Country Portfolio Model Considering Market Uncertainties in Construction Industry

Size: px

Start display at page:

Download "Country Portfolio Model Considering Market Uncertainties in Construction Industry"

Alicia Little
5 years ago
Views:

1 CCC 2018 Proceedigs of the Creative Costructio Coferece (2018) Edited by: Miroslaw J. Skibiewski & Miklos Hajdu Creative Costructio Coferece 2018, CCC 2018, 30 Jue - 3 July 2018, Ljubljaa, Sloveia Coutry Portfolio Model Cosiderig Market Ucertaities i Costructio Idustry Wo Ji Woo a, Seug-Heo Ha b *, Woosik Jag c, a Departmet of Civil ad Evirometal Egieerig i Yosei Uiversity, Seodaemu-gu, Seoul ad 03722, Republic of Korea b Korea Istitute of Civil Egieerig ad Buildig Techology, Ilsaseo-gu, Goyag-si, Gyeoggi-do ad 10223, Republic of Korea c Departmet of Civil ad Evirometal Egieerig i Chosu Uiversity, Dog-gu, Gwagju ad 61452, Republic of Korea Abstract I recet decades, market ucertaities such as upredicted ecoomic recessios ad expasios have sigificatly affected to the iteratioal costructio market. These ucertaities arise simultaeously either at the coutry level or more broadly ad traditioal project-based risk maagemet has limited to maage a cotractor profit. From the perspective of maagig the market ucertaities, therefore, this study proposes coutry portfolio model that provides a optimized coutry portfolio solutio through cosiderig o the market outlook, coutry risk ad expected profitability. These are evaluated by coutry-specific data related costructio market ad actual project performace data usig mote-carlo simulatio ad geetic algorism. It is expected that the proposed coutry portfolio model will support to decide better decisio about eterig ew iteratioal costructio market by givig the ideal coutry portfolio cosiderig market ucertaities The Authors. Published by Diamod Cogress Ltd. Peer-review uder resposibility of the scietific committee of the Creative Costructio Coferece Keywords: Coutry Portfolio; Markrt ucertaity; Iteratioal costrcutio market. 1. Itroductio I recet decades, market ucertaities such as upredicted ecoomic recessios ad expasios have sigificatly affected to the iteratioal costructio market. I geeral, these ucertaities arise simultaeously either at the coutry level or more broadly [1] (e.g., the Asia fiacial crisis ad the global fiacial crisis); I additio, sice the costructio busiess is a order-based idustry, ulike other busiesses such as maufacturig, the success or failure of a project is highly iflueced by the ucertaities which is caused by characteristics of the host coutry ad its cliets. Thus, traditioal project-based risk maagemet has limited to maage a cotractor s profit [1]. I such rapidly chagig ad highly ucertai host coutry eviromets, diversificatio is oe of the strategic decisios made by iteratioal costructio firms to maage market ucertaities. There are may studies is coducted the costructio market risk ad diversificatio. However, there are still questios about how to maage market ucertaities. Therefore, there is a eed to study which is to cotrol market ucertaities through a better coutry portfolio. The purpose of study is to propose the coutry portfolio model cosiderig market ucertaities i costructio idustry by usig the mote-carlo simulatio ad geetic algorithm. This model ca provide weight portio of coutries amog portfolio. Correspodig author: Author shh6018@yosei.ac.kr 411

2 CCC 2018 Proceedigs 2. Research Methodology 2.1. Modelig Frameworks This study proposes coutry portfolio model that provides a optimized coutry portfolio solutio based o the market outlook, coutry risk ad expected profitability. The market outlook meas how the market will chage. I this model, it is assessed by costructio market s growth rate ad volatility of the market size. The coutry risk meas the risk caused by ature of the host coutry which cosists of the political risk, executio risk ad istitutioal risk. it is assessed by the coutry risk score ad volatility of the coutry risk. The expected profitability meas what the profit will be ad how the profit will chage. it is assessed by the expected profit rate ad volatility of profit performace. The proposed model utilizes a mote-carlo simulatio ad geetic algorithm. The mote-carlo simulatio is utilized to determie the probability distributio of coutry risk ad expected profitability. From the probability distributio, the coutry risk score, volatility of coutry risk, expected profit rate ad volatility of the profit performace is obtaied. The geetic algorithm is utilized to provide a optimized coutry portfolio solutio which is to maximize market growth rate ad expected profit rate ad while to miimize volatility of the market size, coutry risk score, volatility of coutry risk ad volatility of profit performace Data Collectio To calculate the market outlook, coutry risk ad expected profitability, this study collects the secodary data o cadidate coutries ad actual project performace data. Coutry-specific data related costructio market were collected from iteratioally reputable istitutios, such as the World Bak [2], IHS Global Isight [3]. The project performace data were obtaied from the Iteratioal Cotractors Associatio of Korea (ICAK) [4] which has iformatio o iteratioal project performace of Korea compaies. Furthermore, this study selected 14 coutries which are Uited Arab Emirates, Taiwa, Germay, Malaysia, Uited States, Vietam, Saudi Arabia, Sigapore, UK, Oma, Kuwait, Thailad, Pakista, Philippies. There are two reasos for choosig these coutries. Oe is that these coutries are beloged as mai or promisig coutries i Korea. The other reaso is the data availability. 3. The Proposed Coutry Portfolio Model 3.1. Market Outlook The market outlook is dealt with how the market will chage. It cosists the costructio market s growth rate ad volatility of the market size The costructio market s growth rate meas the expected costructio spedig growth [5]. This study utilizes the forecasted data of the size of each costructio market i the Global Costructio Database provided by IHS Global Isight [3]. The costructio market s growth rate is measured by the compoud aual growth rate (CAGR) from Volatility of the market size is a cocept that is cotrary to costructio market stability. Volatility of the market size [6] is measured by the percetage of costructio market volatility (σ) usig IHS Global Isight [3] data for each costructio market size over te years ( ); Volatility(σ) = 1 (x 1 i=1 i x ) 2, ( x i = l S i ) (1) S i 1 Where S i = the costructio market size at i year, x = average of x i ad = the size of the costructio market size data set The costructio market s growth rate (CAGR) ad volatility of the market size are derived as followed Table

3 CCC 2018 Proceedigs Table 1. Market outlook, Coutry risk ad Expected profitability of cadidate coutries Coutry Growth Rate (CAGE) Market outlook Coutry risk Expected profitability Volatility (σ o) Risk Score Volatility (σ r) Profit Volatility (σ p) Uited Arab Emirate (UAE) Taiwa Germay Malaysia Uited Stated of America (USA) Vietam Saudi Arabia Sigapore Uited Kigdom (UK) Oma Kuwait Thailad Pakista Philippies Coutry Risk I geeral, coutry risk has oly bee evaluated or cosidered as a discrete value provided by WGI, DBI, BMI ad the other data source. However, whe the coutries have similar risk levels, it is ecessary to evaluate the coutry by cosiderig the volatility of coutry risks rather tha the subtle risk differeces. To take accout of this ucertaity, coutry risk is calculated with the political risk, executioal risk ad istitutioal risk by usig mote-carlo simulatio. The calculatio process is cosisted 4 steps. First step is to calculate the political risk value which explais each coutry s level of maturity ad political stability. It is evaluated by usig the worldwide goverace idicators (WGI) provided by the World Bak [2]. WGI provide the estimate score (the high estimate score meas high quality) ad the stadard deviatio of six factors: voice ad accoutability; political stability ad absece of violece; govermet effectiveess; regulatory quality; rule of law; ad cotrol of corruptio. I order to express the ucertaity of political risk, mote-carlo simulatio is coducted. The political risk value is derived by summig the probability distributios for each six dimesio factors by applyig the ormal distributio(~(factor s estimate score, factor s stadard deviatio)). Secod step is to calculate the executioal risk value which explais each coutry s level of busiess efficiecy ad flexibility. This risk is measured by usig the doig busiess idicators provided by the World Bak [2]. It cosists te factors: startig a busiess; dealig with costructio permits; gettig electricity; registerig property; gettig credit; protectig miority ivestors; payig taxes; tradig across borders; eforcig cotracts; ad resolvig isolvecy. I order to express the ucertaity of executioal risk, mote-carlo simulatio is also coducted. However, sice DBI do t provide the stadard deviatio ulike WGI, the probability distributio is derived by data fittig which is fittig the overall distace to frotier (DTF) scores from (See Fig. 1). The DTF meas the busiess efficiecy ad flexibility (the high DTF score meas high efficiecy). The criterio of choosig the probability distributio is the Akaike Iformatio Criterio(AIC) ad subjective opiio based o the form of data set. 413

CCC 2018 Proceedigs Fig. 1. The probability distributio of Vietam executio risk value. Third step is to calculate istitutioal risk value which explais each coutry s level of istitutioal maturity.

Fial step is to calculate the coutry risk value. The coutry risk value is derived by summig the three risk values which is political, executioal ad istitutioal risk values ad it is subtracted from 10.

4 CCC 2018 Proceedigs Fig. 1. The probability distributio of Vietam executio risk value. Third step is to calculate istitutioal risk value which explais each coutry s level of istitutioal maturity. This risk is measured by usig istitutioal profiles database (IPD). Sice there is o data to defie the probability distributio, this risk is used the discrete value to calculate the coutry risk. Fial step is to calculate the coutry risk value. The coutry risk value is derived by summig the three risk values which is political, executioal ad istitutioal risk values ad it is subtracted from 10. The equatio as follows: Coutry risk value = 10 (w p S p + w e S e + w i S i ) (2) w p + w p + w p = 1 (3) Where S P = value of political risk, w p = weight of political risk value S e = value of executioal risk w e = weight of executioal risk value S i = value of istitutioal risk ad w i = weight of istitutioal risk value From the coutry risk value s probability distributio, the coutry risk score which is the mea of the probability distributio ad volatility of the coutry risk which is the stadard deviatio of the probability distributio are derived as followed Table 1. (Weight is assumed that the w p is 0.3, w p is 0.6 ad w p is 0.1) As show i Fig. 2, you ca see why cosiderig the volatility is ecessary. If decisio-maker simply compare the risk scores, Sigapore may be a better coutry tha the UK. However, i terms of volatility, the UK is more stable tha Sigapore because Sigapore is more likely to higher risk scores tha the UK. Fig. 2. The coutry risk probability distributio (Sigapore, Uited Kigdom). 414

The mai reaso for this result is that the weight of executioal risk is higher tha that of political risk. The sesitivity aalysis results will chage with each weight chage. 3.

5 CCC 2018 Proceedigs The sesitivity aalysis is also coducted (see Fig. 3). Although sesitivity aalysis results were differet for each coutry, i most coutries (except Germay, Kuwait), the sesitivity of executioal risk is greater tha that of political risk. The mai reaso for this result is that the weight of executioal risk is higher tha that of political risk. The sesitivity aalysis results will chage with each weight chage Expected profitability Fig. 3. The result of Sesitivity aalysis (UAE, Germay) Expected profitability is measured by usig the actual iteratioal egieerig ad costructio project profit data [4] executed by Korea cotractors across 14 coutries over the past 25 years. I order to obtai the expected profit rate ad volatility of profit performace, the probability distributio is derived by data fittig (Fig.4). This study removes some data which are too much higher, lower ad zero profit rate to improve the Data fittig quality. The criterio of choosig the probability distributio is the Akaike Iformatio Criterio(AIC) ad subjective opiio based o the form of data set. Fig. 4. The probability distributio of UAE profit performace From this probability distributio, the expected profitability which is the mea of the probability distributio ad volatility of project performace which is the stadard deviatio of the probability distributio are derived as followed Table

6 CCC 2018 Proceedigs 3.4. Coutry portfolio solutio from Geetic algorithm This study develops the objective fuctios to fid the best portfolio solutios that maximize market growth rate ad expected profit rate ad while to miimize volatility of the market size, coutry risk score, volatility of coutry risk ad volatility of profit performace. Before coductig geetic algorithm, this study uses the stadard deviatio method to calculate variables with the same dimesio (stadardized score). The stadardized score of variable (Z Vi ) is calculated as follows: Z Vi,j = V i,j μ i,j σ i,j (4) Where V i,j = the origial value of variable i i coutry j, μ i,j = the average value of variable i i coutry j, ad σ i,j = the stadard deviatio for variable i i coutry j. The objective fuctio as show i the followig equatios: Miimize f(x) = i=1 x i Z V1,i + i=1 x i Z V2,i + i=1 x i Z V3,i + i=1 x i Z V4,i i=1 x i Z V5,i + x i Z V6,i Where f = market portfolios (x) x i = weight portio of coutry i amog the portfolios Z V1,i = The stadardized score of market growth rate Z V2,i = The stadardized score of volatility of market size Z V3,i = The stadardized score of coutry risk score Z V4,i = The stadardized score of volatility of coutry risk Z V5,i = The stadardized score of expected profit rate Z V6,i = The stadardized score of volatility of profit performace ad = Number of coutry The Costrait coditios as show i the followig equatios: i=1 (5) i=1 x i = 1 (6) 0 x i 1 (7) Uder these coditios, the geetic algorithm is coducted. This model ca fid out optimal solutio withi 70 iteratios (see Fig.5) ad the weight portio of each coutry is derived (see Table 2). The result meas that if decisiomaker wat to maximize market growth rate ad expected profit rate ad while to miimize volatility of the market size, coutry risk score, volatility of coutry risk ad volatility of profit performace, He or she should allocate highest weight portio of Vietam amog the portfolios. Fig. 5. The result of Geetic Algorithm (plot the fitess value) 416

7 CCC 2018 Proceedigs Table 2. Weight portio of coutry amog the portfolios Coutry Weight portio Uited Arab Emirate (UAE) 0.20% Taiwa 3.00% Germay 5.10% Malaysia 2.90% Uited Stated of America (USA) 2.10% Vietam 67.70% Saudi Arabia 2.80% Sigapore 2.40% Uited Kigdom (UK) 3.50% Oma 0.20% Kuwait 0.00% Thailad 2.50% Pakista 3.50% Philippies 4.30% 4. Discussio ad Coclusio I rapidly chagig ad highly ucertai host coutry eviromets, diversificatio is oe of the strategic decisios made by iteratioal costructio firms to maage market ucertaities. From the perspective of maagig the market ucertaities, this study proposes coutry portfolio model that provides a optimized coutry portfolio solutio based o the market outlook, coutry risk ad expected profitability by usig mote-carlo simulatio ad geetic algorithm. The model s fial output is weight portio of coutry amog the portfolios. There are some advatages. First advatage is that this model is possible to support decisio makers by presetig probability distributios rather tha simple discrete values through mote-carlo simulatio. Secod it model ca help to make a more accurate portfolio solutio cosiderig the ucertaity of couty risk. Furthermore, sice it uses accessible data such as WGI, DBI, GI ad so o, it has a advatage that it ca be easily used by ayoe. However, there are may limitatios. The variables to be cosidered are too few ad simple to make a coutry assessmet. For example, competitiveess of order ad itesity of competitio are very importat factors i coutry assessmet, but they do ot reflect these. Also, sice the model is costructed by relyig o geeral data, it is less usability i practical area. Fially, the objective fuctio ad costrait coditios are too simple. There are a lot of variables ad ucertaities which affect the coutry assessmet. I the future study, if a model is developed that ca be aalyzed by addig uobserved variables such as competitive stregth ad the characteristics of coutry such as experiece, host govermet, I thik it ca be used as a support tool. Also, from the optimizatio poit of view, I thik that it ca be a good model if a more realistic optimal fuctio ad costrait is set. Ackowledgemets This work was supported by a grat fuded by Miistry of Lad, Ifrastructure ad Trasport of Korea govermet (18SCIP-C ) Refereces [1] Jug, W., Ha, S. H., & Lee, K. W., Coutry portfolio solutios for global market ucertaities. J. Maage. Eg. 28 (2011), [2] World Bak. (2018). World Bak ope data. hhttp://data.worldbak.org/i [3] IHS Global Isight. Global costructio outlook, (2017). Lexigto, MA. [4] ICAK (Iteratioal Cotractors Associatio of Korea). Iteratioal costructio iformatio service.,(2018) <hhttp:// [5] Lee, K. W., & Ha, S. H., Quatitative aalysis for coutry classificatio i the costructio idustry. J. Maage. Eg. 33 (2017), [6] Lueberger, D. G. (1998). Ivestmet sciece, Oxford Uiversity Press, New York. 417

Optimizing of the Investment Structure of the Telecommunication Sector Company

Optimizing of the Investment Structure of the Telecommunication Sector Company Iteratioal Joural of Ecoomics ad Busiess Admiistratio Vol. 1, No. 2, 2015, pp. 59-70 http://www.aisciece.org/joural/ijeba Optimizig of the Ivestmet Structure of the Telecommuicatio Sector Compay P. N.