Demand Shocks for Public Debt in the Eurozone

• Published date: 01 October 2022
• Author: ANDRAS LENGYEL,MASSIMO GIULIODORI
• Date: 01 October 2022
• DOI: http://doi.org/10.1111/jmcb.12891

DOI: 10.1111/jmcb.12891

ANDRAS LENGYEL

MASSIMO GIULIODORI*

Demand Shocks for Public Debt in the Eurozone

In this paper we use intraday government bond futures price changes around

German and Italian Treasuryauctions to identify unexpected shifts in the de-

mand for public debt. Estimates show that positive demand shocks lead to

large negative movements in Treasury yields. Evidence shows signicant

spillover effects into Treasury bond, equity, and corporate bond markets of

other eurozone countries. We nd interesting differences in the effects of

demand shocks between the two countries, consistent with the “safe-haven”

status of German bonds versus the “high-debt” status of Italian Treasuries.

Results suggest that these effects are stronger during periods of high nan-

cial stress.

JEL codes: E43, F4, G15

Keywords: high-frequency identication, primary market, sovereign

bonds, yield curve

G     are usually

considered the safest and most liquid assets. They have a key role in savers’ port-

folio decisions, investors’ risk-management activities, and banks’ repo operations.

Consequently, private sector demand for the largest European economies sovereign

bonds was initially stable following the introduction of the single currency.Volatility

was low, with yields exhibiting strong comovements, up until the Global Financial

*The authors would like to thank the editor, twoanonymous referees, Roel Beetsma, Raffaele Corvino,

Robin Döttling, Gabriele Galati, Jakob de Haan, KostasMavromatis, Julia Schaumburg, Sweder van Wijn-

bergen and seminar participants at 2nd Lancaster Summer School on Applied Macroeconometrics, Eras-

mus Universiteit Rotterdam, De Nederlandsche Bank, Universiteit van Amsterdam, and the 12th PhD

Workshop at Collegio Carlo Alberto for their useful comments and suggestions. Andras Lengyel would

like to acknowledge the hospitality of the De Nederlandsche Bank where part of this research was con-

ducted. The Online Appendix is available at: https://www.uva.nl/en/prole/l/e/a.lengyel/a.lengyel.html

A L is a PhD candidate at the University of Amsterdam and TinbergenInstitute (E-mail:

a.lengyel@uva.nl). M G is a Professor of Empirical Macroeconomics at the University

of Amsterdam and TinbergenInstitute (E-mail: m.giuliodori@uva.nl).

Received May 9, 2020; and accepted in revised form October 27, 2021.

Journal of Money, Credit and Banking, Vol. 54, No. 7 (October 2022)

© 2021 The Authors. Journal of Money, Credit and Banking published by Wiley Periodicals

LLC on behalf of Ohio State University.

This is an open access article under the terms of the Creative Commons Attribution-NonCom-

mercial-NoDerivs License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no modications or adaptations

are made.

1998 :MONEY,CREDIT AND BANKING

Crisis.1The crisis broke this pattern and credit risk became an increasing problem

in many euro area countries. Countries on the periphery of the euro area have seen

private sector demand for their bonds drying up. The European Central Bank (ECB)

took up an active role and entered the buy side of the market, in order to rst facilitate

liquidity with the Securities Markets Programme (SMP), and later to inuence prices

with the Public Sector Purchase Programme (PSPP). The aim of this latter program

was to decrease long-term interest rates, by purchasing large quantities of long-term

euro area public debt securities.

These shifts in the demand for public debt provide the motivation for our study.

Our goal, in a broader sense, is to identify unexpected demand shocks for German and

Italian government bonds and analyze how nancial markets react to these shocks.

The choice of these two countries stems from the fact that a shift in the demand for

German and Italian debt is associated with markedly different sentiments in nancial

markets. German government bonds are some of the safest traded securities, expe-

riencing large inows during times of high nancial stress. Consequently, investors

attach large “safety” and “liquidity” premiums to them (Ejsing and Sihvonen 2009,

Klingler and Sundaresan 2020). Investors’ attitude toward Italian bonds, on the other

hand, is substantially different. Italy has one of the highest public debt-to-GDP ratios

in the euro area. Its sustainability is a topic of ongoing debate and investors require a

substantial risk premium for holding Italian bonds.

The rst contribution of our paper is to identify unexpected shifts in the demand for

public debt in these two countries, by exploiting the institutional setup of government

bond auctions. The identication strategy follows Gorodnichenko and Ray (2017),

who proposed it in the context of U.S. auctions. These auctions are important events

where institutional investorscan accommodate partof their security needs. Therefore,

auction results can give hints about changes in market demand for these securities.2

Furthermore, the prevailing demand at an auction shows investors’ perception of the

health of a country’s economy and creditworthiness of its government.3Importantly,

the timing and the setup of these events are such that they allow to capture price

movements that can mainly be attributed to demand-side factors. Debt management

agencies disclose information about the supplied securities and its quantity well in

advance of an auction. On the auction day, therefore, investors are already well in-

formed of the supply side of the market. The demand side is, however, unknown up

until the agency releases the results of the auction. Hence, when these results are

published, investors receive new information solely on the demand conditions. By

1. See Figure A1 in the Online Appendix.

2. News agency Bloomberg wrote on December 27, 2018: “The Treasury in Rome plans to auction

as much as 5 billion euros of debt Friday, including benchmark ve-year and 10-year securities. (...) The

results will provide an indicationof the underlying demand for Italian bonds next year.”

3. As news agency Reuters wrote on July 30, 2018: “Italy’s scheduledbond auctions, which included

the sale of a new 10-year benchmark, was seen as a test of investorrisk appetite amid political tensions in

the euro zone’s thirdlargest economy.”

ANDRAS LENGYEL AND MASSIMO GIULIODORI :1999

looking at high-frequency price movements of government bond futures contracts

around the rst releases of the results, we can isolate price variations that are mainly

attributed to the strength of the demand side.

The second contribution of the paper is on the estimation of the effects of the iden-

tied Treasury demand shocks on domestic bonds and the yield curve. Our estimates

show that a one-standard-deviation demand shock in Germany decreases home yields

by around 1.6 basis points. In Italy a similar shock has an effect of 3.3 basis points.

These effects are found to last up to 30 trading days. Using our estimates we provide

back-of-the-envelope calculations of the effects of the PSPP on German and Italian

yields, which are in line with the ndings of the existing literature (Altavilla et al.

2015, De Santis 2020). We also investigate the effects of a location-specic demand

shock on the Treasury yield curve, distinguishing between short-term and long-term

Treasury demand shocks. We show that in Germany the shocks have local effects, in

the sense that nearby (i.e., similar maturity) yields respond stronger. Our ndings for

Germany are in line with Gorodnichenko and Ray (2017), who nd local effects of

demand shocks for the U.S. Treasury yield curve. Our results are also consistent with

studies bringing evidence of bond market segmentation, such as Boermans and Ver-

meulen (2018). In Italy, on the other hand, a positivedemand shock always decreases

short-term interest rates more, regardless of the shock location. We nd that German

and Italian demand shocks have spillover effectson the Treasury bond yields of other

euro area countries. While the German shocks have more sizable effects on France

and the Netherlands, the responses to the Italian demand shocks are mostly centered

on the Spanish Treasury yields. We also show that Treasury demand shocks lead to

reactions of the corporate bond markets. More specically, euro area corporate bond

yields drop in response to German demand shocks, whereas they are rather unaffected

by Italian demand shocks.

Of particular interest are the results of the spillover effects of demand shocks on

equity markets and Treasury credit default swap (CDS) spreads. The seemingly sim-

ilar information (increased demand for government debt) has vastly different effects

between the two countries. More specically the main euro area stock indices drop

following German demand shocks, while Treasury CDS spreads increase. On the

other hand, Italian demand shocks lead to positive responses of equity prices and de-

creases in CDS spreads. We reconcile these results as follows.The information about

increased demand for bonds alters investors’ risk preferences and expectations in two

different ways. A sudden demand increase for German bonds is associated with nan-

cial markets turning to a “ight-to-safety” mode. Willingness to take risk decreases

and investors rebalance their portfolios from equities to bonds. On the other hand, the

market movements associated with a positive Italian demand shock are the result of

a higher risk appetite and a better outlook for the Italian economy leading to positive

effects on the Italian stock market and decreases in Treasury CDS spreads.

Finally, our last contribution is to test for the presence of state dependence. We

show that during times of high nancial stress the estimated responses documented

above tend to be larger relative to periods of low nancial stress. We also study if

positive and negative demand shocks for Treasury bonds have asymmetric effects.