Relationship Lending and Firms’ Leverage: Empirical Evidence in Europe

Date01 September 2017
AuthorRoberto Guida,Valentina Sabato
Published date01 September 2017
Relationship Lending and Firms
Leverage: Empirical Evidence in
Roberto Guida
International University at Rome, Via G. Baracconi 5/A, 00161 Rome, Italy
Valentina Sabato
LUMSA University of Rome, Rome, Italy
Using a novel measure of relationship lending based on the kind of information
banks use to assess borrowers, we investigate the role of relationship lending in
rmscapital structure. Using a unique dataset of European manufacturing rms,
we measure relationship lending based on three dimensions (closeness, soft
information, exclusivity) and relate them to rmsleverage. Overall our results
support the hypothesis that supply factors matter. We nd that the actual use of soft
information increases leverage and only rms without soft information-intensive
relationships increase their leverage through multiple relationships. However, the
effect of relationship lending on leverage varies across countries.
Keywords: relationship lending, soft information, cap ital structure, leverage,
financial systems
JEL classification: D45, G15, G21, G32
1. Introduction
In this paper, we use a novel relationship lending measure based on the kind of information
banks use to assess borrowerscreditworthiness to investigate the role of relationship
lending in rmscapital structure. In particular, we are interestedin determining whether
having relationships with banks that are able to gather relevant soft information about the
prospects and creditworthiness of borrowers affects rmsleverage ratios.
Relationship lending is a common practice in credit nancing. According to the 2014
Survey on the Access to Finance of Enterprises in the Euro Area(European Central
The authors are grateful to two anonymous referees and John Doukas (the Editor) for their
helpful suggestions. Furthermore we would like to thank Stefania Cosci and Valentina
Meliciani for many comments and discussions.
European Financial Management, Vol. 23, No. 4, 2017, 807835
doi: 10.1111/eufm.12109
© 2016 John Wiley & Sons, Ltd.
Bank, 2014), bank-related products are the most relevant sources of nancing for small-
and medium-sized enterprises (SMEs) in the eurozone, whereas market-based sources
are less often reported as being relevant; thus, banks are the main actors and banking
relationships the main instruments in nancing for SMEs in Europe. However, in the last
few decades, the European nancial system has witnessed a process of transformation
that has moved it somewhat toward an arms-length system, stimulating and promoting
competition between nancial institutions.
To meet Basel II capital requirements, many
banks have been adopting new screening technologies, based on standardised internal
rating models that use hard information(quantitative information such as credit history,
balance sheet data, rating and scoring), that have progressively substituted pre-existing
technologies that are also based on soft information(qualitative information consisting
mainly of words expressing subjective judgements, opinions and perceptions). These
recent trends in the European nancial markets, along with better information
processing, more sophisticated rating tools and the growth of the securitisation market,
have pushed banks to reduce their relationship lending activities. Hence, strong bank-
rm relationships seem to become less important (Issing, 2003; Memmel et al., 2008).
How does this impact on rmsability to nance their investments, especially for rms
that theory predicts may be credit constrained?
Since the work of Modigliani and Miller (1958), the capital structure literature has
tried to answer Myers(1984) capital structure puzzle. In this regard, different theories
that give alternative explanations for rmscapital structure decisions have been
developed. The empirical literature, which mainly tests the validity of the alternative
theories, has examined the role of several rm-specic factors, under the implicit
assumption that rms are not credit constrained (Faulkender and Petersen, 2006).
However, the same market frictions that make capital structure choices relevant also
imply that rms may be unable to raise sufcient capital to fund all of their good projects;
i.e., rms may be rationed by their lenders (Stiglitz and Weiss, 1981). In this case, supply
factors become relevant: a rms capital structure may depend on the amount of credit
available to the rm.
Asymmetric information and agency cost problems may be overcome by lenders that
specialise in collecting information about borrowers. By interacting with borrowers over
time and across different products, the nancial intermediary may be able to mitigate the
information asymmetry that is the cause of the market failure, and these nancial
relationships have been empirically demonstrated to be important in relaxing capital
constraints. In particular lending relationships should be most valuable where the
information about a rm and potential investment opportunities is most uncertain (SMEs
and innovative rms).
Whenever the amount of soft information produced about a rm
Rajan and Zingales (2003) classify nancial systems as relationship-based and arms-length
systems rather than bank- and market-based systems. There is not an exact coincidence
between the two classications, although they overlap to some extent. On the validity of the
distinction between bank- and market-oriented systems, see also Schmidt and Tyrell (1997).
Although the literature suggests that banks have an advantage in nancing informationally
opaque rms by screening and monitoring borrowers, banks actually nance a smaller
fraction of total debt when adverse selection and risk-shifting problems are potentially large
(e.g., for start-ups) (Huyghebaert and Van de Gucht, 2007).
© 2016 John Wiley & Sons, Ltd.
808 Roberto Guida and Valentina Sabato

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT