Any old rule won’t do

The Quick Take

  • Over the past four decades, more than 100 countries have adopted fiscal rules to strengthen fiscal frameworks and improve debt sustainability
  • A fiscal rule enforces boundaries on fiscal policy and is aimed at course-correcting past fiscal measures which have generated unintended consequences
  • National Treasury is exploring the option of proposing a binding fiscal anchor for South Africa
  • A high-quality fiscal anchor is a much-needed sign of commitment to a more sustainable fiscal path

Marie Antelme is an economist with 23 years of experience as a market economist.

Mounting debt levels are often a sign of looming trouble. Invariably, escalating debt increases economic fragility as the burden of repayment crowds out investment, hindering growth and limiting potential. High levels of debt also inhibit governments’ ability to allocate funds where they are most needed or respond effectively to unexpected economic shocks.

Global government debt has risen at a historically unprecedented rate to unprecedented levels since 2010. History shows us that more than half of the periods of aggressive debt accumulation since the 1970s have ended in some form of crisis.

Prompted by these crises, and recognising the need to ensure fiscal sustainability, many governments have over the last four decades adopted ‘fiscal rules’ aimed at strengthening fiscal frameworks and improving debt sustainability. The IMF Fiscal Rules Dataset shows that from 1985 to 2021, 106 countries adopted at least one fiscal rule. Initially concentrated in developed markets, the adoption of fiscal rules proliferated in emerging and developing economies from the 2000s. The adoption of these rules was frequently in response to crises, but was also influenced by factors such as the adoption of the Maastricht Treaty in Europe, as well as the impact of volatile commodity prices on government revenues.

Against this background, and with South Africa’s debt now officially at 73.9% of GDP, National Treasury’s proposal to implement a binding fiscal rule is a welcome indication of an intervention. In February, with the tabling of the 2024/25 Budget, National Treasury made a new commitment, “(T)o chart a sustainable long-term path for the public finances, government will, after extensive consultation, propose a binding fiscal anchor. This will secure the benefits of fiscal consolidation and ensure that permanent fiscal imbalances do not reappear in a way that requires painful future adjustments. Over the medium term, the debt-stabilising primary budget surplus will anchor fiscal policy.”

Examining fiscal rules from around the world and assessing their efficacy offers insights into the impact that a potential fiscal anchor could have on South Africa’s fiscal sustainability.


A fiscal rule imposes a long-lasting constraint on fiscal policy through numerical limits on budgetary aggregates. The aim is to create a framework which corrects distorted incentives or time inconsistencies (those policies put in place at a specific time that may not always be appropriate as economic circumstances change) and limits procyclicality in fiscal policy. In other words, it enforces boundaries on fiscal policy aimed at course-correcting past fiscal measures, which have generated unintended consequences by limiting the ability of governments to over/underspend sufficiently to ensure that debt reduces relative to GDP.

The IMF identifies four types of fiscal rules:

  • Budget balance rules limit the influences on the debt ratio, which are under the control of policy makers. This may be the main budget or primary balance or some measure of cyclically adjusted balance.
  • Debt rules set an explicit ceiling, usually as a ratio to GDP.
  • Expenditure rules set limits on primary, total or current government spending, usually in absolute terms or by a reference growth rate.
  • Revenue rules set caps or floors on revenue aimed at boosting collection and/or managing ‘windfall’ revenue use.

All four rules have challenges. A budget balance rule can be hard to communicate and monitor. While debt rules are easy to communicate, they are more challenging to achieve because government is not always in full control of the things which add to the debt stock. For example, in-year ‘stock-flow adjustments’, which relate to the impact of exogenous factors like exchange rates or inflation on the valuation of debt instruments, can complicate this objective. Revenue and expenditure rules are sometimes politically difficult to agree because they limit the reach of the incumbent government. The most common rule is a debt rule coupled with operational limits on expenditure and/or the budget balance.[1]


Early iterations of fiscal rules had shortcomings which limited their effectiveness. Some fiscal rules were inflexible, without or with limited escape clauses. This shortcoming was particularly problematic in the global financial crisis (GFC), when governments needed to expand spending massively, but had limited ability to do so quickly. The Maastricht Treaty’s Stability and Growth Pact limits government main budget deficits to 3% of GDP. The pandemic prompted a further re-think of escape clauses, broadening the scope by which rules can be suspended. As a result, fiscal rules have been refined over time, and their design and implementation strategies have also been improved – which stood many countries in better stead when the Covid pandemic hit.

Several key features are recognised to enhance effectiveness, and the IMF uses the following criteria to measure the ‘strength’ of fiscal rules:

  1. The legal basis of the rule is an important feature of strength. Approximately 60% of countries embed the fiscal rule at or above its statutory level, either within fiscal responsibility frameworks, budget laws or, as is the case in Brazil and Denmark, even in the Constitution. Fiscal rules fortified by robust legal foundations are less likely to be changed when they become politically expedient.
  2. Effective implementation and the manner of implementation are also key determinants of rule’s robustness. Sufficient monitoring and oversight, and in some cases a correction mechanism, bolster the rule. Formal enforcement embeds the rule in the budgeting process, ensuring that governments are accountable for ex-ante compliance. The establishment of independent ‘fiscal councils’ to monitor and measure adherence to rules has also gained popularity and helped enhance performance.

While fiscal rules are popular, there is mixed evidence and even less agreement on how effective these have been in meeting their stated objectives. As a result, opponents often argue that countries which have adopted fiscal rules have either changed them when politically expedient, that the fiscal pain they inflict outweighs their benefits or that the rules did not guarantee better economic performance.

While evidence of countries achieving lower debt is mixed, studies show that the quality of the rule has a material impact on desired outcomes. It is not enough to have just any kind of rule. Using the IMF’s criteria, well-developed, high-quality ‘strong’ fiscal rules can significantly lower debt stock – particularly in an economic upswing – but fiscal rules which incorporate expenditure rules can also help reduce debt when the economic cycle turns down.[2] [3]


South Africa has witnessed a steady deterioration in its fiscal position since 2009, a trend that became more pronounced from 2012 onwards. Gross government debt troughed in Q3 2008 at 23.5% of GDP. Over the next three years, South Africa added R560bn to the debt stock, which rose to 41.1% of GDP. In the 2012 Budget, National Treasury committed to a programme of fiscal consolidation to stabilise the debt ratio over the medium term. Specifically, the 2012 Budget Review committed “(O)ver the next three years, growth in the economy will outpace growth in non-interest expenditure, and spending as a proportion of GDP will decline.” In the 2012 Medium-Term Budget Policy Statement, National Treasury implemented an expenditure ceiling, committing itself to real expenditure growth of 2.9% per annum.

However, it didn’t hold. A combination of much weaker-than-budgeted revenue collections and large expenditure outruns – in spite of the ‘rule’ – saw double digit growth in annual debt accumulation, starting in 2008. A dramatic increase in the wage bill was the biggest expenditure contributor to the excess, while support for Eskom started in 2008/09, with allocations to SAA and a range of other SOEs all contributing to the mounting spending.

More recently, the cost of debt has grown significantly. Combined expenditure on these three components over the past 15 years (from 2008 – 2023) grew at a compound annual rate of 9.6%. From the start, South Africa’s fiscal rule failed both ‘strength’ acid tests – it had no legal basis, it was not durably embedded in the budgeting process and so it was hard to implement without broader political commitment, and as a result demanded no accountability.

This period was also characterised by very weak growth. The average real GDP from 2008 to 2023 was 1.3%, despite the expansionary stance of fiscal policy. There are many reasons why growth was weak, but the economic impact of fiscal policy is undoubtedly one of them. The persistence of large deficits, as spending failed to adjust fast enough to weak revenue outcomes, not only drove debt stock up but also raised the cost of that debt. Rising risk premia has resulted in higher long-term interest rates, which crowd out investment. For the state, the rise in debt service costs has also crowded out spending in other areas, most noticeably on capital stock, which has had a long-term negative impact on service delivery. At the same time, the collapse of Eskom (which became evident in 2009 and where most of the SOE-related spending was directed) depressed activity and undermined confidence. This confluence of negative and reinforcing developments undermined the ongoing budgetary commitment (the ceiling features in all budget-related communication), and as of fiscal 2023/24, gross government debt is now officially 73.9% of GDP.

National Treasury’s commitment to a more effective fiscal anchor is a very necessary intervention. A numerical debt ceiling range (60%-65% of GDP, for example) coupled with some form of budget balance target (main budget or primary) seems the most likely recommendation, although evidence suggests an expenditure target is the most effective at lowering debt levels over time. The statement from National Treasury’s Budget Review 2024, quoted above, acknowledges the need for a binding fiscal anchor to avoid a painful outcome. It also implicitly highlights the need for both broader government buy-in and accountability. These features were notably absent from past efforts, hampering sustained consolidation of the deficit and curtailment of debt accumulation.

Another important consideration, which is not often explicitly mentioned, is that the adoption of a binding fiscal anchor should boost confidence, not only amongst citizens concerned by the economic and social implications of the unsustainable rise in government debt, but also among foreign investors whose capital is a necessary source of fiscal funding.

Similar to how the adoption and delivery of an inflation target cemented the credibility of the central bank, a high-quality fiscal anchor is a much-needed sign of commitment to a more sustainable fiscal trajectory, which could materially bolster confidence and improve the prospects of stronger, more inclusive growth.

[1] “Fiscal Rules and Fiscal Councils: Recent Trends and Performance during the COVID-19 Pandemic” IMF Working Paper, WP/33/11 Davoodi, Elger, Fotiou, Garcia Macia, Han, Lagerborg, Lam and Medas
[2] “Fiscal Rules in Times of Crisis”, World Bank Group Research and Policy Briefs, July 16, 2020

[3] “Fiscal Rules and Economic Cycles: Quality (Always) Matters”, IDB Working Paper Series, January 2023. Andrian, Hirs-Garzon, Urrea and Valencia

SA institutional readers
Global (ex-US) readers
US readers

Marie Antelme is an economist with 23 years of experience as a market economist.