By Anton Rühling, OpenOil and David Mihalyi, Natural Resource Governance Institute
This post originally appeared on openoil.net and resourcegovernance.org on August 12, 2016
Many oil-rich countries’ governments reap huge and notoriously opaque production revenues. In the past, transparency initiatives were limited to exposing tax payments. However, EITI reports and company disclosures are exposing more details on these deals. We are seeing a growing database of incisive price data emerging.
Many observers want to know how much the oil is worth and whether a fair price is paid for it. A common assumption is that the value of a barrel of oil is easy to determine. In fact, it’s quite complicated. The oil price’s daily fluctuations are closely monitored. These are done using the so-called “reference prices”: the value of a barrel of a particular oil quality sold in a particular hub. Every barrel, though, is worth a different amount. How light the oil is, how expensive it is to transport it to main hubs and how it is sold (on a spot or long-term contract) all factor in to its value.
Source: Screenshot from Page 349 of NEITI report, Original source is EIA.
A person wanting to know the value of their country’s oil once had three options: using one of the global free benchmarks, which lack granularity, as a best approximation; paying large subscription fees to access granular market data; and looking for revealing industry reports.
The recent release of EU mandatory payment disclosure reports have provided a granular and standardized new data source on oil prices. Oil companies are now disclosing the value of the production entitlements they pay in kind to governments, both in volume and in price terms. Assembling these we built a small table of oil prices.
Data source: BP, Shell, Statoil 2015 mandatory payment disclosures.
Data from the disclosures is available on resourceprojects.org.
The above graph displays the price of oil in 2015 across BP, Shell and Statoil projects. Prices spread across a wide range, with most observations between USD 20 to USD 55 per barrel. Reported oil price in Libya is surprisingly high. There are also wide variations in the price of oil for projects in the same country. The price reported by Shell from SPDC East is much lower than for other Nigerian projects, a finding that led to civil societyraising questions. We also found significant differences between the prices reported by BP and Statoil for the same projects and period in Azerbaijan.
In the case of Statoil, we could calculate the change across two years of disclosures. We found that prices dropped by an average of 47 percent from 2014 to 2015—roughly equivalent to the drop in the Brent price, a popular benchmark. But again, there was a significant variation in the proportions of reported price drops.
New data is also beginning to reveal the prices governments sell their oil for. This information should be published more systematically by both governments and in-country operators. In Azerbaijan, for example, this information was disclosed in its 2014 EITI report.
The above graph shows the price Azerbaijan national oil company SOCAR sold its own share of crude for in 2014. The graph above shows that Azeri state oil was sold at prices between USD 69 and USD 113. The average price (weighted by quantity) was USD 99 per barrel.
These figures are then compared with the reported production entitlement values discussed above. For the same year, Statoil reported on the two producing projects in Azerbaijan. They reported a price of USD 103.4 per barrel for the Azeri, Chirag and Gunashli (ACG) complex, the country’s largest oil project, contributing three-fourths of national oil production. They also reported a much lower USD 52.9 price for the Shah Deniz gas and condensate field. Taking a weighted average of the two yields a price of USD 97.5 per barrel.
To summarize, we find that in 2014, the Azeri government sold its oil at an average price USD 1.5 per barrel higher (at 99 USD) than the market value as reported by Statoil (at 97.5 USD). Of course, we don't know many further details to evaluate these transactions, for example the timing difference between production and sales, or the risks involved in the oil trade. Access to oil sales contracts could shed light on how much of the price risk is borne by the oil trader and how much is on SOCAR.
The price data we present remains very limited, both in company and country coverage. However, we can already explore the range in prices, looking for patterns and outliers in both reported market value and government oil sale prices. As reports on payments to governments continue growing, we are gaining valuable data on oil prices for free. We can start better exploiting these data to monitor taxation and promote better policymaking and informed discussions on the sector and on the broader debate on trade misinvoicing.
This is why initiatives such as Publish What You Pay’s “Data Extractors” program, as part of which this methodology was initially piloted, are crucial to build up more open data capacity among citizens across the globe.
Anton Rühling is the program manager at OpenOil, overseeing the development of all open data applications. David Mihalyi is an economic analyst with NRGI.
By David Mihalyi and Chris Perry, Natural Resource Governance Institute
This post originally appeared on www.resourcegovernance.org on April 1, 2016
NRGI is excited to launch the public alpha version of ResourceProjects.org.
ResourceProjects.org is an open-source repository of data on oil, gas and mining projects across the world. It provides a platform to collect, display, download and search extractive project information using open data. It aims to harvest data on project-by-project payments to governments—based on recent mandatory disclosure legislation in the EU, U.S. and Canada as well as EITI reports—and link it to associated information about the project from a variety of sources. The platform will make it easier for journalists, CSOs, researchers and government officials to search, access and download relevant data.
As we continue to develop the platform and connect it to new data sources, we are inviting contributors and collaborators to get involved.
Why does project-level data matter?
Projects are the physical, tangible presence of extractive operations in a country. A project is the mine that people see out of their window or the oil field along their coastline. But a project also has a concession area where it is located, one or more participating companies, contract documents detailing their obligations and payment information giving an insight into their economic contribution.
Governments and citizens groups can also use project data to model revenues and forecast budgets, such as in Ghana, where all interested parties could see how different oil prices affected the money available for the budget. Others, such as CCSI, Global Witness and Open Oil have modelled contracts to evaluate extractive deals, while IMF economists routinely use project-level information for fiscal design and technical assistance using their publicly available FARI model. Project information has a multitude of applications beyond fiscal modeling. It can be tied to spatial data to help better understand local impacts or environmental consequences, as highlighted by recent academic papers.
Why did NRGI build this tool?
Information on extractive projects are scattered across different company and government websites, in EITI reports, as well as databases compiled by regulators, international organizations and civil society. It comes in multiple formats: PDF, spreadsheets and in computer queryable databases. These are rarely linked to each other at all.
ResourceProjects.org brings this information into one place. We are also working on linking the data gathered to other repositories on related entities, such as OpenCorporates for associated companies; ResourceContractsfor oil and mining contracts; and Open Oil`s concession map. All information on the platform is stored with details on what source it came from and how it was retrieved. By bringing this information together in a standardized and accessible format, we are allowing users to explore extractive projects with greater depth.
How to get involved?
We are now looking for people who are interested in getting involved in the site. By the end of April, we will have added company disclosures from the U.K. that are starting to be released. Beyond the U.K., many companies are beginning to release project-by-project tax payment data. We would welcome any organisations or individuals who wish to lead on sourcing data from specific countries from upcoming mandatory disclosures.
Additionally we are inviting feedback as well as interested collaborators to help develop the site and its content. Further features and enhancements will be rolled out in the coming weeks and we are looking for partners who want to get more closely involved.
Finally, we are seeking to support the growing community of data users. Please sign up to the ResourceProjects mailing list if you want to keep up to date with what’s happening and how different organizations are using project-level information for improving resource governance.
If you are interested in getting involved, please contact NRGI economic analyst David Mihalyi at email@example.com.
David Mihalyi is an economic analyst and Chris Perry is an open data analyst with NRGI.
With new disclosure laws in effect in the European Union, Norway, Canada, and now the United States, there has never been a better time to be on the front lines of the fight for transparency and accountability in the extractive industries. Project level payments are being disclosed by oil and gas companies like BP, Statoil, and Shell. However, as any seasoned data extractor will know, this information is often released in PDF format, making it difficult to transfer the data tables into a spreadsheet application where it can be put to use. While companies listed and registered in the UK are required to disclose information in XML format (company reports are made available for download in .csv files), not all countries require this, making data extraction seem an arduous task.
Tabula is a powerful and extremely useful open-source web application for extracting data locked in tables in PDF documents. Similar to the Google Scraper application we introduced in an earlier post, Tabula can expedite the process of getting information into a usable format when copying and pasting is not an option.
NOTE: Tabula will only work on optical character recognition (OCR) enabled PDF documents, not image-based documents. Put simply, PDF reports with OCR make the content searchable and interpretable by software. Even if you don’t know which type your document is, Tabula will prevent you from uploading the wrong kind.
Tabula can be downloaded at Tabula.technology and works for Windows and Mac users. Follow the instructions on the page and make sure to have a version of Java installed.
Once you have Tabula installed, double click the "tabula" application file in the Tabula folder and it will open up to a page in your web browser.
NOTE: the command prompt will open and run for a few seconds before the page opens in the browser
The first step is to import the PDF from which you want to extract data into Tabula. For the remainder of this post, we will be using the BHP Billiton Economic contribution and payments to governments Report 2015, which the company released voluntarily in September, 2015. Some company reports include this information in .csv files, but so far BHP Billiton has only provided PDF reports.
If we try a simple copy and paste operation into a spreadsheet application with any of the tables in the document, we will find that all the information is imported into a single column or a single cell.
Download the report from the company site linked above, then use the “Browse” button in the Tabula page to find the saved PDF file. Select it and click “Import”.
For today’s post we are sharing an easy way to see the full list of mandatory disclosure reports submitted to Companies House Extractives Service in the United Kingdom (UK). This site houses the project-level payments reports filed in compliance with the UK implementation of the EU Accounting and Transparency Directives, which requires oil, gas and mining companies listed on a UK stock exchange (or large companies incorporated in the UK) to report their project-level payments to governments.
International oil majors BP and Shell, as well as mining giant Rio Tinto, have now made their reports publicly available on the UK register, in a machine-readable format (excel files). However, there is a slight problem with how the Companies House site presents the reports. Upon navigating to their page you are met with the following:
As you can see above, there is nothing more than a search tool, and there is no way to browse the list of companies that have filed reports. In order to find a company report, you need to search for that name specifically.
Fortunately, we have found a way to access the full list of companies that have reported!
There are two ways to do this:
Inputting this search query gives you a list of the currently available reports in no discernible order. There are dates attached to each report but you will have to click on each company to find them:
Mandatory disclosure reporting is new and it is important that companies report at the appropriate level of granularity, and that governments make these disclosures easily accessible and usable for the public. The best practice for presenting mandatory disclosure data is to provide machine-readable files as is found on the UK site. This allows stakeholders to skip the steps of scraping and cleaning the data as was necessary with the Total report, which was filed under France’s transposition of the EU Directives.
In a future post we will describe the process for accessing extractives data in the countries where reports are available--information on where to find company disclosure data can be found on our resources page.
Click here for the archives to see our full list of posts.