Why Do Firms in Customer–Supplier Relationships Hold More Cash?

• Published date: 01 December 2015
• Author: Kee‐Hong Bae,Jin Wang
• DOI: http://doi.org/10.1111/irfi.12058
• Date: 01 December 2015

Why Do Firms in

Customer–Supplier Relationships

Hold More Cash?*

KEE-HONG BAE†AND JIN WANG‡

†Schulich School of Business, York University, North York, ON, Canada, and

‡School of Business and Economics, Wilfrid Laurier University, Waterloo, ON,

Canada

ABSTRACT

A f‌irm is in customer–supplier relationships when its business depends on a

small number of major customers/suppliers. In this paper, we provide evi-

dence that relationship-specif‌ic investments undertaken by f‌irms in

customer–supplier relationships are associated with high cash holdings in

these f‌irms. The evidence is consistent with the prediction of Titman’s

stakeholder theory that a f‌irm relying on relationship-specif‌ic investments

maintains a high cash reserve as a cushion to sustain its relationship-specif‌ic

investments when negative shocks occur. Our f‌indings suggest that

relationship-specif‌ic investments are important determinants of the precau-

tionary motive to hold cash.

I. INTRODUCTION

A f‌irm is in customer–supplier relationships when its business depends on a

small number of major customers/suppliers. In this paper, we provide evidence

that relationship-specif‌ic investments undertaken by f‌irms in customer–

supplier relationships lead to more cash holdings in these f‌irms.

Titman (1984) argues that a f‌irm in customer–supplier relationships often

undertakes relationship-specif‌ic investments and thus faces high liquidation

costs (Titman 1984). The concern is that the f‌irm may have to liquidate

relationship-specif‌ic assets when adverse shocks occur and access to the capital

markets is too costly. Thus, the concern about the possibility of liquidation is

likely to strengthen the f‌irm’s precautionary motive for holding cash. By main-

taining a high level of cash, the f‌irm can reduce the liquidation risk.

We test several implications of the stakeholder theory on cash-holding

motives using a sample of 116,041 f‌irm-year observations covering the period

* Bae gratefully acknowledges research support from the Schulich School of Business at York

University and the Social Science and Humanities Research Council of Canada. Wang gratefully

acknowledges research support from Wilfrid Laurier University and the Social Science and

Humanities Research Council of Canada. All errors are our own.

bs_bs_banner

International Review of Finance, 15:4, 2015: pp. 489–520

DOI: 10.1111/irf‌i.12058

© 2015 International Review of Finance Ltd. 2015

1982–2006. In our empirical tests, we use the ratio of sales to principal custom-

ers over total sales as a proxy for the importance of customer–supplier relation-

ships to a f‌irm’s business and R&D intensity as the measure of relationship-

specif‌ic investments.

We f‌irst present the evidence that f‌irms in customer–supplier relationships

hold more cash. For instance, in our basic regression model, we f‌ind that cash

holdings increase by 15.3% of the sample mean when a f‌irm’s status changes

from average f‌irm without principal customers to average f‌irm with principal

customers. We note that the theory of Titman (1984) links the importance of

relationship-specif‌ic investments with f‌irms’ cash-holding motives, suggesting

that the positive relation between customer–supplier relationships and cash

holdings arises because f‌irms in customer–supplier relationships make more

relationship-specif‌ic investments. We investigate this prediction. Consistent

with the view that relationship-specif‌ic investments drive the positive relation

between customer–supplier relationships and cash holdings, we f‌ind that the

extent of customer–supplier relationships is signif‌icantly related to both the

extent of relationship-specif‌ic investments and the level of cash holdings. More

importantly, we f‌ind that the positive relation between customer–supplier rela-

tionships and cash holdings is more pronounced for f‌irms with higher

relationship-specif‌ic investments than for f‌irms with lower relationship-specif‌ic

investments, which is precisely the result one would expect if relationship-

specif‌ic investments are the main driver of the positive relation.

After presenting evidence of the positive relation between customer–supplier

relationships with relationship-specif‌ic investments and cash holdings, we

attempt to identify the mechanism through which such a positive relation

occurs. Titman’s argument implies that f‌irms in customer–supplier relationships

maintain high cash reserves as a cushion to sustain their relationship-specif‌ic

investments. A f‌irm that is close to default is less able to cope with a liquidity

shortage by raising cash from the capital markets and thus is more likely to

liquidate its assets. Thus, if the positive relation between customer–supplier

relationships with relationship-specif‌ic investments and cash holdings is driven

by f‌irms that hold more cash to mitigate the high f‌inancial distress costs

associated with relationship-specif‌ic investments, then the positive relation

should be more pronounced for f‌irms that face a higher risk of bankruptcy. We

use Altman’s Z-score and credit ratings as proxies for the probability of bank-

ruptcy. Consistent with the argument of Titman (1984), we f‌ind that the

positive relation between a f‌irm’s dependence on customer–supplier relation-

ships and cash holdings exists mostly when a f‌irm undertakes relationship-

specif‌ic investments and faces a higher risk of bankruptcy at the same time.

Following Banerjee et al. (2008), we also examine the effect of government

customers and nongovernment customers on cash holdings separately. Govern-

ment customers are often not prof‌it driven and therefore may be less concerned

about their suppliers’ liquidity shortage. In fact, government customers may

even purchase from distressed suppliers to save jobs. The f‌inancial distress costs

are thus lower when a government agency is a major customer, suggesting that

International Review of Finance

490 © 2015 International Review of Finance Ltd. 2015

the positive relation between customer–supplier relationships and cash hold-

ings should be weaker or nonexistent. We f‌ind that the positive relation

between customer–supplier relationships and cash holdings indeed exists only

for nongovernment customers and not for government customers. Further-

more, the positive relation between the dependence on nongovernment cus-

tomers and cash holdings is more pronounced for f‌irms associated with higher

extent of relationship-specif‌ic investments.

Finally, we conduct a number of additional tests to verify the robustness of

our f‌indings. First, we examine how f‌irms save cash around the year when they

enter a customer–supplier relationship. We f‌ind that f‌irms with high extent

of relationship-specif‌ic investments tend to save more cash either out of cash

f‌lows or from reductions in capital expenditures. To the extent that entering

customer–supplier relationships is exogenous to the f‌irm’s cash holdings, the

results mitigate the concern over the endogenous relation between customer–

supplier relationships and cash holdings. Second, we use a dummy for the

supplier f‌irm’s patents citing its customer f‌irms’ patents as a relationship-level

indicator of undertaking relationship-specif‌ic investments. We again f‌ind that

the positive relation between customer–supplier relationships and cash hold-

ings is more pronounced when relationship-specif‌ic investments exist. Finally,

we examine how customer f‌irms’ business risk, measured by the weighted

average of customers’ cash f‌low volatility, affects the positive relation between

customer–supplier relationships and cash holdings in supplier f‌irms. We f‌ind

that such a relation is stronger when customer f‌irms are faced with greater

business risk, suggesting that concern over customers’ business prospects and

hence the value of relationship-specif‌ic investments drives the positive relation.

The main contribution of our paper is that we present evidence that a f‌irm’s

relationships with nonf‌inancial stakeholders affect its cash holdings. Our study

is closely related to the literature on how f‌irms’ dependence on nonf‌inancial

stakeholders affects f‌irms’ f‌inancing decisions (Titman 1984; Titman and

Wessels 1988; Maksimovic and Titman 1991; Kale and Shahrur 2007; Banerjee

et al. 2008; Bae et al. 2011; Li and Tang 2015). More broadly, our study contrib-

utes to the literature on how the interaction between f‌irms and nonf‌inancial

stakeholders is related to a wide spectrum of corporate strategies, such as the

design of CEO compensation (Arora and Alam 2005), information disclosure

(Almazan et al. 2006), earnings management (Raman and Shahrur 2008),

payout policy (Wang 2012), supply-chain management (Cen et al. 2014).

More broadly, our paper is also related to the recent literature on how a f‌irm’s

customer-base concentration affects its performance and cost of capital.

Patatoukas (2012) documents a positive association between customer-base

concentration and return on equity, while Irvine et al. (2013) f‌ind opposite

results for younger, less prof‌itable f‌irms where customer concentration impairs

prof‌itability and increases distress risk. Dhaliwal et al. (2015) f‌ind that concen-

trated customer base increases a supplier’s risk and results in a higher cost of

equity. Cen, Dasgupta, Elkamhi, and Pungaliya (2015) f‌ind that f‌irms relying

on principal customers experience relatively low loan spreads, which they

Customer–Supplier Relationships and Cash Holdings

491© 2015 International Review of Finance Ltd. 2015