Cosatu’s plan to get Eskom out of debt
South Africa is facing its worst postapartheid economic crisis with state-owned companies holding crippling debt, but decisive use of pension funds could grow the economy.
Author:
14 February 2020
After two years of the government kicking the can of the restructuring of Eskom’s balance sheet down the road, the Congress of South African Trade Unions (Cosatu) has entered the debate with a novel proposal for a social compact for the public utility that could drastically reduce its debt and help stabilise its finances.
This intervention, which was accepted “in principle” by stakeholders at the National Economic Development and Labour Council last week, has started a discussion about the role of the Public Investment Corporation (PIC), the asset manager of the Government Employees Pension Fund (GEPF), in the economy.
At the end of September 2019, Eskom had debt of R454 billion. During the year to the end of March 2019, Eskom paid interest of R30.2 billion on its debt, which was equivalent to 16.8% of its revenues. In 2007, Eskom had debt of R40.5 billion.
But it spent R614 billion during the next 12 years until 2019 on its capital investment programme without a plan on how it would be financed. The Medupi and Kusile power stations accounted for R244 billion or 40% of the total. The government provided support of R83 billion towards the investment programme. The balance of the R614 billion was financed through debt and tariff increases.
It eventually became clear that Eskom could not trade its way out of its debt crisis. In February 2019, Eskom chief financial officer Calib Cassim said management had proposed that the government should take over R100 billion of Eskom’s debt.
In Parliament in September, former Eskom chairperson Jabu Mabuza said the company could only support debt of R200 billion. This meant that debt of R254 billion had to be taken off Eskom’s balance sheet, the financial statement that outlines what a company owns (its assets) and owes (its liabilities) and what its shareholders have invested in it (shareholders equity).
But the government has provided no support for the Eskom balance sheet. Instead, the National Treasury is supporting Eskom’s income statement, which shows the company’s revenues and expenses, or costs. The October 2019 medium-term budget policy statement allocated R138 billion to Eskom that will be paid in three tranches of R49 billion for 2019-2020, R56 billion for 2020-2021 and R33 billion for 2021-2022. This is meant to help Eskom with its interest expenses.
Crowding out other expenditure
This addresses the symptoms and not the root cause of the debt crisis. Eskom’s costs, excluding foreign exchange losses, were about 116% of revenues in 2019. The gap between revenues and costs is almost the same as the interest payments. So, Eskom keeps borrowing to pay off its debt.
With a debt of R200 billion instead of R454 billion, about 8% of revenues would get off the income statement. The drop in interest payments would be about R17 billion. The debt would increase at a slower rate until the company could take another 8% of revenues off its income statement to achieve a balance between revenues and expenses.
At this point, it is important to point out that Eskom employees are not the main reason for the company’s financial crisis. In 2007, Eskom made a profit of R6.5 billion. Over the next 12 years until 2019, Eskom’s average selling price of electricity increased five times. But its costs increased at a faster rate. Primary energy costs (coal and independent power producers) increased 7.6 times to 55.3% of revenue in 2019 from 32.5% in 2007.
Depreciation and amortisation, the losses of the value of a company’s tangible and intangible assets respectively that are charged against its income statement, increased 6.3 times to 16.5% of revenues from 11.8%. Net finance costs increased 17.7 times to 15.3% of revenues from 3.9%, as the debt increased 10.9 times to R440.6 billion. The only cost item that decreased was the employee benefit cost, which fell to 18.5% of revenues from 23.6%.
Although an Eskom social compact will require sacrifices from all stakeholders, reductions in employee benefit costs will not shift the dial. For example, a 25% cut over four years would only result in savings of about 1.6% of revenues a year. It makes no sense to focus only on employee benefit costs, which account for 18.5% of revenues. The other costs, the bigger fish to fry, which account for 97.5% of revenues, should also be part of the discussion.
Lifting R250 billion of Eskom debt to the national balance sheet would add about five percentage points to the country’s 60% debt-to-GDP or gross domestic product ratio. This ratio is not high by international standards. But the problem with this option is that it would have no impact on economic growth and investment or create a single job. The government must borrow to invest and create jobs and not to pay for investment that happened in the past.
A reason to not use the national budget to support Eskom’s income statement is that it crowds out other expenditure, which is then used to justify austerity policies or budget cuts that reduce the rate of economic growth, the Alternative Information and Development Centre says.
Opinions to consider
Other options must be considered. The PIC’s board discussed the issue of finding ways of reducing Eskom’s debt and interest burden during late 2017, according to a former board member. The asset manager has Eskom bonds of about R90 billion. One idea was for the PIC to convert its Eskom loan into shares in the company. Under this scenario, Eskom would no longer have to pay interest on its loan. But the PIC would get a share of Eskom profits, which would be paid out as a dividend. However, dividends have a lower return than interest payments on a loan. Eskom will not declare a dividend for a few years.
The PIC would make up for the lower return from an increase in the value of the investment. The talk at the time was that the PIC could eventually exit the investment through a listing of Eskom on the JSE.
“There were two arguments in favour of such a move. First, if Eskom failed, the country would also fail. There would be a risk that the pensions would not be paid. Second, there was an opportunity to buy Eskom at a low value and exit at a high price. The argument against the debt-for-equity swap is that it would replace a guaranteed high return investment with a less secure investment in Eskom Holdings,” the former PIC board member says.
PIC’s management later presented the idea to GEPF’s trustees, but it never went further because the asset manager did not follow up with a firm proposal, according to a source who was at the meeting. Last year, Magda Wierzycka, the billionaire chief executive of Sygnia, an asset management company, proposed that Eskom should issue a zero coupon bond to the GEPF, which is a loan with no interest payments. In 2019, President Cyril Ramaphosa’s sustainability task team proposed that R200 billion of Eskom debt should be taken off its balance sheet and placed in a special purpose vehicle (SPV), a legal entity created for a specific purpose under which investors can be pooled as a single entity.
The SPV would refinance the debt after raising $11 billion from foreign climate financiers and other funders in return for commitments to accelerated decommissioning of Eskom power stations. After blending cheap donor funds with commercial finance, the SPV would have a lower cost of capital. The problem with this proposal is that Eskom and the government would share payments on the interest in the SPV. Effectively, the transaction would not take the entire R250 billion off the Eskom balance sheet as the utility would still have to pay interest.
Cosatu’s plan
Cosatu has proposed a debt-for-equity swap to take R250 billion off the Eskom balance sheet. It will mobilise the PIC, the Unemployment Insurance Fund (UIF) and the Development Bank of Southern Africa to pool their funds in an SPV and invite other private-sector Eskom bond holders, according to spokesperson Sizwe Pamla. The SPV will then exchange its Eskom bonds for shares in the power utility. The UIF has assets of about R174 billion.
The PIC is the UIF’s asset manager. Eskom owes the Development Bank about R21.5 billion. This is equivalent to 24% of the bank’s R90 billion loan book. This is an excessive level of exposure. It crowds out infrastructure investment, which is the bank’s mandate. Enoch Godongwana, the head of the ANC’s economic transformation committee, says the proposed SPV should buy shares in a carved out profitable subsidiary of Eskom rather than at the level of the holding company, whose prospects are uncertain. In this way, SPV shareholders would get dividends. The problem is the time it would take to implement this proposal.
There are two ways to fund pensions schemes, through tax revenues (pay-as-you-go) or accumulated funds or savings invested in financial markets (pre-funding). Private-sector pension funds are pre-funded because a company can go bankrupt and have to pay all employee pensions on the same day. However, there is no scenario in which the South African government could close shop and have to pay the pensions of 1.3 million public servants on the same day. There will always be teachers, nurses and police officers. Therefore, internationally, most pension funds for state employees operate on a pay-as-you-go basis or with partial funding.
There are also two ways of designing pension schemes. In a defined-benefit scheme, the pension benefits are specified upfront and are not related to the value of a member’s contributions or the performance of a fund. In a defined-contribution scheme, the pension benefits depend on the value of the member’s contributions and the performance of a fund.
In South Africa, public-sector employee schemes were funded on a pay-as-you-go basis until the late 1980s. The contributions of the employer and the employee were always enough to pay pensions on a cash basis. But the apartheid rulers, fearing that a black government would not be able to honour pension payments, switched to a system of pre-funding.
The GEPF is a defined-benefit scheme. Pension benefits are guaranteed – based on years of service and final salary – and are not dependent on investment returns or the level of employer and employee contributions. This means that the workers do not benefit or make losses if the value of the assets in the PIC increase or decrease.
As former finance minister Trevor Manuel has pointed out: “Given that the GEPF is a defined-benefit fund, it would be inappropriate to consider any returns accruing from such investments to be benefitting the beneficiaries. This is simply because the pension benefits are predetermined. Such investments are essential to the extent that the employer [government] is able to meet its obligations to employees.”
This means that the PIC’s assets belong to the government and not the workers. The PIC is the government’s means of financing its obligation to employees. There is no evidence that a promise to pay that has a backing of financial assets is stronger than one that is backed by employer and employee contributions to a fund or through the tax system. Until 2013, employer and employee contributions to the GEPF were sufficient to pay all pensioners. Since then, there has been an increase in the number of pensioners and improved benefits for them. There is now a R28 billion gap between contributions and benefit payments.
The decisions on whether to pre-fund pension obligations to public-sector employees and the level of funding are political. The PIC had assets of R2.2 trillion at the end of September 2019, according to the Reserve Bank’s latest Quarterly Bulletin publication. This included shares worth R1 billion, government bonds worth R520.3 billion, state-owned company bonds worth R172.9 billion and cash of R143.9 billion. According to the latest actuarial valuation, the GEPF has a funding level of 108% compared with a target of 90% that was set by its trustees.
This means it has an excess funding of about R300 billion. This level of funding is obscene in a country that has such high levels of poverty. There must be a debate about the need for pre-funding. International pensions policy expert Nicholas Barr says in an International Monetary Fund paper: “I know that I will need to buy food for the rest of my life, but I do not accumulate a food fund. I intend to pay groceries out of future earnings.”
Making tough decisions
Cosatu has achieved a breakthrough in terms of finding a solution to the Eskom balance sheet.
But with South Africa in its worst postapartheid economic crisis, other options must also be considered to grow the economy and reduce Eskom’s debt. The government must look at the entire SA Inc. balance sheet. South Africa has a net loan debt of R2.8 trillion. The Road Accident Fund (RAF) has a net asset value (assets minus liabilities) of minus R242 billion, according to the treasury’s Budget Review publication.
However, SA Inc. also has PIC assets of R2.2 trillion and foreign exchange reserves of R834.1 billion. The UIF has a net asset value of R157.5 billion. The Compensation Fund has a net asset value of R41.4 billion. The three social security funds – the UIF, the Compensation Fund and the RAF – have a combined net asset value of minus R42.9 billion, according to the Budget Review.
Looking at this situation, the Reserve Bank should be part of the solution. It can provide the foreign reserves or other financing to plug some of these holes. There could be a mechanism for the two solvent social security funds to offset the RAF’s liabilities. If that is not possible, the UIF surplus should be used to reduce Eskom debt. The PIC could liquidate most of its assets and write off state and state-owned company debt, including the Eskom debt, worth almost R700 billion.
It could also dispose of some of its R1 trillion share portfolio – gradually to prevent share prices from collapsing – and convert the proceeds into infrastructure that will grow the economy and create jobs. Under this scenario, the GEPF could remain partially funded to cover the gap between contributions and payments. Alternatively, there could be an increase in contributions and/or a reduction of benefits. Under this scenario, there would be no need for a fund.
South Africa cannot afford another year of government dithering over making decisive interventions to grow the economy and reduce Eskom’s debt. The alternative – of pointless austerity, job losses and a collapsing economy – is too ghastly to contemplate.