KPMG’s July 2002 Oil Diagnostic report, known as the “Inception Report,”28 provided the first third-party scrutiny of how Angola’s oil revenue is managed by the government. It was not an audit, but it did reveal how the government mismanaged the country’s principal source of income. The government has not agreed to publish the full report, nor has it agreed to publish any of the subsequent quarterly reports under the Oil Diagnostic’s terms of reference with KPMG. It did publish an executive summary of the Inception Report on July 17, 2003. Human Rights Watch has obtained a copy of the full July 2002 Inception Report. The difference between the July 2002 full report and July 2003 executive summary are minimal: some information on oil-backed loans is altered, for example, but the analysis and criticism of the Angolan oil sector is identical. However, the July 2002 report has much more detail on the state of the Angolan oil industry and the government’s management of oil revenues.
The Oil Diagnostic is especially significant because oil revenue is the Angolan government’s principal source of income. Angola is the second largest oil producer in sub-Saharan Africa and enabled the government to pursue vigorously its conflict with Jonas Savimbi’s rebel National Union for the Total Independence of Angola (União Nacional para a Independência Total de Angola, UNITA) movement until April 2002, after Savimbi was killed and the war with UNITA ended. Between 1995-2002, oil revenues comprised approximately 70 to 89 percent of government revenues and from 85 to 92 percent of exports, according to the IMF. In 2000, when the Oil Diagnostic was announced, oil accounted for 89 percent of government revenue, more than U.S. $4 billion. In 2001, oil accounted for about 81 percent of government revenue and was estimated to be at least 75 percent of government revenue in 2002.29
The report reveals serious defects in Angola’s oil revenue management. The report makes clear that in 2000 the state-owned oil company Sonangol did not follow Decree 30/95 that requires all oil revenue to be deposited in the central bank, passing over two billion dollars through other accounts; the government did not have meaningful procedures to verify payments by companies, including Sonangol; Ministry of Finance records were unreliable because in many cases they were based on paper transactions rather than actual transfers; and the government lacked or failed to provide sufficient information to allow KPMG to reconcile hundreds of millions of dollars in discrepancies. KPMG, dependent on information that the government was willing to provide, found that, in many cases, the government did not have adequate information or procedures to reconcile conflicting information. The government’s failure to make these reports public makes it very difficult, if not impossible, for the Angolan public to exercise adequate oversight over the government’s use of public funds. Moreover, the report’s findings that the government could not adequately account for billions of dollars of revenues makes it extremely difficult to plan and allocate sufficient resources to those activities that could improve human rights, such as increased assistance to internally-displaced persons (IDPs) and demobilized soldiers, or strengthening the weak judiciary. Furthermore, the opaqueness of the government’s activity impedes Angola’s emerging civil society from exercising government oversight, educating the public, or critiquing government decision making, all of which are crucial for increased respect for human rights.
From the start, it was clear that the key to the Oil Diagnostic’s success would be government cooperation since the data KPMG required would have to come from the government, cross referenced by information from companies. Past efforts by international financial institutions to monitor oil revenues in other countries had been unsuccessful because of the inability or unwillingness of governments to provide adequate information. This problem also plagued the Angolan Oil Diagnostic. For example, the Inception Report was due in April 2001, but KPMG did not actually finish it until July 2002 due to delays in receipt of data from the government.
Ideally, the Angolan government would have agreed to an audit of its oil revenue and would have allowed it to be retroactive, since there were many allegations of opaque arms purchases and misuse of oil revenue. However, the government only agreed to the forward-looking Oil Diagnostic study. Nevertheless, the Inception Report was critically important since it provided the initial baseline data to compare against subsequent assessments of the oil sector. The Inception Report examined oil revenue flows for 2000, even though it was not published until 2002 and only made public in summary form in 2003.30 The IMF reported that the principal reason for delays was the government’s failure to provide information to KPMG.31 The difference between the 2002 Inception Report and the 2003 executive summary was a revision of aggregate oil revenues. In its July 2002 report, KPMG presented the following table showing aggregate incoming oil revenue:
Table 1: Sources of Incoming Oil Revenue 200032
KPMG acknowledged, however, that it could not accurately determine overall government oil revenues, primarily because of incomplete data on oil-backed loans the Angolan government had received. Oil-backed loans are up-front loans that are paid for by future oil production. In effect, Angola mortgages future oil production for immediate cash. These loans play a critical role for the Angolan government since it has very few sources of financing and routinely uses these loans to raise hard currency. However, the number, amount, and use of those loans by the government have rarely been disclosed. KPMG needed this data to assess whether the government had received additional revenue as a result of such loans and what portion of its oil was pledged to repay loans. But as the report phrased it, KPMG could not “determine the amount of loan facilities which were actually received during the year 2000,” nor could KPMG “establish the amount repaid in either cash or in settlement through oil.”33 Unable to determine precisely how much of the money was received from loans, KPMG based the figures in the table on the amount of all loans active in 2000.
KPMG appeared to have obtained better data from the government a year later, however, when the July 2003 executive summary was published. In that report the amount of revenue from loans was revised downward to approximately U.S.$1 billion. The following table shows the revised incoming oil revenue for 2000:
Table 2: Revised Incoming Revenue in 200034
In its study, KPMG found major discrepancies in incoming oil revenue, poor record keeping, and numerous areas where data was missing.Revenue discrepancies broadly fell into two categories: the discrepancy between the amount Sonangol was due to deposit in the Banco Nacional de Angola (BNA), the Angolan central bank, as required under Angolan law Decree 30/95, versus what it actually deposited in the BNA; and discrepancies between the amounts that different departments within the Ministry of Finance have recorded for the same transactions involving Sonangol including for funds that have illegally bypassed the central bank. It took a comprehensive look at these activities for 2000 as the starting year for the Oil Diagnostic reports.
Under Decree 30/95, all taxes and royalties of all oil companies, including Sonangol, must be paid into a special Petroleum Account housed within the BNA. The account is also supposed to receive all of Sonangol’s uncommitted export oil sales. Those funds are used by the BNA to repay public debt, to credit the Ministry of Finance for Sonangol’s and other oil companies’ tax payments, and to set aside funds that can be used by Sonangol to meet cash calls. These sums are received in dollars and the BNA is supposed to credit back Sonangol with equivalent sums in Angolan Kwanzas so that it can meet any obligations it may have in Angola. Any surplus amounts after these obligations are met are transferred back to Sonangol and are supposed to be deposited in Sonangol’s accounts in Angolan commercial banks. If a deficit in the Petroleum Account occurs, the BNA is supposed to make withdrawals from Sonangol’s accounts in Angolan commercial banks. Through the Petroleum Account, oil revenues and expenditures revenues and expenditures can be adequately recorded and tracked by the BNA on behalf of the Ministry of Finance.35
Private companies were paying the BNA account during 2000 as the following table shows:
Table 3: 2000 Tax Receipts Received by the Angolan Central Bank from Private Companies36
While private companies apparently abided by the system of paying into the account, Sonangol did not. KPMG reported a discrepancy of U.S. $2.0 – 2.6 billion between the revenues the Ministry of Finance claimed it had received and what the BNA said it had received.37 KPMG found five different sums for incoming oil revenues: figures of the Accounting Department of the Ministry of Finance; figures of the Tax Department for the Ministry of Finance; a revised Ministry of Finance total that subtracted 1999 taxes received in 2000, taxes and Profit Oil paid during 2000, and 2000 taxes and Profit Oil that was going to be paid in 2001; figures from Ernst & Young’s (E&Y) fiscal reports on oil revenues conducted on behalf of the Ministry of Finance; and the amount of money that the BNA reported it had received.38 The Ernst & Young figures were not an audit, but a compilation of taxes paid, taxes payable, sales exports, volumes, and sales prices of companies taken from the companies’ own data.39 KPMG used the E&Y reports extensively because they provided a comprehensive set of data.40 The following table illustrates the discrepancies:
Table 4: Total Taxes And Profit Oil Received By All Companies In 2000 (U.S.$ And Kwanzas Millions)41
KPMG was unable to reconcile the approximately U.S.$192 million discrepancy between the figures provided by the Accounting Department and the figures provided by Tax Departments within the Ministry of Finance. Nor could it find any procedures by the Ministry of Finance to reconcile those figures.42 KPMG was also unable to reconcile the approximately U.S.$394 million to U.S.$587 million difference between the various Ministry of Finance figures and the Ernst & Young figures. KPMG was unable to resolve this discrepancy by the time it published the July 2002 Inception Report and still had not resolved the discrepancy by July 2003 when the Executive Summary was made public.
A major controversy surrounding oil revenue has been the role of private companies. Their inability or unwillingness to publish their payments to governments has drawn substantial criticism from the press and NGOs, in particular. Such scrutiny has also led to international initiatives, such as the Publish What You Pay Campaign or the Extractive Industries Transparency Initiative, that promote greater corporate and government transparency (see Section VI below). However, the Inception Report determined that the primary source of oil revenue discrepancies was not foreign companies, but Sonangol itself. In the case of royalties and taxes, KPMG found that the Ministry of Finance reported that it had received U.S.$114 million to U.S.$418 million more than Sonangol’s Fiscal Report said it had paid. The following table illustrates the discrepancies:
Table 5: Discrepancies Between Sonangol’s Fiscal Report and Two Sets of Figures from the Ministry of Finance Tax Directorate (U.S.$ Millions) 43
According to the Inception Report, the Ministry of Finance did not attempt to reconcile these discrepancies, nor did it have any procedures in place to reconcile the discrepancies.44
Sonangol’s Profit Oil, the amount of oil it sells for profit after companies have taken their share and Sonangol has paid its expenses, was another source of discrepancies. Sonangol receives Profit Oil from its production sharing agreements in various oil blocks. Under Decree 15/89, it is required to transfer the specified amounts, minus a 10 percent commission, to the Ministry of Finance. The Profit Oil it remits to the Ministry of Finance is an aggregation of Sonangol’s own Profit Oil, along with any Profit Oil that private companies are required to pay to the government under certain production sharing agreements. However, KPMG was unable to disaggregate these sums or determine whether the total amount paid was equal to the amount owed. The Inception Report states:
Despite these limitations, KPMG was able to analyze the aggregated Profit Oil reportedly paid to the Ministry of Finance. KPMG found that the Ministry of Finance received approximately U.S.$135 million or U.S.$323 million less than what Sonangol reported it had paid. KPMG used data from Sonangol’s Fiscal Report, compiled by Ernst & Young, and two sources of data from the Ministry of Finance’s Tax Directorate. The following table illustrates these discrepancies:
Table 6: Comparison between Profit Oil Paid to Ministry of Finance versus Profit Oil Received in 2000 (US$ Millions) 46
KPMG could not reconcile the approximately U.S.$188 million difference in the two figures the Ministry of Finance’s Tax Directorate provided for the same amount of Profit Oil it had reportedly received. Similarly, KPMG was unable to reconcile the discrepancies between the outstanding balances due from Sonangol. KPMG noted that the Tax Directorate attributed some of these funds balances owed from previous years, but did not provide a detailed analysis of this factor and “were therefore unable to apply them either against our other information on Profit Oil receipts for 2000 or the receipts of the other tax categories.”47 As a result of the lack of information, KPMG reported that it could not:
Moreover, KPMG said that without a master reconciliation spreadsheet and better bookkeeping by the government, KPMG or the government could not:
KPMG essentially stated that there was no way to reconcile the various discrepancies in order to accurately determine how much money the government should receive or even how much it did receive based on the available data. KPMG noted:
Despite these limitations and numerous discrepancies between royalties, taxes, and Profit Oil owed versus what was actually paid, KPMG was able to approximate how much money was sent to the government and how much illegally bypassed the central bank. KPMG’s analysis showed that throughout 2000, Sonangol had underpaid the BNA by more than U.S.$2.1 billion, based on the Ernst & Young reports. Even though it owed over $2 billion in taxes and Profit Oil, Sonangol paid less than U.S. $20 million, less than 1% of what it owed, to the BNA.50 KPMG attributed the discrepancies to two major factors: an underpayment of taxes by Sonangol to the BNA and its failure to remit any of its Profit Oil to the central bank.51 KPMG could not determine the overall accuracy of the Ministry of Finance records nor could it reconcile discrepancies within various Ministry of Finance figures for the same transactions, raising further questions as to how the money actually was used.
By January 2001, Sonangol had formally announced that it would stop paying into the Petroleum Account altogether. According to the Oil Diagnostic report, Sonangol said that it had stopped paying because the BNA would delay crediting Sonangol in Kwanzas for the funds it received since it only paid Sonangol after it had paid other government debts. Sonangol also claimed that BNA took more money in dollars from Sonangol than it should have and delayed crediting Sonangol in Kwanzas, leaving the Petroleum Account with insufficient funds. Because of high inflation, the Kwanza devaluates quickly against the dollar and Sonangol was actually receiving less than it paid in dollars and was unable to meet its tax obligations.
Following its decision to withdraw from the Petroleum Account system, Sonangol reportedly retained a portion of the dollar export sales proceeds and did not deposit them into the BNA. Instead, it deposited the funds into Sonangol accounts in Angolan commercial banks and reportedly paid its taxes directly to the Ministry of Finance, bypassing the BNA altogether, in violation of the law.52 The balance is reportedly paid to the BNA as proscribed under Decree 30/95 so that the BNA can pay off government debts.53 However, this explanation does not explain how the company could violate existing laws without consequence or why Sonangol only paid U.S.$18.712 million in 2000 to the BNA.
Moreover, KPMG could not adequately determine whether sums that were reported as having been paid to the Ministry of Finance were actually paid or whether they were paper transactions that never went to the Ministry, but went to pay off other debts or were used for other purposes. Unlike the BNA, which records revenues that are actually deposited, the Ministry of Finance records inflows even when it has not actually received them, raising further questions as to the veracity of those transactions. For example, if Sonangol used oil proceeds directly to pay an oil-backed loan, the Ministry of Finance would record this as a paper transaction claiming that it received those funds, even if it never physically received any money.54 But, as noted, KPMG could not determine the veracity of the paper transactions.55
Various forms of bonus payments were included in KPMG’s July 2002 Oil Diagnostic report and the July 2003 Executive Summary. These payments included Signature Bonus Payments that companies paid to the central government once they were awarded oil blocks, exploration bonuses that were designated by Sonangol for community and development projects, and commercial discovery bonus payments companies made once fields were declared commercially viable. Signature Bonus Payments were not paid in 2000, but were detailed for previous years. However, exploration and commercial bonuses were paid, but it is not clear whether these were solely paid to Sonangol or intended for the central government.
Signature bonus payments were particularly important since these were large cash payments from oil companies in exchange for lucrative offshore oil blocks. The government has rarely disclosed the amount and use of those funds. In the past, approximately U.S.$970 million in bonus payments were used for opaque arms purchases.56 Even though no signature bonus payments were made in 2000 it was extremely important to provide a historical record of signature bonuses to determine how much money the government had previously received.
All of the oil blocks awarded in Angola, except for Block 0 and Block 4 have included a signature bonus payment from the oil companies to the government. While no signature bonuses were paid in 2000 and did not constitute part of government revenue for this year, KPMG did provide the amounts for signature bonuses paid in prior years, while the bonus payments for Blocks 16 and 34 were disclosed from other sources. The following table details these payments.
Table 7: Signature Bonus Payments by Block, Year, and Amount57
The payment for Block Thirty-Four was not part of KPMG’s study, but the IMF reported that this was the amount paid, along with an additional U.S.$100 million for social projects in 2002.58 In 2003, it revised downward the payment to U.S.$278.6 million, but it did not address the reported U.S.$100 million social bonus payment or how the government used those funds.59
Later, the government was forced to disclose the U.S.$30 million bonus payment made by Ranger Oil, Odebrecht, and Sonangol for Block 16 on June 5, 2002. The disclosure was in response to parliamentary inquiries about the deal; to the knowledge of Human Rights Watch, this was the first time that the government had disclosed such a payment to parliament.60 Additionally, Energy Compass, a specialty publication that covers the international oil and gas industry, reported that ChevronTexaco and its partners had agreed to pay an approximately U.S.$500 million bonus payment on June 27, 2003. The payment was agreed upon as part of the companies’ renegotiation for the lucrative Block 0 concession in offshore Cabinda.61 It is not clear when the companies would make this payment, but Human Rights Watch believes that both the companies and the government should fully disclose the amount and use of the bonus payment.
The actual amount of oil production was another area marred by discrepancies and crucial since this figure provides the central bank a way to determine taxes due. During the course of its analysis, KPMG examined how much oil was actually produced and sold by private companies and the government through Sonangol by comparing data from the companies with an independent analysis conducted by Saybolt, a Netherlands-based (but founded in the United States) auditing and inspection firm. Saybolt concluded that 268,550,010 barrels were lifted in 2000, whereas the total amount of liftings from the companies’ fiscal reports was 272,525,440 barrels. The net discrepancy was 3,975,430 barrels. While comparatively small in terms of production volume—about 1.5 percent, it would be quite large in financial terms since the 3.9 million barrel difference would total approximately $87,459,460 to $98,590,664, depending on the price of oil.62 These discrepancies were not attributable to any one company, but a combination of overreporting and underreporting by all of the companies that lifted oil during 2000. However, TotalFinaElf and Sonangol’s refinery stood out for their sizable discrepancies. In the case of TotalFinaElf, Saybolt recorded 16,081,071 barrels less than what the company reported in its fiscal reports. The Sonangol Refinery reported some 8,636,037 barrels less than what Saybolt found. KPMG was “unable to establish” whether the central bank or Ministry of Finance had any way of auditing or comparing oil liftings to determine how much tax should be paid.63
The portrait the Inception Report painted of Angola’s management of oil revenues was indeed bleak. Billions of dollars illegally bypassed the BNA, hundreds of millions of dollars in discrepancies were found within the Ministry of Finance, and no government institutions had procedures in place to determine why this had happened. But the shortcomings in Angola’s management of incoming oil revenue are not unique. As the next section details, the IMF determined that its management of expenditures was equally problematic.
28 The full title of the report is: Current Assessment of the Angolan Petroleum Sector: Inception Report by KPMG for the Ministry of Finance, Government of Angola, July 2002.
29 “Angola: Staff Report for the 2002 Article IV Consultation,” p. 32.
30 KPMG, Current Assessment of the Angolan Petroleum Sector: Inception Report, July 2002, p. 103.
31 “Angola: Staff Report for the 2002 Article IV Consultation,” p.18.
32 The source for this table is KPMG, Inception Report, p. 22.
33 KPMG, Inception Report, p. 22.
34 KPMG, Avaliação do Sector Petrolífero Angolano Sumário Executivo Relatório Inicial, July 2003, p.18.
35 KPMG, Inception Report, p. 111.
36 Ibid., p. 134.
37 Ibid., p. 107.
39 Ibid., p. 129.
41 Ibid., p. 130.
42 Ibid., p. 137.
43 Ibid., pp. 138-140.
44 Ibid., p. 138.
45 Ibid., pp. 146-147.
46 Ibid., pp. 144-145.
47 Ibid., p. 143.
48 Ibid., p. 143.
50 Ibid., p. 132.
52 Ibid., p. 113.
54 Ibid., p. 134.
55 Ibid., p. 156.
56 See: Angola Unravels., pp. 94-98; and Human Rights Watch interview with Foreign Minister Venâncio de Moura, Luanda, December 9, 1998.
57 All information taken from KPMG, Current Assessment of the Angolan Petroleum Sector: Inception Report, July 2002, p. 124, unless otherwise noted.
* $1 million was designated for the improvement of Kissama National Park.
58 “Angola: Staff Report for the 2002 Article IV Consultation,” pp.19-20.
59 International Monetary Fund (IMF), “Angola: Staff Report for the 2003 Article IV Consultation,” statistical appendix, July 14, 2003, pp. 77-78.
60 Human Rights Watch interviews, June 10, 2002; and the Economist Intelligence Unit, “Angola: Country Report,” August 2002, p.22.
61 Christina Katsouris, “Money-spinners,” Energy Compass, June 27, 2003.
62 KPMG, Inception Report, p.114. Oil price data for the dollar estimates was taken from the U.S. Department of Energy, Energy Information Agency, “Selected Crude Oil Prices, Beginning of the Year: 1991-2001,” available at: www.eia.doe.gov/emeu/iea/table71.html.
63 KPMG, Inception Report, p.115.