium was strengthened, and he successfully pressured for the Consortium to pay more tax. Furthermore, his government became more authoritarian, for example hampering the movement of local NGOs in the oil producing area, remodelling the committee for the management of funds allocated in accordance with its own interest and abolishing the Coordination Nationale du Projet Pétrole.
In February 2008, a major rebel attack on N’Djamena, the capital of Chad, again threatened the Déby regime, but again French military and logistical intervention prevented the government from being overthrown. This time, the government was shored up further by international support, especially from France. In view of all this, the World Bank’s position became untenable in the face of the evidence that the model it had designed had collapsed, and in August of that year it expressed its concern regarding compliance with the Memorandum of Understanding. Finally, on 5 September, Chad paid back its debt to the World Bank in respect of the project (US$ 65.7 million), thereby putting an end to the Bank’s involvement. It was obvious that the Chadian army’s capacity was increasing, as the funds available to the government and the lack of public spending control were enabling it to increase. This increased capacity was confirmed by the army attack on rebel bases in May 2009, which ended in a clear victory for the government.
Some lessons can be drawn from the case of Chad which may certainly help in future debates concerning the design of institutional mechanisms for managing external revenue in Africa. Here we refer to oil revenue, but this could equally apply to revenue from other mineral resources (diamonds, gas, etc), and some authors analyse development aid in the same terms:
1.‘Good governance’ is difficult to export and apply successfully. Chad is not as extreme a case as Iraq, but it does show the difficulties in establishing from abroad a political-administrative model in such a short time, which does not respond to the country’s own development or to real demand there. Following the failure of the ‘minimum state’ model used in the first structural adjustment plans, the ‘good governance’ model implemented from the mid-1990s by the World Bank and the IMF also had limitations. At the time of enacting Loi 001/PR/99 some complained of the foreign interference it implied in the country’s internal affairs.
2. In relation to the above point, one of the specific aspects of ‘good governance’ in Chad has been the policies of transparency in managing revenue. And in general there has indeed been plenty of transparency. Information regarding revenue inflows, allocation of resources from these inflows, extractive activity and the socio-economic impacts has been plentiful and relatively easy to access:
- On a quarterly basis, EssoChad released a detailed financial report.
- At the request of local and international NGOs, in 2001 the Office of the President of the World Bank set up an independent ad-hoc monitoring committee for this project (the International Advisory Group), in order to perform an on-the-ground audit of compliance with the socio-environmental terms of the contracts and the management of oil revenue, and eventually to advise the World Bank and the governments of Chad and Cameroon in this connection. This committee has regularly produced reports, in which it has outlined the problems in the model of managing revenue.
- At the same time, the IFC also commissioned an external consultancy firm (D’Appolonia SpA) with the task of socio-environmental monitoring within the framework of what was called the External Compliance Monitoring Group (ECMG). The ECMG has conducted a number of visits to the project, and has periodically released reports evidencing a number of problems which needed to be tackled in order to meet the IFC’s environmental standards, most of which were not resolved. Nevertheless, in August 2004 the ECMG issued a certificate of compliance, which was widely criticised by local and international NGOs.
- Although in the end it proved to be a body manipulated by political power, the Collège de Contrôle et Surveillance des Ressources Pétrolieres (involving Chadian civilian society), has released monthly financial reports regarding the use of oil revenue under its control and, at the time, it also denounced some administrative irregularities relating to public tenders.
The point of all this is that although informative transparency is desirable in itself, it does not automatically pre-empt the problems highlighted in the literature concerning the ‘resource curse’. Admirable initiatives such as Publish What You Pay and the Extractive Industries Transparency Initiative (EITI) are necessary, but not sufficient.
3. The project scheduling was not appropriate. The speed of development of institutional capacities at all levels (cash, revenue management and control bodies, territorial administration, ministries in the priority sectors, etc) was overtaken by the arrival of revenue a few months earlier than scheduled, and high oil prices from 2006, which caused major bottlenecks and administrative headaches. In June 2009 (less than six months into the project), revenue for the government of Chad deriving from the project totalled US$4.536 billion, when the projection in 2000 was for US$1.7 billion in 25 years. For example, despite the repeated calls by the CCSRP regarding its limited management capabilities, it is still ill-equipped to manage a huge budget. It is worth pointing out that the World Bank paid little attention to the need to underpin oil revenue management capacity before the revenue actually started flowing in. Capacity-building projects launched in parallel by the World Bank were aimed at ‘[having]at least some capacity in place when the project is implemented’. In this regard, it might be worth considering the application of a kind of Hotelling rule, whereby the resources must be left underground until it has been verified that there is an institutional apparatus in place able to guarantee that the current net value of their exploitation in terms of development is positive.
4. Chad is a classic case of the theory of obsolescing bargaining. This theory establishes that in a negotiation between a developing country’s government and a transnational company to seal a contract for exploiting natural resources, the bargaining power at the outset is held by the company, but once investment and infrastructure are in place this power passes to the government. Beforehand, the developing country has natural resources that cannot be exploited unless it is through the external injection of capital and technology from transnational companies. In this situation, the government holds little bargaining power, and it is willing to accept often unfair contractual conditions. However, once the investment is made, and having verified its profitability, the level of risk perceived by the company diminishes, and the government has a greater scope for manoeuvre to renegotiate the contract in its favour. In the case of Chad, barely a year after beginning to export oil, the government began to pressure the Consortium to revise the contractual terms and, at the end of 2005, it told the World Bank that it intended to change the previously-agreed model for using the revenue. The specific characteristic of Chad is that ‘obsolescing bargaining’ not only happened in respect of the Consortium, but also in relation to the World Bank.
5. Another of the lessons refers to the structure of revenue envisaged in Loi 001/PR/99. This distinguished between direct revenue (royalties and dividends on the shareholding in TOTCO, the company that owns the oil pipeline), and indirect revenue (tax on the Consortium, permits and other fees). The Loi (both the original and the amended versions) only envisaged intervening in regard to direct revenue, leaving the use of indirect revenue entirely to the government’s discretion. In principle, it was estimated that indirect revenue would not exceed 40% of the total, but one of the consequences of the greater bargaining power explained above has been higher taxes on the Consortium, so that in June 2009, of the US$4.536 billion received, 62% (US$2.819 billion) correspond to indirect revenue, and are not therefore covered by the Loi, vs. 38% (US$1.717 billion) of direct revenue. As Pegg indicates, a simpler design covering all revenue would probably have been more effective, although this would not have prevented the model from failing.
6. The resource curse theory places the emphasis on the internal dysfunctions triggered by the inflow of external revenue. One of the limitations of this theory is the small weighting afforded to external stakeholders, and in particular transnational companies. In the case of Chad, there has never been a threat to the oil inflow at companies belonging to the Consortium. While it is true that the government has made a show of piling pressure on the Consortium, in fact this was merely a redressing of the imbalance in bargaining terms, which were unfair in the first place (royalties were set at a low 12.5%). As the government’s financial capacity increases, it may consider gaining a foothold in the Consortium (or other oil projects), to the detriment of some of the current participants, but this is unlikely to happen without compensation. In short, in Chad the Consortium has been the party least damaged by the model’s demise.
Conclusions: The World Bank has certainly been damaged by the failure of this model. It has failed in its bid to redeem its image by linking oil operations to reducing poverty. The difficulty of putting into the same bag such complex, controversial and unproven policies and discourses as good governance, managing external revenue and reducing poverty came up against a scenario in which stakeholders (the government, the rebellion, the Consortium) had conflicting and dynamic interests, and in which certain uncontrollable aspects have also emerged, such as high international oil prices, which made revenue soar. Furthermore, the World Bank was subjected to not inconsiderable pressure in 2006 by the US government and France to ease its demands on the Déby government, in order to achieve a compromise solution with regard to repealing the Loi, which ended up becoming the Memorandum of Understanding. Furthermore, the World Bank was too confident, perhaps naively so, that the model’s design would resist the possible lapses of the Déby government; it failed to take into account its authoritarian track record. The need to make the model work coupled with foreign pressure led the World Bank to fail to correctly factor in a number of variables.
It is certainly no coincidence that Chad’s oil project coincides with the US’s energy interests, adding another external element to the equation. The change in the US’s geo-energy strategy since the first Gulf War (1991) among other things envisages an increase in oil supply from the Gulf of Guinea and a decrease in supply from the unstable Middle East. This in part explains the World Bank’s involvement in this project.
To continue with the external factors, it is interesting to see how the case of Chad illustrates the strategic rivalry between China and the US in Africa, and the shape this rivalry might take in the future. The Déby government, like so many others in the region, is currently facilitating operations for Chinese oil companies in new oil projects (especially the China National Petroleum Corporation –CNPC–), and it is beginning to consider laying an oil pipeline to link up the Chadian fields with the Sudan pipeline (Muglad-Port Sudan, majority-owned by the CNPC). There is even speculation that it might eventually link up Kribi and Port Sudan. Although relations between the Sudanese and Chadian governments are currently sour, these scenarios cannot be ruled out.
In short, some of the predictions of the resource curse theory, in particular the strengthening of authoritarian behaviour, may be applicable to Chad, but should be duly qualified by the presence of external factors.
Artur Colom Jaén
Department of Applied Economics (Universidad Autónoma de Barcelona), member of the African Studies Group (Universidad Autónoma de Madrid) and the African Societies Study Group (University of Barcelona)
 For more background on structural political instability, see S. Massey & R. May, ‘The Crisis in Chad’, African Affairs, vol. 105, nr 420, p. 443-449.
 For further background, see Colom (2008), ‘Petróleo, gobernanza y conflicto en el Chad’, ARI 73/2008, Elcano Royal Institute; S. Pegg (2005), ‘Can Policy Intervention Beat the Resource Curse? Evidence From The Chad-Cameroon Pipeline Project’, African Affairs, vol. 105, nr 418, p. 1-25; S. Massey & R. May (2005), ‘Dallas to Doba: Oil and Chad, External Controls and Internal Politics’, Journal of Contemporary African Studies, vol. 23, nr 2, p. 253-276; and Crisis Group (2009), ‘Tchad: sortir du piège pétrolier’, Briefing Afrique, nr 65.
 J.F. Bayart (2008), ‘Obscenité franco-tchadienne’, Le Monde, 12/II/2008.
 T. Harford & M. Klein (2005), ‘Aid and the Resource Curse’, Public Policy for the Private Sector, note nr 291.
 See R. Abrahamsen (2001), Disciplining Democracy. Development Discourse and Good Governance in Africa, Zed Books; and G. Harrison (2004), The World Bank and Africa: The Construction of Governance States, Routledge.
 Ibid, Crisis Group.
 Its full name is the International Advisory Group for the Chad-Cameroon Petroleum Development and Pipeline Project, also known by its French name, which is Groupe International Consultatif du Projet d’Exploitation Pétrolière et d’Oléoduc Tchad-Cameroun, http://www.gic-iag.org. In September 2009 the IAG completed its mandate, and no other structure is expected to replace it now that the World Bank is no longer formally linked to the project.
 K. Horta, S. Nguiffo & D. Djiraibe (2007), ‘The Chad-Cameroon Oil & Pipeline Project. A Project Non-Completion Report’, Environmental Defense, http://www.edf.org/documents/6282_ChadCameroon-Non-Completion.pdf (accessed November 2009).
 For a broader discussion, see I. Kolstad & A. Wiig (2009), ‘Is Transparency the Key to Reducing Corruption in Resource-Rich Countries?’, World Development, vol. 37, nr 3, p. 521-532.
 Essochad, ‘Project Update No. 26. Mid-Year Report 2009’, http://www.essochad.com/Chad-English/PA/Files/26_allchapters.pdf (accessed December 2009); and International Advisory Group, ‘Final Report’, 3/IX/2009, http://www.gic-iag.org/doc/PR93125.033_Final_Report.pdf (accessed December 2009).
 S. Pegg (2009), ‘Chronicle of a Death Foretold: Evidence From The Chad-Cameroon Pipeline Project’, African Affairs, vol. 108, nr 431, p. 311-320.
 J. Gould & M. Winters (2007), ‘An Obsolescing Bargain in Chad: Shifts in Leverage between the Government and the World Bank’, Business and Politics, vol. 9, nr 2.
 Ibid. S. Pegg (2009).
 For a more extensive criticism of the resource curse theory, see J. Di John (2007), ‘Oil Abundance and Violent Political Conflict: A Critical Assessment’, Journal of Development Studies, vol. 43, nr 6, p. 961-986; and M. Basedau (2005), ‘Context Matters. Rethinking The Resource Curse in Sub-Saharan Africa’, GIGA Working Papers, nr 1, German Institute of Global and Area Studies, http://www.giga-hamburg.de/dl/download.php?d=/content/publikationen/pdf/wp01_basedau.pdf (accessed December 2009).