Ruling party electoral bruising put spotlight on the MTBPS

Budgets don’t happen in isolation. They reflect the political priorities of the government of the day, which are usually linked to expectations of voters.

However, these choices are constrained by the domestic and global economic outlook, and since governments borrow large sums of money in the bond market, prevailing market conditions. Given these realities, it is all about trade-offs: between competing visions of how the economy should be run, the demands of different interest groups, maintaining continuity and accelerating change and the needs of today and tomorrow.

Last week’s Medium Term Budget Policy Statement (MTBPS) was no different. What made it more interesting is the electoral bruising the ruling party suffered in the recent municipal polls. However, the MTBPS would have been largely finalised before the 1 November vote and it contained no move towards populism. This might also reflect the fact that the election itself showed no shift to the left on the part of voters, who were presumably preoccupied with crumbling municipalities, rather than clamouring for giveaways.

In fact, the MTBPS maintains a strong commitment to fiscal discipline, putting state finances back on a sustainable path and implementing reforms to remove obstacles to faster economic growth. Though delivered (capably) by a new Finance Minister, Enoch Godongwana, this represents continuity, something investors will welcome for now. Taxpayers will be pleased that no additional tax measures were announced. At this stage, a lack of surprises is a good thing.

Assumptions
The MTBPS updates the February Budget by presenting planned spending and revenue for the next three fiscal years. As such, it is based on an assumed economic growth outlook. Though the future is always hard to predict, getting it more or less right is important. Being too optimistic in tax revenue assumptions can lead to persistent increases in borrowing. In fact, every year since 2008/09, spending has exceeded tax revenues, which were consistently projected to grow faster than they actually did.

The economic outlook as it stands today is much better than a year ago, though new headwinds have emerged recently: disruptive loadshedding as Eskom’s coal-fired power stations again show their age; the squeeze on business and consumer incomes from surging fuel prices; and softer export prices.

National Treasury estimates that the economy will expand by 5.1% in 2021 from -6.4% in 2020. However, growth numbers for 2022 to 2024 are quite conservative around 1.7% per year.

As has been widely reported, this year’s growth spurt has substantially boosted the state’s coffers compared to what was expected at the start of the year. Tax revenues for the current fiscal year are now expected to come in R120 billion ahead of what was pencilled in in February. This was largely, but not exclusively, due to the surge in dollar prices of a range of commodities. Efficiency gains at SARS also seem to have played a role. Several key commodity prices, notably rhodium and iron ore, have come down to earth again. Looking at the total export basket, South Africa’s main commodity prices are down a third since April, but still almost twice as high as before the pandemic.

Chart 1: Index of the dollar prices of SA’s main commodity exports

Source: Refinitiv Datastream

It would be foolish to base future planning on the assumption that commodity prices will remain at these levels. They are inherently volatile and largely unpredictable. Too many fiscal disasters in emerging markets stem from making the mistake of projecting windfall commodity prices into the future. The good times never last forever. Fortunately, Treasury is very aware of this risk, partly because it made this mistake before.

While tax revenues were better than expected compared to the 2021 Budget forecast, they are still lower than what was projected before Covid hit in February 2020. In other words, the recovery is going better than expected, but we are still recovering nonetheless. Talk of a ‘tax windfall’ to be spent on pressing social needs is therefore somewhat misplaced. We were lucky to be bailed out by commodity markets, but our economy and its tax-generating capacity is still worse off compared to a no-Covid scenario.

On the spending side, there is a broad commitment to maintaining discipline without resorting to deep cuts that can end up being counterproductive. But there is an almost endless list of compelling needs to test that commitment. The two biggest remain government’s wage bill and social security spending. A third, more SOE support, has emphatically been ruled out.

Moderation in public sector wage increases remains crucial. Treasury pencilled in only 1% average annual growth in the wage bill over the next three years, but negotiations with public sector unions has barely begun. Meanwhile, the Constitutional Court still needs to rule on whether last year’s wage freeze was legal. If not, around R40 billion in backpay will have to be found.

The calls for a form of basic income grant grew louder following the July unrest, but with the extension of the R350 per month special relief of distress grant to March, there is space to leave the matter until February. It is reasonable to assume that it will be extended again until a final decision is made. At around R40 billion a year, this is affordable, but will require spending reprioritisations and possibly somewhat higher taxes. Anything more than that implies a fundamental restructuring of the Budget. The MTBPS makes it clear that any expansion of the social safety net cannot be financed by more borrowing. Either it will be funded by closing existing programmes, or by raising new taxes in an economy that is already taxed to the hilt.

Better debt profile
The difference between spending and revenue is the budget deficit. Thanks to the tax revenue overrun and StatsSA’s rebasing of its gross domestic product (GDP), it has improved markedly. From 10% GDP last year, it is estimated to be 6.6% of GDP in the current fiscal year. This is set to decline to below 5% by 2024/25. In other words, it is better, but by no means good.

The primary deficit excludes interest payments from the equation. It is expected to dwindle from 2.3% this year to a small surplus in 2024/25. This means that most of the overall deficit will go towards interest payments. In fact, of every R1 in tax revenue, 21 cents will go to paying interest on average over the next three years. This is not so much because overall debt is excessive, but rather because interest rates are so high. Interest payments are therefore crowding out other areas of spending. It is this simple fact – not ideology, not politics - that necessitates the stabilising of overall debt.

Chart 2: Primary balance and interest payments as % of GDP

Source: National Treasury

If all goes according to plan, Treasury now expects overall debt to peak at 78.1% of GDP in 2025/26, earlier and at a lower ratio than previously estimated. In rand terms, this amounts to around R5.5 trillion. That sounds like a scary number, but remember the flipside: it is R5.5 trillion in interest-bearing assets, 66% of which is held by local institutions including banks, pension funds, asset managers and insurers.

Chart 3: South Africa gross government debt to GDP ratio

Source: National Treasury

Ultimately, faster economic growth is needed to square the budget circle. That is pretty much the message from the market too. South Africa’s government debt level has increased alarmingly over the past decade, but is not high by global standards. But where other countries can “grow into” their debt, there is little confidence we can, especially not with our high interest rates. We are stuck in a vicious circle: investors charge high interest rates because they are worried about debt, and debt cannot decline fast enough because interest rates are too high. Again, sustained economic growth can interrupt this negative spiral. This requires more progress on the various reform initiatives that are underway. Few of these programmes are under Treasury’s direct control and require other government departments to act with far greater urgency. However, Treasury’s Operation Vulindlela, a select unit that works with these departments to remove growth obstacles, seems to be achieving success.

Global markets crucial
Finally, global market conditions are extremely important. These can potentially become more unfavourable in the coming year as the US Federal Reserve and other major central banks reduce pandemic-related emergency support. This is widely expected and therefore should not upset markets, unless it happens at a faster pace than what is priced in, for instance if these central banks suddenly become more worried about inflation than they have been up to now. Fortunately, the first big test, the Fed’s announcement of a scaling back of its monthly bond purchases, has gone off without a hitch. However, last week’s US inflation numbers were much worse than expected, and caused market jitters, including a weaker rand.

On the plus side, South Africa’s current account surplus means we are less dependent on global capital flows compared to the previous Fed tightening cycle. However, this surplus is expected to shrink to a small deficit by 2023 as the export surge fades and imports normalise.

Chart 4: South African 10-year local currency bond yield, %

Source: Refinitiv Datastream

Risks, and opportunities
In a nutshell, the fiscal outlook has improved substantially over the past year, but is far from rosy. Another economic shock could easily knock the plans off course. Meanwhile, the government’s political commitment to taking the bitter medicine of fiscal consolidation (or rather, dishing it out) will continuously be questioned, particularly in the face of electoral pressures. However, none of this is news.

South Africa’s elevated long bond yields reflect these risks and more. Even when commodity prices were at their most extreme levels earlier this year, local bonds failed to rerate materially. The risk of the government defaulting on its debt – i.e. failing to repay interest or capital – is very low. South Africa does not have a history of defaults on its bonds, and most debt is issued in rands. The profile of government debt maturities over the coming years is not particularly alarming, though 2029 will see R350 billion in bonds that will need to be repaid or rolled over.

Chart 4 shows that the gap between South Africa’s 10-year bond yield and the US equivalent has declined from the pandemic-panic levels, but remains high by historical standards. There is thus a wide margin of safety in local bonds, offering one of the best fixed interest investment opportunities globally.