Fiscal Rules, OK? Managing the Public Finances After Covid-19

At over £350 billion, the budget deficit this year will break all peacetime records for public borrowing as a proportion of national income. With the economy in need of fiscal support for some time yet, and ultra-low interest rates reducing the burden of debt servicing, a growing stock of national debt will not pose a problem in the short term. But a debate has sprung up around how much fiscal restraint is required in the longer term.

With the budget looming, the Treasury is expected to seek to revise its fiscal targets to govern borrowing in the years ahead. Fiscal rules have been used in recent decades to reassure voters and market participants of governments’ commitment to careful stewardship of the public finances. But their record has been patchy, and many past rules have failed in the wake of economic shocks at precisely the moment that uncertainty about the public finances is heightened.

Over the past year there has been a vibrant debate, in the UK and internationally, about the appropriate design of fiscal frameworks. On the one hand, some have tended to focus on switching targets for the stock of debt for ceilings on debt-interest payments. These approaches are a step in the right direction, but they tend to lead to rules that are either too tight to allow optimal policy in the face of an economic shock or too loose to impose much control on borrowing.

On the other hand, several eminent economists have proposed looser sets of “standards” or “guidelines”. However, such a move towards using discretion rather than rules in the management of public borrowing seems unsuited to the UK context, not least because they would not achieve the goal of making government clearly accountable for keeping the public finances sustainable.

We believe that it is possible to combine clear accountability with a rules-based fiscal framework that also allows fiscal policy the much greater flexibility that a low-interest-rate world demands.

Our proposed framework involves four elements:

1. Governments would choose a long-term debt anchor, which is translated into an annual deficit ceiling.

2. A “real-time affordability test” is then used to adjust the deficit limit up (when borrowing is cheap) or down (when real interest rates are high) according to the economic context.

3. An “escape clause” would suspend these rules to allow fully flexible fiscal policy in a downturn.

4. An overarching goal of increasing public net worth, which would ensure prudent spending and investment decisions.

One turbulent year on from the onset of the pandemic, our proposed fiscal framework still seems like the approach to balancing the competing objectives of fiscal policy in the 2020s.

The budget deficit for the current fiscal year looks set to rise to over £350 billion, breaking all peacetime records for public borrowing as a proportion of national income. The figures presented in the forthcoming budget will be dire, and it is likely that the red ink will continue to flow for several years.

It looks unlikely that there will be immediate spending cuts or tax measures at the budget in response. Indeed, an extension of pandemic support measures is likely to be announced which will lead to debt levels peaking higher than the 110 per cent of GDP currently forecast. A slower recovery will only add to the need for more borrowing.

With the economy in need of fiscal support for some time yet, and ultra-low interest rates reducing the burden of debt servicing, a growing stock of debt will not pose a problem in the short term. But a debate has sprung up around how much fiscal restraint is required in the longer term. The government will need to demonstrate that public debt remains on a sustainable path.

In recent decades, fiscal rules have been used to reassure voters and market participants of governments’ commitment to careful stewardship of the public finances. But their record has been patchy, with numerous frameworks abandoned in the face of changing economic circumstances. And with debt-service costs now so low, some see such constraints as no longer necessary and actively harmful to good macroeconomic policy.

In this paper we argue that a framework of rules remains important to good fiscal management but that it is possible and desirable to take a very different approach to those adopted in the past. The new fiscal framework must allow sufficient flexibility in the face of shocks and provide scope for fiscal policy to play a macroeconomic stabilising role, while also maintaining long-term debt sustainability.

In light of the past year, in which the Covid-19 pandemic would have tested any proposed framework to the limits, we discuss different proposals that have been made and assess their suitability.

A new fiscal framework will need take into account that the role of fiscal policy has changed. Before the 2008 global financial crisis, there was a consensus among economists and policymakers that monetary policy should be the primary tool for stabilising the macroeconomy, with fiscal policy providing some support through automatic stabilisers (that is, the tendency for tax revenues to fall and spending on unemployment benefits to rise in the face of a downturn) but generally focused on keeping public debt sustainable. The situation where interest rates are at their lower bound of zero (meaning monetary policy is unable to fully respond to a downturn) was discussed for the most part only as a theoretical possibility in economics textbooks. Even after the Bank of England cut interest rates to 0.5 per cent in 2009, many economists were supportive of plans for aggressive deficit reduction, believing, in line with market expectations, that interest rates would rise again relatively swiftly and that monetary policy would be able to adjust to keep the recovery on track.

But after ten years of ultra-low interest rates and poor economic performance, the picture looks very different. The need for fiscal policy to play a more active role in stabilising the economy is now much more widely understood. Tomorrow’s fiscal rules must therefore both provide the flexibility that governments require to respond to shocks and sufficiently constrain fiscal policy during normal times to ensure public finances remain sustainable.

With these considerations in mind, the chancellor announced a review of fiscal rules in the March 2020 budget, with the intention of publishing the outcome in the autumn budget last year. Although the budget was delayed, the result is still expected in this year's March budget. In this paper, we examine the fate of past fiscal rules and consider how existing and proposed fiscal rules have stood up to the challenge presented by the pandemic. We go on to outline our own proposal for a new approach that addresses the weaknesses of past attempts. We also consider how these different proposals fulfil the objectives the government has set out for the next generation of fiscal rules.

The Conservative party indicated its intention to reset the fiscal framework during the 2019 general election campaign. In any case, the pre-existing rules, set by Philip Hammond as chancellor in 2016, would have been shattered by the Covid-19 pandemic: Far from falling as a share of GDP in 2020–21, public debt will likely jump by almost 20 per cent of national income. But before we turn to suggestions for the future, it is worth examining how we got here. Table 1 sets out the different fiscal rules that have existed in the UK since they were first introduced in 1997.

What lessons can we learn from the track record of fiscal rules in the UK? First, fiscal rules have successfully tied government's hands at least to some extent, with a number of rules having been met, others surviving for a while and others coming close to being achieved. Even if there has been some fudging to meet the rules – for example, around changing the start and end dates of the economic cycle under Labour^[_] and the coalition government delaying payments to international organisations to prevent the deficit rising in 2012–13^[_] – there has been a genuine effort to meet the targets.

As a result, fiscal policy has tended to help stabilise the economy (that is, cyclically adjusted deficits were larger when the economy was weaker) since fiscal rules were first introduced in 1997, whereas in the previous 20 years it tended to exacerbate booms and busts.^[_] Moreover, fiscal rules have on the whole only been abandoned in the face of economic shocks where they would have required large pro-cyclical changes to fiscal policy – that is, massive tax rises or spending cuts in response to an economic downturn (e.g., in 2008 or following the Brexit referendum) – which would clearly not have been a sensible way to stabilise the economy. The U-turn on tax credit cuts which caused the benefit cap to be breached and the decision to increase spending in the 2019 spending round are the only exceptions where it can arguably be said that rules were breached as a result of a discretionary loosening of fiscal policy. Even then, there are mitigating factors: The benefit cap was set based on forecast spending, so it was unsurprising that it was breached when benefit cuts were reversed, and the 2019 spending round seemed to be (just) consistent with the rules at the time,^[_] even if it seemed likely that this would change as forecasts were revised in line with weakening in the economy. So even in this case, fiscal rules seemed to be constraining the government’s hand to some extent.  

Nonetheless, the fact that fiscal rules have been broken during challenging economic times is worrying since that is when they are most needed. It is most important to maintain confidence among voters and market participants that public debt is sustainable when debt is increasing rapidly during a downturn. Rules that have to be rewritten in periods of unpredictability are failing in their central purpose.

Therefore, rather than concluding that fiscal rules are not a useful invention, the challenge should be to devise different rules that allow fiscal policy to respond flexibly to shocks while holding the government to account for keeping the public finances sustainable. New rules should also take the changing post-Covid-19 environment into account. As we have argued elsewhere,^[_] the big increase in the debt-to-GDP ratio does not pose an immediate risk to the sustainability of the public finances, at least in part because interest rates are so low that the burden of these higher debt levels in terms of interest payments is very small compared to almost any time in the past. In the next section, we consider how existing proposals might fare in the face of these challenges. 

The government’s current fiscal rules have been found wanting during the pandemic. To meet the requirements of reaching a current budget balance within three years would require a substantial fiscal consolidation to be announced in the budget to take place by 2023–24: The Office for Budget Responsibility (OBR) estimated in November that the government was on track to miss this target by £29 billion, so a tightening of this magnitude would be necessary to meet the target.^[_] To introduce such a severe fiscal tightening in just a year or two when interest rates are expected to remain close to zero would not be a sensible response: Monetary policy would not be able to offset the tightening of fiscal policy.

But the government now has more headroom against its limit on debt-interest spending, which is now expected to account for 2.7 per cent of revenues in 2020–21, significantly lower than the pre-pandemic forecast of 3.3 per cent and well below the government’s 6 per cent limit: Ultra-low interest rates have reduced debt-servicing costs more than higher debt levels have increased it. This in turn suggests that there is scope for increasing investment spending beyond the government’s self-imposed limit of 3 per cent of GDP (which it is still just on target to hit) – but under the current framework it is not permitted to respond to low interest rates in this way.

What about rules that were suggested by other commentators prior to the pandemic? Jonathan Portes and Simon Wren-Lewis suggested targeting the overall budget deficit^[_] at the end of a five-year rolling horizon with a “knock out” that would mean that the rules did not apply when interest rates were at their zero lower bound.^[_] This rule would certainly allow for a flexible response to the pandemic, but might run into problems later on. Without any restrictions on the debt (as opposed to the deficit), governments could keep deferring tightening measures without worrying about breaching the rule. Portes and Wren-Lewis argue that the OBR should be given the responsibility of calling out this behaviour if it arises. But this does not seem an ideal solution: The government might disagree with the OBR and as an unelected, technocratic body the OBR might be reluctant to make what would undoubtedly be a controversial judgement that could not be definitively proved with hard data. Worse, the government might be tempted to choose members of the Budget Responsibility Committee who they felt would not make this call. Having more concrete rules to prevent debt rising in good times would also seem to be desirable.

The Resolution Foundation (RF) built on these rules with their own proposal in 2019.^[_] Although it also has a knock out and a deficit target, both are slightly different. The knock out would apply when the economy is operating at least 1 per cent below capacity and bank rate is below 1.5 per cent, rather than when there was a chance of hitting the zero lower bound. The deficit target would be to aim for a cyclically adjusted current budget balance at a fixed point five years from when the rules were announced (though this would be reset if the knock out were activated), with a margin of error of 1 per cent of GDP either way. They suggest that chancellors aim for the upper end of this range in each budget to ensure they do not end up missing their target through forecasting errors. RF also adds two further rules: The government must increase public-sector net worth (that is, total government assets, including non-financial assets, less total liabilities) over a five-year parliament, and keep debt-interest payments below 10 per cent of tax revenues. The last of these conditions would remain even if the knock-out clause were triggered. The aim of these two conditions is to incentivise borrowing for productive investment – the value of assets purchased would (hopefully more than) offset the additional debt needed to pay for it in the net-worth calculation – but not to the extent that a rise in interest rates on government debt would threaten the sustainability of the public finances.

A little over a year and nearly one pandemic on from publication, the RF proposals still seem broadly sensible. Starting whatever fiscal consolidation was necessary only when the economy is operating close to its potential level and then reaching a cyclically adjusted current budget balance five years after that would allow fiscal policy to give the economy the support it will need in the years ahead. Similarly, a focus on net worth rather than net debt would allow the government to take advantage of low interest rates to make worthwhile investments. But the debt-interest payment ceiling now seems totally irrelevant: Debt-interest payments are set to fall to 2.2 per cent of tax revenues by 2024–25, even after the big increase in debt levels seen during the pandemic, way below the 10 per cent ceiling. There would therefore be essentially no limit to the level of borrowing to pay for investment, at least in the short term. This could lead to difficulties if interest rates started to rise. If debt levels were allowed to increase a lot more – which would not be inconsistent with these fiscal rules – debt-interest payments would become very sensitive to interest rate rises. Previous TBI analysis has shown that if the government were to continue to run budget deficits that were high by historical standards in the 2020s and then interest rates were to start to rise to reach historically normal levels by 2040, interest payments would rise rapidly during the 2030s and it would be difficult to prevent the 10 per cent ceiling being breached.^[_] Another mechanism to limit investment spending, linked to prevailing measures of interest rates, would seem to be desirable.

All of this highlights one of the central lessons from past fiscal rules: that contemporaneous ceilings on debt or debt-interest payments are too rigid to be useful. Either they are set too high, making them irrelevant, or too low, making them not credible in the wake of inevitable economic shocks that occur from time to time.

Low interest rates across the world, particularly since the start of the pandemic, have sparked proposals in other advanced economies too. Rather than seeking to make fiscal rules more state-dependent and thus more complicated, others have recommended a shift to “guidelines”. In the US, Jason Furman and Lawrence Summers recently advocated a shift of fiscal policy to focusing on economic growth and financial stability while ensuring that debt-interest payments do not either rise rapidly or rise above 2 per cent of GDP. They propose the following guidelines:

• There should be emergency spending financed by borrowing in response to economic downturns.

• Most long-term expenditure programmes should be paid for by taxes, other than those that can plausibly be thought to pay for themselves.

• The composition of government spending should be improved to make it more supportive of demand and more efficient.

Olivier Blanchard, Alvaro Leandro and Jeromin Zettelmeyer of the Peterson Institute for International Economics have developed a proposal to replace fiscal rules in the Eurozone.^[_] The “standards” they propose are that member states should ensure that there is a high probability that debt will remain stable, and that those that had a material risk of debt not becoming sustainable would have to reduce their deficits. Stochastic modelling would be used to estimate the probability that debt would reach an unsustainable level.

With fiscal-rule proposals growing more and more complicated to allow more optimal responses to prevailing economic conditions, moving to a simpler system of standards sounds appealing. In countries with different institutional setups – such as the US, where Congress plays a much larger role in designing the budget, or the Eurozone, where the European Commission has more power to hold Member States to these standards – these ideas may be more appropriate than a more precise set of fiscal rules.

But it is hard to see how these would work in the UK, where the executive is much more powerful. Presumably it would be for the OBR to decide whether these standards were being adhered to. But this is ultimately a matter of judgement on which the government and the OBR could (most of the time at least) reasonably disagree: The economic payoffs of particular programmes, whether spending is becoming more or less efficient, and whether the risk of public debt becoming unsustainable is 10 per cent or 1 per cent are all questions on which there is far from a settled consensus. It is therefore unlikely to place much of a constraint on government actions in practice and may even lead to the process of appointing members of the Budget Responsibility Committee becoming politicised.

A looser set of guidelines or standards therefore seems to fall short of a key function that rules have sought to play in making government clearly accountable to voters and market participants for keeping public debt sustainable.

Is there then a way of refining existing fiscal-rule proposals to take account of the latest developments? Can we design an approach that reflects the need for a bigger macroeconomic stabilisation role for fiscal policy, but without either being so vague as to be unenforceable or so stringent as to be unhelpful? TBI has proposed an approach that we think achieves this balance and we argue that the “all-weather” fiscal framework we proposed last February does still live up to its name.^[_] The framework has four elements.

1.  A Long-term Government Debt Objective and Implied Deficit Ceiling

As a first step, the government would specify a long-term objective or “anchor” for the debt-to-GDP ratio. Unlike previous rules, which have set ceilings on debt a handful of years ahead which lack credibility, such a long-term target would represent what the debt ratio would be expected to converge towards in the very long term – say, several decades.

The OBR would then convert the long-term debt objective into an implied underlying structural deficit limit based on its forecast of long-run nominal GDP growth, i.e., the sustainable real GDP growth rate (the growth of potential output) plus inflation.^[_] This gives the baseline deficit limit. This would be the highest deficit that would be consistent with the target debt-to-GDP ratio over the long term. Naturally, a chancellor could opt for a lower deficit if they wished to speed the otherwise gradual adjustment towards their debt objective.

In being explicit about their debt objective, the chancellor should set out the considerations that supported this decision. For example, if the chancellor believed it was reasonable for debt to remain at its current levels of around 100 per cent of GDP even if interest rates were back at more historically normal levels, they would have to explain why this was justified. By contrast, if the chancellor wished to pursue a debt-reduction strategy, they should be explicit about what a more optimal level of debt is and why. For credibility, this debt objective should ideally remain fixed for some time, although a new government could reasonably adopt a different objective, provided it spelled out its reasoning.

This approach would strengthen democratic accountability since the chancellor would have to justify a simple level of debt to which he or she aspires, while allowing the technocrats to translate this into an implied maximum annual borrowing limit.

2. A Real-Time Affordability Test

The deficit limit would then be adjusted to reflect the affordability of additional borrowing. The adjustment would be calculated based on the difference between the real interest rate on government debt and the long-run growth rate, such that the government’s scope to borrow is expanded when borrowing is cheaper and reduced when it is more burdensome.^[_]

This adjustment reflects the fact that, irrespective of the government’s longer-term objective for debt, there are times when additional debt is much more sustainable. For instance, since borrowing costs are currently low it would make sense to take the opportunity to increase investment spending or improve key public services. By contrast, if borrowing costs were higher, the government ought to be more cautious about taking on more debt.

An implication of this adjustment is that the debt-to-GDP ratio – and hence the government’s deficit limit – would be allowed to remain above the longer-term objective if borrowing costs stayed low for a long period. Indeed, if interest rates were to remain as low as they are now in perpetuity, the adjusted deficit limit would be consistent with debt remaining at 100 per cent of GDP even if the long-run debt anchor was 75 per cent (see Figure 1). But if interest rates started to rise from 2025 to reach historically normal levels by 2040, deficits would have to be reduced and debt would return to a glide path towards the long-run target. Similarly, if we went through a long period of interest rates being higher than the long-run growth rate, the rule would reduce the deficit limit and, if this persisted for long enough, debt would move below the long-term objective level.

Figure 1

Adjusted deficit limit forces debt back to anchor if interest rates rise

Source: OBR, TBI calculations.

An advantage to this method over other proposals is that it links borrowing headroom explicitly to the affordability of new debt. It would put the brakes on deficit bias much more effectively than a debt-interest rule, as it would bite much more quickly if interest rates increased, since it focuses on the marginal affordability of new debt rather than the cost of the existing stock of debt, which only changes slowly.

By linking the deficit limit to the prevailing, rather than average, government-bond yield there is a greater chance that reckless spending which caused a spike in bond yields would be much more quickly reflected in this test than in one based on the average interest rate on government debt. At the same time, the adjustment of spending plans to higher interest rates would be smoother under this proposal than under a debt-interest rule, which might require rapid consolidation if the limit looked set to be breached.

A further advantage of adjusting the deficit limit in this way is that it provides government with an incentive to implement growth-friendly policies. By taking account of the ability to repay via potential output and increasing the deficit limit if the sustainable growth rate rises, this rule rewards policies that raise growth rates and penalises those that do the opposite.

3. An Escape Clause

Similar to Portes and Wren-Lewis’ knock out, our proposal contains an escape clause to enable counter-cyclical fiscal policy by suspending the deficit limit at times when output was at least 1 per cent of GDP below estimates of its potential. This would enable active counter-cyclical policy to operate at times when the economy was weak. At present, it would allow fiscal policy to support the economy during the pandemic until the point where output had almost returned to its potential level.

4. A Net Worth Goal

Like the RF’s proposal, under our framework the government would also be assessed against its progress towards increasing public-sector net worth over five years. Conceptually, a net-worth objective is appealing as it is more closely related to sustainability than the net-debt concept that is typically used. Moreover, it does not create perverse incentives to incur liabilities that are not included in the narrower debt definition. A net-worth constraint would complement the need for the government to spell out its rationale for the debt objective by limiting the degree to which short-termist governments could cut investment in favour of current spending. But the addition of an overall deficit limit also places a limit on investment spending.

We believe that the government should seriously consider introducing fiscal rules along these lines in its next budget. In the 2020 budget, the Treasury set out the considerations it would take into account in reviewing them; these are summarised in Figure 2, where we also consider how well the various rules and standards we have discussed address the criteria the government has set out (green = well, yellow = partially, red = badly or not at all).

The framework outlined here stands up well against all of these criteria as it allows more flexibility for fiscal policy to stabilise the economy in bad times, but still keeps the public finances on a sustainable path. As a result it more closely meets the various objectives we have for fiscal policy in the low interest rate world. Other fiscal frameworks would also represent a big step forward but do not manage to do all of these to the same degree: Existing rules would require premature fiscal tightening, and other proposals do not offer such reliable protection of fiscal sustainability. A shift to guidelines rather than rules appears attractive, but implementing them would be complicated in the UK context. Forcing an independent body like the OBR to make controversial judgements as to whether standards had been met runs the risk of weakening its approval by the public if the government took a different view. Moreover, completely departing from a rules-based framework in favour of much more discretion seems particularly unwise in polarised political times.

Figure 2

Assessing fiscal rule proposals against the government's criteria

After an examination of the details of different proposals for fiscal rules, it is worth reminding ourselves what the aim of fiscal rules is in the first place; namely, to hold governments accountable for achieving sustainable public finances and intergenerational fairness. TBI’s proposed all-weather fiscal framework aims to achieve fiscal sustainability by limiting deficits except when the economy is operating significantly below potential, while also allowing the government to boost public investment to take advantage of interest rates being unusually low, with the aim of reaching a set level of debt in the long run. How successful it would be at achieving this objective depends on how often the escape clause is invoked and the extent of fiscal stimulus that is required on each occasion. It may turn out that the deficit limit during good times, which only moves debt towards its target level very slowly, is insufficient to prevent debt gradually creeping up over the course of decades if there are frequent crises that require fiscal policy to support the economy. Other rules that include an escape clause or knock out – a necessary provision given that the zero lower bound on interest rates seems likely to constrain the ability of monetary policy to respond to economic downturns for the foreseeable future – do not deal with this issue either.

The same concerns apply to intergenerational fairness: Repeated crises that lead to debt increasing over time could lead to future generations being overburdened with debt-interest payments in spite of the commitment to increasing public-sector net worth during good times. A tighter deficit limit may ultimately be required. Furthermore, there are other ways that current policy could unfairly burden future generations that are not prevented by existing fiscal rules. For example, promising more generous state pensions to current workers creates an obligation that future generations might have to fulfil that is not included as a liability in the calculation of public-sector net worth. An interesting possible addition to the rules might be a “sustainable commitments rule”, which ensures that pre-committed spending on debt interest, state and public-sector pensions, and PFI contracts was not forecast to increase too rapidly.^[_]

One turbulent year on, our proposed fiscal framework still seems like the right approach to balancing the competing objectives of fiscal policy in the 2020s.  Its policy implications – that fiscal tightening should wait until the economy has recovered, then slow adjustment to a long-run target debt level through a limit on deficits in normal times and taking advantage of low interest rates to invest more when appropriate – are a sound prescription for the years ahead. The flexibility it gives to make appropriate adjustments to the fiscal stance in light of developments in the economy would hopefully reduce the need for further tweaks to the rules in the future. This would be a welcome development after a decade or more of instability, perhaps finally ending the cycle of boom and bust in fiscal policy thinking.