April 27, 2020

Macro series – The burden of Sisyphus

BY John Llewellyn, Russell Jones

Share on twitter
Share on whatsapp
Share on facebook
Share on linkedin
Share on email
Share on reddit

Report Contents

Listen to our reports with a personalized podcasts through your Amazon Alexa or Apple devices audio translated into several languages

en flag
zh flag
pt flag
es flag
Press play to listen
( 17 mins)


  • The pandemic and the reaction to it will result in a surge in public sector liabilities.
  • Gross G-7 debt is projected to rise to an all-time high of 140% of GDP or more.
  • Individual countries have in the past coped with much higher debt burdens.
  • Today’s indebtedness is sustainable provided nominal growth exceeds interest costs.
  • Nevertheless, some countries will probably tighten fiscal policy too much, too soon.

Shock therapy

Fiscal policy has been relaxed dramatically …

That in current circumstances a rapid loosening of fiscal policy – indeed a wholesale setting-aside of fiscal rules – was both necessary and appropriate is self-evident. Even in the most fiscally conservative countries it is accepted that, in the event of a large adverse exogenous shock to private sector demand, the public sector should step in and act as a stabilising force. And the principle was particularly appropriate going into the present crisis, given that the leeway of conventional monetary policy in the major economies had already largely been exhausted. There was no alternative but to increase health spending, and attempt to put a floor under the income of households, individual firms, and whole industrial sectors.

What the true extent of fiscal policy support will ultimately evolve into is unknowable, and nor is it at all clear what it will mean: it is impossible at the moment to separate, for any country, what will turn out to have been a direct claim on GDP (such as wage payments); a potential claim on GDP (such as guarantees); or the purchase of an asset that can later be sold back to the private sector (such as various forms of conservatorship).

That said, on the basis of a simple adding up of headline programmes, the size of the fiscal response has already been immense. According to tentative estimates by the IMF in its latest Fiscal Monitor, the G-20 economies taken as a whole have announced direct fiscal support measures equivalent to some 3.5% of GDP. For the G-7 economies, the figure is around 6%, together with loan and guarantee schemes that regularly amount to a further 10% or more of GDP.1

… inflating deficits and debt burdens unprecedently

Such interventions are unprecedented in modern times. Certainly they are in excess of those made during the global financial crisis (GFC). They are likely to leave the average G-20 and G-7 broadly- defined budgetary shortfall in the region of 12% of GDP this year; and average G-20 and G7 primary balances – arguably the best proxy of a government’s immediate influence on the economy – are also forecast to swell to a double digit proportion of GDP.2 In the US, the estimated figure could extend to more than 15% of GDP. In the case of the UK, the fiscal expansion has already been the largest and fastest in peacetime, and it is likely to be supplemented further in due course.3

Macro series - The burden of Sisyphus 1

Dead weights

The gross public debt burdens, which had been in the region of 85% of GDP for the G-20, and 90% of GDP for the G-7 at the onset of the GFC, and have been running some 20 ppts of GDP higher over recent years, are now expected to jump towards 140% of GDP over the coming two years. Again, the IMF’s figures are perforce only tentative, and many other forecasters expect both the economic downturn, and the subsequent impact on the public finances, to be considerably more dramatic and enduring.4

G-7 debt is now higher than at the end of WWII

Whatever the final reckoning, it is already evident that debt ratios are historically extremely elevated – and not just in a peacetime context. The prospective average gross debt burden for the advanced economies taken together will comfortably exceed the World War II peak of around 120% of GDP (although there are many individual historical country cases of deficits and debt ratios having been higher, indeed significantly higher, than is currently projected).5

Evaluating the consequences

High public debt is believed to be economically costly …

High public debt is often presented as economically, if not morally, destructive. Beyond the philosophical arguments linking debt with notions of sin, a number of economic arguments are advanced as to why excessive government liabilities are damaging. Higher public debt is likely, inter alia, to:

  • Complicate countercyclical policy and, to the extent that it does so, lead through hysteresis effects to slower growth of economic potential.
  • Engender higher future distortionary taxes, which will generally weigh on economic performance, not least through constraining disposable incomes and thereby investment and productivity.
  • ‘Crowd out’ private capital accumulation, also lowering future potential output.
  • Increase the potential for solvency crises. Indeed, there is already the suggestion that some European countries have soft-peddled on budgetary support because of concerns about affordability.6

… but much depends on growth and rates

Debt dynamics

The strength of these arguments has long been hotly debated, and depends to a significant degree on two considerations. As illustrated in Box 1, public sector debt dynamics are governed essentially by: the primary balance and the so-called ‘snowball’. The former largely speaks for itself: it is self- evident that large, sustained, primary deficits quickly swell public debt. Less discussed, and less understood, is the all-important ‘snowball’.7

Macro series - The burden of Sisyphus 2

Box 1: Debt dynamics

The following equation describes how debt as a proportion of GDP evolves over time: Dt/Yt = (1+r) Dt-1/(1+g) Yt-1 + bt


Dt is government debt at time t; Yt is GDP at time t;

r is the nominal interest rate;

g is the nominal growth rate of GDP (compromised of the real growth rate and inflation); bt is the primary fiscal balance as a share of nominal GDP at time t.

If r is greater than g, the government will need to run a primary fiscal surplus to prevent the debt to GDP ratio from rising. The larger is r relative to g, and the higher the debt to GDP ratio, the larger the primary surplus will need to be.


The ‘snowball’ has two sub-components: first, an expression that contains the difference between the implied interest payable on the outstanding stock of debt and the rate of growth of nominal GDP; and second, the starting level of public debt as a proportion of GDP, which acts as a ‘scalar’. If the ‘snowball’ is unfavourable, primary surpluses are required to prevent debt from increasing. The more unfavourable the snowball, and the higher the initial debt burden, the larger the primary surplus required. On the other hand, a favourable snowball means that primary deficits can be consistent with stable or indeed falling debt burdens.

There is no single threshold for debt sustainability

This changing interaction between growth and interest rates is one reason why it is so difficult to know the precise costs and benefits of different debt levels, or indeed to determine the maximum debt that a country can sustain.8 Historically, formal defaults have been rare in large economies that borrow in their own currencies and print their own money. This is especially the case outside major wars and their aftermaths.

What is also true is that such defaults as have occurred, both in the larger economies and beyond, have happened at varied levels of debt. There is apparently no strict maximum level, or even maximum range, of debt that can truly be said to be ‘sustainable’.

Crises come when growth is slow and rates are high

However, what is evident is that fiscal crises tend to be engendered by a combination of high interest rates and low nominal GDP growth, especially when the debt ratio was already high. In contrast, when interest rates are low, and nominal GDP growth is reasonably rapid – either because of strong real growth, high inflation, or both – then even initially high levels of public debt can be brought down quickly, and are of much less consequence.

Latterly, because of a range of overlapping considerations, extending from low investment demand to higher savings rates and widening inequality, interest rates have been historically low in both nominal and real terms, and hence government borrowing costs have tended to fall short of both actual and expected growth, even as the latter have also tended to fall short of historical norms.

But rates today are historically low …

More than one-fifth of all government bonds traded with negative yields at the end of 2019, and the proportion has risen over the past four months. For most advanced economies, debt service costs are low relative to GDP. The average figure is currently significantly below 2%, as against more than 3% 25 years ago. It is moreover, difficult to argue that the cost of capital is holding back private investment.

… and growth will gradually return

With inflation expectations subdued, and central banks acting through their asset purchase and other unconventional monetary policies to depress the entire term structure of interest rates, there is good reason to believe that the snowball will remain favourable when growth returns, and debt ratios, even if they temporarily surge, can be brought to heel and encouraged to decline once the pandemic passes. It is also worth noting that a tendency for nominal GDP growth rates to exceed interest rates is not exceptional. Historically, this condition has held for long periods in the advanced economies.

Rules of thumb


High and rising debt cannot be ignored …

All this is not to suggest that governments can, or should, try to ignore fiscal constraints entirely. Debt cannot be allowed to grow ad infinitum, nor can governments set fiscal policy without any limiting principles as to what is, and what is not, desirable or possible. Higher debt does have potential downsides, and the optimal modus operandi is to keep debt relatively low over time.

However, this is different from saying that the major economies would be well advised to act rapidly and seek to reverse the accumulation of debt that will eventuate from the COVID-19 pandemic to the exclusion of other policy goals such as improving infrastructure networks, addressing climate change, and reducing at least the more socially-destructive inequalities. The risks associated with historically-high debt burdens in the years ahead could well prove less damaging than would aggressive, blanket, post-crisis fiscal consolidation.

… but it can be managed

In looking beyond the immediate crisis, our judgement is that government debt burdens can be managed, provided that nominal GDP growth is sustained, and in particular at a level in excess of interest costs. In a perfect world, this nominal growth should be skewed towards real growth, underlining the importance of good structural policies, but a little inflation can grease the wheels of debt adjustment. Deflation, especially in the context of sluggish real growth, on the other hand, makes the process of debt management much harder, and central banks would be warranted in doing all that they can to avoid it.

Excessive austerity can be self-defeating

Seeking to slash budget deficits and reduce debt burdens when an economy is still in recession, or even when growth is weak, would likely prove not only challenging, but also self-defeating, leaving the burden of debt even higher than before. It would also risk encouraging social distress and political turbulence.

Hard and inflexible balanced budget rules can prove absurdly pro-cyclical when the fiscal stance is so endogenous to the business cycle. The monetary stance and, the value of the exchange rate in the case of individual countries, can provide a powerful offset to the deflationary effects of fiscal restraint. Monetary policy, not least its unconventional variants, can also play an important role in reducing debt service costs, and keeping them at very low levels.

Some will resort monetisation

Some governments and central banks may decide to monetise a significant part of the increase in public debt as a way of both stimulating their economies and reducing the ultimate obligation of the public debt on taxpayers.

Box 2: The UK’s experience with an intermittently high burden of public debt

After the Napoleonic Wars, the debt ratio was at an all-time high, some 260% of GDP. But over the course of the next 98 years, it declined to negligible proportions (25% of GDP). This was achieved by running persistent, and for a period high (5-6% of GDP), primary budget surpluses at a time when real GDP growth, although persistently faster than in previous epochs, tended to be lower than real interest rates, apart from during the relatively fast- growing 1860s. During this period, the price level, although volatile from year to year, actually fell.

During this near century-long fiscal adjustment, the nominal debt stock fell by around 20%. However, World War 1 saw it rise twelve-fold, and it took a doubling of the price level during the hostilities to constrain the increase in the debt ratio to 127% of GDP. In the early 1920s, the governments of the day subjected the economy to a punishing programme of public spending cuts in a just about successful effort to balance the budget, bring taxation levels down to pre-war levels, and drive wartime inflation out of the system. This was also a time of tight monetary policy and high nominal and real interest rates. The associated deep recession, large rise in unemployment, and sharp fall in the price level actually resulted in a further rise in the debt ratio, equivalent to almost fifty percentage points of GDP. Moreover, this era saw mounting industrial unrest as nominal wages were cut, culminating in the General Strike of 1926.

The mid-late 1920s witnessed something of a recovery, however, and the debt ratio began to decline, although it failed to return to its post-WW1 level. In 1929, the Great Depression began. Despite the deterioration in the economic environment, the government responded with successive rounds of expenditure cuts aimed at keeping the budget in balance, which it again just about managed to do, apart from in 1932. But as output and prices tumbled, and unemployment soared, as had been the case a decade earlier, fiscal deflation failed to stabilise the debt ratio. Indeed, the burden of debt once again escalated, reaching almost 180% of GDP in 1933.

Despite two rounds of budgetary austerity, Britain’s debt ratio was substantially higher than it had been at the end of WW1.

It was only in the mid-1930s that the burden of debt began to fall again. Balanced budgets remained the order of the day, but GDP growth, both real and nominal, responded to the monetary policy flexibility afforded by Britain’s exit from the gold standard, and a sharp depreciation of sterling. But by the outbreak of WWII, Britain’s burden of liabilities had still not returned to its 1919 level.9

Not surprisingly, WWII saw another huge increase in the amount of outstanding government debt both in absolute terms and relative to an inflated level of GDP. The immediate post WWII debt ratio, at around 250% of GDP, was more than twice its 1919 level, and significantly higher than in any other economy that did not formally default, suffer a bout of hyperinflation, or undergo a currency conversion.

However, a combination of sustained expansion of real activity during the post-war boom, persistent, if hardly traumatic, inflation (averaging around 4% during the period), and of financial substantial repression had cut this figure by more than two-thirds by 1970.10 Even though this epoch saw the dramatic expansion of the Welfare State, a doubling of the proportions of both public spending and taxation in GDP, a greater willingness to budget for a deficit if output fell short of potential, and a rise in absolute level of debt of more than 50%, the debt ratio dropped back to a level last seen in WW1 (some 70 % of GDP).

Since 1970, the UK’s public sector debt ratio has ebbed and flowed from decade to decade. It initially declined, dropping to less than 40% of GDP at one stage in the 1990s, although in the wake of the Global Financial Crisis it rose sharply, stabilising only in the late-2010s, just above 85% of GDP. Protracted efforts to narrow the budget deficit through public spending restraint provided at best mixed results as, notwithstanding historically low borrowing costs, the austerity contributed to a sluggish pace of economic recovery, and stirred social and political discontent. Over this 50-year period, inflation made an especially large contribution to nominal GDP growth following the two 1970s oil shocks, but it became a much less important consideration from the 1990s, with the arrival of formal inflation targeting and operational Bank of England independence.

In historical terms, therefore, Britain’s prospective debt ratio, although much higher than it was fifteen years ago, does not look especially abnormal, or high. With appropriate policies once growth is solidly underway again, it can almost certainly be brought down.

Better than the alternative

Governments have responded appropriately

No doubt many people will believe that the fiscal response even to date has been excessive; but it is necessary to take account of the counterfactual, which would almost certainly have been even more ghastly, especially as it would have led to more long-term scarring and damage to future tax raising capacity.

That said, it is clear that policymakers will in future have to walk a narrow path between excessive indebtedness and insufficient policy stimulus. History however does suggest that, with a little good luck and a following wind, a satisfactory trade-off can be achieved.

Watch fors

  • Yet more fiscal stimulus; but ideally to not too dissimilar extent across countries.
  • Deficits and debt ratios overshooting the IMF’s projections; but ideally not by a great deal.
  • Partial direct monetisation in some economies; but hopefully within pre-defined rules.
  • Resort to financial repression.
  • Growing pressure to reverse fiscal policy course in the years ahead.
  • Permanently higher levels of public spending and tax revenues in GDP.


1 IMF. 2020. Fiscal Monitor. April.

2 IMF. 2020. ibid. Primary balances are net of interest rate costs.

3 There is good reason to believe, as the IMF has suggested, that the effectiveness of fiscal stimulus will be greater once the recovery starts.

4 IMF. 2020. ibid.

5 IMF. 2013. A modern history of fiscal prudence and profligacy. January.

6 It has been argued, for example, that Spain and Italy fall into this category.

7 In reality, a third term, the stock-flow adjustment, ensures consistency between net borrowing (a flow) and the variation in the stock of gross debt. This ‘catch all’ term includes, inter alia, realised losses/gains from interventions in the banking sector, and can lead to large one-of increases in the level of public debt, but in and of itself has less dynamic relevance than primary deficits or the ‘snowball’, and so has not been explored here.

8 Some studies have found – or at least have been interpreted as finding – that beyond a certain threshold – estimates range between 67 and 95% of GDP, higher public debt lowers potential growth. By contrast, others have found that such thresholds are non-existent, or highly country specific.

9 This period also saw perhaps the closest that Britain has come to a formal default in the modern era. In 1932, the government instituted a ‘voluntary’ debt restructuring of £2.1bn (more than a quarter of the debt stock) of the 5% 1917 War Loan into 3.5% perpetual bonds (callable from 1952) on terms that were unambiguously unfavourable to bond holders. Ultimately, more than 75% of holders accepted the restructuring (92% of the outstanding value) thanks largely to the government’s appeal to patriotism in the face of economic adversity, together with some heavy handed ‘moral suasion’.

10 Financial Repression is perhaps best defined as the creation of a captive market for government debt via regulatory controls on the ability of the private sector to invest in alternative assets.

More by John Llewellyn, Russell Jones