The ECB is also here to close governments' financing gap

May 19 2020

In response to the Covid-19 crisis, gross financing needs could amount to EUR800-900bn for the Eurozone, and only part of it will be covered by the issuance of long-term bonds. Next to state guarantees — depending on how much they translate into actual financing needs — these needs include discretionary spending (investments, tax cuts etc.), and additional expenditure on social security (automatic stabilizers). As a result, public deficits will increase. In Germany, we expect the deficit to reach 7% of GDP, in France 11%, in Italy 13.6%, and in Spain 7.8%. In addition, countries are expected to maintain their mix of short- and long-term funding: short-term debt instruments account for an average of 50% of total issuance in the Eurozone. The already revised funding plans of some sovereign debt agencies show a tendency to align the additional issuance of long-term bonds with the level of discretionary spending, while the planned issuance of short-term instruments seems to be more in line with the level of additional deficits linked to automatic stabilizers. Governments’ expectations of a temporary shock in 2020 followed by a significant recovery in 2021 contextualize the adopted strategy: to minimize the costs of transitory deficits using short-term financing. After all, the very short end of the yield curve is in negative territory for almost all euro countries.

For this year we expect a gross issuance of long-term government bonds of EUR1250bn, of which EUR450bn are attributable to the Covid-19 crisis. This corresponds to an increase by 40% compared to last year and 50% compared to the initial funding plans. By the end of April, 35% of this volume had already been placed on the market, in line with the average completion rate of the last five years. Thus, despite the massive fiscal support programs, governments have not yet been tapping the bond market hastily. However, there are considerable differences between countries. Portugal and Spain in particular have engaged in significant frontloading, while Germany and Italy are lagging behind their usual issuing speed.

This massive increase in gross issuance might suggest a significant expansion in the supply of long-term euro government bonds. Indeed, after accounting for redemptions, the outstanding amount would rise by EUR770bn this year (net issuance). This would represent an increase of 11% of the outstanding volume. Again the situation varies greatly from country to country. While the increase in Germany and Greece is close to 20%, it is below 5% in Belgium or Portugal.

Figure 1 – Gross issuance of long-term government bonds in 2020

(estimates, in % of amount outstanding)

Figure 1 – Gross issuance of long-term government bonds in 2020
Sources: National debt agencies, ECB, Refinitiv, Allianz Research

Figure 2 – Long-term government bond issuance and ECB purchases

(estimates, in € bn)

Figure 2 – Long-term government bond issuance and ECB purchases
Sources: National governments and debt agencies, ECB, Refinitiv, Allianz Research

Figure 3 – Net issuance of long-term government bonds in 2020 

(estimates, in % of amount outstanding)

Figure 3 – Net issuance of long-term government bonds in 2020
Sources: National debt agencies, ECB, Refinitiv, Allianz Research

The unprecedented ECB purchases are there to ensure the « free float » stays roughly the same (-EUR30bn or -0.5% for the outstanding amount). See Figure 2. Indeed, the ECB's increasing influence on the Eurozone government bond market calls for net issuance adjusted for the central bank's net purchases as a key metric of the Eurozone bond market. This measure reflects the bond supply available in the market i.e. the “free float” (see Figure 4). With the expansion of the PSPP and the introduction of the PEPP this year, the ECB's net purchases should largely neutralize the additional net supply of government bonds. We expect net bond purchases under these two programs to reach EUR800bn, and thus -EUR30bn as resulting free float. Despite the massive increase in gross issuance, there will be no supply shock on the euro government bond market this year. Country to country, we expect the free float to increase somewhat more markedly in Germany, Italy and Spain, while we expect a decline in the Netherlands, Belgium and Portugal (see Figure 3). The 18 May French-German proposal for a EUR500bn rescue fund does not change this outlook, as the ECB should proportionally increase the purchases of the rescue fund bonds at the expense of national governments’ bonds.

However, we estimate the dampening effect of QE on 10y yields at currently -130 bps. It could reach up to -200bps — and a flatter yield curve overall — if the PEPP program is, as we expect, extended until end of 2021 (see Figure 5). In addition to price-insensitive central bank purchases, we expect strong demand for safe assets from increasingly risk-averse private savings and regulatory-driven demand related to excess liquidity in the banking sector. This situation will lead to a further decline of the compensation for holding longer-term bonds (term premium). By reducing duration risk, this will exert downward pressure on yields. While the rise in issuance is unlikely to cause a parallel upside shift of yield curves, we might however witness a certain flattening of the curve between the very short and the middle to long end. The increased issuance activity in the short-term segment is likely to put upward pressure on yields there. At least as long as no further rate cuts are priced into forward rates (which we do not expect). At the same time, the short maturities are less exposed to buying pressure from the ECB as its purchase programs at least partially exclude assets with maturities below one year.

Figure 4 – Evolution of Euro government bond supply and ECB impact

(long-term government bonds, in €  mn)

Figure 4 – Evolution of Euro government bond supply and ECB impact
Sources: ECB, Refinitiv, Allianz Research
Figure 5 – Dampening effect on 10y yield through ECB purchases  
Figure 5 – Dampening effect on 10y yield through ECB purchases
* Term premium estimated by ACM Model, QE impact based on Eser et al. (2019)
Sources: ECB, Refinitiv, Allianz Research
Patrick Krizan
Senior Economist
patrick.krizan@allianz.com