The european journal 2018 vol 15 no 2

Vol. 15 No. 2

w w w. j ou r na l.sing idunum.a c .rs

Vol. 15 No. 2 Publisher: Singidunum University E d it o r ia l B o a r d

Professor Milovan Stanišić, Singidunum University mstanisic@singidunum.ac.rs Emeritus Slobodan Unković, Singidunum University unkovic@singidunum.ac.rs Professor Francesco Frangialli, UNWTO frangialli@gmail.com Professor Gunther Friedl, Technische Universität München, München gunther.friedl@wi.tu-muenchen.de Professor Karl Ennsfellner, IMC University of Applied Sciences, Krems karl.ennsfellner@fh-krems.ac.at Professor Gyorgy Komaromi, International Business School, Budapest gyorgy@komaromi.net Professor Vasile Dinu, University of Economic Studies, Bucharest dinu_cbz@yahoo.com Professor Ada Mirela Tomescu, University of Oradea, Oradea ada.mirela.tomescu@gmail.com Professor Radojko Lukić, University of Belgrade rlukic@ekof.bg.ac.rs Professor Alexandar Angelus, Lincoln University angelus@lincolnuca.edu Professor Nemanja Stanišić, Singidunum University nstanisic@singidunum.ac.rs Professor Verka Jovanović, Singidunum University vjovanovic@singidunum.ac.rs Professor Milan Milosavljević, Singidunum University mmilosavljevic@singidunum.ac.rs Professor Olivera Nikolić, Singidunum University onikolic@singidunum.ac.rs Professor Goranka Knežević, Singidunum University gknezevic@singidunum.ac.rs Professor Mladen Veinović, Singidunum University mveinovic@singidunum.ac.rs Professor Jovan Popesku, Singidunum University jpopesku@singidunum.ac.rs Professor Zoran Jeremić, Singidunum University zjeremic@singidunum.ac.rs Professor Vesselin Blagoev, Varna University of Management blagoev@vum.bg Professor Michael Minkov, Varna University of Management minkov@iuc.bg Associate Professor Christine Juen, Austrian Agency for International Mobility and Cooperation in Education, Science and Research, Wien chrisine.juen@oead.at Associate Professor Anders Steene, Södertörn University, Stockholm/Hudinge anders.steene@sh.se Associate Professor Ing. Miriam Jankalová, University of Zilina, Prague miriam.jankalova@fpedas.uniza.sk Associate Professor Bálint Molnár,Corvinus University of Budapest, Budapest molnarba@inf.elte.hu Associate Professor Vesna Spasić, Singidunum University vspasic@singidunum.ac.rs Associate Professor Michael Bukohwo Esiefarienrhe, University of Agriculture, Dept. of Maths/Statistics, Markurdi esiefabukohwo@gmail.com Associate Professor Goh Yen Nee, Graduate School of Business, Universiti Sains Malaysia yngoh@usm.my Research Associate Professor Aleksandar Lebl, Research and Development Institute for Telecommunications and Electronics, Belgrade lebl@iritel.com Roberto Micera, PhD, Researcher, National Research Council (CNR) Italy r.micera@iriss.cnr.it Assistant Professor Patrick Ulrich, University of Bamberg patrick.ulrich@uni-bamberg.de Assistant Professor Jerzy Ładysz, Wrocław University of Economics, Poland jerzy.ladysz@ue.wroc.pl Assistant Professor Konstadinos Kutsikos, University of the Aegean, Chios kutsikos@aegean.gr Assistant Professor Theodoros Stavrinoudis, University of Aegean, Chios tsta@aegean.gr Assistant Professor Marcin Staniewski, University of Finance and Management, Warsaw staniewski@vizja.pl Assistant Professor Gresi Sanje, İstanbul Bilgi Üniversitesi, Istanbul gresi.sanje@bilgi.edu.tr Assistant Professor Michał Biernacki, Wrocław University of Economics, Poland michal.biernacki@ue.wroc.pl Assistant Professor Piotr Luty, Wrocław University of Economics, Poland piotr.luty@ue.wroc.pl Assistant Professor Blazenka Hadrovic Zekic, Faculty of Economics in Osijek, Croatia hadrovic@efos.hr E d it o r ia l O f f ice

Editor in Chief: Managing Editor: Technical Editor: English Language Editor:

Professor Nemanja Stanišić, Singidunum University Associate Professor Gordana Dobrijević, Singidunum University

Aleksandra Stojanović, Singidunum University Marija Popović, Singidunum University

nstanisic@singidunum.ac.rs gdobrijevic@singidunum.ac.rs astojanovic@singidunum.ac.rs mpopovic@singidunum.ac.rs

Prepress: Jelena Petrović Design: Aleksandar Mihajlović ISSN: 2406-2588 The European Journal of Applied Economics is published twice a year.

Contact us: The European Journal of Applied Economics 32 Danijelova Street, 11010 Belgrade, Serbia Phone No. +381 11 3094046, +381 11 3093284 Fax. +381 11 3093294 E-mail: journal@singidunum.ac.rs Web: www.journal.singidunum.ac.rs Printed by: Caligraph, Belgrade

Access to full text articles: Singipedia (www.singipedia.com), SCindeks (www.scindeks.ceon.rs).

Copyright © 2018 Singidunum University, Belgrade All rights reserved.

This is an open access journal which means that all content is freely available without charge to the user or his/her institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles in this journal without asking for prior permission from the publisher or the author. This is pursuant to the BOAI definition of open access.

CONTENTS

1 - 16

17 - 28

The influence of demographic characteristics of consumers on decisions to purchase technical products Adis Puška, Ilija Stojanović, Sead Šadić, Haris Bečić

The relationship between taxes and economic growth: evidence from Serbia and Croatia Branimir Kalaš, Vera Mirović, Nada Milenković

Trade, transportation and environment nexus in Nigeria 29 - 42

43 - 57 58 - 73

Adedayo Emmanuel Longe, Kayode Daniel Ajulo, Olawunmi Omitogun, Emmanuel Olajide Adebayo

Inflation rate impact on the share returns of real sector companies in AP Vojvodina Goran Anđelić, Nenad Penezić, Vilmoš Tot, Marko Milošević

Constraints to small and medium-sized enterprises development in Bangladesh: results from a cross-sectional study

Md. Shahidul Islam, Md. Faruk Hossain

Capital structure in emerging markets: evidence from Nigeria

74 - 90

Yinusa Olumuyiwa Ganiyu, Ismail Adelopo, Yulia Rodionova, Olawale Luqman Samuel

III

EJAE 2018, 15(2): 1-16 ISSN 2406-2588 UDK: 659.113.25 314.17:339.13(497.15) 005.311.12:519.237 DOI: 10.5937/EJAE15-16576 Original paper/Originalni naučni rad

THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS Adis Puška1,*, Ilija Stojanović2, Sead Šadić3, Haris Bečić4 Independent researcher, Brcko, Bosnia and Herzegovina 2 European University Brcko District, Brcko, Bosnia and Herzegovina 3 PHI Health Center Brcko, Spokesman, Brcko, Bosnia and Herzegovina, 4 College of eMPIRICA, Brcko, Bosnia and Herzegovina 1

Abstract: The term consumer behavior has become very popular in recent studies. It is characterized by various processes, with consumers purchasing decisions being one of them. This paper explores the way in which demographic characteristics influence purchasing decisions with focus on technical products including household appliances, computers, TV sets and similar technical products that cost more than 400 BAM. The empirical study was conducted in the region of northeast of Bosnia Herzegovina and 192 respondents were included in the study to express their allegations of purchasing decisions. Factor analysis was used to identify broader constructors as a basis for observation of the variables. In total, six variables were identified. Based on the calculation of Cronbach's Alpha indicators, it has been established that there was a low internal affiliation of claims in two variables, upon what they were discarded.

Thus, four variables were used to study purchasing decision-making.

The results of multivariate analysis of variance (MANOVA) showed that gender, income level and employment status of the respondents significantly influence purchasing decisions among consumers. The analysis of variance (ANOVA) has further shown that females Under the age of 25 and whose income is less than 400 BAM with less than 25 years are generally dedicating to purchasing decisions. The empirical findings showed that the most satisfied with their purchasing decisions are male respondents, whose income varies between 400 BAM and 800 BAM and who belong to the category of students under 25 years.

* E-mail: adispuska@yahoo.com

Article info: Received: April 14, 2018 Correction: May 5, 2018 Accepted: May 9, 2018

Keywords: consumer behavior, buying decisions, demographic factors, factor analysis, multivariate analysis of variance, analysis of variance.

1

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

INTRODUCTION

Nowadays more focus is being given to study the importance of consumer behavior in marketing.

The business environment is increasingly changing and it is very difficult for companies to fight for customer and market share (Khaniwale, 2014). Each company must adapt to changes in the market to survive. Companies must be familiar with their customers’ needs and their purchasing decisions.

Consumer behavior is a complex process that involves a variety of activities including: search, selection, purchase, use, evaluation of products and services to meet their needs and desires (Belch & Belch, 2004). Several internal and external factors, which can range from short-term to long-term emotional feelings, are affecting customer behavior (Hirschman, 1985). Marketing experts need to understand the dynamics of internal factors that are influencing consumers’ decisions. These factors vary from person to person, from situation to situation; therefore, it is important to draw some generalizations of consumer behavior (Komal Prasad & Jha, 2014).

Main decisions for the consumer are: what he buys (which products and services), how much (amount), where (place of purchase), the time spent shopping and payment methods (Khaniwale, 2014). This study is focused on purchasing decision with the focus on technical goods. The aim of this research is to find out how consumers behave and how they are deciding on buying technical equipment.

This research is focused on studying consumer purchasing decisions on the example of technical goods. This research has been conducted in order to find out which demographic factors have the greatest impact on consumer buying behavior and how they influence purchasing decision-making. Based on the results, understanding of customer behavior in purchasing decision-making will be improved which will assist the sellers of the technical devices to promote their products in order to make more impact on purchasing decisions as a most important segment in buying process. In this paper, beside the introduction section, the theoretical framework will be provided and hypotheses of research will be set up in the second part of the paper, research methodology will be given in the third part, the results will be processed and the discussions will be conducted in the fourth part, and finally, the most important conclusions from the research will be given in the fifth part.

LITERATURE REVIEW Consumer behavior includes an analysis of individuals by understanding which method they use to select products, and how they use products and services to fulfill their desires. Consumer behavior refers to all thoughts, feelings and actions that the individual has or had before or during purchasing of the product, service or idea. Main activity in studies of consumer behavior is to understand the process of purchasing decision. The whole process of purchase decision making involves a consideration what to buy, what brand is good or appropriate, where to shop from and when, how much time to spend and at what intervals. Therefore, the end result of customer behavior is making a final decision on the product choice, brand choice, choice of retailers, purchase time, purchase amount and frequency of purchase (Khaniwale, 2014).

Demographic, behavioral and psychographic factors help to understand consumers and their needs (Kotler & Armstrong, 2007). In marketing surveys demographic factors such as age, number of household members, sex, income level and social class are used extensively, and are considered as 2

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

good indicators for the study of consumer behavior (Iqbal, Ghafoor & Shahbaz, 2013). Behavioral factors refer as a way how consumer behaves, how they accept certain products, why they accept it, etc.

Psychographic factors are used for determining and evaluating the lifestyle of consumers, the ways they use their activities, interests and opinions (Tam & Tai, 1998). The study is particularly focused on demographic characteristics of respondents.

When deciding on purchase of various products the consumers make various efforts. The least effort is invested in purchase of food products, because decision on purchasing these products is done automatically and without much thinking. When buying certain products like cars and some technical equipment, the purchase is preceded by a long deciding and consideration of various options to make the final purchasing decision (Markovina, Kovačić & Radman, 2004). Decisions that require a strategic approach are very specific and they are characterized by: high involvement in decision-making, longterm resources acquiring budget available for the purchase of other goods and services (Kos Koklič & Vida, 2009). Sometimes the purchase of technical equipment requires major efforts in the decisionmaking process.

The level of decision making, which comes before the purchase, is a very important factor in marketing research. Based on the obtained data about the level and complexity of decision making for individual product, it is possible to conduct market segmentation and determination of the targeted market where each company will operate. During market segmentation consumers can be divided according to the degree of involvement in the process into the consumers of high, medium and low involvement. Using that approach, it is possible to create marketing messages to targeted consumers that will have the greatest impact on purchasing decision. Also, the data on the involvement of consumers may be used to customize products. Previous studies have shown different behaviors of consumers when purchasing depending on their involvement (Beatty & Smith, 1987).

The purchase decision-making process that a consumer goes through includes the following phases (Engel & Blackwell, 1978): ◆◆ recognition of the problem, ◆◆ search for information, ◆◆ estimating alternatives, ◆◆ purchasing decision and ◆◆ behavior after the purchase.

The first phase begins with the need or the recognition of the problem. This is followed by a search for alternatives that include seeking information from various sources, as well as internal and external environment, such as experience. The third phase involves consumer criteria in calculating benefits subjected to evaluating alternatives. When a decision is made, the consumer enters the fourth phase where the purchase of selected alternatives takes place. The final step involves post-purchase evaluation and consumer behavior after the purchase (SueLin, 2010).The assessment of experience may be influenced by time-dependent parameters that do not have to be directly related to the service, but will lead to an accumulation of experience using this product or service (Dulleck & Kerschbamer, 2006). In recent years, for many reasons, the value of user experience appears to be an important issue in marketing research. Creating experiential value is crucial for customer satisfaction and loyalty (Echchakoui, 2016).

In the service, sector customer experience helps retail environments to create a sustainable competitive position (Srivastava & Kaul, 2014).

3

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Many factors influence consumer’s purchasing decision making. Demographic factors play an important role in purchasing process. Income, age, occupation, and other demographic factors may influence decision-making (Anderson & Gaile-Sarkane, 2008). Iqbal Ghafoor & Shahbaz (2013) showed in their study that the following demographic factors influence the selection of shops to purchase goods: the level of education, occupation, income level, number of household members. Sharma & Kaur (2015) proved in their study that sex and marital status significantly affect the way of purchasing. Alooma & Lawan (2013) showed that demographic factors, such as age, sex, marital status, occupation, education and income, are key variables that influence consumer behavior. Mazloumi et al. (2013) demonstrated that gender, education, marital status, activity and age play an important role in the buying behavior of consumers. A large number of studies have examined individual values, personal attitudes, ethnicity, normative pressure and cognitive bases, including functional background and educational qualifications, as well as their impact on purchase (Mansi & Pandey, 2016). Based on this, demographic characteristics, social and individual values and personal attitudes affect individuals in their purchasing decisions.

Multidisciplinary research worksconcluded that there are gender differences in purchasing behavior and socially responsible behavior (Homburg & Giering, 2001). Studies have shown that women are prone to impulsive shopping and are more loyal to brands (Tifferet & Herstein, 2012) and therefore more and more retailers turn to women as a target group. In some studies, the female sample is more likely to adopt environmentally-friendly purchasing practices (Mainieri et al. 1997; Liu et al. 2012).

The following hypothesis is based on previous conclusions: ◆◆ H1 - There is a significant difference in the purchasing decision-making with regard to gender As the income level determines consumers’ purchasing power, consumers with high-income can afford to buy real estate, life insurance policies, expensive cars, travels, etc. In contrast, low-income consumers will be satisfied with basic living needs and choose affordable products (Štulec et al. 2017).

Professionals influence socially responsible purchasing, and in particular the qualifications and competences of procurement experts probably have a strong impact on environmental, safety and philanthropic activities (Kacprzak & Pawłowska, 2017). Based on this, the following hypothesis was set: ◆◆ H2 - There is a significant statistical difference in the purchasing decision making with regard to the amount of household income Research has shown that there is a difference in customer behavior depending on their qualifications (Chan, 1996). Hambrick and Mason (1984) claim that the educational background is a useful indicator of knowledge and skills. The meta-analysis carried out by Hines, et al. (1987) state that there is a difference in behavior between highly educated and less educated individuals. In contrast, Olli, et al.

(2001), does not find such significant differences between highly educated and less educated individuals. The conclusion is that the level of education is positively linked to the acceptability of innovation (Kimberly & Evanisko, 1981) which tells us about customer acceptability of some new technologies.

Based on this, the following hypothesis was set: ◆◆ H3 - There is a significant statistical difference in the purchasing decision-making with regard to consumer’s level of education The current status of respondents plays a major role on their income. Unemployed respondents have lower income levels and they will find purchasing decision making more important and select products that have lower prices (Štulec et al. 2017). It is, therefore, important to recognize how the status of respondents affects purchasing decisions, whether employees pay less attention to buying decisions from unemployed people and how they decide. Based on this, the following hypothesis was set: 4

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

◆◆ H4 - There is a significant statistical difference in the purchasing decision-making with regard to consumer’s employment status

Studies have shown that age has impact in relation to sustainable purchasing and environmental intentions (Anderson & Cunningham, 1972; Samdahl & Robertson, 1989). According to Carlsson and Karlsson (1970), people with different age enable different responsiveness to the force called “stimulatory pressure” having in mind that younger people are changing faster than older people. Previous studies (Parment, 2013; Kacprzak & Pawłowska, 2017) have shown that younger generations can handle a large amount of information easier. Based on this, the following hypothesis was set: ◆◆ H5 - There is a significant statistical difference in the purchasing decision-making with regard to consumer age Family as a spending unit is often the most interesting to marketing experts due to the greatest spending and the role of a woman in the family who often makes the purchasing decision. Nordstroem and Ridderstrale (2002) say that things have become very personal and that freedom of choice has become a key element in the present time. Children, both young and teenagers, can have a significant impact on the budget allocation and purchasing decisions. Childbirth is also a major event that creates the need for a new and wide range of products that future parents were not buying before (Peter, 2005). In accordance with the abovementioned information, the following hypothesis on household members is formed: ◆◆ H6 - There is a significant statistical difference in the purchasing decision-making with regard to the number of household members among consumers

METHODOLOGY The empirical study was conducted in the region of northeast of Bosnia Herzegovina during 2016 in the period from January to May. The aim of this study was to examine how consumers make their purchasing decisions when buying technical products. Convenience sample was used for data collection.

An online questionnaire placed on the scientific portal 1ka.si, that is promoted trough Facebook pages of the largest portals in the region, has been used. In total, 2,084 respondents accessed the online questionnaire, and 192 respondents completed the questionnaire, which represents a response rate of 9.21%.

The questionnaire consisted of two parts. The first part contained general questions about demographic characteristics of respondents: sex, household income, education, employment status, age and number of household members. The second part of the questionnaire contained 21 statements which used the Likert scale of 5 levels interval from “strongly disagree” to “strongly agree”. The claims used in this group of questions are adapted from the following studies: Beatty & Smith (1987), Jeyakumar & Paul Robert (2010), Bui, Krishen and Bates (2011) and Waheed, Mahasan & Sandhu (2014). Based on these studies the claims for consumer behavior during purchase of technical equipment was adjusted.

Statistical analysis of the data obtained in this study was performed using the SPSS 20 software tool.

In this the work following steps have been used: 1. Presentation of the demographic characteristics of the respondents, 2. grouping claims using factor analysis, and testing reliability of the measurement scale of collected the data, 5

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

3. testing hypotheses using multivariate analysis of variance(MANOVA), and

4. analysis of the influence of certain demographic characteristics of the respondents in purchasing decisions via analysis of variance (ANOVA).

In order to examine the results, the factor analysis indicators of Kaiser-Meyer-Olkin’s (KMO) adequacy of the sample and Bartlett’s test of sphericity were used. The KMO value ranges in closed interval from zero to one. If its value is less than 0.6, than the correlation matrix is not acceptable for factor analysis. In Bartlett’s test, it is preferred that the level of significance is less than p <0.05 (Puška, Maksimović & Fazlić, 2015). During the implementation of factor analysis varimax rotation of factors and Kaiser Normalization was used. Reliability of the scale of collected data was tested using Cronbach’s alpha coefficient whose results range from zero to one. If the value of this indicator is close to zero, then these data are unreliable, and if they are close to one then they are very reliable (Kozarević & Puška, 2015). In order to accept some factors, it is necessary that the value of Cronbach’s alpha is higher than 0.7 (Tavakol & Dennick, 2011).

MANOVA and ANOVA were used. These analyzes are used to answer the question: Do the changes in the independent variables have significant effects on change in the dependent variables (Grbić & Puška 2015)? While conducting these analyzes it was examined how the demographic characteristics of respondents influence purchasing decisions when buying technical products.

The research hypotheses testing was conducted at the level of inferential statistics of 0,05 (p < 0,05), which means that if the significance level is lower than the set level the null hypothesis is accepted, otherwise the alternative hypothesis is accepted and the null hypothesis is rejected.

RESULTS AND DISCUSSION Our first step is to analyze the basic characteristics of the respondents included in the study that is showed in Table 1.

Demographic variables Sex

Household income in BAM

Level of education:

Employment status:

6

Frequency

Percentage

1. Male

118

61.5%

2. Female

74

38.5%

1. Less then and equals to 400

25

13,2%

2. 401-800

57

30.0%

3. 801-1200

44

23,2%

4. More than 1201

64

33.7%

1. Primary education

6

3.1%

2. Secondary Education

86

44.8%

3. Higher education

32

16.7%

4. University degree

68

35.4%

1. Student

53

28.0%

2. Employed

99

52.4%

3. Unemployed

37

19.6%

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Age:

Number of household members:

1. 15-24

64

33.3%

2. 25-40

103

53.6%

3. 41-54

22

11.5%

4. 55 and more

3

1.6%

1. 1-2

40

21.5%

2. 3-4

106

57.0%

3. 5-6

43

18.3%

4. 7 and more

6

3.2%

Table 1. The results of demographic characteristics of the respondents

Out of a total number of 192 respondents, 61.5% of respondents are males, while 38.5% are females.

Observing the household income of respondents, most of them have a monthly household income of more than 1200 BAM which is 33.7% of the respondents, followed by the respondents with a monthly household income between 401 and 800 BAM which is 30.0%, followed by respondents having a monthly household income between 801 and 1200 BAM 23.3%. The last group of respondents have a monthly household income of less than 400BAM 13.2%.

In relation to the educational structure, most of the respondents possess a high school education with 44.8%, followed by the respondents who have a university degree with 35.4%, then the respondents with higher education with 16.7%, while the least respondents possesses a lower level of education who are represented with 3.1%. Observing the employment status of respondents, the most of respondents are employed with 52.1% followed by students with 27.9%, and unemployed persons with 19.5%.

In our sample, most of the respondents are aged between 25 and 40 years, with 53.6%, then the respondents aged between 15 and 24 years with 33.3%, followed by respondents between 41 and 54 years of age with 11.5%, while the least respondents have more than 55 years with 1.6%. Most respondents belong to a household with 3 or 4 members representing 57.0% of the sample, followed by those who have one or two members in the household with 21.5%, and then followed by respondents who have 5 or 6 members in a household with 18.3%, while the least respondents have 7 or more members of the household with 3.2% in the sample.

After presenting the basic demographic characteristics of the respondents, the next analysis indicates whether the collected data are acceptable for further analysis. By using factor analysis, whose results are presented in Table 2, specific grouping statements were analyzed. Apart from this analysis, the table contains calculated values of Cronbach’s Alpha indicators that will be used to determine the reliability of measurement scales of the collected data.

The results were obtained by performing factor analysis and are shown in Figure 1. As shown from Figure 1, six factors have an eigenvalue greater than one. Thus, based on the Kajzer criterion, the factor analysis has grouped used claims into six factors. Table 2 shows values of the factor coefficients for each factor and the amount of individual variance factors used for explanation.

7

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Figure 1. Scree plot Factor 1 I accept tips given to me by salesperson

2

3

4

5

.74

.74 I ask salesperson for additional information .72 Salesperson plays a significant role in purchase of technical products .68 I am looking for alternatives by visiting various shops .56

When I cannot decide what to buy I ask salesperson

Factor 1. Role of salesperson in purchasing process, explained variance = 27.46, Cronbach’s alpha = .76 All hard work that I put in process of deciding on purchase of technical products has paid off

.78

When I buy a device, I feel very satisfied

.78

After purchasing I do not feel guilty conscience that I could buy another product

.69

I’m confident in my decision and I am not changing it

.60

Factor 2. Satisfaction with purchasing decision, explained variance = 11.88, Cronbach’s alpha = .75 I ask my friends and acquaintances for help when buying technical products

.66

I am looking for alternatives on the Internet

.62

It takes me days to decide on purchase of technical products

.56

When I collect all possible alternatives I make decision at home

.52

Factor 3. Shopping information, explained variance = 7.19, Cronbach’s alpha = .70

8

I make decisions on purchase of technical products at home

.77

Usually I am not making decisions to purchase technical products on my own

.74

6

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

I am buying a new technical product only when old one is broken or is outdated

.60

Factor 4. Decision to purchase, explained variance = 5.83, Cronbach’s alpha = .73 I have shopping only at certain stores where I have had positive experiences from past purchases

.76

When buying technical products I buy trusted brands

.55

When shopping I know exactly which product I will buy

.51

Factor 5. Previous experience, explained variance = 5.26, Cronbach’s alpha = .65 Price plays a key role in purchase of technical products

.78

Warranty length has a very important role in purchasing of technical products

.62

Factor 6. Price and warranty, explained variance = 4.79, Cronbach’s alpha = .49 KMO = .800, χ2 = 1077.79, Bartlett’s Test of Sphericity = .00, explained variance = 62.40 Table 2. Factor analysis of data acceptability

The results of the factor analysis (Table 2) grouped the claims into six factors. These factors explained 62.40% of the variance of the basic set which is a customary percentage that is present in social research (Kurnoga Živadinović, 2004). The value of the KMO indicator is higher than the required 0.6 (KMO = .80), while the value of Bartlett’s test is lower than the set level of significance (p = .00), which confirms the results of the factor analysis.

The first factor,indicated as “Role of salesperson in purchase process”, explains 27.46% of the variance. This factor grouped the five statements that are mostly related to the role of salesperson in making purchasing decisions. The value of Cronbach alpha indicator is greater than 0.7 (.76), thus confirming the reliability of the scale of collected data for this factor. The second factor, indicated as “Satisfaction with purchasing decision”, included the four statements related to satisfaction of the respondents with their decision with purchase of technical equipment. This factor explains 11.88% of the variance of the basic set. The results of Cronbach’s alpha for this factor are greater than 0.7 (.74) which confirms that the collected data are reliable for further analysis. The third factor, indicated as “Shopping information”, included the four statements that are related to the way of looking for alternatives by the respondents.

This factor explains 7.19% of the variance of the basic set. The value of Cronbach’s alpha is very close to the set level of confidence (.70). Thus, the data that are grouped by this factor are reliable and acceptable for further analysis. The fourth factor, indicated as “Purchase decision”, included the three claims that are related to the way in which the respondents decide about the purchase of technical equipment.

This factor explains 5.83% of the variance of the basic set and the resulting value of Cronbach’s alpha indicators is higher than the set level of reliability (.73) which proves that the data grouped by this factor are reliable for further analysis.

The fifth factor, indicated as “Previous experience”, grouped the three claims related to previous experience with the shop or device that has a positive impact on the purchasing decision. This factor explains 5.26% of the variance of the basic set, and the value of Cronbach’s alpha for this factor is low (.65). Having in mind this result, the conclusion is that the data grouped by this factor are unreliable for further 9

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

analysis. This value of Cronbach’s alfa could be caused by a small number of questions, weak correlation between the questions or heterogenic questions. The sixth factor, indicated as “Price and warranty”, included the two items related to the impact of price and product warranty on the purchasing decision.

This factor explains 4.79% of the variance of the basic set and the value of Cronbach’s alpha is very low for this factor (.49). It is concluded that the data are unreliable and will not be used for further analysis.

The results of the factor analysis have reduced the total number of claims from 21 to 16 that are grouped into four factor groups. In the further analysis only the first four factors will be used because they have a reliable measurement scale, while the last two factors are discarded from further analysis.

Furthermore, MANOVA analysis was used to examine the hypotheses identified in this study.

Respondents’ characteristics

F-test

Sig.

Status of hypothesis

Sex

5.17

.001

Accepted

Household income

1.97

.025

Accepted

Level of education

.99

.462

Discarded

Employment status

2.34

.018

Accepted

Age

1.05

.403

Discarded

Number of household members

1.13

.336

Discarded

Table 3. Testing the hypothesis using MANOVA analysis

The results of MANOVA analysis are shown in Table 3. According to the empirical findings, three hypotheses are accepted, and three hypotheses are not accepted. This analysis showed that gender (F = 5.17, p = .001), household income (F = 1.97, p = .025) and employment status (F = 2.340, p = .018) play an important role in purchasing decisions of the respondents. However, this analysis showed that the level of education (F = .99, p = .462), age (F = 1.05, p = .403) and the number of household members (F = 1.13, p = .336) do not play a significant role in purchasing decisions among the respondents included in this study, since there are no statistically significant indicators to claim the opposite.

After testing the hypothesis with MANOVA, an ANOVA analysis was conducted to examine the impact of certain factors within the individual characteristics of the respondents to determine what factors our respondents use when deciding about the purchase of technical equipment.

Respondents’ characteristics

Sex

Household income

10

Factor

Variance

F-test

Sig.

Ratio

Role of the salesperson in purchasing process

4.55

7.47

.007

1<2

Satisfaction with purchasing decision

.95

1.50

.223

2<1

Shopping information

.35

.53

.467

1<2

Decision to purchase

7.50

8.70

.004

1<2

Role of salesperson in purchasing process

2.28

3.78

.012

3<4<2<1

Satisfaction with purchasing decision

.18

.28

.838

3<1<4<2

Shopping information

1.67

2.60

.054

2<3<4<1

Decision to purchase

2.39

2.72

.046

2<3<4<1

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Level of education

Employment status

Age

Number of household members

Role of salesperson in purchasing process

.85

1.36

.257

4<2<3<1

Satisfaction with purchasing decision

.20

.32

.811

2<4<3<1

Shopping information

1.02

1.58

.196

3<2<4<1

Decision to purchase

1.46

1.64

.181

4<3<2<1

Role of salesperson in purchasing process

.27

.43

.651

3<1<2

Satisfaction with purchasing decision

2.28

3.78

.025

3<2<1

Shopping information

.83

1.32

.271

2<3<1

Decision to purchase

3.45

4.04

.019

2<3<1

Role of salesperson in purchasing process

1.05

1.69

.171

4<2<3<1

Satisfaction with purchasing decision

.64

1.00

.392

3<2<1<4

Shopping information

.27

.40

.750

3<2<4<1

Decision to purchase

2.20

2.51

.060

2<3<1<4

Role of salesperson in purchasing process

.42

.68

.564

3<4<2<1

Satisfaction with purchasing decision

.79

1.27

.285

4<1<2<3

Shopping information

1.25

2.01

.115

4<2<3<1

Decision to purchase

.78

.89

.447

4<2<1<3

Table 4. Factor dependency on basic characteristics of the respondents

The results of the ANOVA analysis (Table 4) for „gender” as one of the basic characteristic of the respondents has shown that there is a significant statistical difference between the two factors: “Role of salesperson in purchasing process” (F = 7.47, p = .007) and “Decision of purchase” (F = 8.70, p = .004), while in the other two factors there is no significant difference.

The ratio represents the arithmetic mean of the offered answers for certain dimensions for the basic characteristics, which is in this case “gender”. These relationships between factors showed that female respondents have consulted more salespersons, have looked for more alternatives, and got along with more statements when making decisions on purchase of technical goods, while male respondents were more satisfied with their purchase decision.

For the basic characteristic „household income”, the results have shown that there is a significant statistical difference for the factors: “Role of salesperson in purchasing process” (F = 3.77, p = .012) and the “Decision to purchase”(F = 2.72, p = .046). For other factors, there is no significant difference in the answers. The ratio indicated that the statements related to buying decision are mostly used by the respondents with a monthly household income below 400 BAM, while the second place was taken by the respondents with income rates higher than 1200 BAM. These findings suggest that the household income does not play a decisive role in the purchasing decision because the respondents who have income levels between 400 and 1200 BAM are the least satisfied with their purchasing decisions since they also use purchasing decisions the least.

The results related to the level of education of the respondents indicated that there is no significant statistical difference in the answers for any of the factors. What is the characteristic about these findings is that purchasing decision is most used by those respondents who have lower levels of education, while the order in other dimensions ranges differently in comparison to other factors. It has been found that those respondents with a university degree are least satisfied with their purchasing decision, the 11

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

respondents with higher education contact salespersons least and do not make decisions to purchase at home, while respondents with higher education look for alternatives in the form of technical goods the least in relation to the other.

Using the basic characteristic „Employment status of the respondents“, it has been found that there is a statistically significant difference for the following factors: “Satisfaction with purchasing decision” (F = 3.78, p = .025) and “Decision to purchase” (F = 4.04, p = .019), while the other factors do not have significant differences in the answers. The results of the relationships indicated that students are most satisfied with purchasing decisions, they are looking for alternatives the most and make decisions at home, while employed people contact salespersons the most. Furthermore, unemployed persons contact salespersons the least. They are least satisfied with their buying decision, while employed respondents look for alternatives the least and do not make decisions to purchase at home.

The results for the primary characteristic „Respondents age” have shown that there is no significant difference in any of the factors. The respondents who have less than 25 years contact salespersons the most and look for alternatives, while the most satisfied with the decision are the respondents who have more than 55 years. The findings should be taken with some caution because there were only three respondents with more than 55 years. The respondents aged between 41 and 55 years are least satisfied with purchasing decision, and look for alternatives the least, while respondents aged between 25 and 40 years make decisions to purchase at home the least.

After a closer look at the results for the basic characteristic „Number of household members” ,it has been found that there is no significant difference for any of the factors. Thus, the number of household members does not particularly effect the purchasing decision. Moreover, it has been found that the respondents who have one or two household members use salesperson opinion most, while the respondents who have five or six household members are the most satisfied with their purchasing decision. The respondents with seven or more household members use purchasing decisions the least.

CONCLUSION The results of MANOVA analysis have shown that there is a significant statistical difference for the following basic demographic characteristics including gender, income level and status of the respondents thus playing a significant role in purchasing decisions. Three hypotheses are proven in this study indicating that the following basic characteristics of the respondents are significant: sex, household income and employment status. The other three hypotheses are discarded because it is proven that there is no significant statistical difference in the following basic characteristics of the respondents: level of education, age and number of household members.

The importance of this research was to investigate how demographic characteristics of respondents in B&H influence purchasing decision making. The importance of research is reflected in the fact that B&H residents are at the bottom of the European scale when it comes to per capita income. Therefore, the obtained results are particularly important because they are about buying technical equipment that require significant amount of money. The obtained results will help manufacturers and retailers of technical goods understand the behavior of consumer in countries with small per capita income levels.

This will allow them to adapt to any demand, regardless of the country they are located in. Therefore, it was important to conduct this research in B&H.

12

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

ANOVA proved that female respondents consult more salesperson, look for more alternatives, and decisions to purchase make at home. Male respondents are more satisfied with their own purchase decision. It has been proved that people with income less than 400 BAM and people with low level of education spend most of their time on purchasing decision. The empirical findings have shown that students are the most satisfied with purchasing decisions. They look for most alternatives and make decisions to purchase at home, while employees contact salespersons the most. It has been noticed that respondents who are younger than 25 years contact more salesperson, and look for more alternatives.

Number of household members has no effect on purchasing decisions.

Based on these findings it can be concluded that females, who have income levels less than 400 BAM with lower level of education and less than 25 years, spend most of the time in searching for alternatives.

People, who are male and have income levels higher than 800BAM, older than 55 years use purchasing decisions the least. The most satisfied with their purchasing decisions are male respondents, who have income levels between 400BAM and 800BAM and have a lower level of education.

This study has its limitations which can be resolved in future studies. The biggest drawback was the questionnaire since the claims about purchase decision-making were too similar. Thus, it was not possible to group the reliable data as shown in the results of Cronbach’s Alpha. In future studies, it is necessary to include more subjects in research and to cover larger territories. Despite all these shortcomings, this paper represents one of the first papers that explore purchasing decisions in Bosnia and Herzegovina giving significant guidelines for future research in this area.

REFERENCES Alooma, A.G., & Lawan, L.A. (2013). Effects of Consumer Demographic Variables on Clothes Buying Behaviour in Borno State, Nigeria. International Journal of Basic and Applied Science, 1(4), 791-799.

Anderson, W.T., & Cunningham, W.H. (1972). Gauging foreign product promotion. Journal of Advertising Research, 12(1), 29‐34.

Andersone, I., & Gaile-Sarkane, E. (2008). Influence of factors on consumer behavior. 5th International Scientific Conference (pp. 331-337). Vilnius: Technika.

Beatty, S.M., & Smith, S.M. (1987). External Search Effort: An Investigation Across Several Product Categories.

Journal of Consumer Research, 14(1), 83-95. DOI:10.1086/209095 Belch, G.E., & Belch, M.A. (2004). Advertising and Promotion: An Integrated Marketing Communications Perspective. New York: McGraw Hill.

Bui, M., Krishen, A.S., & Bates, K. (2011). Modeling regret effects on consumer post-purchase decisions. European Journal of Marketing, 45(7/8), 1068-1090. DOI:10.1108/03090561111137615 Carlsson, G., & Karlsson, K. (1970). Age, cohorts and the generation of generations. American Sociological Review, 35(4), 710-718. DOI:10.2307/2093946 Chan, T. (1996). Concerns for environmental issues and consumer purchase preferences: a two-country study.

Journal of International Consumer Marketing, 9(1), 43-55. DOI:10.1300/J046v09n01_04 Dulleck, U., & Kerschbamer, R. (2006). On Doctors, Mechanics, and Computer Specialists: The Economics of Credence Goods. Journal of Economic Literature, 44(1), 5-42. DOI:10.1257/002205106776162717 Echchakoui, S. (2016). Addressing Differences Between Inbound and Outbound Agents for Effective Call Center Management. Global Business and Organizational Excellence, 36(1), 70-86. DOI:10.1002/joe.21757 Engel, J.F., Blackwell, R.D., & Kollat, D.T. (1978). Consumer Behavior. Hinsdale, IL: Dryden Press.

Grbić, N., & Puška, A. (2015). Utjecaj etnocentrizma na kupovno ponašanje potrošača na području Brčko distrikta BiH. Zbornik Ekonomskog fakulteta u Zagrebu, 13(2), 103-120.

13

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Hambrick, D.C., & Mason, P.A. (1984). Upper Echelons: The Organization as a Reflection of Its Top Managers.

The Academy of Management Review, 9(2), 193-206. DOI:10.2307/258434 Hines, J.M., Hungerford, H.R., & Tomera, A.N. (1987). Analysis and synthesis of research on responsible environmental behaviour: A meta-analysis. Journal of Environmental Education 18(2), 1-8. DOI:10.1080/009589 64.1987.9943482 Hirschman, E.C. (1985). Cognitive processes in experimental consumer behavior. Research on Consumer Behavior, 1, 67-102.

Homburg, C., & Giering, A. (2001). Personal characteristics as moderators of the relationship between customer satisfaction and loyalty-an empirical analysis. Psychology & Marketing, 18(1), 43-66. DOI:10.1002/1520-6793 Iqbal, H.K., Ghafoor M.M., & Shahbaz, S. (2013). Impact of Demographic Factors on Store Selection: An Insight in Pakistani Society. Journal of Marketing Management, 1(1), 34-45.

Jeyakumar, K., & Paul Robert, T. (2010). Joint Determination of Price, Warranty Length and Production Quantity for New Products under Free Renewal Warranty Policy. International Journal for Quality Research, 4(1), 51-58.

Kacprzak, A., & Pawłowska, A. (2017). Work and shopping overflow: Consequences and differentiation among selected psychological and demographic characteristics. European Management Journal, 35(6), 755-765.

DOI:10.1016/j.emj.2017.06.003 Khaniwale, M. (2014). Consumer Buying Behavior. International Journal of Innovation and Scientific Research, 14(2), 278-286.

Kimberly, J.R., & Evanisko, M.J. (1981). Organizational Innovation: The Influence of Individual, Organizational, and Contextual Factors on Hospital Adoption of Technological and Administrative Innovations. The Academy of Management Journal, 24(4), 689-713. DOI:10.5465/256170 Komal Prasad, R., & Jha, M.K. (2014). Consumer buying decisions models: A descriptive study. International Journal of Innovation and Applied Studies, 6(3), 335-351.

Kos Koklič, M., & Vida, I. (2009). A Strategic Household Purchase: Consumer House Buying Behavior. Managing Global Transitions, 7(1), 75-96.

Kotler, P., & Armstrong, G. (2007). Principles of Marketing. Upper Saddle River, NJ: Prentice Hall.

Kozarević, S., & Puška, A. (2015). Povezanost primjene lanca snabdijevanja, partnerskih odnosa i konkurentnosti malih i srednjih kompanija. Ekonomska misao i praksa, 10(2), 579-596.

Kurnoga Živadinović, N. (2004). Utvrđivanje osnovnih karakteristika proizvoda primjenom faktorske analize.

Ekonomski pregled, 55(11-12), 952-966.

Liu, X., Wang, C., Shishime, T., & Fujitsuka, T. (2012). Sustainable consumption: green purchasing behaviours of urban residents in China. Sustainable Development, 20(4), 293-308. DOI:10.1002/sd.484 Mainieri, T., Barnett, E.G., Valdero, T., Unipan, J., & Oskamp, S. (1997). Green buying: the influence of environmental concern on consumer behavior. The Journal of Social Psychology, 137(2), 189-204. DOI:10.

1080/00224549709595430 Mansi, M., & Pandey, R. (2016). Impact of demographic characteristics of procurement professionals on sustainable procurement practices: Evidence from Australia. Journal of Purchasing and Supply Management, 22(1), 31-40. DOI:10.1016/j.pursup.2015.06.001 Markovina, J., Kovačić, D., & Radman, M. (2004). Uključenost pri donošenju kupovnih odluka - primjer tri prehrambena proizvoda. Journal of Central European Agriculture, 5(3), 151-159.

Mazloumi, S.S.S., Efteghar, A., Ghalandari, A., Saifi, B., & Aghandeh, I. (2013). Evaluating the effect of demographic differences on consumers’ purchasing behavior (Case Study: Tetra Pak Consumers). International Research Journal of Applied and Basic Sciences, 4(7), 1866-1867.

Nordstroem, K., & Ridderstrale, J. (2002). Funky business. Zagreb: Differo.

Olli, E., Grendstad, G., & Wollebaek, D. (2001). Correlates of environmental behaviors: bringing back social context. Environment and Behavior, 33(2), 181-208. DOI:10.1177/0013916501332002 14

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

Parment, A. (2013). Generation Y vs. Baby Boomers: Shopping behavior, buyer involvement and implications for retailing. Journal of Retailing and Consumer Services, 20(2), 189-199. DOI:10.1016/j.jretconser.2012.12.001 Peter, J.P. (2005). Consumer behavior and marketing strategy. Boston, MA: McGraw Hill.

Puška, A., Maksimović, A. & Fazlić, S. (2015). Utjecaj kvalitetenazadovoljstvoilojalnoststudenata. Poslovnaizvrsnost. 9(2), 101-119.

Samdahl, D.M., & Robertson, R. (1989). Social determinants of environmental concern: Specification and test of the model. Environment and Behavior, 21(1), 57-81. DOI:10.1177/0013916589211004 Sharma, K.C., & Kaur, S. (2015). The Impact of Demographic Factors on Impulse Buying Behaviour of Online and Offline Consumers (A Case Study of Punjab, Haryana, New Delhi and Chandigarh). International Journal of Engineering Technology, Management and Applied Sciences, 3(9), 63-69.

Srivastava, M., & Kaul, D. (2014). Social interaction, convenience and customer satisfaction: The mediating effect of customer experience. Journal of Retailing and Consumer Services, 26(6), 1028-1037. DOI:10.1016/j. jretconser.2014.04.007 Štulec, I., Petljak, K. & Rakarić, J. (2017). Utjecaj demografskih karakteristika potrošača na proces donošenja odluke o kupovini. Ekonomska misao i praksa, 12(1), 381-404.

SueLin, C. (2010). Understanding Consumer Purchase Behavior in the Japanese Personal Grooming Sector.

Journal of Yasar University, 17(5), 2821-2831.

Tam, J., & Tai, S. (1998). The Psychographic Segmentation of the Female Market in Greater China. International Marketing Review, 15(1), 61-77. DOI:10.1108/02651339810205258 Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53-55. DOI:10.5116/ijme.4dfb.8dfd Tifferet, S., & Herstein, R. (2012). Gender differences in brand commitment, impulse buying, and hedonic consumption. Journal of Product & Brand Management, 21(3), 176-182. DOI:10.1108/10610421211228793 Waheed, A., Mahasan, S.S., & Sandhu, M.A. (2014). Factor That Affects Consumer Buying Behavior: An Analysis of Some Selected Factors. Middle-East Journal of Scientific Research, 19(5), 636-641. DOI:10.5829/idosi. mejsr.2014.19.5.13623

15

EJAE 2018  15 (2)  1-16

PUŠKA, A., STOJANOVIĆ, I., ŠADIĆ, S., BEČIĆ, H.  THE INFLUENCE OF DEMOGRAPHIC CHARACTERISTICS OF CONSUMERS ON DECISIONS TO PURCHASE TECHNICAL PRODUCTS

UTICAJ DEMOGRAFSKIH KARAKTERISTIKA POTROŠAČA NA ODLUKE O KUPOVINI TEHNIČKIH PROIZVODA

Rezime: Izraz ponašanje potrošača postao je veoma popularan u novijim studijama. Odlikuju ga različiti procesi, pri čemu odluke potrošača o kupovini predstavljaju jedan od njih. Ovaj rad istražuje način na koji demografske karakteristike utiču na odluke o kupovini sa fokusom na tehničke proizvode, uključujući kućne aparate, računare, televizore i slične tehničke proizvode koji koštaju više od 400 KM. Empirijska studija sprovedena je u regionu severoistočne Bosne i Hercegovine, a 192 ispitanika uključeno je u studiju kako bi izrazile svoje navode o kupovnim odlukama. Analiza faktora korišćena je za identifikaciju šireg konstruktora kao osnove za posmatranje varijabli. Ukupno je identifikovano šest varijabli. Na osnovu izračunavanja Cronbachovih alfa indikatora, utvrđeno je da postoji niska interna afilijacija izjava u dve varijable, na osnovu čega su iste i odbačene. Stoga su četiri varijable korišćene za proučavanje odluke o kupovini.

Rezultati multivarijantne analize varijanse (MANOVA) pokazali su da pol, nivo prihoda i radni status ispitanika značajno utiču na odluke o kupovini kod potrošača. Analiza varijanse (ANOVA) je dalje pokazala da se žene ispod 25 godina i čiji su prihodi manji od 400 KM, uglavnom baveu odlukama o kupovini, tj. tome posvećuju najviše vremena Empirijski nalazi pokazuju da su najviše zadovoljni odlukama o kupovini muški ispitanici čiji se prihod kreće između 400 i 800 KM i koji pripadaju kategoriji studenata mlađih od 25 godina.

16

Ključne reči: ponašanje potrošača, odluke o kupovini, demografski faktori, analiza faktora, multivarijantna analiza varijanse, analiza varijanse.

EJAE 2018, 15(2): 17-28 ISSN 2406-2588 UDK: 336.228:330.34(497.11+497.5) 330.34 DOI: 10.5937/EJAE15-18056 Original paper/Originalni naučni rad

THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA Branimir Kalaš*, Vera Mirović, Nada Milenković University of Novi Sad, Faculty of Economics, Subotica, Serbia

Abstract: This study presents an empirical analysis of taxes and economic growth in Serbia and Croatia in the period 2007-2016. In order to identify the impact of tax forms on economic growth and their relationship, the authors decided to set up a panel regression where gross domestic product is the dependent variable, while corporate income tax, value added tax, social security contributions and excises are independent variables. The results of random effect model have shown that corporate income tax, value added tax and social security contributions have a positive impact on the gross domestic product, while excises affect the gross domestic product negatively. However, only value added tax has a statistically significant impact on economic growth in these countries, with each increase in revenue from this tax contributing to the growth of gross domestic product in the observed period.

Article info: Received: July 29, 2018 Correction: August 24, 2018 Accepted: August 29, 2018

Keywords: taxes, economic growth, relationship, Serbia, Croatia.

INTRODUCTION – THEORETICAL BACKGROUND Economic growth is the basis of increased prosperity (Myles, 2000). Taxes are important tool for the economy. They represent the crucial component in contemporary business and their relevance is manifested through stability and predictability (Kalaš et al. 2016). There are many definitions of taxes (Anyanwu, 1993; Bhartia, 2009; Appah and Oyandonghan 2011; Angahar and Alfred, 2012; Chigbu et al. 2012; Salami et al. 2015). Appah (2010) defined tax as a compulsory levy imposed on a subject or his/her property by the government to provide social amenities and create conditions for the economic prosperity of the society. Likewise, Chigbu and Njoku (2015) emphasize that taxation is a major source of revenue for every economy and it’s usually an instrument used in reducing the gap between the rich * E-mail: branimir.kalas@ef.uns.ac.rs

17

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

and the poor. Taxes should be “good” for the economy and its development. Mitra and Stern (2003) determined an optimal tax level and structure that could contribute to the more efficient growth of the economy. Likewise, Mankiw et al. (2009) defined optimal taxation theory in terms of the fact that an adequately designed tax system can maximize a social welfare function. In the academic world, there is a voluminous literature on taxes and their growth features, as well as on widely varying methodologies and results (Gale et al. 2015). For example, McBride (2012) shares results of Congressional Research Service, which has found support for the theory that taxes have no effect on economic growth by relying on the U.S. experience since World War II, where they found that a rapid economic growth occurred in the 1950s when the top rate was more than 90%. Table 1 reflects empirical studies which examined the effects of tax forms on economic growth.

18

Year

Time period

Country

Effect

Result

Helms

1985

1965-1979

United States

Negative

Revenue used to fund transfer payments slows growth

Padovano and Galli

2001

1951-1990

23 OECD countries

Negative

Effective marginal income tax rates are negatively correlated with GDP growth

Tomljanovich

2004

1960-1990

United States

Negative

Higher tax rates negatively affect short-run growth, but not long-run growth

Lee and Gordon

2005

1980-1997

70 countries

Negative

Reducing corporate income tax by 1% raises annual growth by 0.1% to 0.2%

Tosun and Abizadeh

2005

1980-1999

OECD countries

positive/ negative

Shares of personal and property taxes have responded positively to economic growth, while the shares of payroll and goods and services taxes have reflected a relative decline

Bania, Gray and Stone

2007

1962-1997

United States

Negative

Taxes directed towards public investments were first added then subtracted from GDP

Reed

2008

1970-1999

United States

Negative

Robust negative effect of state and local taxation

Alesina and Ardagna

2010

1970-2007

OECD countries

Negative

Fiscal incentives based on tax cuts enhance growth more than increased consumption

Gemmell, Kneller and Sanz

2011

1970-2004

17 OECD countries

Negative

Taxes on income and profit are most damaging to economic growth over the long run.

The second in line are consumption taxes

Romer and Romer

2010

1945-2007

United States

Negative

Tax increase of 1% GDP leads to a fall in output of 3% after about two years

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

Arnold, Brys, Heady, Johansson, Schwelnuss and Vartia

2011

1971-2004

21 OECD countries

Negative

Corporate taxes are most harmful, followed by personal income tax, consumption and property tax

Barro and Redlick

2011

1912-2016

United States

Negative

Cut in the average marginal tax rate of 1% raises GDP per capita by around 0.5 in the next year

Ferede and Dahlby

2012

1977-2006

Canada

Negative

Reducing corporate income tax by 1% raises annual growth by 0.1 to 0.2%

Positive

Value added tax and petroleum profit tax have a positive and significant relationship on the gross domestic product

positive/ negative

There is a significant positive relationship between petroleum profit tax, company income tax and economic growth. Insignificant relationship was perceived between customs, excises and economic growth

Ibadin and Oladipupo

Onakoya and Afintinni

2015

2016

1981-2014

1980-2013

Nigeria

Nigeria

Table 1. Empirical studies about the effects of taxes on economic growth Source: Adapted from McBride (2012)

TAX TRENDS IN SERBIA AND CROATIA Tax systems of modern countries differ from each other in terms of constituent elements and the share of tax forms (Aničić et al. 2012). Perre and Hashorva (2011) argued that tax systems in West Balkan countries are similar and founded on three main types of taxes: personal income tax, corporate income tax and value added tax. Šimović et al. (2014) defined Croatian tax system as a hybrid where the elements of income-based and consumption-based taxation concept are present. Right on, Arbutina (2000) cited that value added tax has been applied for a short period of time in Croatia, but at the moment it’s the cornerstone of the state revenue. For Serbia, Ranđelović (2008) determined the significance of direct taxes in transition countries, while personal income tax is perceived to be a fundamental element of modern tax systems in developed countries. On the other hand, Stevanović and Gajić (2013) noticed that tax system in Serbia is not efficient enough and needs to be characterised by simplicity in terms of efficient administration, low costs electronic business and flexibility through the fact that tax legislation should follow the economic justification. Also, the existence of political responsibility and fairness represents an essential principle which provides a clear and equal approach in terms of paying taxes and economic power. Nerre et al. (2014) argued that Serbian tax system has been continuously improving since the start of the transition in 2000, where total revenue expressed as a percentage of GDP has risen from 33% in 2000 to 42% in 2012. Also, they determined social security contributions as the main source of tax revenues.

19

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

Figure 1. GDP growth rate in Serbia and Croatia from 2007 to 2016 Source: Authors based on http://www.imf.org/external/country

Figure 1 manifests the annual rate of GDP growth in selected countries from 2007 to 2016. First, in Serbia, rates were above 5% in 2007 and 2008, while the largest drop of 3.12% was recorded in 2009, as a result of decreasing economic activity in the world and escalating global economic crisis. Then, in the next two years, there is a slight increase of this indicator after which Serbian economy achieved a negative rate of 1.01%. A similar trend was reported in 2014 when the growth rate of gross domestic product amounted 1.83%, but in the last two years, the average growth rate was 1.24% in Serbia. On the other hand, Croatia had a similar trend of gross domestic product, while in 2009 their decrease was higher than in Serbia, and recorded a negative rate of 7.38%. Further on, Croatian economy had negative rates and average decline amounted 2.16% until 2015 when a slight growth of 1.65% was recorded.

Figure 2. Tax revenues in Serbia and Croatia from 2007 to 2016 Source: Authors based on http://www.imf.org/external/country

Looking at the tax revenue’s share of GDP, it is higher in Serbia compared to Croatia. At the beginning, tax revenues constituted 35.8% of gross domestic product in Serbia and 25.7% in Croatia, where the share of tax revenues declined until 2010 and 2011. Also, there was an increase of tax revenue’s 20

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

share of 0.4% in Serbia, while in 2012 it recorded a similar growing trend by 0.9% in Croatia. Comparing the average share of tax revenues, it amounts to 34.61% and 24.49% of the gross domestic product respectively, which indicates that there is a disparity of around 10% in observed variables between the selected countries.

Figure 3. Comparative review of tax forms in Serbia and Croatia Source: Authors based on http://www.imf.org/external/country

Figure 3 represents a comparative review of tax form’s share in the gross domestic product in Serbia and Croatia from 2007 to 2016. First, personal income tax share is around 4% in Serbia and 6% in Croatia, with a noticeable declining trend in both countries. In fact, since the beginning of the period, the share of this tax form is reduced to 1.2% in Serbia and 0.8% in Croatia. On the other hand, corporate income tax share increased by 0.6% in Serbia, while in Croatia it recorded a decline of 0.4%. Ranđelović (2010) emphasizes that average share of the corporate income taxes in GDP of Serbia is 2, 5 to 3 times lower than in the European Union. Value added tax is one of the most abundant tax forms in selected countries and share of this tax exceeds 10% of the gross domestic product. At the end of 2016, the share of value added tax amounted 10.6% in Serbia and 13% in Croatia. Further, social security contributions represent tax form with a share of around 11% and 12%. However, there was a reduction of this tax for 21

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

0.4% in Serbia and 1.6% in Croatia, in the observed period. Finally, excise share is higher in Serbia than Croatia, where it has a completely different trend. Since 2007, the share of this tax form is increased by 1.9% in Serbia, while in Croatia it has recorded a decline of 0.2%. Looking at the observed tax forms, it is evident that Croatia has a greater share of taxes in the gross domestic product, except in case of excise, which is higher in Serbia. On the other hand, an interesting fact is that tax revenues in Serbia make almost 40% of the gross domestic product, while in Croatia it is only 30%. It can be assumed that other tax types in Serbia have a greater share in the gross domestic product.

METHODOLOGY AND DATA The research is focused on tax forms and their relation to the gross domestic product, which is defined as a proxy for economic growth. The analysis is conducted for the period covering 10 years as of 2007 to 2016 in Serbia and Croatia, where the authors used percentage share of GDP from the official data of International Monetary Fund. The paper used a multiple regression by means of which they studied the nature of the presented relationship. Likewise, there is an analysis of variance and multicollinearity between the independent variables through Variance Inflation Factor test (VIF).

Variable

Notation

Calculation

Source

GDP

% annual growth rate

IMF report

Corporate income tax

CIT

% share of GDP

IMF report

Value added tax

VAT

% share of GDP

IMF report

Social security contributions

SSC

% share of GDP

IMF report

Excises

EXC

% share of GDP

IMF report

Dependent variable Gross domestic product Independent variables

Table 2. Variable definition Source: Authors review

The paper represents a regression model, which includes one dependent variable and four independent variables. First, a gross domestic product was used as a proxy for economic growth while tax forms were defined by the next schedule: corporate income tax as CIT, value added tax as VAT, social security contributions as SSC and excises as EXC. In the presented study we have used a random effects model due to its capability to account for the correlation among the residuals. Hausman test confirmed that this type of modeling option is appropriate for the analysis. The random effects model is specified as follows: GDPit = β0 +µi + β1CITit+ β2VATit + β3SSCit + β4EXCit + εit

(1)

where GDPit stands for gross domestic product of country i [Serbia, Croatia] for period t [2007-2016], CITit stands for corporate income tax as % of GDP, VATit for value added tax as % of GDP, SSCit for 22

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

social security contributions as % of GDP, EXCit for excises as % of GDP, β0 for the grand intercept, βi for the coefficients with the independent variables, εit for the error term and µi for the random effect pertaining to the countries.

RESULT RESEARCH Taking into account that research examines tax effect on economic growth in Serbia and Croatia, the authors applied panel regression model with the gross domestic product as the dependent variable and four independent variables such as corporate income tax, value added tax, social security contributions and excises. At first, we presented descriptive statistics of explanatory variables, as well as mean, standard deviation and minimum and maximum level.

Country

GDP

CIT

VAT

SSC

EXC

Serbia Mean

1.34

1.43

10.49

10.99

5.08

Std. Dev.

2.92

0.36

0.38

0.29

0.72

Min.

-3.12

0.7

9.8

10.6

4

Max.

5.89

1.9

11.1

11.5

6.3

Croatia Mean

-0.07

2.49

12.13

11.78

3.76

Std. Dev.

3.47

0.29

0.74

0.21

0.48

Min.

-7.38

2.3

11.1

11.3

3.2

Max.

5.15

3.1

13.2

12

4.6

Total Mean

0.64

1.96

11.31

11.38

4.42

Std. Dev.

3.20

0.63

1.02

0.48

0.9

Min.

-7.38

0.7

9.8

10.6

3.2

Max.

5.89

3.1

13.2

12

6.3

Table 3. Descriptive statistics Source: Authors based on SPSS

Table 3 shows mean, minimum and maximum value of explanatory variables in Serbia and Croatia in the period 2007-2016. The results reflect that Serbia reached mean GDP by 1.34 compared to Croatia which had negative growth rate by 0.07. Further on, mean share of tax forms are higher in Croatia in relation to Serbia, except excises. Based on the obtained results, the highest standard deviation of gross domestic product and value added tax was in Croatia, while corporate income tax and social security contributions have the lowest variations of the observed variables in Serbia.

23

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

Variables

GDP

CIT

VAT

SSC

EXC

1

-.214

.135

.142

.227

.365

.570

.551

.337

20

20

20

20

20

Pearson Correlation

-.214

1

.726**

.767**

-.480*

Sig. (2-tailed)

.365

.000

.000

.032

20

20

20

1

.799

Pearson Correlation GDP

Sig. (2-tailed) N

CIT

N VAT

20

20

Pearson Correlation

.135

.726

Sig. (2-tailed)

.570

.000

20

20

Pearson Correlation

.142

.767

.799

Sig. (2-tailed)

.551

.000

.000

20

20

Pearson Correlation

.227

-.480

-.536

-.489

Sig. (2-tailed)

.337

.032

.015

.029

20

20

20

20

N SSC

N EXC

N

**

20 **

**

-.536*

.000

.015

20

20

1

-.489* .029

20 *

**

20 *

20 *

1 20

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

Table 4. Correlation matrix Source: Authors based on SPSS

Based on the results, it can be concluded that there is no significant relationship between tax forms and gross domestic product. Also, there is a strong positive and significant correlation between corporate income tax (CIT), value added tax (VAT) and social security contributions (SSC). On the other hand, there is a statistically significant correlation between excises and observed tax forms.

Variable

VIF

1/VIF

CIT

5.64

0.177303

VAT

3.41

0.292864

SSC

2.19

0.457391

EXC

2.18

0.458307

Mean VIF

3.36

Table 5. Multicollienarity test

Table 6 includes VIF test for independent variables where the results show an absence of multicollinearity between them. Namely, the average value of VIF test is 3.36, which is less than the reference value of 10.

24

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

Variable

RE model

FE model

Coef.

Std. Err.

Coef.

Std. Err.

CIT

1.59 (0.552)

2.61

-2.13 (0.333)

2.20

VAT

3.09 (0.035)

1.32

1.15 (0.043)

1.09

SSC

1.54 (0.656)

3.37

2.16 (0.565)

3.76

EXC

-0.30 (0.827)

1.36

1.44 (0.242)

1.23

_cons

-53.69 (0.166)

36.72

-39.19 (0.332)

40.38

R-square

0.3474

Hausman test Observation

0.1457 0.3012

20

20

Table 6. Model estimation

In order to select an adequate model, Hausman test is included in the analysis, and the results of this test show that random effect model is quite appropriate (p-value = 0.3012). The results of random effect model emphasize that corporate income tax, value added tax and social security contributions have a positive impact on the gross domestic product, while excises negatively affect the gross domestic product. Also, value added tax is an only tax which has a significant impact on the gross domestic product in the analyzed period. Looking at the character of tax effects, value added tax causes the highest change of gross domestic product compared to other taxes. It means that 1% increase of value added tax raises gross domestic product by 3.09%. These findings are logical because value added tax is the most generous tax in these countries.

CONCLUSION Taxes have a fundamental role and place in the economy of each country and they have to be determined at an optimum level in order to provide contribution and prosperity for the economy. The role and impact of taxes on the gross domestic product is widely discussed in the world. There are many types of research which reflected the negative impact on GDP and a small number of research that manifested a positive relationship between taxes and GDP. Using a panel regression model for Serbia and Croatia this paper has shown that there is a positive impact of corporate income tax, value added tax and social security contributions on the gross domestic product, while excises have a negative impact on the gross domestic product. Value added tax is the only tax which has statistically significant impact on the gross domestic product. It is the logical result because value added tax is the most generous tax in Serbia and Croatia. The research confirms previous analysis of observed variables in the world, and the novelty of this paper is reflected in the fact that main tax forms such as personal income tax and corporate income tax do not have significant impact on the gross domestic product.

Enabling informative support to policymakers about tax importance and its effects on economic growth 25

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

in the presented countries represents the main contribution of this study. Bearing in mind that value added tax significantly affects the gross domestic product, governments should make additional efforts to increase revenues of this tax form in order to enhance economic growth. Likewise, the presented model gives an empirical contribution to previous studies and provides a possibility to be applied in other countries in the region.

REFERENCES Alesina, A., & Ardagna, S. (2010). Large changes in fiscal policy: taxes versus spending. Tax Policy and the Economy, 24, 35-68. DOI: 10.3386/w15438 Angahar, P. A., & Alfred, S. I. (2012). Personal income tax administration in Nigeria: challenges and prospects for increased revenue generation from self employed persons in the society. Global Business and Economics Research Journal, 1(1), 1-11.

Aničić, J., Laketa, M., Radović, B., Radović, D., & Laketa, L. (2012). Tax policy of Serbia in the function of developing the economic system. UTMS Journal of Economics, 3(1), 33-43.

Anyanwu, J. C. (1993). Monetary Economics: Theory, Policy and Institutions. Onitsha, Nigeria: Hybrid Publishers.

Appah, E., & Oyandonghan, J.K. (2011). The Challenges of Tax Mobilization and Management in the Nigerian Economy. Journal of Business Administration and Management, 6(2), 128-136.

Arbutina, H. (2000). Value Added Tax in Croatia - An (Almost) Perfect Tax in an Imperfect Environment. Revenue Law Journal, 10(1), 107-118.

Arnold, J., Brys, B., Heady, C., Johansson, A., Schwelnuss, C., & Vartia, L. (2011). Tax Policy for Economic Recovery and Growth. The Economic Journal, 121(550), 59-80. DOI:10.1111/j.1468-0297.2010.02415.x Bania, N., Gray, J., & Stone, J. (2007). Growth, taxes and government expenditures: growth hills for U.S. states.

National Tax Journal, 60(2), 193-204.

Barro, R., & Redlick, C. (2011). Macroeconomic Effects of Government Purchases and Taxes. Quarterly Journal of Economics, 15369, 51-102. DOI:10.3386/w15369 Bhartia, H. L. (2009). Public Finance. New Delhi: Vikas Publishing House PVT Ltd. Chigbu. E.E, Akujuobi, L. E., & Appah, E. (2012). An Empirical Study on the Causality Between Economic Growth And Taxation in Nigeria. Current Research Journal of Economic Theory, 4(2), 29-38.

Chigbu, E.E., & Njoku, C.O. (2015). Taxation and the Nigerian Economy: (1994-2012). Management Studies and Economic Systems, 2(2), 111-128.

Ferede, E., & Dahlby, B. (2012). The impact of tax cuts on economic growth: Evidence from the Canadian provinces. National Tax Journal, 65(3), 563-594.

Gale, W., Krupkin, A., & Rueben, K. (2015). The Relationship between Taxes and Growth at the State Level: New Evidence. National Tax Journal, 68(4), 919-942.

Gemmell, N., Kneller, R., & Sanz, I. (2011). The Timing and Persistence of Fiscal Policy Impacts on Growth: Evidence from OECD Countries. The Economic Journal, 121(550), 33-58. DOI:10.1111/j.1468-0297.2010.02414.x Helms, L. J. (1985). The Effect of State and Local Taxes on Economic Growth: A Time Series Cross-Section Approach. Review of Economics and Statistics, 67(4), 574-582. DOI:10.2307/1924801 Ibadin, P. O., & Oladipupo, A. O. (2015). Indirect taxes and economic growth in Nigeria. Ekonomska misao i praksa, 2, 345-364.

International Monetary Fund. (2018). IMF country information. Retrieved April 17, 2018, from http://www.imf. org/external/country Kalaš, B., Pjanić, M., Milenković, N., & Andrašić, J. (2016). Comparative Analysis Paying Taxes Indicator: Evidence from Western Balkans Countries and Turkey. International Journal of Management, Accounting and Economics, 3(4), 222-232.

26

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

Lee, Y., & Gordon, R. (2005). Tax Structure and Economic Growth. Journal of Public Economics, 89(5-6), 10271043. DOI:10.1016/j.jpubeco.2004.07.002 Mankiw, G., Weinzierl, M., & Yagan, D. (2009). Optimal Taxation in Theory and Practice. Journal of Economic Perspectives, 23(4), 147-174. DOI:10.3386/w15071 McBride, W. (2012). What is the Evidence on Taxes and Growth? Tax Foundation, Special Report 207, December 18, 2012.

Mitra, P., & Stern, N. (2003). Tax systems in transition. World Bank Policy Research Working Paper 2947. Washington, DC: The World Bank.

Myles, G. (2000). Taxation and Economic Growth. Fiscal Studies, 21(1), 141-168. DOI:10.1111/j.1475-5890.2000. tb00583.x Nerre, B., Dragojlović, A., Ranđelović, S., & Đenić, M. (2014). Tax reform in Serbia: Experiences and perspectives.

In Tax Reforms: Experiences and Perspectives: Conference Proceedings, 20 Jun 2014 (pp. 79-96). Zagreb: Institute of Public Finance.

Onakoya, A.B., & Afintinni, O.I. (2016). Taxation and economic growth in Nigeria. Asian Journal of Economic Modelling, 4(4), 199-210.

Padovano, F., & Galli, E. (2001). Tax rates and economic growth in the OECD countries (1950-1990). Economic Inquiry, 39(1), 44-57. DOI: 10.1111/j.1465-7295.2001.tb00049.x Pere, E., & Hashorva, A. (2011). Tax Systems in West Balkans Countries. Romanian Economic and Business Review, 6(2), 80-94.

Ranđelović, S. (2008). Dual Income Tax - An Option for the Reform of Personal Income Tax in Serbia? Economic Annals, 178/179, 183-197.

Ranđelović, S. (2010). Unapređenje performansi poreskog sistema Srbije kroz reformu poreza na dobit. In Poreska politika u Srbiji: Pogled unapred (pp. 75-94). Beograd: USAID Sega projekat. (In Serbian).

Reed, R. (2008). The robust relationship between taxes and U.S. state income growth. National Tax Journal, 61(1), 57-80.

Romer, C., & Romer, D. (2010). The Macroeconomic Effects of Tax Changes: Estimated Based on a New Measure of Fiscal Shocks. American Economic Review, 100(3), 763-801. DOI:10.3386/w13264 Salami, G.O., Apelogun, K.H., Omidiya, O.M., & Ojoye, O.F. (2015). Taxation and Nigerian economic growth.

Research Journal of Finance and Accounting, 10(6), 93-101.

Stevanović, M., & Gajić, A. (2013). Analysis of development of the tax system in Serbia. Annals of the Oradea University, 2, 300-304. Šimović. H., Blažić, J., & Štambuk, A. (2014). Perspectives of tax reforms in Croatia: expert opinion survey. Financial theory and practice, 38(4), 405-439. DOI:10.3326/fintp.38.4.2 Tomljanovich, M. (2004). The role of state fiscal policy in state economic growth. Contemporary Economic Policy, 22(3), 318-330. DOI:10.1093/cep/byh023 Tosun, M.S., & Abizadeh, S. (2005). Economic growth and tax components: An analysis of Tax changes in OECD. Applied Economics, 37(19), 2251-2263. DOI:10.1080/00036840500293813

27

EJAE 2018  15 (2)  17-28

KALAŠ, B., MIROVIĆ, V., MILENKOVIĆ, N.  THE RELATIONSHIP BETWEEN TAXES AND ECONOMIC GROWTH: EVIDENCE FROM SERBIA AND CROATIA

ODNOS IZMEĐU POREZA I EKONOMSKOG RASTA: PRIMER SRBIJE I HRVATSKE

Rezime: Ova studija predstavlja empirijsku analizu poreza i ekonomskog rasta u Srbiji i Hrvatskoj u periodu 2007-2016. Kako bi se utvrdio uticaj poreskih oblika na ekonomski rast i njihov odnos, autori su postavili panel regresioni model gde je bruto domaći proizvod zavisna varijabla, dok su porez na dobit preduzeća, porez na dodatu vrednost, doprinosi za socijalno osiguranje i akcize nezavisne varijable. Rezultati modela slučajnog efekta prikazuju pozitivan uticaj poreza na dobit preduzeća, poreza na dodatu vrednost i doprinosa za socijalno osiguranje, dok akcize negativno utiču na bruto domaći proizvod. Međutim, samo porez na dodatu vrednost ima statistički signifikantan uticaj na ekonomski rast u ovim zemljama, pri čemu svako povećanje prihoda po osnovu ovog poreza doprinosi rastu bruto domaćeg proizvoda u posmatranom periodu.

28

Ključne reči: porezi, ekonomski rast, odnos, Srbija, Hrvatska.

EJAE 2018, 15(2): 29-42 ISSN 2406-2588 UDK: 658.286.4:502/504(669)"1984/2014" 005.591.1 DOI: 10.5937/EJAE15-17360 Original paper/Originalni nauÄ?ni rad

TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA Adedayo Emmanuel Longe1,*, Kayode Daniel Ajulo1, Olawunmi Omitogun2, Emmanuel Olajide Adebayo3 1 Center for Petroleum, Energy Economics and Law, University of Ibadan, Ibadan, Oyo State, Nigeria 2 Department of Economics, Olabisi Onabanjo University, Ago-Iwoye, Ogun State, Nigeria 3 Department of Economics, Bowen University, Iwo, Osun State, Nigeria

Abstract: This study investigates trade, transportation and environment nexus in Nigeria. The main objective of the study is to incorporate transportation activities and trade into the EKC model seeing transportation as an important factor that determines environmental degradation in Nigeria. Annual data from 1984 to 2014 is used in the study, whereas ARDL estimation technique was adopted to analyse the data. The results show that all the variables have a positive significant correlation with carbon emissions in Nigeria except trade that was negative and insignificant. According to ARDL estimates,trade, import transport services and GDP per capita have positive impact on CO2 emissions in the long-run while in the short-run, the result shows that trade, GDP per capita, energy consumption and transport services are capable of correcting about 74% deviation of carbon emissions back to long-run equilibrium. Based on the findings, drawn conclusion stipulates that energy consumption and trade are the main determinant factors that contribute to carbon emission and appropriate attention should be given to energy consumption by considering alternative efficient energy sources to curb the increasing emission in Nigeria.

Article info: Received: May 8, 2018 Correction: May 28, 2018 Accepted: June 1, 2018

Keywords: trade, transportation, CO2, ARDL.

INTRODUCTION The impact of ozone layers depletion has been a matter of great concern in recent years. Many factors are considered to contribute significantly to environmental degradation, but economic activities and energy consumption are recognized as the initiators of the phenomenon (Avetisyan et al. 2010).

The increase in economic activities was traced to high influx of people to the urban centers as a result of more opportunities available in the cities (Achike et al. 2012). The migration act positively contributes * E-mail: longeemmanuel28@gmail.com

29

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

to the number of vehicles in the urban centers which results in traffic congestion, as transport services experienced a positive change in developing countries such as Nigeria where it occurs (Fenger, 2009).

Studies on the trade, transport and environmental degradation are mixed as they are mostly considered separately. Among the studies on the nexus between trade and environmental degradation, Islam, Cheng and Rajib (2012); Ahmad and Wyckoff (2013); Kozul-Wright and Fortunato (2012); Keho (2016); Ali, Zaman and Ali (2015); and Fernandez-Amador, Francois, & Tomberger (2016) support theoretical assumption that trade activities increase carbon emissions. On the other hand, Hasson and Masih (2017), and Athula (2011) argued against the assertion that in the long-run, there is a high probability that trade will reduce carbon emissions through the use of improved technology in the transport sector.

This, however, proves that transport is an important determinant of environmental degradation, and confirms the assumptions of Xie, Fang and Liu (2017); Wang, Xie and Yang (2016) who claimed that transportation activities significantly contribute to carbon emissions, out of which most emissions are from trucks (Konur and Schaefer, 2014).

In Nigeria, the use of fossil fuel is common in the transportation sector, being the major source of the transport sector power. It was also documented that a larger percentage of the imported oil products are consumed in the transport sector through trade activities (Achike et al. 2012). As well as other developing countries, Nigeria was also found to be a country that consumes more finished goods, and exports less (Achike et al. 2012). This, however, implies that both import and export activities in the country contribute more to environmental degradation. As documented by World Bank (2017), transport services in the Nigerian economy contribute more to environmental degradation than trade activities. Between 1977 and 1987, the percentage of trade to GDP, percentage of transport services to commercial export and import services and carbon emissions were recorded to be 41.65%, 16.91%, 41.09% and 41.07% respectively. Between 1988 and 1997, trade and carbon emissions increased to 76.86% and 52.96% respectively, while transport services from export and import increased to 11.55% and 15.86%. As trade experienced a fall between 1998 and 2007 to 64.46%, carbon emission also reported a fall to 44.05%, while transport service through export and import increased to 75.59% and 32.16% respectively. In 2014, it was recorded that trade and carbon emission decreased further to 30.89% and 35.89%, while transport services in the export sector increased to 51.37% and from the import sector increased to 37.89% (WDI, 2017). This implies that the correlation between trade and carbon emissions is high, while it’s lower between transport services and carbon emissions (See figure 1). On this note, it is important to investigate the relationship between trade, transportation services and the environment in the context of Nigeria.

Figure 1. Trend Analysis of Trade, CO2 and Transport Services Source: Authors, (2018)

30

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

The rest of the study is partitioned into four parts; section two give details on the literature review, section three presents the data source and model specification, section four reveals the analytical framework, while section five entails the conclusion and recommendation(s).

LITERATURE REVIEW Theoretically, argument on the factors responsible for increased environmental degradation is noted through the submissions of Environmental Kuznets Curve (1955), Environmental Daly Curve Hypothesis (1973), and Brundtland Curve Hypothesis (WCED 1978) and has grabbed the attention of researchers on the phenomenon in developed and developing countries. Empirically, Avetisyan et al. (2010) incorporated global data on trade, transportation models, transport emissions, and output emissions in North-America and noted that 90% of global trade related emissions in machinery exports are due to international transportation. Similarly, Cristea et al. (2013) using U.S. database observed that trade contributes less to emission compared to the amount of emissions from transportation.Using a mathematical model to analyze the relationship between international trade and carbon emissions in Bangladesh; Islam et al. (2012) argued that the trade activities in Bangladesh significantly contribute to emissions.

By adopting conservative approach to investigate carbon emissons embodied in international trade of good in OECD countries; Ahmad and Wyckoff (2013) noted that the amount of carbon emissions embodied in gross flow of imports and exports is relatively significant. Giving instances, the study revealed that the emissions embodied in imports and exports of Sweeden’s economy amounts to 50% of its total domestic production of emissions, while the emission embodied in imports to the United States represents 2.5% of the global carbon emissions from fossil fuel combustion. In China, Feng et al. (2013) examined carbon emissions and export trade nexus in China using input-output model.

The study verified that larger proportion of emissions in the economy are from export trade activities such as textile, chemicals and plastic products among others. Similarly in the U.S. economy, Weber et al. (2007) investigated the import trade, transportation and carbon emissions nexus using a multiregional input-output analysis. The study confirmed that the supply chain of importation, transport and retail activities positively relates to carbon emissions. The study suggested two approaches which are: environmental supply chain management by industrial sectors, and international rationalization of environmental policy by the government.

Using STIRPAT model to incorporate the effect of transportation infrastructure on urban carbon emissions in China; Xie et al. (2017) discovered that construction of transportation infrastructure, population size, per capita GDP, energy intensity, and industrial structure contribute to rises in urban carbon emissions and intensity. In clarifying this, they further revealed that in large-scale cities, construction of transport infrastructure positively affect urban carbon emissions and intensity and it is the same for medium cities, while for small scale cities their impact is insignificant. Wang et al. (2016) ascertained from their findings for the economy of Jiangsu that growth in GDP and transportation contribute masively to the increase in carbon emissions. Integrating inventory control and transportation decisions under carbon emissions regulatios, Konur and Schaefer (2014) discovered that emission generated from vehicles (trucks) forms majority of the emissions from the transportation sector.

Takarada et al. (2014) used two model general equilibrium model to analyse international transport sector and welfare nexus. It was noted that there was a unilateral reduction in the number of pollution permit from international transport sector and individual welfare.

31

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

Also, Kozul-Wright and Fortunato (2012) studied the relationship between international trade and carbon emissions. They presented that a positive relationship exists between trade and emssions in the developed countries and likewise for the less industrialised countries. Keho (2016) determined the effect of trade openness on environmental among eleven (11) ECOWAS countries using co-integration bounds test and granger causality. The study detected a positve longrun relationship between carbon emissions and trade, and a bi-directional long-run causality running from pollution to economc expansion in four (4) ECOWAS countries. This is attributable to energy use by high-polluting industries, and importation of used car. In Pakistan, Ali, Zaman and Ali (2015) also confirmed that trade openness which is a result of foreign direct investment inflow, causes environmental degradation through old technogies. While in South Africa, Hasson and Masih (2017) used ARDL approach to investigate the relationship between energy consumption, trade openness, economic growth, carbon emissions, and electricity consumption. The study revealed that there is a cointegrating relationship between energy consumption, trade openness, economic growth, carbon emissions, and electricity consumption. The study argued that in the longrun, higher level of trade openness reduces carbon emissions via the promotion of environmental technology development that brings about less pollution and energy efficiency. In Sri Lanka, Athula (2011) argued that there is no longrun causality between trade openness and carbon emissions, but a uni-directional causality running from trade to carbon emissions was identified in the short-run. This implies that an increase in trade will cause carbon emissions to rise in the short-run.

Fernandez-Amador et al. (2016) examined the relationship between carbon dioxide emissions and international trade. The study identified a link between global supply and demand of goods and services, international trade, and carbon emissions. Investigating the effect of carbon footprints, fuel subsidies and market-based measures on transport, trade and climate change in Germany, Monkelbaan (2011) argued that a good carbon footprint approach is reducing carbon intensity without inhibiting compliance cost. The study supported fuel subsidy removal in order to relieve governmental fiscal burden during times of economic crisis and free up resources for spending on other priorities. Peters et al.

(2009) revealed that it is possible to reduce carbon emisisons through international trade by 60% fossil fuel combustion and processed emissions from transport.

Likewise, Kim, Janic, & Wee (2010) adopted linear programming-based algorithm to evaluate the trade-off between carbon dioxide emissions and logistics cost. The study revealed that trade-off curves exhibit a linear relationship which indicates that the rise in freight costs represent a prerequisite for carbon emisisons reduction. The study also established that an increase in the lower carbon emitting system’s capacity would further reduce carbon emissions. In addition, Shapiro (2016) used a structural general equilibrium model and noted that gains from international trade were greater than the environmental cost of international trade. Zhang, Liu, & Bae (2017) studied the effect of trade openness on carbon emissions in 10 newly industrialized countries. It was ascertained from the findings that in the long-run, there is correlation between international trade, carbon emissions, and real GDP, and trade activities reduces carbon emission through tecnological transfer. In China, Gozgor & Can (2016) argued that energy consumption and trade openness positively influence carbon emissions, while export quality has an adverse effect on carbon emissions.

In Nigeria, Achike et al. (2012) examined the determinants of greenhouse gas emission and its impact on trade, climate change mitigation and adaptation of policies using Zellner’s seemingly unrelated regression model. The study revealed that agricultural land expansion in all forms, farm technology based on increasing use of farm machineries and tractors, deforestation (forestry commercial activities), and the use of fossil fuel as a means of energy all contributed significantly to the level of carbon emissions in Nigeria.

32

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

From the reviewed studies, it is noted that most of them do not have a unite conclusion due to the scope, methodology and the country focus. While some studies (Xie et al. 2017; Islam et al. 2012; Wang et al. 2016; Achike et al. 2012; Gozgor and Can, 2016; among others) argued a positive link between trade activities and emissions, studies of Perers et al. (2009); Cristea et al. (2013), Hasson and Masih (2017) argued a negative nexus on the phenomenon. This study however deviates by using the ARDL model to test the EKC curve and considering transportation activities as one of the factors that significantly contributes to increasing environmental degradation in Nigeria.

METHODOLOGY AND DATA USED The study used annual secondary data sourced from the World Development Indicators (WDI), (2017). The data scope spans between 1984 and 2014. This is due to the data available on used energy and carbon emissions. The data used are as follows: Carbon emissions (Kt) proxy for environment, Energy use (kg oil equivalent per capita), transport services (% of commercial services export), transport services (% of commercial services import), trade (calculated as the ratio of import plus export to GDP), and GDP per capita (at current LCU). The data are analysed using E-views statiscal package, 9th version. See Appendix 1 for the data used.

The study follows the model of existing studies such as Avetisyan et al. (2010); Cristea et al. (2013); Gozgor and Can (2016) and Zhang et al. (2017); on the validation of the EKC model. Our model also proposes that trade and transport services in the economy are important determinants of carbon emissions. Thus, the model is written as; CO2t = f (TRt , GDPt , EU t , TREt , TRI t )

(1)

Taking the natural logarithm of equation (1), the function is re-specified as follows; InCO2t =+ α 0 β1TRt + β2GDPt + β3 InEU t + β 4TREt + β5TRIt + ε t

(2)

From equations (2), InCO2t implies log form of carbon emissions, TRt is trade captured as the ratio of trade to GDP, InGDPt is log form of gross domestic product per capita, InEUt is the log form of energy use in Kg oil equivalent per capita, TREt is the log form of transport services as a percentage of commercial services export, TRIt is transport services as a percentage of commercial services import and ε t is the error term. All at time t. α 0 is the model intercept, while β1 − β5 are the coefficients of the parameters.

ARDL estimation technique is adopted to estimate the parameters of the model. The ARDL estimation model for the short-run and long-run relationship among the variables is specified as; n

n

n

n

∆InCO2t = ϑ0 + ∑ ρ1∆CO2t −k + ∑ ρ2 ∆TRt −k + ∑ ρ3 InGDPt −k + ∑ ρ 4 ∆InEU t −k q 1 = q 1= q 1 =

n

+ ∑ ρ ∆TRE

n

q 1 =

+ ∑ ρ6 ∆TRI t −k + β1CO2t −1 + β 2TRt −1 + β 3GDPt −1

t −k 5 q 1= q 1 =

(3)

+ β 4 InEU t −1 + β 5TREt −1 + β 6TRIt −1 + δ ecmt −1 + ε t 33

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

From equation (3), the ∆ denotes the changes in the variables in the short-run, n is the optimal lag length, ε t - error term at time. The parameters ρ ( i = 1, 2, 3, 4, 5, 6) are the corresponding long-run multiplier, and the parameters β (1, 2, 3, 4, 5 and 6)are the short-run dynamic of the ARDL model. δ ecm is the parameter estimate of the error correction model.

ANALYTICAL FRAMEWORK The study carried out a pre-estimation test to understand the behaviour of the data used in the study.

The tests include; correlation test, descriptive statistics, unit root, and ARDL Bounds test.

Correlation Test The correlation test shows that there is a weak negative insignificant correlation between trade and carbon emission. There is a very low positive insignificant correlation between import transport services and carbon emission. There is a strong positive significant correlation between export transport services, GDP, energy consumption and carbon emission. Since the correlation values are not up to 0.95, this implies that multicollinearity problem is absence among the variables. Table 1 presents the result.

InCO2 InCO2

TR

TRI

TRE

InGDP

InEU

1.0000 -----

TR TRI TRE InGDP InEU

-0.1381

1.0000

(0.4666)

-----

0.2506

-0.5995

1.0000

(0.1817)

(0.0005)

-----

0.7036

-0.3591

0.4151

1.0000

(0.0000)

(0.0513)

(0.0226)

-----

0.5583

0.2711

-0.2320

0.4639

1.0000

(0.0013)

(0.1473)

(0.2174)

(0.0098)

-----

0.7028

0.1067

0.1436

0.5853

0.8634

1.0000

(0.0000)

(0.5748)

(0.4490)

(0.0007)

(0.0000)

-----

Note: The probability values are in parenthesis ( ) and * denotes significance level at 1%.

Table 1. Correlation Matrix Test Source: Authors Computation (2018)

Descriptive Statistics The result in table 2 shows that, on average, changes in environmental quality, trade activities, transportation activities, gross domestic products and energy use, range between 11.1%, 0.53%, 33.9%, 10.1% and 6.6% respectively as a result of policy adjustments among other measures. Also, it can be 34

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

deduced that the variables reaction to economic shocks are not outrageous or far from the expected equilibrium. The result also confirms that all the variables are normally distributed through the JarqueBera statistics which has a probability value greater than 10%.

InCO2

TR

TRI

TRE

InGDP

InEU

Mean

11.10033

0.533829

33.278

35.89968

10.1386

6.577303

Maximum

11.57184

0.818128

53.21

93.35

13.14374

6.682488

Minimum

10.46879

0.236089

10

2.88

6.671149

6.509513

Std. Dev.

0.397434

0.158609

11.7779

29.08345

2.070137

0.049624

Jarque-Bera

3.666481

1.370994

1.601042

3.721383

1.982951

2.179895

Probability

0.159895

0.50384

0.449095

0.155565

0.371029

0.336234

31

31

31

31

31

31

Observations

Table 2. Descriptive Statistics Source: Authors Computation (2018)

Unit Root Test The variables follow a unit root process and stationary after first differencing except energy use stationary at level using the Augmented Dickey Fuller (ADF) and Phillip Perron (PP) test. This implies that the empirical model is not appropriate for cointegration analysis. This facilitates the use of ARDL bounds test and ARDL estimation. The result is presented in table 3 below.

ADF Variables

PP

Levels

First difference

Levels

First Difference Order of Integration

InCO2

-1.987

-5.216*

-1.987

-5.216*

I(1)

InEU

-3.246**

-4.734

-2.436

-7.096*

I(1)

InGDP

-1.798

-5.588*

-1.812

-7.390*

I(1)

TR

-1.368

-8.654*

-2.035

-26.570*

I(1)

TRE

-2.611

-5.633*

-2.592

-5.657*

I(1)

TRI

-1.898

-5.159*

-1.808

-12.004*

I(1)

*, ** implies 5% and 10% level of significance Table 3. Unit Root Test Result Source: Authors Computation (2018)

Optimal Lag Test The study carried out an optimal lag test to select the appropriate lag length for the data used in the study. From the criteria available in the test, this study used the Akaike Information Criterion (AIC) result to select the optimal lag length. The result reveals an optimal lag length of 2 which is applied for this study (see table 4).

35

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

Lag

LogL

LR

FPE

AIC

SC

HQ

0

-172.18250

NA

0.03613

13.70634

13.99667

13.78995

1

-52.20761

175.3478*

6.14e-05*

7.24674

9.279049*

7.831971*

2

-13.55360

38.65402

0.00008

7.042584*

10.81687

8.12944

* implies the selected lag length for each criterion Table 4. Optimal Lag Length Results Source: Authors computation (2018)

ARDL Bounds Test The bounds test result is provided in table 5. The result shows that there is a strong statistically significant cointegration relationship among the variables at 5% level. This is confirmed by the superior value of the F-statistics over the critical values of the lower and upper bounds class.

Model for Estimation

F-Statistics

Lower-Upper bound at 5%

Fco (InCO2t / InTRt /InGDPt /EUt /TRIt / TREt )

4.78*

3.12-4.25

*implies cointegrating factors at 5% level of significance.

Table 5. ARDL Bounds Test Source: Authors Computation (2018)

ARDL Estimation Table 6 provides the short-run and the long-run estimate of the parameters. In the long-run, the result shows that the effect of trade is negative and elastic (-1.91). Import transport services are negative and elastic (0.079). Export transport services are positive and significant (0.010). GDP effect is negative and elastic (-2.45). Energy consumption is positive and elastic (13.22).

In the short-run, the model uses an optimal lag of 2. This implies that carbon emission reacts to changes in the independent variables after 2years. The result shows that trade impact is positive and elastic (0.99). Import transport services are positive and elastic (0.011). Export transport services are positive and elastic (0.008). GDP and Energy use effect are positive and inelastic (0.29) and (2.49) respectively. The error correction model result shows that CO2 converges to equilibrium in the long-run by 0.74% at 5% significance level through the channels of trade, import and export transport services, GDP and energy use.

The stability and validity of the model is also tested using the serial correlation test, heteroscedasticity test and Recursive test. The result verified a serial correlation problem among the variables, but the absence of heteroscedasticity problem and assurance of the model stability at 5% significance level through the CUSUM and CUSUM squares test result retains the model (see table 6 and figure 1 & 2).

36

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

ARDL Long-run Estimate Selected Model: ARDL (1, 2, 2, 2, 2, 2)

Nigeria

InCO2

TR

-1.914 (0.0267)**

TRI

-0.079(0.0028)*

TRE

0.010(0.0476)**

InGDP

-2.452(0.0007)*

InEU

13.217(0.0153)**

∆TRt-1

0.127(0.7494)

∆TRt-2

0.986(0.0486)**

∆TRIt-1

-0.032(0.0147)

∆TRIt-2

0.011(0.0793)***

∆TREt-1

0.014(0.0727)

∆TREt-2

0.008(0.0629)***

ARDL Short run Estimate Selected Model: ARDL (1, 2, 2, 2, 2, 2)

∆InGDPt-1

-0.313(0.2095)

∆InGDPt-2

0.286(0.4001)

∆InEUt-1

7.194(0.0088)

∆InEUt-2

2.486(0.3429)

ECMt-1

-0.743(0.0062)*

R2

Overall

0.97

F-Statistic

Overall

13.02[0.0004]

Breusch-Godfrey Serial Correlation LM Test:

F-Stat

6.617424[0.0304]

Heteroskedasticity Test: Breusch-Pagan-Godfrey

F-Stat

1.988655[0.1484]

Stability Tests

CUSUM

Stable at 5%

Stability Tests

CUSUM of Squares

Stable at 5%

Note: *, **, *** implies significance level at 1%, 5%, and 10% respectively, and the probability values are in parenthesis ( ) and [ ] Table 6. ARDL Estimates and Diagnostic Test Results Source: Authors’ Computation (2018)

Figure 2. CUSUM Test

Figure 3. CUSUM of Square 37

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

Discussion of Findings From our empirical findings, firstly, the EKC model is validated in the context of Nigeria in the long-run. Similar to the study of Gozgor and Can (2016) for China, it was established that a positive linkage exists between growth per capita and carbon emissions in Nigeria. This implies that growth per head intensifies CO2 emissions in the long-run. Though, the variable shows a negative impact in the short-run after an adjustment of two years period, but in the long-run where increase in growth is needed by most developing countries to meet the increasing population needs, carbon emission increases and exposes the environment to more hazzards.

Second, energy consumption exerts same sign effect on carbon emissions. The result is in consonance with the study of Gozgor and Can (2016). The effect was significant in the short-run, and insignificant in the long-run. Given that Nigeria consumes more of fossil fuel product, it is not surprising that the result turns out to be similar to that of the economy of China. Adopting policies to reduce the consumption of fossil fuel product may not be the best, but considering alternative source of energy such as the renewable source, may be the reason for the insignificant effect in the long-run.

Third, trade in the short-run has a weak negative elastic impact on carbon emissions and in the long-run the positive impact is very strong and inelastic. This implies that trade activities at the earlier stage of the economy are tailored towards reducing carbon emissions, but as the activities expand, more technologies that consume more emission associated energy products are used. The findings also confirmed the proposition of the EKC model in the long-run for Nigrian economy on trade activities.

Fourth, import transport services in the short-run are elastic and negative, while they are positive and elastic in the long-run. This result shows that transportation of import goods in the economy significantly contributes to climate change experienced in recent years in the country through the emission they release.

Fifth, export transport services also exert a weak positive significant impact on CO2 emissions in Nigeria. This implies that the variable is a strong determinant of CO2 emissions increase in the economy through its activities.

Our findings also confirm the posibility of correcting the divergence of CO2 emissions by 74% back to long-run equilibrium if appropriate measures are taken.

CONCLUSION AND RECOMMEDATION(S) The paper examines the relationship between environment, trade, energy consumption, export and import transport services and GDP per capita in Nigeria between 1984 and 2014. The study uses ARDL estimation technique to validate the relationship among the variables. It was observed that trade, import transport services and GDP per capita all have a negative and positive impact on CO2 emissions in the short-run and long-run respectively in Nigeria. Export transport services and energy consumption both show a positive relationship with CO2 emission in the short-run and long-run. The study concludes from the findings that trade, energy use, GDP and tranport services are important determinant of CO2 emissions in Nigeria and are capable of correcting emissions in the environment to fit the long-run equilibrium expectations if appropriate measures. Therefore, a major recommendation from the findings is that, energy consumption in the economy should be given utmost attention as it 38

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

is an important variable to be used in all the sectors of the economy. This can be done by considering alternative energy sources (such as the renewables) that are enviromentally friendly, to reduce the positive impact of energy consumption in the short-run and long-run of the economy via its usage.

The work can be used as a basis for further research on the impact of trade, transport and the environment in other countries of the same level of development as Nigeria (African continent or other continents) and the findings can be compared with the circumstances in Nigeria. The study is however limited in terms of scope through data availabilty.

REFERENCES Achike, A.I., Onoja, A.O., & Agu, C. (2012). Green house gas emission determinants in Nigeria: implications for trade, climate change mitigation and adaptation policies. Retrieved from http://www.trapca.org/workingpapers/revised-version-of-paper-by-achike-onoja-and-Agu.pdf Ahmad, N., & Wyckoff, A. (2013). Carbon dioxide emissions embodied in international trade of goods. OECD Science, Technology and Industry Working Papers, 2003/15 1-66.

Ali, Z., Zaman, Z., & Ali, M. (2015). The Effect of International Trade on Carbon Emissions: Evidence from Pakistan. Journal of Economics and Sustainable Development, 6(9), 289-299.http://www.iiste.org/Journals/ index.php/JEDS/article/viewFile/22639/23309 Athula, N. (2011). Does Trade Openness Promote Carbon Emissions? Empirical Evidence from Sri Lanka. The Empirical Economics Letters, 10(10), 972-986. Retrieved from http://hdl.handle.net/10072/42748 Avetisyan, M., Cristea, A., Hummels, D., & Puzzello, L. (2010). Trade and the Greenhouse Gas Emissions from International Freight Transport. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.364 .3822&rep=rep1&type=pdf Cristea, A., Hummels, D., Puzzello, L., & Avetisyan, M. (2013). Trade and the greenhouse gas emissions from international freight transport. Journal of Environmental Economics and Management, 65, 153-173.

DOI:10.1016/j.jeem.2012.06.002 Feng, Z., Xue, J., & Song, Y. (2013). Research on carbon emissions in China’s export trade based on input-output model. Chinese Journal of Population Resources and Environment, 11(1), 1-9. DOI:10.1080/10042857.2013.

777202 Fenger, J. (2009). Air pollution in the last 50 years – From local to global. Atmospheric Environment, 43, 13-22.

DOI:10.1016/j.atmosenv.2008.09.061 Fernandez-Amador, O., Francois, J. F., & Tomberger, P. (2016). Carbon dioxide emissions and international trade at the turn of the millennium. Retrieved from https://www.wto.org/english/res_e/reser_e/gtdw_e/ wkshop16_e/francois_e.pdf Gozgor, G., & Can, M. (2016). Does export product quality matter for CO2 emissions? Environmental Science Pollution Resources, 1-10. DOI 10.1007/s11356-016-8070-6 Hasson, A., & Masih, M. (2017). Energy consumption, trade openness, economic growth, carbon dioxide emissions and electricity consumption: evidence from South Africa based on ARDL. Retrieved from https://mpra. ub.uni-muenchen.de/id/eprint/79424 Islam, M. R., Cheng, Y., & Rajib, M. S. (2012). International Trade and Carbon Emissions (CO2): The case of Bangladesh. Journal of Economics and Sustainable Development, 3 (5), 18-27.http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.855.7442&rep=rep1&type=pdf Keho, Y. (2016). Trade Openness and the Environment: A Time Series Study of ECOWAS Countries. Journal of Economics and Development Studies, 4(4), 61-69. DOI:10.15640/jeds.v4n4a6 Kim, N. S., Janic, M., & Wee, B. v. (2010). Trade-Off Between Carbon Dioxide Emissions and Logistics Costs Based on Multiobjective Optimization. Transportation Research Record, 107-116. DOI: 10.3141/2139-13 39

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

Kozul-Wright, R., & Fortunato, P. (2012). International Trade and Carbon Emissions. European Journal of Development Research, 24(4), 509-529.

Li, J., Lu, Q., & Fu, P. (2015). Carbon Footprint Management of Road Freight Transport under the Carbon Emission Trading Mechanism. Mathematical Problems in Engineering, 1-14.

Monkelbaan, J. (2011). Transport, Trade and Climate Change: Carbon Footprints, Fuel Subsidies and Marketbased Measures. International Centre for Trade and Sustainable Development, 1-60.

Peters, G.P., Marland, G., Hertwich, E. G., Saikku, L., Rautiainen, A., & Kauppi, P. E. (2009). Trade, transport, and sinks extend the carbon dioxide responsibility of countries: An editorial essay. Climatic Change, 97, 379-388. DOI:10.1007/s10584-009-9606-2 Shapiro, J.S. (2016). Trade Costs, CO2, and the Environment. Retrieved from http://www.econ.yale.edu/~js2755/ Trade_CO2_Environment.pdf. 1-64.

Takarada, Y., Ogawa, T., & Dong, W. (2014). Trade, Transportation, and the Environment: Welfare Effects of Emissions Reductions and International Emissions Trading. Retrieved from http://www.etsg.org/ETSG2014/ Papers/188.pdf.

Weber, C. L., Matthews, S. H., Corbett, J. J., & Williams, E. D. (2007). Carbon Emissions Embodied in Importation, Transport and Retail of Electronics in the U.S. A Growing Global Issue. in: Proceedings of the 2007 IEEE Symposium on Electronics and the Environment, 174-179.

Zhang, S., Liu, X., & Bae, J. (2017). Does trade openness affect CO2 emissions: evidence from ten newly industrialized countries? Environmental Science Pollution Resources. DOI:10.1007/s11356-017-9392-8

APPENDIX 1 Data Used

40

Year

Trade (calculated as the ratio of import plus export to GDP) proxy as (TR)

1984

0.236088825

789.3021425

677.7652322

45.52

59.19

69625.329

1985

0.259000637

879.5493062

682.8194438

44.71

73.97

69893.02

1986

0.237167563

872.867951

671.4990085

38.93

27.13

73505.015

1987

0.416466623

1270.270851

676.8560962

41.09

16.91

59343.061

1988

0.353119785

1635.607191

678.8559452

53.21

36.3

70747.431

1989

0.603917611

2460.585406

684.4482507

43.69

2.88

42441.858

1990

0.530302209

2955.288186

697.192079

33.63

3.86

39196.563

1991

0.648765987

3367.268305

712.2481627

45.19

11.26

42273.176

1992

0.610309731

5542.175689

721.9703508

52.98

14.6

46614.904

1993

0.581098489

6960.196381

715.4377996

31.81

17.17

45137.103

1994

0.4230887

8974.894335

680.710066

25.3

13.63

35199.533

1995

0.597678343

18595.84082

682.2695813

22.44

16.4

35841.258

1996

0.576909942

25277.36571

693.778301

10

10.42

39665.939

Transport Energy GDP per capservices (% use (kg oil ita (at current of commerequivalent per LCU) proxy as cial services capita) proxy (GDP) import) proxy as (EU) as (TRI)

Transport services (% of commercial services export) proxy as (TRE)

CO2 Emission (Kt) proxy as (CO2)

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

1997

0.76859991

25603.90787

699.6506998

15.86

11.55

42328.181

1998

0.66173245

24198.89041

687.1179161

17.36

12.84

37869.109

1999

0.558463914

27757.6646

694.1712967

19.85

12.03

40285.662

2000

0.713805312

38555.41187

703.2447175

19.84

12.03

76057.247

2001

0.818128491

39131.13426

720.047237

19.84

12.03

85734.46

2002

0.633836372

55400.52357

724.6112534

19.84

12.03

93677.182

2003

0.752189025

66245.95627

746.612206

22.46

10.4

101616.237

2004

0.484481307

86219.73905

748.3413177

20.17

104304.148

2005

0.507483592

106055.7028

757.9586631

44.11

93.35

106067.975

2006

0.646093139

131191.7073

744.5452271

27.36

88.82

98891.656

2007

0.644629088

143022.3776

750.7831193

32.16

75.59

95055.974

2008

0.649729738

164055.0215

752.8597869

30.6

65.94

96148.74

2009

0.618028542

163443.6517

721.4533582

37.08

62.41

76735.642

2010

0.426513849

349791.642

755.9891516

42.81

75.2

91517.319

2011

0.527941049

391174.5039

778.4993881

35.92

68.6

95694.032

2012

0.443801366

433955.8196

798.3031153

43.45

67.45

98502.954

2013

0.3104886

471456.0508

779.8515426

42.55

57.36

98136.254

2014

0.308851936

510805.4436

763.3913679

38.75

51.37

96280.752

Source: WDI, 2017.

41

EJAE 2018  15 (2)  29-42

LONGE, A. E., AJULO, K. D., OMITOGUN, O., ADEBAYO, E. O.  TRADE, TRANSPORTATION AND ENVIRONMENT NEXUS IN NIGERIA

POVEZANOST TRGOVINE, TRANSPORTA I ŽIVOTNE SREDINE U NIGERIJI

Rezime: Ova studija istražuje odnos trgovine, transporta i životne sredine u Nigeriji. Glavni cilj studije jeste da uključi aktivnosti transporta i trgovina u EKC model, pošto autori smatraju da je transport važan faktor koji utiče degradaciju životne sredine u Nigeriji. Za potrebe ispitivanja korišćeni su godišnji podaci od 1984. do 2014. godine. Tehnika procene ARDL-a je usvojena u cilju analizea podataka. Rezultati pokazuju da sve varijable imaju pozitivnu značajnu korelaciju u odnosu na emisije ugljenika u Nigeriji, osim trgovine koja je bila negativna i beznačajna.

Prema procenama ARDL-a, trgovina, usluge uvoza i BDP po glavi stanovnika pozitivno utiču na emisije CO2 dugoročno, dok u kratkom roku rezultati pokazuju da trgovina, BDP po glavi stanovnika, potrošnja energije i transportne usluge mogu da koriguju čak oko 74% odstupanja emisije ugljenika prema dostizanju dugotrajne ravnoteže. Na osnovu nalaza, zaključak predviđa da su potrošnja energije i trgovina glavni faktori koji doprinose emisiji ugljenika i treba posvetiti pažnju potrošnji energije uzimajući u obzir alternativne efikasne izvore energije kako bi se suzbila rastuća emisija u Nigeriji.

42

Ključne reči: trgovina, transport, CO2, ARDL.

EJAE 2018, 15(2): 43-57 ISSN 2406-2588 UDK: 336.748.12:658.155(497.113) 658.1 DOI: 10.5937/EJAE15-17997 Original paper/Originalni naučni rad

INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA* Goran Anđelić1, Nenad Penezić1, Vilmoš Tot2, Marko Milošević1,* 1 Educons University, Faculty of business Economy, Sremska Kamenica, Serbia 2 Union University-Nikola Tesla, Faculty for strategic and operational management, Belgrade, Serbia

Abstract: Taking into account the current trends in the domestic financial market, the subject of this research is the analysis, testing and quantification of the inflation rate impact on the daily share returns of the real sector companies in AP Vojvodina. The aim of the research is to generate concrete, practically tested and quantified knowledge about the possibilities and efficiency of the GARCH models application to quantify the inflation rate impact on the share returns of the observed companies.

The period covered by the survey is from 2006 to 2016 and it includes the real sector companies of AP Vojvodina whose shares are quoted within the stock exchange index BELEXline. The research results show the exact correlation between the daily return rates of the observed companies and the financial risk factor - the inflation rate. The results also show a positive impact of the variable inflation rate on the returns of companies NIS a.d (0.013876), Sojaprotein a.d (0.019167) and Vital a.d (0.051056), and a negative impact on the returns of the company Veterinary Institute a.d (-0.000183). The research results confirm the role and significance of the application of econometric models in order to quantify financial risk factors of returns from investment activities in the real sector companies of AP Vojvodina.

Article info: Received: September 9, 2018 Correction: September 27, 2018 Accepted: October 3, 2018

Keywords: inflation rate, share returns, GARCH models, real sector, risk, investment.

* The work is a part of a research project titled: “Analysis of Financial Risk Factors in Sustainable Development of AP Vojvodina” (Project Number 142-451-3554/2017-01/02) that was financially supported by the Provincial Secretariat for Higher Education and Scientific Research.

* E-mail: milosevicm1@gmail.com

43

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

INTRODUCTION Stock exchanges are considered important institutions of the financial system because they encourage the growth of all economic sectors by allocating savings from surplus entities to financing the deficit units and enable optimum distribution and use of scarce capital resources, thus providing the basis for long-term sustainable economic growth and development. As the stock market is considered to be a key to economic progress, the research focus is on defining the factors of financial risk that influence and determine the share returns. Although financial theory provides specific factors for companies and industries, there is a growing recognition among financial researchers that macroeconomic variables play a key role in determining a stock market performance.

Among macroeconomic variables, inflation is considered as one of the most important factors that influence share returns. Inflation is an increase in the general level of prices of goods and services in the economy that leads to a purchasing power decrease and/or a money value decrease. The earliest conclusions about the relationship between inflation and share returns are based on hypotheses presented by Irving Fisher in 1930. Based on the Fisher hypothesis, it can be concluded that the share return is directly proportional to the expected inflation rates. Therefore, there should be a positive correlation between the inflation rate and the share returns, whereas the nominal investment share returns should be accompanied by an inflation increase and thus protect investors from the decrease of the real money value.

On the other hand, the opposite view of the negative relationship between inflation and the share returns is also present. Fama (1981) explained that the negative correlation between share returns and inflation was caused by a positive correlation between the share returns and the real activity and the negative correlation between inflation and real activity.

Secondary market shares are sold on the basis of supply and demand, and therefore reflect the quality of a company or a joint-stock company operation. In addition, the economic environment of a company should be taken into account, i.e. the financial risk factors, such as inflation, foreign exchange rate, the state credit rating and similar affect the share returns. According to Adamović (2008), it is very important that a state reduces the hyperinflation to a low level (one-digit inflation) – otherwise, there will be no serious economic progress. But the stabilisation policy must not interact with a policy of economic growth. Stabilisation reduces inflation, but its long-term operation slows down the pace of economic growth and brings poverty. A solid monetary policy that implies an overvalued exchange rate of national currency and high interest rates can not trigger new jobs. Such a policy only slows down every economic development (Kitanović and Krstić, 2010).

Being one of the basic determinants of the time value of money, the expected inflation must be taken into account when investing in stocks or making financial decisions on investments. Therefore, investors have the task to predict factors affecting the portfolio performance and accordingly make decisions based on their own expectations, knowing that the inflation is one of those factors that influence the portfolio. The factors that affect the price of shares are divided into external and internal factors.

External factors determine the macroeconomic environment in which the company operates. The main external factors are economic activity - GDP, inflation, exchange rate, the balance of payments, external debt, conditions in the industrial branch, existing market characteristics, political factors, etc.

Internal factors are the result of the company’s operation (Damnjanović, 2017).

44

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

The market price of shares reflects the quality and performance of the company’s operation, and it is, therefore, an indicator of the justified share investments. Therefore, it is very important for investors in the financial market to approach a detailed financial analysis before assessing the value of the shares in which they intend to invest, as well as to quantify the impact of financial risk factors, such as inflation.

Testing and analysing the impact of the inflation rate on the share return of real sector companies in AP Vojvodina through GARCH econometric models in the domestic financial market, not only provide qualitative information on the impact effectiveness, but also analyse the differences between the observed impacts. The following hypotheses were tested in this research: Basic hypothesis H0: GARCH model successfully tests the inflation impact on the share return of real sector companies in AP Vojvodina.

Furthermore, the additional hypothesis tested in the research goes as follows: H1: The inflation rate positively influences the returns of the real sector companies in AP Vojvodina.

The paper is structured in the following way: the research subject, the goal, and the hypotheses are defined in the introductory notes. The next part of the paper presents relevant research literature. The third part deals with the methodology and sample used in the research. The results and discussion are presented in the last section, which is followed by the conclusions and references. The industrial companies examined in this study are headquartered in AP Vojvodina and they operate (sell their services, merchandise or products) all over the territory of the Republic of Serbia, yet, the research results are reliable and unrestricted.

LITERATURE REVIEW The main tasks of the central bank are to maintain the price stability and the stability of the financial system. In addition, recent experience explicitly indicates that inflation is the result of the interaction of monetary objectives and market expectations and that it must be seen as a credibility problem.

Kitanović and Krstić (2010) state that the inflation-measuring statistics are conceptualised in the way that they almost always overestimate the developmental scores, and they rarely, or never, show them in real terms. The total growth of prices would be a signal for overheating of the entire economy, with the resulting growth of interest rates and decreased production investments and employment.

Additionally, Serbian financial market has been characterized by insufficient development, low trading volume stocks, “shallowness”, lack of continuous stock trading, low liquidity, lack of market transparency, high transaction costs, incomplete application of international accounting standards and low level of corporate governance.

According to Štajner, Ivanišević, Katić, and Penezić (2015), numerous factors, such as macroeconomic, political, socio-economic, demographic, technological, and environmental, affect the performance of companies. The service sector has gained a dominant role in Serbia’s economic development after 2001.

Moreover, the authors conclude that, whether it is about an expansion of the existing organisation, an establishment of a new one or a particular investment project, general regional information provides a quality basis and starting point for further analyses that should result in quality investment decisions.

The data have proven that higher inflation is mainly related to developing countries, and the volatility of stocks is higher in those countries than in developed markets (Barakat, Elgazzar and Hanafy, 2016). Since the 1930s, the research has shown that almost every country has had the worst stock 45

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

returns during high inflation periods. Real stock returns have been equal to the actual returns minus inflation. The evaluation results of the stock market index of S&P 500 over a ten-year period show that the highest real returns occurred when the inflation rate was between 2 and 3%. Inflation greater than or equal to 2 to 3% usually means that the U.S. macroeconomic environment has got problems that have varying impacts on the stock prices. More important than the actual returns is the volatility of returns caused by inflation and knowing how to invest in such an environment because of the financial risk. Financial market research has shown that at a high inflation time, the stock prices rise, while in the deflation period with low inflation rates, returns are higher; however, it may not be necessarily true for all financial markets.

Tripathi and Kumar (2014) studied the relationship between inflation and stock returns in BRICS countries1 using panel analysis in the period from 2000 to 2013. The results show that there existed a significant positive correlation between inflation rates and stock returns in India and China and a significant negative correlation in Russia and Brazil. The panel results for individual markets did not reveal any long-term equilibrium in Russia, India and South Africa, while in case of Brazil and China, the results showed an established long-term equilibrium of the inflation rate and stock returns. Further integration tests on the panel found no long-term correlations between stock returns and inflation rates.

Engle, Ghysels and Sohn (2013) used GARCH models to assess the volatility of macroeconomic variables on the stock market. They got encouraging results in terms of long-term forecasting. The models included a long-term component forecast driven by inflation and industrial production growth. They concluded that the ideas of the short- and long-term component models guided by economic sources could be potentially extended to multivariate settings.

According to Duarte (2013), a stock’s inflation risk can be written as the product of the market price of inflation risk and the stock’s quantity of risk. In scientific research, it is proven that stocks whose returns covary negatively with inflation shocks have unconditionally higher returns. This implies that the average market price of the risk of inflationary shocks is negative: periods with positive inflationary shocks usually tend to be poor states of nature, and investors are willing to pay insurance in the form of lower mean returns when holding an inflation-mimicking portfolio. Furthermore, the paper concludes that the negative inflationary risk arises because high inflation today predicts low growth in future real consumption.

The authors Ang, Briere and Signori (2012) studied the inflation hedging ability of individual stocks in their research work. While the poor inflation hedging ability of the aggregate stock market has long been documented, there is considerable heterogeneity in how individual stock returns covary with inflation. Stocks with good inflation-hedging abilities have had higher returns, on average, than stocks with low inflation betas and tend to be drawn from the energy and technology sectors.

In their scientific research, Campbell and Vuolteenaho (2004) proved that high inflation was positively correlated with rationally expected long-term real dividend growth. It was further considered that inflation was a subjective risk premium. However, inflation is largely related to the wrong price, supporting Modigliani-Cohn’s (1979) thesis that investors have subjective growth forecasts extrapolating previous nominal growth rates without taking into account the impact of time-varying inflation.

1

46

BRICS is an acronym for the association of the five major emerging economies: Brazil, Russia, India, China and South Africa. BRICS members are all developing countries, but they are distinguished by large, rapidly growing economies and significant impacts on regional and global affairs. Since 2014, five BRICS countries represent nearly 3 billion people, which account for 40% of the world’s population and have a combined nominal GDP of $ 16,039 trillion, i.e. 20% of the world GDP.

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

Since the aforementioned studies on the correlation between the inflation rate and stock returns clearly prove that the inflation rate shows either a positive or a negative effect depending on certain factors (e.g. the development of the financial market), the authors were motivated to examine the nature of these connections in the case of real sector companies in AP Vojvodina.

The review has shown diverse results of the inflation impact on stock markets and company return rates. The paper will present whether the impact of inflation is significant and to what extent, in relation to the returns of real sector companies in AP Vojvodina.

METHODOLOGY The research sample included daily values, as well as the calculated return rates of the real sector companies of AP Vojvodina, listed in the Belexline Index: NIS a.d Novi Sad, Sojaprotein a.d Bečej, Vital a.d Vrbas, Veterinarski zavod a.d Subotica and Budućnost a.d Bačka Palanka. The period covered by the survey is from January 1, 2006, to December 31, 2016. The inflation rate impact was also observed in the ten-year period from 2006-2016.

According to Brooks (2008), the rate of returns can be shown as follows:

rt = (InPt / Pt −1 ) *100

(1)

where rt is the logarithmic return rate of the observed shares at time t, while Pt and Pt and Pt-1 are the empirical values of the shares of the observed series in the period t and in the previous period, i.e. in the period of the first delay.

In this paper, an appropriate methodology for modelling volatility and testing of research hypotheses has been used. The application of the GARCH (Generalized Autoregressive Conditional Heteroskedasticity) model is used to confirm the basic H0 and the auxiliary H1 hypotheses in the research; the most favourable GARCH model, showing the significance of the inflation rate impact, was selected for each company share and each observation period.

The GARCH model describes processes in which volatility changes are presented in the following way (Brooks, 2008). In the work on the sample, we use the GARCH (1,1) model for the time series Yt yt = ε t = εt

(2)

IID htηt , ηt  → N (0,1)

(3)

2 h= a0 + ∑ i 1 a= t = i ε t −i + ∑ j 1 b j ht − j

(4)

ht = C0 + C1ht −1 + C2ε 2t −1

(5)

Yt = α 0 + α1Yt −1 +  + α qYt −q + ε t

(6)

q

p

where ht is a conditional variance, i.e. deviation from εt according to the information available in a time t. The GARCH(1.1) model connects a conditional variance ht with the past squared errors and 47

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

past conditional variances. The basic version of the GARCH (1.1) model includes the inflation rate to measure the impact, so the GARCH (1.1) model is now as follows:

Yt = C0 + C1Yt −1 +  + CqYt −q + STINFL . + ε t

(7)

In order to select the best variations of the GARCH model, the following types of GARCH models have been used: log = (ht ) = a0 + ∑ i 1 a= i g (ηt −1 ) + ∑ i 1 bi log (ht −i ) q

p

(8)

the EGARCH model in the form:

θηt + γ [|ηt | −E|ηt |] which are the pondered values of innovation in a where ε t = htηt and g (ηt ) = model with the asymmetric effect between the positive and negative returns of the financial asset, while θ and γ are constant. The basic version of the EGARCH model includes the inflation rate to measure the impact, so the EGARCH model is now as follows: log ht 2 = C + β ln (ht −12 ) + α (ε t −1 ) + C1STINFLAC .

(9)

and the TARCH model that has the following form: 2 ht= = w + ∑ i 1α ε 2 + ∑ j 1 β= ht − j 2 + ∑ i 1γ i I t −iε t −i 2 = i t p

q

p

j

(10)

1 ako je ε t −1 < 0 where is I t −1 =  0 ako je ε t −1 ≥ 0 where the function indicator is It-i , while α and β represent non-negative parameters that satisfy the condition α + β < 1. Also, in the TGARCH model, the conditional volatility ht 2 is positive if α + γ ≥ 1 , γ while the process is stationary in a covariance if and only if  α + 2  + β < 1 . The parameter γ measures the asymmetric or leverage effect in the sense that the artificial variable takes the value 1 if the residuals are negative or the value 0 if the residuals are non-negative. The basic version of the TGARCH model includes the inflation rate to measure the impact, so the TGARCH model is now as follows:

ht 2 = C0 + (α + γ I t −1 ) ε 2t −1 + β ht −12 + C1STINFLAC .

(11)

where α and β represent non-negative parameters satisfying the condition α + β <1. Also, in the custom TGARCH model, the conditional volatility ht 2 is positive if α + γ ≥ 1 , while the process is stationary in covariance if and only if α +  γ  + β < 1 . The parameter γ measures the asymmetric or leverage effect 2 in the sense that the artificial variable takes the value 1 if the residuals are negative, or the value is 0 if the residuals are non-negative.

48

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

All models in this paper were estimated using EViews, by the Marquardt algorithm optimization and Bollerslev and Wooldrige method for standard errors estimates. GARCH model parameters are estimated using the quasi-maximum likelihood – QML. In the study, the selection of the adequate model was based on AIC (Akaike Information Criterion), SICS (Schwarz Information Criterion) and HQC (Hannan-Quinn) information criteria that were used to choose the most favourable models and confirming the research hypotheses. According to Gujarati (2010), the used criteria are calculated as follows:

2k AIC = l n (σ2 )+ T

(12)

k SIC = l n (σ2 )+ l n T T

(13)

2k HQC = l n (σ2 )+ l n (l n (T)) T

(14)

where σ2 is the residual variance, which is equivalent to the residual sum of the squares divided by the number of observations in the series, k=p+q+1 is the total number of estimated parameters, and T is the sample size. From the above-mentioned criteria, the strictest penalties are imposed by SIC criterion, AIC has the mildest penalties, while HQC is somewhere in-between. Although according to Brooks (2008), the best criterion cannot be claimed, the most favourable models were selected according to the lowest SIC information criterion.

In order to quantify the impact of the inflation rate, GARCH 1.1, TGARCH, and EGARCH models including the inflation will be applied for all the observed returns of companies from AP Vojvodina during the monitoring period. Then, using the AIC, SIC and HQC information criteria, the most favourable GARCH model will be selected between GARCH 1.1, TGARCH and EGARCH for the observation period, and separately for each AP Vojvodina company whose returns were observed. Finally, the comparison of the results of the selected most favourable GARCH models will be presented in order to quantify the impact of inflation rate and test the hypotheses.

The link to the data used in the research for stock prices time series and inflation rate is: https:// data.mendeley.com/datasets/9gw8jwm6kd/1

RESEARCH RESULTS AND DISCUSSION This part of the paper will present the research results of the applied GARCH model for analyzing, testing and quantification of the impact of the inflation rate on the returns from the investment activities in the real sector companies of AP Vojvodina. In the following section, the best model for each of the observed companies in the monitored period will be first selected, and then the residual motion will be graphically presented. The following tables refer to the presentation of the best models and the display of the normal distribution of the sample.

49

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

Figure 1. Inflation rate movement in the period from 2006-2016 Source: The authors’ work based on data from the NBS, 20172.

Figure 1 shows the inflation rates in Serbia in the observed ten-year period from 2006 to 2016. As a factor of financial risk, the inflation rate showed cyclical movement, due to the structural problems of the Serbian economy inherited from the previous period. During the observed period, the inflation rate had been steadily growing and declining, until the middle of 2013, when the inflation rate stabilized at the level of one to three per cent on average.

Figure 2. Stock prices data of the observed company in the period from 2006-2016 Source: Calculation by the authors

The prices of shares were observed at the daily level (daily rates of return) while the inflation rate was observed on a monthly basis. The relationship between the inflation rate and the daily rates of return was observed over a ten-year period, which implied medium and long-term effects.

Table 1 presents a comparative overview of the GARCH model results for the impact of the inflation rate on the share returns of the observed companies of AP Vojvodina: NIS a.d. Novi Sad, Sojaprotein a.d. Becej, Vital a.d. Vrbas, Veterinarians Zavod a.d. Subotica and Budućnost a.d. Bačka Palanka, and in accordance with the above criteria, the optimal chosen models3.

2 Source: Central Bank of Serbia, http://www.nbs.rs/ 3 It was not possible to establish a model for measuring the inflation rate impact on the returns of Buducnost

a.d. Bačka Palanka (BDBP) due to the statistical significance of the probability density function (Log likelihood) estimated in the observed data values (for log likelihood = -1,996,392 => F-test=0.028 that is <0.05).

50

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

Kompanije APV

GARCH 1.1 AIC

SIC

TGARCH HQC

AIC

SIC

EGARCH HQC

AIC

SIC

HQC

NIIS

-4.863995 -4.754255 -4.819403 -4.849108 -4.717420 -4.795597 -4.857902 -4.857902 -4.857902

SJPT

2.055199

2.055199

2.055199

2.059074

2.059074

2.059074

2.048371

2.048371

2.048371

VITL

2.738930

2.738930

2.738930

2.886774

2.886774

2.886774

2.726628

2.726628

2.726628

VZAS

-2.802415 -2.802415 -2.802415 -2.872138 -2.872138 -2.872138 -2.852639 -2.852639 -2.852639

Table 1. Representative criteria for selecting optimal models Source: Calculation by the authors

Based on the AIC, SIC and HQC criteria in Table 4.1, the most favourable GARCH models in the observation period were selected. The lower the AIC, SIC and HQC criteria, the more favourable the model. The results of the criteria applied for the optimal choice of GARCH models showed that EGARCH model was most favourable for measuring the inflation impact on the returns of the companies NIS, SJP and VITL, while TGARCH model was most favourable for VZAS.

NIIS

SJPT

VITL

VZAS

Graph 1. Movement of the residual return rates of real sector companies in APV for the observed period from 2006 to 2016 Source: Calculation by the authors

Graph 1 shows the movement of the residual return rates of the observed companies in AP Vojvodina: NIS a.d Novi Sad, Sojaprotein a.d Bečej, Vital a.d Vrbas, Veterinarski zavod a.d Subotica in the observed period. The entire observation period clearly shows the extremely large fluctuations in the residual returns in the period marked by the global financial crisis. In the lower part of the graph, the 51

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

blue frame lines mark the average residual values of the daily return rates of the observed companies.

Showing the movement of the residual returns, the graphs SJPT, VITL and VZAS represent fewer oscillations that indicate reduced trade volume. In the final observation years, small residual oscillations were also noticed, which indicates a slow recovery of the Serbian financial market from the consequences of the global financial and economic crisis.

Y - NIIS

Y - SJPT

Y - VITL

Y - VZAS

Variance Equation EGARCH

Variance Equation EGARCH

Variance Equation EGARCH

Variance Equation TGARCH

C(1)

-0.963087

C(1)

-1.053873

C(1)

-0.412522

C

0.003220

C(2)

0.438051

C(2)

0.551809

C(2)

-0.272910

RESID(-1)^2

0.047202

C(3)

0.064212

C(3)

-0.346755

C(3)

-0.379705

RESID(-1)^2*<0

-0.111806

C(4)

0.931301

C(4)

0.328761

C(4)

0.668852

GARCH(-1)

0.513685

ST_INFL

0.013876

ST_INFL

0.019167

ST_INFL

0.051056

ST_INFL

-0.000183

where: C and C (1) are constants, RESID (-1)^2 is the square of the standardised residuals, i.e. the coefficient of the 1st order delays of the asymmetric adapted TGARCH model, GARCH (-1) - the GARCH effect of the univariate custom TGARCH model, RESID (-1)^2*<0 – the asymmetric or leverage effect, C(2) – the ARCH effect, C(3) – the leverage effect, C(4) – the GARCH effect, ST_INFL - represents the independent variable of inflation rate models.

Table 2. The estimated parameters of the optimal GARCH model for measuring the inflation rate impact on the real sector shares of AP Vojvodina Source: Calculation by the authors

Table 2 presents the most favourable evaluated GARCH models for measuring the inflation rate for the real sector companies of AP Vojvodina: NIS a.d Novi Sad, Sojaprotein a.d. Bečej, Vital a.d. Vrbas, Veterinarski zavod a.d Subotica in the observed period from 2006-2016.

The evaluated EGARCH model shows the positive impact of the inflation rate (0.013876) on the daily returns of NIS shares. The obtained results of the EGARCH model show that if the inflation rate was changed by 1 unit, the daily return rate Y of NIS a.d. Novi Sad shares would be 0.484353. This would further mean that if 1,000,000 funding units were invested in the shares of NIS a.d, the daily return would be 4,843.53 units.

The evaluated EGARCH model shows the positive impact of the inflation rate (0.019167) on the daily returns of Sojaprotein shares. The obtained results of the EGARCH model show that if the inflation rate was changed by 1 unit, the daily return rate Y of Sojaprotein a.d Bečej shares would be -0.50089.

This would further mean that if 1,000,000 funding units were invested in the shares of Sojaprotein a.d, the daily return would be -5,008.91 units.

The evaluated EGARCH model shows the positive impact of the inflation rate (0.051056) on the daily returns of Vital shares. The obtained results of the EGARCH model show that if the inflation rate was changed by 1 unit, the daily return rate Y of Vital a.d Vrbas shares would be -0.34523. This would further mean that if 1,000,000 funding units were invested in the shares of Vital a.d, the daily return would be -3,452.29 units.

The estimated TGARCH model shows the negative impact of the inflation rate (-0.000183) on the daily returns of Veterinarski zavod shares. The obtained results of the TGARCH model show that if the 52

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

inflation rate was changed by 1 unit, the daily return rate Y of Veterinarski zavod a.d Subotica would be 0,.452118. This would further mean that if 1,000,000 funding units were invested in the shares of the Veterinary Institute a.d, the daily return would be 4,521.18 units.

The research did not deal with marginal changes of the inflation rate (an increase from 1% to 2% might have a positive impact on stocks, but an increase from 9% to 13% may trigger a decline) and the impact on the market price of the shares. The paper tested only influence the rate of inflation as a factor influencing the daily rates of return. Future research could be focused on testing the marginal changes in inflation rates and impact on stock prices.

NIS a.d Novi Sad

Sojaprotein a.d Bečej

Vital a.d Vrbas

Veter. zavod a.d Subotica

Series: Standardized Residuals

Series: Standardized Residuals

Series: Standardized Residuals

Series: Standardized Residuals

Sample of data (adjusted) Sample of data (adjusted) Sample of data (adjusted) 21603 22767 22699

Sample of data (adjusted) 12771

Included Observations of standardized residuals after adjustments 1602

Included Observations of standardized residuals after adjustments 2766

Included Observations of standardized residuals after adjustments 2698

Included Observations of standardized residuals after adjustments 2771

Mean

-0.028170

Mean

0.024182

Mean

-0.093564

Mean

-0.049842

Median

-0.168407

Median

0.038538

Median

-1.62E-05

Median

0.032568

Maximum

3.956099

Maximum

3.701462

Maximum

5.301347

Maximum

3.968955

Minimum

-2.380546

Minimum

-3.347844

Minimum

-6.517267

Minimum

-4.350492

Std. Dev.

1.007914

Std. Dev.

1.003339

Std. Dev.

1.050697

Std. Dev.

1.017090

Skewness

0.663569

Skewness

0.181282

Skewness

-1.737992

Skewness

0.038116

Kurtosis

4.933591

Kurtosis

6.054484

Kurtosis

21.70036

Kurtosis

10.53410

Jarque-Bera

30.02123

Jarque-Bera

51.64307

Jarque-Bera

1974.748

Jarque-Bera

312.2269

Probability

0.000000

Probability

0.000000

Probability

0.000000

Probability

0.000000

Table 3. Distribution of standardized residuals daily returns of real sector companies of AP Vojvodina for the observed period from 2006 to 2016 Source: Calculation by the authors

Table 3 shows the value of basic statistical information for returns of investment in the real sector companies of AP Vojvodina: NIS a.d Novi Sad, Sojaprotein a.d Bečej, Vital a.d Vrbas, Veterinarski zavod a.d Subotica in the observed period from 2006-2016. First, the average daily returns had a negative sign for NIS (-0.03), Vital (-0.09) and Veterinary Institute (-0.05), while Sojaprotein shares had an average positive daily return (0.02). The table also shows the maximum and minimum return values of the real sector companies of AP Vojvodina: NIS a.d Novi Sad, Sojaprotein a.d Bečej, Vital a.d Vrbas, Veterinarski zavod a.d Subotica in the observed period, whose shares are quoted at the Belgrade Stock Exchange, Belexline. Furthermore, the amount of standard deviation shows the potential risk of investing in the observed shares of the real sector companies of AP Vojvodina. The standard deviation for NIS was (1.01), for Sojaprotein (1.00), for Veterinary Institute (1.02), while the largest deviation from 53

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

the average daily returns was recorded for Vital shares (1.05). The obtained results of the standard deviation in the observed Serbian financial market can be interpreted as a lack of larger trading activities within the Belexline index. Also, the asymmetry values of skewness and the values of the flattening of distribution (kurtosis - tells about the thickness of distribution tails, i.e. the possibilities for extreme events, the extreme movements in returns) of the observed real-sector companies of the APV speak of an abnormality in the distribution of the return. The values of the asymmetry of the distribution function in the observed period for NIS (0.66), SJPT (0.18) and VZAS (0.04) were slightly positive, which means that the right (positive) tail of the daily return distribution was longer with more positive than negative movements, which was not the case for VITL shares (-1.74). The distribution flattening was in values above 3 for NIS (4.93), SJPT (6.05), VITL (21.70) and VZAS (10.53) and it shows that the probability of extreme return movements of the observed shares was high.

CONCLUSIONS The research results undoubtedly point to the importance of the research subject through the prism of validation, quantification and optimization of the inflation impact on investment activities in modern business conditions. The research tests the place, role and significance of the inflation rate on the assessment of the daily returns of real sector companies in AP Vojvodina: NIS a.d Novi Sad, Sojaprotein a.d Bečej, Vital a.d Vrbas, and Veterinarski zavod a.d Subotica. As researchers, the authors emphasized the importance of analysis and optimisation of the model performance in investing activities in the real sector of AP Vojvodina in the observed financial market, whose main features are instability and low efficiency.

The basic H0 hypothesis was set up, assuming that the application of the GARCH model can successfully test the inflation impact on the share returns of real sector companies in AP Vojvodina. The research and results, obtained by using the GARCH model, confirmed the basic hypothesis. On the given sample, the daily returns trends of the real sector companies in AP Vojvodina and the financial risk factor - the inflation rate in the observed period were directly correlated. The exception was the return of the shares of Budućnost a.d Bačka Palanka, where it was not possible to create a model because of the statistical significance of the probability density function (Log likelihood) estimated in the observed data values. In practice, this means that the share trade of Budućnost a.d recorded an extremely small volume in the observed period from 2006 to 2016, which resulted in equal (same) prices over a longer period, so the daily returns were equal to zero. Due to the high illiquidity and low volume of trading, the shares of Budućnost a.d. Bačka Palanka, the authors were not able to establish a model, i.e. to measure the inflation rate impact on returns.

Assuming that the inflation rate positively influenced the returns of the real sector companies of the AP Vojvodina in the observed period, the H1 hypothesis is partially confirmed. The research results show the positive impact of the inflation rate variable on the returns of the companies NIS a.d Novi Sad (0.013876), Sojaprotein a.d Bečej (0.019167) and Vital a.d Vrbas (0.051056), and a negative impact on the returns of Veterinarski zavod a.d Subotica (-0.000183). Company-specific factors like price elasticity of demand for the company products, financial leverage, currency exposure, etc. may affect the nature of relationship between inflation and stock prices, and this may help explain the heterogeneity observed in the results (the case of VZAS where the impact of inflation was different) and this should certainly be taken into account in the further research.

54

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

It can be concluded that the common feature of all observed shares of real sector companies in AP Vojvodina in the period from 2006 to 2016 is that there was no distribution normality and that different sample asymmetries and distortions were manifested in different periods.

The aforementioned facts point to the importance of testing the differences between various models of inflation impact evaluation and their different “behaviour” in certain market circumstances in the observed period. It was necessary to perform a holistic, comprehensive and systematic approach to the analysis, quantification and validation of investment expectations for the observed financial market of Serbia and the real sector companies of AP Vojvodina whose shares are listed within the Belexline index. The special quality of the research results stems from the fact that the study focused on the financial market of the developing country (Republic of Serbia), with a relatively small number of studies published with this topic.

The scientific contribution of work is reflected in the quality and significance of the research results and the possibilities of efficient application of methods that quantify the significance of the inflation rate influence on the level of daily returns of the real sector companies of AP Vojvodina. The practical contribution of the work is in the expanded possibilities of efficient application of the assessment of the inflation rate impact on the daily rates of return in everyday business decision-making on investment activities.

Despite the above contributions, on the one hand, the authors dealt with the problems and challenges that arose from the specificities of returns listed under the stock exchange index Belexline (developing financial market) and, on the other, the need for adjusting the tested GARCH models to the specificities of these markets. The largest research challenge was to apply econometric models and adjust the inflation rate impact on the observed returns of the shares of real sector companies in AP Vojvodina, thus enabling them to successfully apply and obtain results that are based on science and practice. The research did not deal with marginal changes of the inflation rate and the impact on the market price of the shares. The paper tested the inflation only as a factor influencing the daily rates of return. The directions for future research could be focused on testing the marginal changes of inflation rates and impact on stock prices.

Future research in this area should focus on expanding research focus on other company shares within the financial markets of developing countries, as well as extending the impact of financial factors. This will increase the flexibility of the tested econometric (GARCH) models in the light of the maximization of the effects from the investment activities. In this sense, the focus of future research will be extended and the applied methodology significantly modified to a higher level of flexibility and adaptability, taking into account the dynamics of changes in the financial markets caused by globalized trends.

ACKNOWLEDGEMENT The authors acknowledge the financial support of the Provincial Secretariat for Higher Education and Scientific Research, within the Project No. 142-451-3554/2017-01/02.

55

EJAE 2018  15 (2)  43-57

ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

REFERENCES Adamović, S. (2008). Ekonomski dijalozi. Beograd: Službeni glasnik. In Serbian.

Ang, A., Brière, M., & Signori, O. (2012). Inflation and individual equities. Financial Analysts Journal, 68(4), 36-55.

Barakat, M.R., Elgazzar, S.H., & Hanafy, K.M. (2016). Impact of macroeconomic variables on stock markets: Evidence from emerging markets. International Journal of Economics and Finance, 8(1), 195. DOI:10.5539/ ijef.v8n1p195 Belgrade Stock Exchange. (2017). Retrieved from http://www.belex.rs/ Brooks, C. (2008). Introductory Econometrics for Finance, 2nd ed. New York: Cambridge University Press.

Campbell, J.Y., & Vuolteenaho, T. (2004). Inflation illusion and stock prices. NBER Working Paper No. w10263.

Retrieved from https://ssrn.com/abstract=1645104 Central Bank of Serbia. (2017). Retrieved from http://www.nbs.rs/ Damnjanović, R. (2017). Matematički model za određivanje cene akcija – uključivanje uticaja inflacije. Oditor časopis za Menadžment, finansije i pravo, 3(2), 107-124.

Duarte, F.M. (2013). Inflation risk and the cross section of stock returns. FRB of New York Staff Report No. 621.

DOI: 10.2139/ssrn.2273666 Engle, R.F., Ghysels, E., & Sohn, B. (2013). Stock market volatility and macroeconomic fundamentals. Review of Economics and Statistics, 95(3), 776-797.

Fama, E.F. (1981). Stock returns, real activity, inflation, and money. The American Economic Review, 71(4), 545-565.

Fisher, I. (1930). The theory of interest. New York: Macmillan Company.

Gujarati, N.D, & Poreter, D. (2010). Basic Econometrics. New York: McGraw-Hill.

Kitanović, D., & Krstić, M. (2010). Uticaj deviznog kursa na privredni rast Srbije. Ekonomske teme, 48(1), 13-27.

In Serbian.

Modigliani, F., & Cohn, R.A. (1979). Inflation, rational valuation and the market. Financial Analysts Journal, 35(2), 24-44. Štajner, S., Ivanišević, A., Katić, I., & Penezić, N. (2015). Comparative perspectives on the development of economic power in Serbia and countries in the region. Poslovna ekonomija, 9(1), 119-140.

Tripathi, V., & Kumar, A. (2014). Relationship between Inflation and stock returns–evidence from BRICS markets using Panel Co integration Test.

56

EJAE 2018  15 (2)  43-57 ANĐELIĆ, G., PENEZIĆ, N., TOT, V., MILOŠEVIĆ, M.  INFLATION RATE IMPACT ON THE SHARE RETURNS OF REAL SECTOR COMPANIES IN AP VOJVODINA

UTICAJ STOPE INFLACIJE NA PRINOSE OD AKCIJA U KOMPANIJAMA REALNOG SEKTORA U AP VOJVODINI

Rezime: Uzimajući u obzir aktuelne trendove na domaćem finansijskom tržištu, predmet istraživanja u radu je analiziranje, testiranje i kvantifikacija uticaja stope inflacije na dnevne stope prinosa akcija kompanija realnog sektora AP Vojvodine. Cilj istraživanja jesu konkretna, u praksi testirana i kvantifikovana saznanja o mogućnostima i efikasnosti primene GARCH modela u funkciji kvantifikacije uticaja stope inflacije na stope prinosa akcija posmatranih kompanija. Vremenski period obuhvaćen istraživanjem je od 2006. do 2016-te godine i uključuje kompanije realnog sektora AP Vojvodine čije se akcije kotiraju u okviru berzanskog indeksa BELEXline. Rezultati istraživanja pokazuju tačnu korelacionu vezu između dnevnih stopa prinosa posmatranih kompanija i faktora finansijskog rizika - stope inflacije. Rezultati takođe pokazuju pozitivan uticaj varijable stope inflacije na prinose akcija kompanija NIS a.d (0.013876), Sojaprotein ad (0.019167) i Vital a.d (0.051056), a negativan uticaj na prinose akcija kompanije Veterinarski zavod a.d (-0.000183).

Rezultati istraživanja potvrđuju ulogu i značaj primene ekonometrijskih modela u svetlu kvantifikacije faktora finansijskih rizika na prinose od aktivnosti investiranja u kompanije realnog sektora AP Vojvodine.

Ključne reči: stopa inflacije, prinosi akcija, GARCH modeli, realni sektor, rizik, investiranje.

57

EJAE 2018, 15(2): 58-73 ISSN 2406-2588 UDK: 005.342:334.713(549.3) 330.34 DOI: 10.5937/EJAE15-17015 Original paper/Originalni nauÄ?ni rad

CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY Md. Shahidul Islam1,2,*, Md. Faruk Hossain1 Bangladesh Knitwear Manufacturers and Exporters Association (BKMEA), Dhaka, Bangladesh.

2 Bangladesh Institute of Social Research (BISR) Trust, Dhaka, Bangladesh.

1

Abstract: Small and Medium Enterprises (SMEs) are the significant contributor to the Bangladesh economy as they have the competency to create higher employment, improve local technology, and create ground for the future sustainable industrialization and corporate sectors. The aim of this study is to identify major constraints faced by the SMEs in Bangladesh.

Article info: Received: May 28, 2018 Correction: June 24, 2018 Accepted: June 27, 2018

A cross-sectional study was conducted in Narsingdi Municipality, Bangladesh and social survey method was applied to collect the data.

Two-stage cluster sampling method was utilized to select respondents and data were collected through face-to-face interview. Principal Component Analysis (PCA) was used to determine significant constraints to SMEs. The study found that poor infrastructure and electricity supply are the main constraints to the development of SMEs. Limited access to credit, lack of proper business knowledge and plan, high domestic market competition, lack of skilled manpower and technology, and high cost of raw materials and equipment are the major constraints next to main constraint.

Keywords: small and medium-sized enterprises (SMEs), constraints, principal component analysis (PCA), Bangladesh.

INTRODUCTION There is growing evidence that nurturing Small and Medium Enterprises (SMEs) has been playing a vital role in the economic development, especially in developing countries. SMEs have been considered as a powerful tool for boosting up the economic growth by creating new employment opportunities, improving local technologies, developing entrepreneurships, integrating with large-scale industries and international trade (ACMA, 2015). These have been considered as the main driver of businesses in developing countries, created ground for the future sustainable industrialization, developed 58

* E-mail: shahidulsoc@gmail.com

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

entrepreneurial skills and fostered private ownerships (Moktan, 2007). Development of entrepreneurship and human skills can be considered as two driving forces of sustainable economic and social development (Tambunan, 2009) where establishing SMEs is the main strategy to develop entrepreneurship and human skills. Evidence supports that Japan has achieved its financial development based on SMEs (Chowdhury, 2003).

Despite having sufficient manpower Bangladesh cannot properly utilize it. Through SMEs, it is possible to create more employment opportunities with small capital comparing to large industries where its manpower can be utilized for productive outcome. SMEs can be developed in backward industry that will generate job opportunities for the semi-skilled and unskilled labour (Chowdhury, 2013).

SMEs have the ability to survive in poor economic condition because they are flexible in nature and can adopt with the changing market trend compared to larger industries (Zaman and Islam, 2011).

Evidence supports that SMEs are more innovative than larger farms, facilitate the market with new innovation, and lead to technological changes (Acs et al. 1999). In addition, SMEs require relatively small capital, less infrastructural support and consumption of utilities, and less time to start up business and have quick returns (Chowdhury, 2013). Moreover, SMEs do not necessitate highly sophisticated expertise and technology. They can sustain using local resources and can be operated by trained manpower rather than sophisticated technology. SMEs are suitable for domestic market that can be run by utilizing local limited resources.

The productivity of SMEs largely depends on its ability to innovate, adopt new technologies and apply them to local conditions. The competitive advantage and ability of SMEs to trade in a global environment largely depend on the innovation (Berry et al. 2002). SMEs provide the nursery and pre-background of innovation (Wolf, 2006). Evidence supports that many scientific innovations were originated by small organizations like as photocopier, helicopter, insulin, zipper, stenographic pen, etc. (Longenecker et al. 1997).

SMEs can be run in local areas where the large enterprises cannot. The evidence also supports that most of the large enterprises, even the international multi-million dollar enterprises, have their origin in SMEs (National Credit Regulator, 2011). SMEs have great impact on rural economy because of their widely spreading presence across the rural areas and also have the capacity to employ large amount of labour force (Tambunan, 2011).

From the economic perspective, enterprises are considered for both suppliers and consumers. So, they can play an important role in the market if they have enough power to control those capabilities.

SMEs are the main consumer for the raw materials produced domestically or internationally. Any increase in their demand for consumer goods will stimulate the activity of their suppliers, thus this activity is stimulated by the demand of their clients (Berry et al. 2002). If SMEs create demands in the market, they also play the role of supplier as well as consumer. So, generating effective demand is an important role of SMEs for establishing new SMEs and new income generation activity. SMEs contribute to the national economies by producing manufacturing goods of value or by fulfilling the demand of consumers or other enterprises (Abor and Quartey, 2010).

Although SMEs are the driving force of the economy, the mounting of a new SME development is not innate proceeding. Even the development of SMEs is not easy; it is often hampered by different external and internal constraints. These constraints may differ from area to area, rural to urban, among sectors and, among entrepreneurs within a sector (Tambunan, 2011). In Bangladesh, the establishment of SMEs is facing various problems related to raw materials, power, land, marketing, transport, technical facilities and finance. Poor capital structure and other constraints hamper the growth and function of SMEs 59

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

(Quader and Abdulla, 2009). These constraints are very complex and interrelated also. There are a small number of studies in Bangladesh that have investigated constraints to SMEs development. This study is designed to determine the most relevant constraints hampering the SMEs development in Bangladesh.

Definition of the SMEs Definition: There are so many definitions of SMEs that differ from country to country, depending on their size, scope and sector. Bangladesh Bank (2011) has defined the SMEs as follows:

Sectors

Fixed Asset other than Land and Building (BDT)*

Employed Manpower (not above)

Service

5.00 lakh – 1.00 crore

10-25

Business

5.00 lakh – 1.00 crore

10-25

Industry

50.00 lakh– 10.00 crore

25-99

Table 1. Small Enterprise Source: Bangladesh Bank (2011),*BDT=Bangladesh Taka, 1 lakh = 100 thousands

In service unit, an enterprise is treated as small if in the current market prices, the replacement cost of its fixed asset (excluding land and buildings) is BDT 5 lakh to 1 crore, with employment size ranging from 10 to 25. In Business unit, an enterprise is treated as small if in the current market prices, the replacement cost of its fixed asset (excluding land and buildings) is BDT 5 lakh to 1 crore, with employment size up to number of employee ranges from 10 to 25. In Industrial unit, an enterprise is treated as small if in the current market prices, the replacement cost of its fixed asset (excluding land and buildings) is BDT 50 lakh to 10 crores with employment size ranging from 25 to 99.

Sectors

Fixed Asset other than Land and Building (BDT)

Employed Manpower (not above)

Service

1.00 crore- 15.00 crore

50-100

Business

1.00 crore- 15.00 crore

50-100

Industry

10.00 crore – 30.00 crore

100-250

Table 2. Medium Enterprise Source: Bangladesh Bank (2011)

In service units, an enterprise is treated as medium if in the current market prices, the replacement cost of its fixed asset (excluding land and buildings) is BDT 1 crore to Tk. 15 crores with employment size up to number of employee ranges from 50 to 100. In Business unit, an enterprise is treated as medium if in the current market prices; the replacement cost of its fixed asset (excluding land and buildings) is BDT 1 core to Tk. 15 crore with employment size up to number of employee ranges from 50 to 100.

In Industrial unit, an enterprise is treated as medium if in the current market prices; the replacement cost of its fixed asset (excluding land and buildings) is BDT 10 crore to Tk. 30 crore with employment size up to number of employee ranges from 100 to 250.

60

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Contribution of SMEs to the economy Activity /sector

20072008

20082009

20092010

20102011

20112012

20122013

20132014

20142015

20152016

20162017

Industry

17.77

17.90

17.94

17.42

18.96

19.54

19.47

20.16

21.01

21.73

Large scale and medium scale

12.63

12.71

12.68

13.20

13.73

14.28

15.95

16.58

17.37

18.02

Small scale

5.14

5.18

5.26

5.22

5.23

5.27

3.51

3.58

3.64

3.71

Table 3. Sectoral share in GDP at constant prices (1995-1996) Sources: Statistical pocketbook Bangladesh, FY2016-17

Data presented in Table 3 reveal that share of industry in the GDP had been increased between FY2007-08 to FY2016-2017. Industrial contribution was 17.77% in FY2007-08 which was increased by 21.73% by FY2016-2017. It is also noticeable that major contribution has come from larger and medium scale industry which was gradually increasing. But the GDP share has been decreasing in small enterprise.

Year

Growth Rate of Small & Cottage Industry

Medium to Large Industry

2001-02

7.69 %

4.6%

2002-03

7.21 %

6.6%

2003-04

7.45 %

6.95%

2004-05

7.93 %

8.30%

2005-06

9.21 %

11.41%

2006-07

9.69 %

10.80%

2007-08

7.1 %

7.38%

2008-09

7.30%

6.54%

2009-10

8.17%

6.27%

2010-11

5.67%

11.11%

2011-12

6.58%

10.76%

2012-13

8.81%

10.65%

2013-14

6.33%

9.32%

2014-15

8.54%

10.70%

2015-16

9.06%

12.26%

2016-17

9.21%

11.32%

Table 4. Yearly Growth Rate of Small & Cottage and Medium to Large Industry Source: Bangladesh Economic Review, FY2016-17

61

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Data presented in Table 4 reveal yearly growth rate of small & cottage and medium to large industry. In 2001-02, the growth rate of small enterprises was 7.69%. In 2006-07, the growth rate of small enterprises had increased to 9.69%. After this year the growth rate of small enterprises had decreased.

In 2013-14, the growth rate of small enterprises was 6.33% and that was the lowest growth rate ever. In 2016-17, the growth rate of small industries was 9.21%. Data presented in Table 4 clearly denote that the growth rate of the small enterprise has increased in recent years.

REVIEW OF LITERATURE Constraints to SMEs development There are various constraints identified to the development and operation of SMEs in Bangladesh.

Quader and Abdulla (2009) identified financial constraints, regulatory constraints, and constraints on physical technical, and marketing input to SMEs’ development in Bangladesh. They found that high lending rate, unavailability of financing, cost of high equipment, collateral need and small domestic market size, lack of technically skilled workers, lack of protective measures and uncertainty are major constraints to the development of SMEs in Bangladesh. They also added that high-interest rate is the major constraint to invest in SMEs of Bangladesh. Moktan (2007) examined major constraints of SMEs in Bhutan. He found that problems of business policies and regulatory, infrastructure and geophysical constraints, and finance are major constraints to SMEs development in this country. He also found that there were different constraints in urban and rural as per size, sector and ownership. Chowdhury (2007) investigated constraints to establish SMEs in Bangladesh. He found that inadequate infrastructure, lack of financial support, and political instability were major constraints. He suggested establishing political stability and rule of laws, enhancing infrastructure facilities and providing financial help to the entrepreneurs for developing SMEs in Bangladesh. Mambula (2002) investigated factors that influence the start-up, growth, and performance of SMEs in Nigeria. He found that lack of finance, poor infrastructure, access to raw materials were main constraints to SMEs’ development in Nigeria. Moreover, lack of qualified and experienced entrepreneurs and untrained management were also constraints to SMEs’ development. Kshetri (2011) examined various factors related to the entrepreneurship development and entrepreneurial performance in India. He found that regulatory framework, market conditions, access to finance, R&D and technology related factors, physical infrastructures, entrepreneurial capabilities and entrepreneurial culture are affecting entrepreneurship performance and small business development.

SMEs and Financing Chowdhury and Ahmed (2011) examined problems of SMEs’ financing in Bangladesh. They explored that non-availability of sufficient credit, complex loan granting procedure, poor infrastructure facilities, troubles of collateral requirements, scarcity of working capital, lack of skilled workforce, poor salary structure, lack of coordination among SMEs related organizations, weak marketing strategies were constraints to SMEs’ development in Bangladesh. They suggested that commercial banks and other financial institutions should make the SMEs’ loan procedure easier. The Bangladesh Government may establish one specialized SME Bank to give collateral free bank loans to SMEs’ entrepreneurs. Quartey (2015) examined the relationship between finance and SMEs in Ghana. He found that access to finance and location in a commercial town were the major determination of SMEs’ growth. On the other hand, 62

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

large initial capital, good exporting records and good profit of the firms were major determinations of access to finance for SMEs’ development. Tambunan (2015) examined the recent development of SMEs in Indonesia with special consideration to identify constraints to SMEs’ development. He found that lack of access to credit is the main constraint to SMEs’ development in this country. Moreover, local entrepreneurs faced problems to explore their resources in market because there was low facility to capture market. They depend on their local network partners to reach the market. Hasan and Jamil (2014) identified that access to finance is the main constraint to SMEs’ development in Bangladesh.

Most of the banks in Bangladesh are less likely to give the loan to the SMEs’ entrepreneur because they consider SME financing risky investment. Moreover, SMEs’ entrepreneurs have insufficient knowledge on the procedure of seeking institutional finance. Abor and Biekpe (2006) found that access to finance is not only the main constraint to SMEs’ development but also hampers the growth and performance of SMEs in Ghana. He found that access to credit is hampered by various factors like collateral requirement, poor knowledge of finance providers, stringent eligibility criteria, lack of knowledge about lending criteria and bureaucracy. Bhaird and Lucey (2010) examined determinations of the capital structure of SMEs in Ireland. He found that age, size, level of intangible activity, ownership structure and the provision of collateral are important determinants of the capital structure in SMEs. Aziz and Siddique (2006) explored that access to finance is the major driver for the development of SMEs in Bangladesh.

They argued that pro-SME policy and friendly regulatory framework is essential for the sustainable SMEs’ development. They found that Bangladesh Government and Bangladesh Bank have taken some policies that help to access the credit for SMEs’ development.

SMEs development and economic growth Peltier and Naidu (2012) investigated the effect of social networks on improving the organizational lifecycle performance in small to medium-sized enterprise (SME) in India. They found that advice from personal networks (family and friends) is helpful to start up a new SME as well as its functions. Hoque (2015) explored the role of information and communication technologies (ICTs) in SMEs’ development in Bangladesh. They found that ICTs facilitate SMEs entrepreneurs to be more qualified in decisionmaking process. Using ICTs in SME has many benefits such as improving and reducing transaction costs, improving resource allocation, and shifting the production function. They found that the consciousness of benefits, support from government and financial support are important determinants of ICTs’ use in SME in rural areas. Lack of electricity is one of the major obstacles to successful growth of SMEs in developing countries. Khandker (2014) identified that electricity supply is the prerequisite for SMEs’ development as well as its success in both Bangladesh and Pakistan. Similarly, access to electricity and finance is not the only obstacle to SME’s growth in South Asian countries, but it also has a negative impact on the growth of sales of the SMEs. He suggested that government should take necessary steps to provide financial support so that they start up new SMEs. Size of firms has also significant effect on the growth of SMEs. Middle sized firm grows faster than the small size farm. Abor and Quartey (2010) examined the contributions of SMEs to economic development and constraints to SMEs development in Ghana and South Africa. They found that 92% business is SMEs that contribute about 70% to Ghana’s GDP. They recognized some factors constraints to SME’s development such as lack of access to appropriate technology; limited access to international markets, the existence of inefficient laws, regulations and rules, poor institutional capacity, poor management skills and proper training, and finance. Beck et al. (2005) explored the relationship between SME’s development and economic growth, poverty reduction in 45 countries across the world. They found that there is a strong positive 63

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

association between SME’s development and economic growth. But their results showed that there is no significant relationship between SME’s development and poverty alleviation. Moreover, SME’s development has no significant effect to reduce inequality.

DATA AND METHOD Study area: Narsingdi Municipality was selected as the study area. It is 50 km north east from the capital of Bangladesh. It is a densely industrial area and famous for textile industry. This area is potential for various types of SMEs like fishery, fish feed, poultry feed, flat construction and plot trading, power loom, operation of construction firm and light engineering workshop (Bangladesh Bank , 2011).

Sampling: The cross-sectional research design was applied to the study because the survey was conducted at one point in time. Two-stage cluster sampling method was applied due to non-availability of sample. Narsingdi Municipality consists of nine wards. A ward is the lowest administrative tier of a city or town in Bangladesh that consists of several villages or mohallas (Islam, 2013). Three wards (wards no 2, wards no 7 and wards no 8) were randomly selected from nine wards (employment between10 and 250). At the second stage, 132 SMEs were selected randomly from these wards where the acceptable margin of error was .05 and t-value for the alpha level of .05 was 1.96.

Data collection: Social survey method was applied to the study and the data was collected through face-to-face interviews with respondents. Both closed-ended and open-ended questions were included in the questionnaire and data was collected through face-to-face interviews. The questionnaire had two parts: first part of the questionnaire contained the socio-economic information of respondents and at second section, questionnaire focused on the constraint to SMEs. There were 21 questions in this section, related to constrain to SMEs and respondents were given five options (strongly agree to strongly disagree) according to Likert scale.

Data analysis The SPSS (Statistical Package for Social Science) 19 program was used to analyze the data. At first, factor analysis by extracting the Principal Components Analysis (PCA) using varimax normalized method was applied to identify the significant constraint to this study. Then, constraints were detected by ranking, according to factor loading, so that constraints could be identified according to their severity.

Principle components analysis There are some rationales to use PCA: (1) it reduces and summarizes the 21 variables in fewer number factors that contains the major information of the original variable.(2) to find the amount of each common factor possessed by each observation(the factor scores). (3) it explores the relationship between observed variable(measurements) and the underlying latent factors. (4) It also calculates the variance of each variable and identifies the structure in correlation between the variables (Moktan, 2003).(5) Finally, to find and determine which of the variables contained in these factors are the most important of all constraints in the SME’s development in Bangladesh.

The principal component variables y1, y2, …, yq are defined to be linear combinations of the original variables x1, x2, …, xq that are uncorrelated and account for maximal proportions of the variation in the original data, i.e., y1 accounts for the maximum amount of the variance among all possible linear 64

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

combinations of x1, …, xq, y2 accounts for the maximum variance subject to being uncorrelated with y1 and so on. Explicitly, the principal component variables are obtained fromx1, …, xq as follows:

Y1 = a11 X 1 + a12 X 2 + ... + a1n X n Y2 = a21 X 1 + a22 X 2 + ... + a2 n X n − − − − − − − − − − − − − − − −

Yn = an1 X 1 + an 2 X 2 + ... + ann X n Where, the coefficients aij (i = 1, …, q, j = 1, …, q) are chosen fulfilling the conditions of maximal variance and no correlation ( Vyas and Kumaranayake, 2006).

Variable of the study No

Variable

1

Low profit due to domestic competition

2

Limited access to market/small domestic market

3

Lack of long standing customer relationship

4

Duration too short/repayment schedule

5

Hartal (strike)

6

Limited access to credit/inadequate loan size

7

Collateral requirement/Insufficient collateral possessed by most SMEs

8

Interest rate high

9

Lack of government support and assistance

10

Lack of feasibility study and proper business plan

11

Lack of information on technical and marketing aspects of SMEs

12

Lack of management knowledge

13

Legal and regulatory framework

14

Scarcity of technical skill

15

Inability of skilled manpower

16

Lack of both public and private sector backed credit guarantee schemes for SME financing

17

High cost of equipment

18

High cost of raw materials

19

Lack of entrepreneurship education and training

20

Poor infrastructure

21

Problem of power supply

Table 5. Variable of the study

65

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

RESULT OF THE STUDY

Table 6. Correlation of the variable

Kaiser-Meyer-Olkin(KMO)

0.746

Bartlett’s Test of Sphericity Approx. Chi-Square

5249.038

Degree of freedom

210

Significant

0.000

Table 7. Measure of the sample adequacy

Bartlett’s Test of Sphericity was applied to measure the sampling adequacy for PCA analysis. The KMO measures the sampling adequacy which should be greater than 0.5 for a satisfactory factor analysis to proceed. Looking at the table above, the KMO measure was 0.746. High value generally indicates that a factor analysis may be useful with sample data. The same table shows that the Bartlett’s test of sphericity was significant. The test’s associated probability was less than 0.05. In fact, it was actually 0.000. This means that the correlation matrix was not an identity matrix.

Component

66

Total

% of Cumulative Variance %

Rotation Sums of Squared Loadings

Extraction Sums of Squared Loadings

Initial Eigenvalues Total

% of Cumulative Variance %

Total

% of Cumulative Variance %

1

6.088

28.991

28.991

6.088

28.991

28.991

3.429

16.328

16.328

2

3.109

14.804

43.794

3.109

14.804

43.794

3.172

15.106

31.434

3

2.461

11.719

55.513

2.461

11.719

55.513

2.616

12.456

43.890

4

1.887

8.987

64.500

1.887

8.987

64.500

2.494

11.876

55.766

5

1.371

6.528

71.028

1.371

6.528

71.028

2.397

11.412

67.178

6

1.288

6.134

77.162

1.288

6.134

77.162

2.097

9.984

77.162

7

.946

4.507

81.668

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

8

.809

3.852

85.520

9

.755

3.597

89.117

10

.567

2.700

91.817

11

.436

2.077

93.894

12

.425

2.026

95.920

13

.306

1.455

97.375

14

.165

.788

98.163

15

.132

.630

98.793

16

.084

.400

99.193

17

.066

.314

99.506

18

.043

.204

99.710

19

.030

.142

99.852

20

.027

.129

99.980

21

.004

.020

100.000

Extraction Method: Principal Component Analysis.

Table 8. Total variance explained

A six-factor solution explained 77.162 of the total variance (Table 8). The first factor explained the 28.991% of total variance. The second factor explained 14.804 % of the total variance. The third factor contained items that explained 11.719 % of the total variance. The fourth factor explained 8.897of the total variance. The fifth factor explained 6.528%of the total variance. The six factors explained 6.13%3of the total variance. All the remaining factors are not significant.

Figure 1. Scree plot showing the eigenvalues associated with each factor

67

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Figure 1 shows that there are 6 factors with an eigenvalue of more than 1. Moreover, Table 8 also shows that factor 7 has an eigenvalue of less than 1. The figure showed that the curve begins to flatten between factors 6 and 7. So, only six factors (with an eigenvalue of >1) have been retained.

Component 1

2

3

4

5

6

Low profit due to domestic competition

.911

.057

.041

.176

.015

.215

Limited access to market/small domestic market

.906

.034

.029

.195

.073

218

Lack of long standing customer relationship

.875

.113

.055

.007

.325

.160

Duration too short/repayment schedule

.643

.297

.225

.081

.000

.193

Hartal (strike)

.514

.181

.026

.215

.509

.144

Limited access to credit/inadequate loan size

.149

.932

.020

.095

.153

.016

Collateral requirement/Insufficient collateral possessed by most .171 SMEs.

.910

.004

.071

.206

.060

Interest rate high

.201

.893

.016

.031

.100

.098

Lack of government support and assistance

.091

.487

.068

.260

.041

.049

Lack of feasibility study and proper business plan.

.139

.010

.904

.131

.141

.010

Lack of information on technical and marketing aspects of SMEs

.062

.046

.901

.150

.151

.014

Lack of management knowledge

.010

.025

.640

.207

.192

.008

Legal and regulatory framework.

.151

.371

.452

.040

.307

.078

Scarcity of technical skill

.226

.088

.104

.891

.212

.045

Unavailability of skilled worker

.220

.081

.131

.894

.209

.035

Lack of both public and private sector backed credit guarantee schemes for SME financing.

.045

.236

105

.579

.065

.243

High cost of equipment

.088

.137

.057

.258

.876

.121

High cost of raw materials

.176

.214

.067

.029

.792

.101

Lack of entrepreneurship related education and training

.101

.094

.480

.417

.518

.234

Poor infrastructure

.120

.109

.001

.098

.079

.944

problem of power supply

.156

.090

.019

.145

.116

.929

Extraction method: Principal Component Analysis.

Rotation method: Varimax with Kaiser Normalization.

Table 9. Rotated Component Matrix

Here, Factor 1 explains variables; low profit due to domestic competition, small domestic market, lack of long standing customer relationship and this factor interpreted as domestic market competition.

68

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Factor 2 explains variables having limited access to credit, insufficient collateral possessed by most SMEs, and high interest rate and therefore, theses three variables are grouped into one.

Factor 3 explains variables having lack of feasibility study and proper business plan and lack of information on technical and marketing aspects of SMEs —and therefore, it is grouped as a factor related to lack of proper business knowledge and plan of SMEs. Factor 4 explains variables related to scarcity of technical skill and unavailability of skilled worker —and therefore, this is related to lack of skilled workers and technology.

Factor 5 explains two variables, high cost of equipment and high cost of raw materials —and so it may be inferred as a factor related to the high cost of raw material and equipment.

Factor 6 explains variables having poor infrastructure and problem of power supply —and it is related to poor infrastructure and electricity.

Factor

Factor loading Mean

Rank

Domestic market competition

0.897

4

Problem access to credit

0.911

2

Lack of proper business knowledge and plan of SMEs

0.903

3

Lack of skilled workers and technology

0.893

5

High cost of raw material and equipment

0.834

6

Poor infrastructure and electricity problem

0.937

1

Table 10. Summary of the factors

Finally, the results were ranked according to factor loading mean value (Table 10). Firstly, the result showed that poor infrastructure and electricity is the top constraint to the development of SMEs in Bangladesh. Secondly, problem access to credit affirmed the second highest constraint to SMEs’ development. In addition, lack of proper business knowledge and plan of SMEs and domestic market problem competition hold the third and fourth constrain to SMEs. Lastly, lack of skilled workers and technology; high cost of raw material and equipment are the fifth and sixth problem to SMEs’ development.

DISCUSSION This study tried to identify major constraints to SMEs’ development in Bangladesh. This study revealed that poor infrastructure and electricity are major constraints to SMEs’ development in this country. The previous results showed that the lack of the electricity is the main constraint to SMEs’ development in this South East Asia, especially in India, Pakistan and Bangladesh (Khandker, 2014).

In addition, lack of consistency of the electricity supply is the most important problem that negatively affected not only SMEs’ development but also hampered its productivity and sales.

For this reasons, SMEs’ development in rural areas is more difficult than in urban area. In rural area, electricity supply is quite poor; some rural areas even have no electricity supply.

In urban area shortage of electricity supply hampers the productivity of the SMEs and sell of the product (Khandker, 2014). Vila and MacDonald (2013) showed in their study that about 13.2% of total annual sales have been lost due to electricity supply.

69

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Our study also found that poor infrastructure is also another vital constraint to SMEs’ development in Bangladesh. The transportation infrastructure in Bangladesh is not properly developed and is especially poor in rural area. Product delivery is not possible in due time because of the poor infrastructure. For example, garments industry in Bangladesh need one month longer time compared to other competitive countries for purchasing order to delivery of the product (Chowdhury,2007). Poor transportation infrastructure and insufficient transport services result in high transportation cost that reduces the profit opportunity of the SMEs (Chowdhury, 2007).

Problems with access to credit, especially insufficient access to institutional and formal source of financing are considered as another major constraint to the growth of SMEs (Haque and Mahmud,2003).

Financial institutions are less likely to give loans to SMEs due to high risk level associated with SMEs. Evidence supports that all the 60.4% SMEs certain financial problems, whereas 68.6% microenterprises faced problems in terms of accessing finance in Bangladesh (Vila and MacDonald, 2013). It is also observed that 74.9% SMEs faced financial problems while they were initializing new SMEs and repairing the older one (Vila and MacDonald, 2013).

High interest rate is also major constraint to SMEs’ development. It also reduces the profit of the SMEs. All SMEs in Bangladesh have to pay average 15.6% interest rate (Vila and MacDonald, 2013).

It is also noted that banks are the major sources of formal loan of SMEs but the bank requires higher interest rate compared to other sources like as microfinance institutions, government related sources and private individuals. SMEs have to pay average 16.5% interest rate to the Bank whereas the rate is lower ,15.4%, to other institutions (microfinance institutions and government related sources) and the interest rate is lowest to private individual -11.4% (Vila and MacDonald, 2013).

Lack of proper business knowledge and plan of SME is another reason of failure to success in SMEs’ development, especially microenterprises that have been established for a short time goal, and have no proper plan for the long run. The main purpose of the growth of some small enterprise is to fulfill the demand of the present market situation. They have no proper long time goal even have no idea on how to adopt if the market condition changes. As a result, most small enterprises have been closed because they could not adapt to the new market situation. Large industries have strong R&D that is capable to measure present market condition and evaluate the future market condition, and can take the long time business plan.

Domestic market competition is also constraint to SMEs’ development in Bangladesh. Manufacturing and SMEs of the RMG sector also have to take the challenges of global competition. But within the country, the formal SMEs have to be competitive with informal agents to supply the same market segment (Vila and MacDonald, 2013). Most micro-companies do not have a clear strategy to deal with more fierce competition beyond price competition (Vila and MacDonald, 2013). Small companies have lack of syndicate to negotiate with larger companies (Vila and MacDonald, 2013).

Larger industries have the policy to adopt new technologies in commercial intelligence. But SMEs do not have those types of efficiency and are not able to proficiency in technology development.

Large industries published their activities in public media through website or monthly, yearly in the form of books. One can easily trace their growth, profit, manpower, property from these sources.

When they go to bank for demanding loan, the bank can easily observe the database and can recognize the large industry as a reputed organization. They generally choose the large industry for the loan. But SMEs have no record about their activities and may not publish their activities and as a result they are not identifiable in market easily. When they go to bank or other finance institution, the bank may not take the risk to give the smart loan to the unknown organization.

70

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Large industry has a research and development unit and a strong human resource management unit.

They maintain updated information about the loan policy of government institutions and financial institutions, and even keep the proper information about entire bank’s loan policy. They have proper knowledge about the process of loan and demand for loan - which banks’ loan scheme is suitable for them. Moreover, large industry has been taking loan for the long time from the bank. They have good relation with the bank and the bank recognizes them as a trustworthy organization. Large industry can easily get the loan compared to small and medium industry. So, the poor management and accounts practices of the SMEs are also the reason to limited access to finances.

Banks require collateral for various types of loan. The small firms generally have not enough collateral or have no goodwill for taking the loan from any financial institutions. So, access to finance is one of the vital constraints to develop SMEs in Bangladesh. Without enough capital, it is very hard to start a business in Bangladesh. Lack of skilled workers and technology is also one of the constraints to SMEs’ development. Most of the workers in SMEs are not properly skilled, and most of the small SMEs have no sophisticated technology.

CONCLUSION This paper identified constraints to development of SMEs in Bangladesh. This study revealed that poor infrastructure and electricity are the main constraints to start up new SME. Access to finance is the second major problem to SME’s development. Lack of proper business knowledge and plan of SMEs is also a major problem in Bangladesh. Most of the SMEs have been established without proper business knowledge even they have no long term business plan. Domestic market competition, lack of skilled manpower and high cost of raw materials also remain the constraints of SMEs’ development in Bangladesh.

REFERENCES Abor, J., & Biekpe, N. (2006). Small business financing initiatives in Ghana. Problems and Perspectives in Management, 4(3), 69-77.

Abor, J., & Quartey, P. (2010). Issues in SME development in Ghana and South Africa. International Research Journal of Finance and Economics, 39(6), 215-228.

Acma, M.Q. (2015). Productivity and Performance Evaluation of SME Sector in Bangladesh: Evidence from the Historical Data. Journal of Islamic Finance and Business Research, 3(1), 14-22.

Acs, Z.J., Morck, R., & Yeung, B. (1999). Productivity growth and firm size distribution. In Z.J. Ács, B. Carlsson & C. Karlsson (Ed.), Entrepreneurship, Small and Medium-Sized Enterprises and the Macroeconomy (pp.

367-396). Cambridge: Cambridge University Press.

Ahmed, K., & Chowdhury, T.A. (2009). Performance evaluation of SMEs of Bangladesh. International Journal of Business and Management, 4(7), 126. DOI:10.5539/ijbm.v4n7p126 Aziz, T., & Siddique, M.N.E.A. (2016). The Role of Bangladesh Bank in Promoting SMEs’ Access to Finance in Bangladesh. International Journal of SME Development, 3(2), 103-118.

Bangladesh Bank. (2011). Small and Medium Enterprise (SME) Credit Policies & Programmes. SME & Special Programmes Department. Retrieved September 20, 2017, from https://www.bb.org.bd/sme/smepolicye.pdf Bangladesh Bank. (2017). Industrial Credit Report. Retrieved September 20, 2017, from https://www.bb.org.bd/ openpdf.php

71

EJAE 2018  15 (2)  58-73

ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Bangladesh Bureau of Statistics. (2017). 2017 Statistical Year Book Bangladesh. Retrieved September 20, 2017, from http://bbs.portal.gov.bd/sites/default/files/files/bbs.portal.gov.bd/page/b2db8758_8497_412c_a9ec_ 6bb299f8b3ab/S_Y_B2017.pdf Beck, T., & Demirguc-Kunt, A. (2006). Small and medium-size enterprises: Access to finance as a growth constraint. Journal of Banking & Finance, 30(11), 2931-2943. DOI:10.1016/j.jbankfin.2006.05.009 Beck, T., Demirgüç‐Kunt, A. S. L. I., & Maksimovic, V. (2005). Financial and legal constraints to growth: does firm size matter? The Journal of Finance, 60(1), 137-177.

Beck, T., Demirguc-Kunt, A., & Levine, R. (2005). SMEs, growth, and poverty: Cross-country evidence. Journal of Economic Growth, 10(3), 199-229. DOI:10.1007/s10887-005-3533-5 Beck, T., Demirgüç-Kunt, A., & Maksimovic, V. (2008). Financing patterns around the world: Are small firms different? Journal of Financial Economics, 89(3), 467-487. DOI: 10.1016/j.jfineco.2007.10.005 Centre for Policy Dialogue. (2017). Bangladesh Economy in FY2016-17: Interim Review of Macroeconomic Performance. Retrieved September 20, 2017, from https://cpd.org.bd/wp-content/uploads/2017/06/BangladeshEconomy-in-FY2016-17-Interim-Review-of-Macroeconomic-Performance.pdf Chowdhury, M. S. (2007). Overcoming entrepreneurship development constraints: the case of Bangladesh.

Journal of Enterprising Communities: People and Places in the Global Economy, 1(3), 240-251. DOI: 10.1108/17506200710779549 Chowdhury, M. S. A., Azam, M. K. G., & Islam, S. (2015). Problems and prospects of SME financing in Bangladesh. Asian Business Review, 2(2), 51-58.

Chowdhury, T. A., & Ahmed, K. (2011). An Appraisal of the Problems and Prospects of Small and Medium Enterprises (SMEs) Financing in Bangladesh: A Study on Selected Districts. Retrieved September 20, 2017, from http://dspace.ewubd.edu/handle/123456789/415 Dinhucha Gonçalves Fumo, N., & Jose Chiappetta Jabbour, C. (2011). Barriers faced by MSEs: evidence from Mozambique. Industrial Management & Data Systems, 111(6), 849-868. DOI: 10.1108/02635571111144946 Haque, A. K. E., & Mahmud, S. (2003). Economic policy paper on access to finance for SMEs: Problems and Remedies. Retrieved September 20, 2017, from http://www.dhakachamber.com/economic_policy/Access_to_finance.pdf Hasan, F., & Jamil, G. M. H. (2014). Financing Small and Medium Enterprises in Bangladesh-Issues and Challenges. The Asian Journal of Technology Management, 7(1), 45-54.

Hoque, M. R., Saif, A. N. M., AlBar, A. M., & Bao, Y. (2016). Adoption of information and communication technology for development: A case study of small and medium enterprises in Bangladesh. Information Development, 32(4), 986-1000.

Hossain, N. (1998). Constraints to SME Development in Bangladesh. College Park, MD: Institutional Reform and the Informal Sector, University of Maryland.

Khandke, A. (2014). Constraints and Challenges of SME Development in the Developing Countries: A Case Study of India, Pakistan and Bangladesh. International Journal of SME Development, 1(1), 87-118.

Kshetri, N. (2011). The Indian environment for entrepreneurship and small business development. Cluj-Napoca: Babes-Bolyai University.

Longenecker, J. G., Moore, C. W., Petty, W., & Palich, L. E. (2005). Small business management: An entrepreneurial emphasis. Mason, OH: SWC Publishing.

Mac an Bhaird, C., & Lucey, B. (2010). Determinants of capital structure in Irish SMEs. Small Business Economics, 35(3), 357-375. DOI:10.1007/s11187-008-9162-6 Mambula, C. (2002). Perceptions of SME growth constraints in Nigeria. Journal of Small Business Management, 40(1), 58. DOI:10.1111/1540-627X.00039 Moktan, S. (2007). Development of small and medium enterprises in Bhutan: Analysing constraints to growth.

South Asian Survey, 14(2), 251-282. DOI:10.1177/097152310701400205 72

EJAE 2018  15 (2)  58-73 ISLAM, M. S., HOSSAIN, M. F.  CONSTRAINTS TO SMALL AND MEDIUM-SIZED ENTERPRISES DEVELOPMENT IN BANGLADESH: RESULTS FROM A CROSS-SECTIONAL STUDY

Peltier, J. W., & Naidu, G. M. (2012). Social networks across the SME organizational lifecycle. Journal of Small Business and Enterprise Development, 19(1), 56-73. DOI:10.1108/14626001211196406 Quader, S. M., & Abdullah, M. N. (2009). Constraints to SMEs: A Rotated Factor Analysis Approach. South Asian Studies, 24(2), 334-350.

Quartey, P. (2003). Financing small and medium enterprises (SMEs) in Ghana. Journal of African Business, 4(1), 37-55.

Syed Manzur, Q. (2008). Constraints to SMEs: A Rotated Factor Analysis Approach. A Research Journal of South Asian Studies, 24(2), 334-350.

Tahi Hamonangan Tambunan, T. (2011). Development of small and medium enterprises in a developing country: The Indonesian case. Journal of Enterprising Communities: People and Places in the Global Economy, 5(1), 68-82. DOI:10.1108/17506201111119626 Vila, J., MacDonald, M. (2013). The state of the SME sector-the manufacturing SME sector in Bangladesh (2013): An overview. Draft working paper 3.

Vyas, S., & Kumaranayake, L. (2006). Constructing socio-economic status indices: how to use principal components analysis. Health Policy and Planning, 21(6), 459-468.

Zaman, A.H., & Islam, M.J. (2011). Small and medium enterprises development in Bangladesh: Problems and prospects. ASA University Review, 5(1), 145-160.

OGRANIČENJA ZA RAZVOJ MALIH I SREDNJIH PREDUZEĆA U BANGLADEŠU: REZULTATI IZ UNAKRSNE STUDIJE

Rezime: Mala i srednja preduzeća (MSP) značajno doprinose ekonomiji Bangladeša jer imaju potencijal da kreiraju veću zaposlenost, poboljšaju lokalne tehnologije i stvore osnove za buduće održive industrijske i korporativne sektore. Cilj ove studije je da identifikuje glavna ograničenja sa kojima se suočavaju mala i srednja preduzeća u Bangladešu.

Unakrsna studija sprovedena je u opštini Narsingdi, dok je za prikupljanje podataka primenjena metoda anketiranja javnog mnjenja u Bangladešu. Metod uzorkovanja nasumično selektovanih kandidata u dve faze korišćen je za odabir ispitanika, a podaci su prikupljeni putem intervjua licem u lice. Analiza glavne komponente korišćena je za utvrđivanje značajnih ograničenja za MSP. Studija je pokazala da su slaba infrastruktura i snabdevanje električnom energijom glavna ograničenja za razvoj malih i srednjih preduzeća.

Ograničen pristup kreditima, nepostojanje odgovarajućeg poslovnog znanja i plana, velika konkurencija na domaćem tržištu, nedostatak kvalifikovanih radnika i tehnologija i visoki troškovi sirovina i opreme su glavna ograničenja.

Ključne reči: mala i srednja preduzeća (MSP), ograničenja, analiza glavnih komponenti, Bangladeš.

73

EJAE 2018, 15(2): 74-90 ISSN 2406-2588 UDK: 330.142 339.13:336.763(669)"1998/2016" DOI: 10.5937/EJAE15-17631 Original paper/Originalni nauÄ?ni rad

CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA Yinusa Olumuyiwa Ganiyu1,*, Ismail Adelopo2, Yulia Rodionova3, Olawale Luqman Samuel1 1 Department of Accounting, Banking and Finance, Olabisi Onabanjo University, Ago-Iwoye Ogun State, Nigeria 2 Department of Accounting, Economics and Finance, University of the West of England United Kingdom 3 Department of Accounting and Finance, Leicester Business School, De Montfort University, Leicester United Kingdom

Abstract: This paper examines the impact of firm specific factors as determinants of capital structure choice of Nigerian firms based on the data of 115 non-financial firms listed on the Nigerian stock exchange, for the period 1998-2016.The study employed two-step system generalized method of moment in a dynamic panel framework. The findings of the study reveal positive relationship between profitability, firm risk, firm dividend and leverage. Asset tangibility, growth opportunities, size and age are found to be negatively related to leverage. The study therefore concludes that variables identified in the agency cost theory that provide explanations for capital structure of firms in the developed and some emerging countries, are relevant but not fully applicable in the Nigerian context.

This study shows that managers tend to use more debt as they prefer higher free cashflow because it facilitates the consumptions of perks.

The use of long-term debt may reduce the opportunistic behaviour of managers. Managers may strive to ensure debt repayment promptly to avoid bankruptcy which can be very costly for the firm, and managers may lose their job and reputation. Managers would prefer the use of less debt in firms where there are high growth opportunities. Managers of firms may prefer to employ less debt when they see that debt can restrict them to explore future opportunities because of the commitment and covenants associated with debt.

74

* E-mail: yinusa2016@gmail.com

Article info: Received: August 25, 2018 Correction: September 16, 2018 Accepted: September 18, 2018

Keywords: capital structure, agency theory, generalized method of moment, dynamic panel.

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

INTRODUCTION Studies that have examined the capital structure decisions in the emerging countries context are basically divided into two streams. First, studies that report full portability of the capital structure theories in the emerging countries context without modification to account for the various market imperfections that characterise emerging market such as lack of contract enforcement, poor corporate governance, and high asymmetric information. These studies established that the traditional capital structure theories provide full explanation for the financing choices of firms in emerging markets (Sheik and Wang, 2011; Akinlo, 2011). Second, there is a large number of studies that suggest that the traditional capital structure theories are applicable in emerging countries, but they are not fully portable and therefore require modifications to account for the specificity of emerging countries in terms of institutional and macroeconomic framework that are quite different from the developed economies (Li et al. 2007).

Most of the previous studies on capital structure in the emerging countries in Africa (Akinlo, 2011 Barine, 2012; Chandrasekharan, 2012; Matemilola, Bany-Ariffin and Azman-Saini, 2013; Fosu, 2013; Muhtar and Ahmad, 2015) are focused on joint analyses of the different capital structure theories particularly in a static framework. None of these studies has considered the agency cost theory of capital structure in an emerging country context. Emerging market in Africa serves as excellent laboratory to test the agency theory of capital structure because of the profound misaligned interests between managers of firms in emerging market and their shareholders (Harvey, Lins and Roper, 2003). This study therefore addresses the gap by examining the portability of the agency cost theory in a market that defers from the developed market in terms of institutions and macroeconomic environment.This study complements the existing studies on emerging market by focusing specifically on firms in Nigeria; a low income emerging economy in sub Saharan Africa that is pervaded with several market imperfections (Gwartidzo, 2009 ).

The study contributes to the empirical literature on capital structure in three ways. First, the study represents one of the few studies which analyse the capital structure choices of firms within the agency cost theoretical framework using data from an emerging stock market in Africa. Second, the study uses two step dynamic generalized methods of moments which capture the dynamic effects of the capital structure as part of the determinants of capital structure rather than a static framework usually considered by past studies in the literature. The rest of the study is structured as follows- section 2 presents the literature review and hypotheses development. Section 3 deals with data and methodology, whereas section 4 presents the results and the paper concludes in section 5.

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT This section of the study reviews relevant literature and develops hypotheses tested in the study.

Theoretical underpinning: Agency cost theory Modigliani and Miller’s (1958) seminal paper on the irrelevance of capital structure on firm value (hence performance) laid the foundation for other differing theoretical predictions. The trade-off theory relaxed the perfect market assumptions of Modigliani and Miller (1958) and made theoretical prediction that capital structure is relevant for firm performance for reasons such as tax deductibility of debt interest and agency costs (Fosu, 2013).

75

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Jensen and Meckling (1976) pioneering research on agency cost theoretical model builds on the previous study by Fama and Miller (1972). Jensen and Meckling (1976) argued that the observed capital structure of a firm should have the central objective of minimising the potential for opportunistic behaviour. Modern firms are organised in such a way that ownership is separated from control.

Managerial ownership further complicates the conflict of interests concerns (Jensen and Meckling, 1976; Florackis, Kanas and Kostakis, 2015). As part owner in the firm, managers have dual access to firm information which give them advantage above the residual claimants in the firm (Magaritis and Psillaki, 2007). This compounds the asymmetry in the informational wealth of the manager compared to other owners of the firm (Flannery, 1986; Ross, 1977). Agency theoretical underpinning also suggests that managers may be caught up in a classical moral hazard situation where their best intention produces negative value to the firm (De Miguel et al. 2005). This is what Jensen and Meckling (1976) described as the Agency cost, which is regarded as the reduction in the value of the firm as a result of the opportunistic behaviour by management of the firm.

One area of firm’s operation that exposes both managers and shareholders to conflict of interests is their capital structure decision. In this situation, managers have a choice to either act in the interest of the firm or in their own interest. Agency theoretical frame argues that managers, as rational and utility maximising economic actors, are more likely to take decisions that maximises their personal returns than the firm’s return. In order to ensure their wealth is maximized, the shareholders impose measures to ensure that the agency cost is mitigated. One of such measures that reduce manager’s opportunistic behaviour is the use of capital structure, especially debt. Jensen and Meckling (1976) posit that firms can use their capital structure to mitigate the agency problem that arises from the opportunistic behaviour of managers. They identified two kinds of conflict in the firm due to agency problem and agency cost.

They have noted that on one hand there is a conflict between the shareholders and managers and on the other hand there is a conflict between the shareholders and debt holders.

The conflict between shareholders and managers arises because managers do not have full residual claim in the firm. Therefore, managers may not act fully to protect the interest of shareholders but rather they may waste free cash flow on perquisites and bad investment. To forestall this, shareholders (Principal) create appropriate incentives for managers (agent) and incur monitoring costs to reduce the self-seeking behaviour of managers (Michalca, 2011). In order to resolve the conflict between managers and shareholders, firms tend to employ more debt in their capital structure. This often results into more debt repayment by the firms. This reduces the available cash flow in the firm and thus helps to control the opportunistic behaviour of managers (Manos, 2001; Michalca, 2011).

The second conflict identified by Jensen and Meckling (1976) is related to the conflict that arises between debt holder and equity holders, because the debt employed by the firm to mitigate agency problem creates opportunities for shareholders to invest in a suboptimal manner which can result in risk shifting (Harris and Raviv, 1991; Sun, Ding, Guo and Li, 2015). Risk shifting relates to the tendency of debt employed by firms to induce equity holders to engage in higher risk investment than the debt holder envisaged (Harris and Raviv, 1991). This would cause changes to the cash flow and reallocate wealth from debt holders to equity holders if the risky investment is successful. This is possible because the amount of interest payable to the debt holder must have been fixed in the debt contract before the risk shifting behaviour of the firm (Ismail, 2006). The extra gains from the successful risky investment become accruable to the equity holders. This risk shifting behaviour therefore could make debt to become more expensive, more constraining and less available in future as a source of finance (Manos, 2001). This implies that the use of debt becomes more of cost to the firm than benefit.

76

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

The explanations above implies that what drives firms to employ debt rather than equity in their capital structure is simply their aim to mitigate the agency cost that arises from separation of ownership from control, rather than to take the tax advantage of using debt and trade off the tax benefit of debt against the cost such as financial distress cost as posit in the trade-off theory ( Sheik and Wang, 2011).

The agency cost theory also signifies that debt is not used by the firm to signal to investors about the quality of the firm in terms of capacity to use debt and make repayment after exhausting the internal funds. Equity is however employed after the debt has also been exhausted and there is still financing needs by the firm as posit in the pecking order hypothesis (Belkhir, Maghyereh and Awartani, 2016).

The agency cost theoretical model by Jensen and Meckling (1976) assumed that firms have optimal capital leverage position that they strive to achieve. The optimal capital structure of the firm in the agency cost theoretical model is the capital structure level that minimizes the agency cost and maximizes the value of the firm (Gwartidzo, 2009). This implies that capital structure choice of the firm is not static but dynamic. The dynamic nature of capital structure suggests that capital structure of firm’s changes across firms and across time i.e. each firm in an industry for example can change its capital structure over time to ensure the agency cost is minimized and value of the firm is maximized (Korajczyk and Levy, 2003).

It could be argued that the Agency cost theoretical model is more relevant in an environment where the rights of creditors and shareholders are not well protected, where institutional quality in terms of laws and its enforcements are very weak, and where financial development is still at infant stage because of the fact that these features perpetuate information asymmetry and do not make the market to be efficient in any meaningful way (Kesister, 2004). These features are persistent features in most emerging markets (Myers, 2003; Haron, 2014).

It is against the backdrop of the presence and persistence of these features that promotes agency related problems such as weak institutions and other market imperfections in a country like Nigeria that motivate this research to employ the agency cost theory as the main theoretical basis. The study empirically examines the full portability of agency cost theoretical model in the Nigerian context with the aim of providing empirical validity for the theory in an economy that is quite different in terms of operating environment of firms that Jensen and Meckling (1976) had in mind when they developed the agency cost theoretical model.

Jensen and Meckling (1976) identified some agency related theoretical variables as proxies for agency problem that influence the capital structure of firm. These variables are measured by firm characteristics such as: size, growth opportunities, asset tangibility, ownership, risk and profitability. The theoretical predictions of the agency cost based model on the relationship between capital structure and these agency theoretical variables differ from one variable to the other. Similarly, empirical studies that have considered these factors have also reported mixed and inconclusive findings. Based on the predictions of the agency cost theory and the empirical findings of studies in the literature, this study developed hypotheses to test the empirical validity of the agency cost theory in the Nigerian context. The next section develops hypotheses for this study.

Determinants of Leverage There are several firm specific variables that influence capital structure choice of firms according to the agency cost theory. Several empirical studies on emerging market capital structure have provided mixed empirical results regarding the determinants of capital structure. Based on the firm specific 77

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

variables used in empirical capital structure model of Frank and Goyal (2007), this study employs the following firm specific factors as determinants of capital structure of firms using the agency cost theoretical framework.

Profitability The expected theoretical relationship between leverage and profitability is positive. This theoretical prediction is hinged on the fact that debt is used by the firm as a measure to prevent managers to have access to excess cash flow that can enhance their opportunistic behaviours (Belkhir, Maghyereh and Awartani, 2016). As firms become more profitable, they have the capacity to obtain debt and repay it on due date. Free cash flow can make managers engage in other activities that are not value enhancing for the firm. Firms therefore use debt to forestall these (Charkraborty, 2010). Several empirical studies (including, Shehu, 2011; Chandrasekharan, 2011; Barine, 2012) have found support for the agency cost theory prediction of a positive relationship between profitability and leverage. However, other studies such as Al-Sakran (2001), Chen (2004), Chakraborty (2010), Sheik and Wang (2011), and Akinlo (2011) have reported negative relationship between profitability and leverage. They argue that profitable companies have less need for debt since new investment can be undertaken from retained earnings.

Due to the conflicting findings in the literature, our null hypothesis is that: H1: There is no relationship between profitability and leverage of firm.

Asset Tangibility Agency cost theory predicts positive relationship between asset tangibility and firm leverage. This theoretical prediction is based on the fact that firms with tangible asset tend to use more debt in their capital structure. This is because the tangible asset can serve as collateral to secure the debt in the event of bankruptcy. Tangible asset could also provide additional economic benefit to the firm in terms of their tax-shield effects. Empirical findings in respect of the relationship between firms’ asset tangibility and leverage have been inconclusive. While some studies report positive relationship (Shehu 2011; Chandrasekharan 2012; Drobetz et al. 2013), others have either reported negative or no relationship at all (Sheik and Wang 2011; Akinlo 2011; Michalca 2011, and Joeveer 2013). Studies that have reported negative or no findings argue that firms with more tangible assets are likely to have less need for debt financing since they could use their superior assets quality to generate better performance. However, there are far more studies that have found support for the agency cost prediction of positive relationship between firm leverage and asset tangibility. Consequently, we hypothesise that: H2: There is a positive relationship between asset tangibility and leverage of firm.

Growth opportunities Previous studies have argued that growth opportunities that a firm faces could impact on the propensity to use debt compared to equity (Manos, 2001). Firms in high growth potential industry require higher investment than firm in saturated industry. The decision on the sources of fund to explore growth 78

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

opportunities is a strategic one that directors have to make. Furthermore, previous studies argue that debt financing is cheaper than equity (Jensen, 1989 ) and consequently it should be expected that firms operating in high growth industry will use more debt rather than equity compared to firms operating in less growth opportunities. Looking from an agency theoretical frame, it is reasonable to expect a positive relationship between growth opportunities and leverage. Due to growth opportunities managers have high tendency to engage in opportunistic behaviour, since such behaviour may be covered by the need to exploit market opportunity which may be more significant than worrying about management opportunistic behaviour. Thus, firms may use debt to curtail managers’ opportunistic behaviour due to repayment commitment (Jensen and Meckling, 1976; Margaritis and Psillaki, 2010).

However, managers would prefer the use of less debt in firms where there are high growth opportunities especially where the level of managerial entrenchment is high. Managers may prefer to use less debt when they see that debt can restrict them to explore future opportunities because of the commitment and covenants associated with debt and the fear of consequences associated with bankruptcy (Jensen and Meckling, 1976). Studies (such as Akinlo, 2011) document negative relationship between debt and growth opportunities. In order to mitigate the opportunistic behaviour that can arise due to future growth opportunities, firms are likely to employ more debt in order to prevent opportunistic behaviour of managers using the debt repayment commitment to reduce the free cashflow of managers. This suggests that the more the opportunities for growth the more debt firms are likely to employ. Therefore positive relationship is expected between growth opportunities and debt. Shehu, (2011) and Chandrasekharan, 2012 reported positive relationship between debt and growth opportunities. Based on these theoretical predictations and mixed empirical findings. The study formulates null hypotheses as below: H3: There is no relationship between growth opportunities and leverage

Firm risk The relationship between firm risk and leverage is inconclusive in extant literature. The positive relationship between firm risk and leverage is hinged on the fact that firms employ debt as disciplinary device to prevent moral hazard and other opportunistic behaviour of managers especially the tendency of the managers to take actions that can increase the variation of earnings of the firm (Jensen, 1986).

To forestall this, the firms employ more debt in their capital structure to increase the debt commitment of the firm to debt holders. The higher the tendency for managers to engage in opportunistic behaviour, the more debt likely to be employed by the firm. However, the use of debt in an effort to reduce the opportunistic behaviour of managers can increase the level of volatility of the earnings of the firm and the tendency of going bankrupt if debt obligations are not met on due date (Jensen and Meckling, 1976; Jensen, 1986; Margaritis and Psillaki, 2010). Therefore risky firms have a dilemma in their capital structure decisions. This also depends on their current level of leverage. Highly leveraged firm may be symptomatic of high risk and in which case additional leverage could tilt the firm into bankruptcy when it is unable to meet repayment scheduled (Qian et al. 2009). Consequently, such a firm may reduce their level of leverage. This may lead to negative relationship between leverage and firm risk (Chen, 2004).

The literature gives inconclusive evidence on the relationship between leverage and firm risk. Positive relationship was reported between firms risk and debt ratio by Wiwattanakantang, (1999). However, 79

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Alderson and Betker (1995), Qian and Wirjanto (2009), Sheik and Wang (2011), Qian et al. (2008), Drobetz et al. (2013) documented negative relationship. Therefore the study hypothesized that: H4: There is no relationship between firm risk and leverage.

Dividend Firms pay dividend out of profit as a reward for shareholder and to signal good performance including strong capital base and reliable management. Consequently, it is not expected that firms will use debt to pay dividends. The relationship between leverage and dividend payout is linked to firm performance in terms of profitability. Profitable firms are more likely to have excess cash generated from trading and will therefore be able to pay dividend.

Extant literature suggests that such firms will rationally have less demand for debt financing. It is therefore reasonable to expect a negative relationship between firms’ dividend payout and leverage.

Empircal evidence on this relationship is scanty. Notable existing studies include the work of Florackis, Kanas and Kostakis, 2015). We hypothesise that: H5: There is a negative relationship between dividend payout and leverage.

Firm size and leverage Agency cost theory predicts dual relationship between firm size and leverage. The positive relationship is based on the fact that firms employ long term debt to mitigate managers’ opportunistic behaviour.

This usually applies in large firms where managers do not have controlling interest. Furthermore, large firms tend to have access to debt at cheaper cost than small firms because they are diversified and have reputation as well as capacity to repay their debt than small firms (Byoun, 2008).Thus a positive relationship is expected between firm size and leverage. Agency problem is also more prominent in large firms than small firms as the principal (shareholders) are usually separated from the agent (managers).

However, where the managers have controlling interest in the firm, they tend to grow the firm to large size and ensure their interest is well protected so that they have continuous access to perks and perquisites, as well as opportunities to engage in empire building (Margaritis and Psillaki, 2010). The controlling managers have the incentive to avoid the use of debt even when it is available at a cheaper cost to prevent bankruptcy and taking over of the firm by the outsider (Jensen and Meckling, 1976; Margaritis and Psillaki, 2010; Chakraborty, 2010; Fosu, 2012), and to foster the access to consumptions of perks which may become constrained if the firm has high leverage. It is therefore reasonable to expect a negative relationship between leverage and firm size.

Empirical studies also report mixed findings. For example studies such as Chakraborty (2010), Shehu (2011) and Chandrasekharan (2012) report negative relationship between leverage and firm size.

Based on the theoretical prediction of the agency cost theory and the mixed findings in the empirical literature. This study therefore hypothesized thus: H6: There is no relationship between firm size and leverage.

80

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Age Firm that have been in existence a long time may have built reputation for credit worthiness. Such firms have better opportunities and tendencies to employ debt as a device to mitigate agency problem of managers. However, young firms may find it very difficult to use debt as a mitigating device of agency problem because of the absence of debt repayment reputation that requires time to build. It is against this backdrop that this study hypothesises thus: H7: There is a positive relationship between age and leverage.

DATA AND METHODOLOGY This section of the study explains the study design and data used in the investigation. Justifications are provided for the variables and scope of the study. Data for the investigation were gathered from the Facts Book of the NSE, and the annual reports and accounts of the companies. The study includes only non-financial firm with three or more years of consecutive observations to enable a robust analysis.

The listed firms in Nigeria were classified into 13 new industrial classifications by the NSE as at 2012.

The study excludes firms in the financial services and investment industries due to their significantly different reporting and regulatory requirements. Table 1 below shows the sector classification, number of listed firms on the Nigerian Stock Exchange, and sample selection for this study. This study is based on data from 115 Nigerian out of 184 non-financial firms listed on the NSE. 69 firms were excluded due to lack of data availability. The study uses an unbalanced panel data framework consisting of 115 Nigerian firms listed on the Nigerian Stock Exchange (NSE) from the period 1998 to 2016 with at least three year consecutive data for each firm. The study used unbalanced panel due to lack of data. Adelopo (2011) highlights the significant data challenge that accounting and finance researchers in Africa face.

The next section presents the empirical model.

Industry Classification

Number of firms

Agriculture

8

Services

28

Consumer goods

43

Alternative Securities Market (ASEM)

15

Healthcare

16

Industrial Goods

30

Oil and Gas

10

Natural Resources

9

ICT

10

Construction and Real Estates

9

Conglomerates

6

Financial Services

97

Memorandum Quotation

27 81

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Total Companies listed on the Nigerian Stock Exchange as at 2012

305

Exclude: Financial services companies

97

Memorandum companies (Investment companies)

27

Actual Working population of the study

184

Exclude companies without data point for at least three years

69

Actual sample firms for the study

115

Table 1. Firms distribution of companies listed on the Nigerian Stock Exchange and sample selection

Empirical Model This study used a dynamic panel model of the form stated below to capture the persistence of leverage and to identify the optimal speed of adjustment in leverage level: yit = ∅yi ,t −1 + x 'it β + δ i + ε it Where yit is the current period leverage and yi,t-1 is the leverage a year ago with ∅ as the speed of adjustment of leverage to it former level. x 'it is the vector of independent variables including profitability, Asset tangibility, size, growth opportunities, age, dividend and risk. δ i is the fixed effect whilst ε it is the disturbance term.Using GMM model in a dynamic panel framework solves two important problems that empirical investigations in finance often confront. These are the endogeneity and omitted variable problems (Roodman, 2009). A Dynamic panel requires the presence of the lagged dependent variable as part of the independent variables. However, this renders both OLS and the usual static panel estimator bias and inconsistent. The presence of the lagged dependent variable as part of the explanatory variable leads to correlation between the error term in one period and the dependent variable and the error terms of another period. To resolve this problem Arellano and Bond (1991) suggest the use of Generalised method of moments (GMM) which essentially uses the idea of instrumental variables to correct for correlations between the error terms and the dependent variable. They suggested the use of lagged exogenous variables as instrument both at level and at first difference. However, Blundell and Bond (1998) showed that this approach does not provide optimal solution and instead suggest the use of system GMM in circumstance with panel that contain limited time (T) and large cross section(N).

Blundell and Bond (1998) argue that system GMM estimator explores more moment conditions on the lagged and difference level using the lagged difference of the exogenous variables as instruments in the level equation.

Two fundamental tests are undertaken to show the suitability of system GMM estimator. First whilst the rejection of the null hypothesis of no autocorrelation in the difference residual is not problematic, second order rejection of the null of no auto-correlation is problematic. Furthermore, Sargan/Hansen test is conducted to test the suitability of the instruments. The study used a stepwise regression approach whereby dependent variable is first regressed against its lagged value, then include the other firm specific variables in the model. Table 2 below presents detailed information about the variables including their sources, definitions, and expected signs between the variables.

82

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Variable

Definition

Sources

sign

Total Leverage Ratio (TLR)

Total Debt/ Total Debt plus Total Equity.

Frank & Goyal (2007); Handoo & Sharma (2014).

Positive

Long Term Leverage Ratio (LTLR)

Long Term Debt/ Total Debt plus Total Equity

Frank & Goyal (2007); Handoo & Sharma (2014).

Positive

Short Term Leverage Ratio (STLR)

Short Term Debt/ Total Debt plus Total Equity.

Frank & Goyal (2007); Handoo & Sharma (2014).

Positive

Profitability (PROF)

Earnings Before Interest and Tax/ Total Assets

Kayo & Kimura (2011); Chang et al. (2014); Gwatidzo & Ojah (2014); Handoo & Sharma (2014); Alves & Francis (2015); Belkhir et al. (2016).

Positive/ Negative

Assets Tangibility (TANG)

Tangible Fixed Assets/ Total Assets.

Kayo & Kimura (2011); Chang et al.

(2014); Handoo & Sharma (2014); Alves & Francis (2015); Belkhir et al.

(2016).

Positive

Growth Opportunities (GO)

Percentage change in the log of Total Assets.

Kayo & Kimura (2011); Chang et al.

(2014); Belkhir et al. (2016).

Positive/ negative

Risk

Standard deviation of the Earnings Before Interest and Tax / Total Asset.

Chang et al. (2014).

Positive/ negative

Dividend

Dividend/Profit after tax

Flokaris et al. (2015).

Negative

Size (SIZE)

Natural logarithm of Total Assets.

Kayo & Kimura (2011); Chang et al.

(2014); Gwartidzo & Ojah (2014); Alves & Francis (2015).

Positive/ negative

Age

Number of years that the firm have been in existence

Gwatidzo & Ojah (2014); Belkhir et al. (2016).

Positive

Table 2. Definition of Variables

EMPIRICAL RESULTS AND DISCUSSION OF FINDINGS This section presents the empirical results. First, it provides a detailed explanation of the descriptive statistics and the correlations analysis. This is then followed by the presentation of the two step system GMM regression results.

Descriptive statistics Table 3 below presents the descriptive statistics of the variables. During the entire period, the mean of total leverage ratio was 0.50. This is greater than the mean of long term leverage ratio of 0.19 but both are less than the mean of short term leverage of 1.0097. These indicate that on average, the sample firms employ more short term leverage ratio and total leverage ratio. The standard deviation for the total leverage ratio of the sample firms is 1.19. This suggests that total leverage ratio of sample firms has high variability as evidence in the coefficient of standard deviation that is greater than one. Long term leverage ratio has standard deviation of 0.86.Short term leverage ratio has the highest standard deviation (11.23) than total leverage and long term leverage ratios. The average age of the sample firms in the study was 33 years.

83

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Mean

Standard Deviation

Minimum

Maximum

Total leverage ratio

0.2462

0.2919

0.0000

0.9970

Long term leverage ratio

0.0874

0.1546

0.0000

0.9928

Short term leverage ratio

0.1588

0.2170

-0.3658

0.9744

Profitability

0.0363

0.0908

-0.3137

0.6213

Asset Tangibility

0.2018

0.3408

0.0000

10.4400

Growth Opportunities

0.2125

2.0104

-28.7904

33.1651

Risk

6.2020

74.3085

-166.856

2657.495

Dividend

0.1071

0.8614

-2.241

15.4936

Size

8.0335

5.5004

-4.2929

20.0598

Age

31.7790

19.4241

0.0000

93.00

Variable

Total Leverage Ratio (TLR) = total debt/Total Debt and total equity. Long Term Leverage Ratio (LTLR) = Long term debt/total debt and total equity. Short Term Leverage Ratio (STLR) = Short Term Debt/Total Debt and Total Equity. Profitability (PROF) = Earnings before Interest and Tax/Total Assets. Assets Tangibility (TANG) = Fixed tangible assets/total assets. Growth opportunities (GO) =Percentage change in the log of total assets. RISK = Standard deviation of the earnings before interest and tax to total asset.Size (SIZE) = Natural logarithm of total assets. Dividend pay-out= Dividend/Profit after tax. AGE = number of years on the Nigerian Stock Exchange.

Table 3. Descriptive statistics

The variability of the age of the sample firms is 19.42 as shown by the standard deviation. The average growth opportunities of the sample firms are 0.21.The standard deviation of growth opportunities for the sample firms was 2.01. This indicates high degree of variability of growth opportunities of the sample firms. The average fixed asset as a percentage of total assets (asset tangibility) of the sample firms is 0.20.

The standard deviation is 0.34. The mean value of firm size is 8.03 while the standard deviation is 5.50.

The average dividend of the sample firms from 1998 to 2012 is 0.11 with standard deviation of 0.86.The mean value of risk for the 115 firms is 6.20 and the variability of risk among the sample firms is 74.30.

This is consistent with findings from studies (Frank & Goyal, 2009; Kayo & Kimura, 2011; Chang et al. 2014) for emerging economies. They found book value of leverage and market value of leverage to be between 14.2% and 27.2%, 12.6% and 12.5% respectively establishing that firms value equity more than debt for their capital structure in those economies.

TLR

84

LTLR

STLR

PROF

TANG

GO

RISK

DIV

SIZE

TLR

1.0000

LTLR

0.6869

1.0000

STLR

0.8557

0.2117

1.0000

PROF

-0.0259

-0.0675

0.0132

1.0000

TANG

-0.0861

-0.0860

-0.0546

0.1154

1.0000

GO

-0.0566

-0.0378

-0.0492

0.0498

0.0719

1.0000

RISK

0.0676

-0.0237

0.1078

0.0213

-0.0133

-0.0084

1.0000

DIV

-0.0275

-0.0415

-0.0075

0.2081

0.0082

-0.0018

-0.0083

1.0000

SIZE

-0.1294

-0.1145

-0.0926

0.0835

0.2359

0.0813

-0.0410

-0.0375

1.0000

AGE

-0.2153

-0.1432

-0.1875

0.1187

0.1195

0.0281

-0.0422

0.0506

0.2478

AGE

1.0000

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Total Leverage Ratio (TLR) = total debt/Total Debt and total equity. Long Term Leverage Ratio (LTLR) = Long term debt/total debt and total equity. Short Term Leverage Ratio (STLR) = Short Term Debt/Total Debt and Total Equity. Profitability (PROF) = Earnings before Interest and Tax/Total Assets. Assets Tangibility (TANG) = Fixed tangible assets/total assets. Growth opportunities (GO) =Percentage change in the log of total assets. RISK = Standard deviation of the earnings before interest and tax to total asset.Size (SIZE) = Natural logarithm of total assets. Dividend pay-out= Dividend/Profit after tax. AGE = number of years on the Nigerian Stock Exchange.

Table 4. Correlation matrix

Table 4 shows the pairwise correlation between leverage ratios (short term, long term and total) and firm specific variables (Age, growth opportunities, Asset tangibility, profitability, size dividend and risk ).The correlation results does not reveal any high level of correlation among the variables therefore there is no problem of multicollinearity.

STLRT-1

0.2825 (0.000)***

LTLRT-1

0.2003 (0.000)***

TLRT-1

0.3146 (0.000)***

Prof

0.3737 (0.001)***

0.1832 (0.001)***

0.5413 (0.001)***

Tang

-0.1532 (0.001)***

-0.1777 (0.001)***

-0.2994 (0.000)***

Go

-0.0068 (0.000)***

-0.0065 (0.000)***

-0.0117 (0.000)***

Risk

0.0015 (0.000)***

0.0003 (0.000)***

0.0012 (0.000)***

Div

0.0194 (0.002)***

-0.0046 (0.000)***

0.0122 (0.000)***

Size

-0.0004 (0.000)***

0.0068 (0.000)***

0.0057 (0.000)***

Age

-0.0152 (0.000)***

-0.0053 (0.000)***

-0.0188 (0.000)***

Arrelano and Bond AR(2) Probability

0.503

0.470

0.437

J-Statistic

93.217

94.331

96.51

Total Leverage Ratio (TLR) = total debt/Total Debt and total equity. Long Term Leverage Ratio (LTLR) = Long term debt/total debt and total equity. Short Term Leverage Ratio (STLR) = Short Term Debt/Total Debt and Total Equity. Profitability (PROF) = Earnings before Interest and Tax/Total Assets. Assets Tangibility (TANG) = Fixed tangible assets/total assets. Growth opportunities (GO) =Percentage change in the log of total assets.

Dividend= RISK = Standard deviation of the earnings before interest and tax to total asset.Size (SIZE) = Natural logarithm of total assets. Dividend pay-out= Dividend/Profit after tax. AGE = number of years on the Nigerian Stock Exchange.

Table 5. Two step system Generalized method of moments results Short Leverage ratio(Stlr) Long term Leverage ratio(Ltlr) Total Leverage ratio (Tlr)

85

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Signficance level 10% * 5% **1%*** Table 5 above presents the results of three regression models. Model 1-3 is the regression with short term leverage, long term leverage and total leverage as the dependent variables respectively. The results in Table 5 indicate that the relationships between the lag leverage ratios (short term, long term and total leverage) and current leverage is positive and significant for all three measures of leverage.

This indicates that as immediate year financing of firms increases, the current leverage also increases.

This suggests that the capital structure choice (short term, long term and total leverage) firms make in previous year influences their current year capital structure choice. The positive significant coefficient confirms the presence of optimal capital structure of firms which supports the prediction of the agency cost theory of the existence of optimal capital structure that firms strive to achieve. This suggests that firms in Nigeria adjust their capital structure to reach a target level.

The results for all the three measures of leverage indicate that profitability is positive and significantly related to leverage. The results imply that as firms become more profitable they tend to employ more debt in their capital structure coupled with the fact that they have capacity to repay as evidenced in their profitability. The positive relationship also supports the perspective that managers tend to use more debt as they prefer higher free cashflow because it facilitates the consumptions of perks. The positive relationship is consistent with the expected theoretical positive relationship between profitability and leverage posits by the agency cost theory of capital structure. The finding complies with the positive relationship between profitability and leverage reported in past studies such as Qian and Wirjanto (2009) and Chakraborty (2010) for emerging market. Our study accept the hypothesis (H1)and suggests that there is a significant positive relationship between leverage and firm profitability.

The result from Table 5 above also shows negative but statistically insignificant relationship between tangibility and leverage for short and total leverage ratio. However, the finding indicates statistically significant positive relationship between asset tangibility and long term leverage ratio. This conforms to the theoretical prediction of the agency cost theory. This suggests that firms with tangible asset tend to use more long term debt and that there is a positive relationship between the tangible assets and leverage level. This is because the tangible asset may serve as collateral to secure the debt in the event of bankruptcy. Whereas creditors may not consider tangible assets for short term borrowing. The use of long term debt may reduce the opportunistic behaviour of managers. Managers may strive to ensure debt repayment promptly to avoid bankruptcy which can be very costly for the firm, and managers may lose their job and reputation in a situation where the firm ceases to exist due to the inability of the firm they manage to meet up with debt obligations or the firm’s assets were taken over by the debt holders as a result of non repayment of debt. The finding conforms with the reported results in Salawu and Agboola (2008), Abor and Biekpe (2009) for countries in emerging market. Our findings therefore supports H2 of a positive relationship between tangibility and leverage.

The result in Table 5 further shows that there is a statistically significant negative relationship between leverage and growth opportunities for short term leverage and total leverage ratio but we found a statisitically significant positive relationship between growth opportunities and long term leverage. The negative result support the theoretical prediction of the agency cost theory and indicates that managers would prefer the use of less debt in firms where there are high growth opportunities. Managers of firms may prefer to employ less debt when they see that debt can restrict them to explore future opportunities because of the commitment and covenants associated with debt. The negative finding conforms with the reported findings in studies such as Salawu and Agboola, 2008; Qian, Tian and Wirjanto, 2009; Akinlo, 2011. On the other hand, the positive relationship reported in this study between long term 86

EJAE 2018 ď&#x201A;˛ 15 (2) ď&#x201A;˛ 74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L. ď&#x201A;˛ CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

leverage ratio and growth opportunities supports similar findings reported by De Miguel and Pindado, 2001; Salawu, 2007; Karadeniz, 2008; Shehu, 2011; Chandrasekharan, 2012. These findings reflect the mixed findings in existing literature.

The results presented in Table 5 also show statisitically significant positive relationship between firm risk and leverage. This suggests that firms use more debt when the level of volatility of earnings is high. Therefore high earning volatility is associated with high firm leverage. Though, earnings volatility is associated with low long term debt. This finding is contrary with reported findings from Qian and Wirjanto (2009), Sheik and Wang (2011), Qian et al. (2008), Drobetz et al. (2013) that have reported negative relationship between firm risk and leverage. Based on the positive relationship found between risk and leverage the study accept hypothesis H4.

Furthermore, our empirical results indicate positive relationship between short term leverage, total leverage and dividend. While long term indicates statistically significant negative relationships with dividend. These suggest that firms tend to employ more short term debt when they have less dividend obligations therefore owners may will be able to use short term debt to mitigate the agency problem at the firm level. However, the results indicate positive relationship between long term. This suggests that dividend obligation of firms make them use more debt in their capital structure such that debt can be used to mitigate the agency problem at the firm level. Based on these findings the study thus accepts H5.

Table 5 also shows that the relationship between firm size and leverage is significantly negative with short term debt. These results indicate that larger firms tend to use more debt than equity compared to smaller firms. The positive relationship conforms with the findings of studies such as Haung and Song (2008), Salawu and Agboola (2008), Qian et al. (2008), Abor and Biekpe (2009), Qian and Wirjanto (2009), Sheik and Wang (2011), Akinlo (2011) and Michalca (2011) that have reported positive relationship between firm size and leverage. However, in this investigation the relationship is statistically significant therefore H6 is rejected. In the same vein, the study rejects H7 based on the negative relationship between age and leverage (total leverage ratio). This signifies that firms that have been in existence for long time may have built reputation especially debt repayment reputations to access debt for the purpose of mitigating the opportunistic behaviours of managers. The results from Table 5 above suggest that firm specific factors are crucial in the analysis of capital structure of firms in Nigeria.

CONCLUSION This paper presents an analysis of the firm specific determinants of capital structure choices of Nigerian firms based on the data on 115 Nigerian non-financial firms listed on the Nigerian stock exchange, for the period 1998-2016 in a dynamic panel framework. Findings from the study reveal positive relationship between leverage and profitability and firm risk and firm dividend. Asset tangibility, growth opportunities, size and age are found to be negatively related to leverage. The study therefore concludes that variables identified in the agency cost theory that provide explanation for capital structure of firms in developed and some emerging countries, are relevant but not fully portable in the Nigerian context.

Some of the variables stand as factors driving capital structure of firms in Nigeria as predicted by the agency cost theory but several other factors identified by the agency cost theory do not provide direct explanation for capital structure choice of firms in Nigeria as posits in the theoretical prediction of the agency cost theory. It is against this backdrop that this study suggests that further studies that capture the role of country factors alongside firm specific factors as determinants of capital structure of firms in emerging markets seems important to better appreciate the impacts of institutional factors on firm behaviour including their capital structure decisions.

87

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

This study show that managers tend to use more debt as they prefer higher free cashflow because it facilitates the consumptions of perks. The use of long-term debt may reduce the opportunistic behaviour of managers. Managers may strive to ensure debt repayment promptly to avoid bankruptcy which can be very costly for the firm, and managers may lose their job and reputation. Managers would prefer the use of less debt in firms where there are high growth opportunities. Managers of firms may prefer to employ less debt when they see that debt can restrict them to explore future opportunities because of the commitment and covenants associated with debt.

REFERENCES Abor, J., & Biekpe, N. (2009). How do we explain the capital structure of SMEs in sub-Saharan Africa? Evidence from Ghana. Journal of Economic Studies, 36(1), 83-97.

Adelopo, I. (2011). Voluntary disclosure practices amongst listed companies in Nigeria. Advances in Accounting, 27(2), 338-345.

Agrawal, A., & Jayaraman, N. (1994). The dividend policies of all‐equity firms: A direct test of the free cash flow theory. Managerial and Decision Economics, 15(2), 139-148.

Akinlo, O. (2011). Determinants of capital structure: Evidence from Nigerian panel data. African Economic and Business Review, 9(1), 1-16.

Alves, P., Couto, E.B., & Francisco, P. M. (2015). Board of directors’ composition and capital structure. Research in International Business and Finance, 35, 1-32.

Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies, 58(2), 277-297.

Barine, M.N. (2012). Capital structure determinants of quoted firms in Nigeria and lessons for corporate financing decisions. Journal of Finance and Investment Analysis, 1(2),61-81.

Belkhir, M., Maghyereh, A., & Awartani, B. (2016). Institutions and Corporate Capital Structure in the MENA Region. Emerging Markets Review, 26, 99-127. DOI: 10.1016/j.ememar.2016.01.001 Bokpin, G. A. (2009). Macroeconomic development and capital structure decisions of firms: Evidence from emerging market economies. Studies in Economics and Finance, 26(2), 129-142.

Byoun, S. (2008). How and when do firms adjust their capital structures toward targets? The Journal of Finance, 63(6), 3069-3096.

Chakraborty, I. (2010). Capital structure in an emerging stock market: The case of India. Research in International Business and Finance, 24(3), 295-314.

Chandrasekharan, C.V. (2012). Determinants of capital structure in the Nigerian listed firms. International Journal of Advanced Research in Management and Social Sciences, 1(2), 108-133.

Chang, C., Chen, X., & Liao, G. (2014). What are the reliably important determinants of capital structure in China? Pacific-Basin Finance Journal, 30, 87-113. DOI:10.1016/j.pacfin.2014.06.001 Cotei, C., Farhat, J., & Abugri, B. (2011). Testing Trade-off and Pecking Order Theories: Does Legal System Matter? Managerial Finance, 37(8), 53-69.

Deesomsak, R., Paudyal, K., & Pescetto , G. (2004). The determinants of capital structure: evidence from the Asia Pacific region. Journal of Multinational Financial Management, 14(4), 387-405.

Drobetz, W., Gounopoulos, D., Merikas, A., & Schroder, H. (2013). Capital structure decisions of globally-listed shipping companies. Transportation Research Part E: Logistics and Transportation Review, 52, 49-76.

Fama, E.F., & Miller, M.H. (1972). The theory of finance (Vol. 3). Hinsdale, IL: Dryden Press.

Florackis, C., Kanas, A., & Kostakis, A. (2015). Dividend policy, managerial ownership and debt financing: A non-parametric perspective. European Journal of Operational Research, 241(3), 783-795.

88

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Fosu, S. (2013). Capital structure, product market competition and firm performance: Evidence from South Africa. The Quarterly Review of Economics and Finance, 53(2), 140-151.

Frank, M.Z., & Goyal, V.K. (2009). Capital structure decisions: which factors are reliably important? Finance Management, 38(1), 1-37. DOI:10.1111/j.1755-053X.2009.01026.x Gwatidzo, T., & Ojah, K. (2014). Firms’ debt choice in Africa: Are institutional infrastructure and non-traditional determinants important? International Review of Financial Analysis, 31, 152-166. DOI:10.1016/j. irfa.2013.11.005 Haron, R. (2014). Capital structure inconclusiveness: evidence from Malaysia, Thailand and Singapore. International Journal of Managerial Finance, 10(1), 23-38.

Handoo, A., & Sharma, K. (2014). A study on determinants of capital structure in India. IIMB Management Review, 26(3), 170-182. DOI:10.1016/j.iimb.2014.07.009 Jensen, M.C. (1989). Active investors, LBOs, and the privatization of bankruptcy. Journal of Applied Corporate Finance, 2(1), 35-44.

Jensen, M.C., & Meckling, W.H. (1976). Theory of the firm: managerial behaviour, agency costs and the ownership structure. Journal of Financial Economics, 3(4), 305-360. DOI:10.1016/0304-405X(76)90026-X Jensen, G. R., Solberg, D. P., & Zorn, T. S. (1992). Simultaneous determination of insider ownership, debt, and dividend policies. Journal of Financial and Quantitative Analysis, 27(02), 247-263.

Jõeveer, K. (2013). What do we know about the capital structure of small firms? Small Business Economics, 41(2), 479-501.

Jõeveer, K. (2013). Firm, country and macroeconomic determinants of capital structure: Evidence from transition economies. Journal of Comparative Economics, 41(1), 294-308.

Jong, A., Kabir, R., & Nguyen, T. T. (2008). Capital structure around the world: The roles of firm-and countryspecific determinants. Journal of Banking & Finance, 32(9), 1954-1969. DOI:10.1016/j.jbankfin.2007.12.034 Karadeniz, E., Kandir, Y., Balcilar, M., & Onal, Y. (2009). Determinants of capital structure: evidence from Turkish lodging companies. International Journal of Contemporary Hospitality Management, 21(5), 594-609.

Kayo, E.K., & Kirma, H. (2011). Hierarchical determinants of capital structure. Journal of Banking & Finance, 35(2), 358-371. DOI: 10.1016/j.jbankfin.2010.08.015 Lin, C., Ma, Y., Malatesta, P., & Xuan, Y. (2013). Corporate ownership structure and the choice between bank debt and public debt. Journal of Financial Economics, 109(2), 517-534.

Margaritis, D., & Psillaki, M. (2007). Capital structure and firm efficiency. Journal of Business Finance & Accounting, 34(9‐10), 1447-1469.

Matemilola, B.T., Bany-Ariffin, A.N., & Azman-Saini, W.N.W. (2012). Financial leverage and shareholder’s required returns: evidence from South Africa corporate sector. Transition Studies Review, 18(3), 601-612.

Modigliani, F., & Miller, M. (1958). The cost of capital, corporation finance and the theory of investment. The American Economic Review, 48(3), 261-297.

Modigliani, F., &Miller, M. (1963). Corporate income taxes and the cost of capital: a correction. The American Economic Review, 53(3), 433-443.

Ojah, K., & Gwatidzo, T. (2009). Corporate capital structure determinants: evidence from five African countries.

The African Finance Journal, 11(1), 1-23.

Onofrei, M., Tudose, M. B., Durdureanu, C., & Anton, S. G. (2015). Determinant Factors of Firm Leverage: An Empirical Analysis at Iasi County Level. Procedia Economics and Finance, 20, 460-466. DOI:10.1016/ S2212-5671(15)00097-0 Qian, Y., Tian, Y., & Wirjanto, T. S. (2009). Do Chinese publicly listed companies adjust their capital structure toward a target level? China Economic Review, 20(4), 662-676.

Rajan, R.G., & Zingales, L. (1995). What do we know about capital structure? Some evidence from international data.The Journal of Finance, 50(5), 1421-1460.

89

EJAE 2018  15 (2)  74-90

GANIYU, Y. O., ADELOPO, I., RODIONOVA, Y. SAMUEL, O. L.  CAPITAL STRUCTURE IN EMERGING MARKETS: EVIDENCE FROM NIGERIA

Roodman, D. (2009). Estimating fully observed recursive mixed-process models with cmp. Retrieved from https:// www.cgdev.org/publication/estimating-fully-observed-recursive-mixed-process-models-cmp-workingpaper-168 Salawu, R.O., & Agboola, A.A. (2008). The determinants of capital structure of large non-financial listed firms in Nigeria. The International Journal of Business and Finance Research, 2(2), 75-84.

Shehu, U. H. (2011). Determinants of Capital Structure in the Nigerian Listed Insurance Firms. China-USA Business Review, 10(12), 81-98.

Sheikh, N. and Wang, Z. (2011). Determinants of capital structure: An empirical study of firms in manufacturing industry of Pakistan. Managerial Finance, 37(2), 117-133.

Yeh, H.H., & Roca, E. (2010). Macroeconomic conditions and capital structure: Evidence from Taiwan. Discussion papers in Finance, 2010-14. Griffith University, Department of Accounting, Finance and Economic Zou, H., & Xiao, J.Z. (2006). The financing behaviour of listed Chinese firms. The British Accounting Review, 38(3), 239-258.

STRUKTURA KAPITALA NA TRŽIŠTIMA U RAZVOJU: DOKAZI IZ NIGERIJE

Rezime: Ovaj rad ispituje uticaj specifičnih faktora preduzeća kao determinante izbora strukture kapitala firmi iz Nigerije, na osnovu podataka od 115 nefinansijskih firmi navedenih na nigerijskoj berzi u periodu od 19982016. U radu je primenjen dvostepeni sistem generalizovane metode momenata u dinamičkom panelu. Rezultati ispitivanja pokazuju pozitivan odnos između profitabilnosti, čvrstog rizika, čvrste dividende i leveridža. Utvrđeno je da su opipljivost imovine, mogućnosti za napredovanje tj. razvoj, veličina i starost preduzeća negativno povezani sa leveridžom. Studija stoga zaključuje da su varijable identifikovane u teoriji troškova agencije koje pružaju objašnjenja za kapitalnu strukturu firmi u razvijenim i nekim zemljama u razvoju, relevantne ali nisu u potpunosti primenjive u nigerijskom kontekstu.

Ova studija pokazuje da rukovodioci imaju tendenciju da više koriste zaduživanje jer preferiraju slobodni tok gotovine što olakšava potrošnju bonusa. Korišćenje kredita sa dugim periodom otplate može smanjiti oportunističko ponašanje menadžera. Rukovodioci mogu nastojati da osiguraju otplatu duga, kako bi se izbegao stečaj koji može biti vrlo skup za firmu, dok menadžeri mogu izgubiti svoj posao i reputaciju. Menadžeri bi voleli da manje koriste zaduživanje u preduzećima gde postoje velike mogućnosti za rast. Menadžeri firmi radije izbegavaju zaduživanja kada vide da ih dug može ograničiti u istraživanju i iskorišćavanju budućih prilika, zbog obaveza koje zaduživanje povlaći sa sobom.

90

Ključne reči: struktura kapitala, teorija agencije, generalizovani metod momenata, dinamički panel.

CIP - Каталогизација у публикацији Народна библиотека Србије, Београд 33 The EUROPEAN Journal of Applied Economics / editor-in-chief Nemanja Stanišić. Vol. 12, No. 1 (2015)- . - Belgrade : Singidunum University, 2015- (Loznica : Mobid). - 28 cm Dva puta godišnje. - Је наставак: Singidunum Journal of Applied Sciences = ISSN 2217-8090 ISSN 2406-2588 = The European Journal of Applied Economics COBISS.SR-ID 214758924