Total’s affiliate list sheds little light on what really matters
Total today announced it has published what looks like a relatively comprehensive list of its affiliate structures around the world. Some 902 affiliates incorporated in 99 jurisdictions and operating in 117 jurisdictions around the world.
It is interesting reading, if you are an open data geek. Some company structures definitely seem as though they yield insight into how operations around the world get converted into tradable, product, money, and market capitalisation. More of that in a moment.
But unfortunately – and I really don’t like to put a damper on things – it’s important to note, before this gets widely greeted as a major advance in transparency, that the list of affiliates is “flat”. It simply lists a legal name, a country of incorporation and a country of operations. Here’s a sample from the PDF… (Oh, by the way, another problem was that it was published as a PDF, making it hard to suck down the data and play with it… but luckily it turned out to be easily convertible into this Excel file which also meant we could run a few basic analytics)
That’s important because for most governance purposes you actually need the hierarchy of affiliates, starting with a company that interacts with the other partners in a project, whether governmental or private sector, and going right the way up through as many chains as it takes to get to the “mother company”, or the publicly traded entity in this case – Total SA on the Paris stock exchange.
It is the precise relationship between one affiliate and another all the way up the chain which creates the fiscal environment by which this or that affiliate is or is not liable for taxes of various kinds, due to a complex web of bilateral investment treaties, transfer pricing rules and the rest of it. To give an indication of how much “hierarchy” is involved, we found during our mapping of the BP group last year chains of up to 12 layers of ownership before they arrived at the mother company BP Plc.
So for Total, let’s take the example of Angola. Total has 13 affiliates operating in Angola, though none incorporated there – there are in fact 20 countries around the world where Total affiliates operate without any local incorporations, including Iraq, Iran, Yemen, Azerbaijan, Libya, Myanmar.
Of these 13, eight are incorporated in France. This means they could jump back up to the mothership within one or two stages, and the tax implications at least may all happen simply as a result of whatever arrangements exist between Angola and France, either bilaterally or as part of broader arrangements.
What about the other five? Two are registered in the Netherlands, two are in Bermuda and one is in the Bahamas. We only know two more things for sure: first, that these are all high financial secrecy jurisdictions, and often used for financial structuring. Second, money flowing into any of these companies does not stop there. It has at least one more step to travel, to at at least one more jurisdiction, before it is integrated into the consolidated group accounts Total SA reports to its shareholders in Paris.
But as to the specifics we can only speculate. Why would six exploration and production vehicles be incorporated directly in France whereas one is registered in Bermuda and another in the Bahamas? We don’t know. What’s the relationship between “Angola LNG Supply Services LLC” registered in the USA, Angola in the middle, and “Angola LNG Limited” in Bermuda? The names alone suggest there is likely to be one, and indeed that such a relationship will have a bearing on assessment of Total’s upstream development costs in what will be a very capital intensive development of Liquefied Natural Gas. And that in turn would be likely to affect how much cost recovery came out of LNG production and back to Total and at what stage in the project, and therefore ultimately what the profit split between Total and the government of Angola is likely to be.
It’s a reasonable hypothesis – but exactly that. We don’t know anything at all.
Total’s flat list leaves us guessing. So there are some interesting nuggets there for data geeks like ourselves – this pie chart of top countries of incorporation, for example, or how Total’s activities in Iraq and Kurdistan are structured. But it is very far from being a system-level view into how the company interacts with the global financial system.