South Africa is in a strange situation with regard to the supply of petroleum fuels. Sasol produces about 23% of the country’s requirements at, what seems to be, a considerably lower cost than the other refineries — yet it is required by law to sell its products at prices as if the stuff was being imported from international refineries. If Sasol produced 100% of our fuel requirements, we would enjoy prices at about R1,60 a litre. If this were the case, there would be no debate about Sasol’s windfall tax.
We have two problems with a windfall tax on Sasol. If it is backwardÂÂlooking (back to 1999 when the agreement with the government expired) it will probably degenerate into a slanging match between Sasol and the Treasury as to whether Sasol has repaid past subsidies. If it is forward-looking, it will deter investors and it is likely that tax consultants will soon find ways to worm out of it. Former United States president Jimmy Carter found this out when he introduced a windfall tax on oil companies in the 1980s: he expected to get $393-billion over 10 years, but only got $80-billion.
A windfall tax is not the solution because the Treasury is looking at the wrong problem. In South Africa, Sasol has two major entities: firstly, it is a coal miner and this entity sells about R5-billion worth of coal to the synfuels operation and another R1-billion to other buyers. Secondly, there is the synfuel operation — built on proprietary technology bought with taxpayers’ money from Germany, but evolved over the years with taxpayers’ and shareholders’ money.
The mining entity is relatively easy to deal with. The Minerals Royalty Bill, which caused such a flurry a few years back when early drafts were leaked, does one thing right. It recognises the country’s mineral wealth as a patrimony of the people of South Africa. Accordingly, it wants a royalty from those who mine, extract and sell these minerals. Simply put, this is a percentage on sales, and this is what the fuss was about.
Some analysts have suggested that the royalty be changed to a progressive formula tax, similar to that in use in the gold-mining industry since 1936. In this formula, tax is based on a relationship between sales and profits: taxes are high for high-grade producers and even higher when the gold price rises, but much lower for marginal producers and nothing when the gold price is low.
We support this formula system (certainly initially) for all miners, including Sasol mining. In the long term, however, we prefer a resource rental on all mineral wealth.
One of the problems with both the royalty system and the progressive formula tax is that they encourage companies that just sit on leases, or are inefficient producers, and punish those who produce efficiently and profitably.
A resource rental, on the other hand, would be based on factors such as the type of mineral, grade, ease of mining and location (distance from local users or ports); in other words, as any other fixed-property rental would be calculated. The rental would be payable from the signing of the lease or granting of the right to mine, with grace periods allowed during development. This would stimulate mining activity, probably leading to a new minerals boom in South Africa. We add that this rental would be in lieu of company taxes and preferably all other taxes as well.
When we come to Sasol synfuels, the problem is a little more complex. Sasol is said to produce ”crude oil” from coal at between $15 and $20 per barrel. This is considerably less than the price other oil companies are paying for crude oil. Assuming the costs of refining are similar for all refineries in South Africa, Sasol has a huge cost advantage. If there was a free market in petrol products, Sasol could easily undercut its competitors during times of high oil prices (higher than the cost of converting coal to oil).
The question (especially in relation to a proposed windfall tax) is whether Sasol’s technology is a national asset, or whether it belongs solely to Sasol. Sasol, of course, has exclusive use of this technology, but an argument could be made for saying that because of its antecedents, it is really a national asset similar to South Africa’s mineral patrimony, and for which a rental should be paid.
How this rent is calculated is another matter. In ideal circumstances where it could — and in our opinion, should — replace taxation, the formula-based resource rental may be a useful starting point. If the formula were to be based on a value-added-to-revenue ratio –or even an earnings before taxes, interest, depreciation and amortisation (EBITDA) to profit ratio — rather than a profit-to-revenue ratio, this may deal with the situation where Sasol was making minimal profits because of a low crude oil price.
Sasol already pays a considerable sum each year in taxes. In company taxes and the special tax on companies alone, Sasol paid R4,6-billion in 2005, R3,2-billion in 2004 and R4-billion in 2003. Including all other taxes, these amount to R5,7-billion, R5,6-billion and R6-billion, respectively. Using an EBITDA-based formula, these numbers would have been R6,6-billion, R4,9-billion and R5,7-billion respectively.
However, this is using the present gold mining formula, which is probably wholly inappropriate. But whatever formula is used, it should be based on a situation where the tax replaces all other taxes. Taxation is a payment to a factor of production like capital and labour, and there should be some relationship between taxation and value added. Currently, taxation is fairly arbitrary and is based on cash flows and what the government thinks it can get away with. This is probably why there are so many different types of taxes, and many are hidden to mask the full extent of taxation on a company.
The important aspect is that a resource rental concept is recognised in both the discussion document and the Mineral Royalty Bill. This concept now needs to be widened to include all resources that make up South Africa’s patrimony. Sasol is part of this patrimony — it needs to be both protected and pay its fair share of its ”rental”.
Stephen Meintjes is a stockbroker on the JSE. Michael Jacques is an honorary lecturer in the Wits school of accountancy. Both write in their personal capacities