This column was first published on Business Day
Contagion from SA’s state-owned enterprises crisis is heading for its development finance institutions.
It started with a plan tabled by Cosatu two weeks ago that has inexplicably gained some level of support from business to government. It is, however, totally unworkable.
It proposes to shift R250bn-worth of Eskom’s R450bn of debt into a special purpose vehicle funded by the Development Bank of Southern Africa (DBSA), the Industrial Development Corporation (IDC) and the Government Employees Pension Fund (GEPF).
But no-one seems to have considered that this is mathematically impossible.
Let’s think for a moment about what the balance sheets of the DBSA, IDC and GEPF can accommodate.
The DBSA, which has a mandate to fund development infrastructure, has a balance sheet of R89.5bn in assets. Like any bank, it has risk management policies that include the concentration risk it can take to any one entity.
According to its most recent annual report, it has a limit of R500m of exposure to a single government entity. Its board recently allowed an exception to that to lend R3.5bn to SAA with a sovereign guarantee, a move that has already weakened the overall risk position of the bank.
While the DBSA is not subject to the Banks Act, the concentration risk guidelines set by Basel 3 would be a line it can’t cross without risking a run on its own funding. Basel 3 limits the maximum exposure a bank can take to 25% of its capital, which works out to R9.2bn for the DBSA. Given that the DBSA already holds Eskom debt, some of this exposure is already deployed.
Moving on, the IDC has a mandate of funding industrial development. It has total assets of R144.6bn. Its own risk framework specifies, in line with Basel 3, that it can take no exposure to a single industry of more than 25% of its capital. That works out to R24bn.
But given that the IDC has been a major funder of power projects, it is already heavily exposed to energy generation, so less than half this limit is likely to be available for further exposure to Eskom.
The GEPF seems to have serious firepower with over R2-trillion in assets. But it also has to apply prudent risk management.
According to its latest annual report, at end-March 2019, it already holds R84.5bn worth of Eskom bonds, which represents half of the public sector debt it holds, and 15% of the entire debt asset allocation of its portfolio. This is already a substantial concentration. The GEPF cannot absorb more exposure without stepping far from acceptable investment risk management practice.
The GEPF is a defined benefit pension fund. This means that its members are not exposed to the financial performance of its investments. The government guarantees the fund. So any shortfall it faces must be covered by the national budget and therefore, the taxpayers. While Cosatu has suggested that using the GEPF represents the workers coming to the party, it is nothing of the sort.
It merely transfers additional risk onto the government balance sheet. One suggestion has been to convert the R84.5bn of debt the GEPF holds into equity. But the resulting equity would have to be valued at zero, representing a claim on a company that has been losing billions, so the government would face an immediate contingent liability of an equivalent amount to make good GEPF’s funded status.
But even if the GEPF doubled its Eskom debt exposure, taking it into unacceptable risk territory, the three institutions between them would not be able to absorb even R100bn of Eskom debt. That is not even half what Cosatu imagines.
It is astounding that the plan has gone as far as it has, despite this obvious maths.
Cosatu has also suggested that prescribed assets could be used to compel the private sector to take on exposure. Prescription is required when the normal risk and return features of an asset do not justify investment by institutional investors.
While the GEPF is a defined benefit scheme, many of those that would face prescription are not, so workers would be facing a loss in the value of their pensions. Prescription is a stealth tax on savers.
The question is whether it is optimal to use such a stealth tax, or alternative taxes including VAT and income tax instead. Given the damage prescription will have to normal financial market incentives and to the savings of pension members, alternatives should be considered.
Cosatu should be concerned about the existential threat to the state’s solvency. The one sure way that pension members will lose is if the government tips into bankruptcy. There would be a dramatic curtailment of worker’s benefits.
Cosatu has clear motive to bring ideas to the table, but they need to focus on what matters: government solvency. A good suggestion would be, for example, to accept a wage freeze in the public sector for three years. That is the kind of move that would contribute to avoiding total calamity and eventual collapse into the arms of the IMF.
So what should be done about Eskom’s finances? The one option the government seems to be ignoring is to reschedule the Eskom debt. That means forcing Eskom’s debt holders into more favourable terms. It could do that by switching Eskom bonds into sovereign bonds with different terms, removing debt from the Eskom balance sheet onto the national balance sheet.
Simultaneously, it could seriously embrace green financing facilities from around the world that would provide it with discounted funding while following through on the Eskom restructuring it has promised.
At least such an approach would avoid adding the development finance institutions to the list of financial disaster areas in the public sector.
- Theobald is chairperson at Intellidex