Countercyclical Foreign Currency Borrowing: Eurozone Firms in 2007–09

• Published date: 01 February 2022
• Author: PHILIPPE BACCHETTA,OUARDA MERROUCHE
• Date: 01 February 2022
• DOI: http://doi.org/10.1111/jmcb.12818

DOI: 10.1111/jmcb.12818

PHILIPPE BACCHETTA

OUARDA MERROUCHE

Countercyclical Foreign Currency Borrowing:

Eurozone Firms in 2007–09

Using syndicated loan-level data, we document and explain the causes and

implications of a new and surprising stylized fact. In the midst of the nan-

cial crisis, dollar borrowingby leveraged Eurozone (EZ) corporates rose dra-

matically relative to their euro borrowing. Weshow that this resulted from a

shift from EZ to non-EZ banks, mainly U.S. banks. This was combined with

an increase in the proportion of dollar lending by non-EZ banks, explained

by a rise in the relative cost of euro wholesale funding and the disruptions

in the FX swap market. Non-EZ banks thus dampened the 2007–09 credit

crunch in Europe.

JEL codes:E44, G21, G30

Keywords:money market,swaps, credit crunch, corporate debt, foreign banks

A      

been observed during the recent nancial crisis. Although gross capital ows de-

clined sharply in general (e.g., Broner et al. 2013), the decline has been particularly

steep for banking ows among developed economies (Milesi-Ferrettiand Tille 2011).

The literature shows evidence of a ight home effect in syndicated bank loans (Gi-

annetti and Laeven 2012a) and of nancial protectionism in bank lending (Rose and

Wieladek 2014). We also observe that global banks have increased the use of their

local currency in their lending (e.g., Ivashina, Scharfstein, and Stein 2015). In this

context, it is surprising that dollar borrowing by many Eurozone (EZ) nonnancial

corporates increased dramatically from an average of 5 billion USD during 2004–06

The authors would like to thank Linda Goldberg, StevenOngena, and Clemens Otto, as well as seminar

participants at the NewYork Fed, the IMF,the World Bank, the Swiss National Bank, the University of Lau-

sanne, University of Geneva, UniversityParis Dauphine, Aix-Marseille School of Economics, University

of Zurich, and participants at the 1st RELTIFconference held at Oxford University for helpful comments

and suggestions. Financial support from the ERC Advanced Grant #269573 is gratefully acknowledged.

Philippe Bacchetta is Swiss Finance Institute professor of economics, Faculty of Business and Eco-

nomics, University of Lausanne (Email: philippe.bacchetta@unil.ch).Ouarda Merrouche is professor

of economics, Faculty of Business and Economics, EconomiX, University of Paris-Nanterre (Email:

ouarda.merrouche@alumni.eui.eu).

Journal of Money, Credit and Banking, Vol. 54, No. 1 (February 2022)

© 2021 The Ohio State University

204 :MONEY,CREDIT AND BANKING

Fig 1.Syndicated Loan Issuance Denominated in U.S. Dollar.

N: The sample includes all nonnancial Eurozone borrowers.

S: Thomson-Reuters Dealscan, European Central Bank.

to a peak of 40 billion in 2008 (solid line in Figure 1). Although the proportion of

dollar borrowing was about 5% in 2004–06, it increased to 35% for noninvestment

grade borrowers and to 15% for investment grade borrowers in the second half of

2008 (Figure 2).1Importantly, we do not observe a similar pattern in bonds. The in-

crease in dollar bond issuance was rather transitory (it lasted on average 1 quarter)

and much smaller (4 percentage points for investment grade rms and 8 percentage

points for noninvestment grade rms).

The purpose of this paper is to document this surprising episode in international

banking ows and identify the factors that led to that development. We argue that the

increase in dollar borrowing by EZ leveraged2rms can be explained by the distinct

reaction of EZ and U.S. banks (the main foreign lenders in the EZ syndicated loan

market) to the nancial crisis. Moreover, these reactions are related to two main (and

perhaps related) symptoms of the global nancial crisis: the domestic credit crunch

and the drying up of global interbank markets. Figure 1 depicts a high positive cor-

1.This increase cannot be attributed to a valuation effect: the euro appreciated against the dollar by

about 20% during the period when the increase was strongest, that is, Q2-2007 and Q3-2008.

2.Throughout the paper, we use the terms leveraged, noninvestment grade, low-credit quality, and

risky interchangeably. Noninvestmentgrade rms in our sample have an average leverage ratio (or a ratio

of long-term debt over total debt) of 19.5% against 5.7% for investment grade rms.

PHILIPPE BACCHETTAAND OUARDA MERROUCHE:205

Fig 2.Percentage of Syndicated Loan Issuance Denominated in U.S. Dollars by Borrower Risk Type.

N: Risky borrowers are rated belowinvestment grade. The sample includes all nonnancial Eurozone borrowers.

S: Thomson-Reuters Dealscan, authors’ calculations.

relation between the amount of dollar debt issued by EZ corporates and an indicator

of the tightness of the EZ credit supply.3

Our baseline sample includes quarterly syndicated loan issuances obtained from

Thomson-Reuters Dealscan for the period Q1-2004 to Q4-2009. We observe the na-

tionality of lead lenders and the currency denomination of loans. In our baseline set-

ting we dene non-EZ banks as arrangers headquartered outside the EZ. Moreover,

we dene a non-EZ bank loan as a syndicated loan with at least one arranger bank lo-

cated outside of the EZ; alternatively, we measure the participation of non-EZ banks

by their percentage in a syndicate.

During the nancial crisis, U.S. banks partly replaced EZ banks in loans to riskier

rms. A potential explanation is the difference in banking regulation. During this

period and until 2014, U.S. lenders operated fully under Basel I, while EZ banks

were under Basel II.4Under the Basel I framework, the risk weight on risky and safe

3.This indicator is the net percentage of domestic banks surveyed by the European Central Bank

(ECB) that report tightening in credit standards to large rms in the past 3 months.

4.While the US ofcially adopted Basel II in 2007, the date of expected compliance was delayed to

2012. See Daryl Getter, US implementation of the Basel Capital Regulatory Framework, Congressional

Research Report 7-5700, April 9, 2014. On June 7, 2012, the Federal Banking Regulators announced the

nal rule for implementation of Basel II. First approvals for the use of Basel II capital rules were issued in

February 2014.