Summary
With over €674Bn of debt maturing in 2022 and 2023 and secular rises in interest rates, Italy has been in the news. The ECB is keen to tame inflation, with annualised HICP rates rising rapidly to over 8%.
But the requirement for Italy (and other European countries) to roll their maturing debt at sustainable levels places constraints on the ECB's ability to tame inflation.
This blog takes a closer look at some of the historical context of debt, GDP, taxation rates & debt servicing costs of several European countries.
It uses this data to gain an understanding of the size of the task facing the ECB (and Italy), and opines on what the likely policy outcome of the ECB's July 21st. meeting will be.
Monetary policy has only 'niche appeal'
I wrote in earlier posts about the US Federal Reserve’s response to rapidly rising inflation and in another post compared that to the actions of the ECB. In the world of monetary policy and central banking we are frequently reminded by policy makers that affairs should be boring. Boring they may be, but when trying to balance growth and inflation it is rare for affairs to be completely straightforward.
We discussed the additional constraints that the ECB was under. Namely that the principle tool to fight inflation - that of raising rates - is likely to create an additional set of problems. These problems would see some European Governments forced to incur higher borrowing costs when rolling over maturing bond issues.
Higher yields for particular European Governments
The market had begun to factor in these higher borrowing costs. Prices fell and spreads over German Bunds rose for may European Government Bonds. Figure 1 compares BTP and Bund yields over the past 15 years. One can see the recent rise in this figure below.
Figure 1 - A comparison of Bund & BTP yields
The main focus of attention in both the financial and mainstream press has been Italy. With spreads on BTPs over Bunds reaching 236bps, levels not seen since the beginning of the pandemic. More significantly BTP yields rose to over 4% - levels not seen since the end of the sovereign debt crisis. This hike in BTP yields came after the ECB announced that it would end its bond-buying at the beginning of Q3 and would consider raising rates to stave off inflation.
Anti-fragmentation
Subsequently the ECB pacified markets with an announcement of an ‘anti-fragmentation’ tool, after an emergency meeting on Wednesday June 15th. This sent both Italian spreads & yields lower. While all the language is on ‘spread targeting’ it is worth reflecting that absolute yields are the main focus. The yield determines the debt servicing costs for that sovereign nation. While spreads grab the headlines, absolute yields determine what must be paid; as we shall explore in the remainder of this blog.
The amount of Italian debt maturing.
As the figure below shows there is a sizeable amount of Italian Government debt to roll in the next eighteen months.
Figure 2 - Italian debt maturity ladder
Italian Government Bond maturities by year can be seen here. Around €200Bn (out of a total of €369Bn) remain to mature in 2022, with a further €305Bn next year. 2023 also happens to be an election year in Italy. The remaining debt set to mature until the end of 2023 counts for roughly 25% of the outstanding Italian debt issuance.
The figure below shows the GDP of Germany and France - who along with Italy are the three largest economies in the Eurozone. The chart also shows as Spain, Portugal and Greece - with whom Italy are often compared in terms of indebtedness. Italy is one of the G10 countries with a GDP of around $2Trn - note figure 3 is in EUR (€), not USD ($).
Figure 3 - Selected European countries GDP per year since 1998 - in (€Bn)
While Italy’s GDP has grown since the advent of the Euro it has not seen as robust growth as other countries in Europe as can be seen in figure 4 below.
Figure 4 - Growth in selected European countries' GDP since the advent of the Euro
A figure widely discussed with respect to debt servicing is the debt to GDP ratio. While there is nothing particularly special about any specific level of this ratio; it does capture a country's ability to generate sufficient taxation to service debt interest payments or pay down existing debt.
Figure 5 - Debt to GDP ratios of selected European countries since 1961
Perhaps too much significance is attached to the level of '1'. Debt to GDP ratios less than 1 being looked upon as manageable, and ratios over 1 as being riotously out of control. In reality of course there is nothing special about '1' here at all. Nevertheless the higher the debt to GDP ratio, then the greater the concern about repayment and servicing.
The figure above shows the debt to GDP ratios of our group of six European countries. The data is from the early 1960s. In that time Italy has always run a high level of indebtedness relative to its economy. Being overtaken by Greece just after the introduction of the Euro.
Projecting tax revenues.
The chart below shows the actual and projected tax revenues as a percentage of GDP for our selected countries.
Figure 6 - Historic and projected tax revenues as a percentage of GDP
It can be seen that Italy has enjoyed taxation revenues above 40% of GDP for the past decade and a half. (I will confess to being a little surprised by this, but being pleasantly surprised is never a bad thing. Check out the excellent database OECD Stats for this and a host of other very useful information).
Cost of debt servicing
Since the advent of the Euro it has cost Italy 6% or less, as a percentage of GDP to service its debt. This has fallen to around 3% in 2020, as can be seen in the figure below.
Figure 7 - Cost of debt servicing for Italian Government, as a percentage of GDP.
Given Italy's current GDP this figure implies Italy's debt servicing costs as between €55Bn and €60Bn a year. Given the total outstanding Italian debt this implies a blended rate of between 2.2% and 2.4%. This is a little higher (but not much) with where we have seen Italy's average 10Yr BTP yield since the advent of the Euro - which is when nearly all of the outstanding stock was issued. See figure 8 below.
Figure 8 - Market yield of 10Yr BTP since 1991
Given the amount of debt to be refinanced in the next couple of years (see the bar chart in figure 2 and the same data represented as a pie chart below in Figure 9) one can get grasp the magnitude of the increase to debt servicing cost.
Figure 9 - Pie chart of Italian debt maturity ladder
At a yield of 4.5% Lucidate calculates that this would increase Italy's debt servicing from €55bn to €75Bn (that is to say the still to mature 75% of the debt serviced at the rate implied above, and the 25% of new debt issued at 4.5%). Should spreads widen to somewhere near 7% then this means that the cost of debt servicing increases to €91Bn.
While 7% may seem unlikely, it is a sobering thought that those levels were seen during the sovereign debt crisis. (The BTP yield was 7.05709% on 1st November 2011 - see St. Louis Fed Economic data: Fred). The 7% number is not plucked from the air...
While an increase of debt servicing to these levels is not existential, it is uncomfortable, particularly in an election year and particularly at a time with inflation-linked wage demands from public sector workers leading to industrial action in many parts Europe.
In this context one can see the dilemma facing the ECB and their desire to look closely at anti-fragmentation policy tools.
Likely policy responses.
We thus have some quantitative foundations to understand the impact of an increase in sovereign yields. This blog is focussing on the likely policy of the ECB. What might its ‘anti-fragmentation’ toolkit look like?
There will likely be some use of the Pandemic Emergency Purchase Program, an asset purchase facility. This will no doubt help. However it will be limited in size. The ECB has a formula requiring that the allocation of assets purchased converge to an amount according to a country's GDP and population. This formula could of course be tweaked to provide more assistance, but there will be a hard limit at some point to purchases under PEPP.
A tool without a hard limit that is already in the ECBs toolbox is the OMT - Outright Monetary Transactions. This was proposed during that last Sovereign Debt crisis that occurred in 2011 (which if you recall spawned the now forgotten portmanteau ‘Grexit’). The OMT comes with some strict budgetary controls and conditions attached (examples of which can be seen here -> https://www.esm.europa.eu/financial-assistance/programme-database/conditionality , which perhaps explains European states’ reluctance to embrace it.
So a ‘third’ way seems inevitable. A combination that almost certainly relaxes the formula in a PEPP-like structure to allow for greater quantities of asset purchases; as well as a diluting of some of the budgetary conditions of an OMT-like regime to make it more digestible for a country’s electorate.
Conclusions
We continued where we left off in the last blog post on the unique challenges faced by the ECB.
We took a closer look at debt to GDP ratios of several European countries. We also examined their history of economic growth and the amount of taxation revenue they have been able to realise as a percentage of GDP.
We examined the debt maturity profile of Italy. This showed approximately 25% (a quarter of a trillion euros) of Italy’s outstanding debt maturing in the next eighteen months.
We calculated that if spreads were to rise to levels seen in the Sovereign Debt crisis this could lead to an increase of €36Bn (a 65% increase) in the cost of servicing debt for the Italian government.