ERSA’s fiscal future webinar series (II)
22 October 2020
The first brief introduced the themes, the participants and then the first two topics; first Treasury and the South African Reserve Bank’s Response to COVID, and second the fiscal deficit and fiscal sustainability. In this brief we cover the conversation around government debt, including a presentation made by Treasury. Brief three discusses economic growth, brief four the credibility of South Africa’s financial institutions and complacency risk, and brief five covers the policy reforms which South Africa needs to make.
INTRODUCTION
Government’s debt is closely linked to the fiscal deficit and fiscal sustainability. This is because other than tax revenue, government bonds are a major source of funding for government’s expenditure requirements. As such, the seminar participants were once again grouped into two general camps: one which thought government debt was too high or heading that way and urgently needs to be reined in, while the other believes the debt level is not too high and bonds should be issued at an increased rate.
In order to provide context, we will first cover the points from a presentation made by Tshepiso Moahloli from government’s Treasury department.
TREASURY’S PRESENTATION
As a result of the COVID crisis government’s borrowing requirement has more than doubled since February 2020, from R344 billion to R777 billion.
There has been a rise in government borrowing costs with the rise in uncertainly, market volatility and a lack of trading and liquidity. While the downgrades imposed by the international ratings agencies have negatively affected South Africa’s credit quality and made funding more expensive. A significant amount of revenue is consumed by debt servicing/interest costs, which is expected to consume over 20% of government’s total expenditure in 2020/21. A level of 25% would indicate a high probability of debt default.
South Africa’s debt-to-GDP ratio is also at the worse end of the spectrum when compared to its peers, and the growth in debt has constantly been outstripping growth in GDP, crowding out service delivery. Government is also not earning enough revenue to pay back accumulating debt and borrowing has to be used for this, which may cause a vicious cycle that could lead to a foreign debt crisis.
The funds raised specifically for COVID cannot be expected to be raised indefinitely. With regards to the possibility of tapping South Africa’s private sector investment funds to fill the gap, the private sector’s assets-under-management (AUM) are not growing as fast as government’s borrowing costs. And while AUM is significant (R15.5 trillion) it is all invested and there are limits to the various instruments they are legally allowed to invest in. The proportion of government debt held by non-residents also continues to decline.
When interest rates are higher than growth as is currently the case in South Africa, the government debt to GDP ratio increases. When growth is higher than interest rates it falls and that is what we should be aiming for.
ARGUMENTS FOR INCREASING GOVERNMENT DEBT
It is not the level of debt that is an issue: instead it is what you do with the money raised and the cost of the debt that matters, i.e. the interest charged on it. When there is lack of certainty about the future you cannot expect long-term rates to reflect future outcomes. We have to solve the high long-term rates problem or we will run into a debt trap. Money creation is a stimulus to the economy. The inflation outlook has improved and therefore the point is to help the recovery. There is a lack of imagination with regards to the bridge needed to stimulate the economy.
There is a disconnect between South Africa’s low inflation and high interest rates, and we need to somehow decrease interest costs without decreasing expenditure – which will result if government issues less debt and fewer funds become available. While some countries increase inflation in order to help pay away their debt (inflation increases tax revenue as wages and prices increase), that is not the attractive option. Better options would be to decrease borrowing costs through quantitative easing (QE), borrowing more short-term since these interest rates are lower than long-term rates, using support from the World Bank and IMF, or greater intervention by the South African Reserve Bank (SARB). This last option would allow the SARB to make loans at a subsidised rate compared to the development banks, and also buy a much larger amount of government bonds.
We currently have higher debt than otherwise due to past inflation miscalculations by Statistics South Africa. Inflation was in fact 2% lower than we thought and budgeting was done on this basis.
Not all government debts are equal. We need to take into consideration maturity, currency and ownership. Short-term maturities command a lower interest rate in South Africa and rand denominated debt does not carry exchange rate risk; therefore we should issue more short-term debt in rands and if it is South Africans who own the debt they will be less likely to sell their holdings.
A sectoral understanding is also important. In South Africa household and private sector has been deleveraging (decreasing their debt levels) and it is up to government to pick up the slack.
Lastly, the rand is a fiat currency[1] and therefore there is no default risk.
ARGUMENTS FOR REINING IN GOVERNMENT DEBT
Of the measures proposed to decrease interest rates, financial repression would be one of the less good ways and hard to exit from. Financial repression is when government channel funds from the private sector as a form of debt reduction which results in the government being able to borrow at lower interest rates, thereby obtaining low-cost funding for government expenditures.[2]
South Africa’s debt levels are an issue because of low economic growth and the associated debt service costs are becoming an increasing portion of government’s expenditure. If government has to spend 25% of its revenue on servicing the interest attached to its bonds, its ability to provide sufficient service delivery will be placed under extreme pressure. Thus we are at a point where we must respond to a debt crisis.
South Africa’s tax revenue is not lower than norms but its debt is higher than the emerging market average, and once debt is beyond a certain point debt service costs increase rapidly.
A saving grace for South Africa’s government debt has been that the majority of it is long-term – which allows more time to pay it back – and is mostly denominated in rands (around 75%) – which lowers exchange rate risk should the value of our currency depreciate. However, the composition of the debt is changing from longer-term to short term and the level of foreign currency debt is also rising. Should the change happen quickly it will cause big problems.
Government’s borrowing needs are currently greater than national savings, which is shocking because if you want economic growth you must rely on foreign savings, and now it looks like there is a probability of default which discourages foreign investors. We must also not take market access/attracting investors to buy our bonds for granted. Each notch we are moved below investment grade adds 1% to 1.5% to the cost of servicing our debt, versus being downgraded a number of notches while rated investment grade, which adds 0.5%. Liquidity crises arrive suddenly, and our increasing debt level is increasing the probability of turning investors away.
Increasing debt decreases long-term consumption over time and deleveraging (lowering debt levels) will have to happen sooner than later, especially since GDP growth is at 1%. And if government inefficiency is crowding out the private sector, how would more government borrowing increase growth?
By Charles Collocott, Researcher, HSF, 22 October 2020
[1] Fiat money is a government-issued currency that is not backed by a commodity, such as gold. This gives central banks the potential to exert greater control over the economy because they can control how much money is printed.
[2] https://www.investopedia.com/terms/f/financial-repression.asp
