The Choice of Debt Source by UK Firms

• Author: Laura Mccann,Andrew Marshall,Patrick Mccolgan
• Published date: 01 May 2016
• Date: 01 May 2016
• DOI: http://doi.org/10.1111/jbfa.12194

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 43(5) & (6), 729–764, May/June 2016, 0306-686X

doi: 10.1111/jbfa.12194

The Choice of Debt Source by UK Firms

ANDREW MARSHALL,LAURA MCCANN AND PATRICK MCCOLGAN∗

Abstract: We examine the choice of borrowing source among public debt, syndicated bank

loans, bilateral bank loans and non-bank private debt. Using a sample of 400 non-financial firms

over the period 2000–2012, we find strong support for the reputational theory of borrowing

source. Larger firms are more likely to borrow in public debt markets. Bank dependent firms are

less likely to borrow in public debt markets and choose between bank and non-bank private debt

based on maturity, collateral available to lenders and other firm characteristics. These results are

consistent with the role of borrower reputation being the primary determinant of borrowing

source for UK listed firms.

Keywords: debt policy, public debt, bilateral loans, syndicated loans, non-bank private debt

1. INTRODUCTION

Theoretical research assigns an important role to the debt source in a firm’s capital

structure decision. First, Diamond’s (1991) reputational theory proposes that firms

build reputation in credit markets through borrowing and repaying private debt

initially, and that repeating this process over time allows firms to build a reputation

in credit markets. This results in lower borrowing costs. Second, bank monitoring

can reduce asymmetric information costs for firms that produce less information

than lenders in public debt markets optimally demand, as banks can more closely

scrutinize the firm to monitor cash flows (Fama, 1985). Finally, bank debt is more

easily renegotiated in the event of financial distress given that bank lenders have more

information on the firm’s cash flows and investment opportunities to decide whether

liquidation is optimal (Chemmanur and Fulghieri, 1994; Yosha, 1995).1

In this paper we empirically investigate theories of debt source choice by examining

their determinants in a UK setting. We use the distinctive features of the UK market

for corporate lending to extend the literature on the choice of borrowing source. UK

firms have historically placed a greater dependence on bank debt than public bond

∗Andrew Marshall and Patrick McColgan are from the Department of Accounting and Finance, University of

Strathclyde, UK. McCann is from Aberdeen Business School, University of Aberdeen, UK. The authors are

grateful to Dick Davies, Paul Draper, Robert Faff, Alan Goodacre, David Mauer, Krishna Paudyal, Gordon

Phillips, Leilei Tang, and to seminar participants at the 2013 Midwest Finance Association Annual Meeting

(Chicago) for helpful comments and advice on earlier versions of this work. All errors remain our own.

(Paper received June 2015, revised revision accepted February 2016).

Address for correspondence: Laura McCann, University of Aberdeen Business School, University of

Aberdeen, Aberdeen, AB24 3QY, UK.

e-mail: l.mccann@abdn.ac.uk

1 Johnson (1997) and Denis and Mihov (2003) provide detailed reviews of the theoretical determinants of

borrowing source.

C

2016 John Wiley & Sons Ltd 729

730 MARSHALL, MCCANN AND MCCOLGAN

issues. Data from the Bank of England (2009, 2012) illustrate that UK firms continue

to borrow predominately from banks and show a large increase in the use of syndicated

bank loans, relative to bilateral loans, in recent years. Moreover, the underdeveloped

nature of the public bond market in the UK creates a potentially important role for

non-bank private debt as an alternative bond source in a firm’s financing choice in

the UK (Breedon, 2012). Diamond (1991) refers to the importance of non-traditional

debt sources for low credit quality firms screened out of bank lending and public bond

markets. Given the description of non-bank private debt as combining the features of

low-grade public debt and bank debt in Denis and Mihov (2003), we expect that non-

bank private debt can represent an important financing source for UK firms.

Using a sample of UK non-financial firms in the FTSE-350 index of the London

Stock Exchange (LSE) over the period 2000 to 2012 we consider the determinants of

both the existing balance sheet debt and the source of finance for new debt issuance.

In doing so, we distinguish between borrowing from public debt markets, bank loans

(both bilateral and syndicated) and non-bank private debt. Focusing on existing and

new debt issues allows us to consider the importance of firm and debt characteristics,

credit market conditions and a firm’s financing history in determining incremental

financing choices.

Our study extends the literature on the choice of debt source in a number of

areas. First, we consider a broader range of bank lending sources for both existing

balance sheet debt and marginal debt financing choices than has been examined

in the literature. Examining incremental financing decisions allows for an improved

understanding of the role of firm, loan and economic characteristics, and a firm’s

borrowing history in determining the source of borrowing than is possible when ex-

amining previously issued debt on the firm’s balance sheet. In addition to splitting debt

financing between bonds and loans, we separate bonds between public and non-bank

private bonds, and bank loans between syndicated and bilateral lending agreements.

Second, we use this expanded range of borrowing sources to examine the impact of

borrower reputation on a firm’s choice of debt source. The initial empirical research

building on Diamond’s (1991) asymmetric information-based model of borrower

reputation proposes that firms borrow from public and private debt sources. This

research has developed over time to consider the role of reputation in accessing non-

bank private debt (Johnson, 1997; Denis and Mihov, 2003; Arena, 2011). However

few studies distinguish between bilateral and syndicated bank loans and none, to our

knowledge, in the context of the choice across a range of debt sources.2We arg ue

that syndicated loans combine elements of traditional relationship lending associated

with bilateral bank loans and transaction based lending more commonly associated

with raising capital in bond markets. In this sense, syndicated loans can be viewed as

an extension of Diamond’s (1991) reputational theory of borrowing source choice.

Syndicated loans are more easily monitored than public bonds, but less so than

bilateral lending agreements.

2 See Houston and James (1996) and Hadlock and James (2002) for US evidence on the choice between

public and private securities and Barnes and Cahill (2005) and Antoniou et al. (2008) for UK firms. A small

number of studies have examined the choice of public bond financing relative to syndicated bank loans

(Altunbas et al., 2009) and syndicated loans relative to bilateral loans (Dennis and Mullineaux, 2000; Marsh,

2006; Sufi, 2007). Gomes and Phillips (2012) examine the choice between public and private markets,

and amongst straight debt, equity and convertible securities. However, their private category aggregates

syndicated and bilateral loans with Rule 144-A privately placed bonds. None of these studies examine

syndicated loans across a broad spectrum of debt choices.

C

2016 John Wiley & Sons Ltd

THE CHOICE OF DEBT SOURCE BY UK FIRMS 731

Finally, we extend the literature by examining the impact of the global financial

crisis on the determinants of a firm’s borrowing source. Our sample covers the period

of credit expansion prior to 2008 and the reported collapse in the availability of bank

financing worldwide. Given the expected reduction in bank lending surrounding this

period we examine whether the financial crisis had a direct impact on balance sheet

debt source and the choice of marginal borrowing source. We also consider whether

and how the determinants of borrowing source have changed around the financial

crisis.

When choosing amongst debt securities, we find that the largest debt issues are

loans syndicated across a number of lending banks. However, our results also show

that syndicated and bilateral loans have similar maturities. Public bonds and non-bank

private debt have longer maturities. We find that the maturity of a firm’s existing debt

is closely related to its marginal financing choices as firms with a higher proportion of

existing long-term debt are more likely to borrow in bond markets in their incremental

financing decisions.

Consistent with Diamond’s (1991) theoretical predictions, we find that firm size

and collateral are the most important firm characteristics in determining debt source.

Larger firms are more likely to borrow in public debt markets and firms with little

collateral are screened out of traditional sources and forced to borrow in non-bank

private markets. Our findings on the importance of fixed assets as collateral also

support theories of debt source based on efficiency of liquidation where firms with

difficult to value intangible assets are more likely to choose private debt sources. We

find other firm characteristics affect existing debt source or incremental financing

choices, but they are not consistent across our results.

Reputation effects in credit markets are also evident in the impact of a firm’s bor-

rowing history on marginal financing choice. We find that firms that are characterized

as bank dependent are significantly less likely to borrow in public debt markets relative

to other financing sources. Prior to the global financial crisis we find that syndicated

bank loans act as an intermediate step to public debt markets for bank dependent

firms, but this effect is not present in the post-crisis period where bilateral loans

increased in popularity for our sample of large firms.

We find that the financial crisis has impacted on the determinants of borrowing

source. Firm size and age, which are associated with borrower reputation and the

need for ease of renegotiation of lending agreements, are important determinants

of borrowing source in the post-crisis period, but were unrelated to borrowing source

pre-crisis. Moreover, firms dependent on banks for existing debt finance are restricted

to bank loans for incremental debt financing decisions post-crisis, which is not the case

in the pre-crisis period. This supports Diamond’s (1991) theory of borrower reputation

leading to firms repeatedly borrowing from similar lending sources, but suggests that

bond market lenders were more relaxed in their lending decisions in the lead up to

the global financial crisis.

The remainder of this paper is structured as follows. Section 2 provides a discussion

of the sample data. Our results for existing debt financing are in section 3. Section 4

reports our findings for incremental debt financing decisions. Section 5 examines the

impact of the financial crisis on firm’s debt sourcing decisions. Section 6 discusses

further analysis to confirm the robustness of our core findings. Finally section 7

provides a summary and concludes.

C

2016 John Wiley & Sons Ltd