University of Groningen Trade credit in the rice market of ... · Dao Thi Ba ), my aunties (Y Uong,...
Transcript of University of Groningen Trade credit in the rice market of ... · Dao Thi Ba ), my aunties (Y Uong,...
University of Groningen
Trade credit in the rice market of the Mekong Delta in VietnamNguyen, Lam Thu Uyen
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2011
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Nguyen, L. T. U. (2011). Trade credit in the rice market of the Mekong Delta in Vietnam. University ofGroningen, SOM research school.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 06-11-2020
Nguyen Lam Thu Uyen
Trade Credit in the Rice Market of
the Mekong Delta in Vietnam
Theses in Economics and Business
ii
Publisher: University of Groningen, Groningen, The Netherlands
Printed by: Ipskamp Drukkers B.V., The Netherlands
ISBN: 978-90-367-4863-6
978-90-367-4864-3
©2011 Nguyen Lam Thu Uyen
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system of any nature, or transmitted in any form or by any means,
electronic, mechanical, now know or hereafter invented, including
photocopying or recording without prior written permission of the
publisher.
Trade Credit in the Rice Market
of the Mekong Delta in Vietnam
Proefschrift
ter verkrijging van het doctoraat in de
Economie en Bedrijfskunde
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. E. Sterken,
in het openbaar te verdedigen op
donderdag 21 april 2011
om 13.15 uur
door
Nguyen Lam Thu Uyen
geboren op 17 september 1978
te Soc Trang, Vietnam
iv
Promotores: Prof. dr. B.W. Lensink
Prof. dr. C.L.M. Hermes
Copromotor: Dr. C.H.M. Lutz
Beoordelingscommissie : Prof. dr. Christopher Woodruff
Prof. dr. Arjun. Bedi
Prof. dr. J. Van De Meer- Kooistra
Acknowledgements
This dissertation would not have been possible without the help and support I have
received from colleagues, friends and family. In this section, I express my sincere
gratitude to them.
I would like to express my deepest gratitude to my supervisors, Prof.
Robert Lensink, Prof. Niels Hermes and Dr. Clemens Lutz. As an academic
supervisor of NPT project - a corporation between Faculty of Economics and
Business, University of Groningen and School of Economics and Business
Administration (SEBA), Cantho University- Robert offered me the chance to write
this PhD dissertation. He was also very helpful with his useful feedback and
support while I was working on this project. My deepest gratitude also goes to
Professor Niels Hermes. Niels helped me to find joy in doing research into trade
credit issues since I worked on my master thesis, and I could not have wished for a
better coach to guide me through the process of writing a dissertation. His
accessibility, enthusiasm and open-mindedness made it a real pleasure to work
with him. In addition, I am very grateful to Dr. Clemens Lutz. With a lot of effort,
he provided me invaluable assistance and support during the five years it took me
to write this dissertation.
Furthermore, I greatly appreciate the efforts and useful comments of the
members of the reading committee consisting of Prof. Christopher Woodruff, Prof.
Arjun Bedi and Prof. Van De Meer- Kooistra.
There are so many people who contributed by providing me both
emotional and practical support. I greatly appreciated the assistance from several
members of the International Relations Office, especially Anita Veltmaat, Gonny
Lakerveld, Wiebe Zijlstra, Erik Haarbrink. My thanks also go to all staff members
of SOM, the Center for Development Studies for the arrangements during my
vi
study and especially to Mr. Arthur De Boer for the instructions in the layout of this
book. I would like to express my deep gratitude to the SEBA staffs that arranged
necessary conditions for me to complete this dissertation i.e. the official procedures
and data collection (Dr. Mai Van Nam, Mr. Trung Tin, and Ms. Truc. Lien, Ms.
Linh Giang). Also, my great thanks goes to the staffs of statistic bureaus in
TienGiang, Angiang, VinhLong, CanTho, HauGiang and SocTrang provinces for
providing the opportunity to conduct the large number of enterprise interviews in
2007.
Next, I would like to thank my colleagues and friends providing me with
both emotional and practical support. My thanks go to Thong, Tra and Scott for
spending time discussing on trade credit issue. In addition, my special thanks go to
Tu for being my paranymphs and Xuan for her great help while I was in hospital in
2007. I thank Aljar, Binh, Tu, Dut, Khuong, Tran, Tri, Kadek, Ilko, chi Phuong and
other friends who have supported me but it is not possible to mention each one
individually.
I want to warmly acknowledge my parents in law (Nguyen Van Lanh and
Dao Thi Ba), my aunties (Y Uong, Y La, K. Yin), my uncles (Tia Xiem and Cu
Phen), my cousins (My Linh, My Phuong, My Tran, Kathy, Stijn and Kristiof) for
their emotional support. During the 5 years, they made my life in Groningen easy
and enjoyable.
To end with, I would like to express my deepest thankfulness to my
parents (Nguyen Tan Nhan and Lam Thi Lien), my husband (Nguyen Minh Phung)
and my sister (Nguyen Lam Phuong Thao and Huynh Nguyen Huy Hoang) for their
constant love, support and encouragement. By always standing behind me,
providing both practical and mental support, they have considerably contributed to
the completion of this thesis.
Uyen
Groningen, March 2011
Table of Contents
Chapter 1 Introduction 1
1.1 Introduction 1
1.2 Aim of the Study 4
1.3 Structure of the Thesis and Summary of the Main Findings 6
1.4 Limitations and Suggestions for Further Research 10
Chapter 2 Trade Credit Supply and Firm Sales Growth 13
2.1 Introduction 13
2.2 Review of Literature 14
2.2.1 Literature on Firm Growth 14
2.2.2 Trade Credit in Marketing Theory 22
2.3 The Data, Empirical Strategy 25
2.3.1 The Sample 25
2.3.2 Empirical Strategy 27
2.3.3 Empirical Measures 28
2.4 Empirical Results 30
2.4.1 Trade Credit Provision 30
2.5 Conclusions 34
Chapter 3 A Survey in the Rice Market of the Mekong Delta37
ii
3.1 Introduction 37
3.2 A Brief Description of the Rice Market of the Mekong Delta 38
3.2.1 Overview of the Paddy/Rice Production 38
3.2.2 Characteristics of Rice Traders in the Mekong Delta 40
3.3 Description of Data Collection 43
3.3.1 The Pilot Survey 43
3.3.2 The Official Survey 44
3.4 Descriptive Statistics of the Sample 47
3.4.1 Overview of the Sample 47
3.4.2 Trade Credit Provision 49
3.4.3 Firm Competitiveness 52
3.4.4 Financial Constraints 55
3.5 Summary 57
Chapter 4 Trade Credit Supply and Firm Competitiveness 59
4.1 Introduction 59
4.2 Trade Credit Supply and Firm Competitiveness: A Survey 61
4.3 The Vietnamese Rice Sector 65
The Survey 66
4.4 Methodology and Data 68
4.4.1 The Regression Model 68
Type of Firm Dummy Variables 72
4.4.2 Descriptive Statistics 73
4.5 Empirical Results and Discussion 77
4.5.1 Firm Competitiveness and Trade Credit Supply 78
4.5.2 Customer Bargaining Power and The Role of Perceived
Competition 80
iii
4.6 Summary and Conclusions 82
Chapter 5 Trade Credit Supply and Customer Characteristics
89
5.1 Introduction 89
5.2 Related Literature 90
5.3 Hypothesis and Methodology 93
5.3.1 Hypothesis Development 93
5.3.2 Empirical Methodology 94
5.3.3 Descriptive Statistics 99
5.4 Estimation Results and Discussion 104
5.4.1 The Impacts of Customers’ Characteristics 105
5.4.2 The Effects of Market Characteristics 106
5.4.3 The Effects of Supplier Characteristics 107
5.5 Conclusion 112
Chapter 6 Trade Credit Supply in Different Market Segments
113
6.1 Introduction 113
6.2 Empirical Methodology 114
6.3 Why May Trade Credit Differ Across Segments? 118
6.3.1 Market Context Characteristics and Trade Credit Supply 118
6.3.2 Structural Differences Across Market Segments 119
Market Structure Across Market Segments 121
Access to External Finance 123
Customer Characteristics 124
6.4 Estimation Results 126
6.4.1 Differences Between Market Segments: General Overview 126
iv
6.4.2 A Close Look at Each Segment 127
Wholesale-Miller Segment 127
Wholesaler Segment 128
Miller Segment 130
Retailer Segment 130
6.5 Summary of the Results 131
v
List of Tables
Table 2.1: Summary of studies on determinants of firm growth 18
Table 2.2: Sample firms by industry 26
Table 2.3: Descriptive statistics 29
Table 2.4: Results of the fixed-effect models 32
Table 2.5: Results from the one step system GMM estimator 33
Table 2.6: Results from the random effect and the OLS
regressions 34
Table 3.1: Distribution of rice firms according to main business
activities and firm size (measured by total assets) 48
Table 3.2: Size and age of the surveyed rice firms 48
Table 3.3: Descriptive statistics of profitable variables 49
Table 3.4: Distribution of the surveyed rice firms providing credit 50
Table 3.5: Experience with defaulting by clients 51
Table 3.6 Trade credit received 51
Table 3.7: Firm competiveness measures 53
Table 3.8: Mean and median of competitive measures across
different market segments 55
Table 3.9: The importance of financial constraints 56
Table 3.10: Firms using bank loans in 2006 57
Table 4.1: Descriptive statistics 76
Table 4.2: Correlation matrix 77
Table 4.3: Tobit regressions of the relationship between trade
credit supply and competitiveness 84
Table 4.4: Tobit regressions of the relationship between trade
credit supply and competitiveness 86
Table 5.1: Variable definition 101
Table 5.2: Summary statistic of the sample 103
Table 5.3: Correlation matrix 104
Table 5.4: The impact of customer characteristics on trade
vi
credit provision 110
Table 6.1: Mean Differences across different Market
Segments 120
Table 6.2: Types of customers 124
Table 6.3: Determinants of trade credit supply in the four
market segments: w-millers, wholesalers, millers and
retailers 134
List of Figures
Figure 3.1: Marketing channel of the rice market in The
Mekong Delta 37
Chapter 1
Introduction
1.1 Introduction
Since Vietnam conducted economic reforms in 1986, it has transformed itself from
a centrally planned economy to a market-oriented economy. One of the most
important features of the economic reform was the legalization of the private
sector. Consequently, the private sector has grown dramatically during the last 20
years. In particular, the number of private firms has increased rapidly, and most of
them are small and medium-sized enterprises (SMEs). For example, Nguyen et al
(2008) showed that 99 per cent of private firms in Vietnam in 2004 were SMEs.
These SMEs face many challenges, however. The major barrier is the limited
access to capital for future growth (Rand, 2007; Nguyen and Ramachandran, 2006;
Sakai and Takada, 2000).
The financial sector in Vietnam is characterized by an immature financial
market and a banking system in which state-owned commercial banks still play a
dominant role. The official Vietnamese stock market was launched in July 2001
with only two individual stocks on the first trading day. The list of companies
increased to 32 enterprises at the end of 2005 and 620 enterprises in 2010. Yet,
according to HCMC Institute for Development Studies, in 2010, Vietnam has
500,000 enterprises of which SMEs account for 97 per cent (Saigon Time 10,
2010). SMEs in Vietnam have not had access to the stock market, and therefore
they had to rely on the banking system for their capital needs. The four main state-
owned commercial banks accounted for 73 per cent of total bank assets and
provided 72 per cent of total loans in the economy in 2001. However, half of the
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
2
state-owned commercial bank loans (45 per cent) are channeled to state-owned
enterprises (SOEs) (Asian Development Bank, 2002). Consequently, SMEs, and
especially private SMEs, find it very difficult to access bank loans though bank
financing is among the most important sources of external financing for SMEs.
While facing financial constraints is an important problem of Vietnamese
firms, Vietnamese suppliers provide much trade credit to their trading partners.
Trade credit occurs when a supplier delivers the goods or services to a customer at
t=0 and allows the customer to delay payments for a certain period. Thus, the
customer pays for these goods or services at t=1. We observe that in general trade
credit provision is popular in Vietnam (McMillan and Woodruff, 1999), but also in
the case of the rice market of the Mekong Delta (Luu, 2003). Given the fact that
Vietnamese SMEs as well as rice firms have faced financial constraints, we find
Vietnam to be an interesting case to study in terms of the underlying motivations
for firms to supply trade credit.
Existing studies on trade credit supply concentrate on two different
views. According to the finance theory, suppliers know their trading partners better
than banks and other financial institutions. Therefore, the information asymmetry
between lenders and borrowers is less of an issue for suppliers than for banks and
financial institutions. As a result, firms facing bank-borrowing constraints tend to
turn to suppliers for credit. Accordingly, suppliers with good access to external
finance may be willing to supply credit.
The marketing theory of trade credit stresses the importance of trade
credit provision as a marketing device to support sales. This category of studies
particularly focuses on the impact of market characteristics, i.e. market competition
on trade credit supply. It argues that when firms have difficulties raising sales due
to fierce competition, they will supply trade credit. Several explanations for this
have been suggested. In fact, trade credit enables suppliers to allow their clients to
verify product quality before making any payments. Consequently, trade credit can
Chapter1: Introduction
3
work as a signal or guarantee of product quality. Also, by granting credit, suppliers
provide financial help to clients to finance their purchases. Indeed, trade credit is a
short-term finance source that can be used to attract clients, especially financially
constrained customers.
The literature that examines access to external finance as the main
determinant of trade credit supply is extensive. Yet, the number of empirical
studies investigating the role of trade credit supply as a marketing device is
relatively limited. In addition, recent studies show that the finance theory may not
provide a full explanation of trade credit supply. In fact, it is shown that small,
young and financially constrained firms seem to provide more credit, especially in
dealing with large clients (Van Horen, 2005, 2007). This poses an interesting
research question: what are the factors that drive firms to supply trade credit? To
contribute to an answer to the question, this dissertation concentrates on studying
the factors that determine trade credit supply for Vietnamese firms.
This dissertation will focus on a particular industry – the rice industry in the
Mekong Delta of Vietnam – for several reasons. First, rice is one of the main
commodities in Vietnam, especially in the Mekong Delta. In fact, during the last
decade Vietnam has been the second largest rice exporter in the world. In
particular, the Mekong Delta has been Vietnam’s rice bowl, producing about 50
per cent of the country’s total rice output and providing more than 90 per cent of
the volume of Vietnam’s rice export. Consequently, the rice market of the Mekong
Delta has been a large market that consists of different market segments. The four
main market segments of this particular market are wholesale-millers (w-millers),
millers, wholesalers and retailers. Wholesale-millers process paddy into rice and
sell rice as their final products, while millers only provide milling services and will
charge a milling fee. Wholesalers distribute milled rice from wholesale-millers to
retailers, who sell rice to the final consumer. According to Luu (2003), market
competition varies substantially across these four market segments. Thus, the rice
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
4
market is an interesting case for studying the impact of market competition on
trade credit provision, which is one of the main targets of this thesis.
Second, concentrating on one industry in one country allows us to measure
firm competitiveness and market competition more carefully. Third, the rice
market contains a few large rice exporters and state-owned enterprises, and a large
number of private firms that are mostly SMEs. As mentioned above, a shortage of
capital is one of the greatest difficulties Vietnamese SMEs meet. Rice firms,
however, provide a lot of credit to their trading partners (Luu, 2003). The rice
market is therefore interesting and suitable to investigate the factors that cause
firms to supply credit.
1.2 Aim of the Study
The ultimate objective of this thesis is to contribute to answering the following
research question: what are the factors that persuade firms to provide trade credit to
their clients? In order to answer this research question, the thesis will be examining
the following issues.
Trade Credit Supply and Sales Growth
According to the marketing theory, suppliers provide credit with the aim of
promoting sales. However, whether trade credit provision indeed enhances sales
growth is still unknown, since no empirical study has investigated the impact of
trade credit on sales growth.
Trade Credit Supply and Firm Competitiveness
We will be taking the marketing view to investigate the relation between market
competition and trade credit supply. It argues that firms facing fierce competition
Chapter1: Introduction
5
find it difficult to increase sales. Therefore, they will provide more trade credit to
support sales. The impact of market competition on trade credit provision,
however, is still an unresolved issue in the literature. Empirical studies on this
issue are rare and produce conflicting results. On the one hand, Fisman and Raturi,
(2004) find a positive effect of market competition on trade credit provision. On
the other hand, McMillan and Woodruff (1999) show a negative influence of
market competition on trade credit provision. These studies use similar approaches
and apply a cross-country and/or cross-industry analysis with a single or a dummy
variable measuring competition and/or market structures. Yet, it is difficult for a
dummy variable to capture the differences in the market structures and/or the
degree of market competition across different countries and/or industries. In
addition, these studies may not control for differences in product characteristics
and/or industry-specific characteristic that influence the use of trade credit
(Summer and Wilson, 2003; Giannetti et al, 2009). For example, Giannetti et al
(2009) find that service firms are more willing to offer credit than manufacturers.
With the aim of examining the relation between market competition and trade
credit supply, we focus on one industry in one country to eliminate various
products/industry characteristics that may affect trade credit provision. Therefore,
we are able to focus on the market structure in one industry as the major factor
influencing trade credit supply. At the same time, we plan to use several variables
to carefully measure different factors driving a firm’s competitiveness. This
approach is expected to provide us a better insight into the impact of market
competition on trade credit supply.
Trade Credit and Customer Characteristics
Trade credit may work as a marketing device to attract and keep customers to
generate sales (Long et al, 1993; Fisman and Raturi, 2004). Yet, suppliers do not
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
6
provide credit to all customers. For example, some studies suggest that firms are
motivated to give more credit to large clients with strong bargaining power (Van
Horen, 2007; Fabbri and Klapper, 2009). Other studies find that suppliers are more
willing to offer credit to credit-worthy customers (McMillan and Woodruff, 1999).
This poses an interesting question: what determines the amount of trade credit
supplied to customers? One of the objectives of this thesis is to study how
individual customer characteristics affect the amount of trade credit provided at the
transaction level. Using our own survey, we obtain details on specific trade
relations between suppliers and customers. Consequently, we are able to
investigate which characteristics provide customers with better access to trade
credit.
Trade Credit Supply and Market Segments
The discussion of the data in chapter 3 shows that trade credit supply varies
significantly across the main market segments; the substantial differences in trade
credit supply across market segments are confirmed in chapters 4 and 5. Therefore,
one of the targets of this project is to understand these differences across market
segments such as w-millers (processing firms), wholesalers, retailers, and millers
(milling-service firms).
1.3 Structure of the Thesis and Summary of the Main Findings1
Chapter 2 aims at examining whether the view of trade credit as a marketing device
is relevant to the case of Vietnamese firms. The chapter concentrates on
1 The thesis is a collection of papers. There is some overlap. Those who have read the previous
chapters can skip the section 4.3, section 5.2 and section 6.2 since these matters have also been dealt
with in chapter 3, section 4.2 and section 5.3 respectively.
Chapter1: Introduction
7
investigating whether trade credit supply enhances sales growth. It uses a panel of
Vietnamese listed firms in the country’s two stock markets: Ho Chi Minh Stock
Exchange and Hanoi Stock Exchange, for the period 2004-2007. Although several
theoretical studies suggest that trade credit functions as a marketing device to
support sales, our study is one of the first to empirically examine the impact of
trade credit supply on firm sales growth. The significant effect of trade credit
supply on sales growth does indeed reveal the necessity of investigating the role of
trade credit as a marketing device, which is one of the main issues of the empirical
studies presented in chapters 4, 5, and 6.
Chapter 3 discusses the rice market in the Mekong Delta. It describes the
survey used to collect the data for the empirical studies presented in chapters 4, 5,
and 6. Next, the chapter discusses the data and provides an in-depth overview of
the rice market of the Mekong Delta. This is important in terms of providing an
overall picture of the rice industry and insights into the causes and nature of the use
of trade credit by rice traders in the Mekong Delta.
The focus of chapter 4 is the relation between firm competitiveness and
trade credit supply. Existing empirical evidence on this issue is based on cross-
industry and/or cross-country analyses. We apply a different approach in this
thesis. We concentrate on one industry, namely the rice industry in the Mekong
Delta, to study the effect of firm competitiveness on trade credit supply. By
following this approach, we expect to remove the effect of the differences in
product characteristics, the various industries (Giannetti et al, 2009; Summer and
Wilson, 2003), and the differences in the development in financial markets banking
system (Fisman and Love, 2003), and legal system (Lu, 2009) on trade credit
supply in different countries. Indeed, these factors may influence trade credit
supply differently, and thus they influence the relation between trade credit supply
and firm competitiveness. Moreover, this approach allows us to apply Porter’s five
forces model, which suggests that firm competitiveness is influenced by five
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
8
factors in five distinct dimensions (Porter 1979, 1980, 1985, and 2008). We plan to
explicitly investigate the different dimensions that drive firm competitiveness and
their impact on the supply of trade credit. We concentrate on two different
dimensions of competitiveness: the competitive pressure from other suppliers
(rivals), and bargaining power vis-à-vis the customer.
We find strong evidence that competition causes suppliers to grant credit
to their trading partners to support sales and avoid customer switching. In addition,
concerning a firm’s bargaining power vis-à-vis clients, the results show that
suppliers also offer more credit in dealing with clients that have strong bargaining
power. The reason is that the threat of switching to another supplier is higher for a
client with strong bargaining power, and this provides suppliers with incentives to
grant credit to keep the customer. In fact, this chapter adds convincing evidence for
the relation between trade credit supply and market competition. It shows that
suppliers may be highly motivated to use credit to enhance their sales.
Chapter 5 uses transaction level data on suppliers and customers involved
in specific trade relations, and proceeds to identify the customer characteristics that
improve access to suppliers’ trade credit sources. Empirical analyses on this topic
are rare, and the available studies leave out important variables like customer
bargaining power and market competition. In chapter 5, we cover a broader range
of variables that reflect customer characteristics. We include several proxies
measuring customer creditworthiness, characteristics of trading relations and
market contexts. In line with our expectations, the study reports a larger amount of
credit offered to creditworthy clients. In addition, suppliers are very much inclined
to give more credit to large clients and/or clients with strong bargaining power.
The empirical evidence in this chapter suggests that suppliers use trade credit to
build trade relations with creditworthy clients and high potential customers, i.e.
large buyers. In doing so, suppliers expect to reduce the risk of client default and
increase their present and future sales.
Chapter1: Introduction
9
We observe significant results for the market segment dummy variables in
all scenarios in chapters 4 and 5. This evidence indicates that there are structural
differences in trade credit supply across the four main market segments. Therefore,
chapter 6 aims at understanding the differences in trade credit behavior across
different market segments. We investigate the differences in determinants of trade
credit supply across market segments. The empirical findings show that the market
segment context significantly influences trade credit supply. In the segments of the
wholesale-miller and wholesaler, the marketing role of trade credit is among the
main motivations driving suppliers to grant credit in these segments. Second, in the
market segment of retailers and millers, the set of variables representing the
marketing theory does not show significant results. Thus, the marketing role may
not be among the major factors driving trade credit supply in the miller and retailer
segments. Third, the information advantage theory of trade credit, which mentions
that suppliers grant credit since suppliers know their trading partners much better
than banks and financial institution, also appears to be a relevant determinant. Rice
firms appear to be willing to offer more credit to customers with whom they have
close relations because of better information. Yet, information advantage theory
appears to be much more relevant in the segment of millers, in which suppliers
select to grant credit to closely related customers. This is different from the other
segments, where traders are willing to grant credit to large customers/customers
with strong bargaining power to generate sales. Intuitively, we provide a number of
explanations for these results. First, the small value of transactions may decrease
the attractiveness of the trade credit option to clients. At the same time, the cost of
collecting information concerning clients may be too high compared to the value of
transactions. This discourages suppliers to employ credit as a marketing instrument
to support sales in these segments.
In general, this thesis has brought a better understanding of the role of
trade credit supply, especially concerning the relation between firm
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
10
competitiveness and trade credit supply on the one hand. On the other hand, this
project also provides fruitful information on the impact of customer characteristics
and market contexts on trade credit supply.
A remarkable result of our study is that even though we concentrate on one
industry, major differences in trade credit supply still exist across the four main
market segments. This study shows that trade credit supply in different markets
depends on different combinations of market context characteristics like the
customer bargaining power, the product, the main business activity, the banking
system, etc. One of our main findings is that the role of trade credit supply may
differ even among the different segments of one industry. Therefore, this study
implies that not only country and industry characteristics but also market segment
specificity should be taken into account when studying the role of trade credit.
1.4 Limitations and Suggestions for Further Research
This study is subject to several limitations. For example, the survey data gave us
numerous useful insights that may help improve our understanding of the role of
trade credit. However, we realize that data collected through surveys may be
biased. For example, a significant number of the participants in the interviews are
retailers, who may have a low average level of education. This may lead to a lower
reliability of their answers, since they may not understand all survey questions. Or
perhaps respondents have incentives to give incorrect answers. To deal with these
difficulties, in 2006 we conducted a pilot survey to test our questionnaires and
make them right for rice traders. In addition, we employed the staff of statistic
bureaus in each province, as they are well-trained interviewers. They were capable
and willing to further illustrate and clarify questions. Therefore, we think that we
have done everything we could to reduce biases to a minimum.
This thesis focuses on one industry in one country to gain an insight into the
impact of market structure on trade credit supply. It is interesting that the
Chapter1: Introduction
11
determinants of trade credit supply differ in different market contexts; we find that
the main determinants of trade credit differ throughout the four market segments.
Firms in the processing and wholesale sector appear to have stronger incentives to
use credit to generate sales than firms in the retail and service sector. This fact
offers directions for further research, investigating the role of trade credit as a
marketing tool to enhance sales growth in different sectors.
We also find strong evidence that competition and buyer bargaining power
are associated with trade credit supply. However, very little theory has been
developed on this issue. Another direction for further research may therefore be to
develop a theory to explain the impact of market structure, including competition
and buyer bargaining power, on trade credit supply.
Chapter 2
Trade Credit Supply and Firm Sales
Growth
2.1 Introduction
Trade credit has been used extensively both in developed and developing countries.
Several theoretical and empirical studies argue that suppliers employ trade credit as
a marketing device to generate sales. Yet, no study has investigated whether trade
credit can actually enhance firm sales. Therefore, this study aims at investigating if
trade credit can help to enhance sales growth. We employ a firm-level data set of
the listed firms at the Ho Chi Minh Stock Exchange and the Hanoi Stock Exchange
to examine the impact of trade credit provision on firm sales growth. In addition,
we investigate whether the impact of trade credit provision on firm sales varies
across different types of firms: i.e. young firms and old firms.
Vietnam is an interesting case to examine the impact of trade credit
provision on firm sales for several reasons. Vietnam is a transition economy with
an inefficient banking system and infant financial markets. Vietnamese enterprises
face several administrative difficulties in formal and informal credit markets
(Rand, 2004). According to Rand, 14 to 25 per cent of Vietnamese enterprises are
credit-constrained. These firms would increase their debt holdings 40 to 115 per
cent if borrowing constraints were relaxed. Yet, although facing financial
constraints, Vietnamese suppliers provide a lot of trade credit to their trading
partners (McMillan and Woodruff, 1999; Luu, 2003). In our sample, 97 per cent of
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
14
the firms provide credit to their customers. We find it interesting to study the
underlying motivations to supply credit in these cases.
This chapter analyzes the impact of trade credit provision on firm sales
growth. The chapter is organized as follows. Section 2.2 provides an overview of
the relevant literature. Section 2.3 continues with the data and empirical strategy,
while section 2.4 presents the discussion of the empirical results. Finally, section
2.5 concludes the chapter.
2.2 Review of Literature
2.2.1 Literature on Firm Growth
Much attention has been devoted to examining the determinants of firm
growth in the literature. Previous studies have focused primarily on initial firm-
specific conditions such as firm size and age, innovation, financial structure,
ownership structure, and human capital.
Concerning the relationship between firm size and firm growth, Gibrat’s
theory (1931) suggests that the expected growth rate of a given firm does not
depend on its size. The principle behind Gibrat’s “law of proportional effects” is
that on average a firm’s expected growth in the next period is proportional to its
current size, implying that the expected growth rate is independent from its size.
Yet, several subsequent empirical studies demonstrate a significant relation
between firm size and growth. Although some studies report positive impacts
(Prais, 1974; Singh and Whittington, 1975; Wijewardena and Tibbits, 1999), the
negative relationship between firm size and growth has been reported by most
recent studies. According to Beck et al (2006), small firms in countries with more
developed financial institutions grow faster in comparison with large firms. The
reason is that information asymmetry is an important problem for small firms with
potentially profitable growth opportunities that want access to finance. In countries
Chapter2: Trade Credit Supply and Firm Sales Growth
15
with developed credit markets, this problem is less serious and financial constraints
are far more relaxed, so that small firms can achieve their growth potential. For
example, Dunne and Hughes (1994) find a negative impact of size on growth for
the case of listed UK manufacturing firms; Bottazzi and Secchi (2003) use data of
the Italian manufacturing firms and show a negative influence of size on firm
growth. Goddard et al (2002) use data from quoted Japanese manufacturing firms
and show a similar result: a negative effect. Yashuda (2005) also points at the
negative effect of size on firm growth for Japanese manufacturing firms.
McPherson (1996) finds that small Southern African firms grow faster. Calvo
(2006) shows the same results for Spanish manufacturing firms. Brown et al (2005)
find that small firms have higher growth rates in Romania.
The effect of a firm’s age on growth has also been discussed widely.
According to Jovanovic’s model (1982), firms learn about their efficiency through
production activities. Consequently, if output is a decreasing convex function of
managerial inefficiency, Jovanovic concludes that young firms grow faster than
older firms. The negative and significant relation between firm age and growth is
also found to be prominent in empirical studies: (Evans (1987a, b) uses data from
US manufacturing firms and shows that young firms grow faster; Variyam and
Kraybill (1992) point at a negative influence of age on firm growth in the case of
US manufacturing and services firms; Geroski and Gugrle (2004) show a negative
correlation between firm age and growth for large European companies; Brown et
al (2005) employ data from small businesses in Romania and find a negative effect
of age on firm growth; Wijewardena and Tibbits (1999) show similar results
regarding the negative impact of age on firm growth for small Australian
companies; Yasuda (2005) finds that young firms experience higher growth rates
than older firms in the case of Japanese manufacturers.
Several theoretical papers stress the important role of innovation for firms
that want to expand sales and market shares. For example, Hey and Kamshad
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
16
(1994) stress that investment in product innovation is the main strategy for
expansion – the reason being that investment in innovation (R&D) may lead to
technical changes, causing (partly) economic growth (Griliches and Lichtenberg,
1984; Solow, 1957). However, empirical studies find it hard to prove the impact of
innovation on sales growth. For example, Freel (2000) uses firm-level data from
228 US manufacturing firms and does not find that innovators are experiencing
higher growth rates. Similarly, Bottazzi et al (2001) do not find a significant effect
of innovation and technology on sales growth. Only a few studies manage to
demonstrate a positive relationship between innovation and sales growth. Geroski
and Toker (1996) employ firm-level data from UK firms and show that innovation
has a significant positive impact on sale growth. Another notable study that
stresses the positive effects of innovation on sales growth is Roper (1997).
Next, turning to the financial structure, several studies focus on the
relationship between firm expansion and financial performance. These studies
stress that credit constraint problems negatively affect firm growth. Indeed,
financial constraints may hinder investments in potentially profitable projects and
influence firm growth negatively. For example, Brown, Earle and Lup (2005) show
that credit constraint relaxation (measured by access to bank loans) has a positive
effect on firm sales growth. Bottazzi et al (2008) also find significant effects of
financial performance on firm growth even though the magnitude of the coefficient
is quite small. Musso and Schiavo (2008) prove that access to external finance
funds has a positive effect on firm growth in term of sales and employment
numbers.
Ownership structure is also argued to affect growth rates. According to
Ehrlich et al. (1994), owners invest capital in a firm to maximize the value of the
firm. However, managers may pursue private objectives. The degree to which
manager’s private objectives deviate from the firm profit maximization objective
may be related to firm ownership structure. In particular, this study shows that state
Chapter2: Trade Credit Supply and Firm Sales Growth
17
ownership can effect firm growth negatively in the long run. When firms are state-
owned, managers may pursue more easily their own goals rather than the
objectives of the state, since the government may not be able to monitor managers
properly anyway. Backer and Sleuwaegen (2003) indicate that foreign ownership is
an additional source of firm heterogeneity affecting firm productivity. Beck et al.
(2005) find that foreign-owned firms experience faster growth while state-owned
companies seem to experience lower growth rates.
Human capital also appears to be a factor that contributes to firm growth.
Education and training can influence knowledge, skills, motivation, and the ability
to provide suggestions to short- and long-term business plans that affect firm
productivity growth. In fact, Almus (2002) shows the positive effects of human
capital on growth in the case of German companies. McPherson (1996) also
emphasizes the positive and significant influences of human capital level on the
growth of micro and small businesses in five Southern African countries.
Final determinants for firm growth as proposed by some studies refer to the
manager: the manager’s age, sex, and competencies. For example, Ibrahim and
Goodwin (1986) show that a manager’s managerial skills are key factors that
contribute to the growth of small firms. Similar findings were described by Huck
and McEwen (1991): they identify entrepreneurs’ competencies in management,
planning and budgeting, and marketing as most crucial to the successful operation
of a small business.
Although several studies investigate the determinants of firm sales growth,
not one investigates the impact of trade credit provision on firm sales growth rates.
This is remarkable since several theoretical studies argue that trade credit actually
works as a marketing device to promote sales for suppliers.
18
Tab
le 2
.1:
Sum
mar
y o
f st
udie
s on d
eter
min
ants
of
firm
gro
wth
No.
Au
thor
V
ari
ab
les
C
ou
ntr
y/d
ata
O
utc
om
es
1.
Alm
us
(2002)
Gro
wth
rat
es i
n
emplo
ym
ent
Ger
man
fir
ms
Posi
tive
effe
ct o
f hum
an c
apit
al o
n
gro
wth
2
Bott
azzi
et
al (
2001)
Gro
wth
in s
ales
T
he
worl
d t
op 1
50
phar
mac
euti
cal
firm
s
No s
ignif
ican
t ef
fect
of
innovat
ion o
n
firm
gro
wth
3.
Bott
azzi
and S
ecch
i
(2003)
Gro
wth
in n
et a
sset
s T
he
Ital
ian m
anufa
cturi
ng
firm
s
Neg
ativ
e im
pac
t of
firm
siz
e on g
row
th
4
Bott
azzi
et
al (
2006)
Gro
wth
in t
ota
l sa
les
A p
anel
of
Ital
ian
man
ufa
cturi
ng a
nd s
ervic
e
firm
s.
Posi
tive
effe
ct o
f fi
nan
cial
per
form
ance
on f
irm
gro
wth
5
Bro
wn e
t al
(2005)
Gro
wth
in e
mplo
ym
ent
and s
ales
Sm
all
busi
nes
s in
Rom
ania
N
egat
ive
impac
t of
firm
siz
e on g
row
th
Neg
ativ
e im
pac
t of
firm
age
on g
row
th
Posi
tive
effe
ct o
f cr
edit
const
rain
ts
rela
xin
g o
n f
irm
gro
wth
.
1
9
No.
Au
thor
V
ari
ab
les
C
ou
ntr
y/d
ata
O
utc
om
es
6
Cal
vo (
2006)
Gro
wth
in e
mplo
ym
ent
Span
ish m
anufa
cturi
ng f
irm
s
in t
he
per
iod 1
990-2
000
Neg
ativ
e im
pac
t of
firm
siz
e on g
row
th
7
De
Bac
ker
and
Sle
uw
aegen
(2003)
Gro
wth
in l
abor
pro
duct
ivit
y t
hat
is
def
ined
as
the
rati
o o
f
val
ue
added
to t
he
num
ber
of
emplo
yee
s
22.4
52 B
elgia
n
man
ufa
cturi
ng f
irm
s fo
r th
e
per
iod 1
990-1
995
Posi
tive
effe
ct o
f fo
reig
n o
wner
ship
on
gro
wth
.
8
Dunne
and H
ughes
(1994)
Annual
gro
wth
in n
et
asse
ts
The
UK
fir
ms
in t
he
per
iod
1975-1
985
Neg
ativ
e im
pac
t of
firm
siz
e on g
row
th
9
Ehrl
ich e
t al
(1994)
Gro
wth
in t
ota
l to
n-
kil
om
eter
per
form
ed
by t
he
airl
ines
23 i
nte
rnat
ional
air
lines
.
Neg
ativ
e im
pac
t of
stat
e-o
wned
shar
es
on f
irm
gro
wth
.
10
Evan
(1987 a
, b)
Gro
wth
in e
mplo
ym
ent
U
S m
anufa
cturi
ng f
irm
s N
egat
ive
impac
t of
firm
age
on g
row
th
20
No.
Au
thor
V
ari
ab
les
C
ou
ntr
y/d
ata
O
utc
om
es
11
Fre
el (
2000)
Gro
wth
in s
ales
and
emplo
ym
ent
228 U
S m
anufa
cturi
ng f
irm
s N
o s
ignif
ican
t ef
fect
of
innovat
ion o
n
firm
gro
wth
12
Ger
osk
i an
d T
oker
(1996)
Gro
wth
in t
urn
over
The
UK
fir
ms
Posi
tive
effe
ct o
f in
novat
ion o
n f
irm
gro
wth
13
Ger
osk
i an
d G
ugle
r
(2004)
The
log o
f th
e num
ber
of
emplo
yee
s
Euro
pea
n c
om
pan
ies
firm
wit
h m
ore
100 e
mplo
yee
s
Neg
ativ
e im
pac
t of
firm
age
on g
row
th
for
the
case
14
Goddar
d e
t al
(2002)
Annual
gro
wth
rat
es i
n
tota
l as
sets
Japan
ese
man
ufa
cturi
ng f
irm
s N
egat
ive
impac
t of
firm
siz
e on g
row
th
15
McP
her
son (
1996)
Gro
wth
in t
he
num
ber
of
work
ers
Mic
ro a
nd s
mal
l en
terp
rise
s
in s
outh
ern A
fric
a
Neg
ativ
e im
pac
t of
firm
siz
e on g
row
th
Posi
tive
effe
ct o
f hum
an c
apit
al o
n
gro
wth
16
Sin
gh a
nd
Whit
tingto
n (
1975)
Annual
gro
wth
in n
et
asse
ts
Quote
d c
om
pan
ies
in t
he
UK
P
osi
tive
impac
t of
firm
siz
e on g
row
th
2
1
No.
Au
thor
V
ari
ab
les
C
ou
ntr
y/d
ata
O
utc
om
es
17
Var
iyam
and
Kra
ybil
l (1
992)
Gro
wth
in e
mplo
ym
ent
US
man
ufa
cturi
ng a
nd
serv
ice
firm
s
Neg
ativ
e im
pac
t of
firm
age
on g
row
th
18
Wij
ewar
den
a an
d
Tib
bit
s (1
999)
Gro
wth
in t
ota
l as
sets
S
mal
l A
ust
rali
an c
om
pan
ies
Neg
ativ
e im
pac
t of
firm
age
on g
row
th
19
Yas
huda
(2005)
Gro
wth
rat
es i
n t
ota
l
asse
ts
Japan
ese
man
ufa
cturi
ng f
irm
s N
egat
ive
impac
t of
firm
siz
e on g
row
th
Neg
ativ
e im
pac
t of
firm
age
on g
row
th
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
22
2.2.2 Trade Credit in Marketing Theory
Trade credit is used extensively in trade transactions among firms all over the
world. Trade credit occurs when a supplier delivers the goods or services to a
customer at t=0 and allows the customer to delay payments for the goods for a
certain period. Thus, the customer pays for these goods or services at t=1. A
common idea that runs through several studies on trade credit supply is that the
main purpose of granting credit is to increase sales (Nadiri, 1969; Blazenko and
Vandezande, 2003; Long et al., 1993).
Theoretical Studies
Nadiri (1969) was the first to develop a neoclassical model in which firms consider
the cost of providing trade credit as a type of selling costs to influence product
demand. In his theoretical framework, firms are assumed to select output prices,
the costs of providing trade credit (opportunity costs of investing in trade credit)
and the volume of output in order to maximize net profit. The outcome of this
model is that trade credit provision is positively related to the level of sales as well
as profit margins. This study shows that firms use trade credit as a type of selling
instrument to increase sales and profitability. Therefore, firms that spend more on
trade credit provision will gain a higher level of sales and profitability.
Sharing this view, Blazenko and Vandezande (2003) extend Nadiri’s
theory by including the cost of bad debts caused by trade credit provision. In
particular, they assume that the higher the prices of products, the larger the amount
of trade credit needed, the more likely it is that the trader is not paid by clients –
and therefore the higher the costs of bad debts caused by losses from client default.
In this theoretical framework, firms seek to maximize profit margins while the
marginal cost increases. This model arrives at a relation between trade credit
Chapter2: Trade Credit Supply and Firm Sales Growth
23
provision and profit margins that can be either positive or negative, depending on
the price elasticity of product demand. For example, when the marginal cost
increases, firms should increase selling prices; thus, firms with very high price
elasticity of product demand will experience a large drop in sales. Consequently,
firms with a high price elasticity of products may attempt to support revenues by
using the two following measures at the same time: (1) controlling the degree of
increasing selling prices strictly, and (2) providing additional trade credit. In this
case, while trade credit use increases, the profit margin may decrease since firms
manage to keep the increase of selling price to a minimum. This outcome adds to
the prediction of Nadiri (1969) that the relation between trade credit provision and
profit margins can also be negative, depending on the price elasticity of product
demand. In short, according to Blazenko and Vandezande (2003), under the
condition of facing an increase in marginal costs, trade credit can be used to
encourage sales and profitability. Yet, if a firm’s price elasticity of product demand
is too high, trade credit supply may not help to increase the firm’s profitability
though it may support sales.
In this stream of research, several theoretical studies attempt to go one step
further to explain how trade credit functions to support sales. For example, Long et
al (1993) show that trade credit may be like a guarantee of product quality to
customers. In particular, they explain how trade credit can function as a marketing
device to promote sales. By providing trade credit to buyers, sellers allow their
clients to verify the quality of a product before making any payment. When the
quality is poor, buyers can return products without making payments.
Consequently, it is only profitable for sellers that produce high-quality products to
grant credit to customers, while producers of low quality will choose not to provide
credit. Long et al (1993) suggest that as a result trade credit can work as a product
guarantee as well as being a sign of product quality to increase product demand. In
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
24
particular, this product guarantee can promote sales of firms that produce high-
quality products and have not yet gained a well-established reputation of product
quality.
Hyndman and Serio (2006) have developed an alternative theoretical model
that explains how trade credit works as a marketing device to support sales. In this
theoretical framework, the authors consider the number of competitors as the main
factor that drives market competition. At the same time, they assume that the seller
can decide to sell on cash or on credit under the constraints of incurring monitoring
costs if they sell on credit. The outcome of this study implies that when there is a
single supplier, the seller does not need to attract customers and thus they only sell
cash. However, increasing the number of suppliers first induces suppliers to grant
credit to develop a trading relation with customers in order to generate sales. Yet,
when the number of suppliers increases too much, customers find it easy to switch
to another supplier, and suppliers will decrease the amount of trade credit provided.
The reason is that it is difficult for suppliers to enforce the repayments from
customers. Indeed, this study implies that trade credit can be used to encourage
sales. However, firms in a monopolistic market do not need to use trade credit to
support sales. At the same time, it is very costly to provide credit in highly
competitive markets, and consequently firms may be less willing to provide credit
to support sales when facing fierce competition.
Furthermore, several studies argue that suppliers provide credit to encourage
customers to order large quantities. These studies concentrate on discussing the
optimal economic quantity order under conditions of delayed payment. For
example, Rachamadugu (1989), Chung (1998), and Huang (2007) have developed
theoretical models to determine the optimal economic order quantity when delayed
payment is permitted. In these theoretical models, firms attempt to determine the
optimal economic order quantity in order to minimize the total annual costs, which
include (1) the annual costs of placing orders, (2) the annual costs of stock holding,
Chapter2: Trade Credit Supply and Firm Sales Growth
25
and (3) the annual costs of opportunity capital. In other words, the objective of the
firm is to maximize the net present value of the shareholder wealth by minimizing
the net present value of the cash outflows resulting from the total annual costs. The
solution of these models shows that the optimal order quantity is an increasing
function of the permitted delay in payment.
To conclude, several theoretical studies argue that trade credit works as a
marketing device to stimulate sales growth. However, we do not know of any
empirical study that investigates whether trade credit provision enhances firm sales
growth. In this chapter, we examine the impact of trade credit supply on firm sales
growth rates to gain an insight into whether trade credit provision can help to
support firm sales growth.
2.3 The Data, Empirical Strategy
2.3.1 The Sample
The dataset for this study is a panel for the period 2004-2007. We derive the basic
sample from the official websites of the Ho Chi Minh Stock Exchange and the
Hanoi Stock Exchange, as well as annual reports and websites of listed companies.
In total, 320 firms are listed at the two stock trading centers in September 2008. In
our sample we do not include firms in the financial sector since they have different
accounting rules. In addition, we select only listed firms that released annual
reports at least in the last three years. This means that we exclude several firms that
were listed in 2008 since these firms released annual reports only in the last two
years. The variable of interest is a measure of the sales growth rate, and we need to
observe at least two observations of sale growth rate per firm. Due to these
restrictions, our sample was reduced from 320 firms to 260 firms that were listed in
the period 2000-2007.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
26
The sample represents large manufacturing Vietnamese firms: in total 260
firms in 20 industries. The number of firms per industry is shown in table 2.1. In
our sample, 252 of the 260 firms (97 per cent of the sample) provide credit to
customers. The average proportion of account receivables in total assets is about 18
per cent (see table 2.2)
Table 2.2: Sample firms by industry
Ord.
no
Industry Number of
Firms
1 Basic metals 1
2 Chemical and all allied products 6
3 Clothing 4
4 Commercial machinery and computer equipment 12
5 Construction 80
6 Electronic and other electrical equipment products 12
7 Food and drink 39
8 Furniture and fixtures 7
9 Gasoline, diesel and electricity wholesale 9
10 Medical products 5
11 Mining and quarrying of nonmetallic minerals 2
12 Oil mining services 1
13 Printing and publishing 16
14 Rubber and miscellaneous products 18
15 Services (hotel and entertainment services) 7
16 Stone, clay, glass and concrete products 18
17 Textile mill products 1
18 Tobacco products 1
19 Transportation equipment products 6
20 Transportation service 15
Total 260
Chapter2: Trade Credit Supply and Firm Sales Growth
27
2.3.2 Empirical Strategy
The main research objective of this study is to investigate whether the size of trade
credit supply enhances firm sales growth. If trade credit functions as a marketing
device that influences product demand, then firms that provide more credit will
achieve higher sale growth rates.
In this section, we explore the impact of trade credit supply on firm sales
growth rates. At the same time, we also control for the influence of other relevant
factors such as firm age, firm size, human capital, ownership structure and leverage
on firm growth rates. The regression is based on a panel of Vietnamese listed firms
during the period 2004-2007.
The estimated equation is the following:
SAGROWit = αXit + βTDPROit + εit (2.1)
In equation (2.1) the dependent variable SAGROWit is defined as the annual growth
rate in firm sales at time t in the four-year period 2004-2007. The Xit includes a
constant term and characteristics of the individual firm such as firm size, firm age,
ownership structure, financial structure, and human capital. The core variable of
interest is a measure of trade credit provision. We employ two popular measures of
trade credit supply in the existing literature: (1) TDPRO1it – measured by the ratio
of account receivables to firm total assets (Deloof and Van Overfelt, 2009); (2)
TDPRO2 – measured by the ratio of account receivables to firm sales (Giannetti et
al, 2009).
We use a fixed-effects analysis to estimate the impact of trade credit
provision. The fixed-effect estimator offers several advantages. First of all, it
disentangles the impact of trade credit provision from the observed attributes that
affect the outcome variables. Second, it controls for relevant unobserved firm and
industry characteristics that affect sales growth and do not change over time.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
28
Yet, one of potential criticisms of our study is the potential endogeneity of
trade credit supply variable since causality may run in both directions – from trade
credit provision to sales growth and vice versa. For example, it is possible that
firms having low sales growth in the past may offer more credit to archive higher
sales growth rates or firms with high sales growth may provide more trade credit.
Another concern to consider are unobserved firm characteristics that can influence
both trade credit supply and firm growth. For example, firm competitiveness may
be correlated with trade credit and sales growth. However, based on the data we
have, we are not able to have a good instrumental variable measuring firm
competiveness to solve this problem. In response, we employ the method of GMM
system (Arellano and Bover,1995; Blundell and Bond, 1998), which is expected to
overcome the problem of reverse causation and omitted variable bias mentioned
above.
2.3.3 Empirical Measures
Control Variables
The relations of firm size and firm age to firm growth have been studied
extensively in the literature. Consequently, we control for the effects of firm size
and firm age in our study. FSIZE is measured as the log of firm sales, while FAGE
is the age of a listed firm, defined as the number of years since a firm has been
established. To control for the impact of the financial structure on firm growth, we
introduce leverage, LEV, measured as the ratio of total debt to the value of equity
as our proxy for capital structure. This variable reflects the degree to which a firm
utilizes borrowed money. Furthermore, in order to capture the effects of human
capital on firm growth rates, we employ the ratio of the number of employees with
a university degree (bachelors/masters/doctors) to the total number of employees,
DEGREE, as our measure of human capital (Brown et al,. 2005). Another
Chapter2: Trade Credit Supply and Firm Sales Growth
29
important factor that can contribute to firm growth rates is ownership structure. We
include the variable STSHARE, defined as the percentage of a firm’s shares held by
the state. We also use FRSHARE, measured as the percentages of a firm’s shares
held by foreign investors. Besides, we include the profitable variable, PRORATIO,
measured as the ratio of total profit from the main business activities before tax
(excludes profit from financial activities) to total sales. Finally, trade credit
obtained, TDOBi, is measured as the ratio of account payables to total assets,
measuring the proportion of total assets that firms financed with trade credit from
their suppliers. Due to the data restriction, we cannot include a measure of
innovation. Yet, in the previous discussion, it was suggested that characteristics of
firm managers do influence the growth of small business. However, our data only
includes large firms. Also, characteristics of a board of directors in the sample have
not changed much during 2004-2007. Therefore, the fixed-effect regression is
removed from the variables that measure the characteristics of a board of directors,
such as sex, and education; these variables have not been included in our model.
Table 2.3: Descriptive statistics
Variables Mean Max Min Std Obs
SAGROW 0.366 5.77 -0.944 0.86 662
TDPRO1 0.17 1.00 0 0.16 779
TDPRO2 0.185 0.71 0 0.14 776
PRORATIO 0.089 0.698 -0.201 0.0740 778
LOGSALES 12.104 16.461 7.703 1.316 784
STSHARES 0.300 0.85 0 0.232 871
FRSHARES 0.065 0.882 0 0.155 868
AGE 12.71 53 1 10.10 883
DEGREE 0.237 0.942 0.012 0.179 871
LEV 2.37 39.82 0.022 3.52 772
TDOB 0.156 0.822 0.0001 0.142 773
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
30
2.4 Empirical Results
2.4.1 Trade Credit Provision
Results of the fixed-effects and the one step system GMM models for determinants
of sales growth are presented in table 2.4 and table 2.5. We start by estimating the
general model, which includes all our possible variables.
Our discussion first concerns the impact of trade credit provision on firm
sales growth, since this is the main focus of our study. Indeed, trade credit
provision has a significant and positive impact on sales growth (table 2.4, columns
1 and 2; table 2.5). The estimated coefficient indicates that firms that offer more
trade credit do realize higher sales growth rates. Consequently, the empirical
evidence of a positive association between trade credit provision and firm sale
growth supports the view of trade credit as a marketing instrument.
In the fixed effect regression we also use the interaction term of the trade
credit provision variable, TDPRO, and firm age, FAGE (table 2.4, columns 3 and
4). This allows us to test whether the impact of trade credit provision on sales
growth is different for young firms that have not yet established their reputation of
product quality as compared to old firms with a reputation. The negative
coefficients of the interaction terms between FAGE and TDPRO confirm that the
impact of trade credit provision is stronger for young firms. The reason may be that
young firms do not have an established reputation, and therefore the need of testing
quality before making any payment is higher in their case. When young firms
provide trade credit, clients can return the inferior products without incurring any
costs. This is like a guarantee, which is very important to clients of new suppliers.
Thus, trade credit supply affects sales growth strongly.
In general, the fixed-effect and the system GMM estimators show that
trade credit can help to enhance sales growth rates. In addition, the fixed effect
Chapter2: Trade Credit Supply and Firm Sales Growth
31
model indicates that trade credit effect appears to be more important to young
firms than to older firms. This finding is consistent with Long et al (1993), which
suggest that trade credit functions as a marketing device to support sales by
offering a guarantee and/or to signal product quality.
As a robustness test, the random effect and OLS regressions also provide
similar results for the positive and significant impact of trade credit supply on sale
growth. However, we do not find a significant impact of the interaction term
between trade credit supply TDPRO and firm age AGE on firm sales growth.
Control Variables
Our next concern is the effect of control variables on firm sales growth. First, we
do not find support for the Gibrat law, since large firms in our sample appear to
experience higher sales growth rates than small firms. In addition, we find
supportive evidence that young firms also realize higher sales growth rates than old
firms. The data also suggests that highly profitable firms experience higher sale
growth rates than firms that are less profitable. However, our findings do not
confirm that firms that utilize a higher ratio of debts to equity reach a higher sales
growth rate.
Moreover, our measure of human capital shows no statistically significant
correlation with firm sale growth. However, we find supportive evidence that
ownership structure affects firm sale growth rates. In fact, the estimated results in
this study review the negative and significant effects of the proportion of shares
hold by the states and the proportion of shares hold by foreign investors on firm
sales growth.
In short, our findings suggest that in our sample firm size, firm age, trade
credit provision, and firm profitability are the main determinants of firm sales
growth.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
32
Table 2.4: Results of the fixed-effect models
Variables (1)
(2) (3)
(4)
TDPRO1 1.049
(2.01)**
3.38
(3.030)***
TDPRO2
1.269
(2.84)***
3.440
(2.65)***
TDPRO1*FAGE
-1.276
(-2.79)***
TDPRO2*FAGE
-1.034
(-1.99)**
FAGE -0.809
(-3.71)***
-0.699
(-3.21)***
-0.595
(-2.58)***
-0.502
(-2.14)**
FSIZE 0.768
(3.68)***
0.753
(3.66)***
0.753
(3.67)***
0.744
(3.63)***
PRORATIO 2.078
(2.49)**
1.874
(2.31)**
2.066
(2.44)**
1.903
(2.32)**
LEV 0.018
(1.08)
0.019
(1.13)
0.013
(0.72)
0.018
(1.03)
DEGREE 0.932
(0.90)
0.980
(1.00)
0.981
(0.98)
0.943
(0.96)
FORSHARE -0.032
(-0.14)
0.119
(0.53)
-0.074
(-0.34)
0.074
(0.33)
STSHARE 0.586
(0.73)
0.452
(0.57)
0.365
(0.50)
0.395
(0.53)
TDOB 0.12
(0.27)
-0.044
(-0.09)
0.373
(0.86)
0.066
(0.15)
Conts -8.01
(-3.61)***
-8.067
(-3.67)***
-8.173
(-3.78)***
-8.354
(-3.84)***
R SQUARE 0.2641 0.2702 0.2844 0.2811
Observations 660 660 660 660
The fixed-effect regressions of sales growth on trade credit provision, trade credit obtained,
firm age, firm size, employee degrees, and leverage and ownership structure. Standard
errors are robust and have been adjusted for cluster effects. The T-statistics are given in the
parentheses above. *, ** and *** denote a significant level at 10, 5 and 1 per cent,
respectively.
Chapter2: Trade Credit Supply and Firm Sales Growth
33
Table 2.5: Results from the one step system GMM estimator
Variables (1) (2)
L. salesgrowth 0.008
(0.18)
0.009
(0.12)
TDPRO1 1.537
(1.98)**
TDPRO2
3.256
(2.51)** FAGE -0.110
(-1.32)
-0.207
(-1.71)*
FSIZE 0.386
(3.04)***
0.631
(3.18)***
PRORATIO 1.993
(2.47)**
1.618
(1.69)*
LEV -0.014
(-0.60)
-0.019
(-0.68)
DEGREE -0.221
(-0.31)
-1.127
(-1.01)
FORSHARE -0.683
(-2.69)***
-0.857
(-2.40)**
STSHARE -0.337
(-1.46)
-0.145
(-0.41)
TDOB 0.404
(0.37)
-0.147
(-0.10)
Conts -4.401
(-3.33)***
-6.929
(-3.32)***
Obs. 406 406
Number of firms 257 257
Sargan test
P value 11.89 0.752
19.54
0.242
Wald test 27.70
0.002
19.88
0.030
The one step system GMM regression of sales growth on trade credit provision, trade credit
obtained, firm age, firm size, employee degrees, and leverage and ownership structure.
Standard errors are robust and have been adjusted for cluster effects. The Z-statistics are
given in the parentheses above. *, ** and *** denote a significant level at 10, 5 and 1 per
cent, respectively. 2
2 This result may be suffer from first order serial correlation
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
34
Table 2.6: Results from the random effect and the OLS regressions
Variables (1)
(2) (3)
(4)
TDPRO1 0.828
(2.81)***
0.674
(2.62)***
TDPRO2
0.622
(2041)**
0.455
(2.15)** FAGE -0.078
(1.98)**
-0.071
(-1.76)*
-0.064
(-1.75)*
-0.058
(-1.57)
FSIZE 0.091
(3.71)***
0.070
(2.76)***
0.061
(2.95)***
0.043
(2.03)**
PRORATIO 1.654
(3.07)***
1.429
(2.66)***
1.352
(3.21)***
1.156
(2.74)***
LEV 0.019
(1.10)
0.024
(1.38)
0.021
(1.09)
0.025
(1.44)
DEGREE 0.327
(2.32)**
0.430
(2.86)***
0.340
(2.59)***
0.422
(3.09)***
FORSHARE -0.129
(-0.96)
-0.038
(-0.33)
-0.108
(-0.80)
-0.028
(-0.24)
STSHARE -0.413
(-2.76)***
-0.430
(-2.82)***
-0.417
(-2.97)***
-0.429
(-3.04)***
TDOB -0.1027
(-0.34)
-0.130
(-0.45)
-0.058
(-0.21)
-0.063
(-0.25)
Conts -0.859
(-2.90)***
-0.601
(-1.92)**
-0.494
(-2.03)**
-0.261
(-1.05)
R SQUARE 0.137 0.126 0.108 0.093
Observations 660 660 660 660
The random effect regression (1) and (2), and the OLS regression (3) and (4) of sales
growth on trade credit provision, trade credit obtained, firm age, firm size, employee
degrees, and leverage and ownership structure. Standard errors are robust and have been
adjusted for cluster effects. The T-statistics are given in the parentheses above. *, ** and
*** denote significant level at 10, 5 and 1 per cent, respectively
2.5 Conclusions
In this chapter, we have used a panel of listed Vietnamese firms to identify the role
of trade credit supply in determining firm sales growth. Our major finding is a
Chapter2: Trade Credit Supply and Firm Sales Growth
35
positive and significant relationship between firm sales growth and trade credit
provision. In particular, we observe that the impact of trade credit provision on
firm sales growth is stronger for young firms than it is for old firms. This finding
supports the view that trade credit works as a marketing device to promote sales.
Chapter 3
A Survey in the Rice Market of the
Mekong Delta
3.1 Introduction
As mentioned in chapter 1, this thesis aims to investigate the role of trade credit
provision in the rice market of the Mekong Delta. The literature on trade credit
shows that data sets at firm-level and transaction-level are needed to study this
issue. In order to collect data, we conducted a survey in 2007: 626 rice firms
operating at the six main market segments of the rice market in the Mekong Delta
were interviewed. The survey was conducted in the six provinces of An Giang,
Tien Giang, Vinh Long, Can Tho, Hau Giang, and Soc Trang. This survey helped
us to construct a data set that includes several variables that are relevant to our
research.
This chapter describes the procedures of the survey, providing an overview
of the rice market in the Mekong Delta and some statistics on the sample. Section
3.2 will provide an overview of the rice market in the Mekong Delta; it will
describe the process of rice production and characteristics of the traders. Section
3.3 describes the data collection procedures. Next, section 3.4 presents some
descriptive statistics of the data collected, and finally section 3.5 provides a
summary of the main findings.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
38
3.2 A Brief Description of the Rice Market of the Mekong Delta
3.2.1 Overview of the Paddy/Rice Production
The Mekong Delta is the largest delta in the south of Vietnam. With an area of
39,554 km2
(12 per cent of the entire country), the Mekong Delta has produced
about 20 million tons of paddies or over 5 million ton of rice in 2008. This region
has been considered as the rice bowl of Vietnam, as it accounts for 50 per cent of
the total of rice produced in the country and provides about 90 per cent of the rice
for export from Vietnam.
Details in the process of producing paddy/rice are described in figure 3.1.
In general, there are several phases in producing paddy/rice in the Mekong Delta.
First, farmers produce and dry the paddy immediately after harvesting. Next,
farmers sell the paddy to assemblers/gatherers or they sell directly to processing
firms; gatherers/assemblers are intermediaries who only purchase paddy from
farmers and transport to sell to processing firms without processing the paddy. In
the next phase the paddy is processed into rice. Processing firms check the standard
requirements on moisture. After checking, the dried paddy is stored in warehouses
while the paddy that is too moist is dried by dryers and/or by the sun. After this,
the paddy will be processed further: it will be milled into brown or white rice. In
the case of export rice, the brown rice is polished before packaging. The finished
product will be stored in the warehouse until it is sold on the market.
The flows of paddy/rice in the Mekong Delta are presented in the figure
3.1. Paddy/rice is transported to several marketing agents in the same province and
in other provinces in the Mekong Delta. Figure 3.1 shows that the key marketing
agents in the regional rice market are gatherers, w-millers, millers, wholesalers,
retailers, and SOEs.
Chapter 3: A Survey in The Rice Market of The Mekong Delta
39
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
40
3.2.2 Characteristics of Rice Traders in the Mekong Delta
Farmers
Farmers in the Mekong Delta generally cultivate relatively small irrigated farms
with three main cropping seasons: winter-spring (October-February), spring-
summer (February-May), and summer-autumn (June-September). Rice farmers in
the Mekong Delta have relatively small farms that are 1 to 3 hectares on average.
Farmers produce paddy and dry it immediately after harvesting. After storing some
paddy for home consumption and keeping seeds for the next production cycle,
most farmers sell the surplus to gatherers or processing firms i.e. SOEs and w-
millers. In the Mekong Delta, about 75 per cent of the total paddy production is
sold after harvesting. The proportion of marketable surplus is highest for farmers in
this region (Minot and Goletti, 1995).
Gatherers/Assemblers
Gatherers usually buy paddy from farmers and transport it to processing firms such
as w-millers and/or SOEs for immediate sale. They are not involved in any
processing activity. Gatherers often operate in a relatively small area of about 10
kilometers. Since they work in one area for a long time, they have a good
knowledge of the quality of the paddy from the farmers in the surrounding villages.
According to the 1995 IFPRI survey of rice farmers, gatherers account for more
than 95 per cent of the paddy purchased from farmers and transported and sold to
processing firms.
Processing Firms
Processing firms in the Mekong Delta include private millers and the milling
factories of state-owned enterprises. The difference in processing capacity among
Chapter 3: A Survey in The Rice Market of The Mekong Delta
41
traders at the milling sector is big. In Vietnam, there are about 620 large millers
with a capacity of 15 to 200 tons per day and 30,000 small and medium millers
(Huynh et al, 2009). Small millers have a producing capacity of 0.2 to 1 ton per
day, while the capacity of medium-scaled millers is 1 to 15 tons per day shift.
In this study, we refer to millers as processing firms that provide milling
services and charge a certain price per kg output. They are usually small millers
that own a small factory with a small processing capacity (less than 1 ton of rice
per day). They mainly provide milling services to farmers nearby for home
consumption. These small millers exist in almost every village in the Mekong
Delta. There are also a few large-scale millers/polishers in the Mekong Delta that
provide milling and polishing services to other rice exporters and/or w-millers.
W-millers (medium and large-scale private millers) have a capacity of at
least are larger than 1 ton of rice per day. They usually purchase paddy from
farmers/gatherers and process this paddy into rice. They also store and sell milled
rice as a final product to wholesalers and/or other w-millers. At the same time, w-
millers also sell a large amount of milled rice to SOEs and private rice exporters.
Millers/w-millers account for the milling and polishing of about 80 per cent of the
total rice in the region (www.stp.vn).
There are 11 SOEs (state-owned enterprises) in the Mekong Delta of
Vietnam. This are provincial food companies in each province. Each state-owned
enterprise usually has a few modern factories with modern and high capacity
machines to process high quality rice for exports. For example, the state-owned
food company in the province of Tien Giang has four factories with an average
production capacity of about 100 ton per day. Since SOEs have high capacity
equipment to process and store rice, they are the main rice exporters of the region.
They usually purchase milled rice from w-millers and polish rice to meet the
requirements of foreign traders. Rice exporters also purchase paddy from farmers
to produce rice; however, this accounts for a relatively small part of their exporting
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
42
volumes. They mill and polish about 20 per cent of the total of rice produced in the
region.
Although the government tries to remove external and internal restrictions
so that all rice firms are equal in exporting activities, the number of private firms
that export their products is still very small. Moreover, the total number of rice
exporters in the Mekong Delta remains small.
Wholesalers
Wholesalers play an important role in distributing rice for consumption in the
region. In fact, wholesalers act as intermediaries who distribute rice from w-millers
to final consumers in the Mekong Delta, by using rice retailers/small wholesalers
in the region. According to Minot and Goletti (1995), rice wholesalers trade about
2 million tons of rice to be consumed in the region.
Retailers
Retailers usually have shops at marketplaces. Retailers mainly purchase rice from
wholesalers located nearby, they store the rice and sell it to final consumers at local
markets.
Gatherers and exporters (SOEs) are excluded from the empirical analyses
in chapters 4, 5 and 6.3 We focus on the four market segments – w-millers, millers,
wholesalers, and retailers – since these traders only produce and trade rice at the
regional market. In these four segments, traders play a role that is different from
that of the processing firms, service firms, and wholesale and retail firms. Due to
3 Gatherers mainly transport and trade paddy, which is different from rice, the main product of other
market segments. Besides, SOEs concentrate on processing rice for exporting purposes and are
mainly involved in exporting activities. SOEs play a dominant role in long distance trading, e.g. on
the interregional and international market. The market of SOEs is very large since rice exporters may
not only compete with rice exporters inside the country but also with rice exporters from other
countries such as Thailand and India. As a consequence, it can be difficult to measure the market
structure in this market segment.
Chapter 3: A Survey in The Rice Market of The Mekong Delta
43
the differences in the main business activities, we notice that the degree of market
competition varies across these market segments (Luu, 2003).4 At the same time,
trade credit is popular in this particular market. Therefore, we find that the rice
market is an interesting case to investigate the impact of firm competitiveness on
trade credit supply.
3.3 Description of Data Collection
Our data collection procedures consist of two stages: a pilot survey was conducted
in 2006, and the survey was finalized in 2007.
3.3.1 The Pilot Survey
A pilot survey was carried out in June and July 2006 with the aim of field-testing
the questionnaires and survey procedures. The pilot survey took place in the four
provinces: Tien Giang, Vinh Long, Can Tho and Soc Trang in the Mekong Delta.
Participants of the pilot included a small group of agents involved in the rice
market of the Mekong Delta. They are two SOEs in Tien Giang and Vinh Long,
five millers, five wholesales, and five retailers and five gatherers in Can Tho and
Soc Trang. At this stage of the research, the quality of information and a good
insight into different businesses as well as business relationships in marketing was
the main issue.
The participants were selected through the network that had previously
been established. The interviewees were former students of the School of
Economics and Business Administration of Can Tho University and were at that
time working at the interviewed rice firms. The discussions with these interviewees
were open and helpful. The interviewees provided useful feedback on our measures
about firm competitiveness and trade credit use, and they allowed us to test the
4 See section 3.4.3.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
44
validity of our questionnaires. After completing the pilot, much was revised on the
basis of the feedback we got from the interviewees, e.g. we rephrased several
questions to make them more accessible to rice traders. The interviewees also
suggested that we use some measures of competition that would be easier to apply
for the rice traders.
3.3.2 The Official Survey
The final survey was conducted in the summer of 2007. The key criterion to select
the provinces for our survey was that the provinces had to be the main rice
producers in the Mekong Delta: the number of rice traders in these provinces is
high, just like the volume of trade between rice traders.
The six selected provinces are: An Giang, Tien Giang, Vinh Long, Can
Tho, Hau Giang, and Soc Trang. An Giang was selected because recently this
province has been the largest rice producer in the Mekong Delta; for example, in
2006, when the Mekong Delta produced 18,229,200 tons paddy, An Giang was the
largest producer with a total amount of 2,923,200 tons.
The other five provinces in the survey are all large rice producers. One of
the advantages that these provinces have over others, is that they are located close
to one another and along national highway 1A; transport between these provinces
is convenient and can be via roads and waterways. Inter-province trade transactions
are large, and market integration between these provinces is expected to be high.
Traders in these provinces can easily exchange information on the rice market such
as prices, policy and other rice firms.
The Enterprise Survey
In Vietnam, nationwide enterprise surveys are conducted every five years to collect
information on every corporation. This enterprise survey collects data for the
government to evaluate the performance of current economic policies and to make
Chapter 3: A Survey in The Rice Market of The Mekong Delta
45
plans for the future economic development of the country. The latest enterprise
survey, which was announced by a Prime Minister’s Decision on August 15, 2006,
took place in July 2007.
Early July we learned about the enterprise survey. We found that it thought
it would be a great advantage if the staff of statistic bureaus in each province were
to conduct our survey together with the enterprise survey. Firstly, the interview
team appeared to be highly qualified; the staff of statistics bureaus has a lot of
experience with enterprise interviews. Secondly, the government had authorized
the statistic bureaus to do the interviews. That means that firms will be more
cooperative and will sooner make an appointment; this is important since it can be
very difficult to make appointments to interview a large number of firms. Thirdly,
the staff of the statistics bureaus has been working with these firms for quite some
time. Every year, firms in each province have to provide all kinds of reports such
as balance sheets, cash flow statements and profitability statements, since the
statistics bureau in each province has to prepare data for the province yearbook. As
a result, firms will provide relatively consistent answers since big differences in the
figures/numbers when compared to the previous year will lead to many questions.
In addition, firms also try to provide useful answers to other questions. Fourthly,
the two questionnaires of our survey and the enterprises survey contained several
similar questions. For example, both questionnaires include questions on
background information: main business activities, and accounting figures as total
sales, assets, resources of capital, bank loans, employees etc., and business results
for the previous year (2006). Therefore, it is convenient for staff of the statistics
bureaus in each province to conduct the two surveys in parallel: the enterprise
survey and our survey.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
46
Survey Procedures
In June 2007, with the introduction from Can Tho University, we contacted
statistics bureaus in each of the six selected provinces. We proposed a cooperation
by conducting our survey together with the enterprises survey. Soon after reaching
an agreement with the statistics bureaus in 2007, a short training was organized to
explain our questionnaire to the staff of the bureaus. Our survey was conducted in
the period July to September 2007 and led to 626 interviews with rice firms in the
Mekong Delta.
Sample Selection
As mentioned, our sample mainly includes rice firms at the four main market
segments of the rice industry in the Mekong Delta. They are millers, w-millers,
wholesalers and retailers. In addition, we interviewed some rice exporters and
gatherers to enable us to better describe the rice market.
In each province, the statistics bureau prepared a list of rice firms from
different districts. The lists included registered rice firms that are millers, w-
millers, wholesalers and SOEs. Each list included all firms located in a particular
district, and the firms were listed in alphabetical order. We decided to select the
first firm of every five firms in the list (numbers 1, 5, 10 etc. of the lists) to take
part in the interview.
Rice retailers are not registered firms. They usually have shops at
marketplaces. These shops are often located in the main street of a local (district)
market. We picked the first shop out of every 10 shops for our interview.
In total, 626 rice firms in the Mekong Delta were interviewed within three
months. As the focus of the study is trade credit use among rice firms in the rice
market of the Mekong Delta, in our interviews we focused on important traders in
domestic markets as processing firms (w-millers/millers) and trading firms
(wholesalers and retailers).
Chapter 3: A Survey in The Rice Market of The Mekong Delta
47
The interviewers made an appointment and sent the questionnaires to the
interviewees one week prior to the interview. Next, the interviewers did interview
firm owners (managers) face to face. In each province, we attended the interviews
for two days to observe and/or check whether the interviewees had any questions
regarding the questionnaire and whether the interviewers could explain things
properly. At the same time, we also checked whether the finished questionnaires
were filled in properly. Early October 2007, we again visited each province for two
days, we worked together with the interviewers to check all the information of the
completed questionnaires and collected them.
3.4 Descriptive Statistics of the Sample
The core target of this project is to investigate the relevance of trade credit supply
as a marketing instrument. Existing studies on this topic mainly focus on the
impact of market characteristics on trade credit supply such as market competition,
bargaining power. We therefore concentrate on gathering information on market
context, i.e. firm competitiveness as a measure of market competition at the firm
level, a firm’s bargaining power. This section presents descriptive statistics of the
measures of trade credit supply: firm competitiveness. We also show statistic
figures of other control variables that may influence trade credit supply, such as
firm size, firm age, firm profitability, and the firm’s financial constraints.
3.4.1 Overview of the Sample
Size and Age of Firms In the Sample
The sample includes 158 w-millers, 183 millers, 101 wholesalers, and 145
retailers. It includes six local state-owned enterprises (exporters); the rest of the
sample is private firms. Categories of the survey firms are presented in table (3.1).
In our survey, the size of firms is measured by its total assets. On average,
the value of total assets of the firms in our sample is about 2.5 billion VND
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
48
(155,097 USD); the firm age is about eight years. Regarding the distribution of the
surveyed firms of the sample, table 3.1 reveals that firms with capital of less than 1
billion VND (62,039 USD) account for 70 per cent of the sample, while firms with
capital more than 1 billion VND make up 30 per cent of the sample.
Table 3.1: Distribution of rice firms according to main business activities
and firm size (measured by total assets)
Categories of the surveyed firms Number of firms Percentages
State-owned enterprises 6 1
Private firms 620 99
Types of firms
Retailers 145 23.3
Wholesalers 101 16.3
Millers 184 29.23
W-millers 158 25.08
Gatherers 17 2.88
Exporters 21 3.35
Size of firms
Less than 100 million VND 204 32.59
From 100 million to 1 billion VND 232 37.06
From 1 to 5 billion VND 131 20.93
More than 5 billion VND 59 9.42
Source: survey in 2007
Table 3.2: Size and age of the surveyed rice firms
Obs Max Mean Min St.dev
FAGE (years) 625 29 8.37 1 5.20
FSIZE(millions VND) 626 269,000 2,520 1 12,300
FSIZE at different
market segments
SOEs/private exporters 21 269,000 31,900 6,560 55,600
W-millers 158 73,300 4,220 18 8,580
Millers 184 6,110 669 15 807
Wholesalers 101 15,000 1,010 11 2,550
Retailers 145 410 35.3 1 55.5
Gatherers 17 2,500 615 170 605
Source: survey in 2007
Chapter 3: A Survey in The Rice Market of The Mekong Delta
49
Profitability
Table 3.3: Descriptive statistics of profitable variables
MARGIN Obs. Max Mean Min St.dev
The whole
sample
626 0.61 0.13 0 0.1183
W-millers 158 0.50 0.13 0 0.1289
Millers 184 0.61 0.23 0 0.1106
Wholesalers 101 0.55 0.06 0.003 0.0776
Retailers 145 0.28 0.07 0.003 0.0397
Source: own survey, conducted in 2007
We measure the firms’ profitability by using the price-cost margin.
5 On average,
the price-cost margin of firms in our sample is 0.13. The price-cost margin also
varies much across different market segments: it is higher for processing firms
(average of MARGIN is 0.23 for millers and 0.13 for w-millers) than it is for
trading firms (average of MARGIN is 0.06 for wholesalers and 0.07 for retailers)
(table 3.3).
3.4.2 Trade Credit Provision
Trade credit appears to be used extensively in this particular market. Overall, the
figures in table 3.4 show that 65 per cent of the firms in the sample (i.e. 410 of our
626 rice firms) grants trade credit to their clients. The average amount of trade
credit granted is about 25 per cent of total sales.
5 The price-cost margin is calculated by
ceaveragepri
tsiableaverageceaveragepri cosvar− . Details are
explained in chapter 4, pages 81
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
50
Table 3.4: Distribution of the surveyed rice firms providing credit
Total
firms
Credit supplying
firms
Amount of trade credit
supply (proportion of
credit sales (%)
Total sample
(%)
626 410
65.49
0.25
W-millers
(%)
158 145
91.77
0.40
Millers
(%)
184 55
28.98
0.13
Wholesalers
(%)
101 74
73.27
0.30
Retailers
(%)
145 101
69.65
0.19
Source: own survey, conducted in 2007
Rice firms at different market segments appear to behave differently with
respect to trade credit provision. The differences in trade credit provision are large
and statistically significant across the four market segments. Trade credit occurs
more frequently in the segments of w-millers, wholesalers and retailers than with
millers. In fact, 92 per cent of w-millers in the sample provide trade credit to
customers, and about 40 per cent of revenues were made on delayed payments in
2006. The percentage of traders at the wholesale segment granting credit decreases
to 73 per cent, while the proportion of credit sales is 30 per cent of the total
revenues. Traders at the retail and miller segments seem to provide less credit to
their clients; on average, 70 per cent at the retail segment offer credit to trading
partners, and the percentage of sales made on credit is about 19 per cent. At the
same time, only 29 per cent of the millers grant credit to their customers, and about
13 per cent of their sales were made on delayed payments in 2006.
Experience with defaulting by clients
On average, 26 per cent of the traders in our sample confirm that they were not
been paid at least once in the last three year by their clients, who have receive
credit. Retailers are most likely to not being paid, since 41 per cent of retailers in
Chapter 3: A Survey in The Rice Market of The Mekong Delta
51
our sample mention that they have not been able to get their money back at least
once. This figure is 24 per cent for w-millers and wholesalers, while it is 19 per
cent in the case of millers. The maximum of times that firms in our sample have
not been paid in the last three year is six.
Table 3.5: Experience with defaulting by clients
Number of firms that were
not paid
Total
firms
Number of
firms that
were not paid
Number of
times firms
were not paid
Total sample
(%)
626 165
26.36
0.79
W-millers
(%)
158
38
24.05
0.59
Millers
(%)
184 35
19.02
0.63
Wholesalers
(%)
101 25
24.75
0.44
Retailers
(%)
145 60
41.37
1.57
Source: own survey, conducted in 2007
Trade Credit Received
Table 3.6 Trade credit received
Trade credit obtained
(BUYDP)
Obs Max Mean Min St.dev
Total sample 626 1 0.13 0 0.2275
At market segment level
W-millers 132 0.80 0.14 0 0.2253
Wholesalers 99 1 0.10 0 0.1882
Retailers 145 1 0.13 0 0.3011
Source: own survey, conducted in 2007
The firms in our sample also receive credit from their suppliers. On average, w-
millers purchase about 14 per cent of their total input on delayed payments. In the
case of retailers, 13 per cent of total inputs is purchased on credit, and at the
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
52
wholesale market this is 10 per cent. Millers are mainly involved in providing
milling services, they seldom purchase paddy/rice and therefore do not receive
credit from suppliers.
3.4.3 Firm Competitiveness
The data in this section reflect firm competitiveness in the rice market. According
to Porter’s five forces, firm competitiveness in an industry is driven by the
following factors: competition with existing rivals, bargaining power vis-à-vis
customers, bargaining power vis-à-vis suppliers, threats of nearby substitutes, and
barriers to new entrants.
In our survey, we use several indicators to measure firm competitiveness.
In this section, we show five indicators that measure the three following factors
that drive firm competitiveness: competition from existing rivals (PERCOMP,
NORIVALS), bargaining power vis-à-vis customer (BPWCUS, SALDEC), and
bargaining power vis-à-vis suppliers (BPWSUP). These factors are very likely to
affect trade credit supply.
Existing competition occurs between businesses in the rice market that
supply products to the same type of clients. We try to measure existing competition
by using two variables: PERCOMP and NORIVALS. PERCOMP is a scale variable
indicating the important role of market competition and its effects on the results of
their rice business, according to the interviewees: (0=no pressure through
competition at all; 1=little competition but not important; 2=competition is
relatively important; 3=competition is important; 4=competition is extremely
important). NORIVALS is estimated by the number of competitors a rice firm uses
as a reference to determine its prices. Although the variation of PERCOMP across
the four segments is relatively low, it is statistically significant. Table 3.7 shows
that on average w-millers consider competition to be more important to their
business activities than other rice traders do. (PERCOMP mean: w-millers 3.32;
Chapter 3: A Survey in The Rice Market of The Mekong Delta
53
wholesalers 2.9; millers 2.86; and retailers 2.83.) In general, w-millers check their
selling prices with more competitors to determine their own selling prices (five
competitors) than other rice traders do (three competitors).
Regarding a firm’s bargaining power vis-à-vis customers, the firm can be
more competitive if it has a high bargaining power over its customers with respect
to determining selling prices and contract terms. We use the two important
indicators: BPWCUS and SALDEC to estimate a firm’s bargaining power vis-à-vis
its customers. BPWCUS is estimated by the percentage of sales for which a firm
can set the selling prices without any negotiations with customers. SALDEC is
measured by the percentage sales would decrease if a firm sets its selling price at 1
per cent higher than the other suppliers. Our rice firms state that they can set
selling prices without negotiating with customers for about 44 per cent of total
sales. In addition, using a selling price that is 1 per cent higher than the prices of
other supplies leads to a decrease in sales of about 62 per cent. This shows that
competition in this particular market is fierce. Note that the Mekong Delta yields
about 4 million tons of exporting rice each year, which is about 90 per cent of the
exported rice of the entire country. Consequently, the number of traders is very
high in this region and they process and trade large volumes. To illustrate, the
average size of an order in the w-miller segment is about 94 tons.
Table 3.7: Firm competiveness measures
Obs. Max Mean Min Median St.dev
PERCOMP 626 4 3.04 0 3 0.9773
NORIVALS 599 20 3.99 0 3 3.4874
SALDEC 609 1 0.62 0 0.75 0.3720
BPWCUS 593 1 0.44 0 0.4 0.3728
BPWSUP 583 100 26 0 20 27
Source: own survey, conducted in 2007
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
54
Moreover, there are large differences in a firm’s bargaining power vis-à-
vis its customers across different market segments. Retailers appear to have more
bargaining power vis-à-vis their clients than traders in other market segments. For
example, retailers can decide on the selling price of 65 per cent of their total sales
(table 3.8) while wholesalers can determine the selling price of 40 per cent of their
total sales. This figure is 43 per cent for millers and 40 per cent for the w-millers.
Regarding a firm’s bargaining power vis-à-vis its suppliers, theoretically it
is less attractive for a firm if its suppliers have strong bargaining power. We use
the variable BPWSUP, measured by the differences in selling prices that induce a
customer to move away from their regular supplier. This indicator aims to capture
the strength of a firm’s bargaining power compared to its suppliers in negotiating
contract terms and prices. Table 3.8 shows that w-millers are very careful with
input prices since a difference of 14 VND in purchasing price per kg paddy may
lead to w-millers switching to another supplier. This indicates that w-millers have a
stronger bargaining power vis-à-vis their suppliers (gatherers). In addition,
bargaining power vis-à-vis the supplier is better for wholesalers than it is for
retailers. A difference of 25 VND per kg rice will cause wholesalers to move to
another supplier, while a difference of 50 VND per kg rice would stimulate rice
retailers to behave similarly.
In short, competition appears to be important to rice traders in the Mekong
Delta. At the same time, the data shows significant differences in the
characteristics of market structure across the four main market segments. W-
millers face the strongest pressure from competition and have the least bargaining
power with customers; as processing firms, w-millers usually buy and sell very
large amounts and therefore their clients may have strong bargaining power. At the
same time, w-millers have strong bargaining power vis-à-vis their suppliers at their
input market. Wholesalers also experience a lot of competition and have a
relatively low bargaining power with their customers. Retailers have the strongest
Chapter 3: A Survey in The Rice Market of The Mekong Delta
55
bargaining power vis-à-vis customers; however, they have very little bargaining
power with their suppliers.
Table 3.8: Mean and median of competitive measures across different
market segments
W-millers Millers Wholesalers Retailers Sig.
PERCOMP ***
Mean 3.32 2.86 2.90 2.83
Median 4 3 3 3
NORIVALS ***
Mean 5.2 3.44 3.72 3.81
Median 4 3 3 3
BPWCUS ***
Mean 0.30 0.72 0.70 0.65
Median 0.15 0.80 0.85 0.75
SALDEC ***
Mean 0.72 0.43 0.40 0.34
Median 0.90 0.4 0.30 0.1
BPWSUP ***
Mean 14 18 25 50
Median 10 10 10 50
Source: own survey, conducted in 2007
The one-way ANOVA test for the mean difference of firms’ competiveness measures
across the four different markets segments: w-millers, millers, wholesalers and retailers.
*** indicate significance at the 1 per cent level.
3.4.4 Financial Constraints
A lack of capital can negatively influence a firm’s incentives to provide trade
credit. With the aim of controlling for the impact of the lack of capital on trade
credit provision, we gather information on firms’ financial constraints and their
access to bank loans. First, we asked rice traders to evaluate the importance of their
shortage of capital on the results of their businesses (LACKCAP: 0=no lack of
capital at all; 1=lack of capital is not important; 2=lack of capital is relatively
important; 3=lack of capital is important; 4=lack of capital is extremely important).
According to Luu (2003), financial constraints can make life very difficult for rice
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
56
traders in the Mekong Delta; the figures in table 3.9 confirm this as the mean of
LACKCAP is 3.
Table 3.9: The importance of financial constraints
LACKCAP Obs Max Mean Min St.dev.
Sample 626 4 3.04 0 0.9904
W-millers 158 4 3.2 0 1.021
Millers 184 4 2.79 0 0.9246
Wholesalers 101 4 3.09 0 1.02
Retailers 145 4 3.15 0 0.9904
Source: own survey, conducted in 2007
We also asked whether rice firms had any difficulty raising bank loans, and if they
had a loan in 2006. 21 per cent of our sample confirmed that they find it difficult to
apply for a bank loan, and 41 per cent of the rice traders in our sample had a bank
loan in 2006. W-millers used bank loans more frequently than traders in other
market segments; for example, 66 per cent of w-millers had a bank loan in 2006.
43.56 per cent of the traders in the wholesaler segment had a bank loan, while 37.5
per cent of the millers used bank loans in 2006. Rice retailers seldom use bank
credit, since only 9 per cent of them had a bank loan. A possible explanation for
this may be that w-millers and millers have factories that they can use as collateral
in applying for a bank loan. Wholesalers may obtain a bank loan by using their
large shops as collateral. Retailers do not have such means, and that explains why
they have fewer bank loans: retailers complained that financial constraints were
making business more difficult.
Chapter 3: A Survey in The Rice Market of The Mekong Delta
57
Table 3.10: Firms using bank loans in 2006
Number of firms Percentages
Sample 256 (626) 40.90
At segment level
W-millers 105(158) 66.46
Millers 68 (184) 37.50
Wholesalers 44(101) 43.56
Retailers 13(145) 8.97
Source: own survey, conducted in 2007
3.5 Summary
This chapter describes the statistical data of our sample of respondents: rice firms
operating in the four main market segments of the rice industry. It sheds some light
on the way rice traders in the Mekong Delta work.
Our sample contains 626 rice firms, including four important types of
traders on the domestic rice market: w-millers, millers, wholesalers and retailers. It
shows that market competition is very fierce in this particular industry. It also
indicates that financial constraints are important obstacles to the rice traders of the
Mekong Delta. In addition, it shows that the use of trade credit, market structures
and several firm characteristics vary significantly across the four main segments of
the rice market. For example, we observe that there is a statistically significant
difference in firm size, firm profitability, competitiveness, and trade credit
behavior across these market segments.
Chapter 4
Trade Credit Supply and Firm
Competitiveness
4.1 Introduction
As mentioned in the introduction chapter, the main target of this thesis is to
investigate the motivations of trade credit supply in the rice market. The literature
on the determinants of trade credit supply is rather extensive. Research suggests
that trade credit is provided because suppliers have better information on
customers than banks do (Schwartz, 1974; Petersen and Rajan, 1997) and/or
because it reduces transaction costs (Ferris, 1981; Schwartz, 1974). Trade credit
supply may also be determined by the extent to which suppliers themselves have
access to external finance (Cook, 1999; Delannay and Weil, 2004; Demirguc-Kunt
and Maksimovic, 2002).
A relatively small number of papers focus on whether trade credit supply is
related to the competitiveness of firms. Most papers in this field rather discuss the
competitive environment in which firms are active. Some of these papers claim
that monopolistic suppliers may provide more trade credit, because they are better
able to enforce payments than suppliers in competitive environments (Petersen and
Rajan, 1997; McMillan and Woodruff, 1999). With strong competition, enforcing
repayment of trade credit may be more difficult, since customers can easily switch
to other suppliers. However, other papers claim that suppliers who are faced with
strong competition use trade credit as a means to sell (more) goods (Fisman and
Raturi, 2004). In a competitive environment, trade credit can be used as a tool to
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
60
reduce the likelihood of customers switching to another supplier. Both positions
are supported by some empirical evidence.
We revisit the relationship between firm competitiveness and the provision
of trade credit but we take a different approach by using data from a single industry
in one single country. This allows for controlling for industry-specific and country-
specific characteristics that may influence the relationship. In particular, we use
data from a large survey among different rice market participants in the
Vietnamese Mekong Delta. The survey contains information on over 600 millers,
w-millers, wholesalers, and retailers. We have chosen Vietnam, because trade
credit is an important source of finance, given that formal financial markets are
relatively underdeveloped. In our sample, 65 per cent of the firms report they
provide trade credit. We focus on one sector (rice) in order to control for the
impact of product characteristics on trade credit use (Wilson and Summers, 2002).
The rice sector plays a prominent role in Vietnam’s economy, especially so in the
Mekong River Delta.
Moreover, when analyzing the relationship between trade credit supply and
firm competitiveness, we take into account the fact that firm competitiveness is a
multi-dimensional phenomenon, determined by interactions between the firm and
its suppliers of inputs, customers and rivals (Porter, 2008). We measure firm
competitiveness in terms of the firm’s perceived competition from its rivals and its
bargaining power vis-à-vis its customers.
Finally, we analyze whether the relationship between some dimensions of
competitiveness and trade credit supply may be dependent on other dimensions of
competitiveness. In particular, we investigate whether the relationship between
customer bargaining power and trade credit supply of a firm depends on the extent
of perceived competition from its rivals.
Our empirical analysis documents that preventing customer switching is a
major reason why firms provide trade credit to their customers. We also show that
Chapter 4: Trade Credit Supply and Firm Competitiveness
61
the impact of customer switching on trade credit supply is stronger in more
competitive market environments.
This chapter is organized as follows. In section 4.2, we present a brief
review of the literature on the use of trade credit and firm competitiveness. In
section 4.3 we explain the country and industry context and show why the issue of
trade credit is of importance in this particular case. Section 4.4 provides a short
discussion of the survey we have used for this study. Section 4.5 then continues
with a discussion of the dataset and the methodology of measuring competition.
The empirical results are described in sections 4.6 and 4.7, after which we provide
a summary and conclusions in section 4.8.
4.2 Trade Credit Supply and Firm Competitiveness: A Survey
The literature focusing on the relationship between trade credit supply and
competitiveness is not very extensive. The available evidence is inconclusive with
respect to the nature of this relationship. Some papers explain the relationship by
pointing at the role of contract enforcement in case a customer defaults on repaying
the trade credit. These papers stress the problems of moral hazard which a trade
credit supplier may face, i.e. he/she may be confronted with the problem of non-
repayment by the customer. Non-repayment may stem from the fact that the
customer lacks the resources to pay, or it may be the result of a deliberate action by
the customer (strategic default) to capture the value of the credit received. The
latter may be prevented by using enforceable contracts. In the absence of such
contracts and/or mechanisms, the supplier may threaten to stop future deliveries.
Yet, the effectiveness of this threat depends on the extent to which the customer
needs future supplies, on the availability of alternative suppliers and/or on the
extent to which suppliers share information on defaulting clients. If alternative
supply is readily available (and assuming that customers need a continuous flow of
supplies for their business and that information sharing among suppliers is low or
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
62
absent), a customer may switch to a new supplier, which seriously reduces the
effectiveness of the threat of stopping ending deliveries. However, if the supplier is
a monopolist, it will be easier to effectively use the threat of non-delivery. This
explains why monopolistic suppliers may be more willing to provide trade credit
(Petersen and Rajan, 1997; Cunat, 2007; Giannetti et al, forthcoming).
Related to the above discussion, some papers stress the fact that
monopolistic suppliers in general have longer lasting relationships and contact their
clients more frequently than suppliers in competitive markets do. These longer
relationships provide suppliers with more information regarding the type of
customer they deal with. Due to the regular trade relationship they have with their
customers, suppliers have a better opportunity to investigate their customers’
creditworthiness and they are better able to monitor them. Therefore, long trade
relations help reduce information asymmetries, and thus the moral hazard problem.
This may again explain why monopolistic suppliers are more willing to provide
trade credit than suppliers in competitive markets (Petersen and Rajan, 1997; Jain,
2001; Atanasova and Wilson, 2003).
McMillan and Woodruff (1999) find empirical evidence for a negative
relationship between trade credit and competition, using information from trade
relations among small private firms in Vietnam based on survey data from 259
manufacturing firms. In particular, they find that suppliers provide trade credit
more frequently when customers have difficulties finding alternative suppliers,
when the supplier has information about the customer’s creditworthiness based on
long trading relations, and/or when the supplier is a member of a network of
suppliers in which information about customers is exchanged and which serves as a
way of sanctioning defaulting customers. Their conclusion is that suppliers only
provide trade credit to their trading partners if they can ensure that customers will
comply with the agreement to pay late or if they can enforce repayments.
A number of papers stress the importance of trade credit as an instrument
to improve the competitiveness of the supplier. When suppliers are confronted with
Chapter 4: Trade Credit Supply and Firm Competitiveness
63
customers who can easily switch to new suppliers because of strong competition,
trade credit can be used as an instrument to prevent customers from switching. This
may work particularly well in cases where access to bank credit for working capital
purposes is low, making trade credit valuable to the customer. One of the few
papers investigating the positive relationship between trade credit and competition
is by Fisman and Raturi (2004). Using multi-industry firm level data from five
Sub-Saharan African countries, they show that firms that are active in more
competitive markets are more likely to obtain trade credit than firms that are active
in monopolistic markets.
Van Horen (2005) in her paper focuses on a related issue regarding the
relationship between trade credit and firm competitiveness, using data on trade
credit supply for firms from different industries in 42 different countries. She
argues that the bargaining power of the supplier vis-à-vis its customers may affect
decisions to provide trade credit. In particular, the weaker the bargaining power of
the supplier in its trade relations with customers, the higher the probability that
trade credit is supplied, since potential customers may only be willing to buy from
a supplier when trade credit is supplied. In the empirical analysis, Van Horen
shows that young and small firms, which are supposed to have relatively low
bargaining power vis-à-vis their clients, have to offer more trade credit in order to
lock in their customers. Thus, her results suggest that trade credit may be used by
suppliers as a competitiveness device.
In a recent study, Fabbri and Klapper (2008) explicitly focus on trade
credit supply and its relation to the extent of a firm’s market power. Using unique
data for about 2,500 Chinese firms in the manufacturing and service sectors,
including detailed information on the market power of firms in their input and
output markets, as well as details on the terms of trade credit, they show that firms
with weak market power extend more trade credit. Moreover, they show that firms
receiving trade credit from their suppliers of input are more likely to extend trade
credit to their customers, and that these firms tend to extend trade credit against
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
64
similar contract terms (“maturity matching”) more often when they face strong
competition in output markets and/or have strong market power in input markets.
On the basis of these results, Fabbri and Klapper (2008) argue that trade credit
plays an important role in the supply chain for financially constrained firms when
coping with competition.
To conclude this overview, the above discussion shows that the empirical
literature on trade credit and the competitiveness of firms is inconclusive and
focuses on different aspects of competitiveness. While MacMillan and Woodruff
(1999) focus on the importance of the market power of the supplier, others like
Fisman and Raturi (2004), Van Horen (2005) and Giannetti et al (forthcoming)
stress the importance of the customer’s bargaining power as a determinant of trade
credit supply. Moreover, all these studies use multi-industry and/or multi-country
datasets. Fabbri and Klapper (2008) are the first to look at the importance of both
market structure and competition in determining trade credit supply, using multi-
industry data.
We follow Fabbri and Klapper (2008) by focusing on market competition
and customer bargaining power when evaluating the relationship between trade
credit supply and firm competitiveness. However, in contrast to all previous work
on this subject, we use data from a single industry (rice) in a single country
(Vietnam). Using industry-specific data allows for controlling for the fact that
product characteristics may affect the supply of trade credit (Wilson and Summers,
2002). Moreover, using country-specific data allows for controlling for the
influence of the institutional setting, such as existing formal banking and law
systems, which have shown to be important determinants of trade credit supply
(Fisman and Love, 2003; Giannetti et al, 2008).
Chapter 4: Trade Credit Supply and Firm Competitiveness
65
4.3 The Vietnamese Rice Sector
In this study we focus on Vietnam, since trade credit is considered to play an
important role in trade relations in this country (MacMillan and Woodruff, 1999).
In particular, we focus on the use of trade credit in the rice sector as rice is one of
the most important goods produced in this country. The Mekong Delta in the south
of Vietnam is an important rice producing region. Although the region covers only
12 per cent of the country’s total area, over the last number of years it has
accounted for approximately 50 per cent of total rice production and 90 per cent of
total rice exports.
Rice in the Mekong Delta is generally produced at small farms; a typical
farm producing rice has about 1-3 hectares of land. Rice farms produce paddy,
which is in most cases sold to gatherers. According to a 1995 survey by IFPRI,
gatherers purchase 95 per cent of the paddy produced by farmers (Minot and
Goletti, 2000). These gatherers then sell the paddy to milling firms. The millers
transform the paddy into rice. They may also polish the rice, which raises the
quality. In the Mekong Delta, millers account for about 80 per cent of the total
milling and polishing activities.6 After the rice has been produced and polished, the
millers sell it to wholesalers or directly to retailers, who then sell it to final
consumers in the domestic market. Next to millers that buy paddy and mill and sell
the rice themselves, there are also several millers that only provide milling
services.
The above described marketing channel focuses on rice trading for the
domestic market. Yet, a substantial part of the rice production is exported. The
marketing channel for exported rice is dominated by state-owned enterprises
(SOEs) and a few privately owned exporting firms. The SOEs and private
exporters may buy paddy directly from farms or from gatherers and mill and polish
the rice themselves. They account for 20 per cent of the total milling and polishing
6 Information obtained from the website www.stp.vn.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
66
activities in the region. Alternatively, they bring the paddy to millers or buy milled
and polished rice from millers.
The Survey
The data used in this research comes from a survey held among 626 firms that are
active in the Vietnamese rice market in six different provinces in the Mekong
Delta. These provinces are Can Tho, An Giang, Vinh Long, Tien Giang, Hau
Giang and Soc Trang. The survey was held in 2007 and has been conducted by
experienced officials of the provincial statistical offices of the National Statistical
Department. The specific questions regarding trade credit were part of a
standardized nationwide enterprise survey these statistical offices carry out every
five years. This survey is used by the government to collect information in order to
evaluate the country’s economic performance and economic policies and to plan
for future economic developments. The latest enterprise survey took place in July
2007.
Our questionnaire was directed towards the four main types of firms that
are active in the rice market, i.e. the rice millers (including pure millers and
wholesale-millers who also sell rice), rice wholesalers, and rice retailers. For each
firm the questionnaire was filled out by the manager/owner. In each of the six
provinces, we used the list of officially registered firms to select the ones that
finally participated in our survey on trade credit. Based on this list, which was
ordered alphabetically using the names of the company, we selected the first of
every five firms (e.g. number 5, 10, 15, etc.) to join the survey.7 This approach
allowed us to select the millers, wholesalers and exporters for our survey.
However, rice retailers in general are not officially registered. Rice retailers usually
have small shops at local marketplaces in towns and cities. Retailers were selected
by visiting these local marketplaces and randomly picking every first one out of ten
7 This approach was used because it turned out to be too costly to have all firms participate in the
survey.
Chapter 4: Trade Credit Supply and Firm Competitiveness
67
shops the officials of the provincial statistical offices of the National Statistical
Department came across. The gatherers officials came across at these markets were
also randomly asked to participate. Rice millers make up the largest part of our
sample (29 per cent); w-millers who also sell rice, and retailers each represent
around 25 per cent of the total sample; wholesalers account for 16 per cent of the
sample.
The survey included a number of questions related to the extent of
competitiveness of firms. All questions related to measuring competitiveness are
based on the manager/owner’s perception of competitiveness. So, in fact what we
measure is perceived competitiveness, rather than actual competitiveness, which is
mostly based on objective measures, such as market size in terms of shares in total
sales, assets, etc. We explicitly chose to focus on measuring perceptions because
objective measures of total market size in the context of local rice markets in
Vietnam are not available. Moreover, in order to understand what drives the
decision of a manager to supply trade credit, it is more important to look at
perceived competitiveness rather than competitiveness based on objective
measures.8
The importance of trade credit in the Vietnamese rice sector is
corroborated by our data. In total, 65 per cent of all firms in our sample (i.e. 410 of
626 firms) report that they provide trade credit. The importance of trade credit
varies for different types of firms. Of all rice millers who also sell rice, 92 per cent
report that they provide trade credit to their customers. For rice wholesalers and
retailers this is 73 per cent and 70 per cent, respectively; for rice millers, it is only
30 per cent.
When looking at the amount of trade credit provided, again millers who
also sell rice take the leading position. The amount of trade credit these firms
8 The downside of this approach is that perceptions can be biased and therefore over- or
underestimated. We have no reason to believe, however, that this bias is systematically related to
specific firm and/or owner characteristics.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
68
provide equals 40 per cent of the total value of their sales. For wholesalers,
retailers and millers the numbers are 30, 19 and 12 per cent, respectively. Exporters
and gatherers provide trade credit equivalent to 28 and 33 per cent of their sales.
4.4 Methodology and Data
4.4.1 The Regression Model
In order to be able to analyze the relationship between trade credit and
competitiveness with the help of data for the Vietnamese rice sector, we estimate
the following regression model:
ininmimkiki TYPEFIRMCOMPTC εββββ ++++= ,,,0 (4.1)
In this model TC is the amount of trade credit supplied (i.e. delayed payments)
divided by total sales, COMP is a vector of k variables, measuring firm
competitiveness as perceived by individual trade credit supplying firms; FIRM is a
vector of m variables, measuring several firm-specific characteristics; i represents
firms. In the empirical analysis we use this simple model to investigate the impact
of different dimensions of competitiveness on the supply of trade credit separately.
Moreover, we investigate the relationship for different types of firms that are active
in the rice market. This is captured by the vector TYPE in equation (4.1). Our
dependent variable TC has a value between 0 and 100 per cent. Since it is a
censored variable, we use a Tobit model with two sides censoring when carrying
out the estimations in this paper. In the estimations, different measures of
competitiveness are included separately in the regression model along with the
firm-specific control and firm-type dummy variables, using data from the entire
sample of firms.
Chapter 4: Trade Credit Supply and Firm Competitiveness
69
Measuring Firm Competitiveness
The literature on the competitiveness of firms, instigated by the work of Michael
Porter (1979, 1980, 1985, and 2008), puts forward the idea that analyses of
competitiveness should take into account the fact that it is a multidimensional
concept. In our analysis, we aim at explicitly addressing the different dimensions
of a firm’s competitiveness and their impact on the supply of trade credit. We
focus on two dimensions of competitiveness of firms: (1) the competitive pressure
from other suppliers (rivalry); and (2) its bargaining power vis-à-vis customers.9
Our vector COMP in equation (4.1) includes two proxies for a firm’s market power
relative to its rivals/competitors and two proxies for its bargaining power vis-à-vis
its customers. The measures we use in the analysis are defined as follows:
(1) PERCOMP is a variable that measures rivalry from other competitors in
the market as perceived by the firm owner. The competitive force from other
suppliers in the market is rated on a scale that runs from 0 (no competition) to 4
(severe competition).
(2) NORIVALS measures rivalry from other suppliers using the number of
competitors in the market the firm uses as a reference group when setting its own
selling price.
(3) BPWCUS is a measure of the bargaining power of the customer relative to
his/her supplier. It measures the percentage of total sales for which the firm can set
9 In Porter’s framework, three other dimensions are discussed: the bargaining power of a firm vis-à-
vis suppliers of inputs, the threat of new entrants, and the existence of substitutes (Porter, 1979, 1980,
and 1985). With respect to the case of the Vietnamese rice market, we have not included these three
dimensions in the analysis. First, the impact of a firm’s bargaining power vis-à-vis its own suppliers
of inputs on the supply of trade credit to its customers is indirectly taken into account by the fact that
we have information about the amount of trade credit the firm has received from its input suppliers.
The logic is that stronger bargaining power vis-à-vis suppliers of inputs increases trade credit
received, which means a firm has more resources to supply trade credit (see below). Second, the
threat of substitution is low as rice is the staple food in Vietnam, which is firmly rooted in
consumption habits. Third, as the rice market is highly competitive, and market entrance and exit of
competitors takes place on a regular basis, the threat of entrance cannot be easily separated from
market competitiveness measured in terms of rivalry from other suppliers.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
70
the price without negotiating with the customer. Note that the higher the value of
this variable, the weaker the bargaining power of the customer.
(4) SALDEC is also a proxy for supplier’s bargaining power vis-à-vis its
customers. It is measured as the expected decrease in sales due to customer
switching if the firm were to increase its price by 50 VND per kg rice (i.e. on
average a one percentage change of the price of 1 kilogram rice).
As discussed in the earlier literature on trade credit supply and firm
competitiveness, the extent of competitive pressure due to the existence of rivals in
the market may affect trade credit supply in both ways. The nature of the
relationship between the two depends on whether preventing customer switching or
problems related to contract enforcement is the predominant underlying
determinant of trade credit supply decisions. If preventing customer switching is
the predominant factor, we expect a positive association between both variables of
competitive pressure (PERCOMP and NORIVALS) and the supply of trade credit;
if contract enforcement problems prevail, we expect a negative association. With
respect to bargaining power vis-à-vis the customer, the literature suggests that trade
credit supply increases if customers have a stronger bargaining power, again in
order to prevent them from switching to competitors. Given the definitions of both
variables measuring a firm’s bargaining power vis-à-vis its customers, we therefore
expect a negative association between BPWCUS and trade credit supply and a
positive association between SALDEC and trade credit supply.
Firm-Specific Control Variables
The vector FIRM in equation (4.1) consists of a number of firm-specific control
variables. The choice of these variables is based on the existing empirical literature
on the determinants of trade credit supply. This literature shows that the main
determinants are related to access to financial sources, access to information on the
credibility of customers, and past payment performance of customers regarding
Chapter 4: Trade Credit Supply and Firm Competitiveness
71
trade credit received. On the basis of a survey of the literature, we have included
the following list of firm-specific control variables.
(5) AGE: Existing studies of the determinants of trade credit have shown that
the supply of trade credit is seriously and positively related to the availability of
financial sources. AGE is one of the variables we use to measure the availability of
financial sources to firms. Firm age is used in the literature as a measure of the
availability of credit, since older firms are believed to have a better reputation,
giving them better access to bank loans, and thus to more sources to supply trade
credit (Van Horen, 2005). AGE is measured by the logarithm of the number of
years since the firm was established.
(6) LACKCAP is a measure of the firm’s perceived access to financial sources,
ranging from 0 (no lack of capital at all) to 4 (lack of capital is an extremely
important problem). If a firm indicates it lacks financial sources, it is expected to
provide less trade credit.
(7) FSIZE is the logarithm of the value of total sales of the supplier, which is a
measure of its size. In general, larger firms have better access to sources to provide
trade credit to their clients.10
(8) DIFLOAN is a variable that measures the difficulty a firm has in obtaining
a bank loan. This variable takes the value 1 if the firm indicates it is difficult to
obtain a loan, and 0 otherwise. As with LACKCAP, a firm that indicates it is
difficult to obtain a loan is expected to supply less trade credit.
(9) MARGIN is a measure of the profit margin of a firm and is calculated as
i
ii
AP
AVCAP −. APi is the average selling price per ton of rice of firm i in 2006;
10 In other studies, this measure has also been used as a proxy for the bargaining power of the supplier
relative to its clients (Van Horen, 2005). Yet, the size of a firm is a very indirect measure of
bargaining power.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
72
AVCi reflects the average variable costs of firm i in 2006.11
The more profitable a
firm is, the more trade credit it can supply.
(10) If a firm buys inputs for which payments can be delayed – i.e. it receives
trade credit – this increases the financial sources available to supply trade credit.
The variable BUYDP measures the percentage of inputs bought on credit and is
expected to correlate positively with trade credit supplied.
(11) The variable UNPAID measures the number of times a firm has had to deal
with customers who did not paid their trade bill in time (i.e. they did not repay the
trade credit they received) during the last three years. The impact of this variable
on the supply of trade credit is ambiguous. On the one hand, the more a firm has
been faced with non-paying clients in the past, the less it may be inclined to
provide trade credit now or in the future. On the other hand, the firm may feel it
has to provide more trade credit when confronted with non-paying clients, in order
to secure final payment of the trade.
(12) Several studies have shown that the extent to which a supplier has
information on the creditworthiness of its clients influences the amount of trade
credit given (Fisman and Raturi, 2004; McMillan and Woodruff, 1999). The better
the supplier is informed, the more willing he will be to provide trade credit. The
information is obtained by having frequent contact with clients. CONTACT is a
variable measuring the frequency of face to face and/ or phone contact between
supplier and customer, which ranges from 1 (=yearly contact) to 6 (=daily contact).
Type of Firm Dummy Variables
Finally, as mentioned earlier, we investigate the relationship for different types of
firms that are active in the rice market. Given the data we have, the analysis
11 The measurement of the average variable costs differs, depending on the specific market segment
in which a firm is active. For rice millers that also sell rice, the variable costs consist of the costs of
main materials, fuel and labor; for rice-trading firms (wholesalers and retailers) these costs reflect
main materials, transportation costs and labor; and for firms providing milling services the variable
costs consist of fuel and labor.
Chapter 4: Trade Credit Supply and Firm Competitiveness
73
focuses on four different types of firms, i.e. rice milling firms performing milling
services only (MILLER); rice milling firms performing milling services as well as
rice trading (W-MILLERS); rice wholesalers (WHSALE); and rice retailers
(RETAIL). Although all four types are active on the rice market, they perform
different functions and activities. One important difference between firms is the lot
size per transaction. Bulk-breaking is important for retailers; they sell relatively
small amounts of rice to clients and they have to be located close to final
consumers. Competition for these firms is local. In contrast, in the wholesale
market distance becomes less important and competition takes place at the regional
or even national level. The value of wholesale trade transactions is usually much
larger than that of retail transactions. Due to these differences, price levels for
these types of firms differ, although price patterns may be strongly correlated (Lutz
et al, 2006).
Differences in size of transactions also occur between rice milling firms
that perform milling services only and rice milling firms that perform milling
services as well as rice trading. While the first in general have smaller transactions,
the second normally deal with larger volumes. Another difference between the
different types of firms along the rice marketing channel relates to differences in
market conditions, in particular market competition. Competition in retail markets
is usually much stronger than in wholesale markets. Similar differences exist for
the two types of millers: firms that only provide milling services are faced with
much more competition than millers that also sell the rice. These differences may
affect the extent to which different firms provide trade credit. Data on trade credit
supply provided by different firms as discussed in section 4.3.4 corroborates the
above statement.
4.4.2 Descriptive Statistics
Tables 4.1 and 4.2 provide information regarding the descriptive statistics and
correlations of the variables used in the analysis. Table 4.1 first of all shows that on
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
74
average the firms in our sample provide trade credit (see variable TC) equivalent to
25 per cent of their total sales. The amount of trade credit they receive themselves
is roughly half this percentage as the average value for BUYDP is 12 per cent.
Looking at the descriptive statistics of the measures of perceived competition from
rivals, the average value of PERCOMP of 3 (maximum is 4) suggests that
perceived competition is rather high. Moreover, firms state that on average they
focus on the pricing behavior of four rivals when setting their own selling price.
The descriptive statistics of BPWCUS and SALDEC suggest that the customer
bargaining power is rather strong. The average value of SALDEC is 60 per cent,
suggesting that a small rise of the selling price by the firm makes its customers
move to the competition. The average value of BPWCUS of 44 per cent suggests
that firms have to negotiate with the customer on the price for more than 50 per
cent of its total sales.
Table 4.2 provides some first indications of the existing associations
between variables in our dataset. First of all, especially rice millers who also sell
rice, and to lesser extent wholesalers, clearly provide the most trade credit; the
dummy variables W-MILLER and WHSALE show a relatively high correlation with
the trade credit supply variable TC. In contrast, millers, and to lesser extent
retailers, provide far less trade credit, which is borne out by the fact that the
dummy variables RETAIL and MILLER show a negative correlation with TC. Firm
size (FSIZE) correlates positively with TC, suggesting that larger firms provide
more trade credit. Correlation between BUYDP and TC is positive and relatively
high, which suggests that firms that get more trade credit themselves also provide
more trade credit to their customers.12
Most important to the research in this paper, however, is the fact that both
measures of perceived competition due to the existence of rivals in the market
12
The results for the other control variables reported in the table are less strong, as correlation
coefficients remain below 0.20. Therefore, we do not discuss them in this section but come back to
them, when appropriate, in the next section, in which we discuss the results of the econometric
investigation based on the regression model reported in equation (4.1).
Chapter 4: Trade Credit Supply and Firm Competitiveness
75
(PERCOMP and NORIVALS) show a positive correlation with TC. Moreover, the
bargaining variables (BPWCUS and SALDEC) show a negative and positive
correlation with TC, respectively, although these are less strong. These results
strongly suggest that trade credit is supplied in order to prevent customers from
switching to competitors. We further discuss this issue in the next section, when
we present the results of the econometric investigation, using the regression model
specification reported in equation (4.1).
Table 4.1: Descriptive statistics
Variables Mean Max Min Standard dev. Obs
TC 0.251 1 0 0.255 626
WHSALE 0.161 1 0 0.368 626
W-MILLER 0.252 1 0 0.434 626
MILLER 0.294 1 0 0.455 626
RETAIL 0.232 1 0 0.422 626
EXPORT 0.033 1 0 0.180 626
GATHER 0.027 1 0 0.162 626
AGE 1.877 3.36 0 0.787 625
LACKCAP 3.038 4 0 0.990 625
FSIZE 20.32 28.37 15.61 2.322 625
DIFLOAN 0.209 1 0 0.407 626
MARGIN 0.134 0.61 0 0.118 626
BUYDP 0.127 1 0 0.227 567
UNPAID 0.784 6 0 1.621 624
CONTACT 3.856 5 0 3.855 624
PERCOMP 3.004 4 0 0.977 626
NORIVALS 3.988 20 0 3.487 599
BWPCUS 0.440 1 0 0.372 593
SALDEC 62.33 100 0 37.19 609
7
7
Tab
le 4
.2:
Corr
elat
ion m
atri
x
S
AL
DP
W
HS
A
LE
W-
MIL
L
MIL
LE
R
RE
TA
IL
AG
E
LA
CK
C
FS
IZE
D
IFL
O
MA
RG
B
UY
DP
U
NP
A
ID
CO
NT
A
C
PE
RC
O
NO
RI
V
BP
W
C
SA
L
DE
SA
LD
P
1.0
00
WH
SA
LE
0.1
64
1.0
00
W-M
ILL
0.2
94
-0
.221
1.0
00
MIL
LE
R
-0.3
26
-0
.239
-0.3
38
1.0
00
RE
TA
IL
-0.1
31
-0.2
51
-0.3
5
-0.3
84
1.0
00
AG
E
0.0
51
-0.0
08
-0.0
04
0.0
52
-0.0
54
1.0
00
LA
CK
C
0.0
47
0.0
35
0.1
40
-0.2
50
0.0
88
-0.0
35
1.0
00
FS
IZE
0.2
97
0.1
67
0.4
46
-0.3
67
-0.3
74
0.1
00
0.2
95
1.0
00
DIF
LO
-0
.179
0.0
24
-0.1
17
0.0
95
0.0
06
-0.0
41
-0.0
02
-0.0
38
1.0
00
MA
RG
-0
.094
-0.2
31
-0.0
73
0.6
12
-0.2
97
-0.0
69
-0.2
04
-0..
343
0.0
47
1.0
00
BU
YD
P
0.2
76
0.0
22
0.0
06
-0.2
94
0.2
71
-0.0
60
0.1
51
0.0
90
-0.1
49
-0.2
31
1.0
00
UN
PA
ID
0.1
75
0.0
23
-0.1
73
-0.1
50
0.2
96
0.1
16
0.1
28
-0.0
11
0.0
34
-0.1
54
0.2
15
1.0
00
CO
NT
AC
0.0
80
0.0
30
-0.1
17
-0.0
81
0.1
31
0.0
12
0.0
01
-0.0
10
-0.0
40
-0.1
03
-0.0
17
0.1
68
1.0
00
PE
RC
O
0.2
68
0.0
34
0.1
69
-0.1
60
-0.0
89
0.0
05
0.3
27
0.2
40
-0.0
50
-0.0
87
0.1
04
0.0
15
-0.0
68
1.0
00
NO
RIV
0.2
23
-0.0
11
0.1
95
-0.1
08
-0.0
48
-0.0
28
0.1
38
0.0
75
-0.1
73
0.1
023
0.1
67
-0.1
04
-0.2
03
0.2
57
1.0
00
BP
WC
-0
.169
-0.1
10
-0.2
11
-0.0
25
0.3
57
-0.0
01
0.0
10
-0.2
15
-0.1
28
-0.0
46
0.1
56
-0.0
38
0.0
08
-0.0
59
0.0
34
1.0
00
SA
LD
E
0.1
22
0.0
98
0.2
24
0.1
72
0.0
38
0.0
90
-0.0
91
0.2
12
0.1
52
0.1
51
-0.2
27
-0.1
35
0.0
84
0.1
13
-0.0
77
-0.0
30
1.0
0
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
78
4.5 Empirical Results and Discussion
4.5.1 Firm Competitiveness and Trade Credit Supply13
Table 4.3 shows the outcomes of the regressions in which the four competitiveness
variables have been added one by one to a base model specification that includes
all firm-specific and firm-type dummy variables.14
This base model specification is
presented in column 1. Thus, we have four models (presented in columns 2-5),
each including one variable measuring perceived firm competitiveness – the firm-
specific variables and a set of dummy variables referring to the different types of
firms in our dataset.
The results show that all four competitiveness measures are significantly
related to the supply of trade credit. The estimation results suggest the following.
First, we find strong evidence that increased competition from rivals in the market
is positively associated with increased supply of trade credit, as both PERCOMP
and NORIVALS are statistically significant. These results confirm that preventing
customer switching is one of the main reasons why firms to supply trade credit.
Second, we find evidence that the bargaining power of a firm vis-à-vis its
customers is associated with trade credit supply as the coefficient for BPWCUS is
negative and statistically significant, whereas the coefficient for SALDEC is
positive and statistically significant. These results again confirm that preventing
customer switching is an important concern for firms, leading them to grant trade
credit to their customers.
When we include all competitiveness variables in the empirical
specification at the same time (shown in column 6), the results remain the same,
13
We conduct a factor analysis to create new indicators reflecting firm competiveness from the
competitiveness variables used in this section. Applying the new indicators, empirical results on the
relation of trade credit supply and firm competitiveness is presented in appendix 1. 14 We left out a dummy for the millers (MILLER) in order to be able to estimate the models.
Chapter 4: Trade Credit Supply and Firm Competitiveness
79
except for the fact that in this specification the coefficient for SALDEC becomes
insignificant. Overall, the estimation results strongly suggest that firm
competitiveness plays an important role when it comes to providing trade credit to
customers. In particular, the results suggest that firms provide trade credit to their
customers in order to reduce customer switching and to keep their market share.
In almost all cases the results for the control variables are as expected. The
variables that relate to access to financial sources have the expected sign and are
statistically significant in most cases, indicating the importance of financial access
as a determinant of the supply of trade credit. LACKCAP is always negative and
significantly related to trade credit supply in all six specifications, indicating that a
lack of capital reduces trade credit provided to customers. The same results are
found for the variable DIFLOAN, which is always negative and statistically
significant in all specifications, which shows that firms that have difficulty
obtaining a bank loan provide less trade credit. With respect to FSIZE, we find that
it is positive and statistically significant in five of the six specifications, indicating
that larger firms, which are supposed to have better access to finance, provide more
trade credit. The variable AGE always has a positive coefficient but is never
significant, suggesting that older firms, which are supposed to have better access to
finance, do not supply more trade credit. The variable MARGIN is positive and
significant in all six specifications, indicating that more profitable firms generally
provide more trade credit. The variable BUYDP is always positive and highly
significant in all specifications. Thus, firms receiving more trade credit themselves,
supply more trade credit to its own customers.
The coefficient of the variable UNPAID, which measures the number of
times a firm was confronted with customers who did not paid their trade bill in
time, is always positive and highly significant. This indicates that firms provide
more trade credit when faced with non-paying clients, which may be a rational
strategy in order to secure repayment in the future.
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
80
Finally, although the variable CONTACT does have a positive coefficient
as expected, it is almost never significant. This means that in our analysis we
cannot confirm the finding of other studies, which show that firms that contact
their clients more frequently and thus may be better informed about the
creditworthiness of their clients, are more willing to provide credit to customers.
Analysis of the firm dummy variables shows that on average wholesalers,
retailers and millers selling rice provide more trade credit than millers. Moreover,
retailers seem to provide less trade credit than wholesalers and w-millers selling
rice, as the coefficient of the retail dummy is considerably lower than for the other
two dummy variables. These results confirm our earlier discussion of differences
with respect to trade credit supply between different types of firms.
4.5.2 Customer Bargaining Power and The Role of Perceived
Competition
In the previous analysis the impact of perceived competition and the firm’s
bargaining power vis-à-vis its customers on the supply of trade credit have been
investigated separately. However, it can be argued that both dimensions of
competitiveness are, at least potentially, interrelated. In particular, a firm’s
bargaining power vis-à-vis its customers may be weaker if there are more
alternative suppliers in the market. In this situation, the threat of switching to
another supplier is more credible and a firm may then have more reasons to
provide trade credit to retain its sales to existing customers. We may therefore
hypothesize that at higher levels of perceived competition, the effect of bargaining
power on the supply of trade credit may be stronger, whereas at lower levels of
perceived competition the effect may be smaller or even absent.
We test this hypothesis as follows. We split the total sample of firms in
two sub-samples, one including firms with high values of perceived competition
and one including firms with low levels of perceived competition. We create the
Chapter 4: Trade Credit Supply and Firm Competitiveness
81
two sub-samples using our two measures of perceived competition, PERCOMP
and NORIVALS. With respect to the variable PERCOMP firms are in the low
competition sub-sample if they report 0, 1 or 2 for this variable; they are in the
high competition sub-sample if they report 3 or 4 for this variable. With respect to
the variable NORIVALS firms are in the high competition sub-sample if they report
a number of rivals above the average number of rivals reported by all firms in the
sample (i.e. 3.9); they are in the low competition sub-sample if the reported
number of rivals is below the sample’s average.
Next, we run the baseline model specification as shown in column [1] of
table 4.3 for each of these sub-samples and include one of our two bargaining
power variables BPWCUS and SALDEC. Thus, we get eight model outcomes, i.e.
outcomes for the high and low PERCOMP samples using BPWCUS as the measure
for bargaining power (columns [1] and [2]); outcomes for the high and low
PERCOMP samples, using SALDEC as the bargaining power measure in the
regression (columns [3] and [4]); outcomes for the high and low NORIVALS
samples, using BPWCUS in the regression (columns [5] and [6]); and outcomes for
the high and low NORIVALS samples, using SALDEC as the bargaining power
measure (columns [7] and [8]).
The results of the eight models are presented in table 4.4. The results
provide the following general picture. When using PERCOMP to split the sample,
the results in columns [1] and [2] show that for the high and low PERCOMP
samples the coefficient of BPWCUS in both cases is indeed negative; it is
statistically significant only for the high PERCOMP sample. Yet, the size of the
coefficients for both sub-samples shown in columns [1] and [2] are not
significantly different from each other (-0.102 versus -0.092). This does not
support our hypothesis. A similar outcome can be observed when comparing the
results in columns [3] and [4] (high and low PERCOMP samples, using SALDEC
as our bargaining power measure) and [5] and [6] (high and low NORIVALS
samples, using BPWCUS as the bargaining power measure). Only when comparing
Trade Credit in The Rice Market of The Mekong Delta in Vietnam
82
the outcomes for the NORVALS samples using SALDEC as our measure of
bargaining power, we find supportive evidence for the hypothesis. For these
samples, we observe a statistically significant positive coefficient for the high
NORIVALS sample, which is also larger and significantly different from the
coefficient found for the low NORIVALS sample (0.194 versus 0.051; see columns
[7 and [8]).
Overall, then, our results provide hardly any supportive evidence for the
hypothesis that at high levels of perceived competition, the effect of bargaining
power on the supply of trade credit is strong, whereas at low levels of perceived
competition the effect is small or even absent. These results indicate that the
different dimensions of competitiveness we have distinguished in this study
independently from each other are important determinants of trade credit supply. In
other words, high perceived competition as well as strong bargaining power of
customers may both provide strong incentives to firms to provide trade credit to
their customers
4.6 Summary and Conclusions
In this chapter we have investigated the relationship between trade credit supply
and firm competitiveness. We have looked at important dimensions of
competitiveness, i.e. the competitive pressure from other suppliers (rivalry) and a
firm’s bargaining power vis-à-vis its customers. To control for the influence of
country and industry effects, we have used data on trade credit supply for firms in
one industry in one country. In particular, we have used data from the rice market
of the Mekong Delta in Vietnam to study the relationship between competitiveness
and trade credit supply.
Overall, we find strong evidence for the fact that there is a relationship
between trade credit and firm competitiveness. First of all, stronger market
competition leads to a pressure to provide more trade credit to customers in order
Chapter 4: Trade Credit Supply and Firm Competitiveness
83
to retain market shares and avoid customer switching. Second, when focusing on a
firm’s bargaining power vis-à-vis its clients, the results show that trade credit
supply is positively related to strong customer bargaining power. Again, this
supports the idea that firms provide trade credit in order to retain market shares and
avoid customer switching. Further elaborating on these outcomes we hypothesize
that customer bargaining power may be stronger if there are more alternative
suppliers in the market. The threat of switching to another supplier is more
credible, providing stronger incentives to provide trade credit in order to retain its
sales to existing customers. However, we do not find supportive evidence for this
hypothesis.
These results indicate that the different dimensions of competitiveness we
have distinguished in this study independently from each other are important
determinants of trade credit supply. Thus, high perceived competition and strong
bargaining power of customers both provide strong incentives to firms to provide
trade credit to their customers. Generally speaking, therefore, our results indicate
that studies measuring competitiveness as a one-dimensional concept fail to cover
relevant elements of competitiveness that may influence the provision of trade
credit.
Our research has added new evidence to the debate on the determinants of
trade credit. In particular, it makes a convincing case for relating trade credit
supply to the competitiveness of firms. Trade credit supply is used as an instrument
to keep customers from switching in the rice market in the Mekong River Delta of
Vietnam. Future research should verify to what extent our results are corroborated
in other market settings.
84
Tab
le 4
.3:
Tobit
reg
ress
ions
of
the
rela
tionsh
ip b
etw
een t
rade
cred
it s
upply
and c
om
pet
itiv
enes
s (d
epen
den
t var
iable
: per
centa
ge
of
tota
l
sale
s on d
elay
ed p
aym
ents
)
Var
iab
les
[1]
[2]
[3]
[4]
[5]
[6]
WH
SA
LE
0.3
54
(6.9
9)*
**
0.3
57
(7.1
6)*
**
0.3
70
(7.1
1)*
**
0.3
22
(6.2
0)*
**
0.3
49
(6.9
2)*
**
0.3
31
(6.3
2)*
**
W-M
ILL
ER
0.3
98
(8.0
2)*
**
0.3
81
(7.8
1)*
**
0.3
96
(7.8
1)*
**
0.3
74
(7.5
4)*
**
0.3
97
(8.0
3)*
**
0.3
59
(7.2
6)*
**
RE
TA
IL
0.1
78
(3.8
2)*
**
0.1
91
(3.9
5)*
**
0.1
81
(3.6
0)*
**
0.1
76
(3.5
7)*
**
0.1
97
(3.9
1)*
**
0.2
01
(3.9
8)*
**
AG
E
0.0
19
(1.0
8)
0.0
18
(1.0
3)
0.0
19
(1.0
9)
0.0
20
(1.1
8)
0.0
17
(0.9
8)
0.0
16
(0.9
3)
LA
CK
CA
P
-0.0
25
(-1
.73
)**
-0.0
48
(-3
.17
)***
-0.0
31
(-2
.17
)**
-0.0
17
(-1
.19
)*
-0.0
21
(-1
.49
)*
-0.0
46
(-2
.98
)***
FS
IZE
0
.01
6
(1.8
4)*
*
0.0
17
(1.9
1)*
*
0.0
19
(2.1
5)*
*
0.0
14
(1.6
1)*
0.0
13
(1.4
8)
0.0
17
(1.8
8)*
DIF
LO
AN
-0
.12
4
(-3
.42
)***
-0.1
24
(-3
.50
)***
-0.1
09
(-2
.97
)***
-0.1
41
(-3
.84
)***
-0.1
29
(-3
.56
)***
-0.1
30
(-3
.54
)***
MA
RG
IN
0.6
03
(3.9
0)*
**
0.6
07
(4.0
1)*
**
0.5
32
(3.3
0)*
**
0.5
40
(3.5
0)*
**
0.5
66
(3.3
8)*
**
0.4
70
(2.9
7)*
**
BU
YD
P
0.3
14
(5.1
3)*
**
0.2
97
(4.9
4)*
**
0.2
70
(4.3
9)*
**
0.3
23
(5.3
4)*
**
0.3
23
(5.3
1)*
**
0.2
74
(4.6
2)*
**
8
5
UN
PA
ID
0.0
46
(5.2
1)*
**
0.0
44
(5.0
8)*
**
0.0
51
(5.6
9)*
**
0.0
44
(5.0
2)*
**
0.0
46
(5.2
5)*
**
0.0
46
(5.3
6)*
**
CO
NT
AC
T
0.0
06
(0.5
5)
0.0
11
(0.9
4)
0.0
14
(1.2
0)
0.0
17
(1.3
7)
0.0
01
(0.0
8)
0.0
30
(2.3
8)*
*
PE
RC
OM
P
0.0
64
(4.3
2)*
**
0
.06
7
(4.2
9)*
**
NO
RIV
AL
S
0
.01
4
(3.3
6)*
**
0.0
13
(3.1
3)*
**
BP
WC
US
-0.0
93
(-2
.45
)**
-0
.08
1
(-2
.05
)**
SA
LD
EC
0
.08
0
(1.9
5)*
*
0.0
29
(0.6
9)
CO
NS
TA
NT
-0
.51
2
(-2
.72
)***
-0.6
49
(-3
.45
)***
-0.6
28
(-3
.31
)***
-0.4
71
(-2
.45
)**
-0.4
78
(-2
.56
)**
-0.7
39
(-3
.83
)***
Lo
g l
ikel
iho
od
-1
83
.2
-17
3.9
-1
64
.5
-17
2
-18
0.3
-1
42
.2
Pse
ud
o R
2
0.3
62
0
.39
5
0.4
12
0
.37
6
0.3
63
0
.46
9
Ob
serv
ati
on
s 5
26
5
26
5
02
5
09
5
22
4
85
86
Tab
le 4
.4:
Tobit
reg
ress
ions
of
the
rela
tionsh
ip b
etw
een t
rade
cred
it s
upply
and c
om
pet
itiv
enes
s (d
epen
den
t var
iable
: per
centa
ge
of
tota
l sa
les
on del
ayed
pay
men
ts)
Var
iab
les
[1]
= h
igh
PE
RC
OM
P
[2]
= l
ow
PE
RC
OM
P
[3]
= h
igh
PE
RC
OM
P
[4]
= l
ow
PE
RC
OM
P
[5]
= h
igh
NO
RIV
AL
S
[6]
= l
ow
NO
RIV
AL
S
[7]
= h
igh
NO
RIV
AL
S
[8]
= l
ow
NO
RIV
AL
S
WH
SA
LE
0.2
49
(4.2
2)*
**
0.6
28
(6.0
0)*
**
0.2
50
(4.3
3)*
**
0.7
51
(7.1
8)*
**
0.1
68
(2.3
4)*
**
0.4
09
(5.5
2)*
**
0.1
92
(2.6
5)*
**
0.4
58
(6.5
2)*
**
W-M
ILL
ER
0
.30
8
(5.5
9)*
**
0.5
52
(5.1
0)*
**
0.3
28
(6.0
3)*
**
0.6
24
(5.5
7)*
**
0.2
01
(2.8
4)*
**
0.4
91
(6.8
8)*
**
0.3
13
(3.0
2)*
**
0.5
25
(7.4
6)*
**
RE
TA
IL
0.0
81
(1.4
6)
0.6
00
(5.5
9)*
**
0.0
90
(1.5
9)*
0.6
93
(6.0
7)*
**
0.0
26
(0.3
6)
0.2
80
(4.1
4)*
**
0.1
20
(1.3
9)*
*
0.2
88
(4.3
0)*
**
AG
E
0.0
30
(1.5
4)*
-0.0
29
(-0
.87
)
0.0
22
(1.1
4)
-0.0
24
(-0
.70
)
-0.0
42
(1.5
9)
0.0
09
(0.4
0)*
0.0
35
(1.3
6)
0.0
08
(0.3
4)*
LA
CK
CA
P
-0.0
34
(-1
.87
)**
-0.0
52
(-2
.14
)**
-0.0
30
(-1
.69
)**
-0.0
64
(-2
.69
)**
-0.0
25
(-1
.02
)
-0.0
40
(-2
.01
)
-0.0
32
(-1
.35
)*
-0.0
41
(-2
.12
)
FS
IZE
0
.00
9
(0.8
4)
0.0
39
(2.3
5)*
*
0.0
07
(0.6
5)
0.0
47
(2.8
7)*
**
-0.0
06
(-0
.40
)
0.0
23
(2.0
2)
-0.0
04
(-0
.25
)
0.0
21
(1.8
5)
DIF
LO
AN
-0
.12
8
(-3
.06
)***
-0.1
58
(-2
.37
)**
-0.1
10
(-2
.70
)**
-0.1
70
(-2
.52
)**
-0.0
44
(-0
.56
)
-.0
13
3
(-3
.04
)*
-0.0
71
(-0
.90
)
-0.1
13
(-2
.65
)
MA
RG
IN
0.4
18
(2.3
4)*
*
1.4
01
(4.9
6)*
**
0.4
15
(2.3
1)*
*
1.4
53
(4.9
1)*
**
0.2
64
(1.0
8)*
0.6
50
(3.1
2)
0.2
78
(1.1
5)*
*
0.7
04
(3.4
0)*
BU
YD
P
0.3
86
(5.5
5)*
**
0.0
51
(0.5
0)
0.3
86
(5.5
4)*
**
0.0
60
(0.5
5)
0.4
95
(4.9
0)*
**
0.2
31
(3.0
2)*
**
0.4
83
(4.8
6)*
**
0.2
37
(3.0
4)*
**
8
7
UN
PA
ID
0.0
39
(3.8
3)*
**
0.0
65
(4.3
2)*
**
0.0
41
(3.9
9)*
**
0.0
73
(4.7
3)*
**
0.0
27
(1.5
7)*
*
0.0
56
(5.2
6)*
*
0.0
21
(1.1
7)*
*
.05
9
(5.4
6)*
**
CO
NT
AC
T
0.0
14
(1.0
9)
0.0
67
(1.8
0)*
0.0
05
(0.3
6)
-0.0
02
(-0
.07
)
0.0
04
(0.2
3)*
-0.0
36
(-1
.90
)
-0.0
13
(-0
.77
)
-0.0
08
(0.4
7)*
BP
WC
US
-0
.10
2
(-2
.32
)**
-0.0
92
(-1
.41
)
-0.1
14
(-1
.94
)**
-0.1
14
(-2
.26
)**
SA
LD
EC
0
.07
7
(1.5
9)
0.1
16
(1.6
3)
0.1
94
(2.5
3)*
**
0.0
51
(0.9
9)
CO
NS
TA
NT
-0
.20
6
(-0
.93
)
-1.4
14
(-3
.75
)***
-0.2
32
(-1
.06
)
-1.4
51
(-4
.08
)***
0.1
92
(0.5
8)
-0.7
57
(-3
.10
)**
0.0
55
(0.1
8)
-0.7
13
(-2
.99
)**
Lo
g
like
lih
oo
d
-12
3.5
-1
.19
-1
29
.3
-22
.8
-42
.5
-11
9.4
-4
2.6
-1
28
.2
Pse
ud
o R
2
0.3
58
0
.72
7
0.3
41
0
.71
0
0.3
68
0
.41
5
0.3
81
0
.39
3
Ob
serv
ati
on
s 3
72
1
37
3
79
1
43
1
68
3
41
1
69
3
53
Chapter 5
Trade Credit Supply and Customer
Characteristics
5.1 Introduction
In chapter 4 the analysis has been done by using variables at firm level, such as
competition, bargaining power, and profitability, to explain trade credit provision.
This implies that the analysis in the previous chapter can only provide empirical
evidence for variables explaining the average amount of trade credit given by a
supplier. Obviously, using firm level data does not allow us to make a direct link
between the individual user and the individual supplier of trade credit. Therefore,
this approach is broad and does not provide answers that relate to the selection
process: who do suppliers grant more trade credit to?
This chapter uses transaction level data and brings the analysis one step
further by examining the impact of customer characteristics on trade credit
provision. Regardless of the fact that a supplier provides little or a lot of credit, the
supplier does not treat all customers the same. In other words, suppliers usually
decide to grant more credit to some customers than to others. The main aim of this
chapter is to identify the characteristics that give some customers better access to
suppliers’ credit. In particular, we will concentrate at several issues that can be
measured much more precisely with transaction level data than the firm level data.
For example, the length of the trading relation, the number of visits to an individual
Trade credit in the rice market of the Mekong Delta in Vietnam
90
customer before granting credit, frequent contacts, the size of an order, the history
of payments, and the bargaining power of a customer.
Our design is straightforward. We aim to investigate to whom suppliers are
willing to provide more credit. At the same time, we control for several important
suppliers’ characteristics, e.g. market competition, lacking capital, firm size, and
age. This empirical setting is expected to provide an overall view of customer
characteristics that allow customers good access to suppliers’ credit sources.
This chapter contributes to trade credit literature along a few dimensions.
First, existing studies on this topic use cross-industry data. Yet, the use of trade
credit is significantly influenced by the traded goods and/or specific industry
characteristics (Summer and Wilson, 2002; Giannetti et al, 2008). By focusing on
one industry in one country, we eliminate the impact from different industry and/or
products characteristics on trade credit provision. Accordingly, we are able to
measure the impact of customer characteristics on trade credit provision more
carefully. Second, studies on this topic using transaction level data are rare and
they omit some important characteristics that strongly affect trade credit provision
such as customer bargaining power and market competition. We use transaction
level data and cover more customer specific variables in our empirical analysis.
The remainder of this chapter is organized as follows. In section 5.2, we
present a brief review of the literature on trade credit and customer characteristics.
Section 5.3 introduces the hypothesis, the empirical strategy and data. We discuss
empirical results in section 5.4, and section 5.5 concludes the chapter.
5.2 Related Literature
In the literature, studies investigating the impact of customer characteristics on
trade credit provision are rare and can be divided into two categories. The first
category of studies uses firm level data and stresses the importance of customers’
bargaining power on trade credit provision. For example, Summer and Wilson
Chapter 5: Trade Credit Supply and Customer Characteristics
91
(2003) use surveyed data of small businesses in the UK and show that the size of
buyers can very much affect trade credit provision. They employ a dummy variable
that takes the value of 1 if firms agree with the statement that their markets are
dominated by large buyers as the main characteristic of customers. The authors
find that firms provide more credit if large clients dominate their markets. As a
conclusion, the authors argue that selling to large clients induces small suppliers to
provide more credit to attract large buyers.
Another study that focuses on the relation between trade credit provision
and customers’ bargaining power is Van Horen (2007). Using a dataset of 5,164
firms from 20 countries in Eastern Europe and Central Asia, Van Horen (2007)
measures the customers’ bargaining power by a dummy variable that takes the
value of 1 if the percentage of sales to the three largest customers is more than 20
per cent. This study proves that the customers’ market power has a positive impact
on trade credit provision. The author argues that customers with strong market
power would demand to purchase goods on credit to increase their surplus.
Accordingly, suppliers that sell to large firms with high market power are inclined
to extend more credit.
Fabbri and Klapper (2008) use data of 2,500 Chinese small and medium
enterprises and investigate the impact of customer bargaining power and market
competition on trade credit provision. The authors use a dummy variable that takes
the value of 1 if the percentage of total sales accounted by the largest customer is
larger than 5 per cent and 0 otherwise, as an indicator of customer bargaining
power. This study also shows a positive effect of customers’ bargaining power on
trade credit provision.
The second category of studies employs transaction level data and
emphasizes the effect of customer creditworthiness on trade credit provision: e.g.
McMillan and Woodruff (1999), and Aaronson et al (2004). McMillan and
Woodruff (1999) use a sample of 259 Vietnamese firms and analyze trade relations
between each firm and two specific customers. The study investigates how the
Trade credit in the rice market of the Mekong Delta in Vietnam
92
customers’ ability to switch to another supplier and characteristics of trading
relations influence a supplier’s willingness to provide credit. The authors show that
if it is difficult to find an alternative supplier for the traded goods, which can be
interpreted as low market competition, suppliers are more willing to provide credit.
The major finding is a negative relation between market competition and trade
credit provision, which contrasts with the positive relation found in other studies
(Fisman and Raturi, 2004; Van Horen, 2005). In addition, the authors also
document that more trade credit will be granted to a customer if his suppliers
receive more information on customers through business networks or trading
relationships.
Aaronson et al. (2004) use surveyed data from their own survey and study
trade relations of small businesses in Chicago. This study examines the impact of
the ability to collect information about customers, with a special focus on
geographic distance and ethnic ties on trade credit, using data at the transaction
level. The study shows that working with a nearby supplier allows customers to
receive more credit from suppliers. Moreover, it also indicates that Hispanic firms
are more likely to get more credit from firms owned by other Hispanics. The
authors argue that suppliers may know more about customers located nearby
and/or belonging to the same ethnic group.
In general, customers’ bargaining power and creditworthiness are found to
be important factors with respect to trade credit provision. Yet, existing studies on
this issue employ cross-industry analysis and do not control for the impact of the
transacted goods and industry characteristics that affect trade credit provision
strongly (Wilson and Summer, 2002; Giannetti, Burkart and Ellingsen, 2009).
Second, these studies omit several important factors that may influence trade credit
provision. For example, McMillan and Woodruff (1999), and Aaronson et al.,
(2004) do not control for customers’ bargaining power, which significantly
influences trade credit provision.
Chapter 5: Trade Credit Supply and Customer Characteristics
93
The empirical analysis in this chapter will cover all above mentioned
variables that affect trade credit supply. For example, the size of the orders,
customers’ bargaining power, the length of a trading relation with his suppliers, the
number of times a customer does not pay suppliers in time, the frequency of
contact and orders, etcetera, will be included. Consequently, we expect to cover a
broader range of customer specific characteristics than previous studies have.
5.3 Hypothesis and Methodology
5.3.1 Hypothesis Development
Customer creditworthiness can be a significant issue that suppliers need to take
into account when considering whether or not to provide credit to a customer.
Suppliers may prefer to grant credit to customers they know well, so that they can
be sure that clients will repay. Consequently, we hypothesize that firms may supply
more credit to customers they know better through (a) visiting the customer’s
business before providing credit, (b) the length of the trading relation, and (c) the
customer’s habit of repaying in time.
H1a: Suppliers grant more credit to customers who they have visited before.
H1b: Firms grant more credit to customers with whom they have a longer trading
relation.
H1c: Firms grant less credit to customers that in the past were unable to repay in
time.
Several studies also suggest that suppliers offer trade credit to build a
relation with large customers to generate sales. Therefore, firms provide more
credit to customers who order large quantities since they are high- potential
customers. It may also be the case that large customers have high bargaining power
and demand more credit from suppliers. Consequently, suppliers have to grant
Trade credit in the rice market of the Mekong Delta in Vietnam
94
more credit to large buyers/buyers with high bargaining power (Summer and
Wilson, 2002; Van Horen, 2007).
H2a: Firms are more willing to provide credit to clients that order large
quantities.
H2b: Firms are more likely to provide credit to customers that have a high
bargaining power.
5.3.2 Empirical Methodology
As mentioned earlier, this chapter aims to provide an insight into the suppliers’
selection process as to who they will grant credit. Consequently, it examines the
link between trade credit supply and customer characteristics based on data
collected through our questionnaire. The data contains detailed information about
trade credit at the transaction level, characteristics of trading relations among
suppliers and specific customers, customer characteristics and some relevant
supplier characteristics.
We estimate the following regression model:
DPCUS ij = α + βCustomercharacteristicij + δ Xi + ε ij (5.1)
In our specifications, the dependent variable is the proportion of the payments that
customer j is allowed to delay in making payments to supplier i. Customer
characteristics are the core explanatory variables. Important measures of
customers’ characteristics are employed to capture the impact of clients’
characteristics on trade credit supply. These measures will be explained in the next
section. X is a vector of control variables including supplier characteristics and
market characteristics that may affect supplier incentives to offer credit to
customers. In general, the following set of variables will be employed in our study.
Chapter 5: Trade Credit Supply and Customer Characteristics
95
Empirical Measures
Our dependent variable DPCUS15
ij, being the proportion of the payment made after
the delivery of goods by customer j to supplier i, always falls between 0 and 1.
Therefore, we treat this variable as a censored variable and we employ a standard
Tobit model in our study.
Information On Customer Characteristics and Trading Relations
Suppliers may be willing to provide more credit to a customer that they know
better and consider to be a creditworthy client to ensure repayment. The first set of
variables attempts to reflect customer and trade relation characteristics that can
affect a supplier’s incentives to provide credit to a specific customer.
(1) PAYLATEij: the numbers of times customer j has not paid supplier i on time
during the last three years.
(2) LOGLENGREij: the natural log of the length of the trading relation in days
between supplier i and customer j.
(3) VISITij: the number of times supplier i visited the shop of customer j.
(4) CONTACTij: we ask each supplier how often customer j usually contacts
supplier i, either by calling or by actually visiting, to ask about selling prices.
Frequent contact with the supplier may provide more information about the
customer’s business and leads to more trade credit to the customer (Fisman and
Raturi, 2004). In this chapter, we use CONTACTij as a scale variable ranging from
1 to 6. This variable measures the frequency of face to face and/or phone contact
between a supplier and a specific customer:, which ranges from ( 1=yearly;
2=every six months; 3=every three months; 4=monthly; 5=weekly; and 6=daily).
15
We do not define the time gap between delivery and payment in the definition of dependent
variable since it is a common practice that rice suppliers receive the payment of one order when the
next order is delivered (see McMillan and Woodruff, 1999)
Trade credit in the rice market of the Mekong Delta in Vietnam
96
The decision to grant credit to an individual customer may also be
influenced by the size of the customer’s orders and his bargaining power. Some
theoretical studies suggest that suppliers provide trade credit to encourage
customers to order in large quantities. An increasing in the proportion of credit
sales (the proportion of sales on delayed payments) will result in larger order sizes
(Yung, Fu Huang, 2007; Rachamadugu, 1989). Therefore, we include variable
IMCUSij to study whether suppliers prefer to provide credit to customers that place
large orders.
(5) IMCUSij is expected to measure the size of customer j’s average order. It is
measured by the ratio of the average order from customer j to supplier j’s average
monthly revenue.
It may also be the case that customers with high bargaining power demand
to purchase on credit. Accordingly, suppliers selling to clients with high bargaining
power might grant more credit to them (see Van Horen, 2007; Fabbri and Klapper,
2008; Summer and Wilson, 2002).
(6) SUPOWij measures supplier i’s bargaining power vis-à-vis customer j. It is
a dummy variable equal to 1 if supplier i can set the selling prices and contract
terms without negotiating in the trade relation with customer j, and 0 otherwise.
The Measurement of Supplier Characteristics
The decision to provide credit to customers is also influenced by various supplier
characteristics. Our second set of independent variables reflects some general
supplier characteristics that affect a supplier’s willingness to grant credit to
customers. First, some theoretical studies suggest that firms with good access to
external financial markets and/or excess credit at low interest rates are more
willing to provide credit to their trading partners (Schwartz, 1974; Smith, 1988).
Empirical studies use firm size and firm age as proxies for access to external
finance. Large and old firms may have better access to external financial markets,
Chapter 5: Trade Credit Supply and Customer Characteristics
97
and therefore they may be more capable of offering credit to trading partners (Van
Horen, 2005; Fisman and Raturi, 2004). In contrast, firms that face credit
constraints may not be able to provide trade credit. We pick up the effects of being
credit-constrained by using LACKCAPi in our empirical analysis. Second, some
studies also show the impact of supplier profitability on suppliers’ incentives to
provide credit to customers. For example, Petersen and Rajan (1999) show that
profitable firms are more willing to provide credit to customers. Third,
experiencing default in the past may also affect supplier incentives to grant credit.
We control for the supplier characteristics mentioned above by the following
variables.
(7) LACKCAPi is a five-scale variable measuring the importance of a lack of
capital estimated by the firms’ owners. This value can be 0=no lack of capital at
all; 1=lack of capital is not a problem; 2=lack of capital is somewhat problematic;
3=lack of capital is an important problem; and 4=lack of capital is extremely
important.
(8) FSIZE: the natural logarithm of capital is used as a proxy for firm size;
several empirical studies on this topic show the impact of firm size on trade credit
provision (Van Horen, 2005; Fisman and Raturi, 2004).
(9) LOGAGEi: the natural log of firm age, measured in years from the year a
firm has been established until 2007.
(10) BEUNPAIDi: the fact of being unpaid by clients in the past can strongly
affect sellers’ incentives to provide credit to customers. Therefore, we include
BEUNPAIDi, measured by the number of times a firm has not been paid by all of
its customers in the last three years of our study.
(11) MARGINi: we measure a firm’s profitability by estimating the price cost
margins.
i
ii
iAP
AVCAPMARGIN
−=
Trade credit in the rice market of the Mekong Delta in Vietnam
98
where APi is the average selling price per ton rice of firm i in 2006, and AVCi is the
average variable cost of firm i in 2006. The average variable cost is defined as the
total costs including the purchasing price of rice/paddy as a main material, fuel
costs, transportation costs, and labor cost.
The Measurements of Market Circumstances
The third set of variables in our study is a proxy for the characteristics of the
product market in which suppliers operate. We attempt to pick up the effect of
market competition perceived by using the following measure.
(12) PERCOMPi: a five-scale variable reflects the importance of competition as
perceived by a supplier. This value ranges from 0 to 4 (0=no competition at all;
1=competition is not important; 2=competition is relatively important;
3=competition is important; and 4=competition is extremely important).
In addition, several of the above characteristics may differ substantially
across market segments, which may affect trade credit provision. Therefore, we
include a dummy variable for each market segment to control for the market
segment characteristics:
(13) WHSALEi is a dummy variable that takes the value of 1 if a firm is a
wholesaler and 0 if otherwise.
(14) W-MILLERi is a dummy variable that takes the value of 1 if a firm is a
miller that sells rice and 0 if otherwise.
(15) MILLERi is a dummy variable that takes the value of 1 if the firm’s main
business activity is to provide milling services and is not involved in trading, and 0
if otherwise.
(16) RETAILi is a dummy variable that takes the value of 1 if a firm is a retailer
and 0 if otherwise.
Chapter 5: Trade Credit Supply and Customer Characteristics
99
5.3.3 Descriptive Statistics
The data used in this chapter comes from the 2007 rice firm survey in the rice
market of the Mekong Delta. These firms were drawn from the lists of rice firms
provided by the provincial statistic bureau in each province. The data includes rice
firms that operate at the four main market segments of the rice market in our
quantitative analysis: 158 w-millers, 100 wholesalers, 145 retailers, and 184
millers.
The core of our survey was to collect information on trade credit provision
and firm trade relations with two specific customers. We asked rice firms for
details of two trade relations with two specific customers, of whom one is one of
their larger clients while the other is a smaller client. The reason for this selection
method was that one of our targets was to measure the impact of customer
bargaining power and customer size on trade credit provision. For each of the two
specific relationships, firms provided information on the proportion of the
transaction value that was usually paid on delayed payments. Regarding the length
of the credit period, rice firms mentioned that rice suppliers usually allow clients to
settle payments for one order when the next order is delivered. In addition,
suppliers do not explicitly charge any interest on the credit sales.
Table 5.1 presents the definitions of variables and table 5.2 provides
descriptive statistics of trade credit and the above mentioned variables. Trade credit
varies considerably from one customer to another, ranging from 0 to 100 per cent
of all bills. Among the 1252 trade relations we investigated, 58 per cent of firms
provided trade credit, and the average proportion of delayed payments was 37 per
cent.
Trade credit provision also differs a lot across different market segments.
Trade credit appears to be most popular with w-millers. 87 per cent of the trade
relations of w-millers in the sample involve in trade credit. Moreover, our
wholesalers provide more credit to clients than our millers and retailers. About 71
Trade credit in the rice market of the Mekong Delta in Vietnam
100
per cent of the trade relations with rice wholesalers in the sample dealt with trade
credit provision. At the retail segment, the proportion of the trade relations
decreased to 33 per cent, and only 26 per cent of millers in the sample granted
credit to clients.
Another important part of our questionnaires deals with customer
characteristics. We observe that about 49 per cent of the surveyed firms say that
they visit customers at least once before giving them credit. Besides, the customers
of the rice firms appear to contact their suppliers on a monthly basis (mean=4). In
our sample, only 2 per cent of the surveyed firms say that the interviewed
customers used to pay late. This figure shows that the risk of being paid late is
relatively low in this industry.
As mentioned in the previous section, we measure a customer’s bargaining
power through SUPOWij (a dummy variable that is 1 if supplier i can set the selling
price and contract terms without any negotiation with customer j). About 25 per
cent of the surveyed rice firms in our sample say that they have a strong bargaining
power and that they set selling prices without negotiating.
The size of a customer is measured by IMCUSij, the ratio of customer j’s
average order to supplier i’s average total monthly sales. When the value of
IMCUSij is equal to 1, an average order by the client j equals the average sales of
one month of supplier i. In our sample, the average value of IMCUSij is 0.45, which
indicates that the average orders from interviewed clients account for two weeks
sales. This figure is high since several firms purchase in bulk. It is quite usually
that rice exporters sign contracts with foreign importers. In the next step,
SOEs/exporters sign contracts with w-millers, and it takes w-millers several
months to gather and process enough rice for a contract with an exporter. This
explains why the maximum value of IMCUSij is 7.54, which means that the average
order from the customer j equals the sales of 7.5 months of supplier i.
Chapter 5: Trade Credit Supply and Customer Characteristics
101
Table 5.1: Variable definition
Variables Definition
Measures of Trade Credit
DPCUSij
The proportion of the delayed payment, which is paid after the
delivery of goods by customer j to supplier i.
Measures of Customers’ Characteristics
IMCUSij
The ratio of an average order from customer j to his supplier i
‘s average monthly sales
SUPOWij A dummy variable takes the value of 1 if supplier i can set
the selling price in the trade relation with customer j, without
experiencing negotiation and 0 otherwise.
LOGLENGREij Natural log of the length of the relationship between supplier
i and customer j (in days)
VISITij The number of times supplier i visits customer j’s shop
before granting credit to the customer.
CONTACTij a scale variable ranging 1 to 6, showing how often a firm
contact with the specific customer, (1=yearly contact), (2=
contact every half year), (3= quarterly contact), (4= monthly
contact), (5 = weekly contact) (6= daily contact)
PAYLATEij The numbers of times customer j has not paid supplier i on
time and asked for extra extension during the last three years.
Measures of Suppliers’ Characteristics
LOGCAPi
As a measure of firm size, natural logarithm of a firm’s
capital
LOGAGEi Natural logarithm of a firm’s age, the period in years since a
firm was established to 2007
Trade credit in the rice market of the Mekong Delta in Vietnam
102
Variables Definition
LACKCAPi A five scale variable measuring the importance of lacking
capital. This value ranges from: (0 = totally no lacking
capital) (1 = lacking capital is not a problem) (2 = lacking
capital is a little problematic) (3 = lacking capital is an
important problem) (4 = lacking capital is an extremely
important problem).
MARGINi MARGINi =
i
ii
AP
AVCAP − where
APi: the average selling price per ton rice of the firm i in 2006
AVCi: the average variable cost of firm i in 2006.
BEUNPAIDi The number of times that supplier i has been unpaid by its
customers in the last three years
Measures of Market Circumstance
PERCOMPi A five scale variable showing the importance of competition.
This value ranges from (0 = totally no competition), (1=
competition is not important), (2 = competition is relatively
important), (3= competition is important), and (4= competition
is an extremely important issue)
WHSALEi A dummy variable takes the value of 1 if supplier i is a
wholesaler and 0 otherwise
W-MILLERi A dummy variable takes the value of 1 if supplier i is a miller
that sells rice and 0 otherwise
MILLER A dummy variable takes the value of 1 supplier i is a miller
that mainly provides milling service and do not involve much
in any trading activity. Otherwise, this variable takes 0.
RETAILi A dummy variable takes the value of 1 if supplier i is a retailer
and 0 otherwise
1
03
Tab
le 5
.2:
Corr
elat
ion m
atri
x
Dpscus
Imcus
Supow
Contact
Visit
Lengre
Paylate
Lackcap
Fsize
Beunpaid
Margin
Logage
Percomp
Mill-seller
Wholsa
Miller
Retail
Dp
scu
s 1
.00
Imcu
s 0
.28
1
.00
Su
po
w
-0.2
2
-0.2
2
1.0
0
Co
nta
ct
0.0
5
-0.2
3
0.1
0
1.0
0
Vis
it
0.2
3
0.1
4
-0.0
5
0.1
8
1.0
0
Len
gre
0
.07
-0
.04
0
.01
0
.19
0
.04
1
.00
Pay
late
-0
.04
-0
.04
0
.04
-0
.01
0
.07
-0
.05
1
.00
Lac
kca
p
0.0
5
0.0
9
0.0
8
0.0
6
0.1
2
-0.0
9
-0.0
8
1.0
0
Fsi
ze
0.3
1
0.2
6
-0.3
9
-0.1
2
0.2
0
-0.0
3
-0.0
2
0.1
6
1.0
0
Beu
np
aid
0
.17
0
.06
0
.10
0
.12
0
.09
0
.00
-0
.02
0
.10
-0
.15
1
.00
Mar
gin
0
.06
0
.02
-0
.07
0
.06
-0
.07
0
.01
0
.10
0
.04
0
.11
-0
.04
1
.00
Lo
gage
-0.1
1
-0.0
4
0.0
9
0.0
8
0.0
0
0.1
8
0.0
5
-0.0
9
-0.1
0
0.0
2
-0.0
6
1.0
0
Per
com
0
.24
0
.20
-0
.03
-0
.06
0
.16
-0
.02
-0
.07
0
.36
0
.27
0
.03
0
.06
-0
.08
1
.00
W-m
ille
r 0
.35
0
.48
0
.16
0
.11
0
.17
0
.05
0
.03
0
.08
0
.50
0
.06
0
.08
0
.06
0
.18
1
.00
Wh
sale
0
.11
-0
.03
-0
.03
0
.03
0
.07
0
.11
-0
.04
0
.04
0
.01
-0
.04
-0
.08
-0
.04
-0
.02
-0
.25
1
.00
Mil
ler
-0.2
6
-0.2
0
-0.3
1
-0.1
3
-0.2
0
-0.1
0
-0.0
1
-0.1
7
0.0
5
-0.1
5
0.0
7
0.0
2
-0.1
0
-0.3
6
-0.2
9
1.0
0
Ret
ail
-0
.18
-0
.22
0
.51
0
.17
-0
.08
0
.01
0
.07
0
.08
-0
.70
0
.27
-0
.11
0
.06
-0
.10
-0
.32
-0
.26
-0
.37
1
.00
Trade credit in the rice market of the Mekong Delta in Vietnam
104
Table 5.3: Summary statistic of the sample
Obs Mean Min. Max Std.
Measures of trade credit
DPCUSij 1174 0.372 0 1 0.419
Measures of customers’
characteristics
IMCUSij 1135 0.4501 0.00000315 7.537 1.055
SUPOWij 1153 0.251 0 1 0.434
CONTACTij 1170 3.814 0 5 1.365
VISITij 1108 0.921 0 6 1.244
LOGLENGREij 1162 3.304 0 6.897 1.463
PAYLATEij 1172 0.014 0 1 0.116
Measures of suppliers’
characteristics
LACKCAPi 1172 3.039249 0 4 0.983
FSIZE (Logcapi) 1172 19.291 13.815 25.018 2.135
BEUNPAIDi 1170 0.791453 0 6 1.619
MARGINi 1172 0.1313579 0 0.610 0.117
LOGAGEi 1172 1.875233 0 3.367 0.785
Measures of market
circumstances
PERCOMPi 1172 2.982935 0 4 0.976
W-MILLERi 1174 0.170 0 1 0.376
WHSALEi 1174 0.269 0 1 0.444
MILLERi 1174 0.313 0 1 0.464
RETAILi 1174 0.247 0 1 0.431
Chapter 5: Trade Credit Supply and Customer Characteristics
105
5.4 Estimation Results and Discussion
5.4.1 The Impacts of Customers’ Characteristics
We start by estimating the specific to general approach by adding variables one by
one. Next, we include all our possible variables in our model. Thus, we are able to
test the stability of the remaining variables. We also adjust for cluster effects and
we standardize the coefficients in our estimations.
First, we will focus on the effects of customer characteristics and
characteristics of trading relations on trade credit provision. The empirical results
in table 5.4 provide support for the first hypothesis that better information about a
customer results in more trade credit being offered to the customer. First, we find a
significant and positive effect of VISITij on the proportion of payment made after
delivery. The estimated coefficients in table 5.4 show that the level of trade credit
provided to a customer increases with 7 per cent if a supplier visits the customer’s
shop once (β=0.071, t=4.03). These visits may be a way for suppliers to get to
know the customer’s shop and business. Therefore, if rice suppliers visit a
customer several times, the customer will be able to get more trade credit. Second,
we observe that a longer trading relation is significantly and positively associated
with trade credit offered. The length of the trade relation can reveal more private
information about the client’s creditworthiness and the business. In addition,
frequent contact tells the supplier more about the customer. We also observe that
frequent contact enables customers to get more trade credit from suppliers. In
general, the estimation results show that a customer will receive more credit from
his suppliers if the supplier has better information on the customer’s business.
Another important characteristic that influences trade credit supply is
concerned with the customer’s creditworthiness. We proxy for creditworthiness by
the number of times a customer did not pay in time during the last three years.
Unlike our expectation, we do not find a significant and negative impact of
Trade credit in the rice market of the Mekong Delta in Vietnam
106
PAYLATEij on the proportion of trade credit provision. This empirical result
implies that asking for extra time during a credit period does not affect the amount
of trade credit received negatively.
The empirical findings in table 5.4 reveal the importance of the impact of
customer bargaining power on trade credit supply. As discussed above, we proxy
for this by using two variables: IMCUSij and SUPOWij. In addition, we find a
significant and positive relation between the variable IMCUSij and the proportion
of payments made after delivery. This evidence indicates that suppliers prefer to
provide more credit to buyers/clients who order relatively large quantities. In
addition, we observe a negative and significant impact of SUPOWij on the
proportion of the delayed payments made by a customer. The estimation
coefficient shows that if a customer cannot participate in the negotiating process
for better selling prices, the proportion of delayed payments will decrease with 40
per cent (β=-0.399, t=-6.46). This finding confirms that a customer who has little
bargaining power is less likely to obtain credit. In general, the above empirical
findings provide strong evidence for our second hypothesis that rice firms provide
more trade credit to large buyers and/or buyers with high bargaining power. By
doing so, rice firms in the Mekong Delta expect to build up a trading relationship
with large clients, attract and lock in large buyers to support present and future
sales. These empirical findings are consistent with the view of trade credit as a
marketing device to promote sales for rice firms in the sample.
5.4.2 The Effects of Market Characteristics
The estimation results in table 5.4 show that suppliers who perceive a higher
degree of competition, PERCOMPi, are more willing to provide credit. This
supports the view that firms use credit as a marketing instrument to promote sales.
In addition, we observe that rice traders at different market segments
behave differently with respect to trade credit. The amount of trade credit changes
enormously through different market segments. We find significant and positive
Chapter 5: Trade Credit Supply and Customer Characteristics
107
effects of W-MILLERSi, WHSALEi and RETAILi on the proportion of delayed
payments. The regressions show that on average w-millers are likely to grant 54
per cent more credit than millers do (β=0.544, t=6.75). At the same time,
wholesalers provide about 55 per cent more credit than millers (β=0.554, t=6.90)
while retailers provide about 39 per cent more credit than millers (β=0.394,
t=3.90).
5.4.3 The Effects of Supplier Characteristics
Finally, we examine some relevant suppliers’ characteristics that are likely to be
associated with trade credit. We find that large rice firms appear to grant more
credit to clients than small firms. The reason may be that large rice firms (large
capital) have better access to bank loans since they can use their factories or
warehouses as collateral.
In line with theoretical predictions we find that a lack of capital,
LACKCAPi, has a negative and significant effect on trade credit provision. In our
sample, rice firms that lack capital provide 8 per cent less trade credit to their
customers (β=-0.087, t=-3.36). In addition, we investigate the relation between
suppliers’ profit margins and trade credit provision. Profitable firms may have
more available cash flows and thus are able to provide more trade credit to clients
(Petersen and Rajan, 1999). However, this is not confirmed in our study.
In an attempt to measure the default risk incurred by a supplier, we include
UNPAIDi: the number of times that the supplier was not paid by clients in the last
three years. Contrary to our prediction, we observe a significant and positive
association between UNPAIDi and the proportion of delayed payments. This
evidence shows that rice firms that experienced defaults by customers in the past
continue to provide trade credit. Yet, we observe that the risk of not being paid is
rather small. The average number of times that firms were not paid in the last three
years is 0.78. In case a customer is unable to pay his debts in time, suppliers can
offer credit to help these customers survive so that the customer can pay back later.
Trade credit in the rice market of the Mekong Delta in Vietnam
108
At the same time, suppliers can generate future returns from current trade relations.
This also explains why customers who do not pay in time and ask for an extension
of the credit period do not receive less trade credit, since we note no significant
impact of PAYLATEij on trade credit supply.
11
0
Tab
le 5
.4:
The
impac
t of
cust
om
er c
har
acte
rist
ics
on t
rade
cred
it p
rovis
ion:
C
oef
fici
ents
(1)
Coef
fici
ents
(2)
Coef
fici
ents
(3)
Coef
fici
ents
(4)
Coef
fici
ents
(5)
Coef
fici
ents
(6)
Coef
fici
ents
(7)
SU
PO
Wij
-0
.451
(-7.9
7)*
**
-0
.399
(-6.4
6)*
**
IMC
US
ij
0.0
51
(4.0
7)*
**
0.0
51
(2.8
3)*
**
LO
GL
EN
GR
Eij
0.0
44
(3.0
0)*
**
0.0
34
(2.4
3)*
*
VIS
ITij
`
0.0
84
(4.7
2)*
**
0.0
71
(4.0
3)*
**
CO
NT
AC
Tij
0.0
51
(3.1
0)*
**
0.0
33
(1.8
9)*
*
PA
YL
AT
Eij
-0
.076
(-0.4
8)
-0.1
16
(-0.8
8)
FS
IZE
i 0.0
63
(3.6
5)*
**
0.0
95
(5.4
8)*
**
0.0
84
(4.8
4)*
**
0.0
83
(4.6
7)*
**
0.0
84
(4.8
3)*
**
0.0
86
(4.8
9)*
**
0.0
73
(4.1
9)*
**
BE
UN
PA
IDi
0.0
88
(6.1
2)*
**
0.0
91
(6.3
5)*
**
0.0
92
(6.1
4)*
**
0.1
02
(7.3
7)*
**
0.0
86
(5.8
3)*
**
0.0
90
(6.0
7)*
**
0.0
97
(7.0
1)*
**
LO
GA
GE
i -0
.026
(-0.9
5)
-0.0
42
(-1.4
8)
-0.0
67
(-2.3
2)*
*
-0.0
53
(-1.8
4)*
-0.0
55
(-1.9
7)*
*
-0.0
50
(-1.7
6)*
-0.0
44
(-1.5
0)
LA
CK
CA
Pi
-0.0
82
(-3.0
3)*
**
-0.0
94
(-3.3
8)*
**
-0.0
84
(-2.9
7)*
**
-0.0
85
(-3.1
5)*
**
-0.0
92
(-3.3
5)*
**
-0.0
90
(-3.1
7)*
**
-0.0
87
(-3.3
6)*
**
MA
RG
INi
0.2
15
(1.0
0)
0.2
37
(1.0
6)
0.2
42
(1.1
0)
0.3
60
(1.6
0)
0.2
04
(0.9
3)
0.2
48
(1.1
1)
0.2
55
(1.1
6)
PE
RC
OM
Pi
0.1
24
0.1
20
0.1
21
0.1
02
0.1
34
0.1
20
0.1
21
Ch
ap
ter
5:
Tra
de
Cre
dit
Su
pply
an
d C
ust
om
er C
ha
ract
eris
tics
1
11
(4.7
7)*
**
(4.5
1)*
**
(4.5
6)*
**
(3.9
0)*
**
(5.2
1)*
**
(4.5
1)*
**
(4.7
3)*
**
WH
SA
LE
i 0.6
83
(9.0
7)*
**
0.6
11
(7.8
6)*
**
0.6
09
(7.8
0)*
**
0.5
47
(6.8
9)*
**
0.6
30
(7.9
9)*
**
0.6
41
(8.2
6)*
**
0.5
54
(6.9
0)*
**
W-M
ILL
ER
i 0.7
16
(9.9
0)*
**
0.5
75
(7.4
7)*
**
0.6
24
(8.4
0)*
**
0.5
50
(7.1
1)*
**
0.6
39
(8.6
0)*
**
0.6
33
(8.5
0)*
**
0.5
44
(6.7
5)*
**
RE
TA
ILi
0.4
71
(4.9
4)*
**
0.3
33
(3.4
1)*
**
0.3
08
(3.0
8)*
**
0.2
38
(2.4
0)*
*
0.2
99
(2.9
8)*
**
0.3
26
(3.2
8)*
**
0.3
94
(3.9
0)*
**
Const
-1
.581
(-4.8
3)*
**
-2.1
7
(-6.5
2)*
**
-2.0
92964
(-6.3
2)*
**
-1.9
19
(-5.6
3)*
**
-2.1
69
(-6.5
1)*
**
-2.0
01
(-6.0
2)*
**
-1.9
63
(-5.8
1)*
**
Log l
ikel
ihood
-830.5
82
-851.1
39
-879.4
40
-829.8
45
-879.0
31
-891.0
03
-756.8
35
Pse
udoR
2
0.2
27
0.1
97
0.1
88
0.2
01
0.1
92
0.1
833
0.2
48
Obse
rvat
ions
1149
1131
1160
1106
1168
1170
1067
Trade credit in the rice market of the Mekong Delta in Vietnam
112
5.5 Conclusion
In this chapter, we use transaction level data from the survey rice market of the
Mekong Delta and continue to identify the customer characteristics that make
access to supplier credit source possible. We cover a broader range of variables
than previous studies, since these studies either concentrate on customer
creditworthiness or customer bargaining power, whereas we jointly use these
variables in our analysis. In line with our expectations, the empirical findings
indicate that suppliers do indeed have incentives to provide more credit to larger
clients and/or clients with strong bargaining power. At the same time, they are
willing to grant credit to creditworthy clients. This supports the view of the
marketing theory on trade credit, i.e. suppliers seem to provide trade credit to build
trading relations with large buyers to generate present and future sales.
Chapter 6
Trade Credit Supply in Different
Market Segments
6.1 Introduction
One of the starting points of this research project was to concentrate on one
industry in one country to eliminate the various effects on trade credit supply
caused by the differences in the development of financial institutions, and the
banking and legal systems across different countries; for example, Lu (2009)
shows that the legal system has a positive effect on the provision of trade credit.
This approach also allows us to eliminate the effect of the differences in product
characteristics and/or industry characteristics on trade credit supply; for example,
Giannetti et al (2009) show that firms producing standardized products grant less
credit than service firms. Therefore, we focus on market structure (firm
competitiveness) in one market as the major market characteristic affecting trade
credit supply.
Yet, when focusing on one industry with a homogeneous product, we still
observe that trade credit supply differs significantly across the four main market
segments: w- millers, wholesalers, retailers and millers. The coefficients of the
segment dummy variables in the previous chapters show that w-millers and
wholesalers grant much more trade credit to trading partners than millers and
retailers do. This raises an interesting question: what factors cause the large
Trade credit in the rice market of the Mekong Delta in Vietnam
114
differences in trade credit supply across market segments? Do traders who are
active in different market segments behave differently with respect to trade credit
provision?
This chapter tries to explain why trade credit supply varies across market
segments. We examine the determinants of trade credit supply for each market
segment.
The chapter is organized as follows: section 6.2 continues with the
empirical methodology. Section 6.3 describes the structural differences across
market segments that can affect trade credit provision in each segment. Section 6.4
introduces the empirical results on determinants of trade credit supply. Section 6.5
concludes the chapter.
6.2 Empirical Methodology
In this chapter, we employ the transaction-level dataset used in chapter 5, and
investigate determinants of trade credit supply for each segment of the rice market.
Also, we look at the differences in determinants of trade credit supply across
market segments and aim at gaining insights into the differences in trade credit
behavior across different market segments. We estimate equation (6.1) separately
for the four groups of firms at the four market segments: w-millers, millers,
wholesalers and retailers.
DPCUS ij = α + βCustomercharacteristicij + δ Suppliercharacteristici + ε ij
(6.1)
Chapter 6: Trade Credit in Different Market Segments
115
Empirical Measures16
Our dependent variable, DPCUSij,, is the proportion of the payment made after the
delivery of goods by customer j to supplier i. Since it always falls between 0 and 1,
we treat this variable as a censored variable and employ a standard Tobit model.
Customer Characteristics and Trading Relations
Suppliers are expected to provide more credit to a customer whom they know to
ensure repayment. Therefore, the following independent variables should reflect
customer creditworthiness and trade relation characteristics, which may affect a
supplier’s incentives to provide credit to a specific customer.
(1) PAYLATEij: the number of times customer j has not paid supplier i on time
in the past.
(2) LOGLENGREij: the natural log of the length of the trading relation in days
between supplier i and customer j.
(3) VISITij: the number of times supplier i has visited the shop of customer j in
the year before granting credit to customer j.
(4) CONTACTij: we also ask each supplier how often customer j usually
contacts supplier i, either by calling or visiting the supplier, to ask about selling
prices. Frequent contact with his/her supplier may provide more information
about a customer’s business and can result in more trade credit being provided
to the customer (Fisman and Raturi, 2004). In this study, we employ
CONTACTij as a scale variable ranging from 1 to 6. This variable measures the
frequency of face to face and/or phone contact between a supplier and a
specific customer:, which ranges from ( 1=yearly; 2=every six months;
3=every three months; 4=monthly; 5=weekly; and 6=daily).
16
The thesis is a collection of papers. There is some overlap. Those who have read the previous
chapters can skip this section since this matter has also been dealt with in section 5.3.
Trade credit in the rice market of the Mekong Delta in Vietnam
116
(5) According to the marketing theory on trade credit (see Van Horen, 2007;
Fabbri and Klapper, 2008; Hyndman and Serio, 2010), suppliers operating in a
more competitive market are highly motivated to grant credit to support sales. In
addition, the decision to grant credit to an individual customer may be influenced
by the customer order sizes and bargaining power. Several studies suggest that
suppliers provide trade credit to encourage customers to order in large quantities
and/or to establish trade relations with large clients. An increase in the proportion
of sales on delayed payments will result in larger optimal order sizes (Huang,
2007; Rachamadugu, 1989). Moreover, large customers may gain more bargaining
power over suppliers and may therefore demand to purchase on credit to increase
their surplus. Accordingly, suppliers may grant more credit to large clients and/or
clients with strong bargaining power. Therefore, the variables, PERCOMPi,
IMCUSij, SUPOWij, picking up the size and bargaining power of clients, are main
proxies for the marketing theory in our study.
(6) IMCUSij is a measure of the size of customer j’s average order. It is
measured by the ratio of average orders from customer j to the supplier i’s average
monthly revenue.
(7) PERCOMPi: this five-scale variable reflects the importance of competition
perceived by a supplier. This value ranges from 0=no competition at all,
1=competition is not important, 2=competition is relatively important,
3=competition is important, to 4=competition is an extremely important issue.
(8) SUPOWij measures supplier i’s bargaining power vis-à-vis customer j. It is
a dummy variable that equals 1 if supplier i can set the selling prices and contract
terms without any negotiations with customer j, and 0 otherwise (SUPOWij).
Supplier Characteristics
The choice of providing credit to customers is also influenced by various supplier
characteristics. The second set of variables measures some general supplier
Chapter 6: Trade Credit in Different Market Segments
117
characteristics that influence the willingness to grant credit. First, firms with good
access to external financial markets are more willing to provide credit to their
trading partners (Schwartz, 1974; Smith, 1987). In contrast, firms that face credit
constraints may not be able to provide trade credit. In our empirical analysis we
pick up the effects of being credit-constrained with LACKCAPi. Firm size and firm
age may be important determinants of trade credit provision in several studies. The
reasons may be that large and old firms have better access to external financial
markets and therefore they may be more capable of offering credit to their trading
partners (Van Horen, 2005; Fisman and Raturi, 2004). Second, some studies reveal
the impact of suppliers’ profitability on suppliers’ incentives to provide credit to
customers (Petersen and Rajan, 1999). We control for the supplier characteristics
mentioned above by the following variables.
(9) FSIZEi: natural logarithm of capital, used as a proxy for firm size; several
empirical studies on this topic reveal the impact of firm size on trade credit
provision (Van Horen, 2005; Fisman and Raturi, 2004).
(10) LOGAGEi: natural log of firm age, measured in years from the moment a
firm was established until 2007.
(11) LACKCAPi: a five-scale variable measuring the importance of lacking
capital estimated by firms’ owners. This value ranges from 0=no lack of capital at
all, 1=lack of capital is not a problem, 2=lack of capital is a little problematic,
3=lack of capital is an important problem, to 4=lack of capital is an extremely
important problem.
(12) BEUNPAIDi: being unpaid by clients in the past could negatively affect
sellers’ incentives to provide credit to customers. Therefore, we include
BEUNPAIDi measured by the number of times that a firm has been unpaid by all of
its customers in the last three years.
(13) MARGINi: we measure a firm’s profitability by estimating the price cost
margins.
Trade credit in the rice market of the Mekong Delta in Vietnam
118
MARGINi = i
ii
AP
AVCAP −
where
APi = the average selling price per ton of rice of firm i in 2006, and AVCi = the
average variable cost of firm i in 2006. Average variable cost is defined as the total
costs including the purchasing rice/paddy cost as main material, fuel cost,
transportation cost and labor cost.
6.3 Why May Trade Credit Differ Across Segments?
6.3.1 Market Context Characteristics and Trade Credit Supply
According to existing studies such as Summer and Wilson (2003), Giannetti et al.
(2009) and Hyndman and Serio (2010), suppliers differ in their willingness to
extend trade credit for several reasons such as product characteristics
(homogeneous versus differentiated products), business activity (service firms and
non-service firms); market structures (competition, buyers’ market power); intra-
industry trade (suppliers and customers operating in the same section may provide
more information about clients).
Since the characteristics mentioned above vary across different industries
and firms in different industries rely on different combinations of those industry
characteristics, trade credit supply behavior may vary substantially across different
industries. In other words, the context of industry and/or market may induce firms
in different industries/markets behave differently with respect to trade credit
supply. For example, Giannetti et al (2009) finds that service firms and firms
producing differentiated products are willing to grant more credit while firm
producing standardized products offer less credit. It also shows that industry
market structure concerning seller market power vis-à-vis buyer bargaining power
influences trade credit supply significantly.
Chapter 6: Trade Credit in Different Market Segments
119
6.3.2 Structural Differences Across Market Segments
In our study, we observed that structural differences exist even across different
market segments of the rice industry. These differences concern market structure,
customer characteristics and access to external finance, which are supposed to
influence trade credit supply strongly (Wilson and Summer, 2002; Giannetti et al,
2009; and Hyndman and Serio, 2010). The ANOVA results (table 6.1) show that
the mean differences of the variables concerning the three aspects mentioned above
are substantial and significant across the four market segments (w-millers,
wholesalers, retailers and millers). For example, 67 per cent of w-millers used bank
loans in 2006, and the average amount of bank loans was 859 million VND while
only 9 per cent of retailers had bank loans, with an average loan of 2.33 million.
Also, in previous chapters we observe that trade credit supply differs
substantially throughout the four market segments. For example, while 61 per cent
of transactions in the w-millers segment involve trade credit activities, only 20 per
cent of transactions in the miller segment do. The reason may be that the different
contexts of different market segments may induce traders in different segments
behave differently with respect to trade credit supply. This provides us incentives
to extend the analysis further in this chapter on the differences in determinants of
trade credit supply across market. In the remainder of this section, we discuss the
differences of a few variables reflecting market structure, customer characteristics
and access to external finance between the four segments.
Trade credit in the rice market of the Mekong Delta in Vietnam
120
Table 6.1: Mean Differences across different Market Segments
Variables W-miller Wholesaler Retailer Miller Sig
Trade credit supply
DPCUSij 0.61 0.48 0.26 0.20 ***
ADPMENTSi17
0.156 0 0 0 ***
Access to external finance
BANKLOANSi18
(million VND)
840 351 127 2.334 ***
LACKCAPi 3.20 3.10 3.20 2.80 **
Customer characteristics
TRANSACTION VALUEij
(million VND)
777 95.8 0.801 1.534
***
PAYLATEij 0.006 0.06 0.028 0.011 **
Supplier characteristics
FSIZEi 21.289 19.346 16.672 19.609 ***
MARGINi 0.151 0.117 0.109 0.141 **
Market structure
IMCUSij 1.517 0.482 0.072 .119 ***
SUPOWij 0.147 0.213 0.617 0.066 ***
PERCOMi 3.32 2.90 2.83 2.86 ***
NORIVALSi 5.20 3.72 3.82 3.44 ***
The one-way ANOVA test for the mean difference of firms across the four different
markets segments: w-millers, millers, wholesalers and retailers . *, ** and *** denote
significant level at 10, 5 and 1 per cent, respectively.
17 The proportion of advanced payments from customers to total sales. It usually happens when rice
exporters sign a contract and pay their suppliers in advance. 18 The total amount of bank loans on December 31, 2006.
Chapter 6: Trade Credit in Different Market Segments
121
Market Structure Across Market Segments
Market structure appears to vary significantly across the four market segments. In
particular, the degree of competition among existing rivals and customer
bargaining power appears to be less intense throughout the segments: w-miller,
wholesaler, miller and retailer.
Wholesale-Millers
Competition pressure appears to be strongest in the w-miller segment. First, w-
millers consider competition to be a more important difficulty in directing their
business (PERCOM: 3.20). Second, w-millers seem to have the highest number of
rivals compared with other private traders (5.20). Third, w-millers seem to have the
lowest bargaining power vis-à-vis customers compared with other private traders.
Indeed, the figures for IMCUSij show that the ratio of an average order from the
interviewed clients to the average monthly sales is 1.5 in the w-miller segment.
This figure indicates that w-millers’ clients usually order in bulk since the average
order is equal to 1.5 times the monthly revenues. The reason may be that 30 per
cent of the w-millers’ clients are exporters: very large firms that purchase in large
quantities. In fact, it usually takes w-millers several months to process the required
amount of rice for a contract with a rice exporter. Therefore, customers of w-
millers appear to have a strong bargaining power when it comes to negotiating
contract terms and prices. This is confirmed by the variable SUPOWij (table 6.1), in
that w-millers use negotiating procedures to set contract terms and selling prices
for 85 per cent of their sales. In other words, w-millers’ clients have such a strong
bargaining power that w-millers are able to set selling price without negotiating for
only 15 per cent of their sales. Fourth, w-millers may find it easier to collect
information on clients as well as to enforce repayments. The reason may be that 99
per cent of the clients in the w-miller segment are wholesalers (54 per cent), w-
millers (15 per cent), and rice exporters (30 per cent). These clients are registered
Trade credit in the rice market of the Mekong Delta in Vietnam
122
firms with some shops, factories that can be used as collateral to enforce
repayment.
Wholesalers
In the wholesale segment, competition pressure among existing rivals also appears
to be high. For example, wholesalers seem to pay more attention to the effect of
fierce competition on their profitability than millers and retailers (PERCOMP:
2.90). The average number of competitors in this segment is 3.7, which is higher
than for retailers. At the same time, a firm’s bargaining power vis-à-vis customers
is relatively weak in this segment. A large proportion of the wholesalers’ clients
are wholesalers, i.e. relatively large firms (67 per cent). Also, the IMCUSij variable
suggests that an average order is about 50 per cent of the average monthly sales of
wholesalers. This indicates that wholesalers’ clients often order in bulk. The
SUPOWij variable also confirms that wholesalers possess relatively little
bargaining power and have to negotiate to set the price and contract terms for about
78.7 per cent of their revenues. In short, these figures show that wholesalers have
relatively little bargaining power and face a lot of pressure through competition.
Millers
Facing a lot of competition appears to be a significant problem in the miller
segment. Millers seem to have little bargaining power vis-à-vis customers since a
large proportion of clients are w-millers (17 per cent), wholesalers (47 per cent)
and exporters (2 per cent), which are large and formal firms. This is confirmed by
the variable SUPOWij, which indicates that millers negotiate to set prices for about
93.4 per cent of their revenues.
Unlike other private rice traders, millers only provide milling services and
charge a milling fee, which is a small fraction of the total value of rice. In fact, the
average milling fee is about 50 VND/kg, which is about 1 per cent of the average
price of rice. Thus, the average value of the transaction is relatively low (see table
6.1). The low value of transactions may influence the demand for trade credit
Chapter 6: Trade Credit in Different Market Segments
123
negatively, especially since many of the millers’ clients are large firms. The reason
is that clients may not need trade credit to finance the low value transactions.
Retailers
Retailers appear to face less intense pressure through competition than traders in
other market segments. Moreover, retailers seem to have the strongest bargaining
power vis-à-vis customers as compared to other private traders. In marketplaces
retailers sell to a large number of final consumers who purchase only small
quantities. On average, IMCUS indicates that the proportion of sales accounted for
by a customer is only 7 per cent of the average monthly sales. SUPOWij also
suggests that retailers set their own selling prices without any negotiation for about
62 per cent of their total sales. This figure is the highest in this market segment,
which confirms the strongest bargaining power vis-à-vis customers of retailers.
Access to External Finance
Table 6.1 shows that access to external financing varies significantly for rice
traders at different market segments. For example, among private traders, w-
millers have the best access to bank loans. On average, 67 per cent of w-millers
used bank loans in 2006, and the average amount of bank loans was 859 million
VND. At the same time, only 9 per cent of retailers had a bank loan, with the
average loan amounting to 2.33 million VND. This difference may be explained by
the fact that w-millers are usually large and formal firms that own valuable assets
which they can use as collateral. In contrast, retailers are often small and
unregistered firms; it is therefore difficult for them to apply for a bank loan.
In addition, w-millers are the only private traders with access to advance
payments from state-owned enterprises involved in large-scale rice export. Table
6.1 shows that w-millers sometimes receive advance payments from rice exporters;
the average amount of these payments is 15.6 per cent of total firm sales. In fact,
rice exporters usually sign very large exporting contracts with foreign importers. In
Trade credit in the rice market of the Mekong Delta in Vietnam
124
order to guarantee the timely supply of sufficient rice and the required quality, rice
exporters often pay w-millers in advance to collect rice for them. It can also
happen that during the harvesting season, the government provides free credit to
state-owned enterprises to purchase and store rice to maintain the price in the
regional market above the floor price.
Customer Characteristics
Customer characteristics appear to differ fundamentally across market segments.
For example, in the w-miller segment, trade relations are mostly among large firms
since 99 per cent of the w-millers’ clients are exporters (30 per cent), w-millers (15
per cent) and wholesalers (54 per cent). These clients are registered firms and have
very large shops and/or factories. Also, in this segment clients often order in bulk.
For example, the average value of transactions is about 777 million VND.
Consequently, w-millers may find it easier and less costly to gather information
about this type of client and to enforce repayment.
Table 6.2: Types of customers
Supplie
rs
Customers
W-
miller
Export
er
Wholesal
er
Retailer Consu
mer
Registered
firm
Non-
registered
firms or
consumer
W-
miller
(%)
47
15
94
30
170
54
4
1
0
0
301
99
4
1
Miller
(%) 64
17
6
2
173
47
14
4
110
30
243
66
14
34 Wholes
aler
(%)
0
0
0
0
134
67
64
32
2
1
134
67
66
33
Retailer
(%)
0
0
0
0
0
0
38
13
252
87
0
0
290
100
Source: own survey in 2007
Chapter 6: Trade Credit in Different Market Segments
125
In the wholesaler segment, about 67 per cent of customers are other
wholesalers: registered firms with relative large shops. Yet, 32 per cent of the
clients are retailers: informal firms with very small shops and no collateral of any
real value. As a consequence, it is relatively more difficult for wholesalers to
collect information about clients and enforce repayments than w-millers.
In the miller segment, clients can be wholesalers (47 per cent), w-millers
(17 per cent) and exporters (2 per cent), which means that they are also relatively
large firms. However, a large proportion of clients is a farmer from the neighboring
area (34 per cent) – it may not be that costly to collect information about these
clients as most of them live nearby. Yet, it may not be easy for millers to enforce
repayment since farmers may not have valuable assets they can use as collateral.
In the retailer segment, trade relations are mainly between retailers and
final consumers. Again, to collect information and enforce repayments is probably
most complicated and costly for retailers since 87 per cent of retailers’ customers
are individual consumers who purchase in small quantities in marketplaces (see
table 6.2).
In addition, the size of customer orders varies significantly across the
different market segments. Table 6.1 shows that the average value of transactions
is 777 million VND in the w-miller segment and 95.8 million VND in the
wholesale segment, 0.8 million VND in the retail segment and 1.5 million VND in
the miller segment. The low value of transactions in the retail and miller segments
can influence trade credit supply for several reasons. First, the cost of searching for
information about customer creditworthiness and contract enforcement is high
compared to the value of the transaction. Second, the low value of transactions in
the retail and miller segment can also influence trade credit demand, since buyers
do not need trade credit to finance their purchases.
In short, we observe that there are some structural differences across the
four main market segments. The differences refer to access to market structure,
external finance and customer characteristics, which may influence trade credit
Trade credit in the rice market of the Mekong Delta in Vietnam
126
supply significantly. This provides us with the incentive to investigate whether or
not traders in different segments behave differently with respect to trade credit
supply.
6.4 Estimation Results
6.4.1 Differences Between Market Segments: General Overview
The set of independent variables that proxy for the marketing view of trade credit
(PERCOMPi, SUPOWij, IMCUSij) shows more significant results for the w-miller
(table 6.3, column 1) and the wholesaler segments (column 2) than the market
segments of retailers (column 3) and millers (column 4).
In addition, LACKCAPi, measuring access to external finance, shows
robust and significant results for the segments of w-millers, wholesalers, and
millers. Nevertheless, a lack of capital seems to be less important to trade credit
provision for w-millers than for wholesalers and millers. The magnitude of the
coefficients is smaller for w-millers (β=-0.051, t=-1.77) than for wholesalers (β=-
.113, t=-2.35) and millers (β=-0.232, t=-3.16). In fact, w-millers have the best
access to external finance compared to other private rice traders. Furthermore,
LACKCAPi has no significant impact on trade credit supply in the retailer segment.
The reason may be that retailers are very small shops and have limited access to
bank loans: only 9 per cent of the retailers had a bank loan in 2006. In fact, a lack
of capital may be a common problem to most retailers. As a result, this may not be
one of the main factors that determines trade credit supply in this segment.
The set of variables measuring the information on customer
creditworthiness also shows remarkable differences across different market
segments. For example, CONTACTij only shows significant results in the w-miller
and retailer segments. Yet, VISITij is only relevant in the wholesaler and miller
Chapter 6: Trade Credit in Different Market Segments
127
segments. In fact, w-millers sell to very large and formal firms. In this case,
frequent contact is expected to provide one with
sufficient information to make the credit-granting decision. Unlike the case of w-
millers, 33 per cent of wholesalers’ clients are retailers in marketplaces (very small
and informal firms), and 34 per cent of millers’ customers are farmers located in
the neighborhood. Apparently, visiting shops (houses) leads to more information
than contact by telephone. This explains why visits appear to influence trade credit
supply significantly and positively in the wholesaler and miller segments. The
retail segment in particular is different from the other segments. In the other market
segments, clients are rice firms that may have frequent contact to exchange
information about prices, exporting quotas, clients, etcetera. In the retail segment,
individual consumers contact only to purchase. Therefore, when there is more
contact, purchases will be more frequent and the quantity per order will be smaller.
Accordingly, consumers who contact frequently do not need trade credit. This may
explain why frequent contact is negatively associated with trade credit supply in
the retailer segment.
The above outcome indicates that there are major differences in trade
credit determinants across market segments. In order to understand the differences,
the next section will look closer at each segment.
6.4.2 A Close Look at Each Segment
Wholesale-Miller Segment
The most outstanding feature of trade credit supply determinants for the w-miller
segment is the significant results of the set of variables that represents the
marketing theory on trade credit (PERCOMPi, SUPOWij, IMCUSij). First, we
observe a significant and positive effect of PERCOMPi on trade credit supply. W-
millers perceive a higher degree of market competition causing them to grant more
credit to their trading partners. Second, w-millers have incentives to grant more
Trade credit in the rice market of the Mekong Delta in Vietnam
128
credit to large clients and clients with strong bargaining power. The coefficient of
SUPOWij in table 6.3 shows that customers receive about 30 per cent less credit if
they do not have bargaining power and do not negotiate prices and contract terms.
At the same time, a significant and positive association between trade credit supply
and IMCUSij indicates that w-millers provide more credit to large clients. The
explanation may be that large clients and clients with high bargaining power are
usually the clients with a high potential to generate sales. Therefore, w-millers may
grant more credit in order to develop a trading relation with them. These empirical
findings suggest that the marketing theory, which explains trade credit as a mean to
encourage sales, is among the motives that determine trade credit supply in the w-
miller segment.
In fact, it is worth recalling that the transaction value in the w-millers
segment is very large. As a result, trade credit may be an attractive option for
clients to finance their purchases. Accordingly, customers may demand much trade
credit, making trade credit an efficient means for establishing trade relations with
large clients.
As mentioned in the previous section, w-millers may have less difficulty in
gathering information about clients and enforcing repayment, since many of their
clients are relatively large firms that own valuable collateral such as factories and
shops. Yet, w-millers appear to face a relatively lower risk of default. In our
sample, only 0.6 per cent of the w-millers’ transactions were paid late during the
last three years. It may explain why the set of variables measuring information
about customer creditworthiness shows less relevant results with respect to w-
millers. In particular, we do not find a significant effect of the variables VISITij,
LOGLENGREij,, PAYLATEij on trade credit supply.
Wholesaler Segment
The set of variables representing the marketing theory on trade credit shows
significant results in the wholesale segment (PERCOMPi, SUPOWij, IMCUSij,).
Chapter 6: Trade Credit in Different Market Segments
129
The positive and significant impact of PERCOMPi indicates that wholesalers grant
more credit if they perceive considerable market competition. Furthermore,
wholesalers seem to offer credit to large customers (IMCUSij,: β=0.015, t=1.92)
and customers with high bargaining power (SUPOWij: β=-0.251, t=-2.38). This is
consistent with the marketing theory, which suggests that firms give credit to
attract and lock in high potential clients to generate sales now and in the future.
Unlike the w-miller segment, wholesalers may face a higher risk of default
since their clients include relatively many small and unregistered firms (33 per
cent). Consequently, wholesalers have more incentives to invest in assessing a
client’s creditworthiness than w-millers. First, it is shown in table 6.3 that if
wholesalers visit a customer’s shop before granting credit, they will provide more
trade credit. The location and the shop of retailers/wholesalers may reveal
information about the success of a rice retailer. Second, the significant and positive
coefficient of LENGREij shows that wholesalers are willing to grant more credit to
customers who have been in a trade relation for a longer period. Such a relation
may reveal a lot of information concerning customers and may lead to a closer
relation that facilitates trade credit. Third, wholesalers also seem to provide less
credit to clients who have not been able to pay them on time in the last three years.
The variable PAYLATEij only shows significant results for the segment of
wholesalers. In this segment, customers who asked for an extension of the credit
period in the last three years receive about 27 per cent less credit (β=-0.268, t=-
2.02). The reason may be that the risk of default may be more real for wholesalers:
wholesalers’ customers include 33 per cent of all retailers, with very small shops or
seats at marketplaces and who own no valuable assets to use as collateral. Also,
retailers are usually non-registered. As a consequence, wholesalers may have
difficulty enforcing repayments with such clients. This may explain why
wholesalers are more careful in assessing customer creditworthiness than w-
millers.
Trade credit in the rice market of the Mekong Delta in Vietnam
130
Miller Segment
We observe that the set of variables that represent the marketing theory is less
relevant in the miller segment than for w-millers and wholesalers. In particular, the
significant and positive coefficient for PERCOMPi indicates that pressure through
high competition is relevant to millers supplying credit. Yet, we find no significant
effect of IMCUSij and SUPOWij on trade credit provision. This suggests that millers
do not have strong incentives to grant credit to large customers or customers with
high bargaining power, which is different from the case of w-millers and
wholesalers.
A special aspect of the miller segment is that the milling fee is only about 1
per cent of the total value of rice and the average value of transactions is very small
(see table 6.1). This suggests that it is relatively easy for large clients like w-millers
and wholesalers to pay bills immediately. Consequently, large clients may not need
trade credit to finance their purchases. Accordingly, millers do not supply trade
credit to large clients and/or clients with a lot of bargaining power. In fact, large
clients in the miller segment may care more about product quality than about the
possibility of trade credit. This explains why millers provide relatively little trade
credit, which they do not seem to grant to large clients. In contrast, millers grant
credit to clients that have been good trading partners. We observe that in this
segment VISITij is among the relevant determinants of trade credit supply.
Retailer Segment
The results in table 6.3 show that the set of variables representing the marketing
theory also shows less significant results for the retail segment. In column 4 we
note a negative and significant association for SUPOWij. Yet, PERCOMPi and
IMCUSij do not have a significant effect on trade credit supply. Retailers do not
seem to grant more credit when they perceive market competition to be fierce;
Chapter 6: Trade Credit in Different Market Segments
131
retailers do not to offer credit to large clients either; they only seem to grant credit
when their clients have bargaining power and demand credit.
Retailers sell to a large number of final consumers who purchase small
quantities at the marketplace. The value of transactions is low in this segment and
therefore clients may not need trade credit to finance their purchases. Only when
retailers have large clients like restaurants and these large clients demand trade
credit, retailers will grant credit to them. This may explain the negative and
significant impact of SUPOWij. on trade credit supply. Also, it is difficult and
costly for retailers to gather information about their customers’ creditworthiness.
This may be one of the problems that prevents retailers from using trade credit
frequently, even when pressure through competition is high. In fact, it is costly to
collect information on clients and to enforce repayment – especially compared to
the low value of the transactions.
6.5 Summary of the Results
In this chapter, we aimed at gaining a better insight into the causes of the
differences in trade credit supply across the various market segments. We find that
structural differences in market competition and customer characteristics are
among the factors that lead to these large differences. In particular, in the market
segments where there are inter-firm trade relations (clients are rice firms) and the
average value of transactions is large, trade credit supply is higher. In segments
where there is trade between firms and individual traders and consumers and the
transaction value is small, trade credit supply becomes less popular.
In the w-miller segment, trade relations are between relatively large firms
and the value of transactions is high, making trade credit an attractive option for
clients. W-millers also seem to have little difficulty in assessing customer
creditworthiness and enforcing repayments. Also, w-millers have good access to
bank loans. This explains why trade credit is used most extensively in this market
Trade credit in the rice market of the Mekong Delta in Vietnam
132
segment. Concerning the determinants of trade credit supply, the set of variables
representing the marketing theory is among the most relevant determinants of trade
credit supply. The results of these variables suggest that suppliers grant credit to
attract high potential clients and support sales.
In the wholesaler segment, trade relations are inter-firms relations, though
a relatively a lot of clients are retailers, with small and informal firms. Therefore,
wholesalers may find it more difficult to collect information on clients and enforce
repayment. Also, the average transaction value of this segment, which may
influence the demand of trade credit, is lower than that of w-millers. Trade credit
supply is less popular in this segment than in the w-miller segment. With respect to
trade credit determinants, the set of variables that represent the marketing theory
comprises the relevant factors driving trade credit supply in the wholesaler
segment. Suppliers tend to offer more credit to large clients with high bargaining
power. They also provide more credit when they perceive market competition to be
higher. Yet, due to difficulties in enforcing repayment, wholesalers appear to invest
more in assessing customer creditworthiness than w-millers.
Millers are special because they are service firms. Millers charge a milling
fee, which is about 1 per cent of the value of rice. As a result, the value of
transactions is low in this segment, making trade credit a less attractive option.
This may explain why the marketing theory shows less relevant determinants of
trade credit supply in the miller segment than in the segments of w-millers and
wholesalers. In particular, millers do to grant no credit to large clients and/or
clients with strong bargaining power. The reason may be that large clients do not
need trade credit to finance these small transactions. This is in contrast with the
case of the w-miller segment, in which high value transactions make trade credit
attractive to clients. Therefore, trade credit may function less efficiently in building
trade relations in this market segment. This may explain why trade credit supply is
relatively limited in the miller segment. Besides, many of the millers’ clients are
consumers and/or farmers in the neighborhood, and therefore millers may have
Chapter 6: Trade Credit in Different Market Segments
133
difficulty enforcing repayment by such clients. Millers appear to grant credit to
customers with whom they have close trade relations.
In the retailer segment, most trade relations are with final consumers, who
purchase small quantities at local markets. First, final consumers do not need trade
credit to finance these small purchases and therefore the demand for trade credit is
low in this segment. Second, it may be difficult for retailers to gather information
about clients and enforce repayments, e.g. the cost of collecting information and
enforcing repayment may be too high compared to the value of transactions. This
may explain why trade credit is used less frequently in this market segment as
compared to the wholesalers and retailers. The variables representing marketing
theory also show fewer relevant results in this segment. Only when retailers have
large clients like restaurants that demand trade credit, they may provide credit.
In short, this chapter suggests that the market segment specificity such as
the type of clients (firms or individual consumers), the value of transactions, the
nature of the business activity (service firms or non-service firms) are important
factors that influence the determinants and/or the role of trade credit supply.
Table 6.3: Determinants of trade credit supply in the four market segments:
w-millers, wholesalers, millers and retailers
Independent variables
Coefficient
(1)
W-miller
(2)
Wholesalers
(3)
Retailers
(4)
Millers
Marketing theory of trade
credit
SUPOWij
-0.297
(-4.07)***
-0.251
(-2.38)**
-0.539
(-4.41)***
-0.165
(-0.48)
IMCUSij 0.026
(2.26)**
0.015
(1.92)*
0.305
(1.34)
0.142
(0.64)
PERCOMPi 0.091
(2.78)***
0.188
(3.37)***
-0.009
(-0.20)
0.384
(4.70)***
Customer creditworthiness
LOGLENGREij 0.005
(0.35)
0.068
(2.05)**
0.012
(0.29)
0.084
(1.96)**
PAYLATEij 0.102
(1.49)
-0.268
(-2.02)**
-0.221
(-0.13)
CONTACTij 0.056
(3.31)***
-0.004
(-0.13)
-0.170
(-2.39)**
0.031
(0.47)
VISITij -0.009
(-0.42)
0.057
(1.82)*
0.04
(1.10)
0.419
(5.60)***
Supplier characteristics FSIZEi 0.001
(0.07)
0.039
(1.14)
0.310
(5.86)***
0.079
(1.12)
BEUNPAIDi 0.076
(4.72)***
0.055
(1.20)
0.099
(4.39)***
0.314
(6.63)***
LOGAGEi -0.061
(-1.81)*
-0.052
(-0.70)
-0.143
(-2.15)**
0.038
(0.39)
LACKCAPi -0.051
(-1.77)*
-0.113
(-2.35)**
-0.016
(-0.30)
-0.232
(-3.16)***
MARGINi -0.036
(-0.16)
-0.364
(-0.51)
0.729
(1.12)
0.676
(1.17)
Const 0.335
(0.68)
-0.695
(-1.01)
-4.32
(-4.67)***
-3.35
(-2.57)***
Log likelihood -104.698 -136.228 187.82 -187.669
PseudoR2 0.285 0.1646 0.2033 0.2466
Observations 270 190 283 324
Table 6.3 shows the results of equation (6.1) for each of the four groups: w-millers,
wholesalers, millers, and retailers. The standard errors are robust and adjusted for cluster
effects. The T-statistics are given in parentheses above. *, ** and *** denote significant
levels at 10, 5 and 1 per cent, respectively.
135
References:
Aaronson, Daniel and Bostic, Raphael, Huck, Paul and Townsend, Robert, 2004,
“Supplier Relationships and Small Business Use of Trade Credit”, Journal
of Urban Economics, 55, 1, 46-67
Almus, M., 2002, “What characterizes a fast-growing firm?” Applied Economics,
34, 1497-1508
Atanasova, Christina V. and Wilson, Nicholas, 2003, “Bank Borrowing Constraints
and the Demand for Trade Credit: Evidence from Panel Data”, Managerial
and Decision Economics, Vol. 24, No. 6/7, 2003, 503-514
Backer, Koen & Sleuwaegen, Leo, 2003, “ Foreign Ownership and Productivity
Dynamics”, Economic Letters, 79, 2, 177-83
Beck, T. Demirguc-Kunt, A., and V. Maksimovic, 2006, “The Influence of
Financial and Legal Institutions on Firm Size”, Journal of Banking and
Finance, 2995-3015
Blazenko, George W and Vandezande, Kirk, 2003, “The Product Differentiation
Hypothesis for Corporate Trade Credit”, Managerial Decision Economics,
24, 457-469
Bottazzi, G., Dosi, G., Lippi, M., Pammolli, F., Riccaboni, M., 2001, “Innovation
and Corporate Growth in the Evolution of the Drug Industry”, International
Journal of Industrial Organization 19, 1161-1187.
Bottazzi, G. and A. Secchi, 2003, “A Stochastic Model of Firm Growth”, Physica
A, 324, 213-219
136
Bottazzi, G., Secchi, A. and Tamagni, F, 2008, “Productivity, Profitability and
Financial Performance” Industrial and Corporate Change, 17, 711-751
Brown, J David & Earle, John S & Lup, Dana, 2005, What Makes Small Firms
Growth? Finance, Human Capital, Technical Assistence, and The Business
Environment In Romania, Economic Development and Cultural Change,
University of Chicago Press, 54, 1, 33-70,
Calvo, J.L., 2006, "Testing Gibrat's Law for Small, Young, and Innovating Firms",
Small Business Economics, 26, 2, 117-123
Chung, K.J., 1998, “A theorem on the determination of economic order quantity
under conditions of permissible delay in payments”, Computers and
Operations Research, 25, 49-52
Cook, Lisa D., 1999, “Trade Credit and Bank Finance: Financing Small Firms in
Russia”, Journal of Business Venturing, 14, 5-6, 493-518
Cunat, Vicente, 2007, “Trade Credit: Suppliers as Debt Collectors and Insurance
Providers”, Review of Financial Studies, 20, 2, 491-527
Delannay, Anne-France and Weill, Laurent, 2004, “The Determinants of Trade
Credit in Transition Countries,” Economics of Planning, 37, 3-4, 173-193
Demirguc-Kunt and Maksimovic, Vojislav, 2002, Firms as Financial
Intermediaries: Evidence from Trade Credit Data, World Bank, Unpublished
Working Paper
Dunne, P. and A. Hughes, 1994, “Age, Size, Growth and Survival: UK
companies in the1980s”, Journal of Industrial Economics, 42, 2, 115-140.
137
Ebben, Jay and Johnson, Alec, 2005, “Bootstrapping in Small Firms: An Empirical
Analysis of Change over Time”, Journal of Business Venturing, 21, 851-
865.
Ehrlich, Isaac & Gallais-Hamonno, Georges, Liu, Zhiqiang and Lutter, Randall
1994, “Productivity Growth and Firm Ownership: An Analytical and
Empirical Investigation”, Journal of Political Economy, University of
Chicago Press, 102, 5,1006-38
Evans, D.S., 1987a, “Tests of Alternative Theories of Firm Growth”, Journal of
Political Economy, 95, 4, 657-674.
Evans, D.S., 1987b, “ The Relationship between Firm Growth, Size and Age:
Estimates for 100 Manufacturing Industries”. Journal of Industrial
Economics, 35, 567-581.
Fabbri, Daniela and Klapper, Leora F., 2008, Market Power and the Matching of
Trade Credit Terms, World Policy Research Working Paper, 4754,
Washington DC: World Bank
Ferris, J. Stephen, 1981, “A Transaction Theory of Trade Credit Use”, Quarterly
Journal of Economics, 96, 247-270
Fisman, Raymond and Love, Inessa, 2002, “Trade Credit, Financial Intermediary
Development, and Industry Growth”, Journal of Finance, 58, 353-374
Fisman, Raymond and Raturi, Mayank, 2004, “Does Competition Encourage
Cooperation? Evidence from Trade Credit Relationships” Review of
Economics and Statistics, 86, 1, 345-352.
Freel, M.S., 2000, “Do Small Innovating Firms Outperform Non-Innovators?”
Small Business Economics 14, 195-210.
138
Geroski, P. A. and Toker, S., 1996, “The turnover of market leaders in UK
manufacturing industry 1979-86”, International Journal of Industrial
Organization 14, 141-158.
Geroski, P. A., K. Gugler, 2004, “Corporate Growth Convergence in Europe”,
Oxford Economic Papers, 56, 597-620
Giannetti, Mariassunta, Burkart, Mike and Tore Ellingsen, forthcoming, “What
You Sell is What You Lend? Explaining Trade Credit Contracts”, Review of
Financial Studies.
Gibrat R, 1932, Les inegalites economiques , Paris, Sirey
Griliches, Zvi and Lichtenberg Frank R., 1984, R&D and Productivity Growth at
Industry Level: Is There Still A Relationship? University of Chicago Press,
465-502
Goddard, J., J. Wilson and P. Blandon, 2002,” Panel tests of Gibrat’s Law for
Japanese Manufacturing”, International Journal of Industrial Organization,
20, 415-433
Hay, M., K. Kamshad, 1994, Small Firm Growth: Intentions, Implementation and
Impediments, Business Strategy Review, 5, 3, 49-68
Huang, Y. F, 2007, “Optimal Retailer’s Replenishment Decisions in The EPQ
Model Under Two Levels of Trade Credit Policy”, European Journal of
Operational Research, 176, 1577-1591
Huck, J.F. and McEwen, T., 1991, “Competencies needed for Small Business
Success: Perceptions of Jamaican Entrepreneurs”, Journal of Small
Business Management, 29, 4, 90-93.
139
Hyndman, Kyle and Serio, Giovanni, 2010, Competition and Inter-firm Credit:
Theory and Evidence from Firm-level data in Indonesia”, Journal of
Development Economics, 93, 88-108
Ibrahim, B. Ahmed and James R. Goodwin, 1986, “Perceived Causes of Success in
Small Business”, American Journal of Small Business, 11, 2, 41-50
Jain, Neelam, 2001, “Monitoring Costs and Trade Credit,” The Quarterly Review
of Economics and Finance, 41, 89-110.
Jovanovic, B.,1982, “Selection and the evolution of industry”, Econometrica, 50,
649-670.
Long M.S., Malitz I.B. and Ravid S.A.,1993, “Trade Credit, Quality Guarantees,
and Product Marketability”, Financial Management 22, 117-127
Lu, Yi, 2009, Trade credit and Legal System: Evidence from International Data,
Working Paper, School of Business, University of Hongkong, Hongkong
Lutz, Clemens, Praagman, Kees and Luu Thanh Duc Hai, 2006, “Rice Market
Integration in the Mekong River Delta,” Economics of Transition, 14, 3, 517-
546
Lutz, Clemens, Kemp, Ron and Dijkstra S. Gerhard, 2010, “SMEs Perceptions
Regarding Strategic and Structural Entry Barriers”, Small Business
Economics, 35, 1, 19-33.
Luu, Thanh Duc Hai, 2003, The Organization of The Liberalized rice market in
Vietnam, Center of Development Studies, University of Groningen.
McMillan, John and Woodruff, Christopher, 1999, “Interfirm Relationship and
Informal Credit in Vietnam”, Quarterly Journal of Economics, 114, 1285-
1230
140
McPherson, M.A., 2006, “Growth of Micro and Small Enterprises in Southern
Africa”, Journal of Development Economics, Vol.48, 53-277
Minot, Nicholas and Goletti, Francesco, 2000, Rice Market Liberalization and
Poverty in Vietnam, Research Report No. 114, Washington DC: IFPRI
Musso, Patrick and Schiavo, Stefano, 2008, “ The Impact of Financial Constraints
on Firm Survival and Growth”, Journal of Evolutionary Economics, 18, 2,
135-149
Nadiri. M.I, 1969, “The Determinants of Trade Credit in The US Total
Manufacturing Sector”, Econometrica, 37, 408-423
Nguyen, Ngoc Anh, Pham, Ngoc Quang, Nguyen, Chuc Dinh and Nguyen, Nhat
Duc, 2007, “Innovation and Exports in Vietnam’s SME Sector”, The
European Journal of Development Research, 20, 262-280
Nguyen, Tran Dinh Khoi and Ramachandran, Neelakantan, 2006, “Capital
Structure in Small and Medium-Sized Enterprises: The Case of Vietnam”,
ASEAN Economic Bulletin, 23, 2, 192-211
Petersen, Mitchell and Rajan, Raghuram G., 1997, “Trade Credit: Theory and
Evidence” The Review of Financial Studies, 10, 3, 661-691
Porter, Michael E., 1979, “How Competitive Forces Shape Strategy”, Harvard
Business Review 57, 2, 137-145
Porter, Michael, E., 1980, Competitive Strategy: Techniques for Analyzing
Industries and Competitors. New York: Free Press, 1980
Porter, Michael E., 1985, The Competitive Advantage: Creating and Sustaining
Superior Performance. New York: Free Press, 1985
141
Porter, Michael E., 2008, “The Five Competitive Forces That Shape Strategy”,
Harvard Business Review, 86, 1
Prais, S. J.,1974, A New Look at the Growth of Industrial Concentration, Oxford
Economic Papers 26 , 2, 273-288
Rachamadugu, R., 1989, “Effects of Delayed Payments (Trade credit) on Order
Quantity”, Journal of the Operational Research Society, 805-813
Rand, John, 2007, “Credit Constraints and Determinants of The Costs of Capital in
Vietnamese Manufacturing”, Small Business Economics, 29, 1, 1-13
Roper, S., 1997, “Product Innovation and Small Business Growth: A Comparison
of The Strategies of German, UK and Irish Companies”, Small Business
Economics 9, 523-537
Sakai, H and Takada, N, 2000, Developing Small and Medium Scale Enterprises in
Vietnam, NRI Paper No 13, Nomura Research Institute
Smith J.K.,1987, “Trade Credit and Informational Asymmetry”, Journal of
Finance 42, 863-872
Schwartz, Robert A., 1974, “An Economic Model of Trade Credit”, Journal of
Financial and Quantitative Analysis, 9, 643-57
Singh, A. and Whittington, 1975, “The Size and Growth of Firms”, Review of
Economic Studies,42, 1, 15-26
Solow, 1957, “Technical Change and the Aggregate Production Function”, Review
of Economics and Statistics, 39, 312-20
Summer, Barbara and Wilson, Nicholas, 2003, “Trade Credit and Customer
Relationship”, Managerial and Decision Economics, 24, 6-7, 439-455
142
Van Auken, Howard, 2005, “Differences in the Usage of Bootstrap Financing
among Technology-Based versus Nontechnology-Based Firms”, Journal of
Small Business Management, 43, 1, 93-103.
Van Horen, Neeltje, 2005, Trade Credit as a Competitiveness Tool: Evidence from
Developing Countries, Working paper, University of Amsterdam,
Amsterdam
Van Horen, Neeltje, 2007, Customer Market Power and The Provision of Trade
Credit: Evidence from Eastern Europe and Central Asia, Working paper,
University of Amsterdam, Amsterdam
Variyam, J. N. and D. S. Kraybill, 1992, “Empirical evidence on determinants of
firm growth”, Economics Letters, 38, 4, 31-36
Wilson, Nicholas, and Summers, Barbara 2002, “Trade Credit Terms Offered by
Small Firms”, Journal of Business Finance and Accounting, 29, 3, 317-35
Winborg, Joakim and Landstrom, Hans, 2000, “Financial Bootstrapping in Small
Business: Examining Small Business Managers’ Resource Acquisition
Behaviors”, Journal of Business Venturing, 16, 235-254.
Wijewardena, Hema and Tibbits, Garry E., 1999, “Factors Contributing To The
Growth of Small Manufacturing Firms: Data from Australia”, Journal of
Small Business Management, April, 88-95
Yasuda, T.,2005, “Firm Growth, Size, Age and Behavior in Japanese
Manufacturing” Small Business Economics, 24, 1-15
143
Appendix 1:
Factor Analysis on measures of perceived competition and bargaining
power
Factor analysis
Factor analysis/correlation Number of obs = 568
Method: principal factors Retained factors = 2
Rotation: (unrotated) Number of params = 6
Factor loadings (pattern matrix) and unique variances
Variable Factor1 Factor2 Uniqueness
BPWCUS -0.408 0.190 0.798
PERCOMP 0.3468 0.2838 0.799
NORIVALS 0.167 0.379 0.829
SALDEC 0.445 -0.189 0.766
Factor Eigenvalue Difference Proportion Cumulative
Factor1 0.512 0.216 1.293 1.293
Factor2 0.296 0.448 0.747 2.040
Factor3 -0.152 0.108 -0.384 1.656
Factor4 -0.260 0.000 -0.656 1.000
LR test: independent vs. saturated: chi2(6) = 118.26 Prob>chi2 = 0.0000
Rotation
144
Factor analysis/correlation Number of obs = 568
Method: principal factors Retained factors = 2
Rotation: orthogonal varimax Number of params= 6
Factor Variance Difference Proportion Cumulative
Factor1 0.459 0.110 1.159 1.159
Factor2 0.349 . 0.882 2.040
Rotated factor loading and unique variances:
Variable Factor1 Factor2 Uniqueness
BPWCUS -0.448 -0.037 0.798
PERCOMP 0.1602 0.419 0.799
NORIVALS -0.044 0.412 0.829
SALDEC 0.480 0.057 0.766
Based on the rotated factor loading, we name the two variables:
Factor1= CUSPOW (Customer bargaining power vis-à-vis a supplier, formed by
BPWCUS and SALDEC, same sign with SALDEC)
Factor2= COMP (Competition from existing rivals, formed by PERCOMP and
NORIVALS)
145
Table 4. 5:Tobit regressions of the relationship between trade credit supply
and competitiveness (dependent variable: percentage of total sales on
delayed payments)
Variables Coefficients
WHSALE 0.271
(5.87)*** W_MILLER 0.261
(6.94)*** RETAIL 0.199
(3.58)*** AGE 0.002
(0.90) LACKCAP -0.052
(-3.69)*** FSIZE 0.037
(3.85)*** DIFLOAN -0.103
(-3.01)*** MARGIN 0.204
(1.49) BUYDP 0.294
(5.09)*** UNPAID 0.046
(5.68)*** CONTACT 0.033
(2.74)*** CUSPOW 0.055
(2.10)** COMP 0.153
(5.64)*** CONSTANT -0.773
(-3.73)*** Log likelihood -154.886 Pseudo R
2 0.438
Observations 522
147
Samenvatting
De meeste studies over het gebruik van handelskrediet maken gebruik van data van
bedrijven in ontwikkelde landen. Deze studie bestudeert het gebruik van
handelskrediet door bedrijven die actief zijn in verschillende onderdelen van de
supply chain in de rijstsector in de Mekong Delta van Vietnam. Het proefschrift
beoogt een bijdrage te leveren aan de inzichten met betrekking tot de vraag
waarom handelskrediet wordt verstrekt en welke rol het verstrekken van
handelskrediet speelt in het creëren van waarde voor het kredietverlenende bedrijf.
Het grootste deel van het onderzoek is gebaseerd op de resultaten van een
uitgebreide survey onder ruim 600 bedrijven. Het onderzoek bestaat uit vier
empirische studies. De eerste studie onderzoekt de relatie tussen het verstrekken
van handelskrediet en de groei van de afzet van het kredietverstrekkende bedrijf,
waarbij gebruik wordt gemaakt van publiek beschikbare financiële informatie van
beursgenoteerde Vietnamese ondernemingen. De studie laat zien dat er een
positieve correlatie bestaat tussen het verlenen van handelskrediet en de groei van
de afzet. De tweede studie maakt gebruik van de primaire data op basis van de
survey en onderzoekt de relatie tussen het verlenen van handelskrediet in de
rijstsector en de mate waarin een bedrijf onderhevig is aan concurrentie. De
resultaten van deze studie suggereren dat bedrijven die in sterkere mate onderhevig
zijn aan concurrentie meer handelskrediet verlenen. Een mogelijke interpretatie
van dit resultaat is dat bedrijven kredietverlening gebruiken om hun positie ten
opzichte van de concurrentie te behouden dan wel te versterken. In de derde studie
wordt dit resultaat verder onderzocht. Aan de hand van informatie op het niveau
148
van handelstransacties wordt aangetoond dat bedrijven geneigd zijn meer krediet te
verlenen aan grotere en bekendere cliënten. Dit kan worden gezien als een verdere
ondersteuning van de hypothese dat handelskrediet gebruikt wordt als een
instrument om de verkoop te stimuleren. Ten slotte wordt onderzocht in hoeverre
specifieke marktomstandigheden (zoals de mate van concurrentie, de
onderhandelingspositie van handelspartners in onderlinge transacties,
klantkenmerken, en de waarde van de verhandelde goederen) verschillend zijn
voor verschillende onderdelen van de supply chain in de rijstsector van Vietnam.
Uit de analyse blijkt dat marktomstandigheden inderdaad sterk verschillen per
onderdeel van de supply chain en dat dit gevolgen heeft voor het gebruik van
handelskrediet. Toekomstige studies naar het belang en het gebruik van
handelskrediet dienen daarom rekening te houden met de specifieke
marktomstandigheden waarin dit plaatsvindt.