Debt Dynamics

As governments across the world are taking on huge quantities of debt in an effort to support their economies, two questions are repeatedly being asked:  is not all this borrowing going to be inflationary, and how will it be afforded and paid back?

Behind these questions are two widely believed economic “facts”.  The first is that excessive borrowing, and in particular excessive unfunded borrowing, is always inflationary and the more the government turns to the central bank for direct funding (technically known as monetary financing but more colloquially known as printing money), the more a country risks hyperinflation as in Weimar Germany or, more recently, Venezuela or Zimbabwe.  And the second is that when debt as a percentage of GDP reaches a certain level, it risks exploding and causing national insolvency.

We will attempt to show that the risks of inflation from the current much increased fiscal deficits in so many countries will not automatically lead to inflation, and that most developed economies still have considerable borrowing headroom before they risk a runaway debt “snowball”.  Finally we will urge governments, especially the UK government, to not be too urgent and aggressive in seeking to pay the debt back.

Will the borrowing be inflationary?

There is a broad consensus among economists that in the short term, say for the next 3-6 months, the risk of inflation is very limited.  The collapse in consumer demand across so many countries, especially demand in “discretionary expenditure” sectors such as leisure, travel and household goods, leaves a surplus of supply, and the usual expectation would be for this to lead to prices being reduced.  Indeed, early indications are that inflation rates are falling across most of the developed world and many countries may even move into outright deflation. 

How far could this deflation go?  Many economists are comparing the current outlook, and the likely falls in GDP, to the 1930s Great Depression.  Then, central banks were not so active and price levels fell heavily, by more than 10% in some countries.  Today’s central banks are more active, and have shown repeatedly over the last 10 years that they are determined to avoid the extra shock that such deflation would imply.  And it is quite possible that their current direct monetary financing of government will be effective in reducing the fall in prices in the next 3-6 months from say -10% to -2%.

While it is true that this can be interpreted literally as “inflation is 8 percentage points higher than it would have been absent the central bank financing”, it would be a beneficial result and only the pedantic linguists would call the central bank actions “inflationary”.

Looking towards and beyond the end of 2020, though, the consensus evaporates.  Some economists expect to see a combination of reductions in supply, as companies and suppliers go bankrupt, and increased costs, as countries retreat from full globalisation and place more emphasis on the resilience of their supply chains rather than cost efficiencies.  These two combined, they argue, will lead firstly to higher prices to reflect the onshoring, and then quite quickly to demand exceeding supply;  while this should eventually tempt new players into markets, it will in the meantime add to the price level recovery. 

This argument relies quite heavily on a fairly rapid bounce-back in the world economy (the so called “V-shaped recovery”), and on people seeking to satisfy their  suspended demand – for holidays, for white goods etc – as soon as they can.  It also supposes that government support for the economy, once in place, might prove quite difficult to withdraw – temporary measures have a habit of becoming entrenched and beneficiaries never like to lose their benefits – with the result that it could further add to the inflationary mix if they are kept in place longer than is strictly necessary.

Other economists consider a V-shaped recovery increasingly unlikely.  Consumers are losing purchasing power at an alarmingly fast rate – unemployment, both formal and informal, is ballooning in most countries – and the exit from lockdown looks much more likely to be a gradual process, perhaps including backwards steps along the way as the pandemic threatens to restart, rather than a sudden cathartic “Now we are free” moment when animal spirits will return.  And there is also evidence already that those that do still have incomes are looking to save more, both as a precaution against possible misfortune in a more uncertain future and to offset expected higher taxes.

We tend to think this is the more likely scenario – a slow and uncertain recovery, with discretionary spending still kept on quite a tight rein even by those who still have jobs.  And if the multiple rounds of Quantitative Easing since 2010 (different in style only from monetary financing, not substance) did not produce inflation in the last decade, when the world’s GDP was growing, albeit slowly, we do not think the current level of government borrowing and central bank funding will be inflationary against a backdrop of what is likely to be a prolonged period of depressed demand.

How will the borrowing be afforded and paid back?

This is two questions in one.  Firstly, is the debt affordable?  Many people focus on just one statistic, the ratio of government debt to GDP, a focus which was given extra impetus by an influential paper issued some 10 years ago by Carmen Reinhart and Kenneth Rogoff, in which they argued that national debt dynamics worsen notably once this ratio exceeded 90%.

This figure has the merit of simplicity (and because of this was very widely quoted), but is we think too narrow a focus.  For a country which retains creditworthiness (that is, people do not seriously expect it to default on its debt), the relevant figure is not the absolute level of debt but the debt servicing ratio:  how much the debt costs each year.  And today’s extremely low interest rates mean that even very elevated levels of debt can be financed with acceptable and sustainable financial cost.

There are several countries where even before the coronacrisis the current level of debt as a percentage of GDP was well above the Reinhart and Rogoff 90% threshold – in Japan it is well over 200%, and has been for some time without obvious sustainability issues.  And lest current conditions be seen as exceptional, a look at the experience of the United Kingdom shows that at various times in the last 300 years, debt levels have been much higher than they are at current. 

Chart: UK Government debt as a percentage of GDP, 1700-2020
Source: HM Treasury and Llewellyn Consulting [1].

So, the absolute levels of government debt are neither exceptional, nor, we would argue, unaffordable in the short term given current interest rates.  But at some point, the government will want to – and will certainly be under pressure to – reduce the ratio.  What can we learn from looking at past episodes of debt reduction?

The first thing to note is that there is more than one way to reduce the debt-to-GDP ratio.  While politicians and commentators alike might instinctively reach for the toolkit marked “austerity” (lower expenditure, higher taxes, narrower deficits, maybe primary surpluses, etc), it is not the case that the only way to reduce the ratio is to reduce the numerator, the absolute level of the debt.   The ratio is just that, a ratio, and the other way to reduce it is to increase the denominator, nominal GDP.   And in turn this can be done in one of two ways – increasing nominal GDP by increasing real GDP (“grow one’s way out of debt”) or by increasing the price multiplier (“inflate one’s way out of debt”). 

Let us return to the UK’s experiences over the last 200 years, and examine 4 separate episodes of debt reduction.

In 1815, after the Napoleonic Wars, UK government debt was over 260% of GDP.  Over the course of the next 100-odd years, it was reduced to 30% or so of GDP.  The actual stock of debt itself did not change much – it fell in nominal terms by about 20% only – and the big change was the 7 or 8-fold increase in nominal GDP over the century of British Power.  Moreover this was all real growth;  the general price level in 1914 was about half that of 1815.  This was therefore an example of debt reduction by growing the denominator in real terms with considerable success, albeit over an extended period.

In 1918, after the First World War, UK government debt was back up to 130% of GDP.  The post war government introduced an aggressive austerity programme (called the “Geddes Axe”, after Sir Eric Geddes, chairman of the Committee on National Expenditure).  The combination of tight fiscal policy and tight monetary policy, with high real interest rates, was successful in the narrow sense of reducing the debt, but at the expense of a deep recession which firstly caused great social unrest and division in society (including the 1926 General Strike), and secondly undid all the debt reduction as the economy slowed – indeed debt as a percentage of GDP was higher at the end of the 1920s than at the start.  This was an example of debt reduction by reducing the numerator through austerity, with limited results and at considerable cost to social cohesion.

In 1945, after the Second World War, UK Government debt was well over 200% of GDP.  This time, mindful of the bitterness of the austerity 20 years before, the government used higher government expenditure to generate a recovery in GDP.  Inflation, mostly mild, also helped swell nominal GDP, and in the 50 years to 1995, the debt ratio was gradually brought down to about 40%.  This was an example of debt reduction by growing the denominator in nominal terms;  there was real growth, but more than half the contribution came from decades of inflation, usually low (especially in the first 25 years of the period) but persistent.

Lastly, in 2010, after the financial crash, UK Government Debt rose sharply and approached 80% of GDP.  In historic terms this was not an unduly high figure, as the 300-year chart shows, but the combination of the Reinhart and Rogoff paper (then just issued), plus the Euro-area’s fixation on 60% debt as the acceptable maximum (as set out in the Maastricht criteria) changed perceptions on what was tolerable for national debt and forced the Cameron-led coalition government into an aggressive austerity programme.  This was once again then debt reduction by reducing the numerator through austerity as in the 1920s, and as 90 years earlier, it proved both a restraint on GDP and (when combined with QE which inflated asset prices and so the wealth of asset owners) highly socially divisive.

We are well aware that we have here only sketched the outlines of the four most recent high debt periods in the UK’s history, and that circumstances for each were very different – for one thing, Britain will never again be the world’s economic superpower as it was for 100 years after 1815.  But we would draw three conclusions from even this very cursory assessment:

1.  The level of government debt as a percentage of GDP that the UK will face as a result of the current support for the economy is neither particularly high in historic terms nor, given current very low interest rates, unaffordable.  The government should avoid being pressured into over-hasty measures to rein it back.

2.   When the UK government does turn to reducing the debt levels, it should not concentrate exclusively on reducing the numerator of the debt to GDP ratio.  Measures to increase the denominator, nominal GDP, should also be actively pursued, and may prove less damaging to society and social cohesion.

3.  The best outcome would be for nominal GDP to grow through real GDP growth (as in the 19th century).  But growing nominal GDP through long term mild inflation (as in the period after 1945) should not be ruled out of hand completely.

The debt being incurred all over the world will one day have to be addressed. But it is to be hoped that governments, especially the UK government, do not just once again reach for the austerity axe.

[1]              This chart, and the discussion that follows, draw on a paper by Llewellyn Consulting entitled “The Burden of Sisyphus”.  I am grateful to them for permission to do so;  for more information and the full paper contact them at