Sarah Truen
South Africa’s relationship with state owned enterprises is a little unusual. In many other parts of both the developing and developed world, governments and societies have increasingly realised that government is bad at running business, and that removing government from business is generally a good idea. However, in South Africa, privatisation remains a dirty word, for a variety of historical and ideological reasons. Instead we have opted to pursue a development state, where the government keeps its hands firmly on the levers of a large proportion of the economy, ostensibly allowing policymakers more ability to intervene directly in economic outcomes. While state intervention can have an important role to play, for example in sectors where provision of services is important, but difficult to sustain on a commercial basis, there are often ways to achieve these objectives without resorting to state ownership of the organisation delivering the good or service. Smart regulation is one of these ways.
Partly because privatisation has become a dirty word in policy circles, we do not talk enough about exactly what impact our obsession with state control has on our economy. A little analysis of some high level data reveals some interesting insights, which should be more widely discussed. The table below shows some financial data for 17 of our largest state owned enterprises. A rough way of estimating GDP contribution is to add together taxes paid, employee compensation and net profit. On this basis, these SOEs made a contribution to GDP of R78.3 billion in the year to March 2015 (roughly 2% of GDP in that year).
Note: all data is derived from company annual reports, and is for the year to end March 2015, except SAA (March 2014) and Rand Water (June 2015). Tax figures reflect only corporate income taxes, not VAT on goods sold, for example, and are thus an under-estimate of total tax contribution.
As can be seen, the bulk of this GDP contribution was via compensation of employees. These companies paid very little tax, largely because as a whole they made losses, not profits (and because three of them are exempt from corporate income tax).
The opportunity cost of poor performance at SOEs can be easily estimated. Say, for example, profits at SOEs had just managed to match the repo rate, at a 7% return on assets (it should be borne in mind, this was a year when the JSE rose by over 30%, so a 7% return on assets is very, very conservative). The total asset base of the 17 companies shown above was about R1.4 trillion as at March 2015, so a 7% return on that would be about R99.5 billion. Say those companies then paid an effective tax rate of 28% (excluding the tax exempt ones), and for simplicity assume that employee compensation stayed the same. The GDP contribution they made would increase to R196.7 billion. The increase from the actual level would be equivalent to 3.1% of GDP. To put this into perspective, the actual rate of GDP growth we saw in 2015 was 1.3%, so the order of magnitude of the efficiency gains possible are highly significant.
This analysis is greatly simplified, and in practise considerable caution would need to be exercised in setting aggressive profit targets for SOEs. For example, if the SOE has a monopoly over an essential piece of infrastructure (which a number of them demonstrably do), then high levels of profit could reflect extremely problematic abuses of monopoly power. Conversely, where an SOE is meeting a social goal, and doing so reduces its profitability, some allowance must be made for lower sustainable levels of profit than would be expected in the private sector (and this is one of the reasons why several of the SOEs are tax-exempt, to allow them to stay commercially sustainable despite the burdens of their social obligations). However, if an activity is not commercially sustainable, even with tax concessions and subsidies, it is hard to see that it is suitable to be run as an SOE, as the SOE model itself arguably assumes that the activity can be run in a commercial manner.
Even given these considerations, however, it seems clear that simply running our SOEs better would have enormous potential to generate economic growth. And remember, these are the companies that transport our goods, generate our electricity, deliver our post, and build our telecoms networks. Their performance impacts on every other company in the country, because inefficient delivery of these economic basics impacts on the efficiency and competitiveness of the economy as a whole. Given what is at stake, it is a shame that the policy debate around SOEs is still so stifled.