I Read Hundreds of Pages of Africa EITI Reports. Here’s What I Found.

Blog post
23 February 2018
- Patrick Heller

NRGI is building a new global data set that will detail the revenues and activities of national oil companies (NOCs) across the globe. We are compiling most of the data from figures included in the financial reports of the growing number of state-owned entities that are publishing corporate balance sheets, income statements and cash flow statements.

The project is also gathering data from Extractive Industry Transparency Initiative (EITI) reports on revenue flows to state-owned enterprises. Beginning with the 2013 revision to the EITI standard, countries participating in the initiative are required to publish information on the mandates and legal status of any state-owned enterprises (SOEs) active in the extractive sectors, and on the flows of funds to and from these SOEs. EITI participants praised the strengthening of the standard, and the EITI Secretariat and its partners have worked in the intervening years to encourage active implementation, mainstream EITI disclosures into regular government reporting practices, and track their impact on governance debates in implementing countries.

As we are doing our own data-gathering in the effort to improve international benchmarks of NOC performance and revenue management, the figures from EITI reports are proving extremely valuable, especially in a number of African countries where NOCs do not produce detailed financial reports on their own. In a final fit of masochistic number-crunching at the end of 2017, I reviewed our EITI data sets from several African countries that are either producing oil and gas or investing heavily in exploration: Cameroon, Chad, Côte d’Ivoire, the Democratic Republic of Congo, Ghana, Liberia, Mozambique and the Republic of the Congo.

We are looking forward to sharing the results of the broader global data-gathering exercise, so stay tuned for that as 2018 progresses. But for now, I figured it would be valuable to share some observations from my deep dive, about the value of this EITI data and how countries can further refine their approaches to enhance public understanding about the roles and performance of oil-sector SOEs.

In several countries—particularly in Francophone Africa—EITI reports provide the most valuable public information available about what NOCs are doing with public resources. In Chad, Republic of the Congo and DRC, NOCs themselves do not publish detailed information on their finances. Recent EITI reports make it possible for citizens in these countries to begin to understand NOCs’ impact on the oil sector and the economy more broadly. This access to new information is extremely valuable as a mechanism for accountability.

Take the Republic of the Congo, for example. The state-owned Société Nationale des Pétroles du Congo (SNPC) is massively important as a fiscal agent and steward of the oil sector, but it has traditionally managed the sector opaquely, with significant repercussions for governance. Recent EITI reporting sheds light on how much money the company is taking in: more than USD 3 billion in 2014, representing almost all revenues collected by the state in the oil sector (see pages 72 – 76 of the 2014 report). The EITI report also describes how much of that revenue the company is paying to the state, and the different ways in which these funds are transferred. (In this case, SNPC transfers funds both to the treasury and to a special infrastructure fund.)

Data in these reports can be extremely valuable to researchers, public officials and activists asking questions about sector and fiscal management questions. Sticking with the example of the Republic of the Congo, the data in the EITI report can facilitate a better-informed public discussion about issues such as the opportunity costs of the government’s special oil-for-infrastructure arrangements and the value that SNPC is getting for oil it sells on behalf of the state.

In the Republic of the Congo, Cameroon and Côte d’Ivoire, NOCs appear to have separate commercial arrangements for oil they are selling as agents of the state (typically via state entitlements under production-sharing contracts or via state equity) and the NOCs’ “own” oil, to which NOCs gain access via their own at-risk capital. Broadly speaking, the EITI reports indicate that these companies transfer the proceeds of their “state agent” oil, which represents the vast majority of their total revenues, directly to the countries’ treasuries, with some withholding of processing/marketing fees. With the much smaller revenues earned from their “own” oil, the NOCs pay taxes similar to those paid by private oil companies, as well as (in theory) dividends to the state in the event of profits. The table below illustrates the dual-track approach in Côte d’Ivoire and the Republic of the Congo.

NOC revenue flows, 2014 (all figures in USD million)

Country (NOC)	“State agent” oil/gas			“Own” oil/gas				Source
	Gross revenue from sales	Transfers to the state via all taxes	Transfers as percentage of gross revenues	Gross revenue from sales	Transfers to the state via all taxes	(of which dividend payments)	Transfers as percentage of gross revenues	Source
Côte d'Ivoire (PetroCI)	196	196	100%	215	12*	0	6%	2014 EITI Report, pp. 8, 133
Republic of the Congo (SNPC)	2,900	2,900	100%	376	7	0	2%	2014 EITI Report, pp. 72 - 76

Having access to this information across several years gave me a better picture of the role of these companies, and facilitated several hypotheses and questions. Researchers who know the oil sector in Côte d’Ivoire and the Republic of the Congo in detail would certainly be able to provide a deeper assessment than I can, but I share some of these initial responses as an illustration of how EITI data can fuel analysis even for a non-expert:

Hypothesis. The full transfer of all shares of “state agent” oil and gas suggests that the NOC holding on to an excessive share of those revenues does not pose a significant governance challenge, and that the scrutiny of oversight actors is better focused on the company’s sales mechanisms and the management of its own at-risk investments.

Question. In order for citizens to judge how well these NOCs are managing the hundreds of millions in revenues that they are earning via their own at-risk commercial activities, it is necessary to assess not just payments as a share of gross revenues, but the effective tax rates they are paying to the state as a share of their net income, and what kind of profit they are turning. We know from the EITI reports that neither of these companies paid dividends to the state in 2014, but that does not tell us very much about whether they are profitable or how well they are controlling costs. Thus a next step in data analysis would involve asking key questions about NOC expenditures. More on that below.

There may be a community of practice evolving among report-writers. I noted a degree of similarity across reports authored by the same company. The consultancy Moore-Stephens, for example, authored the recent reports for Cameroon, Chad, Côte d’Ivoire, DRC and Liberia, and in spite of some differences presumably arising from the varying approaches of national multi-stakeholder groups, there is a significant degree of consistency in the way the reports are organized and the way figures are presented. As an info-consumer looking across these reports, I was able to get in a rhythm as I read them, to understand what different terms meant in the reconcilers’ parlance and use the commonalities in the reporting structure to illuminate differences in how the different countries manage their NOCs. I suspect that EITI and its participating members could do more to work with the reconciler companies to use common methodologies to draw out lessons and observations at an international level, in furtherance of the EITI Open Data Policy.
EITI reports could take another major step forward by providing additional data on NOC finances. As noted above, reports on these companies would boost their value if they included additional key information to provide a better picture of what NOCs are doing with public resources. Our recent Guide to Extractive Sector State-Owned Enterprise Disclosures offers some guidance on what to include in order to maximize public benefit.

As I reviewed the African EITI reports to gather data on national oil companies, two gaps stood out as particularly salient. First it would be valuable for EITI reports to cover all revenue flows that accrue to NOCs, not just the revenues that the companies collect from certain defined revenue streams. Several of the recent reports provide good information on the revenues that the NOC is collecting from oil sales or from fiscal payments from contractors, for example; but are unclear about whether these represent the only revenue streams that the NOCs access. Including data on all NOC revenue sources – or clarifying that the revenue streams in the report represent the totality of NOC revenues – would make it easier to analyze the overall state of NOC finances and their impact on the economy.

Second, providing data on expenditure would enable stronger analysis of the basis upon which NOCs are (or are not) paying taxes and dividends to the state. This analysis – which connects to the core EITI goal of encouraging dialogue around whether all companies are paying what they should to the state – is virtually impossible to do effectively with only gross revenue figures. Our SOE disclosure guide describes the gold standard for what details to include, but a first step in many countries would be to include top-line reporting on company expenditure, broken down into operational and capital expenditures in upstream, downstream and non-core activities.
Reporting on revenue transfers from the state to the NOC is inconsistent. EITI requirement 4.5 mandates that the EITI reports include material information on “transfers between SOEs and other government agencies” (emphasis added). And while great strides have been made in reporting on payments that national oil companies make to national treasuries, reporting on budgetary and other transfers from the treasuries to the NOCs is far less consistent. Chad’s 2013 EITI report provides the necessary top-line information, providing a breakdown of the budgetary allocation from the state to the Société des Hydrocarbures du Tchad (SHT) and an explanation for the intended uses of the budgeted amounts. Other reports provide less information. The Liberia 2014 report, for example, includes the annual approved budget for state-owned NOCAL (including sources of revenue), but does not detail actual receipts and allocations. And several other countries provide no information at all about transfers from the budget, which makes it harder to assess the overall fiscal relationship. This is particularly important in countries such as Mozambique where in the period before significant production governments tend to fund NOCs primarily by budget allocations.

It may be that several reports lack this information because there are no budgetary allocations to the NOC. In this case, it would be helpful for the report to include a clear statement that this is the case.
Where there is information available directly from state companies themselves, EITI reports are often not well-harmonized. The strongest example of this in our sample was Côte d’Ivoire, where PetroCI publishes moderately detailed financial statements that include income, tax payments, and balance sheets. Taken together, the data included in the company statements and the EITI reports would have painted a more comprehensive picture of the company’s activities and finances. But the reports were developed according to different standards and with inconsistent terminology, rendering the data impossible (for our team, at least) to harmonize across the two documents. Of course, an EITI report is not a substitute for a financial statement produced to international accounting standards, and the reports will not match completely. But EITI multi-stakeholder groups (country-level coalitions responsible for issuing the reports) would enhance their contributions – and avoid the risk of separate “data islands”– if they took advantage of the expertise and access associated with the report-writing process to harmonize available figures as best as possible. At a minimum, these groups could include in the EITI report an interpretive section on “how to read our SOE figures alongside other publicly-available information.” Alternatively, as EITI reporting becomes increasingly mainstreamed, publication of EITI-mandated disclosures directly by the SOE, with a similar explanation of how they relate to other figures, would represent a step forward.
Report writers should aspire to maximum clarity in response to key policy questions. Authors of EITI reports in a number of countries – notably Cameroon, Chad and the Republic of the Congo – have amassed a tremendous amount of information and knowledge about how national oil companies are functioning. But as a researcher seeking information on core questions about NOC financial flows, I found it extremely difficult to piece together the ways in which the information was presented. I struggled to put together a coherent picture as I sorted through different data points scattered across reports, sometimes with different terminology in different sections. I imagine that researchers, journalists and activists within the participating countries have encountered similar challenges.

One problem may be that the precise policy questions to which readers will apply this data are not entirely clear to the authors. Based on our work gathering data to assess SOE performance, I suggest that report writers create a specific section of each report called “Flows of funds to and from the state-owned enterprise.” This would combine information in one place on all of the flows coming into and out of the SOE, with an explanation of the source of those flows and/or the rules according to which they took place. This section could also include additional data on SOE expenditures, employment, activities and accomplishments. As an animating guide, these suggestions should be oriented around helping readers answer the following questions about state-owned enterprises:

How much money is the SOE taking in? What is the basis for the money it received?
What is it doing with that money? How much is it transferring to the treasury? How much is it spending on its own activities? Why?
How effective has the SOE been in converting public resources into results?

EITI reports, on their own, may not provide comprehensive answers to these questions. But authors should keep these queries in mind when writing and designing reports. This will help readers use all of the newly available information to better inform public debate.

Patrick Heller is an advisor with the Natural Resource Governance Institute.