What Is a Prime Bank? A Euribor–OIS Spread Perspective

• Date: 01 March 2014
• Published date: 01 March 2014
• DOI: http://doi.org/10.1111/infi.12044
• Author: Marco Taboga

What Is a Prime Bank? A

Euribor–OIS Spread Perspective

Marco Taboga

Bank of Italy

Abstract

Since the outbreak of the financial crisis in 2007, the level and volatility

of the Euribor–OIS spreads have incr eased significan tly. According to

the literature, this variability is mainly explained by credit and liquidity

risk premia. I provide evidence that part of the variability might also be

explained by ambiguity in the phrasing of the Euribor survey. The

participant s in the survey are a sked at what rate they b elieve interba nk

funds are exchanged between prime ba nks; given the la ck of a clear

definition of a prime bank, this question might leave room for subjec-

tive judgment. In particular, I find evidence that some o f the variabili ty

of the Euribor rates might be explained by changes in the survey

participants’perception of wh at a prime bank is. T his evidence add s to

the difficulties already e ncountered by pr evious studie s in identify ing

and measuring exactly the determinants of the Euribor rates. I argue

that these difficulties are at odds with the clarity, simplicity and

replicability that should be required of a widely used financial

benchmark.

I thank for helpful disc ussions, comments a nd suggestions Paolo Ang elini, Antonio Di

Cesare, Giuseppe Grande, Marcello Pericoli, Francesco Potente, Alessandro Secchi and

seminar participants at the Bank of Italy and at the 2013 French Finance Association

Conference.

International Finance 17:1, 2014: pp. 51–75

DOI: 10.1111/infi.12044

© 2014 John Wiley & Sons Ltd

I. Introduction

Since the outbreak of the financial crisis in 2007, the spreads between the

Euribor and Overni ght Indexed Swap (OIS) rates

1

have been among the

most closely followed gauges of tensions in the interbank market (e.g. Taylor

and Williams 2009; De Socio 2011; Nobili 2012). In the years before the

crisis, these spreads stood at a few basis points and had very limited

variability, but since 2007, when they started moving upwards, they have

been much higher and more volatile, touching peaks of hundreds of basis

points on some occasions.

What do these spreads measure? A consensus has emerged that they

embed both credit risk premia, associated with the default probability of

borrowers of interbank funds, and liquidity premia, due to the fact that

interbank deposits are highly illiquid.

2

Several empirical studies confirm that

both of these components are relevant and help explain the time variation in

the Euribor–OIS spreads (e.g. Schwarz 2010; Filipovic and Trolle 2011).

In this paper, I argue that other factors that have to do with the way the

Euribor rates are c alculated mig ht also be at play. The Euribor rates are

averages of survey responses by banks that have been asked the following

question: To the best of your knowledge, what is the interest rate that a

prime bank would charge another prime bank on an unsecured loan? The

key phrase in this question is ‘prime bank’. Before the crisis began, the

concept of a prime bank was rather unambiguous: There were dozens of

large and internationally active banks that enjoyed high credit ratings and

had tiny credit default swap (CDS) premia (around or below ten basis

points); any one of these banks could have easily been recognized as a prime

bank. However, during the crisis, most of these banks experienced deterio-

rations in their credit ratings and surges in their credit spreads. Which of

them can still be considered prime? In the absence of a standard definition of

a prime bank, this question calls for quite a bit of subjective judgement.

Therefore, after 2007 the Euribor rates might conceivably have been influ-

enced by changes in the survey respondents’perception of what a prime

bank is. This paper provides empirical evidence in favour of this hypothesis.

Existing studies use either averages (e.g. McAndrews et al. 2008, Michaud

and Upper 2008) or quantiles (e.g. Filipovic and Trolle 2011) of the distribu-

tion of the CDS spreads of banks to approximate the credit risk component of

the Euribor–OIS spread. However, if the composition of the set of banks that

1

Or between Libor and OIS rates. In this pape r, I concentrate on Euribor rates o n euro‐

denominated depos its.

2

It can be very costly, if not imp ossible, to withdraw funds employed in an i nterbank deposit

before its expiry.

52 Marco Taboga

© 2014 John Wiley & Sons Ltd