+1 Added to my documents.
Please be aware your selection is temporary depending on your cookies policy.
Remove this selection here

Eurozone government bonds A supportive mix of remarkable funding progress and ECB QE still to come



Funding progress looked quite encouraging at July end for Eurozone government bonds, as roughly 80% of estimated yearly net issuance have been placed, mostly (more than 50%) in just four months, between April and July. Putting remaining supply in perspectives with ECB flows, the technical picture for EZ government bonds looks friendly to the current environment of low core yields and subsequent, persisting search for carry.


September 2020


Septembre 2020


70% and 80% of overall gross and net EZ sovereign supply, respectively, estimated for all of 2020 was placed in the first seven months of the year

2020 funding progress: so far, quite good

On the back of the huge fiscal response needed to counteract the Covid-related global crisis, yearly borrowing requirements have suddenly jumped to record levels for all major sovereign debt issuers of advanced economies. Eurozone countries were no exception, and in a few months the sharp increase in projected fiscal deficits has led to a skyrocketing in gross and net issuance of public debt needed for all of 2020. Just to take a few examples, Italy’s and Spain’s net bond issuance on the year has increased by, respectively, 3.5 and 4 times vs initial yearly targets, while among core countries Germany’s funding needs rose even higher, consistent with its huge fiscal stimulus. Revisions and heightened refunding needs, furthermore, came at the end of a first quarter with, on average, a lower increase in funding. This because the first two months of the year had been run based on previously much lower estimated refunding activity, while March, at the peak of the crisis, was quite challenging in terms of market conditions. This combination led to an even stronger acceleration in primary market activity in the second quarter and in July. In the four months from April to July, in fact, unusual and unprecedented efforts were made by treasuries of all EZ countries to rapidly meet higher funding needs. These efforts were undoubtedly effective thanks mainly to ECB activism on the demand side and to monetary policy’s indirect supportive effects for private investors’ demand for sovereign debt, mainly among banks.

Moving closer into the numbers, 2020 funding progress looked quite encouraging in late July. In a nutshell, 70% and 80% of overall gross and net EZ sovereign supply, respectively, estimated for all of 2020 was placed in the first seven months of the year, with quite broad progress among different countries (see chart 1). While the “big four” – namely Germany, France, Italy and Spain – still have limited remaining net supply to be placed, other countries have already reached their yearly net targets or even surpassed them. As we were anticipating, most of the issuance (more than 50% of yearly volume) came from April to July, as the end of March saw only 28% of EZ public debt net issuance being placed. We guess it is worth considering that this jump in supply did not prevent bond markets from rallying. EGB’s average YTM fell from a 50 bps peak in April to -3 bps as of end-July, meaning that record issuance volumes concentrated in just a few months were more than matched by stronger demand.

Numbers reported in the chart 1 are relative to M/L term bonds and aggregate numbers of the following 10 countries: Germany, France, Italy, Spain, The Netherlands, Belgium, Austria, Finland, Ireland and Portugal.

Graph 1

On the demand side, core banks increased their holdings of both non-domestic and domestic government debt quite evenly

Different funding mixes between core and periphery countries, reflected in different ECB WAM purchases

Extraordinary funding needs produced by the Covid-crisis were matched by quite different funding mixes between core countries on one side and periphery countries on the other. As in the aftermath of the GFC and the sovereign crisis, core countries turned to sizeable issuance of bills and short-dated government bonds, while periphery countries continued to prefer mid and longer dated bonds, in order to keep average debt maturity high and reduce next years’ refunding risk. Reported figures (chart 2) show that year-to-date net debt issuance is split almost 50-50 between bonds & bills for core countries, while periphery mix is 85% weighted in bonds. Details on major countries show Germany even reaching a mix of 25-75 mix in favour of bills so far, while France’s mix looks more balanced between the two. This different mix was reflected in ECB purchases, too, as Germany (and the Netherlands) saw the lowest QE WAM of any Eurozone countries.

On the back of available refunding announcements and numbers released so far, we project that the year-to-date picture is quite representative of the overall expected mix for all of 2020. The remaining months of the year are still likely to show a tilt in favour of short dated-instruments for German debt, while the French balance shows limited needs left in bond refinancing and the “bill side” almost done is not going to change significantly. As far as periphery countries are concerned, we should expect current funding mix to be confirmed around current weights by year end.

The role of MFIs on the demand side

Eurozone banks have purchased record amounts of government debt in the post-Covid months. According to ECB data, MFIs’ holdings of government debt increased by EUR 330bn in H1-2020, with a jump of 22% vs December 2019 levels. Most of the increase, roughly EUR 250bn, took place from March to May alone. During these three months banks bought a volume of public debt not so far from the correspondent EUR 268bn purchases accumulated by the ECB. The pace slowed in June, but there was still an additional increase in banks’ holdings of EUR 37bn.

The role of MFIs in supporting domestic sovereign debt in challenging times is not new. However, the interesting part of the present story is that for the first time in years, core banks increased their holdings of both non-domestic and domestic government debt quite evenly, respectively by EUR 91bn and EUR 89bn in H1-2020.

As the reported figures (chart 3) show, one of the key characteristics of MFIs’ behaviour in recent years has been quite a strong home bias, with banks’ preferring their own domestic government debt. This trend, as we know, started during the Eurozone debt crisis with huge reductions in exposure of peripheral countries’ debt from core banks, a trend that, to some extent, forced periphery banks to be even more tilted to their domestic debt, especially in volatile times. We have now seen six straight months of increases in non-domestic government debt holdings. No doubt, this trend was certainly supported by ECB QE & huge liquidity injections first and more recently by growing expectations of EU deal, both reducing perceived tail risks and representing incentives to cross-country flows in a sort of portfolio rebalancing channel of QE, something hoped for, but hardly achieved by the ECB for quite some time.

Graph 2

Eurozone sovereign bond net issuance, “net” of ECB purchases, should become negative in the remaining part of the year

The past few months have pointed to some changes not only in that aspect but also in the link between banks and Long Term Refinancing with the ECB. Until February this year, LTROs/TLTROs facilities had been very much a sort of “periphery banks story”, with Italian and Spanish banks accounting for the lion’s share of reliance, and core banks on the opposite end, having a limited portion of existing TLTROs and owing most of the excess liquidity in the system, parked (actually charged at negative rates) with the ECB. Under this respect, first partially through March-May weekly LTROs and then mostly with June TLTRO, these roles changed (see chart 4). Overall, core banks moved from only one quarter to almost half usage of their TLTRO overall allowance with the ECB, for the first time surpassing periphery banks’ outstandings. As of end-June, French banks showed higher TLTRO take-up than Italian banks, and German banks were at a higher levels than Spanish banks. Dutch banks, just to take another example, moved from just 26% to 67% of maximum allowance. Something seems changed for the better, for less fragmentation, both in the attitude towards a pure home bias on sovereign debt and on the use of funds made available by the ECB.

The role of ECB demand in the remaining months of the year

Putting remaining supply in perspective with ECB flows, the technical picture for EZ government bond looks friendly to the current environment of low core yields and the subsequent search for carry. So far, according to our estimates, the ECB has covered for most of YtD net supply for many countries, reaching a proportion of 90% of bond net issuance. Assuming that most overall QE potential will be allocated to public debt (80%) through capital key rules, with a proportion of two thirds this year and one third left in H1 next year, the ECB purchases should more than cover for remaining fiscal needs over these last months of 2020. This would be the case for both core and periphery countries as far as M/L term bonds are concerned. In light of deviations accumulated so far, the assumption on capital keys looks to some extent hard to completely achieve by yearend, because that would imply a massive turn of ECB purchases towards German debt in just the five remaining months and quite a sharp slowdown in periphery flows. One implicit assumption of our analysis is that the ECB will make full use of the PEPP amount by June next year. This is also the baseline scenario underlined by the ECB President at the last meeting and also revealed in the minutes, as a condition of the alternative scenario of lower usage (PEPP size as a “ceiling”) would be a significant upside surprise to the mediumterm inflation outlook, a bar quite high in the current environment. Furthermore, an eventual partial “saving” of full PEPP power would likely take place more in Q2 than in Q1 next year. The assumption on time distribution of ECB purchases between this year (two thirds) and 2021 (one third) has both to do with the number of QE months and trends in fiscal needs. First, PEPP is going to last nine months in 2020 and six months in 2021 and PSPP + the EUR 120bn added envelope should be entirely a 2020 story. Second, fiscal deficits are projected to be much lower in 2021 than this year, in the case of some countries even halving from 2020.

Under the above assumptions, a closer look at the numbers of the two biggest periphery countries shows that ECB QE volumes still to come should more than cover for remaining limited net supply. By end-July, together with the additional net placement of more than EUR 20bn of Bills, Italy’s net funding roughly reached EUR 130bn of its estimated yearly target close to EUR 150bn M/L term bonds. 

Graph 3


Likewise, Spain delivered overall net issuance of EUR 82bn of its EUR 102bn target, together with a net bill issuance of EUR 16bn. Each of the two countries is therefore are left with roughly just EUR 20bn in additional net funding needed to be done, while ECB purchases projected under previous assumptions should be close to EUR 40bn for both countries. It is also worth mentioning that these last numbers are calculated under a capital key allocation and that any deviation from that would mean a higher net negative issuance post- ECB purchases.

Among core countries, the demand/supply balance of France and Germany also looks better than so far. Year-to-date data show that cumulative ECB purchases covered for most of French bond net issuance and for most of combined bonds & bills net issuance of German debt. Under previous assumptions, forthcoming ECB demand appears higher than net supply for both countries, as well, with eventual deviations from capital key rules reducing the positive gap between QE flows and supply.


Eurozone sovereign bond net issuance, “net” of ECB purchases, was modestly positive in the first seven months of the year for most Eurozone countries. The sign should turn negative if the ECB doesn’t significantly slow its path of purchases of sovereign debt in the last part of the year. This is true for both core and periphery countries. The PEPP increase, adapting QE overall firepower to cover also for most of next year’s funding needs, made the programme much more effective and quite credible in its time extension. Together with huge TLTRO liquidity injection, its indirect supportive effects on private demand finally reduced the need for the ECB to act directly in support of public debt, therefore gaining flexibility at its disposal for the next months.

This flexibility could be used by the central bank in additional support of private programmes, or kept intact for action next year, or could work as a backstop in case of need, as yields are at lows and spreads have tightened not far from pre-Covid levels. It could as well balance an eventual, partial reduction of holdings by the banking system, as year-end approaches and in light of the sharp increase we referred to previously. Banks still have incentives to keep govies holdings in the current environment of very attractive funding facilities put in place by the ECB and negative rates charged on excess reserves, therefore probably limiting an eventual reduction in exposure.

This demand/supply backdrop is likely to support the present environment of low and negative yields in core countries at the same time to support the search for carry through periphery debt, despite the strong tightening in spreads delivered in the last months.

Finalised on 24/08/2020

BERTONCINI Sergio , Head of Rates and FX Research
Send by e-mail
Eurozone government bonds A supportive mix of remarkable funding progress and ECB QE still to come
Was this article helpful?YES
Thank you for your participation.
0 user(s) have answered Yes.