“Are you quite all right?” someone from the Treasury inquired on Wednesday evening after I had offered a rather fulsome set of praise for the medium-term budget policy statement (MTBPS) on the radio. The solid presentation from the Treasury last week was an interesting opportunity to take a step back and consider wider issues about fiscal policy and policymaking in general.
There were so many numbers in the medium-term budget and together they weave a narrative defined by strategic choices by the cabinet, various committees and individual interaction between parts of the state and people within the Treasury. But at its core every number in the statement has layers of models and interactions and decisions behind it. More so they are all interconnected, between the years and upwards into a small set of headline numbers (like the debt-to-GDP share or the primary balance).
In such fraught debates over the budget within SA’s public square and calls for more spending and a “macroeconomic policy reset” it often gets forgotten that the budget output — in February or October — is a result of long institutionalised processes.
This machine of the budget can be criticised for being risk averse and slow or orthodox or overly conservative or all number of other things. But ultimately the system is a political choice on an ongoing basis and not a conspiracy theory. The cabinet is entirely free to decide to take a different course on macroeconomic policy if it decided it wanted to.
Yet it doesn’t happen. Some of the more easily alarmed might say it’s a mystical spell the Treasury has people under.
The real reason is far more prosaic and dull. It’s that the mass of interconnecting numbers simply don’t provide the flexibility for free lunches. If those free lunches ever existed in the recent years of loose global monetary conditions, they certainly don’t now. SA’s yield curve is actually higher now than it was three months ago, driven by higher global interest rates and outflows of investors from emerging markets.
After the MTBPS the Treasury has perhaps slightly overlaboured the point of “look what happened in the UK” (it’s a rather a low bar really), but it shows the important point that credibility can suddenly disappear in internationalised government bond markets.
The problem with the “reset macro” crowd is that they wouldn’t be able to feed their demands through the machine. When there is serious underspending in the government — not easily fixed until capacity and capability is fixed — there is no way that the state can suddenly take on more. The Keynesian idea that the state spends a load more money is cute but not feasible for a government machine that can’t deliver and can’t spend.
Infrastructure is perhaps the most instructive — this is where everyone across the aisle can agree. We need a huge step-up in the pace and scale of infrastructure build in the country with the existing pipeline simply not moving. This is where a huge amount of public and private sector effort is being undertaken to make it work but with little to show for it in the past five years.
This was an MTBPS in which the social relief of distress (SRD) grant became permanent de facto (if not yet de jure), given buffers put in to pay for it in the outer years beyond March 2024. It was also a medium-term budget in which messes created at state-owned companies (SOEs) were being cleaned up — unavoidably with more money, but now with strict conditionality that the Treasury will transparently apply. Overall, it took the extra revenue it now forecasts in the period ahead and split it roughly evenly between issuing less debt to protect the taxpayer and the fiscus in this global environment, and additional expenditure.
To go further the cabinet needs serious trade-offs. There are choices that could be made with the modest R57bn of buffers (given the risks about the budget) that the Treasury has in the outer year if reforms and growth and so revenue momentum continue. You could spend this on an SRD, you could spend it on basic education quality (given about 15,000-20,000 teachers are expected to retire each year from here — a silent crisis not getting enough attention), you could spend it on infrastructure, or you could spend it on better free basic health care.
Or you could (be forced to) spend it on additional SOE bailouts if political fiddling gets in the way of the Treasury conditionality (perhaps most obviously Transnet, which may well need more bailouts quickly) or use it on the wage bill if the current conservative line the government has taken crumbles.
All this will require a cabinet able to decide between alternatives and have a sense of priorities (which can be political of course).
Or the revenue path may just not materialise. If electricity reforms and the rest of the agenda is too slow to take root there may simply be no extra money available to make such choices and we will be stuck with the status quo. In a sense that is also a choice, albeit a passive one.
This is perhaps the central message of the medium-term budget: larger nominal GDP can lead to a greater number of options being possible, but the extra revenue won’t be big enough to make anything easy. This is why tax hikes may still be required by the Treasury in exchange for making the SRD permanent.
This larger economy, however, is still tenuous with hefty downside risks that could derail it, and many competing expenditure demands — which might force themselves to the top of the agenda as events unfold. There is not just one political demand on extra spending (the basic income grant, or BIG) there are a huge multitude of them. Reform and deepening the tax base would make this all much easier for more demands to be satisfied.