European Journal - Vol 12 No 2 by Singidunum University

Vol. 12 No. 2

w w w.j our nal.sing idunum.ac .rs

Vol. 12 No. 2 Publisher: Singidunum University

Editorial Board

Professor Milovan Stanišić, Singidunum University mstanisic@singidunum.ac.rs Emeritus Slobodan Unković, Singidunum University sunkovic@singidunum.ac.rs Professor Francesco Frangialli, UNWTO frangialli@gmail.com Professor Gunther Friedl, Technische Universität 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.a.crs Professor Jovan Popesku, Singidunum University jpopesku@singidunum.ac.rs Professor Zoran Jeremić, Singidunum University zjeremic@singidunum.ac.rs Associate Professor Christine Juen, Austrian Agency for International christine.juen@oead.at Mobility and Cooperation in Education, Science and Research, Wien Associate Professor Anders Steene, Södertörn University, Stockholm/Hudinge anders.steene@sh.se Associate Professor Ing. Miriam Jankalová, University of Žilina miriam.jankalova@fpedas.uniza.sk Associate Professor Bálint Molnár, Corvinus University of Budapest molnarba@inf.elte.hu Associate Professor Vesna Spasić, Singidunum University vspasic@singiduunm.ac.rs Associate Professor Michael Bukohwo Esiefarienrhe, University of Agriculture, esiefabukohwo@gmail.com Dept. of Maths/Statistics, Markurdi Assistant Professor Patrick Ulrich, University of Bamberg patrick.ulrich@uni-bamberg.de Assistant Professor Konstadinos Kutsikos, University of the Aegean, Chios kutsikos@aegean.gr Assistant Professor Theodoros Stavrinoudis, University of the 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 Dr. Aleksandar Lebl, Research and Development Institute for Telecommunications and Electronics, Belgrade lebl@iritel.com Editorial Office

Editor-in-Chief: Deputy Editor: Managing Editor: Production Editor: Prepress: Novak Njeguš

Professor Milovan Stanišić, Singidunum University Professor Verka Jovanović, Singidunum University Professor Svetlana Stanišić Stojić, Singidunum University Professor Gordana Dobrijević, Singidunum University Design: Aleksandar Mihajlović

sstanisic@singidunum.ac.rs gdobrijevic@singidunuma.c.rs

ISSN: 2406-2588 The European Journal of Applied Economics is published biannually. Contact us: Scan this QR code The European Journal of Applied Economics for Website. 32 Danijelova Street, 11010 Belgrade, Serbia Phone No. +381 11 3093236, +381 11 3093219, Access to full text articles: Fax. +381 11 3093294 Singipedia (www.singipedia.com), E-mail: journal@singidunum.ac.rs SCindeks (www.scindeks.ceon.rs) and EBSCO (www.ebscohost.com). Web: www.journal.singidunum.ac.rs Copyright © 2015 Printed by: Mobid, Loznica

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 1-8 9 - 15 16 - 26 27 - 34

Application of earned value based metrics on small-scale construction projects Radoslav Avlijaš, Goran Avlijaš, Milenko Heleta

The effect of incentive return-to-work policies on single-parent families: A comparative approach Samir Amine, Pénayori Ouattara Stated preferences for improved air quality management in the city of Nairobi, Kenya Hilary Ndambiri, Eric Mungatana, Roy Brouwer

Offer of authentic food as a condition for gastronomic tourism development Bojana Kalenjuk, Dragan Tešanović, Snježana Gagić, Irma Erdeji, Maja Banjac III

35 - 43 44 - 52 53 - 58

De�ining second-tier audit �irms in Serbia: In search of the Small four Nemanja Stanišić, Vule Mizdraković, Tijana Radojević, Nenad Stanić The role of business intelligence in decision process modelling Višnja Istrat, Sanja Stanisavljev, Branko Markoski Instructions for authors

EJAE 2015, 12(2): 1-8 ISSN 2406-2588 UDK: 005.8:728.224(497.11)"2002/2014" 005.52:330.133.1 DOI: 10.5937/ejae12-8515 Original paper/Originalni naučni rad

APPLICATION OF EARNED VALUE BASED METRICS ON SMALL-SCALE CONSTRUCTION PROJECTS Radoslav Avlijaš*, Goran Avlijaš, Milenko Heleta Singidunum University, 32 Danijelova St., Belgrade, Serbia

Abstract: The application of earned value based metrics in Serbia lags significantly behind other developed economies, regardless of the fact that is has proven to be a very useful and flexible tool for prediction of total cost and time of construction projects. This paper aims to contribute to the wider use of earned value management (EVM) as a practical metrics that can be used to track performance of small-scale construction projects. Literature review reveals how the method evolved as a fundamental tool for construction projects monitoring and refers to specific issues concerning its implementation. In order to enable practitioners to realize all advantages of earned value based performance monitoring of small-scale construction projects, the applicability of such methodology is demonstrated through a case study.

INTRODUCTION The main objective of project management is to fulfill all project requirements within budget forecasts and the planned time-line (Avlijaš & Avlijaš, 2010). In order to achieve this goal, project team has to devote significant managerial efforts. One of the most important steps in this process is development of the effective performance monitoring system. In that sense, performance monitoring primarily relates to monitoring of cost, time and quality of the construction project. Regardless of how carefully the project plan was developed, it is not possible to assess the project progress and effectiveness of execution, as well as performance of the entire construction project, without timely and regular performance control (Yang et al., 2010). Project teams strive to control three key performance indicators (KPIs) related to construction projects: cost, time and scope. These KPIs constitute the “triple constraint” of any project, and timely * E-mail: ravlijas@singidunum.ac.rs

Key words: earned value, project management, construction.

monitoring of these values enables project teams to maintain the required level of construction project quality. Namely, project monitoring is an iterative process, during which the actual values are compared to the planned values in order to predict the overall project cost and time and undertake any preventive and corrective measures based on these predictions (PMI, 2013). One of the biggest challenges related to constructive project monitoring is the selection of the appropriate monitoring methodology. In fact, project teams on most of construction projects in the world use the simple comparison of planned to actual costs. This approach does not take into account the value of the work performed, which omits the third dimension of costs: the earned value of the work (Fleming & Koppelman, 2003). The monitoring methodology that enables integration of the work performed with the planned and actual project time and costs, and quantification of project technical performance is Earned Value Management (EVM). 1

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

EVM methodology has been extensively applied in the field of military industry, energy and software engineering, and its use is increasingly growing (Kim et al., 2003). However, the implementation of this very useful and effective control tool in the construction industry lags behind other industries, especially when it comes to small scale projects. The reasons for poor implementation of EVM methodology in the construction industry in European region are not yet fully known, since there are very few academic papers dealing with this issue. Regarding that, it is believed that more significant academic effort should contribute to wider adoption of EVM by practitioners and civil engineers (De Marco & Narbaev, 2013). This paper aims to reduce the gap between theory and practice and contribute to the wider use of EVM methodology in the construction industry in Serbia. In addition, by using a short case study, the paper demonstrates the application of the methodology on a small-scale construction project. The paper is organized as follows. The following section points out to the importance of construction projects efficiency within the global and local economy. The third section provides a short presentation of EVM methodology, which covers the basic concepts, as well as the main advantages and disadvantages. Finally, after demonstration of EVM methodology on a small scale construction project, the paper concludes with a short discussion of the research findings and conclusions. (IN)EFFICIENCY OF CONSTRUCTION PROJECTS Even though the area of construction management has been widely explored for decades, construction industry worldwide is still considered very inefficient. Today, along with the rapid development of technology, higher quality standards of the final product and

increasing competition, there are many obstacles to achieving the project goals. Construction industry performance seems to be considerably lagging behind other industries. Competitive, political and economic pressures contained in the industry, call for changes in the implementation of construction projects. Looking from the aspect of the local economy, the future of the construction industry does not seem promising at the moment. In the period from 2006 to 2011, Serbian construction industry expected annual growth of 15-20%, and a significant increase in construction work was recorded both in the domestic and export markets (Trbović et al., 2007). Sales in 2006 were 120% higher than in 2002, while the value of construction works abroad in 2006 totaled $ 138 million, which is an increase of 64% compared to 2005 (Trbović et al., 2007). However, due to the global financial crisis, the situation in the construction industry worldwide rapidly deteriorated and negative effects did not bypass Serbian companies. As can be seen in Figure 1, the value of construction works significantly declined (Statistical Office of the Republic of Serbia, 2015). According to the official data, 2,000 Serbian construction companies were shut down in 2008, and about 20,000 workers in the construction sector lost their jobs. Competitive pressure has reached a point where only an extremely efficient and profitable enterprise can ensure itself a long-term survival and development. According to the Serbian Chamber of Commerce, construction industry in Serbia accounts for about 11,530 companies and 116,760 workers. Construction firms participate in the total number of companies and in the total number of employees with 13.1% and 11.8%, respectively (Serbian Chamber of Commerce, 2015). Having in mind that construction industry in Serbia supplies more than 30 other

Figure 1. Total value of executed construction works in the Republic of Serbia (in thousands) 2

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

industries with different inputs, it can be concluded that the performance of construction companies and their projects have a significant impact on the national economy (Mladenović, 2013). Serbian construction companies are usually less efficient than foreign companies, as they employ too many workers (Trbović et al., 2007) and domestic laws cannot cope with numerous challenges of modern business (Mladenović, 2013). Poor efficiency is primarily related to significant total project time and cost overrun. In addition to the known threats in the form of technical and external risks, there is an increasing number of recent projects being unsuccessful due to different types of organizational risks. By studying the organizational risks, practitioners report the low level of management skills and insufficient use of modern methods and information systems in the project planning and monitoring phase, as one of the most common causes of failure to meet deadlines and budgets. On a global scale, exceeding of the planned cost represents a trend that has not changed for almost 70 years (Flyvbjerg et al., 2002). According to a survey encompassing 258 capital projects in 20 countries worth a total of $ 90 billion, 9 out of 10 projects recorded cost overruns (average overrun of 28%). The study included America, Europe, Japan and other countries where the average cost overrun was 64%. The results show that 44% of liability is applied to contractors, 24% to investors, 14% to legislature, shared responsibility was observed in 12% of cases, while responsibility applied to consultants was observed in 6% of cases. The most alarming is the fact that 48% of costs were justifiedpayable, 8% justified - unpayable, and 44% completely unjustified (Flyvbjerg et al., 2002). Different types of construction projects in Saudi Arabia were examined in order to determine significant causes of delay. The survey included 15 owners, 23 contractors and 19 consultants, and the results showed that the average of time overrun ranges between 10% and 30% of original duration. The survey concluded that 70% of projects experienced time overrun and found that 45 out of 76 projects considered were delayed (Assaf & Al-Hejji, 2006). Doloi et al. (2012) used regression modeling to examine the critical factors that can affect cost performance in both - the preconstruction and construction phases of Indian projects. Some of the significant factors were: accurate project planning and monitoring, contractors’ efficiency, effective site management, design and communication efficiency on construction projects.

In most of the cases, construction project planning is performed by means of a rough estimate of time frames and budgets, and unidentified risks always affect the degree of accomplishment of project requirements. Practice has shown that monitoring and control of project execution often involve a simple comparison of actual time and costs with the planned values, which does not take into account the actual value of the works completed to a certain point and does not provide a realistic picture of the time and cost of performance. No matter if it is the domestic or foreign market, it is clear that construction companies must adopt a more adequate approach to project management that will be able to respond to all challenges of the constantly changing environment. BENEFITS AND ISSUES IN EVM APPLICATION The Earned value concept was developed in 1967 by the Ministry of Defense of the United States, first as a system of 35 criteria for cost and time control (Cost/Schedule Control Systems Criteria - C/SCSC). At the very beginning, C/SCSC was considered exclusively as a tool for financial control, which could be used in project management. However, in 1989, the Ministry of Defense adopted the criteria system as an official tool for managing programs and procurement activities, and since the mid-nineties, the system has recorded a wider application in public administration (energetics, information technology, space exploration, construction, etc.) and got the new name – Earned Value Management. Earned Value Management (EVM) is one of the most significant methods used in the process of monitoring and control of project performance (PMI, 2011). The original concept of earned value was derived from the industrial engineering methods, which in detail analyzed “cost-performance” approach. In the process of cost-effectiveness assessment, the comparison of planned and actual values of time and cost was carried out. Over the time, the application of this methodology was not limited only to government industrial and military projects, but is still relatively poorly applied to commercial and private projects (Kwak & Anbari, 2012). The importance of the Earned Value Management is that it provides an accurate measure of the cost and time performance in the process of project monitoring and control. Scientific studies conducted by Christensen and his associates in the period 1998-2002, show that the CPI indicator measured at the 15-20% 3

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

of project completion, provides a sufficiently reliable measure for prediction of total project cost with a maximum error of ± 10% (Christensen, 1998, 1999; Christensen & Heise, 1993; Christensen & Templin, 2002). In this regard, the EVM may constitute a means of early warning at just 20% of project completion. Nevertheless, the application of the EVM methodology in the construction industry is still at a very low level (De Marco & Narbaev, 2013). Some of the most common problems in the application of EVM methodology that practitioners state include: the necessity of a detailed plan before the project starts, reliable performance measure and reporting, difficulties in measuring actual physical progress of construction activities, and accurate input data (Wilson et al., 2013). Despite the fact that there are many arguments that the EVM is difficult to apply to small scale construction projects, it has become increasingly accepted as an important and effective method for monitoring various types of projects, regardless of the project size or risk. Therefore, the key factors of successful implementation of EVM methodology are the choice of the right format and its adaptation to monitoring and control of a specific project. The Earned value analysis is based on the three key values: Budgeted Cost of Work Scheduled (BCWS), Budgeted Cost of Work Performed (BCWP), also known as Earned Value (EV) and Actual Cost of Work Performed (ACWP). The fourth extra value refers to the total planned budget of the project (Budget at Completion), which represents the BCWS of the entire project. Based on these values, variances were derived, i.e. the difference between the planned and

actual values, which can be used as indicators of performance and a tool for predicting the total project time and cost. BCWS values for each project activity serve as the basis as BCWP and ACWP values are compared throughout the project’s life cycle. The difference between planned and actual values is measured using two variances that provide accurate positive or negative image of financial project performance. Cost variance (CV) represents a measure of budgetary conformance of AC and is calculated as the difference between the earned value and actual cost (CV=EV-ACWP). Schedule variance (SV) represents the difference between earned and planned value (SV=EV-BCWS). Overall, the positive values of variances mean that the project is implemented within the budget and planned time frame, while negative values mean that the project is over budget and behind schedule. On the other hand, the performance indicators do not provide a monetary value, but can be used as indicators of actual performance. They represent ratios of planned and earned values in case of cost performance index, and actual and earned values in case of schedule performance index. Cost performance index (CPI) can be used to estimate the project cost (CPI=1 means that project costs just as much as planned; CPI<1 project is over budget; CPI>1 project costs less than planned). Similarly, the schedule performance index (SPI) can be used to estimate the project completion time (SPI=1 means that the project is executed in accordance with the time schedule; SPI>1 the project is executed faster than planned; SPI<1 project is late compared to the schedule).

Figure 2. Variances and indexes used in Earned Value Management 4

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

Calculation of two variables - Estimate at Completion (EAC) and the Time Estimate at Completion (TEAC) is a precondition for project execution in a planned way. Already described indicators, CPI and SPI, are used for calculation of the EAC and TEAC values, which represent the projection of the current project progress to the rest of the project execution. The two most common ways of calculating the EAC and TEAC values are: EAC=BAC/CPI and TEAC=D/SPI, where BAC represent the total projected cost of the project, and D total scheduled time of project. The described indicators are shown in Figure 2.

Although these measures are generally accepted as indicators of the overall project costs and finish time, they have often been the subject of criticism, mostly because of the reliability of the CPI and SPI indicators. Some of the criticism includes: the final cost variance is always greater than its value at 20% project completion; cost performance index value does not change by more than 10% compared to its value at 20% of project completion and is usually negative. Finally, the total estimated cost based on the CPI is usually less than actual at the end (Christensen, 1999). This criticism resulted in adjusting the formula for calculating the total cost of the project, so that EAC=BAC/(CPI @ 20% ± 0.1). The issues related to predicting the completion time of the project are usually related to the variance SV, and its SPI indicator. In essence, earned value methodology defines time indicators indirectly through monetary or cost value. Accordingly, it is not possible to unambiguously assess the temporal progress of the project. Problems arise when specific project activities have parallel execution rather than in the order. In such situations, certain non-critical activities can be delayed, and critical activities can be executed faster than planned, which altogether may lead to erroneous conclusions (Russell, 2008). Another drawback associated with the time estimate is that with the progress of execution schedule variance tends to zero, and the schedule performance index to one. This indicates that project execution is on time, although it may be delayed in reality. In this sense, we can say that after a certain percentage of project execution, regardless of the project structure, the usability of SV and SPI indicators decreases significantly. This happens most often after larger part of the project (2/3) is completed, which means that those indicators can be applied with certainty in the period when 15-70% of the project is completed.

In order to cope with these problems, the Lipke (2003) has improved the traditional approach by introducing two additional variables: Earned Schedule (ES) and the Actual Time (AT). ES indicator is calculated by projecting the value of the BCWP on BCWS at a particular moment of realization (AT) the time when the current BCWP according to the plan should have been achieved. This moment can precede or follow AT, depending on whether the execution of the project is delayed or goes forward according to the plan. The formula for calculating the earned schedule is: EC=C+((BCWP(€)-BCWS(t))/ (BCWS(t+1)-BCWS(t)). The difference and the ratio of ES and AT provide two new indicators SV (t) and SPI (t), which can be used throughout the project’s life cycle. In this case, the total estimated time of the project EAC(t)=D/SPI(t) (Lipke, 2003). Several researchers have tried to test the existing approaches and find new ways to improve accuracy of predictions, such as development of statistical and fuzzy models. Vandevoorde and Vanhoucke (2006) have tested several existing methods and revealed that only the earned schedule showed satisfying and reliable results throughout the whole project duration. Ponz-Tienda et al. (2012) presented comprehensive and robust approach to fuzzy scheduling and earned value management of construction projects. Naeni et al. (2011) developed earned value model based on the fuzzy logic in order to develop and analyze the earned value indices, and cost and time estimates at completion under certain level of uncertainty. Narbaev and De Marco (2014) attempted to overcome limitations of the traditional EVM and proposed an earned schedule-based regression model to improve cost estimate at completion. Colin and Vanhoucke (2014) proposed statistical project control procedure which sets tolerance limits to improve discriminative power between progress situations that are either statistically likely or less likely to occur under the project baseline schedule. Lipke et al. (2009) also applied statistical methods to EVM and ES performance indexes and found the proposed forecasting method to be sufficiently reliable. CASE STUDY ANALYSIS This section provides a real case study which illustrates the way in which the earned value methodology, as described in the previous section, can be used as an effective tool for measuring the project performance and prediction of total cost and execution time of the 5

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

construction project. Particular residential/commercial construction project was chosen for demonstration purposes, since in the planning and execution phase, experienced project team fully implemented processes necessary for EVM application, which primarily involves the existence of appropriate organizations and reliable input data. The selected project involved the construction works on the construction of residential/commercial building in the municipality of Zemun, near Belgrade. Most of the work was carried out by the main contractor (including plumbing, electrical and HVAC system), while subcontractors were engaged in individual works such as groundwork, ceramic, floor, facade and paintwork. The total planned value

of the project amounted to € 5,282,117, with the total scheduled duration of 15 months, from March 2013 to May 2014. The project was undertaken under a fixed price (lump-sum) contract, and payments were done on a monthly basis upon completion and receipt of the situation. In order to enable the appropriate planning and monitoring of project execution using the EVM, the project was broken down into almost 900 single project activities in the five-level WBS diagram. Project activity or project phase was considered completed only when fully completed (100%). Based on the monthly progress reports, the project team calculated the key indicators of EVM - variance, performance indexes and assessed the overall costs and time. Table 1 presents

Mar, 2013

Apr, 2013

May, 2013

Aug, 2013

Nov, 2013

Feb, 2014

May, 2014

BCWS

110,228

497,774

1,052,478

2,332,164

3,485,056

4,721,903

5,282,117

BCWP

35,205

248,997

1,049,230

2,198,948

3,286,550

4,650,712

5,282,117

ACWP

140,285

513,197

1,085,507

2,289,053

3,382,895

4,790,741

5,450,125

-105,079

-264,200

-36,277

-90,105

-96,345

-140,029

-168,008

CPI

0.251

0.485

0.967

0.961

0.972

0.971

0.969

EAC

21,048,049

10,886,762

5,464,746

5,498,560

5,436,962

5,441,157

5,450,125

0.32

1.36

2.99

5.71

8.51

11.86

15.00

SV(t)

-0.681

-0.642

-0.006

-0.286

-0.490

-0.140

0.000

SPI(t)

0.319

0.679

0.998

0.952

0.946

0.988

1.000

EAC(t)

46.97

22.09

15.03

15.75

15.86

15.18

15.00

Table 1. Estimated total project costs and time using the earned value method

Figure 3. Earned schedule and time and cost estimate at complete 6

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

the summary of project results achieved during the execution phase. Table 1 reveals that the project was executed on time and over budget by around 170,000 EUR. It can also be seen that after 20% of the project execution (which in this case equals to three months), cost performance index (CPI) stabilizes between 0.96-0.97, which usually tends to unity due to additional eﬀorts to complete the project on time. In all intervals, Estimate at Completion (EAC) was calculated. After three months, 19.86% of the planned work was executed, the actual cost accounted to 20.55% for the total budget, and with 3.46% predicted budget exceedance. In this sense, the overall cost assessment ranged from 102.7% to 106.3% of the original budget, which was exceeded at the end by 3.18%. Similar to monitoring of costs performance, earned value was calculated for schedule indicators - ES and EAC (t). When calculating the value of ES for the first and last period, the value of C for the first period is equal to zero, and for the last is equal to 15, which is the total planned duration of the project. It can be seen in Figure 3 that the estimated time for project execution stabilizes between 15 and 16 months after three months, which proved to be very accurate in relation to the actual project duration. Besides that, estimated total costs at 20% of execution (€ 5,464,746) proved to be very reliable in relation to the total actual cost at the end of the project (€ 5,450,125). CONCLUSION This paper aimed to show the theoretical and practical dimension of performance measurement on a small-scale construction project using the earned value methodology, and thereby contribute to its wider practical application. In addition to presented significance of the method and numerous advantages described in the theoretical part, particular attention was drawn to certain issues and problems that may occur during application and potential ways in which these disadvantages can be overcome. This is primarily related to organizational prerequisites that must be provided before the methodology is applied, as well as certain issues and problems that can occur in forecasting of the total time of project execution. Application and adaptability of the EVM method are illustrated in a form of a case study, which demonstrated that the EVM represents a reliable means for total project cost and duration prediction, already in the early stages of project implementation. Regardless

of the type and size of the construction project, and with a well-structured plan and appropriate monitoring approach, the earned value method can serve as an early warning tool to project teams, showing that certain part of the project is not being executed according to plan, and thus prevent greater consequences in terms of sliding cost, time or scope. REFERENCES Assaf, S.A., & Al-Hejji, S. (2006). Causes of delay in large construction projects. International Journal of Project Management, 24(4), 349-357. doi:10.1016/j.ijproman.2005.11.010. Avlijaš, R., & Avlijaš, G. (2010). Upravljanje projektom. Beograd: Univerzitet Singidunum. In Serbian. Christensen, D.S., & Templin, C. (2002). EAC evaluation methods: do they still work. Acquisition Review Quarterly, 9, 105-116. Christensen, D.S. (1998). Cost and benefits of earned value management process. Acquisition Review Quarterly, 5, 372-386. Christensen, D.S. (1999). Using the earned value cost management report to evaluate the contractor’s estimate at completion. Acquisition Review Quarterly, 19, 283-296. Christensen, D.S., & Heise, S. (1993). Cost performance index stability. National Contract Management Journal, 25, 7-15. Colin, J., & Vanhoucke, M. (2014). Setting tolerance limits for statistical project control using earned value management. Omega, 49, 107-122. doi:10.1016/j.omega.2014.06.001. De Marco, A., & Narbaev, T. (2013). Earned value-based performance monitoring of facility construction projects. Journal of Facilities Management, 11(1), 69-80. doi:10.1108/14725961311301475. Doloi, H., Sawhney, A., Iyer, K.C., & Rentala, S. (2012). Analysing factors affecting delays in Indian construction projects. International Journal of Project Management, 30(4), 479-489. doi: 10.1016/j.ijproman.2005.11.010. Fleming, Q.W., & Koppelman, J.M. (2003). What’s your project’s real price tag. Harvard Business Review, 20-21. doi:10.1225/F0309C. Flyvbjerg, B., Holm, M.S., & Buhl, S. (2002). Underestimating costs in public works projects: Error or lie. Journal of the American Planning Association, 68(3), 279-295. doi:10.1080/01944360208976273. Kim, E., Wells, W.G., & Duffey, M.R. (2003). A model for effective implementation of Earned Value Management methodology. International Journal of Project Management, 21(5), 375-382. doi: 10.1016/j.ijproman.2005.11.010. Kwak, Y.H., & Anbari, F.T. (2012). History, practices, and future of earned value management in government: Perspectives from NASA. Project Management Journal, 43(1), 77-90. doi:10.1002/pmj.20272. 7

EJAE 2015  12 (2)  1-8

Avlijaš, R., Avlijaš, G., Heleta, M.  Application of earned value based metrics

Lipke, W. (2003). Schedule is different. Retrieved Jun 21, 2015, from http://www.earnedschedule.com/Docs/Schedule%20is%20Diﬀerent.pdf Lipke, W., Zwikael, O., Henderson, K., & Anbari, F. (2009). Prediction of project outcome: The application of statistical methods to earned value management and earned schedule performance indexes. International Journal of Project Management, 27(4), 400-407. doi:10.1016/j.ijproman.2008.02.009. Mladenović, I. (2013). Građevinarstvo kao industrijski sektor u funkciji oporavka privrede Republike Srbije. Ekonomski horizonti, 15(3), 245-256. doi:10.5937/ekonhor1303245M Naeni, L.M., Shadrokh, S., & Salehipour, A. (2011). A fuzzy approach for the earned value management. International Journal of Project Management, 29(6), 764-772. doi:10.1016/j.ijproman.2008.02.009. Narbaev, T., & De Marco, A. (2014). An earned schedulebased regression model to improve cost estimate at completion. International Journal of Project Management, 32(6), 1007-1018. doi:10.1016/j.ijproman.2008.02.009. Project Management Institute. (2011). Practice standard for earned value management. Newtown Square, PA: Project Management Institute. Project Management Institute. (2013). A guide to the project management body of knowledge. Newtown Square, PA: Project Management Institute. Ponz-Tienda, J. L., Pellicer, E., & Yepes, V. (2012). Complete fuzzy scheduling and fuzzy earned value management in construction projects. Journal of Zhejiang University SCIENCE A: Applied Physics & Engineering, 13(1), 56-68. doi:10.1631/jzus.A1100160.

Russell, S. (2008). Earned value management: uses and misuses. Air Force Journal of Logistics, 32(2), 85-91. Serbian Chamber of Commerce. (2015). Građevinarstvo, industrija građevinskog materijala i stambena privreda. Retrieved March 20, 2015 from http://www.pks.rs/PrivredaSrbije.aspx?id=7&p=2 Statistical Oﬃce of the Republic of Serbia. (2015). Total value of executed construction works. Retrieved Jun 21, 2015, from http://webrzs.stat.gov.rs/WebSite/ Trbović, A., Vonnegut, A., Brnabić, A., Valentine, S., & Seas, W. (2007). Konkurentni potencijal 11 grana srpske privrede. Ekonomska misao: časopis Saveza Ekonomista Srbije za Pitanja Ekonomske Teorije i Prakse, 40(1-2), 52-66. Vandevoorde, S., & Vanhoucke, M. (2006). A comparison of diﬀerent project duration forecasting methods using earned value metrics. International Journal of Project Management, 24(4), 289-302. doi:10.1016/j.ijproman.2005.10.004. Wilson, B., Ariyachandra, T., & Frolick, M. (2013). Earned value management systems: Challenges and future direction. Journal of Integrated Enterprise Systems, 2(1), 9-17. Yang, H., Yeung, J.F.Y., Chan, A.P.C., Chiang, Y.H., & Chan, D.W.M. (2010). A critical review of performance measurement in construction. Journal of Facilities Management, 8(4), 269-284. doi:10.1108/14725961011078981.

PRIMENA METODE OSTVARENE VREDNOSTI U PRAĆENJU MANJIH GRAĐEVINSKIH PROJEKATA Rezime: Primena metode ostvarene vrednosti u Srbiji značajno zaostaje za drugim razvijenijim privredama, bez obzira na to što se pokazala kao veoma upotrebljiv i prilagodljiv alat za predviđanje ukupnih troškova i vremena realizacije građevinskih projekata. Ovaj rad trebalo bi da doprinese široj primeni metode upravljanja ostvarenom vrednošću (EVM), kao veoma praktičnoj metrici koja se može koristi za praćenje manjih građevinskih projekata. Pregled literature otkriva kako se metoda razvila kao osnovno sredstvo za praćenje građevinskih projekata i upućuje na specifične probleme vezane za njenu primenu. Kako bi se ukazalo na sve pogodnosti metode ostvarene vrednosti u cilju uspešnog praćenja učinka na manjim građevinskim projektima, upotrebljivost metodologije prikazana je u vidu studije slučaja.

Ključne reči: ostvarena vrednost, upravljanje projektima, građevinarstvo. Received: September 2, 2015. Correction: September 7, 2015. Accepted: September 15, 2015.

EJAE 2015, 12(2): 9-15 ISSN 2406-2588 UDK: 331.538 331.101.38-058.832 DOI: 10.5937/ejae12-8672 Original paper/Originalni naučni rad

THE EFFECT OF INCENTIVE RETURN-TO-WORK POLICIES ON SINGLE-PARENT FAMILIES: A COMPARATIVE APPROACH Samir Amine*, Pénayori Ouattara University of Quebec; Center for Interuniversity Research and Analysis of Organizations, Montréal, Canada

Abstract: As part of their struggle against inequality and social exclusion, many countries are trying to reform their redistributive system and low-income support measures to encourage return to work, and reduce inactivity traps. The purpose of this article is to propose a reflection on the social and fiscal policies. The analysis focuses on three measures in three different countries: the Earned Income Tax Credit (EITC) in the United States, the Working Tax Credit (WTC) in the UK and the Active Solidarity Income (RSA) in France. The emphasis is placed on the effects of such policies on the situation of single parents, in particular as part of their transition from welfare (social assistance) to the labor market.

INTRODUCTION For several years, the debate on the equity-eﬃciency dilemma was strongly covered in the theoretical and empirical literature. Nevertheless, socio-economic changes, observed in the labor market and exacerbated by the 2008 crisis, have helped to put this dilemma at the heart of political debate in most of the developed countries (Amine, 2011). In their fight against inequalities and social exclusion, such attempts have been made to reform their redistributive system and more eﬀectively support mechanisms for low-income (Amine, 2014). In this context, Zajdela (2009) poses the problem of inadequate social and fiscal measures with the specific realities of the people and highlights the importance of conducting a diagnosis prior to implementation of the OECD recommendations (2006). As modulated by the OECD, these are the courses of action that should adapt to the realities of each country. Moreover, these policies implemented in accordance with these recommendations are not targeted and usually address the divergent needs of the public. * E-mail: samir.amine@uqo.ca

Key words: single-parent, return to work, labor market participation.

Based on the notion of rationality of individuals, these incentives to return to work emphasize the fact of joining a labor market that need not necessarily produce quality jobs (Zajdela, 2009). In order to make these measures more suited to the needs of persons covered, it would be wise to prioritize policies designed in terms of labor demand and add training policies of labor (Kim, 2011; Zajdela, 2009). In other words, the return to work should not be the only issue to take into consideration, as the socio-fiscal measures should also be associated with the nature of the job (Dubet & Veretout, 2001; Dufour et al., 2003). The purpose of this paper is to propose a reflection on the return-to-work policies. This reflection is based on two elements. On one hand, the analysis focuses on three measures in three diﬀerent countries. It is about the Earned Income Tax Credit (EITC) in the United States, the Working Tax Credit (WTC) in the UK and the Active Solidarity Income (RSA) in France. While these measures have common inspirations, based on the principle of negative tax, their comparison allows for better reporting of institutional characteristics of each economy. On the other hand, the focus is on 9

EJAE 2015  12 (2)  9-15

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

the eﬀects of such policies on the situation of single parents in particular, as part of their transition from welfare to the workforce. CONTEXT AND LITERATURE REVIEW

Unemployment rate has grown considerably in Europe since the end of the post-war economic boom and it also seems to have settled permanently in North America since the beginning of the financial crisis (particularly the US). In this context, European welfare states are bogged down, sealed by excessive debt levels, with budget deficits well above those in the Stability Pact and extremely low growth rates. In addition, social and tax systems are piling on top of each other (since 1945) making fiscal policy (and redistribution policy) properly illegible and ineﬀective; seeking in some cases to make regressive (Landais et al., 2011) at a time when more inequality will stop and even tend to increase significantly (Amine et al., 2012). Moreover, according to the ILO and the World Bank, about 5.1 billion people, i.e. 75% of the world population is not covered by social security adapted, as 1.4 billion people still live on less than $ 1.25 per day (ILO, 2011). This does not seem to spare the developed countries. Indeed, in 2007, around 9.2% of Canadians, or about 2.95 million people, were below the low-income threshold. However, it is important to note that despite this growing inequality, the weight of social benefits in public spending has not necessarily declined in most of the OECD countries. For example, since 1980 the share of the budget France awards within the social system has remained the same (around 33% of public expenditure) (Amine & Dos Santos, 2012). The United States even increased their spending on the labor market policies in 2008 and 2009 (0.2% of GDP instead of 0.1% for all previous years). The UK and Canada have also made increases at the same expenses from 2008 (OECD, 2013). In general, persistent unemployment and budget constraints imposed by the repayment of public debt, largely explain the OECD recommendations directed towards refoundation of social and fiscal measures (Merrien, 2007). These have been the subject of several evaluations and some of them shall be presented in the following paragraph. In a study on Canada, New Zealand and the United States, Cook (2012) focuses on health conditions of social assistance recipients and concludes that the transition to the labor market is a dynamic process 10

that focuses on the socio-economic life and neglects the search for identity of a recipient. In addition, he mentions that single parents as welfare recipients remain in poverty despite obtaining a job. In this sense, Cook (2012) explains that several factors impede the success of labor subsidy programs whose participants are faced with the harsh realities of work-family balance and lack of medical coverage. Sen (2003) also calls for a shift in policy debates to adopt programs that integrate recipients as craftsmen development and involve them in decisions that are supposed to improve their living conditions. In the same vein, Breitkreuz and Williamson (2012) are attempting to understand how the experiences of single parents with young children are aligned with political and ideological concepts. Indeed, based on the experience of Canada, they question the self-sufficiency which is the central notion in recent social policy reforms aimed at reintegrating welfare recipients into the labor market. From a longitudinal research, they manage the results indicating that the objectives of policies do not necessarily reflect the reality in terms of poverty and precariousness, of program participants. Within the same Canadian context, the study of Good-Gingrich (2010) focuses on parent families benefiting from Toronto “Ontario Works” program that administers social assistance in Ontario. The results obtained by Good-Gingrich (2010) confirm the discrepancy between the mode of administration of the programs and the precarious situation suffered by most participants. In another economic context, Dubet and Vérétout (2001) examine in France the reasons explaining the decisions of welfare recipients in their job search. Their questioning concludes that the beneficiaries of minimum income (RMI) would proceed to possible trade-offs between the expected employment income based on their qualifications as to refrain from seeking employment. Providers of social assistance, who have the appropriate qualifications to meet the labor market requirements and take a well-paid job, leave the system. Instead, the low-skilled or unskilled opt for the services of social assistance that are higher than the expected income jobs (Brouillette & Lacroix, 2011). Furthermore, in a study on the single-parent allowance in France, Curraize and Périvier (2009) mention that the personal characteristics of recipients can influence decisions to return, keep or not join the labor market. The designs and subsequent actions of beneficiaries of social assistance differ according to their uncertainties facing the labor market.

EJAE 2015  12 (2)  9-15

Regarding the US context, Kim (2011) analyzes the effects of two programs in the United States and for people receiving emergency assistance: Human Capital Development (HCD) and Labor Force Attachment (LFA). The study uses a combination of the official data on the unemployed and the emergency aid which is the main source of income. Based on these data, Kim (2011) extracted longitudinal observations from 1997 to 2001 on single-parent families (especially women) who participated in the HCD and the LFA. The results indicate that participants who received both programs are more likely to get jobs. Similarly and based on the March CPS data collected from 1978 to 1999, Grogger (2003) concentrates on single-parent families recipients of EITC, of which 93% are women. During the observation period, it appears that the participation of families reached a record rate of 33% in the 90s and then fell to 15% in 1999. Inspired by the French experience and in order to reduce poverty, the Québec government introduced the “prime au travail“ (PT). This is a refundable tax credit for low income. The objectives behind this measure are twofold: insertion and retention in the labor market. The allocation calculation is determined in relation to the income received from a job and it also takes into account the family situation. Fortin et al. (2011) believe that the adoption of this PT should not necessarily be analyzed in terms of purely economic benefits since the early childhood program in Quebec helped to reconcile family and work, and has achieved positive results at the provincial level and federally. Starting from different foreign experiences, Godbout and Arseneau (2005) reach the fact that the PT can be effective for engaging in the labor market of single-parent families with a child as long as the person opts to work full-time at minimum wage. In this line, Cahuc et al. (2008) consider that poverty results from the fact that people are unable to exercise fulltime employment throughout the year. On the same measure, Lafond-Bélanger (2007) analyzes the potential effects of PT on women in couples by structural ex-ante approach. The data used in the analysis come from the Survey of Labour Dynamics and Income 2002. From an econometric discrete choice model and simulations, Lafond-Bélanger (2007) concludes that the PT does not reach women beneficiaries in couples. When compared to an ordinary situation where work is not subsidized, PT encourages inactivity in women in couple, resulting in a decline in their number in the labour market. In the same vein, Parise (2007) adopts a structural approach

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

based on a micro-simulation model to predict the effects that the introduction of the PT could generate as labor supply and hours worked by women alone and single parent families. The results show that the reactions of single women and those of single parent families result in the improvement in the supply of labor by 0.6%. The categories of part-time jobs make up three quarters of the choice of new access to the labor market. This positive effect of PT on participation of single parents in the labor market was also emphasized by Brouillette and Fortin (2008) as part of their experimental study. THE IMPACT ON SINGLE PARENTS Working Tax Credit The UK has engaged very early in the inspired action of the NIT and has repeatedly performed revision of its tax-benefit system; Family Credit and Working Family Tax Credit (WFTC) and Working Tax Credit (WTC) since 2003. Furthermore, new provisions are undertaken in order to establish the “Universal Credit”. In order to qualify for benefits, single parents with one child or more must work for at least 16 hours per week. According to the official data, the implementation of the WTC in 2003 seems to have had the desired effect from 2006, as the single-parent employment rate gradually increased from 49.9% in 2005 to 50.2% in 2006. Such upward trend was maintained up to 52.9% in 2010. Furthermore, in fiscal year 2011-2012, we estimate that 366 000 single parents with jobs, benefited from 30 hours component of the tax credit. They represent 34.76% of recipient families of the component 30 hours of the WTC. Nevertheless, the data give no indication of the job categories, temporary or permanent, occupied by these families. It is worth noting that majority of evaluations of socio-tax schemes in the UK are using the WFTC data. The latter seems to have increased the participation rate of single parents in the labor market by 47% in 1996 to 55% in 2002 (Immervoll & Pearson, 2009; OECD, 2014; Mikol & Remy, 2010). However, some reproach the WTC, predominantly its administration, as it appears to be too complex, cumbersome and unreadable. Such criticism is the main source of change that the WTC “Universal Credit” is undergoing (OECD, 2014). 11

EJAE 2015  12 (2)  9-15

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

Active Solidarity Income

In order to fight against “poverty to work” and encourage return to work, the French government has been implementing a new mechanism called the active solidarity income (RSA) since June 01, 2009. This new measure does not replace the “employment premium” (PPE), but both devices are built with each other. The RSA is a measurement with two components: the RSA base and the RSA activity. The RSA base provides a lump sum to the unemployed and people whose income is below the standard amount. The RSA activity is a financial supplement for workers whose incomes are above the RSA and meet the relevant eligibility criteria for their situations. Périvier (2010) considers that this device gives a new way to the French social policy by passing a design rooted on solidarity to merit a design where labor grant recipients have the requirement to fit in the labor market and remain there. The data from the year 2013 indicate that 33% of single-parent families benefit from the RSA, of which 31% are women and 2% are men. In December 2010, single parents (specifically women) RSA beneficiaries stood out by their participation in the labor market; 40% of single parents with two children against less than 25% for those living in couple with the same number of children (Briard, 2012; Domingo, 2014). As for the number of working hours, the trend is 28 hours per week (Briard, 2012). The statements of the participants in the RSA suggest that part-time work is particularly involuntary and imposed by the employer. Similarly, the comparative study of Mikol and Remy (2010) reveals positive eﬀects of RSA on labor market participation. Based on the quantitative survey of Dares (direction of animation of research, studies and statistics) on RSA beneficiaries 2010-2011, Domingo (2014) determines the characteristics of jobs held by parent families RSA beneficiaries. Indeed, it shows that 55% of single women are permanently employed and the proportion of men in this category would be lower. Relying on the same data, Dares, Briard and Sautory (2012) could not determine the extent of the RSA eﬀects on the labour market and hours worked. In this perspective, Thibault (2014) explains that the RSA encourage return to work, mainly part-time, singleparent families, and more specifically, those with dependent children. However, many critics argue that the two programs (RSA and PPE) cohabit, but do nothing to reconcile their means to achieve the common goal among those 12

targeted. In addition, RSA is not implied, levied on a monthly basis, it is declarative and its complex eligibility requirements may make it incomprehensible to potential beneficiaries, hence indicating the weakness of its application rate, with only 32% of those who would claim the right to RSA. In addition, the RSA in its current structure excludes fringe of workers aged below 25 and increases social exclusion and poverty observed for this category of workers. RSA financing costs are also criticized and because of the growing weight they represent in departmental budgets. The assistance provided to low-income families through the RSA benefits would be more expensive than expected. Indeed, the expenditure in 2013 is more than € 4 billion of the projected amount, thereby raising the total budget to 10 billion. Earned Income Tax Credit The Internal Revenue Service (IRS), which administers the EITC, says the program has evolved significantly since its establishment. In fact, various reforms were introduced in 1986, 1990 and 1993, which has resulted in increased funding for the EITC and made it very generous, especially to families with two children or more (Bontout, 2000). The 1993 reform, which in its structure becomes more favorable for families with two or more children, is considered by Mikol and Remy (2010) as the most ambitious US government since benefits nearly doubled. In this sense, for single parents without a job, the benefits of the EITC make attractive the course of employment and thus encourage integration into the labor market (Bontout, 2000). Indeed, Bontout (2000) mentions that during the period 1993-1996, the EITC provided the basis for an increase in the activity rate for single parents (especially women) raising in from 65.5% to 72.1%. During the period 1985-1996, single parents (specifically women) stood out for their massive comeback in the labor market, a movement that has generated rise in female participation in the labor market, setting their number from 51 to 61.9 million. Several studies in the 1990s contributed to these positive eﬀects of the EITC on participation in the labor market for single-parent families. In a more recent study, Immervoll and Pearson (2009) have shown that about two-thirds of EITC recipients are single parents (Grogger, 2003). By focusing on low-skilled workers, Chyi (2012) extracted the history of their use of social assistance and employment episodes from 1989 to 2000. It shows

EJAE 2015  12 (2)  9-15

that one hand employment has increased by about 30% since 1995, while the number of single-parent families targeted by the study (low-skilled women) receiving social assistance dropped by almost 70%, from 34% to 11% (Chyi, 2012). In general, the EITC provides low-income families with real economic support. Indeed, in 2010, it fought against poverty with the US population by allowing 6.3 million people to improve their financial situation. It is worth noting that 59.5 billion US dollars was spent on the EITC services for the benefit of 27 million eligible families (Tach & Halpern-Meekin, 2014). Hungerford and Thiess (2013) confirm this eﬀectiveness based on the studies that summarize research on the EITC impact on participation of single-parents in the labor market (especially women). However, they emphasize the claim that EITC can be complex because of the supplementary forms to usual forms of tax return. In this sense, Tach and Halpern-Meekin (2014) are attempting to determine how EITC recipients include its structure to the point of use to make trade-oﬀs based on working hours that would allow them to maximize their earnings. Given the participants’ responses to the survey conducted as part of their study, they concluded that recipients do not understand the structure of the EITC and do not differentiate between the portion of the EITC related to their working income and that related to the eligibility of children. The beneficiaries of the EITC translate this incomprehension of its structure by schemes to benefit from higher amounts of credits. CONCLUSIONS

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

to create a harmonized platform that could be more efficient with the individuals concerned. This failure is noted by Martin (2000): Recent studies conducted by the OECD also show that it is of vital importance to consider the interactions between active and passive policies in the labor market if we are to improve the effectiveness of active policies. In the present context where companies have much recourse to forms of employment (atypical, precarious, temporary...) that do not offer individuals the opportunity to lead a stable professional life, resort to social assistance benefits can appear safer (Duvoux, 2010). The inputs and outputs of the labor market on the basis of typical jobs discourage job search of some social assistance providers who actually yearn for permanent employability and wish to invest in programs that help to socially integrate through skilled and rewarding jobs (Duvoux, 2010). In this context, the occupation of job does not put more systematically workers away from poverty. The phenomenon of “working poverty” is increasing and the data on a third of OECD countries indicate that 20% of single-parent families live in such a situation (OECD, 2009). Certainly, measures such as the WTC and EITC fail to encourage the reintegration into the labor market, but no information will give details about the quality and sustainability of jobs. In general, the effect of these programs on participation in the labor market depends on several factors, including the ability of the economy and the labor market to reduce unemployment and provide sustainable, well-paid jobs. Hence, education and training remain the key elements for successful professional integration (Immervoll & Pearson, 2009). REFERENCES

The EITC, commonly cited for its eﬃciency, takes place in a context characterized by relatively low minimum wages and limited access to other social benefits. By contrast, in France, there is a social safety net developed because of the existence of social benefits other than RSA (Mikol & Remy, 2010). The observation made by Mikol and Remy (2010) is that the minimum wage is high and there are several other benefits that could undermine the intended eﬀects of the RSA introduction. Furthermore, the decentralized management of various French social measures and services does not promote real interaction to better achieve the targeted individuals (Martin, 2000). Nevertheless, all incentive return-to-work policies citied herein are limited to financial aid. They do not integrate with the existing measures, called passive,

Amine, S. (2011). Les institutions du marché du travail et les inégalités inter-catégorielles: une comparaison FranceCanada. Revue multidisciplinaire sur l’emploi, le syndicalisme et le travail, 6(1), 5-19. Amine, S., & Lages Dos Santos, P. (2012). L’évolution de la conception du rôle de l’État. In F. Bottini (Ed.), L’État interventionniste: Le rôle de la puissance publique dans l’économie (pp. 59-84). Paris: L’Harmattan. Amine, S., Lages Dos Santos, P., & Rioual, A. (2012). Impôt négatif et allocation universelle, de la théorie à la pratique. In Transformations du monde du travail: Regards interdisciplinaires en relations industrielles (pp. 93-124). Québec: Presses de l’Université du Québec. Amine, S. (2014). Les politiques du marché du travail face à la crise économique: Vers une convergence internationale. Québec: Les Presses de l’Université Laval. 13

EJAE 2015  12 (2)  9-15

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

Bontout, O. (2000). L’Earned Income Tax Credit, un crédit d’impôt ciblé sur les foyers de salariés modestes aux États-Unis. Économie et statistique, 335(5), 27-46.

Breitkreuz, R.S., & Williamson, D.L. (2012). The Self-Suﬃciency Trap: A Critical Examination of Welfare-to-Work. Social Service Review, 86(4), 660-689. doi: 10.1086/668815 Briard, P. (2012). La situation des bénéficiaires du RSA sur le marché du travail fin 2010. Dares Analyses, 2012(04). Retrieved December 6, 2014, from http://travail-emploi. gouv.fr/IMG/pdf/2012-014.pdf Briard, P., & Sautory, O. (2012). Évaluation de l’impact du revenu de solidarité active (RSA) sur l’offre de travail. Dares Analyses, 2012(171). Retrieved December 6, 2014, from http://travail-emploi.gouv.fr/IMG/pdf/DE171.pdf Brouillette, D., & Fortin, B. (2008). Impact de la Prime au travail sur l’effort au travail: Une approche expérimentale. Montréal: CIRANO. Retrieved March 15, 2015, from http://www.cirano.qc.ca/pdf/publication/2008RP-01.pdf Cahuc, P., Cette, G., & Zylberberg, A. (2008). Salaire minimum et bas revenus: Comment concilier justice sociale et efficacité économique. Conseil d’Analyse Économique. Retrieved March 15, 2015, from http://www.cae-eco.fr/ IMG/pdf/079.pdf Chyi, H. (2012). The 1993 EITC expansion and low-skilled single mothers’ welfare use decision. Applied Economics, 44(13), 1717-1736. Cook, K. (2012). Neoliberalism, welfare policy and health: A qualitative meta-synthesis of single parents’ experience of the transition from welfare to work. Health, 16(5), 507-530. doi:10.1177/1363459311434650 Curraize, Y., & Périvier, H. (2009). L’allocation de parent isolé a-t-elle favorisé l’inactivité des femmes. Économie et statistique, 429(430), 159-176. Dubet, F., Vérétout, A., Dubet, F., & Veretout, A. (2001). Une «réduction» de la rationalité de l’acteur Pourquoi sortir du RMI. Revue Française de Sociologie, 42(3), 407-436. doi:10.2307/3323028 Dufour, P., Boismenu, G., & Noël, A. (2003). L’aide au conditionnel. La contrepartie dans les mesures envers les personnes sans emploi en Europe et en Amérique du Nord. Montréal: Presses de l’Université de Montréal. Duvoux, N. (2010). Le travail vu par les assistés: éléments pour une sociologie des politiques d’insertion. Sociologie du travail, 52, 389-408. Domingo, P. (2014). Revenu de solidarité active des hommes et des femmes: des caractéristiques d’emploi diﬀérenciées. Publication mensuelle de la Caisse nationale des Allocations familiales. Retrieved March 15, 2015, from https://www.caf.fr/sites/default/files/cnaf/Documents/ Dser/essentiel/essentiel%20-%20genre%20RSA_ESSENTIEL.pdf Fortin, P., Godbout, L., & St-Cerny, S. (2011). Economic Consequences of Quebec,s Educational Childcare Policy. 14

OISE:Early Years Economics Forum. Toronto, Ontario. Retrieved May 5, 2014 from http://childcarecanada.org/ documents/research-policy-practice/11/06/early-education-economic-forum Godbout, L., & Arseneau, M. (2005). La prime au travail du Québec. Un véritable outil d’incitation au travail ou une simple façon de baisser l’impôt. Montréal: CIRANO. Retrieved March 15, 2015, from http://www.cirano.qc.ca/ pdf/publication/2005s-01.pdf. Good-Gingrich, L. (2010). Single Mothers, Work(fare), and Managed Precariousness. Journal of Progressive Human Services, 21(2), 107-135. Grogger, J. (2003). The effects of time limits, the EITC, and other policy changes on welfare use, work, and income among female-headed families. Review of Economics and statistics, 85(2), 394-408. Hungerford, T.L., & Thiess, R. (2013). The earned income tax credit and the child tax credit: History, Purpose, Goals, and Effectiveness. Economic policy institute. Retrieved March 15, 2015, from http://www.epi.org/publication/ ib370-earned-income-tax-credit-and-the-child-taxcredit-history-purpose-goals-and-effectiveness/ Immervoll, H., & Pearson, M. (2009). A Good Time for Making Work Pay: Taking Stock of In-Work Benefits and Related Measures across the OECD. Paris: OECD Publishing. doi:10.1787/1815199x Kim, J. (2011). The Effects of Welfare-to-Work Programs on Welfare Recipients’ Employment Outcomes. Journal of Family and Economic Issues, 33(1), 130-142. Lacroix, G., & Brouillette, D. (2011). Assessing the impact of a wage subsidy for single parents on social assistance. Canadian Journal of Economics, 44(4), 1195-1221. Lafond-Bélanger, G. (2007). Les effets de la Prime au travail sur l’offre de travail des femmes en couple. Mémoire. Faculté des Sciences Sociales. Université Laval. Landais, C., Piketty, T., & Saez, E. (2011). Pour une révolution fiscale, un impôt sur le revenu pour le XXIème siècle. Paris: Le Seuil. Martin, J.P. (2000). Ce qui fonctionne dans les politiques actives du marché du travail: Observations découlant de l’expérience des pays de l’OCDE. Revue économique de l’OCDE, 30, 85-120. Retrieved March 15, 2015, from http://www.oecd.org/fr/emploi/emp/34529236.pdf Merrien, F.X. (2007). L’État-providence. Paris: P.U.F. Mikol, F., & Remy, V. (2010). Quels effets attendre du RSA sur l’offre de travail et les salaires au vu des expériences étrangères? Un bilan des travaux sur l’EITC et le WFTC. Dares, 2010(153). Retrieved December 6, 2014, from http://travail-emploi.gouv.fr/IMG/pdf/DE153_RSA_ WFTC_-_EITC_FM_VR_def_AM.pdf OECD. (2006). Perspectives de l’emploi de l’OCDE. Retrieved December 6, 2014, from http://www.oecd.org/fr/els/ emp/perspectivesdelemploidelocde2006.htm

EJAE 2015  12 (2)  9-15

OECD. (2009). Perspectives de l’emploi de l’OCDE. Retrieved December 6, 2014, from http://www.oecd-ilibrary.org/fr/ employment/perspectives-de-l-emploi-de-l-ocde-2009_ empl_outlook-2009-fr. doi:10.1787/empl_outlook-2009-fr OCDE. (2014). Connecting People with Jobs: Activation Policies in the United Kingdom. OECD Publishing. doi:10.1787/9789264217188-en OIT. (2011). Socle de protection sociale pour une mondialisation juste et inclusive. Rapport du Groupe consultatif sur le socle de protection sociale. Genève: Bureau international du Travail. Parisé, H. (2007). Impact de la Prime au travail sur l’offre de travail. Une évaluation ex-ante. Mémoire. Faculté des Sciences Sociales. Université Laval.

Amine, S., Ouattara, P.  The effect of incentive return-to-work policies

Périvier, H. (2010). La logique sexuée de la réciprocité dans l’assistance. Revue de l’OFCE, 114, 237-263. doi:10.3917/ reof.114.0237 Sen, A. (2003). L’économie est une science morale. Paris: La découverte. Tach, L., & Halpern-Meekin, S. (2014). Tax code knowledge and behavioral responses among EITC recipients: Policy insights from qualitative data. Journal of Policy Analysis and Management, 33(2), 413-439. Thibault, F. (2014). Ambitions et évaluation du Revenu de solidarité active. Économie et statistique, 467(468), 51-59. Zajdela, H. (2009). Comment et pourquoi activer les inactifs. Travail et Emploi, 118, 69-76.

PODSTICAJNE MERE ZA POVRATAK SAMOHRANIH RODITELJA NA TRŽIŠTE RADA: UPOREDNI PRISTUP Rezime: Kao vid borbe protiv nejednakosti i socijalne isključenosti, mnoge zemlje pokušavaju da reformišu svoje sisteme redistribucije i uvedu stimulativne finansijske mere kako bi podstakli povratak na posao i smanjili neaktivnost stanovništva. Ovaj rad ima za cilj da pruži osvrt na postojeće društvene i fiskalne politike i programe. U istraživanju se prikazuju tri podsticajne mere preraspodele dohotka za siromašne u tri različite zemlje: sniženje poreza po zarađenom dohotku (EITC) u SAD-u; odbitak poreske obaveze za decu i zaposlene osobe (WTC) u Velikoj Britaniji; i odbitak poreza na dohodak slabije plaćenih osoba (RSA) u Francuskoj. U radu se stavlja naglasak na dejstvo takvih stimulativnih programa na finansijsku situaciju samohranih roditelja, naročito kao deo njihovog prelaska iz stanja socijalnog blagostanja (socijalne pomoći) na tržište rada.

Ključne reči: samohrani roditelj, povratak na posao, učešće na tržištu rada. Received: May 25, 2015. Correction: Jun 20, 2015. Accepted: July 21, 2015.

EJAE 2015, 12(2): 16-26 ISSN 2406-2588 UDK: 005.52:330.133.1]:502 502.3:504.5:621.43.064(676.2) DOI: 10.5937/ejae12-9058 Original paper/Originalni naučni rad

STATED PREFERENCES FOR IMPROVED AIR QUALITY MANAGEMENT IN THE CITY OF NAIROBI, KENYA Hilary Ndambiri1, Eric Mungatana1, Roy Brouwer2 Center for Environmental Economics and Policy in Africa, University of Pretoria, South Africa 2 Department of Environmental Economics, Institute for Environmental Studies, Vrije University, Amsterdam, The Netherlands 1

Abstract: The study uses contingent valuation (CV) framework to assess individuals’ preferences for improved air quality management through motorized emission reductions in the city of Nairobi, Kenya. A conventional payment card (PC) is used to draw preferences from individuals in order to estimate the mean and the median willingness to pay (WTP) for air quality improvements in the city. Through interval regression analysis, the study finds that individuals are, on average, willing to pay Kshs. 396.57 ($4.67) and a median of Kshs. 244.94 ($2.88) to improve air quality management in the city. These amounts are found to increase with male gender, individuals’ income, certainty about future income and residence in an urban area. These amounts, however, decline with age, residential distance from nearby roads, and motor vehicle ownership. On the whole, the study shows significant public support towards improved air quality management in the city, which is of vital importance for effective formation and implementation of air quality management programmes.

INTRODUCTION The continuous upsurge in the number of motor vehicles in most urban areas worldwide has raised enormous concerns over the eﬀects of motor vehicle emissions on air quality of urban environments (Cadle et al., 2004; Gwilliam, 2003; Lilley, 2000). Although most transportation modes emit pollutants into the atmosphere through combustion of liquid fossil fuel, the relative abundance of pollutants varies depending on the exact composition of the fuel and details of combustion conditions (Cadle et al., 2004; Colvile et al., 2001). Some of the leading motorized emissions by mass include carbon monoxide (CO), nitrogen oxides (NOx), sulphur oxides (SOx), volatile organic compounds (VOCs) and the particulate matter (PMx) (Asian Development Bank, 2003; Colvile et al., 2001; Fazal, 2006; Schwela, 2004). These emissions are largely harmless if they are in low concentrations, but harmful if their concentrations escalate to levels 16

* E-mail: ndambirihk@yahoo.com

Key words: contingent valuation, willingness to pay, payment card, air quality, Nairobi.

that cause adverse effects on humans and the environment (Ababio, 2003; Amin, 2009). Furthermore, they are mostly ground level sources of pollution highly capable of rendering maximum impact on the general population (Ackerman et al., 2002; Gwilliam & Johnson, 2004; Schwela, 2004; Vliet & Kinney, 2007). In the city of Nairobi, the growth of motorized traffic has been singled out as the major source of low air quality since motorized emissions account for about 90% of total emissions (Odhiambo et al., 2010; Omwenga, 2011). Toxic gases such as CO, NOx, SOx and PMx are in high concentrations as a result of increased urban expansion and motorization linked to rapid economic development experienced over the last decade. Consequently, most city residents are exposed to elevated concentrations of vehicular emissions that potentially pose serious long-term effects on human health and quality of the urban environment (Odhiambo et al., 2010; Kinney et al., 2011; Vliet & Kinney, 2007). For instance, CO causes blood clotting

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

when it reacts with haemoglobin. PMx are carcinogenic, especially those that penetrate deep into the lungs. Exposure to NOx leads to shortness of breath, chest pain and changes in lung function, while SOx causes cardiovascular diseases and bronchitis with detrimental effects to asthmatics (WHO, 2006). Notwithstanding these effects, the authorities in the city of Nairobi are yet to formulate a policy that averts any potential risk from motorized emissions, given that the city has resident and daytime population of 3.1 and 4.2 million people, respectively. The reason behind this problem is that the current studies on air quality in Nairobi (Kinney et al., 2011; Mulaku & Kariuki, 2001; Odhiambo et al., 2010; Vliet & Kinney, 2007) deal only with the technical aspects of measuring air pollutant concentrations. This implies that available policy information is only limited to technical aspects, while sufficient knowledge on the socioeconomic aspects of the policy is also needed for the effective air quality management. Therefore, the purpose of this study is to provide the missing policy information by assessing peoples’ preferences towards improved air quality management in the city of Nairobi. In both developed and developing countries, several studies have already been undertaken to estimate individuals’ preferences for air quality improvements. For instance, Carlsson and Johansson-Stenman (2000) estimate individuals’ preferences for air quality improvement in Sweden using the open-ended question format. They find that individuals are, on average, willing to pay about 2000 SEK/year. The WTP amounts are found to increase with respondents’ income, wealth and education. They are larger for males, people owning a house, people living in more polluted areas and members of environmental organizations, but smaller for the retired people. In Manila, Philippines, Fabian and Vergel (2002), double-bounded discrete choice format is used to estimate commuters’ preferences for improved air quality from public transport vehicles. They show that respondents are willing to pay for an increase in fare by as much as PhP1.24. In Poland, Dziegielewska and Mendelsohn (2005) employ a system of dichotomous choice questions to analyze preferences of Polish citizens towards harmonization of Polish air pollution standards with the European Union (EU) standards. They reveal that Polish citizens value the benefits of managing pollution much less than other citizens from richer EU nations, which means that the suspension of the harmonization process seemed critical and needed to be based on the nations’ economic prosperity. Wang et al. (2006)

use an open-ended question format to estimate peoples’ preferences for improved air quality in the urban areas of Beijing. They find that people are, on average, willing to pay 143 CNY per household per year. The WTP is found to be an increasing function of income and education level, but a decreasing function of age and household population. It is larger for residents in the urban areas than for those in the suburbs. Wang and Zang (2008) evaluate peoples’ preferences for improved air quality based on a variety of hypothetical open-ended questions in Ji’nan, China. They reveal that 59.7% of respondents are, on average, WTP (willing to pay) 100 Chinese Yuan (CNY) per person per annum. Significant factors are found to influence individuals WTP (willingness to pay): expenditure incurred while treating respiratory diseases, individual’s income and the number of workers in the family. In Mashhad, Iran, Firoozzarea and Ghorbani (2011) use open-ended questions and the Heckman procedure to estimate peoples’ preferences for air pollution reduction together with the respective determinants. They find that the total monetary value of air quality improvement in highly polluted region is higher (7.134 billion Rials per month) than in the middle-polluted region (5.242 billion Rials per month). Age, number of children, gender, household income, education level and car owning are significant factors for people who are willing to pay (WTP) for improved air quality. In addition, Du and Mendelsohn (2011) use the double-bounded valuation format with a single follow-up question to evaluate the preferences of Beijing residents to sustain the improved quality of air experienced during the Beijing Olympic Games. The results show that the mean annual WTP per household ranges between CNY 22,000 and 24,000. Finally, Donfouet et al. (2013) analyze peoples’ preferences for air quality improvements in Douala, Cameroon, through a hypothetical referendum scenario. Respondents are also given an overnight time to think about their responses to the valuation questions. The results show that people are, on average, willing to pay (WTP) $US 0.40 per month for improved air quality, which is about 0.02% of their annual income. The overnight time-to-think is found to have negative eﬀects on individuals’ valuation decisions since it lowered the WTP values by 40%. Except for China and Cameroon in the African context, the number of socioeconomic studies dealing with the issues of improved air quality management is insuﬃcient. Such studies, if undertaken, may provide important policy information for addressing air 17

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

quality issues, especially among rapidly urbanizing cities in the developing world. This study, therefore, seeks such policy information by analyzing peoples’ preferences for improved air quality management in the city of Nairobi, Kenya. The rest of the paper is structured as follows: Section 2 presents the theoretical framework, Section 3 describes the survey methodology, Section 4 presents the empirical results and discussion, and Section 5 oﬀers the final conclusion. THEORETICAL FRAMEWORK

Most empirical studies on peoples’ preferences for air quality improvement have previously been conducted using non-market valuation methods. The use of non-market valuation methods has been justified because air quality is an environmental public good, hence unpriced and with no established market for its trading (Du & Mendelsohn, 2011; Wang et al., 2006; Wang & Mullay, 2003; Wang & Whittington, 2005; Wang & Zang, 2008). These methods are broadly classified into two groups (Mendelsohn & Olmstead, 2009; Mitchell & Carson, 1989): a) the revealed preference methods (e.g. travel cost and hedonic pricing) and b) stated preference methods (e.g. contingent valuation and choice experiments). Under the revealed preference methods, proxy markets are used indirectly to attach monetary values on policy proposals by finding correlation between the real market behaviour of individuals and the policy proposal in question. Under stated preference methods, however, hypothetical markets are used directly to attach monetary values on policy proposals by asking people about their WTP for a policy proposal that enhances their welfare or willingness to accept (WTA) compensation for a proposal that decreases their welfare. An important limitation of the revealed preference methods is that they can only attach monetary values on policy proposals through observation of real market transactions. In cases where it is impossible to observe real market transactions, the stated preference methods are preferred (Bateman et al., 2002; Venkatachalam, 2004; White et al., 2001). For that reason, this study implemented contingent valuation framework to assess peoples’ preferences for improved air quality management in the city of Nairobi since real market transactions for air quality improvements are unavailable and can only be 18

proxied through contingent valuation. Moreover, this framework was preferred over the choice experiment approach as it provides the total (use and nonuse) value of the policy proposals in question (Carson, 2000; Mitchel & Carson, 1989). Contingent valuation methodology is deeply rooted in the neo-classical welfare economic theory of consumer behaviour on expenditure minimization (Freeman, 1993; Mitchell & Carson, 1989). In this case, consider the following general expenditure function for an individual living in the City of Nairobi:

e( p, q, u) = y

(1)

where p is a price vector, q is the quality of air in the city, u is the level of utility, and y is the minimum income necessary to allow an individual to maintain utility level in the city. Furthermore, consider the situation where a policy is proposed to improve air quality management through motorized emission reductions. The policy, thus, outlaws all activities that are detrimental to air quality. The individual is then asked about the amount he/she would be willing to pay to reduce motorized emissions. The expenditure function for the initial period before the proposed policy would be:

e( p, qo , uo ) = yo

(2)

where uo is the initial level of utility that an individual can enjoy given prices p, qo is the initial level of air quality in the city and yo represents the minimum level of income required to attain utility level uo. Since the new policy is expected to improve air quality in the city, the new expenditure function would, therefore, be of the form:

e( p, q1 , uo ) = y1

(3)

where q1 is the quality of air after the implementation of the proposed policy and y1represents the minimum income level required to attain utility level uo after the implementation of the proposed policy. The level of utility, uo, is held constant since Hicksian welfare measures assume that utility remains constant. Hence, the individual’s WTP for improved air quality would be a compensating variation measure since an individual would have to part with a certain amount for the improvement to occur. The compensating variation (C) is equal to the individual’s WTP and is given by diﬀerence between the expenditure functions y1 and y0:

EJAE 2015  12 (2)  16-26

C = WTP = y1 - yo

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

C = {e( p, q1 , uo ) - e( p, q0 , uo )}

(4)

The quality of air in the city upon implementing the proposed policy, q1, is supposedly greater than the initial air quality, qo. As utility and prices are held constant, y1 (the minimum income level required to attain utility level uo after implementation of the proposed policy) is less than y0. Therefore, the compensating variation would be negative meaning that an individual has to pay some dollar amount to attain the improved level of air quality management. SURVEY DESIGN The study area The city of Nairobi is located at the south-eastern end of Kenya’s agricultural heartland, at approximately 1° 9’S, 1° 28’S and 36° 4’E, 37° 10’E. It has eight administrative divisions occupying an area of about 696 km2 with an altitude that varies between 1,600 and 1,850 metres above sea level. Although the city covers only 0.1 per cent of Kenya’s total surface area, it carries about 8 per cent of the country’s total population and about 25 per cent of Kenya’s urban population (CBS, 2009). Due to high population growth, the demand for transport in the city has been on the rise, which has resulted in increased motorization and hence, unprecedented levels of vehicle emissions. Population and sample The city’s resident population is estimated at 3.1 million people (CBS, 2009) spread over eight administrative divisions. As such, simple random sampling was used to select 61 respondents from each of the eight administrative divisions constituting the city of Nairobi to make an overall sample of 488 respondents for the entire study. This sampling procedure was chosen because it oﬀered all city residents an equal chance to be part of the study sample. Subsequently, the survey data was collected from each of the administrative divisions, as they could conceptually be regarded as separate populations upon which interviews could be performed independently. Survey technique The study employed personal interviews based on interviewer administered questionnaires to collect in-

formation from respondents. This method was chosen because it could enable the interviewer to motivate respondents to fully participate in the interview process, probe unclear responses and convey intricate information on the study subject to the respondents (Arrow et al., 1993; Dillman, 2000). The questionnaire was divided into five sections, namely: a) a background section that sought respondents’ general knowledge of air pollution in Nairobi; b) a section describing the motorized emission reduction plan; c) a section describing both positive and negative effects of the motorized emission reduction plan; d) a section with the valuation and debriefing questions and; e) a section that sought information on respondents’ socio-demographic characteristics. Survey implementation A pre-test survey was conducted on the survey questionnaire on a sample of thirty respondents using the open-ended value elicitation format as recommended in Haab and McConnell (2002). The respondents were asked to comment on suitability of questions in the questionnaire, paying close attention to wording, clarity, relevance and interpretation of each question in the survey among other anomalies. Bid ranges were also obtained from the pre-test, from which the mean, median, minimum and the maximum WTP values were determined. Based on the responses and comments provided by the respondents in the exercise, a final survey questionnaire was prepared and administered to 488 respondents. Considered environmental good A policy proposal for improved air quality management through motorized emission reductions in Nairobi constituted the public good of interest that was valued in the study. Notably, motorized emissions differ enormously from one vehicle to another, such that an accurate description of some definite level of emission reduction is difficult and can be misleading. As a result, a valuation question that sought for an overall reduction of motorized emissions in the city was posed to respondents and the values they gave were used to estimate mean and median WTP values for the study sample. 19

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

Payment vehicle

Popular payment vehicles used by researchers in CV include fees, taxes and amenity bills. However, Morrison et al. (2000) note that some of these payment vehicles can raise objections and protest responses among survey participants and hence bias the survey results. Following Fonta et al. (2010), this study uses a special trust fund, a neutral kind of payment vehicle, so as to minimize objections and protest responses among participants. In this fund, respondents were hypothetically required to make a one-time contribution to the exclusive purpose of reducing motorized emissions. It was expected that the payment vehicle would enhance the credibility of the hypothetical scenario as opposed to other alternative payment vehicles such as fees, taxes or amenity bills often linked with protest responses in contingent valuation (Morrison et al., 2000; Sayadi et al., 2009). Valuation format

Thus, the valuation question was formulated as follows: “Suppose the presented policy to reduce emissions from motorized vehicles and improve air quality management in the city of Nairobi will actually be implemented to reduce the current amount of emissions, what is the maximum amount of money you would be willing to pay one-oﬀ to the special trust fund to achieve this? (circle or tick a single amount on the card).” The PC included 15 diﬀerent dollar amounts, namely: Kshs. 0, 25, 50, 75, 100, 150, 200, 250, 300, 400, 500, 800, 1000, 1500 and finally Kshs. 2,000, in which case respondents were only required to circle one single amount on the card. Estimation method

Following Cameron and Huppert (1989), the interval regression analysis was used to estimate the mean and median WTP values from responses generated through the PC format. Letting WTPL be the maximum amount that a respondent would pay and WTPU be the lowest amount that a respondent would switch to a ‘No’ rather than a ‘Yes’ response, the individuals’ WTP is therefore taken to lie somewhere in the switching interval (WTPL, WTPU). In order to adjust for the skewed distribution of WTP responses, the lognormal transformation of the WTP responses is preferred, hence:

The study used the PC format to elicit individuals’ preferences based on a comprehensive policy proposal that would control motorized emissions in the city of Nairobi. Under this format, the respondents were given cards where they were asked to circle the highest amount they would be willing to pay for the emission LogWTPi = g i' v + xi (5) reductions. Out of the responses given, inferences were made about their true WTP, which was equal where gi denotes the characteristics of the respondto or higher than the circled value but lower than ent or the valuation good in question, xi stands for the next higher value (Cameron & Huppert, 1989). with zero This format was chosen because of the fact that the the normally distributed random variable mean and standard deviation σ, and v are regresrespondents had the advantage to easily and visually sion coeﬃcients. Assuming that WTP is a random scan through the given set of value intervals (Cameron variable (Welsh & Poe, 1998), the probability that a & Huppert, 1989) and hence determine the range of respondent would select a given monetary amount is: their WTP. Furthermore, the kind of data obtained Prob( yes) = prob(WTPi ³ WTPL ) through this format is less scattered and, therefore, Prob( yes) = 1 - QWTP (WTPL ) (6) does not require larger samples to obtain robust estimates. The format does not suﬀer from yeah-saying where QWTP(WTPL) is the cumulative distribution and starting point bias like other contingent valua- function of the random WTP variable. The probability tion formats (Mitchell & Carson, 1993). Although PC that the WTP would fall between any two monetary questions are theoretically susceptible to range and thresholds is: mid-point bias, there is little empirical evidence of the existence of range or mid-point Prob(WTPU > WTPi ³ WTPL ) = QWTP (WTPU ) - QWTP (WTPL ) (7) bias (Klose, 1999; Ryan et al., 2004). Even which results in the corresponding log-likelihood though the format still has the possibility of yielding function for n number of respondents as: protest zeros, it has not been found to give very high proportion of protest zero responses comæWTP - v g ' ö÷ æWTP - v g ' ö÷ïü ïì n U i÷ L i ÷ï - QWTP ççç pared to other contingent valuation formats Log (L) = å i =1 Log ïíQWTP ççç ÷ ÷ý (8) ïï ÷ ç ç s s è ø è ø÷ïþï î (Hanley et al., 2003; Klose, 1999). 20

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

With further assumption that the stochastic term is normally distributed, v and σ can be estimated and then used to compute the mean and me( g v + s 2) and dian WTP values. Thus, the mean WTP = e g v ( ) median WTP = e . Here, g' is taken as the vector of mean values of explanatory variables, v as the vector of estimated coeﬃcients and σ as the estimated standard variance. ' i

' i

RESULTS AND DISCUSSION Socio-demographic characteristics of respondents Table 1 presents the socio-demographic characteristics of respondents. As shown, the average age of respondents is between 31-40 years with men accounting for the largest share (62%) of respondents. A large proportion of respondents has attained secondary level of education and has an average household size of about 3 people. The mean annual income of respondents varies between Kshs. 20,000 ($253.29) - 30,000 ($352.94) with a large section of the respondents (83%) indicating that they were certain about their future incomes. Research results also show that most respondents resided close (150-200 metres) to nearby roads with a large share living in the urban areas (69%) as opposed to the suburbs (31%). Only a minority of respondents (18%) said they owned a motor vehicle.

Further results also show that people in Nairobi are familiar with air pollution problem, even though they may interpret it differently. While 88% of the respondents define it as the contamination of air with smoke and dust particles, 12% perceive it as bad odour from rotting waste and smoke. Regarding the main sources of air pollution, 55% of respondents mention motor vehicles as the main source of air pollution followed by factories (25%), burning of waste by households (19%) and finally, farming activities (5%). Serious problems that respondents associated with air pollution are respiratory in nature (71%), due to bad odour, dust particles and vehicular smoke. Despite these concerns from the public, the concerned authorities have still done little to address this issue. As a result, some policy measures recommended by the respondents to contain the situation include the use of: quality fuels (51%); fuel efficient motorized vehicles (18%); improved road infrastructure (13%); traffic regulation and control (12%) and construction of bicycle lanes to enhance the use of bicycles (1%). It is found that the majority of respondents (62%) prefer the use of a special fund as the payment vehicle to support their proposed strategies as opposed to donating their labour (17%), being charged a pollution tax (15%) or the use of amenity bills (6%). As such, the use of the special trust fund in the study is found to be widely accepted among respondents.

Variable

Variable description and type

Mean

Std. error

Min

Max

Sign

Age

Age of the respondent in years (categorical: 1= <20; 2=20-30; 3=31-40; 4=41-50; 5=51-60; 6=above 60).

3.04

0.05

Gender

Gender of the respondent (dummy variable: 1=male; 0=otherwise).

0.62

0.02

Education

Education of the respondent (categorical: 1= no education; 2=standard; 3=secondary; 4= tertiary; 5=university).

3.61

0.05

Household income

Average annual earnings in Kshs. (categorical: 1= <10,000; 2=10,001-20,000; 3=20,001-30,000; 4=30,001-40,000; 5=40,001-50,000; 6=50,001-60,000; 6=above 60,000).

2.86

0.07

Household size

Number of family members in the household (categorical: 1=1; 2=2; 3=3; 4=4; 5=5; 6=above 5 members).

3.49

0.06

Distance to nearby roads

Distance from nearby road in meters (categorical: 1= <50; 2=51-100; 3=101-150; 4=151-200; 5=201-250; 6= above 251).

4.78

0.08

Vehicle ownership

Whether or not the respondent owns a vehicle (dummy variable: 1=own; 0=otherwise).

0.18

0.02

Future income certainty

Whether or not the respondent is certain about future incomes (dummy variable: 1=certain; 0=otherwise).

0.83

0.02

Area of residence

Whether or not the respondent resides in the urban area or in the suburbs (dummy variable: 1=urban; 0=otherwise).

0.69

0.02

Table 1. Socio-demographic characteristics of the respondents and their measurement 21

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

Table 2 presents the analysis of various types of WTP responses derived from the study. The survey included a total of 488 respondents. Out of the total number, about 417 respondents (85%) indicated a positive WTP for motorized emission reductions and 71 respondents (15%) gave a zero WTP value. In order to separate protest responses from true zeroes, a closed-ended debriefing question was presented to respondents to justify why they had a zero WTP for air quality improvements. Thus, four possible alternatives were presented to respondents, namely: a) because air quality improvement has no value to me, b) because it is the responsibility of the government, c) because of many other basic financial commitments and, d) because it is the responsibility of motor vehicle owners. WTP Responses

Frequency

Positive willingness to pay responses True zero willingness to pay responses

417

Analytical sample size

435

Protest responses

Total sample size

488

100

Table 2. Summary of willingness to pay responses

Following Strazzera et al. (2003), the first (a) and the third (c) responses were classified as true zero values, while the other two as protest responses since they did not address the value of the public good in question, but some objection as to who should really pay for air quality improvements. Based on the

Figure 1. The distribution of willingness to pay responses 22

above-given classification, 18 respondents (4%) gave a true zero WTP value, while 53 (11%) gave a protest response. In line with the standard practice in valuation studies (Brouwer, 2009; Wang, 1997; Wang & Whittington, 2000; Whitehead et al., 1998), the protest responses were excluded from the analysis. Therefore, only 435 responses, i.e., about 89% of the initial sample size, were subjected to further analysis. Figure 1 presents diﬀerent bid amounts used in the study together with the responses against each bid amount. Out of the 15 bids presented, respondents were supposed to indicate their maximum WTP for air quality improvements in Nairobi. The bids were truncated from below at Kshs. 0 and from above at Kshs. 2,000. As shown in the table, 18 (4%) respondents state a maximum WTP of Kshs. 0, while 4 (1%) respondents state Kshs. 1,500. On the whole, most respondents state Kshs. 500 (14%), Kshs. 100 (13%), Kshs. 200 (9%) and Kshs. 400 (8%) as their individual maximum WTP. Mean willingness to pay and the determinant factors As shown in Table 3, the study found that respondents are, on average, willing to pay Kshs. 396.57 ($4.67) with a median value of Kshs. 244.94 ($2.88) to improve air quality management in the city. In order to assess the factors influencing individuals WTP, several sociodemographic characteristics of respondents, namely, age, gender, income, distance resided by respondents from a nearby road, motor vehicle ownership, respondents certainty about future income and area of residence, were regressed against the grouped data on WTP. Further results of the interval regression analysis are shown in Table 3.

EJAE 2015  12 (2)  16-26

Variable Age

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

Coeﬃcient

Std. errors

-0.118***

0.011

Gender

0.375***

0.028

Household income

0.531***

0.052

-0.020

0.272

-0.192*

0.043

0.196*

0.034

0.442*

0.075

3.872***

0.221

Distance to nearby road Vehicle ownership Certainty of future income Area of residence Constant Log likelihood Number of observations

-994.21 435

LR chi2 (7)

280.17

Probability > chi2

0.0000

Mean WTP (in Kshs.)

396.57

Standard error

17.06

Median WTP (in Kshs.) Bootstrapped 95% confidence intervals

244.94 361.11 - 432.03

Note: *, ** and *** denote p < 0.1, p < 0.05 and p < 0.01.

Table 3. Interval regression results on factors explaining individual’s willingness to pay

As indicated in Table 1, it was hypothesized that age of respondents would either have positive or negative relationship with the individuals’ WTP. This is because older people may sometimes feel susceptible to the eﬀects of motorized emissions and, therefore, are more willing to pay than younger people. Older people may, on the other hand, have a lower WTP because of the trade-oﬀs they may have to make within the household, such as paying school fees. However, the study results show the negative relationship between the age of respondents and their WTP, which implies that younger people are more willing to pay for improved air quality management than older people. These results are similar to those by Carlson and Johansson-Stenman (2000) and Wang et al. (2006), who claim that younger individuals would pay more for air quality improvement plans than older individuals. However, Wang and Mullay (2006) came up with the contrary findings. Positive relationship between gender and the WTP was also hypothesized and the results came out as expected. It means that men are more willing to pay for improved air quality management than women. The most probable reason for this is the fact that men

have more access to information than women and would therefore be more informed about the negative effects of vehicular emissions than women, together with the need for emission controls. Another reason is that they also control budgets within the household. This outcome correlates well with the similar outcome in Carlson and Johansson-Stenman (2000) and Wang and Zang (2008). Household income is another important variable used in the study to explain individuals’ decisions to pay for improved air quality management. It was expected that individuals with higher incomes would be more willing to pay for emission reductions than those with lower incomes, which would conform to economic theory (Loomis & Ekstrand, 1998). The results are positive and they theoretically validate the outcome of the study. A similar outcome is found in many other CV studies including Carlson and Johansson-Stenman (2000), Wang et al. (2006), Wang and Mullay (2006) and Wang and Zang (2008). The distance variable was used to assess whether or not the WTP would vary with distance since the perceived effects of motorized emissions were likely to vary with distance from a nearby road. The findings give evidence of negative, though statistically insignificant relationship meaning that residents living closer to nearby roads are more willing to pay for improved air quality management than their counterparts residing further away from the road. This finding can be attributed to the fact that people residing closer to the road have the perception that they are more affected by emissions than their counterparts residing further away from roads and, therefore, are more willing to pay for air quality improvements. Studies by Carlson and Johansson-Stenman (2000) and Wang et al. (2006) also reveal similar findings that people living in high polluted areas are more willing to pay than those living in the low polluted areas. Negative correlation was expected between motor vehicle ownership and the WTP, because of the public willingness to improve air quality. The results came out as expected as motor vehicle owners are found to be more willing to pay for air quality improvements than non-vehicle owners. However, the results differ from those by Firoozzarea and Ghorbani (2011) who found out that car owners are more willing to pay than non-car owners. Notably, the share of car owners in this study was quite low, which could partly explain why these results differ. The study also examines whether or not individuals’ certainty about their future incomes would have a positive effect on the WTP variable and, as predicted, 23

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

it was found that individuals who are certain about their future incomes are more willing to pay than those with uncertain future incomes. Similar findings are also reported in Fonta et al. (2010). Finally, the study also reveals positive relationship between the area of residence and individuals’ WTP. As in Wang et al. (2006), it means that urban residents are more willing to pay for improved air quality management than their counterparts in the suburbs. CONCLUSIONS

This study is conducted with the aim to analyze individuals’ preferences for improved air quality management through reduced motorized emissions in the city of Nairobi, Kenya, based on the responses from the contingent valuation PC format. The research was inspired by the need to estimate the policy value of implementing air quality improvement programme for the city of Nairobi since air quality problems were on the rise due to the rapid urbanization experienced over the last decade. Initial findings have shown that people in Nairobi are well aware of the air pollution problems in the city upon which they identify motor vehicles as the primary cause of air quality problems owing to the emission of toxic gases and dust particles into the atmosphere. Nairobi residents have also been found to be familiar with adverse health and environment eﬀects of motorized emissions and, as a result, most of them are willing to pay positive amounts towards emission reductions in the city. While a few people are willing to pay true zero amounts towards the same course citing overwhelming financial commitments within the household, others give protest responses against the air quality improvement plan, saying that the government and/or the motor vehicle owners should bear the responsibility of the air clean-up plans. In monetary terms, individuals in the study are, on average, willing to pay Kshs. 396.57 ($4.67) for improved air quality management. The median WTP is Kshs. 244.94, which is equivalent to $2.88. Moreover, age, gender, income, motor vehicle ownership, certainty about future income and the area of residence are found to have significant eﬀects on peoples’ WTP decision for air quality improvements. Since air quality issues continue to worsen in Nairobi due to increased motorization, the city authorities can now use the estimated mean and median WTP to benchmark their budget and policy proposals for motorized emission reductions. Based on the study findings, these budget and policy proposals can also 24

be adjusted to the socio-demographic characteristics of individuals, as they have been found to be important determinants of the peoples’ WTP decision. The valuation estimates can also be used to determine the economic eﬃciency of other air quality improvement programmes in the city of Nairobi and beyond, since peoples’ preferences are evident and determinate. All in all, more studies are required to further our understanding of the policy values of tackling specific issuess (e.g. respiratory diseases, damage to city buildings and contamination of the city water dam) that arise from motorized emissions. Such studies may provide varied additional information to decision makers on how to deal with diﬀerent air quality problems in a developing country context. Acknowledgements: This study was funded by the Center for Environmental Economics and Policy in Africa (CEEPA), University of Pretoria, the Republic of South Africa, and the Centre for International Cooperation (CIC), Vrije Universiteit Amsterdam, the Nethrelands. REFERENCES

Ababio, O.Y. (2003). New school chemistry. Onitsha: Africana-FEP. Ackerman, M., Davies, T., Jeﬀerson, C., Longhust, J., & Marquez, J. (2002). Comparison of diesel and hybrid vehicle emissions by computer modelling. Urban Transport, VIII, 471-480. doi:10.2495/UT020461. Amin, A.T.M.N. (2009). Reducing emissions from private cars: Incentive measures for behavioural change. Nairobi, Kenya: UNEP. Retrieved Jun, 15, 2015, from http://www. unep.ch/etb/publications/Green%20Economy/Reducing%20emissions/UNEP%20Reducing%20emissions%20 from%20private%20cars.pdf Arrow, K., Solow, R., Portney, P.R., Leamer, E.E., & Radner, R.H. (1993). Report on the NOAA Panel on contingent valuations, natural resource damage assessment under the Oil Pollution Act of 1990. Federal Register, 58, 46014614. Asian Development Bank. (2003). Reducing vehicle emissions in Asia. Manila, Philippines: Asian Development Bank. Bateman, I., Carson, R.T., Day, B., Hanemann, W.M., Hanley, N., & Hett, T. (2002). Economic valuation with stated preferences: A Manual. Cheltenham, UK: Edward Elgar. Brouwer, R. (2009). Stated preference uncertainty: signal or noise? Amsterdam: Institute for Environmental Studies, VU University Amsterdam. Cadle, S.H., Croes, B.E., Minassian, F., Natarajan, M., Tierney, E.J., & Lawson, D.R. (2004). Real-world vehicle emissions: A summary of the Thirteenth Coordinating Research Council on-road Vehicle Emissions Workshop.

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

Journal of the Air and Waste Management Association, 54(1), 18-23. Cameron, T.A., & Huppert, D.D. (1989). OLS versus ML estimation of non-market resource values with payment card interval data. Journal of Environmental Economics and Management, 17(3), 230-246. doi:10.1016/00950696(89)90018-1. Carlsson, F., & Johansson-Stenman, O. (2000). Willingness to pay for improved air quality in Sweden. Journal of Applied Economics, 32(6), 661-669. doi:10.1080/000368400322273. Carson, R.T. (2000). Contingent valuation: a user’s guide. Environmental Science and Technology, 34(8), 1413-1418. doi:10.1021/es990728j. Central Bureau of Statistics - CBS. (2009). Economic Survey 2009. Nairobi, Kenya: Central Bureau of Statistics (CBS). Colvile, R.N., Hutchinson, E.J., Mindell, J.S., & Warren, R.F. (2001). The transport sector as a source of air pollution. Atmospheric Environment, 35(9), 1537-1565. doi:10.1016/S1352-2310(00)00551-3. Dillman, D.A. (2000). Mail and internet surveys: the tailored design method. New York: John Wiley & Sons. Donfouet, H., Cook, J., & Jeanty, P.W. (2013). The economic value of improved air quality in urban Africa: results from Douala, Cameroon. Environment and Development Economics, 20(5), 630-649. doi:10.1017/S1355770X14000552. Dziegielewska, D., & Mendelsohn, R. (2005). Valuing air quality in Poland. Journal of Environmental and Resources Economics, 30(2), 131-163. doi:10.1007/s10640004-1515-2. Fabian, H.G., & Vergel, K.N. (2002). Analysis of air pollution exposure of individuals in the road environment. Journal of the Eastern Asia Society for Transportation Studies, 4(5), 55-70. Fazal, S. (2006). Addressing congestion and transport-related air pollution in Saharanpur, India. Environment and Urbanization, 18(1), 141-154. Firoozzarea, A., & Ghorbani, M. (2011). Investigation of citizens’ WTP for Mashhad air pollution reduction: applying Heckit model. 5th Symposium on Advances in Science & Technology, Mashhad, Iran. Retrieved Jun, 15, 2015, from http://profdoc.um.ac.ir/articles/a/1022664.pdf Fonta, M., Ichoku, H.E., & Nwosu, E. (2010). Contingent valuation in community-based project planning: the case of Lake Bamendjim fishery restocking in Cameroon. African Economic Research Consortium, Nairobi, Research Paper 210. Retrieved Jun, 15, 2015, from https://www.africaportal.org/dspace/articles/contingent-valuation-communitybased-project-planning-case-lake-bamendjim-fishery Foster, V., & Mourato, S. (2003). Elicitation format and sensitivity to scope do contingent valuation and choice experiments give the same results? Environmental and Resource Economics, 24(2), 141-160. doi:10.1023/A:1022856329552. Freeman, A.M. (1993). The measurement of environmental and resource values: theory and methods. Washington: Resources for the Future.

Gwilliam, K. (2003). Urban transport in developing countries. Transport Reviews, 23(2), 197-216. doi: 10.1080/01441640309893. Gwilliam, K., Kojima, M., & Johnson, T. (2004). Reducing air pollution from urban transport. Washington, DC: World Bank. Haab, T.C., & McConnell, K.E. (2002). Valuing Environmental and Natural Resources: the Econometrics of Nonmarket Valuation. Cheltenham, UK: Edward Elgar. Hanley, N., Kriström, B., & Shogren, J.F. (2009). Coherent arbitrariness: on value uncertainty for environmental goods. Land Economics, 85(1), 41–50. doi:10.3368/ le.85.1.41 Kinney, P.L., Gatari, M., Volavka-Close, N., Ngo, N., Ndiba, P.K., Law, A., Gachanja, A., Gaita, M.S., Chilrud, S.N., & Sclar, E. (2011). Traﬃc impacts on PM 2. 5 air quality in Nairobi, Kenya. Environmental Science and Policy, 14(4), 369-378. doi:10.1016/j.envsci.2011.02.005. Klose, T. (1999). The contingent valuation method in health care. Health Policy, 47(2), 97-123. Lilley, L.C. (2000). A new approach to emissions inventory modelling - Assessing fuel and vehicle impacts on air quality. Advances in Air Pollution, 8, 389-398. doi:10.2495/AIR000391. Loomis, J., & Ekstrand, E. (1998). Alternative approaches for incorporating respondent uncertainty when estimating willingness to pay: The case of the Mexican spotted owl. Ecological Economics, 27(1), 29-41. doi:10.1016/S09218009(97)00126-2. Mendelsohn, R., & Olmstead, S. (2009). The Economic valuation of environmental amenities and disamenities: methods and applications. Annual Review of Environment and Resources, 34, 325-347. doi:10.1146/annurevenviron-011509-135201. Mitchell, R., & Carson, R. (1993). Using surveys to value public goods: the contingent valuation method. Washington, DC: Resources for the Future. Mitchell, R.C., & Carson R.T. (1989). Using surveys to value public goods: the contingent valuation method. Baltimore: Johns Hopkins University Press. Morrison, M.D., Blamey, R.K., & Bennett, J.W. (2000). Minimizing payment vehicle bias in contingent valuation studies. Environmental and Resource Economies, 16(4), 407-422. doi:10.1023/A:1008368611972. Mulaku, G.C., & Kariuki, L.W. (2001). Mapping and analysis of air pollution in Nairobi, Kenya. International Conference on Spatial Information for Sustainable Development. Nairobi, Kenya. Retrieved Jun, 15, 2015, from https://www.fig.net/resources/proceedings/2001/nairobi/ mulaku-kariuki-TS3-2.pdf Odhiambo, G.O., Kinyua, A.M., Gatebe, C.K., & Awange, J. (2010). Motor vehicles air pollution in Nairobi, Kenya. Research Journal of Environmental and Earth Sciences, 2(4), 178-187. 25

EJAE 2015  12 (2)  16-26

Ndambiri, H., Mungatana, E., Brouwer, R.  Stated preferences for improved air quality management

Omwenga, M. (2011). Integrated transport system for liveable city environment: a case study of Nairobi Kenya. 47th ISOCARP Congress. Retrieved Jun, 15, 2015, from http:// www.isocarp.net/data/case_studies/2022.pdf Ryan, M., Scott, D.A., & Donaldson, C. (2004). Valuing health care using willingness to pay: a comparison of the payment card and dichotomous choice methods. Journal of Health Economics, 23(2), 237-258. doi:10.1016/j. jhealeco.2003.09.003 Sayadi, S., Gonzalez-Roa, M.C., & Calatrava-Requena, J. (2009). Public preferences for landscape features: the case of agricultural landscape in mountainous Mediterranean areas. Land Use Policy, 26(2), 334-344. doi:10.1016/j. landusepol.2008.04.003. Strazzera, E., Genius, M., Scarpa, R., & Hutchinson, G. (2003). The eﬀect of protest votes on the estimates of WTP for use values of recreational sites. Environmental and Resource Economics, 25(4), 461-476. doi:10.1023/A:1025098431440. Venkatachalam, L. (2004). The contingent valuation method: a review. Environmental Impact Assessment Review, 24(1), 89-124. doi:10.1016/S0195-9255(03)00138-0. Vliet, E.D.S., & Kinney, P.L. (2007). Impacts of roadway emissions on urban particulate matter concentrations in sub-Saharan Africa: new evidence from Nairobi, Kenya. Environmental Research Letters, 2(4), 1-6. doi:10.1088/17489326/2/4/045028. Wang, H., & Whittington, D. (2000). Willingness to pay for air quality improvement in Sofia, Bulgaria. World Bank Policy Research Working Paper No 2280. Washington, DC: World Bank.

Wang, H., & Whittington, D. (2005). Measuring individuals’ valuation distributions using a stochastic payment card approach. Ecological Economics, 55(2), 143-154. doi:10.1016/j.ecolecon.2004.11.011. Wang, X.J., Zhang, W., Li, Y., Yang, K.Z., & Bai, M. (2006). Air quality improvement estimation and assessment using contingent valuation method: a case study in Beijing. Environmental Monitoring and Assessment, 120(1), 153168. doi:10.1007/s10661-005-9054-z. Welsh, M.P., & Poe, G.L. (1998). Elicitation eﬀects in contingent valuation: comparisons to a multiple bounded discrete choice approach. Journal of Environmental Economics and Management, 36(2), 170-85. doi:10.1006/ jeem.1998.1043. White, P.C.L., Bennett, A.C., & Hayes, E.J.V. (2001). The use of willingness-to-pay approaches in mammal conservation. Mammal Review, 31(2), 151-167. doi:10.1046/ j.1365-2907.2001.00083.x. Whitehead, J., Blomquist, G., Ready, R., & Huang, J. (1998). Construct validity of dichotomous choice contingent valuation questions. Environmental and Resource Economics, 11(1), 107-116. doi:10.1023/A:1008231430184. WHO. (2006). Air Quality Guidelines: Global Update 2005. Geneva: World Health Organization. Retrieved Jun, 15, 2015, from http://www.who.int/phe/health_topics/outdoorair/outdoorair_aqg/en/

VREDNOVANJE ISKAZANIH PREFERENCIJA ZA UNAPREĐENJE KVALITETA VAZDUHA U GRADU NAJROBI U KENIJI Rezime: U ovom istraživanju koristi se metod uslovnog vrednovanja (CV) kako bi se ispitale preferencije/stavovi ispitanika vezano za efikasnije upravljanje kvalitetom vazduha kroz smanjenje emisija gasova iz motornih vozila u gradu Najrobi u Keniji. Sistem platne kartice (PC) kao poboljšanji oblik upitnika olakšava proces vrednovanja stavova ispitanika u cilju određivanja medijane i srednje vrednosti koju su ispitanici voljni da plate kako bi doprineli poboljšanju kvaliteta vazduha u gradu. Primenom regresione analize ustanovljeno je da su ispitanici spremni da plate Kshs. 396.57 ($4.67) i srednju vrednost od Kshs. 244.94 ($2.88) kako bi se poboljšao kvalitet vazduha u gradu. Primetno je da su ove vrednosti u porastu kod pripadnika muške populacije, osoba sa većim primanjima, izvesnim prihodima na duži vremenski period i osoba koje žive u urbanom području, dok njihova vrednost opada sa godinama, udaljenošću od obližnjih puteva, i upotrebom motornih vozila. Generalno posmatrano, ovo istraživanje ukazuje na ogromnu podršku javnosti kada je reč o unapređenju kvaliteta vazduha u gradu, što je od presudnog značaja za efikasan razvoj i sprovođenje programa upravljanja kvalitetom vazduha.

Ključne reči: kontigentno (uslovno) vrednovanje, spremnost za plaćanje (WTP), sistem platne kartice, kvalitet vazduha, Najrobi. Received: September 9, 2015. Correction: October 2, 2015. Accepted: October 14, 2015.

EJAE 2015, 12(2): 27-34 ISSN 2406-2588 UDK: 338.48-6:641/642(497.113) DOI: 10.5937/ejae12-9139 Original paper/Originalni naučni rad

OFFER OF AUTHENTIC FOOD AS A CONDITION FOR GASTRONOMIC TOURISM DEVELOPMENT Bojana Kalenjuk1*, Dragan Tešanović1, Snježana Gagić2, Irma Erdeji1, Maja Banjac1 1 2

Faculty of Sciences, University of Novi Sad, Serbia College of Professional Studies in Management and Business Communications, Sremski Karlovci, Serbia

Abstract: By perceiving the food as an important part of the tourism offer in terms of biological needs of tourists, the question arises as to how it can be further exploited to contribute to tourism development. In this regard, food prepared using the local ingredients, according to the old methods of processing and serving, and in an authentic way, can be of interest to the tourists of selective forms of tourism, also known as food tourism, gastronomic tourism and gourmet tourism. The research aims to demonstrate the quality of food offer in the hospitality industry of Vojvodina and examine whether it can meet the requirements of foreign tourists who travel for it. The survey was conducted in 147 catering facilities in Vojvodina offering food and beverages, by means of direct and indirect collecting of written offers, and includes the statistical analysis of 10,923 meals. The aim is to indicate the current state of gastronomic offer and all advantages of authentic food and gastronomic tourism.

INTRODUCTION The regions in which the wave of technology and globalization did not change the way food and beverages are served and prepared, including the consumption culture, have become favorable for development of selective forms of tourism, also known as food and drink tourism, culinary tourism, gastronomic tourism, gustatory tourism and gourmet tourism (Hall & Mitchell, 2002). These forms represent a special form of tourism whose aim is to attract tourists motivated by food and drinks in different ways, thus making food an important tool of the tourism industry (Hall & Mitchell, 2000), which in order to develop and grow further has to offer something different. The focus of gastronomic tourists are local and national dishes (Kalenjuk et al., 2012a, 2012b; Kalenjuk et al., 2014), which as such are the reflection of a people’s culture. * E-mail: bojanakalenjuk@yahoo.com

Key words: food, gastronomy, tourism, gastronomic tourism.

Gastronomic tourism is an experience pursued by visiting and tasting food and beverages at primary or secondary producers’, visiting food and beverage festivals and consuming authentic specialties in various restaurants. An increasingly popular form of gastronomic tourism is an activity of tourists who attend diﬀerent multi-day schools of authentic regional cooking (Hall & Mitchell, 2006). The authors define it as a journey during which the purchase or consumption of regional food (and drink), or observation and study of food production (from agriculture to culinary school), represents an important motive or activity (Ignatov & Smith, 2006). Culinary tourists visit places that oﬀer unique culinary experiences. They love to get into their host’s kitchen, inquire about the typical dishes, and learn where they can purchase ingredients and spices needed for a particular meal (Kalenjuk et al., 2012a). 27

EJAE 2015  12 (2)  27-34

Kalenjuk, B. et al.  Offer of authentic food

The popularity of tourism for food and beverage has been known in the world for several decades (Santich, 2004). However, it has recently become a trend in Vojvodina. In the previous development policies and concepts related to tourism development, gastronomy i.e. food of locals and tourists, was not given the adequate development role, as opposed to its basic merits (Kalenjuk et al., 2013). However, the development and standards of food for tourists are not always determined by tourism policy, but national, economic, agricultural and food policies (Hjalager & Corigliano, 2000). Food and tourism are integral parts of agro-tourism, wine tourism, through sale of agricultural, alimentary and gastronomic products. The research will show the quality of food in the hospitality industry of Vojvodina, and whether it is able to please the interests of foreign tourists traveling for its sake. The aim of the paper is to draw attention to the current state of gastronomy, as the basis for development of gastronomic tourism in the region.

LITERATURE REVIEW

Studies have shown that among all the possible costs during the trip, tourists are least willing to reduce the budget intended for food. This suggests that tourist food consumption makes a significant contribution to local restaurants, steakhouses, bakeries, pastry shops, agriculture and food industries, and thus the economy of the destination (Pyo et al., 1991). This is not only about the consumption of culinary products, which are, as such, in their final form, offered to tourists, but also about the consumption of agricultural and food products in the region (Kalenjuk et al., 2011). Food can also be the main instigator and primary motive that meets a multitude of needs and desires (Tikkanen, 2007). Food has been recognized on the tourism market as (Hall & Mitchell, 2001): ◆ part of the local culture, consumed by tourists; ◆ part of the tourism promotion; ◆ potential component of the local agricultural and economic development, and ◆ regional factor aﬀected by consumption patterns and perceived desires of tourists.

Relationship between food and tourism

Gastronomic tourism

Tourism is a branch of economy which has significant social and economic functions, and is an integral factor of socio-cultural development. Tourist activation of certain space contributes to highlighting and better evaluation of other values of the environment in which food is an essential resource (Henderson, 2004; Quan & Wang, 2004). The authors note the particular importance of food as a physiological need, considering that expenses for food and drink in tourism account for one third of total tourist spending of global tourist traﬃc (Meler & Cerović, 2003). Therefore, the relationship between food and tourism is one of the most important issues in the region. Beardsworth and Keil explain that nutritional intake is not the only goal of food consumption, and add that it also includes a gustatory experience and interaction with other people during consumption (Beardsworth & Keil, 1997). Thus, the use of food for special occasions and festivals can represent the cultural and spiritual identity of people. Furthermore, its role is associated with the status and social classes of individuals, which is often closely related to the type of restaurants they visit.

Hall and Mitchell define gastronomic tourism as visiting primary or secondary food producers, food festivals, catering facilities for production and serving of food and certain destinations for tasting and/or experiencing attributes of regional specialties, where food is the primary motive to travel (Hall & Mitchell, 1996). Gastronomic tourism is defined as the activity of exploration and discovery of the culture and history through food, which aﬀects the formation of unforgettable experiences (Long, 2004). Thus, gastronomy connects and incorporates other forms of tourism (Stojanović & Čerović, 2008). Various terms are related to diﬀerent forms of tourist movements motivated by food and beverages, among which are: ◆ gastronomic tourism (Hjalanger & Richards, 2002); ◆ culinary tourism (Wolf, 2006; Ignatov & Smith, 2006); ◆ gustatory tourism (Boniface, 2003) and ◆ food tourism (Hall et al., 2003). By definitions, the “culinary” focuses only on styles of food preparation while “gastronomy” deals with the

EJAE 2015  12 (2)  27-34

consumption of food and beverages in a more general sense and tasting good food and beverages as part of a luxurious lifestyle. The word “culinary” (culinarius) tends to emphasize the style of food preparation and consumption, as well as the social context in which food is ordered, prepared and consumed. Therefore, “culinary” can refer to ingredients, prepared food, beverages, food production, motives, activities, institutional structures and food tourism itself (Ignatov & Smith, 2006). The authors note that gastronomic tourism is different from agro-tourism as it is a subset of cultural tourism (cuisine is part of the culture), and agrotourism is a subset of rural tourism, but also a subset of gastronomic tourism. Surely, gastronomic tourism and agro-tourism are inextricably linked. Similarly, some authors consider wine and beer tourism as subsets of gastronomic tourism because this type of tourism includes the consumption of beverages, both alcoholic and non-alcoholic (Hall, 2003; Wolf, 2006), such as tea tourism, although wine tourism is studied independently. Unlike many other forms of tourist movements, gastronomic destinations are usually available yearround at any time of the day and in all weather conditions (Richards, 2002), noting the importance of authentic food offer in restaurants. Structure of food offer in catering industry Only authentic offer of dishes can satisfy the curiosity of gastronomic tourists, such as local and regional dishes. The authors classify the dishes included in the offer of catering facilities, on the basis of their origin, as: local, national and international, bearing in mind that the ratio of all three should be equal, as it is the only way in which it is possible to meet the demand of all types of consumers (Tešanović et al., 2010; Kalenjuk et al., 2012b). The kind of dishes that should, by all criteria, be included in catering facilities’ offer and attract the attention of tourists motivated by food, are in the first place local dishes, followed by national (Kalenjuk et al., 2012c). Local dishes are either prepared in households or are the specialties of a restaurant and a city, being made using the authentic, local foodstuffs. The examples would be “stuffed steak by chef Marko”, “Leskovački voz”, Novi Sad steak, Temerin steak, a specialty of Salaš 132 etc. These are usually named after a location or the creator of the recipe.

Kalenjuk, B. et al.  Offer of authentic food

National or regional dishes are those which have taken root in a wide area of a country. There are many dilemmas over the food origin. Various nations have historically won other territories and brought along their own eating habits, culinary recipes, starting to make the dishes with ingredients from the conquered territory, but adding to them their own spices. Locals adopted such dishes as their own and passed them on from generation to generation and the dishes thrived. The dishes that have taken hold of a broader territory of a country, regardless of their country of origin, alongside the dishes that were developed on the country’s territory, are considered national dishes. Such dishes include sarma, goulash, stew, roast pork, ćevapčići, Karađorđe’s steak, cheese pie, burek and many other Serbian dishes. It often happens that a local dish is a national one as well, such as Leskovac grill. Moreover, a dish can be local, national and international at the same time, such as Viennese and Parisian steak (Tešanović et al., 2010). International dishes are already affirmed dishes of various international cuisines that are worldwide accepted including: Tournedos Rossini, Italian minestrone, pizza, steak “Esterhazy”, Kiev cutlet and countless other recognizable dishes in the global hospitality industry. They are prepared in restaurants for the tourists without an adventurous gastronomic spirit and those who do not like to experiment and prefer to eat local food during their stay in some other county. Food offer is considered adequate if the facility has at least 30% of local and 30% of national products, while the rest is comprised of those of international origin. In order to ensure the development of gastronomic tourism, it is better to offer more local and national dishes and beverages (Tešanović & Koprivica, 2007). METHODOLOGY In order to obtain necessary information on the structure of food offered by catering facilities, direct and indirect collection of written offers (menus) was conducted by means of random sampling, during the period from March 2011 until March 2013. The analysis and synthesis was performed on a sample of 10,923 meals in 147 catering facilities throughout Vojvodina. Based on the origin, dishes were grouped into local, national and international, after which a statistical observation was performed. Statistically processed data, i.e. absolute frequency, relative frequency, minimum value, maximum 29

EJAE 2015  12 (2)  27-34

Kalenjuk, B. et al.  Offer of authentic food

value, means, standard deviation, coefficient of variation along with discussion are presented in the paper in the form of a table and a graph. RESULTS AND DISCUSSION The analysis of authentic food offer McKercher et al. (2008) emphasize the importance of local, authentic dishes and formation of a unique offer in catering facilities for successful development of gastronomic tourism (McKercher et al., 2008). Table 1 shows the data obtained by analyzing meals per groups in 147 catering facilities, with a minimum and maximum number of dishes, standard deviation, and coefficient of variation for the observed group of dishes, depending on the origin of the recipe. By analyzing the structure, it was concluded that local dishes do not have a satisfactory share in the offer of any food group and that these dishes are represented in a significant percentage in the offer of cold appetizers with 20.86% (n=199), with restaurants often offering homemade dried meat products and vegetables. International appetizers are most frequent with 44.17% (n=429) and are composed of prosciutto served with melon, salmon carpaccio, caviar, cheese platters. National specialties, primarily „Vojvođanska zakuska“, are represented with a share of 34.17% (n=326), which is quite satisfactory. In addition to cold appetizers, ready-made meals of various types of meat have a similar share of 19.40% (n=77), while international dishes are represented in a small percentage of 21.41% (n=407). Ready-made meals offer is dominated by dishes of national origin, 59.19% (n=235), which can be considered satisfactory. National dishes are represented in the offer of fish dishes with significant share of 52.97% (n=472) dominated by different specialties of river fish prepared in different ways, international dishes are represented with 40.74% (n=363), while local dishes account for the minimum of 8.66% (n=56). This is followed by salads with 50.74% (n=888), including „Shopska“ and „Serbian salad“ as the most common ones, although these are not from Vojvodina, but are prepared and offered in addition to chutney, mixed salad and others (international with 45.89% (n=803), local salads are represented by only 3.37% (n=59)). Significant representation of national dishes in the offer of dishes to order with 48,79% (n=1767) (in30

ternational 41.28% (n=1495), local 9.94% (n=360)), soups and broths 49.21% (n=280) (international 38.66% (n=222), local 12.13% (n=407)) and desserts 45.17% (n=533) (international 51.02% (n=602) , local 3.81% (n=45)). The largest share of international dishes is observed in the offer of side dishes and stews with 65.65% (n=407). It is not in favor of the development of culinary tourism that there is a great share of French fries in almost all restaurants, with a small share of other side dishes and stews based on potato, such as mashed potato. National side dishes and stews offer is twice lower and accounts for 29.68% (n=184), whereas side dishes and stews prepared according to the authentic recipes are represented to a very small extent, with a share of only 4.68% (n=29). In the offer of hot appetizers, international specialties account for a considerable share of 68.66% (n=644). The restaurants have a satisfactory offer of cheese, au gratin pancakes and various types of pasta prepared according to Italian recipes, Milanese and Bolognese, ravioli. The research included grouped pizzas as well as pasta dishes, which contribute to the increase in the share of international dishes. National hot appetizers are represented by 25.80% (n=242) and local hot appetizers by only 5.54% (n=52). Based on the number of meals within the catering facilities and coefficient of variation, we have observed considerable differences in the offer of all of the analyzed groups of dishes with a great variation in the offer of local dishes, dominated by the offer of hot appetizers with 294.44%, fish dishes with 282,05% and side dishes and stews with 340.00%, which is not justifiable as Vojvodina is the breadbasket of the country, with rivers full of fish that caterers can use to prepare a lot of authentic, regional specialties. Based on the obtained data, it was noted that none of the groups of dishes within the entire region has satisfactory offer structure of dishes of 30%. The share of national dishes is satisfactory for certain groups, such as soups and broths, ready-made meals and meals to order, but the share of local specialties is not satisfactory in terms of the gastronomic tourism development (Figure 1). The analysis of the overall offer of authentic food By analyzing the overall offer of meals in the surveyed catering facilities in Vojvodina, it was conclud-

EJAE 2015  12 (2)  27-34

Kalenjuk, B. et al.  Offer of authentic food

CATEGORY

COLD APPETIZERS

ORIGIN

Absolute frequency

Relative frequency

Min

Max

international

429

44,17

2,98

3,23

108,39

national

326

34,17

2,26

1,65

73,01

local

199

20,86

1,38

1,43

103,62

Total: SOUPS AND BROTHS

954 international

222

38,66

1,53

1,24

81,05

national

280

49,21

1,94

1,7

87,63

local

12,13

0,48

0,83

172,92

Total: HOT APPETIZERS

571 international

644

68,66

4,47

6,22

139,15

national

242

25,80

1,68

1,53

91,07

local

5,54

0,36

1,06

294,44

Total: READY - MADE MEALS

938 international

21,41

0,59

1,31

222,03

national

235

59,19

1,63

3,07

188,34

local

19,40

0,53

1,19

224,53

Total: MEAT MEALS TO ORDER

397 international

1495

41,28

10,38

6,53

62,91

national

1767

48,79

12,27

6,76

55,09

local

360

9,94

2,50

2,57

102,80

Total : FISH MEALS TO ORDER

3622 international

363

40,74

2,52

2,63

104,37

national

472

52,97

3,28

2,67

81,40

local

8,66

0,39

1,1

282,05

Total: SIDE DISHES AND STEWS

891 international

407

65,65

2,83

2,13

75,27

national

184

29,68

1,28

1,67

130,47

local

4,68

0,20

0,68

340,00

Total: SALADS

620 international

803

45,89

5,58

2,94

52,69

national

888

50,74

6,17

3,89

63,05

local

3,37

0,41

0,73

178,05

Total: DESSERTS Total:

1750 international

602

51,02

4,18

3,01

72,01

national

533

45,17

3,70

2,57

69,46

local

3,81

0,31

0,75

241,94

1180

Table 1. The structure of food oﬀer in the hospitality industry of Vojvodina 31

EJAE 2015  12 (2)  27-34

Kalenjuk, B. et al.  Offer of authentic food

ed that there is a large share of international dishes, and national dishes, 45.73%, and 43.92%, respectively, which is quite positive. The share of local dishes and creations by catering facilities and population is unsatisfactory and accounts for only 10.35% (Fig. 2). Wood and Munoz state that tourists who visit the restaurants with national motives, seek authentic, ethnic and cultural experiences (Wood & Munoz, 2007). Therefore, the authenticity of the offers may have a decisive impact on their overall satisfaction. If we take into account the recommendations of the authors, it can be concluded that the overall offer at the regional level is partially satisfactory, because of the satisfactory number of national specialties in the offer of side dishes (pasta dishes), ready-made meals and dishes to order (meat and river fish).

Figure 1. Food structure according to origin of recipe

Figure 2. The overall structure of meals in Vojvodina 32

CONCLUSION By consulting international literature in the field of gastronomic tourism, gastronomy and hospitality industry, and keeping up with the research conducted in the catering facilities in Vojvodina, we may conclude that the oﬀer of dishes in the entire region has a large share of dishes of international and national origin, with nearly 46%, and 44%, respectively, which is quite satisfactory. The share of local dishes and creations by catering facilities is unsatisfactory and accounts for only 10%. By studying and following trends, places and roles of authentic food of developed tourism destinations worldwide, we may notice that Vojvodina faces with the growing issue of the non-systemic, insuﬃcient

EJAE 2015  12 (2)  27-34

approach to a more successful placement of food resources in the active function of tourism development. By observing the region as a multicultural destination, a breadbasket of the country with extensive culinary heritage, restaurateurs are expected to introduce authentic dishes made with local ingredients, whose combinations of spices and methods of preparation and serving would intrigue and attract foreign tourists motivated by food. In order to ensure a successful development of gastronomic tourism, catering facilities on the territory of Vojvodina are requested to introduce local, regional, authentic delicacies for the sake of achieving the appropriate share of oﬀer at the minimum level of 30%. Acknowledgements: This work is part of the research project: III-046009 and III-046005 funded by the Ministry of Science and Technological Development of the Republic of Serbia. REFERENCES Beardsworth, A., & Keil, T. (1997). Sociology on the menu: An invitation to the study of food and society. London: Routledge. Boniface, P. (2003). Tasting Tourism: Travelling for Food and Drink. Aldershot: Ashgate. Hall, C.M. (1996). Wine tourism in New Zealand. In: Towards a more sustainable tourism: Proceedings of Tourism Down Under II: A tourism research conference, 3-6 December 1996. (pp. 109-119). Dunedin, NZ: Centre for Tourism, University of Otago. Hall, C.M. (2003). Wine, Food, and Tourism Marketing. Binghamton, NY: Haworth Hospitality Press. Hall, C.M., & Mitchell, R. (2000). We are what we eat: Food, Tourism and Globalization. Tourism, Culture & Communication, 2, 29-37. Hall, C.M., & Mitchell, R. (2001). Wine and food tourism. In N. Douglas, N. Douglas, & R. Derrett (Eds.), Special Interest Tourism. Milton, Australia: John Wiley & Sons. Hall, C.M., & Mitchell, R. (2002). The changing nature of the relationship between cuisine and tourism in Australia and New Zealand: From fusion cuisine to food networks. In A.-M. Hjalager & G. Richards (Eds.), Tourism and Gastronomy. (pp. 188-206). London: Routledge. Hall, C.M., & Mitchell, R. (2006). Gastronomy, food and wine tourism. In D. Buhalis & C. Costa (Eds.), Tourism Business Frontiers: Consumers, products and industry. Oxford: Elsevier. Hall, C.M., Sharples, L., Mitchell, R., Macionis, N., & Cambourne, B. (2003). Food tourism around the world: Development, management and markets. Oxford: Butterworth- Heinemann.

Kalenjuk, B. et al.  Offer of authentic food

Henderson, J.C. (2004). Food as a tourism resource: A view from Singapore. Tourism Recreation Research, 29(3), 69-74. Hjalager, A.-M., & Corigliano, M.A. (2000). Food for tourists - determinants of an image. International Journal of Tourism Research, 2(4), 281-293. doi:10.1002/15221970(200007/08)2:4<281::AID-JTR228>3.0.CO;2-Y Hjalager, A.-M., & Richards, G. (2002). Tourism and Gastronomy. London: Routledge. Ignatov, E., & Smith, S. (2006). Segmenting Canadian Culinary Tourists. Current Issues in Tourism, 9(3), 235-255. doi:10.2167/cit/229.0 Kalenjuk, B., & Tešanović, B. (2013). Određivanje pravaca strategijskog razvoja gastronomskog turizma u Vojvodini. Ekonomika, 59(4), 195-202. In Serbian. Kalenjuk, B., Đerčan, B., & Tešanović, D. (2012). Gastronomski turizam kao faktor regionalnog razvoja. Ekonomika, 58(3), 136-146. In Serbian. Kalenjuk, B., Tešanović, D., & Gagić, S. (2014). Regionalni razvoj turističkih destinacija kroz razvoj gastronomskog turizma. In: Od krize prema razvoju: 4. Međunarodna naučna konferencija. (pp. 363-374). BIH: Univerzitet za poslovne studije. In Serbian. Kalenjuk, B., Tešanović, D., Rudić, Lj., Korbatfinski, M., Gagić, S. & Vuksanović, N. (2012). Gastronomski potencijal severne Bačke u funkciji gastronomskog turizma. In: Zbornik radova Departmana za geografiju, turizam i hotelijerstvo. (pp. 337-349). Novi Sad: Prirodno-matematički fakultet. In Serbian. Kalenjuk, B., Tešanović, D., Škrinjar, M., & Đeri, L. (2012). The importance of authentic food in the development of the culinary tourism in Vojvodina. In: Belgrade International Tourism Conference - BITCO Contemporary tourism, wishes & opportunities. (pp. 293-300). Belgrade: College of tourism. Kalenjuk, B., Tešanović, D., Škrinjar, M., & Vuksanović, N. (2011). Gastronomski potencijali Vojvodine u funkciji razvoja turizma. In: Zbornik radova Departmana za geografiju, turizam i hotelijerstvo. (pp. 180-187). Novi Sad: Prirodno-matematički fakultet. In Serbian. Long, L.M. (2004). Culinary Tourism. Lexington: University Press of Kentucky. McKercher, B., Okumus, F., & Okumus, B. (2008). Food tourism as a viable market segment: It’s all how you cook the numbers. Journal of Travel & Tourism Marketing, 25(2), 137-148. doi:10.1080/10548400802402404 Meler, M., & Cerovic, Z. (2003). Food marketing in the function of tourist product development. British Food Journal, 105(3), 175-192. Sung, P.S., Uysal, M., & McLellan, W.R. (1991). A linear expenditure model for tourism demand. Annals of Tourism Research, 18(3), 443-454. doi:10.1016/01607383(91)90051-C 33

EJAE 2015  12 (2)  27-34

Kalenjuk, B. et al.  Offer of authentic food

Quan, S., & Wang, N. (2004). Towards a structural model of the tourist experience: An illustration from food experiences in tourism. Tourism Management, 25(3), 297-305. Richards, G. (2002). Gastronomy: An essential ingredient in tourism production and consumption. In A.-M. Hjalager & G. Richards (Eds.), Tourism and Gastronomy. (pp. 3-20). London: Routledge. Santich, B. (2004). The study of gastronomy and its relevance to hospitality education and training. International Journal of Hospitality Management, 23(1), 15-24. doi:10.1016/ S0278-4319(03)00069-0 Stojanović, T., & Čerović, S. (2008). Značaj gastronomske ponude za razvoj seoskog turizma Srbije. In: Zbornik radova: Geografski fakultet Univerziteta u Beogradu, br. 56. (pp. 165-176). In Serbian. Tešanović, D., Kalenjuk, B., & Vuksanović, N. (2010). Svečani obroci - značajan segment turističkog proizvoda na rečnim brodovima. Turističko poslovanje, 5, 251-262. In Serbian.

Tešanović, D., & Koprivica, M. (2007). Influence of quality definition of regional gastronomic products on formation of touristic offer of Serbia. (pp. 217-223). Ljubljana: Vocational College for Catering and Tourism. Tikkanen, I. (2007). Maslow’s hierarchy and food tourism in Finland: Five cases. British Food Journal, 109(9), 721-734. doi:10.1108/00070700710780698 Wolf, E. (2006). Culinary Tourism: The Hidden Harvest. Dubuque: Kendall/Hunt Publishing. Wood, N.T., & Munoz, C.L. (2007). No rules, just right’ or is it? The role of themed restaurants as cultural ambassadors. Tourism and Hospitality Research, 7(3-4), 242-255. doi:10.1057/palgrave.thr.6050047

PONUDA AUTENTIČNE HRANE KAO PREDUSLOV ZA RAZVOJ GASTRONOMSKOG TURIZMA Rezime: Posmatrajući hranu kao značajan deo turističke ponude, sa aspekta bioloških potreba turista, postavlja se pitanje, na koji način ona može biti dodatno iskorišćena, i tako doprineti razvoju turizma. Sa tim u vezi hrana pripremljena od lokalnih namirnica, po starim metodama obrade i njihovo serviranje na autentičan način mogu biti zanimljivi turistima selektivnog oblika turizma, u svetu poznatog kao turizam hrane, gastronomski turizam i gurmanski turizam. Istraživanje ima cilj da prikaže ponudu hrane u ugostiteljstvu Vojvodine i da li ona kao takva može da zadovolji interesovanja stranih turista koji putuju zbog nje. Istraživanje je sprovedeno u 147 ugostiteljskih objekata za pružanje usluga ishrane i pića na teritoriji AP Vojvodine, direktnim i indirektnim sakupljanjem pisanih ponuda, pri čemu je obuhvatilo statističku analizu od 10.923 jela. Cilj rada jeste da se ukaže na trenutno stanje gastronomske ponude i sve prednosti autentične hrane i gastronomskog turizma.

Ključne reči: hrana, gastronomija, turizam, gastronomski turizam. Received: September 1, 2015. Correction: September 15, 2015. Accepted: October 6, 2015.

EJAE 2015, 12(2): 35-43 ISSN 2406-2588 UDK: 657.631.6 334.728:657.6 DOI: 10.5937/ejae12-9214 Original paper/Originalni naučni rad

DEFINING SECOND-TIER AUDIT FIRMS IN SERBIA: IN SEARCH OF THE SMALL FOUR Nemanja Stanišić*, Vule Mizdraković, Tijana Radojević, Nenad Stanić Singidunum University, 32 Danijelova St., Belgrade, Serbia

Abstract: In this paper, we aim to define the so-called "second tier" audit firms in the Republic of Serbia. In order to accomplish that, we have analyzed 10,555 financial statements of legal entities operating on the territory of the Republic of Serbia for the period 2010-2012. The "Big 4" audit firms have audited 1,101 of the aforementioned financial statements, whereas the rest of the statements have been audited by the remaining 38 audit firms. For the purpose of training a decision tree classification algorithm, a number of calculated financial indicators of audited entities and corresponding audit opinions issued by the "Big 4" firms have been treated as independent and dependent variables, respectively. The resulting set of rules was then applied to the rest of the sample. Based on the observed level of compliance with the "Big 4" criteria (which stands for high quality audit), we have subsequently suggested the "Small 4" audit firms in Serbia. Moreover, the results show that the "Big 4" firms and the other audit firms assign different importance to independent variables. The paper also discusses possible reasons for that and offers recommendations for future research.

INTRODUCTION The recent financial crises have resulted in the bankruptcy of numerous entities and financial institutions globally. However, their financial statements should have shown some indices of higher financial risk prior to their default. For this reason, the public gaze is once again turned to auditors and audit firms. All major stakeholders are aware of the auditor’s role in disclosing whether the entity would carry on with its business or not. Empirical evidence from prior studies also shows that only about half of the public companies in the United States that subsequently entered into bankruptcy received a GCO (going-concern opinion) on the last set of financial statements filed prior to bankruptcy (Geiger et al., 2014). Much higher expectations have been laid upon the Big 4, since it is presumed that those auditors provide opinions of a higher quality due to their size and supposedly * E-mail: nstanisic@singidunum.ac.rs

Key words: decision trees, financial statements, audit opinion.

superior auditing methodologies (DeAngelo, 1981). Compared to other audit firms, they should make fewer Type I and Type II errors. The first type refers to issuance of modified opinion to healthy clients, while the second one refers to issuance of unmodified opinion to clients that have material misstatements in their financial reports or noteworthy going-concern issues. Both types are undesirable with respect to auditor’s reputation. Type I errors might lead to auditorswitching and loss of potential revenue, followed by extra costs for acquiring new clients. On the other hand, in cases of issuance of unmodified opinion to the client that went bankrupt, an auditor may face lawsuit costs. Second-tier audit firms in the United States gained their strength predominantly in the post SOX (Sarbanes-Oxley Act of 2002) era. The SOX has had two eﬀects on the audit service market: firstly, it increased the demand for high-quality auditing, especially from 35

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

publicly traded companies (SEC clients); and secondly, it temporarily eclipsed the Big 4’s reputation for competence. During the period 2003-2005, second-tier auditors observed a 23% (166) increase in SEC clients, whereas the Big 4 faced a decrease of 9% (652) over the same period (Dey & Robin, 2011). The recent financial crisis has provided audit clients with the long awaited excuse to switch from the Big 4 to second or third-tier firms (Hogan & Martin, 2009). Throughout the world, members of the Big 4 (formerly the Big 8 and the Big 6) are considered to be top-level audit firms, and their reputation in the Republic of Serbia does not diﬀer in this respect. Unlike the Big 4, which is comprised of large and reputable international firms, second-tier auditors usually have a more national or local reach. For that reason, they can (and should be) be defined independently for each country according to the available evidence on the quality level of their audit services. To the best of our knowledge, a clear definition of second-tier audit firms in Serbia has not yet been delivered. Therefore, the principal goal of this paper - and hence its key potential contribution - is to identify second-tier audit firms in Serbia. We aim to accomplish that by examining the levels of accordance (discordance) in audit opinions issued by diﬀerent groups of audit firms in the sample (by the members of the Big 4 and by the other 38 audit firms that actively operate in Serbia). Additionally, the methodology and results presented in this paper may serve as a reference for future research with the aim to define the second- tier audit firms in other countries. The structure of the paper is as follows. In the next section, we discuss the theoretical background of this study. This is followed by a description of the research method and calculations. Upon presenting the research results, the paper concludes with a short discussion and provides recommendations for future research. THEORETICAL APPROACH There are numerous definitions of audit quality in the literature. For instance, DeAngelo (1981) defines it as "the joint probability that auditors will detect and report a breach in their client’s accounting system". Dopuch and Simunic (1982) define it as a possibility that audited financial statements are more likely to be free from intentional misrepresentation (DeAngelo, 1981; DeFond, 1992). Based on the definitions provided, a conclusion can be drawn that audit quality is a characteristic that refers to the ability of audit 36

procedures to decrease the agency conflict between the owners of an entity and its management. The aforementioned authors argue that auditors typically specialize in providing auditing services of a certain level of quality to retain their clients and establish distinguishing reputation. Therefore, they will most likely refuse an engagement that demands different levels of audit quality, since such arrangements would be cost-ineffective. However, it should not be disregarded that there are different performance dimensions of audit firms (Causse & Vu, 2012). Still, based on the previous statements, it can be supposed that an auditor’s size can serve as an appropriate proxy for the level of quality of its services. The authors of related research on audit quality (Geiger & Rama, 2006) concluded that both Type I and Type II error rates are significantly lower for the Big 4 firms compared to the non-Big 4 firms. They did not find significant differences between second and third-tier audit firms with respect to either type of reporting error (Geiger & Rama, 2006). Therefore, a higher quality of services provided by first-tier auditors should justify their premium fees. According to Dey and Robin (2011), the Big 4 provide audit services to more than 80% of SEC-listed firms and capture more than 90% of total audit fees paid by these firms. In addition, the Big 4 members account for upwards of 60% of audit revenues in the Republic of Serbia (Jakšić et al., 2012). Research conducted in the United Kingdom by Campa (2013) supports the claim that audit fee premiums are only charged by the Big 4 firms. However, the same author did not find a significant association between audit quality and type of auditor. Besides its main purpose, external auditing ought to be able to inhibit the earnings management practice. Statistically, the estimated effects of earnings management (i.e. discretionary earnings) are frequently used as a proxy for audit quality, and consequently for defining second-tier firms. However, we have not pursued this approach herein because of the specifics of the local economy. More precisely, even though the authors such as Becker et al. (1998) and Palmrose (1988) have proven to be right in the presumption of an auditor’s ability to prevent earnings management practice in developed countries such as the United States, the same does not hold for developing countries such as Turkey and Greece (Tsipouridou & Spathis, 2012; Yaşar, 2013). Therefore, we believe that in this particular case, it is more appropriate to rely on the aforementioned assumption of the "size/international reputation effect" on audit quality.

EJAE 2015  12 (2)  35-43

In order to accomplish the previously defined research goal, we have collected 11,096 financial statements issued by legal entities headquartered in Serbia, specifically those that have a legal obligation to arrange an external audit of their annual financial statements (i.e. all medium-sized and large entities, in terms of the local rules on size classification, as well as public companies). The financial reports of the sample entities have been collected from the official online data service maintained by the relevant national agency. After that, we have calculated a number of widely used financial indicators, which are presented in Table 1. The auditors are advised to do the same when considering a going concern issue. Financial indicator Total revenues Debt-to-total assets ratio Interest-bearing debt-to-total assets ratio Net margin Operating margin Return on equity Return on assets Current ratio Quick ratio Cash ratio Interest coverage ratio Days sales outstanding Days payable outstanding Inventory conversion period Cash conversion period Asset-turnover ratio Fixed assets-to-total assets ratio Foreign ownership Average monthly salary Z' score for private firms implied default probability Market share Table 1. List of financial indicators used in the research

Most of the listed financial indicators are self-explanatory. Nevertheless, it should be pointed out that Altman’s Z-score is a product of discriminant analysis rather than of logistic regression. Consequently, the probabilities of bankruptcy calculated by the transformation function are actually quasi-probabilities that are not exactly equivalent to those obtained in logistic regression procedures. The values obtained

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

in such way are surely directionally correct, and predominantly magnitudinaly correct. Authors in the field of audit quality research, such as Geiger and Rama (2006), typically collect data on a group of business entities and split them into two groups based on whether or not they have ceased operations during the observation period. Subsequently, they examine if audit opinions from previous reporting periods have stressed going concern issues for those that have ceased operations. However, using the aforementioned methodology in specific context of our research would impose certain drawbacks, since frequency of Type I and Type II errors calculated this way may be somewhat misleading. For instance, the owners of a business entity can voluntarily liquidate its assets and cease its operations. In this case, nonmodified audit opinions from the preceding periods are not actual misjudgments. Also, in the context of the Serbian economy and our research sample, it is significantly more likely for medium-sized entities (which would otherwise be classified as small-sized in terms of global standards) to be liquidated than for an S&P 500 listed entity. Moreover, the described approach would considerably reduce the research sample size, since the number of companies that entered into default, and have detailed financial data publicly available, is rather insignificant. Considering previously stated methodological limitations, we believe that this can be avoided in the case of developing countries, by assuming a diﬀerent approach to measuring audit quality. In the following paragraph, we describe a methodological approach that is more pragmatic, as it makes better use of the available financial data. In order to draw relevant conclusions regarding the research subject, we have defined, identified and removed the outliers. Financial statements from the initial sample are left out if they do not meet the following requirements: calculated values of average monthly salary can take values in the range from €30 to €3,000; moreover, net margin is expected to be between -300% and 300%. Upon outlier treatment, the initial sample was reduced to 10,555 financial statements for the reporting period 2010–2012. Subsequently, we gathered audit reports of entities included in the sample, and noted whether they were issued by the Big 4 or not. Then, based on the previous criterion, we split the sample in two parts: the first, which is called the "training sample", included opinions issued by the Big 4; the second, called the "test sample", included the opinions of the rest of the audit firms. It turns out that the members of the Big 4 audited 1,101 financial statements, whereas 9,454 37

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

were audited by the remaining auditors. Finally, a dummy coded dependent variable is defined based on the actual audit opinions stated in the auditor reports (1 = modified opinion; 0 = unmodified opinion). To sum up, calculated financial indicators (see Table 1) are treated as explanatory variables in the process of classification, and audit opinions constitute the binary response variable. Classification process can be performed by means of various statistical (or machine learning) methods. In this research, we have used family of classifiers named "Decision trees". Specifically, we used the IBM SPSS implementation of Exhaustive CHAID algorithm. This is a superior algorithm compared to regular CHAID because it has improved its ability to find an optimal method of grouping the categories for each predictor. CHAID uses a chi-squared statistics (hence the C and H in its name) for node splitting. We have used its likelihood ratio form of estimation, which is more robust than Pearson’s, but also requires more time for calculation. Moreover, we did so in order to take advantage of modern computational power. Nodes in the decision trees are presented in a top-down manner, each with its own corresponding key statistics. In the results section, we also present the classification risk and the gain statistics for each node. The stopping criterion is set to activate if the number of iterations exceeds 100 or the change in expected cell frequencies falls below 0.001. The minimum number of cases per node has been set to 90 for parent nodes and 30 for child nodes. We have used a split-sample valida-

tion sub command for the Big 4 in order to evaluate how well the tree structure will generalize to a larger population (in this case, to the test sample). In order to make the significance level adequately conservative, the Bonferroni adjustment is applied. Above all other reasons, the decision tree family of classifiers was selected for its transparency, as well as for the interpretability of its output (which is in the form of classification rules). This is particularly true when compared to some other advanced machine learning algorithms (e.g. Neural Networks). Furthermore, they have the capacity to detect and account for non-linear eﬀects on the response variable or interactions between predictors (Ritschard, 2010), which is, in view of the problem of interest, another advantageous feature. Linear models are transparent too, but they lack the aforementioned capability of accounting for non-linear eﬀects and interactions between predictors. RESULTS In the initial phase, we have trained the classification algorithm based on the values of financial indicators and corresponding opinions issued by the Big 4 audit firms (the test sample). The tree is grown starting at the root node by repeatedly performing three steps at each node: merging, splitting and stopping. This stage provided key insights into the financial indicators that are considered to have the best discriminatory power (i.e. the strength of association be-

Figure 1. Classiﬁcation rules inferred from the training sample (Big 4 clients) 38

EJAE 2015  12 (2)  35-43

tween independent variable (predictor) and dependent variable) between entities that received an unmodified opinion and those that received a modified one. The first diagram illustrates the fact that the training sample of 1,101 audit opinions (868 unmodified and 233 modified) has been divided at node 0, based on the values of average monthly salary to nodes 1 (less than $300,211), 2 (between $300,211 and $730,679), 3 (between $730,679 and $1,537,860) and 4 (more than $1,537,860). At the second level, cash ratio, Z’- score implied probability of default and return on assets have been used for further classification. By analyzing the node that is first from the left on the second level of the decision tree, we notice that the entities have been further classified based on their cash-ratio values. This leads us to the conclusion that these are the entities with acute financial problems, but that are expected to deal with short-term liquidity issues. The next node of the second level of classification is likely to comprise entities that have significant mid-to-long term solvency issues, since their Z’-score values turn out to be an important discriminator. Again, the next node to the right, which contains 40% of entities, is further classified based on their return on assets and interest coverage ratio values. A possible explanation for this is that companies that are not faced with liquidity or solvency issues are generally more likely to engage in earning(s) management activities if their result (ROA) is closer to the borderline, or if they bear the burden of repaying relatively high loan amounts.

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

In order to identify members of the second tier, we have used the very same rules of classification on the test sample, which consisted of audit opinions issued by other audit firms (other than members of the Big 4). The underlying assumption is that the audit methodology used by the Big 4 is probably superior to that of other auditors. Hence, our goal is to determine the firms which use an audit methodology that seems to be most similar to that of the Big 4. The classification tree is illustrated in the following diagram. Table 2 shows the observed classification consistency. The risk estimate is the proportion of properly classified cases. Sample

Estimate

Standard Error

Training

.202

.012

Test .337 .005 Table 2. The risk of improper classiﬁcation of training and test sample

The risk of wrong classification is higher for the test sample. There are three possible theoretical explanations for this. The first one is model over fitting. This is an error which occurs when a model so closely fits the training sample that, in addition to those characteristics of key interest, it ends up including those that are of a random (noise) nature. As a result, the model is expected to have a significant drop in classification performance when applied to the test sample.

Figure 2. Decision tree and classiﬁcation rules for training sample (remaining audit ﬁrms – non-Big 4) 39

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

The second reason is related to diﬀerent characteristics of clients between the two samples. To be precise, it can be argued that entities that have poor financial performance face somewhat diﬀerent challenges compared to their healthy counterparts. For instance, smaller businesses may more frequently receive a going-concern notice, since they have limited access to additional capital funding (lack of financial flexibility), inaccurate financial reporting (inadequately trained accountants) and lower overall liquidity levels. The same holds for state-owned entities. Research conducted by Radak and Bukvić (2010) shows that only a tiny fraction of state-owned entities are clients of the Big 4 firms. The authors claim that those entities typically have low profitability, messy financial reporting practices and are highly motivated to apply undesirable creative accounting techniques. Consequently, they usually try to evade the "rigorous" controls employed by top auditors, and thus tend to resort to other smaller auditors. Such clients may introduce higher audit risk (predominately more Type II errors) in second and third-tier auditors’ portfolios, which adversely impact their profitability. Therefore, the discriminative power of both cash ratio and Z’-score is greater for the non-Big 4 clients. Regarding the risks that non-Big 4 audit firms face, it is important to notice that the state-owned entities in the Republic of Serbia are found to be particularly risky clients. The third reason, and probably the most significant for this research, is the level of quality of audit service. In other words, the non-Big 4 audit firms may have low classification accordance with the Big 4 firms simply because they provide lower quality audit services (lack of appropriate training, knowledge, resources, etc.). We have compared the actual type of audit opinions assigned to certain entities with predicted values generated by classification rules. The results are summarized in Table 3. Sample

Training

Test

Observed

The classification errors are drastically higher for the test sample (79.8% compared to 66.3% for the training sample). The firms in the test sample (nonBig 4 audit firms) are precise in relation to the issuance of modified opinions (i.e. have less Type I errors), but significantly imprecise in issuing unmodified opinion (i.e. have more Type II errors). We also notice significant diﬀerence in the incidence of classification errors among the non-Big 4 firms. In the following table, we have presented the list of audit firms, listed based on the descending value of accordance with Big 4 classification rules. Specifically, in the context of this research, higher values of accordance metrics are assumed to be more favorable. Table 4 displays the second-tier auditing firms in Serbia (the Small 4): Privredni Savetnik, Stanisic Audit, Vincic and Savica. These are the firms with lowest observed ratio of misclassification (in the context of the research) to total opinions, which means that their audit methodology is relatively similar to that of the Big 4 (at least relatively similar compared to the rest of the firms). It is interesting to notice that the members of the Small 4 are not the most profitable audit firms in Serbia (Jakšić et al., 2012). We should also point out that Privredni Savetnik and Stanisic Audit appear to have greater market share than the other two members of the Small 4. Therefore, since Vincic and Savica have a relatively low number of audits performed, it is plausible that they will lose their status in the foreseeable future, while Revizija and SBV Jankovic have a good opportunity to take over their positions. Also, it is worth noting that Stanisic Audit and Savica have more Type I errors, whereas Privredni Savetnik and especially Vincic have more Type II errors. At this point, it is important to bear in mind that the predicted values are neither based on the percentage of entities that went bankrupt during the observed Predicted

Unmodiﬁed

Modiﬁed

Percent correct

Unmodiﬁed

850

97.9%

Modiﬁed

204

12.4%

Overall percentage

95.7%

4.3%

79.8%

Unmodiﬁed

5053

1,323

79.3%

Modiﬁed

1,862

1,216

39.5%

Overall percentage

73.1%

26.9%

66.3%

Table 3. The impreciseness of classiﬁcation of the training and test samples 40

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

Error

Audit ﬁrm

No error

Type I error

Count

Row N %

Count

Privredni Savetnik

179

86.1%

Stanisic Audit

178

76.1%

Vincic

75.5%

Type II error

Total

Row N %

Count

Row N %

Count

Row N %

5.3%

8.7%

208

100%

14.1%

9.8%

234

100%

3.8%

20.8%

100%

Savica

75.4%

13.1%

11.5%

100%

Revizija

291

74.8%

22.6%

2.6%

389

100%

SBV Jankovic

115

71.9%

25%

3.1%

160

100%

IEF

263

71.9%

14.2%

13.9%

366

100%

Euro Audit

622

71.7%

136

15.7%

109

12.6%

867

100%

Dij Audit

256

70.5%

12.7%

16.8%

363

100%

Alfa Revizija

70%

25%

120

100%

PKF

314

69.8%

12.7%

17.6%

450

100%

Srbo Audit

178

69.3%

22.6%

8.2%

257

100%

Fineks SJ

69.2%

10.8%

20%

100%

Grant Thornton

130

69.1%

10.6%

20.2%

188

100%

Libra audit

69.1%

11.7%

19.1%

100%

Milinkovic audit

105

68.6%

10.5%

20.9%

153

100%

EKI revizija

109

68.6%

1.3%

30.2%

159

100%

Absolute Audit

68.2%

23.3%

8.5%

129

100%

Pan Revizija

163

67.4%

16.9%

15.7%

242

100%

Mdm revizija

169

67.1%

18.3%

14.7%

252

100%

Moore Stephens

316

66.9%

11%

104

22%

472

100%

Konsultant revizija

170

65.9%

7.8%

26.4%

258

100%

MC Global Audit

65.9%

14.6%

19.5%

100%

Revizija plus-pro

177

65.6%

10.7%

23.7%

270

100%

Kapital revizija

65.5%

19%

15.5%

100%

BDO

119

64.7%

1.1%

34.2%

184

100%

Prva revizija

64.5%

6.6%

28.9%

100%

ACA Professional Audit

63.9%

11.1%

25%

100%

Auditor

174

63.5%

9.1%

27.4%

274

100%

NDP

63.2%

29.8%

100%

Bojic revizija

62.8%

20.5%

16.7%

100%

Fin Revizija

120

61.9%

24.2%

13.9%

194

100%

Dst revizija

366

61.2%

149

24.9%

13.9%

598

100%

Rang

60%

23.3%

16.7%

100%

Baker Tilly WB Revizija

113

58.5%

9.3%

32.1%

193

100%

Conﬁda Finodit

602

58.3%

111

10.7%

320

31%

1033

100%

Pan audit

58.2%

3.6%

38.2%

100%

Europoint

56.1%

36.8%

100%

Vizura Invent

54.4%

13.1%

32.5%

160

100%

LB Rev

100

54.3%

7.6%

38%

184

100%

Nova revizija

52.7%

38.2%

9.1%

100%

Auditing

45.5%

54.5%

123

100%

Table 4. The impreciseness of classiﬁcation of the training and test samples 41

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

period, nor on the statistically estimated eﬀects of earnings management practice. For that reason, any potential conclusions regarding audit service quality that might be drawn from the presented results are heavily dependent on the assumptions clearly stated in the previous sections of this paper. We suggest that the results should be viewed as comparative rather than absolute, and consequently interpreted in terms of accordance/discordance, rather than in terms of superiority/inferiority of services quality. SUMMARY In this paper, considerable eﬀort has been devoted to defining second-tier audit firms in the Republic of Serbia. To that end, a large amount of data comprising 10,555 annual financial reports and corresponding audit opinions has been collected and analyzed. Relevant financial indicators have been calculated, and the initial sample has been divided between the training sample, comprising the statements audited by the Big 4, and the test sample, which included financial statements audited by other audit firms. Firstly, an Exhaustive CHAID classification algorithm has been trained on the test sample. The resulting classification rules have then been applied to the test sample, resulting in the predicted audit opinions. Finally, the differences (errors) between the predicted and observed opinions have been analyzed. As expected, second and third-tier auditors had relatively more Type II errors. As a final step, the predicted outcomes have been compared to the actual ones, and the discordance statistics have been calculated and summarized in a form of a table. Based on the results obtained, we have determined that Privredni Savetnik, Stanisic Audit, Vincic and Savica are the Small 4 audit firms operating on the territory of the Republic of Serbia. However, a follow-up analysis reveals that they are not the most profitable ones. Similarly, it seems that there is no clear association between the prevalence of errors and corresponding market share within the second and third-tier auditors, but we recommend that further research should be undertaken concerning this topic. The authors have also pointed out that there are significant methodological limitations to this research (see results section) and that its results should be interpreted with caution.

REFERENCES Becker, C., Defond, M., Jiambalvo, J., & Subramanyam, K. (1998). The eﬀect of audit quality on earnings management. Contemporary Accounting Research, 15(1), 1-24. doi:10.1111/j.1911-3846.1998.tb00547.x Campa, D. (2013). Big 4 fee premium and audit quality: Latest evidence from UK listed companies. Managerial Auditing Journal, 28(8), 680-707. doi:10.1108/ MAJ-11-2012-0784 Causse, D., & Vu, V.H.T. (2012). Les dimensions de la performance des cabinets d’audit légal - Le point de vue des auditeurs. Comptabilité - Contrôle - Audit, 28(18), 97-143. doi:10.3917/cca.183.0097 Deangelo, L. (1981). Auditor size and audit quality. Journal of Accounting and Economics, 3(1), 183-199. Defond, M. (1992). The association between changes in client firm agency costs and auditor switching. Auditing: A Journal of Practice & Theory, 11(1), 16-31. Dey, M., & Robin, A. (2011). Second-tier auditing firms: Developments and prospects. The CPA Journal, 81(6), 32-39. Geiger, M., & Rama, D. (2006). Audit firm size and goingconcern reporting accuracy. Accounting Horizons, 20(1), 1-17. doi: 10.2308/acch.2006.20.1.1 Geiger, M., Raghunandan, K., & Riccardi, W. (2014). The global financial crisis: U. S. bankruptcies and goingconcern audit opinions. Accounting Horizons, 28(1), 59-75. doi:10.2308/acch-50659 Hogan, C., & Martin, R. (2009). Risk shifts in the market for audits: An examination of changes in risk for "second tier’’ audit firms. Auditing: A Journal of Practice & Theory, 28(2), 93-118. Jakšić, D., Mijić, K., & Andrić, M. (2012). Analysis of variations in the performance of audit firms in the Republic of Serbia. Economic Annals, 57(193), 71-91. doi:10.2298/EKA1293071J Palmrose, V. (1988). An analysis of auditor litigation and audit service quality. The Accounting Review, 63(1), 55-73. Radak, I., & Bukvić, L. (2010). Javna preduzeća beže od stroge revizije. Retrieved September 14, 2014, from http:// www.danas.rs/danasrs/ekonomija/javna_preduzeca_ beze_od_stroge_revizije.4.html?news_id=188710 Ritschard, G. (2010). CHAID and Earlier Supervised Tree Methods. Retreived September 14, 2014, from http:// ideas.repec.org/p/gen/geneem/2010.02.html Tsipouridou, M., & Spathis, C. (2012). Earnings management and the role of auditors in an unusual IFRS context: The case of Greece. Journal of International Accounting, Auditing and Taxation, 21(1), 62-78. doi:10.1016/j.intaccaudtax.2012.01.005 Yaşar, A. (2013). Big Four auditors’ audit quality and earnings management: Evidence from Turkish Stock Market. International Journal of Business and Social Science, 4(17), 153-163.

EJAE 2015  12 (2)  35-43

Stanišić, N. et al.  Defining second-tier audit firms in Serbia

ODREĐIVANJE DRUGOG NIVOA REVIZORSKIH FIRMI U REPUBLICI SRBIJI: U POTRAZI ZA “MALOM ČETVORKOM” Rezime: U ovom radu pokušaćemo da definišemo drugi nivo revizorskih firmi u Republici Srbiji. Kako bismo ostvarili cilj rada, analizirali smo 10.555 finansijskih izveštaja pravnih lica koja posluju na teritoriji Republike Srbije, a koji su pripremljeni za period od 2010. do 2012. godine. Od ukupnog pomenutog broja, eksternu reviziju 1.101 izveštaja uradila je "Velika četvorka", dok je reviziju druge grupe izveštaja uradilo preostalih 38 revizorskih firmi. U cilju definisanja algoritma klasifikacije, kao nezavisnu varijablu upotrebili smo izračunate finansijske pokazatelje, dok smo za zavisnu varijablu koristili tip revizorskog mišljenja koji je izdat od strane "Velike četvorke". Primenom metode stabla odlučivanja dobili smo pravila razvrstavanja, koja smo kasnije primenili na uzorku privrednih društava za čiju reviziju su bile zadužene preostale revizorske firme. Na osnovu poklapanja sa kriterijumima "Velike četvorke" (za koju važi da poseduje visok nivo kvaliteta izvršene revizije), od uzorkovanih revizorskih firmi predložili smo one koje bi činile "Malu četvorku" u Republici Srbiji. Pored pomenutog, rezultati istraživanja ukazuju na to da "Velika četvorka" i preostale revizorske firme ne posvećuju podjednaku pažnju pojedinim finansijskim pokazateljima. Mogući razlozi za to su prethodno razmotreni u radu. Na kraju, autori će pokušati da predlože u kom pravcu bi mogla da se kreću buduća istraživanja na ovu temu.

Ključne reči: stablo odlučivanja, finansijski izveštaji, revizorsko mišljenje. Received: April 5, 2015. Correction: April 10, 2015. Accepted: May 2, 2015.

EJAE 2015, 12(2): 44-52 ISSN 2406-2588 UDK: 005.311.6 005.941 DOI: 10.5937/ejae12-8230 Original paper/Originalni naučni rad

THE ROLE OF BUSINESS INTELLIGENCE IN DECISION PROCESS MODELING Višnja Istrat1*, Sanja Stanisavljev2, Branko Markoski2 1 2

Faculty of Organizational Sciences, University of Belgrade, Serbia Faculty of Technical Sciences Mihajlo Pupin, University of Novi Sad, Zrenjanin, Serbia

Abstract: Decision making is a very significant and complex function of management that requires methods and techniques that simplify the process of choosing the best alternative. In modern business, the challenge for managers is to find the alternatives for improving the decision-making process. Decisions directly affect profit generation and positioning of the company in the market. It is well-known that people dealt with the phenomenon of decision making in each phase of the development of society, which has triggered the need to learn more about this process. The main contribution of this paper is to show the significance of business intelligence tools and techniques as support to the decision making process of managers. Research results have shown that business intelligence plays an enormous role in modern decision process modeling.

INTRODUCTION Decision making is the selection of one of the offered alternatives. It is an extremely complex process that should result in making the right management decisions and thus, it is essential to continuously explore and improve modern decision making methods and techniques. There is no ideal solution for the specified problem. Therefore, it should be oriented towards finding the most optimal solution. The manager should be conversant with the basic theory of the decision-making process and should possess necessary experience to make business decisions that will result in maximum profit. There are three dimensions that determine full development of this discipline: qualitative, quantitative and information and communication aspects. These three aspects of the decision-making process fully satisfy all concepts of development of modern decision making on both theoretical and practical level. Quantitative approach to modern decision making is defined as the basic formalism of the decision making problem (Suknović & Delibašić, 2010). According 44

* E-mail: visnja.istrat@gmail.com

Key words: business intelligence, decision-making, decision.

to (Suknović, 2001), the decision making problem is centered around the five items (A, X, F, Θ,≻) where: A: final group of available alternatives (actions) that the session participant ranks with the goal of selecting the most suitable; X: group of possible results that follow as a consequence of the alternative selection; Θ: group of the state of the facts, depends on the unknown state θ∈Θ, because the consequences of the selection of the alternative a∈A can be diﬀerent; F: A x Θ → X determine for each state of the fact v and for each alternative a, resulting in consequence x = F (a,ϖ) ≻: relation of weak influence to X, i.e. binary relation that satisfies the following conditions: (i) x ≻ y or y ≻ x, ∀ x, y∈X (ii) ≻ Is transitive, x ≻ y, y ≻ z ⇒ x ≻ z. Relation ≻ determines the decision maker and is called the relation of the preference. Strong preference x ≻ y, means that x ≻ y, but not y ≻ z. Relation of in-

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

diﬀerence x ~ y means that it is valid x ≻ y and y ≻ z. The most frequent solution to the decision making problems is in transformation of weak item ≻ to X in usual item ≥ over real numbers with the function of utility (Centre for business decision making, 2015). In the conditions of general problem of decision making, it is assumed that the state of the facts v is known, X is multidimensional and known for each alternative as a group of relevant values of the attribute. Based on the level and degree of complexity of the problem, and the number of participants (Suknović & Delibašić, 2010), modern decision making as a human phenomenon is split into the following categories: ◆ Individual decision making. This decision making form is the most simplified and most often explained and used, which is confirmed by numerous references. Activities of the decision making process are dedicated to one decision maker. ◆ Group decision making – team work. It is described by the higher level of the phases and activities of the decision making process. Greater number of participants (of the session) is involved in the selection of the most optimum alternative. It contains significant methods, models and up-to-date tools to support group decision making. ◆ Organisational decision making. This form of decision making is described by the high level of unstructured problems the organizations face. This type of decision making is followed by numerous research experiments, but it is not yet come to systematical knowledge that would significantly improve it. ◆ Metaorganizational decision making. It is the highest level of the application of systematical knowledge in the field of decision making. More information on knowledge discovery can be seen in (Guruler et al., 2010). It is practically noted at the level of one country (state), own national interest through social well-being, culture, income, etc. (Suknović & Delibašić, 2010). The result of successful decision making process is getting the most optimal decision for the defined topic, because there is no perfect solution. Decision is the final outcome of the decision making process, most often choosing the most acceptable alternative from the group. The process should be followed by an

intuition, strong analytical tools and logical thinking. The selection of the alternative is possible by decision making techniques, decision making rules or skills. According to (Suknović & Delibašić, 2010), the decision making moment is the most creative and most critical moment in the entire decision making process. DEVELOPMENT OF THE BUSINESS DECISION MAKING MODEL Data mining is the research and analysis of large amounts of data with the aim to reveal significant patterns and rules. In order to increase productivity of modern companies, the goal is to improve the organization's functions through better understanding of customers. More information on customer value management can be seen in (Verhoef et al., 2007). Data mining techniques and tools have various fields of application – law, astronomy, medicine, industry, etc. Not even one data mining algorithm has been firstly introduced for commercial purposes. More information on data mining strategy can be found in (Hornick et al., 2007). The selection of certain combination of techniques that will be used in certain situation depends on the nature of research, available data and skills and customer preferences. More information on data can be seen in (Verhoef et al., 2010). Data mining can be direct and indirect. Direct data mining explains or categorizes certain fields, like financial income, whereas indirect data mining tries to find the patterns or similarities between target groups of data with no use of certain field or collection or predefined classes. Data mining is mostly oriented towards defining the model. The model is simply an algorithm or a set of rules that connect the collections of input (input elements) with certain goal or output elements. Regression, neural networks, decision making trees and most other techniques are oriented towards model creation. Under the right circumstances, model can result in the explanation on how output elements of certain interest, like the precise order or unsuccessful bill paying, are connected and can be predicted using the available facts. Models are also used to get the results. The result is the way of expressing the findings of the model by numbers. Results can be used to sort the list of customers from the most to least loyal, willing or not to cooperate or to pay back the loan. Data mining process is also called knowledge discovery or knowledge discovery in Databases. One of the synonyms is also knowledge creation. 45

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

Many assignments related to intellectual, economic and business fields that will be solved by data mining can be divided into six categories: ◆ Classification ◆ Estimation ◆ Prediction ◆ Association rules ◆ Clustering ◆ Description and profiling The first three types are the description of direct data mining, where the goal is to find certain target variables. Association rules and clustering are indirect assignments where the goal is to find the data structure, regardless of the variables value. Profiling is the descriptive assignment that can be direct or indirect. The assignment of the association rules is to determine which items go together. A typical example is grouping the items that can be bought together in shopping at a supermarket – market basket analysis. Supermarket chains use association rules to plan the timetable of items on the shelves or in the catalogue so that the items that can be bought together are most often together on the shelves. They are used to identify chances for cross-selling and designing of attractive packaging or grouping of products or services. It is a simple approach on how to create the rules from large databases. If two items, i.e. computer and web camera are often bought together, we can produce two association rules: ◆ Buyers that buy computer can also buy web camera with probability of P1 ◆ Buyers that buy web camera can also buy computer with probability of P2 The assignments and the process of data mining can be diﬀerent. Depending on data mining results, the assignments can be: ◆ Exploratory data analysis: in databases large amounts of data are at disposal. This data mining assignment has two purposes: search of knowledge that customer demands and data analysis. These techniques are interactive and visual for the customers. ◆ Descriptive modeling: describes all data, including models for the whole probability of data distribution, split of multidimensional space into groups and models that describe the relations between variables. ◆ Predictive modeling: this model allows prediction of values of some variables more than known values of other variables. 46

◆ Data mining: this assignment has been used to find the hidden patterns and the rules in cluster. There are diﬀerent sizes at disposal in clusters. The goal of this is to reveal "how it is best to find the rules". This can be achieved by using the rules for induction and many other techniques in data mining algorithms. This is called clustering of algorithms. ◆ Research of the content: the primary goal of this assignment is finding the data set that is often used for audio/video tasks, as well as photos. Such data mining is similar to those that are of interest in dataset. The most frequently used data mining techniques are: ◆ Decision making trees and their rules, ◆ Non-linear regressions and classification methods, ◆ Methods based on examples, ◆ Models of graphical dependence of the probability, ◆ Relational models of learning, etc. In general, we classify these data mining methods as: Online analytical processing (OLAP), classification, clustering, association rules, temporary data mining, time-sheet analysis, space research, web research, etc. More information on classification can be found in (Grabmeier & Lambe, 2007). These methods use diﬀerent types of algorithms and data. The source of data can be data warehouse, database, regular or textual file. Algorithms can be statistical, based on the decision making tree, neural networks, generic algorithms, etc. In general, data mining algorithms are completely dependent on two factors: ◆ type of datasets used ◆ demands of customers Knowledge Discovery Process includes raw data of choosing data mining algorithm, as well as processing of data results. More information on data mining models can be seen in (Johansson et al., 2004). There are programs that are called assistants for knowledge discovery (Intelligent Discovery Assistants - IDA), that help in the application of valid knowledge in the research process. These assistants provide three benefits to the users: ◆ Systematic validation of process for knowledge discovery,

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

GROUPS OF DATA THAT WILL BE TESTED

Outcome

Attributes

Class

Goal

There is the table with three types of data from diﬀerent sources. Those data are then processed and real and important information for the defined problem is obtained. By application of association rules, knowledge is the final outcome of the decision making process. The structure and size of research data, as well as the proper use of business intelligence tools influence the creation of the improved model of business decision making. More information on business intelligence can be found in (Guster & Brown, 2012).

Sub-attributes

Use Data 1

Precision Accuracy Credibility Importance Certainty of source

Data 2

Decision 2

Size of sample Possibility of processing Appropriateness Conﬁdence Up-to-date Availability Data 3

Decision 3

Compact

Improved model of business decision making

Decision 1

Complexity

Creation of model by system of association rules

◆ Effective ranking of valid processes on different criteria, that help in choosing the suitable option and ◆ Infrastructure for knowledge sharing, that leads to externalization of networks. There have been attempts to make this process automatic and design the general tool for data mining that uses business intelligence while selecting data and data mining algorithms, and to a certain extent, knowledge discovery. Knowledge management and data mining are still in the development phase and they represent interesting areas for researchers. Although there is an integrative framework for knowledge management in the context of marketing, there are critical research challenges that should be devoted considerable attention. More information about data mining for marketing can be seen in (Berry & Linoff, 2004). Some of them are connected to data mining techniques and knowledge discovery process, while others are related to knowledge management. Data research through data mining techniques is an interactive process of learning similar to other processes of acquiring knowledge, like scientific research. Selection of data mining algorithms, hypothesis forming, model evaluation and remodeling are the key components of the research process. Since the cycle of attempts and failures for progressive adopting are made of the most valuable knowledge through data mining, the aspect of learning through experiments can be suitable for that. One of the research challenges is to make sure that this process is multi-structured, and therefore to increase the productivity of data mining trials. Furthermore, it is needed to manage the knowledge in the sense that it outlines organizational borders and further distributes towards the other partners. Another challenge is multiple classifications when the customer belongs to more than one category. There is the case of web mining.The Internet is getting the primate as a new channel for the goods distribution, product promotion, transaction management and coordination of business processes and it becomes a valuable and suitable source of data about customers. More information on management can be seen in (Draker, 2003). However, multiple formats of data and distributive nature of knowledge on the Web are the challenge for collecting, revealing, organizing and knowledge management in the way that is suitable for providing support to business decision making. More information on decision support system can be seen in (Kotsiantis, 2011).

Correctness Completeness Actuality Visibility

Table 1. Suggestion of groups of data for testing

According to (Suknović, 2001), there are three dimensions that determine the complete development of modern decision making. These are: qualitative, quantitative aspect and information-communication aspect. Research of all three aspects is the most optimal. Data will be shown in quantitative aspect, as given in the table, and will be processed by numerical 47

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

values. Business intelligence methods and techniques will be applied. Attributes and sub-attributes will be described, as shown in the table, in order to ensure the qualitative aspect of modern decision making. Description of the results will ensure the qualitative aspect. The application of software and visual overview of data will show the information-communication aspect of modern decision making. Table 1 shows the groups of test data with three classes and attributes, each of which contains six sub-attributes. During the research, it is possible to modify the number of classes, attributes and sub-attributes (Mihailović, 2004). Depending on the number of data and relevant criteria, the final data structure of research will be defined. Sub-attributes are values that relate to attributes and are defined by managers based on the significance of certain segments needed for conducting research. Sub-attributes in Table 1 are not of the same importance, and the way of structuring of multiple problem active manager participation in the process of its solving, are in the model of defining the weight of users’ criteria. More information can be seen in (Čupić & Suknović, 2008). Aggregation of criteria (sub-attributes) has been conducted based on the analysis of various current research, opinions from data mining experts and numerous discussions in order to generate the most relevant criteria for data values whose influences cannot be neglected: complexity, accuracy, size of the sample, etc. (table 1).

Table 2. Matrix of evaluation in pairs of criteria 48

For determination of values of criteria in multicriteria decision making, defining of vector coeﬃcients has been used. Criteria defined in the form of vector of coeﬃcients represent the exchange or mutual relation between the criteria. Certain hierarchical structure of problem of comparison of criteria between each other is the matrix of evaluation. For the purpose of testing the group, the software made at the Faculty of Organizational Sciences in Belgrade has been used. More information on software can be found on the web page of the Centre for business decision making. Table 2 shows the comparison of the criteria subattributes from the example of future research. Matrix of evaluation is square dimensions, equal to number of criteria in model. There is no evaluation of criteria between each other and items are visible at the main diagonal. The other values are imported as a result of comparison of each criterion separately. Only direct values are imported, while inverted values are defined on their own, in this example through software. Solution to the problem has the following steps: Matrix of comparison should be adjusted in pairs, so as to put the unit in the field where each criteria is compared to each other. In order to count the sum of elements in each column of adjusted matrix, with following equation: n

l j = å i =1 Pij , j=1, n

(1)

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

Lis the sum of elements of adjusted column, P is the value of the criteria. To divide each element of the column with the sum of values of that column, with the following equation:

hij = Pij ¸ l j , i=1, n, j=1, n

(2)

Counting the sum of elements and then determine the arithmetic mean of each row. Column that has the middle values is actually the normalized vector. n

Wi = å j =1 hij , i=1, n

(3)

search (Promethee, Electre or AHP). For realization of this example, the method of analytical-hierarchical process (AHP) has been used. The method of the analytical-hierarchical process (AHP) was defined by Thomas Saaty in the 1970s. AHP represents the tool and the analysis of decision making, created with the help of managers in solving the complex problems with multiple criteria and greater number of managers. AHP is based on the concept of balance used for determination of the overall relative significance of the group of attributes, activities or criteria, and it relates to the analyzed problem of decision making.

W is the column of values of criteria elements in row, H is the value of criteria. Middle value is determined by data in column W that splits with number of criteria and is shown in row t, because it is a square matrix.

t j = w j ¸ n, j=1, n

(4)

In this manner, the value of each criterion in the model is calculated. Vector of coeﬃcients is normalized and the sum of elements is equal to one. More information about the process of defining the vector of coeﬃcients can be found in (Čupić & Suknović, 2008).

Figure 1. Dialog box for deﬁning the level and attributes of model

Figure 1 shows the initial step of defining the options for creating the level and attributes inside each level of the defined model. It is defined which method of multicriteria decision making will be used in re-

Figure 2. Dialog box for deﬁning the vector of coeﬃcients

Figure 2 shows the dialog box for assigning the significance of each criterion related to others in the predefined table. The predefined table with nine options is Saaty’s most famous scale for assigning the weights that is confident in solving real problems of business systems. Saaty’s scale for conversion of linguistic statements while comparing the significance of pairs of criteria is put as a standard due to its simplicity. The manager most often defines the relations between the criteria, and therefore the subjective aspect of the decision making is apparent. Figure 3 provides an overview for the analysis of hierarchy between the defined levels of the model. There is the influence and interdependence of the defined criteria of business decision making. Visual overview enables data mining to the user and adaptation of the existing elements of the model. More information on data mining trends can be seen in (Kriegel et al., 2007; Krüger et al., 2010; Bramer, 2007). There is the influence of all sub-attributes on all data, the influence of all three types of data on three models of 49

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

Figure 3. Overview of segments of the deﬁned model hierarchy

decisions, and their overall influence on the goal of the model – creating the improved model of business decision making. After data processing by AHP method, there is the graphic with defined vectors of weight of criteria coefficient. Based on the graphical overview of the model, it could be easy to see the criteria that has high vector of coeﬃcients and, therefore, bigger value. Based on the given example, it is easy to see the significance of data (with defined vector 0.14), accuracy (0.122) and possibility of processing (0.121). The plan is to include the axiom of transitivity in future research. According to (Čupić & Suknović, 2008), the axiom of transitivity states: if A is bigger than B, and B is bigger than C, than it follows that A is bigger than C. A > B and B > C => A > C

As could be seen in Table 1, axiom of transitivity could be applied to the choice of class with the most suitable dataset according to the defined criteria by most optimal decision. If the class "Decision 1" is more significant (acceptable, numerically bigger) than class "Decision 2", and class “Decision 2” is more significant that “Decision 3”, it follows that class “Decision 1” is more significant than“Decision 3”. In this manner, the ranking of the final results and classes would be easier, which would lead to the solution to the defined model of business decision making. More information on classes can be found in (Gupta et al., 2010). If the axiom of transitivity is successfully applied to the example of classes and attributes, and if the application is scientifically verified, it follows that it can be successfully used for choosing and comparing the sub-attributes. According to Figure 4, deductive method leads to the following conclusion: if sub-attribute “accuracy” has bigger numerical value

Figure 4. Criteria levels deﬁned by AHP method 50

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

Figure 5. Graphic of criteria level for the most optimal decision of the model

(defined according to the vector of coefficients) than sub-attribute "precision", and "precision" has bigger value than sub-attribute "source confidence" than it follows that sub-attribute "accuracy" has bigger value than sub-attribute "source confidence". More information can be seen in (Istrat, 2014). Figure 5 shows the defined vector of coefficients at the decision level. Decision 2 has the biggest vector (0.717) and the biggest value for the goal of the model. Here, the axiom of transitivity can be applied: if the vector of Decision 1 is bigger than the vector of Decision 2, and the vector of Decision 2 is bigger than vector of Decision 3, then it follows that the vector of the Decision 1 is bigger that the vector of the Decision 3. Decision 1 is of greater importance than the Decision 3. CONCLUSION Based on the current research of domestic and foreign experts in the field of business intelligence, it has been shown that this area has vast potential but is still relatively unrevealed in some segments of application. Therefore, it is the challenge for the researchers and the area where significant scientific and professional benefits can be provided. Innovative research approach to business intelligence is characterized by knowledge and creativity, as well as the use of modern data mining software. Modern scientific methods analyze the results and offer recommendations and guidelines for further research. The primary motivation for development of this model is to show the significance that business intelligence has in the process of decision making for

modern management. The aim is to create the business decision making model that can be applicable to business systems of different aspects and for commercial purposes. The significance of business intelligence for creating the model is shown, which can increase the effectiveness of the decision making process in management. REFERENCES Berry, M.J.A., & Linoff, G. (2004). Data mining techniques: For marketing, sales, and customer relationship management. Indianapolis: Wiley. Bramer, M. (2007). Principles of Data Mining. Springer: London. Centre for business decision making. (2015). Decision-making. Retrieved May 5, 2015, from www.odlucivanje.fon.rs Čupić, M, & Suknović, M. (2008). Odlučivanje. Beograd: FON. In Serbian. Draker, P. (2003). Moj pogled na menadžment : izbor iz dela o menadžmentu Pitera Drakera. Novi Sad: Adizes. In Serbian. Grabmeier, J.L., & Lambe, L.A. (2007). Decision trees for binary classification variables grow equally with the Gini impurity measure and Pearson’s chi-square test. International Journal of Business Intelligence and Data Mining, 2(2), 213-226. doi:10.1504/IJBIDM.2007.013938 Gupta, A., Li, J., Conradi, R., Rønneberg, H., & Landre, E. (2010). Change profiles of a reused class framework vs. two of its applications. Information and Software Technology, 52(1), 110-125. doi:10.1016/j.infsof.2009.08.002. Guruler, H., Istanbullu, A., & Karahasan, M. (2010). A new student performance analyzing system using knowledge discovery in higher educational databases. Computers & Education, 55(1), 247-254. doi:10.1016/j.compedu.2010.01.010. 51

EJAE 2015  12 (2)  44-52

Istrat, V., Stanisavljev, S., Markoski, B.  The role of business intelligence

Guster, D., & Brown, C.G. (2012). The Application of Business Intelligence to Higher Education: Technical and Managerial Perspectives. Journal of Information Technology Management, 23(2), 42-62. Hornick, M.F., Marcadé, E., & Venkayala, S. (2007). Java data mining: Strategy, standard, and practice: a practical guide for architecture, design, and implementation. Amsterdam: Elsevier-Morgan Kaufmann. Istrat, V. (2014). Pristupni rad doktorske disertacije, Unapredjenje modela poslovnog odlučivanja sistemom asocijativnih pravila. Beograd: FON. In Serbian. Johansson, U., Nilsson, L., & Koenig, R. (2004). Accuracy vs. comprehensibility in data mining models. Proceedings of the Seventh International Conference on Information Fusion (pp. 295–300). Retrieved May 5, 2015, from http://www.fusion2004.foi.se/papers/IF04-0295.pdf Kotsiantis, S.B. (2011). Use of machine learning techniques for educational proposes: a decision support system for forecasting students’ grades. Artificial Intelligence Review, 37(4), 331-334. doi:10.1007/s10462-011-9234-x. Kriegel, H., Borgwardt, K.M., Kröger, P., Pryakhin, A., Schubert, M., & Zimek, A. (2007). Future trends in data mining. Data Mining and Knowledge Discovery, 15(1), 87-97. doi:10.1007/s10618-007-0067-9

Krüger, A., Merceron, A., & Wolf, B.A. (2010). Data Model to Ease Analysis and Mining of Educational Data. Retrieved May 5, 2015, from http://educationaldatamining.org/ EDM2010/uploads/proc/edm2010_submission_9.pdf Mihailović, D. (2004). Metodologija naučnih istraživanja. Beograd: FON. In Serbian. Suknović, M. (2001). Razvoj metodologije podrške grupnom odlučivanju. Doktorska disertacija, Fakultet organizacionih nauka, Beograd. In Serbian Suknović, M., & Delibašić, B. (2010). Poslovna inteligencija i sistemi za podršku odlučivanju. Beograd: FON. In Serbian. Verhoef, P.C., & van Doorn, J., & Dorotic, M. (2007). Customer value management: An overview and research agenda. Journal of Research in Management, 2, 51-68. Verhoef, P.C., Venkatesan, R., McAlister, L., Malthouse, E.C., Krafft, M., & Ganesan, S. (2010). CRM in Data-Rich Multichannel Retailing Environments: A Review and Future Research Directions. Journal of Interactive Marketing, 24(2), 121-137. doi:10.1016/j.intmar.2010.02.009.

ULOGA POSLOVNE INTELIGENCIJE U SAVREMENOM ODLUČIVANJU Rezime: Odlučivanje je veoma značajna i kompleksna funkcija menadžmenta koja zahteva metode i tehnike koje olakšavaju proces izbora jedne od više ponuđenih alternativa. U savremenom poslovanju veliki izazov za menadžment jeste pronaći alternativne načine za poboljšanje procesa poslovnog odlučivanja. Odlučivanje utiče direktno na stvaranje profita i pozicioniranje kompanija na tržištu. Činjenica je da su se u svakoj fazi razvoja društva ljudi bavili fenomenom odlučivanja, što uslovljava širenje znanja o ovom procesu. Glavni doprinos ovog rada jeste da ukaže na značaj alata i tehnika poslovne inteligencije kao podrške menadžmentu u procesu odlučivanja. Rezultati ukazuju na ogroman značaj poslovne inteligencije u savremenom odlučivanju.

Ključne reči: poslovna inteligencija, odlučivanje, odluka.

Received: May 6, 2015. Correction: Jun 8, 2015. Accepted: Jun 15, 2015.

INSTRUCTIONS FOR AUTHORS

The European Journal of Applied Economics is a biannual, peer-reviewed journal which publishes significant original scientific research and reviews. The Journal is devoted to the publication of research results in the following areas: economy, management, tourism and hospitality. It is essential that authors write and prepare their manuscripts according to the instructions and specifications listed below. The length and eﬀectiveness of the peer review process will largely depend upon the care used by authors in preparing their manuscripts. Therefore, contributors are strongly encouraged to read these instructions carefully before preparing a manuscript for submission, and to check the manuscript for conformance before submitting it for publication. The manuscripts must be written in English using Microsoft Word (*.doc or *.docx). Make sure that your manuscripts are clearly and grammatically written. Please note that authors who are not native-speakers of English can, to a certain amount, be provided with help in rewriting their contribution in correct English. The Editorial Board expects the editors, reviewers and authors to respect the well-known standard of professional ethics. Types of contributions: ◆ Original papers - (about 10 typewritten pages) report original research which must not have been previously published. ◆ Preliminary communications - (up to 3 pages) report unpublished preliminary results of sufficient importance to demand rapid publication. ◆ Subject reviews - (about 15-20 pages) present an overview of the author’s current research with comparison to data of other scientists working in the field. Manuscripts should be submitted using the Online Submission System. The manuscript must be uploaded as a *.doc or *.docx file (tables and figures should follow the text, each on a separate page). No articles will be published without first undergoing an anonymous refereeing procedure (please read carefully Editorial policy). To facilitate the reviewing process, authors are encouraged to suggest up to three persons competent to review their manuscript. Such suggestions will be taken into consideration but not always accepted. The editor reserves the right to make the final decision with respect to publication. If you have any questions, do not hesitate to contact the Managing Editor at sstanisic@singidunum.ac.rs. 53

EJAE 2015

Instructions for authors

Manuscripts should be typed in English (standard British English) with 1.15 spacing in A4 format leaving 2.5 cm for margins. Manuscript elements: 1. Title page with: 1. title (and short title) 2. name(s) of author(s) 3. name and address of workplace(s) – aﬃliation(s) 4. personal e-mail address(es) 2. Abstract 3. Keywords 4. Introduction 5. Methodology 6. Results and Discussion 7. Reference lists 8. Tables (each on a separate page) 9. Figures (each on a separate page). Each of these elements is detailed below.

1. TITLE PAGE 1.1. Title (and short title) We suggest the title should be relatively short but informative. If a long title is necessary, please prepare an optional short title. 1.2. Name(s) of author(s) A list of all authors of the paper should be prepared. We need full first name and full last name. Initial(s) for middle name(s) is optional. 1.3. Name and address of workplace(s)- affiliation(s) Authors' aﬃliations should be indicated in this section. Either endnote or footnote are not recommended. 1.4. Corresponding author’s e-mail address One e-mail address is needed. It will be used as the corresponding author's email address in all contacts with the authors.

2. ABSTRACT An abstract must accompany every article. It should be a brief summary of the significant items of the main paper. The abstract should give concise information about the content of the core idea of your paper. It should be informative and not only present the general scope of the paper but also indicate the main results and conclusions. The abstract should not normally exceed 200 words. It should not contain literature citations or allusions to the tables or illustrations. All non-standard symbols and abbreviations should be defined. In combination with the title and keywords, the abstract is an indicator of the content of the paper. Authors should remember that online systems rely heavily on the content of titles and abstracts to identify articles in electronic bibliographic databases and search engines. They are therefore requested to take great care in preparing these elements.

3. KEYWORDS List of keywords proposed by the authors, separated by commas. Up to 5 key terms can be selected. We would suggest to avoid repeating the title. 54

EJAE 2015

Instructions for authors

4. TEXT General rules for writing: ◆ use simple and declarative sentences, avoid long sentences, in which the meaning may be lost by complicated construction;

◆ be concise, avoid idle words; ◆ make your argumentation complete; use commonly understood terms; define all nonstandard symbols and abbreviations when you introduce them; ◆ Latin words, as well as the names of species, should be in italic, as for example: i.e. or e.g. ◆ explain all acronyms and abbreviations when they first appear in the text. Generally a standard scientific paper is divided into: ◆ introduction, ◆ main text, ◆ conclusion. Footnotes/Endnotes/Acknowledgements: We encourage authors to restrict the use of footnotes. If necessary, please make endnotes rather than footnotes. Information concerning research grant support or the assistance of colleagues should appear in a separate Acknowledgements section at the end of the paper, not in a footnote.

5. REFERENCE LIST A complete reference should give the reader enough information to find the relevant article. Please pay particular attention to spelling, capitalization and punctuation. The article should contain no fewer than 25 references, preferably published recently. Completeness of references is the responsibility of the authors. Please avoid references to unpublished materials, private communication and web pages. The surname of the author and the year of publication appear in parentheses after the citation, for example (Fisher, 2010). If more than one publication by the same author appear in one year, they must be distinguished by an a, b, etc., for example 2001a, 2001b. In case of quoting the actual words of another author, the page number should be provided, e.g. (Hollard, 2010, p. 23). If the name naturally occurs in the sentence, only the year is given in parentheses, e.g. “Benoliel (1999) thinks…” If there are two authors, the surnames of both should be given (Fisher and Hollard, 2009). If there are more than two authors, the surname of the first author only should be given, followed by et al. (Wilson et al., 2008). However, full names of all authors should be given in the list of references. The original journal title is to be retained in the case of publications published in any language other than English (please denote the language in parenthesis after the reference in the Reference list). 1. Book a. Book (one author) Format: Author. (Year of publication). Book title. Place of publication: Publisher. Example: Baxter, R. (1982). Exactly solvable models in statistical mechanics. New York: Academic Press. b. Book (two or more authors) Format: Author1, Author2, & Author3. (Year of publication). Book title. Place of publication: Publisher. Example: Kleiner, F.S., Mamiya C.J., & Tansey R.G. (2001). Gardner’s art through the ages (11th ed.). Fort Worth, USA: Harcourt College Publishers. 55

EJAE 2015

Instructions for authors

c. Book chapter or article in an edited book Format: Author(s) of chapter. (Year of publication). Chapter title. In Editors of the book (Ed.), Book title (Chapter page range). Place of publication: Publisher. Example: Roll, W.P. (1976). ESP and memory. In J.M.O. Wheatley & H.L. Edge (Ed.), Philosophical dimensions of parapsychology (pp. 154-184). Springfield, IL: American Psychiatric Press. d. Proceedings from a conference Format: Author(s). (Year of publication). Title. In Conference name, Date (Page range). Place of publication: Publisher. Example: Field, G. (2001). Rethinking reference rethought. In Revelling in Reference: Reference and Information Services Section Symposium, 12-14 October 2001 (pp. 59-64). Melbourne, Victoria, Australia: Australian Library and Information Association. e. E-book Format: Author(s). (Year of publication). Title. Publisher. Retrieving date, http address. doi. Example: Johnson, A. (2000). Abstract Computing Machines. Springer Berlin Heidelberg. Retrieved March 30, 2006, from SpringerLink http://springerlink.com/content/w25154. doi: 10.1007/b138965. f. Thesis Format: Author(s). (Year of publication). Title. Information, Place of publication. Example: Begg, M.M. (2001). Dairy farm women in the Waikato 1946-1996: Fifty years of social and structural change. Unpublished doctoral dissertation, University of Waikato, Hamilton, New Zealand. g. Government publication Format: Institution name. (Year of publication). Title. Place of publication: Publisher. Example: Ministerial Council on Drug Strategy. (1997). The national drug strategy: Mapping the future. Canberra: Australian Government Publishing Service. 2. Article a. Journal Article (one author) Format: Author. (Year of publication). Article title. Journal Title. Volume (issue), range of pages. doi. Example: Nikora, V. (2006). Hydrodynamics of aquatic ecosystems: Spatial-averaging perspective. Acta Geophysica, 55(1), 3-10. doi: 10.2478/s11600-006-0043-6. b. Journal Article (two or more authors) Format: Author1, Author2, & Author3. (Year of publication). Article title. Journal Title. Volume (issue), range of pages. doi. Example: Cudak, M., & Karcz, J. (2006). Momentum transfer in an agitated vessel with oďŹ&#x20AC;-centred impellers. Chem. Pap. 60(5), 375-380. doi: 10.2478/s11696-006-0068-y. 56

EJAE 2015

Instructions for authors

c. Journal article from an online database Format: Author(s). (Year of publication). Article title [Electronic version]. Journal Title. Volume (issue), range of pages. Retrieved date of access, from name of database. doi. Example: Czajgucki Z., Zimecki, M., & Andruszkiewicz, R. (2006, December). The immunoregulatory effects of edeine analogues in mice [Abstract]. Cell. Mol. Biol. Lett. 12(3), 149-161. Retrieved December 6, 2006, from PubMed database on the World Wide Web: http://www.pubmed.gov. doi: 10.2478/ s11658-006-0061-z. d. Newspaper article (no author) Format: Article title. (Publication date). Journal Title. page. Example: Amazing Amazon region. (1989, January 12). New York Times, p. D11. 3. Other formats a. Web page Format: Author/Sponsor. (last update or copyright date). Title. Retrieved date of access, from URL. Example: Walker, J. (1996, August). APA-style citations of electronic resources. Retrieved November 21, 2001, from http://www.cas.usf.edu/english/walker/apa.html. b. Lecture note Format: Author(s). (Date of presentation). Lecture title. Lecture notes distributed in the unit, at the name of the teaching organisation, the location. Example: Liffers, M. (2006, August 30). Finding information in the library. Lecture notes distributed in the unit Functional Anatomy and Sports Performance 1102, University of Western Australia, Crawley, Western Australia. c. Patent Format: Author. (Year). Patent number. The location. Issue body. Example: Smith, I. M. (1988). U.S. Patent No. 123,445. Washington, D.C: U.S. Patent and Trademark Office. d. Standard Format: Issue body. (Year). Standard name. Standard number. The location. Example: Standards Association of Australia. (1997). Australian standard: Pressure equipment manufacture. AS4458-1997. North Sydney. e. Computer software Format: Author(s). (Year). Title [computer software]. The location: Company. Example: Ludwig, T. (2002). PsychInquiry [computer software]. New York: Worth. 57

EJAE 2015

Instructions for authors

Tables and figures and/or schemes should not be embedded in the manuscript but their position in the text indicated. In manuscript tables and figures should follow the text, each on a separate page.

6. TABLES Authors should use tables to achieve concise presentation or where the information cannot be given satisfactorily in other ways. Tables should be prepared with the aid of the Word table function, without vertical lines. The minimum size of the font in the tables should be 10 pt. Tables should not be incorporated as graphical objects. Styles and fonts should match those in the main body of the article. Tables should follow the text on the end of the manuscript and should be numbered consecutively using Arabic numbers and their position in the text should be indicated. Each table should have an explanatory caption which should be as concise as possible.

7. FIGURES Authors may use line diagrams to illustrate theses from their text. The figures should be clear, easy to read and of good quality. Styles and fonts should match those in the main body of the article. Lettering and lines should be of uniform density and the lines unbroken. Axis labels should be in bold face. Units should be placed next to variables in parentheses. Figures should follow the text on the end of the manuscript and should be numbered consecutively using Arabic numbers and their position in the text should be indicated. Mathematical equations should be embedded in the text. Complex equations should be prepared with the aid of the Word Equation editor. All equations must be numbered, Arabic numbers, consecutively in parenthesis at the end of the line, as presented:

= NSV

â&#x2C6;&#x2018; NP k =0

â&#x2039;&#x2026; ak

(1)

CIP - Каталогизација у публикацији Народна библиотека Србије, Београд 33 The EUROPEAN Journal of Applied Economics / editor-inchief Milovan 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