The South African government has been heavily involved in energy sector development, as in many countries. The electricity industry is dominated by the state-owned utility, Eskom, while local authorities handle most urban electricity distribution. The state's Central Energy Fund owns the oil and gas exploration company Soekor, the gas-to-liquid fuels producer Mossgas, and was until recent years the owner of coal-to-liquid fuel supplier Sasol. State-owned Petronet runs the national liquid fuel pipeline system, while Portnet controls the ports used for import and export of fossil fuels. The nuclear industry was funded through the parastatal Atomic Energy Corporation (AEC), which still absorbs almost half of the Ministry of Minerals and Energy's budget. Many of these parastatal companies have received loans and subsidies of billions, or tens of billions, of Rands in the past several decades. The national government budget, therefore, only represents a small part of the state's involvement and financial burden of the sector (Trollip 1996). Quantifying the level of state involvement, and tracking this on a regular basis, is challenging for several reasons. First, some enterprises such as the AEC, and even Eskom, have significant non-energy businesses. In fact, the AEC has arguably not made any contribution to electricity production in decades, since the actual nuclear power plant at Koeberg is owned and operated by Eskom and the fuel is imported. Second, much of the investment and expenditure in the industry is only seen in changes in balance sheets, rather than as explicit investment. In 1994/95, for example, the Central Energy Fund increased its provision for non-repayment of loans made to Soekor by more than 110 million Rands - and this is, in essence, a state subsidy (Trollip 1996). Thirdly, it is difficult to distinguish between 'investment' by the state (ie procuring fixed assets) and 'expenditure' (ie operating expenses for government or parastatals) - and it would be misleading to call all of the operating expenses a 'burden', because the state companies may have revenue in excess of these expenses. This is certainly true for Eskom, which has posted healthy profits in recent years even after funding the majority of the electrification programme internally.
The magnitude of this state expense is clear from what information is readily available. In 1995/96, the energy-related budget of the Department of Minerals and Energy Affairs (the former name for the Department of Minerals and Energy) was 515 million Rands, including 489 million (95%!) to the Atomic Energy Corporation. Even removing the AEC repayments for apartheid era loans, total national government energy expense is 320 million Rands, of which 273 million was for the AEC. In the same year, the Central Energy Fund increased provisions for non-payment of loans to state companies by 7.3 billion Rands. Subsidies to the synthetic fuel industry, which are now being phased out, totalled 1.5 billion Rands (Trollip 1996). Capital investment by Eskom was more than 5.4 billion Rands (Eskom 1996). But even this does not capture a range of other institutions - such as the local electricity distributors and the transport/pipeline companies - the investment by the subsidiary state oil companies, research and development in energy that is funded outside the DME (eg from Eskom and the National Research Foundation), or expenditures on infrastructure such as ports for coal handling. For comparison, South Africa's GDP in 1995 was 548 billion Rands (SARB Mar 00). It has not been possible to quantify this indicator for this report, given the limited time and data availability. Ideas for further research needed for quantitative estimates are presented in the notes section below.
Notes to SEW or next year's Observer-Reporter:
To estimate the total non-renewable government investment in the sector, several questions should be clarified:
First, whether the focus should be on all expenditure, capital investment only, or some mix of the two (eg capital expenditure for the parastatal companies but expenditure for government departments).
Second, how to account for the fact that parastatal enterprises may have revenue that offsets expenditure on non-renewable energy.
Third, whether only assets controlled by the South African government should be included - eg oil company Engen is majority owned by Petronas, the Malaysian state oil company.
Once these have been answered, then several steps will be necessary to make an estimate of total investment or expenditure:
A breakdown of the DME budget to identify only those activities related to energy, as well as what share of minerals budgets can be apportioned to coal (if any) - eg, is the AEC budget actually expenditure on energy?
An analysis of what share of parastatal enterprises with non-energy business (eg AEC, Mossgas, Portnet) should be considered energy-related.
A thorough analysis of the Central Energy Fund, Soekor, Mossgas, and Strategic Fuel Fund accounts to assess what actual investment and expenditure is (ie if this is in the balance sheet or even off balance sheet instead of in the income or cash flow statements). Note that the Central Energy Fund has not produced an annual report since 1996/97, but is expected to produce one later this year.
A review of the subsidy polices of government to assess how much explicit or implicit subsidy is going to synthetic fuels manufacture and atomic energy
Check that the funds for regulatory agencies (National Electricity Regulator and National Nuclear Regulator) are captured in the DME budget.