Seldom has the Laffer-curve downside of diminishing-returns been so well illustrated as by Treasury’s mining royalties bill. Forbes magazine’s David Warsh described how Arthur Laffer’s napkin-sketch of a simple convex curve captured an older idea that governments can tax too much, even in their own terms. They get nothing without taxation, they get nothing if they tax everything away at a rate of 100% or more, and in between lie two ill-defined regions.
At low tax rates, raising the rate increases net revenue and at high tax rates, raising the rate reduces net revenue. This tautology cannot be otherwise. But the relevant rates are unknowable, at least in advance. Since 1976, Laffer’s has been a highly influential notion, though regrettably not yet in South Africa.
A decade earlier, James Buchanan speculated on the motivations of ruling groups. They can’t know the tax rate that will maximise revenues while safely skirting the risk that large-scale ‘loss of temper’ will prompt just-surviving subjects to seek to throw them out. In self-interest they’ll keep testing higher tax rates. They may be more motivated by higher short-term revenues that force economic stagnation or decline than by higher long-term revenues expected to flow later from a growing economy. But self-interest also prompts them not to push their looters’ luck too far, and to look out for early warnings of disaster ahead. As should we.
In 1996, Harvard’s Martin Feldstein, president of USA’s National Bureau of Economic Research, contemplated government size. Noting that the central public finance question facing any country is the appropriate level of public spending, and therefore of taxes, he suggested that economists can help by clarifying the concept of ‘deadweight loss’. Here an increased tax rate negatively affects net revenues as the whole economic system adjusts. It may affect labour supply and various other changes that influence taxable income, so that the extra dollar taxed and spent by government costs the economy two dollars or more. Government’s self-interest then diverges from society’s, and redistribution causes economic decline.
So much for dull theory that cuts no ice with interventionist officials. Of course we’re nowhere near overtaxed, they respond. Look at government’s loyal 68% majority – our 3% annual GDP growth – the desperate poor and all those envied fat-cats – so much after-tax income wasted on dividends, booze and gambling – all the good intentions we’ve yet to implement – so many first-world taxes we’ve yet to try.
But Patrick Evans of Southern Era Resources is concerned that mining producers already pay total taxes equal to 38% of profit, compared with Chile’s 15.9% and no royalties. Anglo American’s Michael Spicer reminds us that the total cost of doing business in terms of taxes, levies and empowerment costs has to be kept in line with peer countries. Harmony Gold’s Ferdie Dippenaar reasons that shareholders’ money has to be spent where it gets the best returns. So how will nationalising mineral rights without compensation, followed by a new royalty tax, impact on mining? Helpfully, neutrally or harmfully? Well, do you ever get more of what you tax?
According to the Chamber of Mines, the industry directly contributed R66.8bn, or 7.5%, to GDP in 2001. Treasury’s Martin Grote reckons R4.2bn would have been collected from mining companies in the 2002 tax year by the proposed royalties, and Deutsche Bank’s Andrew Jackson calculates that the 2003 tax take would have been R6bn. That’s for the proposed royalties of 1-8% of gross revenues, including 3% for gold and 4% for platinum.
But at what cost? The Chamber of Mines says the 3% gold royalty potentially cuts the recoverable reserve base by 3.7%, from 16250 tons to 15 650 tons of gold. It does this by increasing fixed costs so that the just-viable “cut-off grade” rises from 4g of gold per ton of ore mined to 4.2g. On recent numbers, that raises working costs from R316 to R330.30 a ton of ore. Which sterilises 600 tons of gold worth R62.5bn or R1.6bn a year that will never be mined, just in order to raise R1bn a year in royalties from gold miners. Tax and spend a rand, and you destroy R1.60! Not to mention lost mining jobs.
What could be clearer? Gold mining tax is already well “over the top” of the Laffer curve and into the area of diminishing returns for the country as a whole. And the tangential corollary should be just as clear reducing existing tax on gold-mining would yield roughly a 1.6-times benefit to GDP.
We lack similar numbers for other mining sectors. But we can probably assume free capital movement between them and hence broad comparability throughout the mining industry, if not also throughout the rest of the economy. Excessive mining tax excessive tax in general already creates Feldstein’s “deadweight loss” to the economy. Additional royalty or other tax will increase that loss.
Is such ‘developmental transformation’, supposedly to bring historically disadvantaged South Africans into the mining industry, worth pursuing at the cost of shrinking the economy? How can it help that we are all made poorer?
But the government apparently doesn’t recognise the state as the problem and is sticking to its recently re-stated commitment to ‘developmental’ redistribution. So, as Buchanan and the Laffer curve suggest, its motivation remains to maximise short-term revenue regardless of moderate damage to the economy, and of course to retain power beyond the 2004 election.
Somehow we need to persuade government to take a longer-term interest in reducing taxes to promote growth and prosperity for all. South Africa desperately needs more growth-oriented policies. Government must remember that just as a rising tide raises all the boats, rapid economic growth increases the general welfare of the entire population better than any tax-and-spend preferential policy it can possibly devise.
Author: Dr Jim Harris is a freelance researcher and writer. This article may be republished without prior consent but with acknowledgement to the author. The views expressed in the article are the authors and they are not necessarily shared by the members of the Free Market Foundation.
FMF Feature Article\6 May 2003
