Sun Hung Kai Properties Ltd, etc. Five in-depth interviews TABLE 1. Risks Associated with Sino-Foreign Construction were conducted in December 1998 and February 1999 that Joint Ventures supported the survey analysis. Rick classification a It is widely accepted that construction activity is particularly (alphabetical order) bject to more risks than other business activities because of its complexity, and a wide range of risks associated with con- struction businesses have been previously identified. A typical (1)Financial risk Bankr classification of risks includes technical risks, management Difficult convertibility of RMB risks, market risks, legal risks, financial risks, and political Loss due to fluctuation of inflation rate risks(Shen 1997). He(1995) presented some examples of Loss due to fluctuation of interest rate risks involved in foreign investment in China. Based on pre Loss due to fuctuation of rmb exchange rate vious works. a list of 58 risk factors. as shown in Table 1 ow credibility of shareholders and lenders were constructed and presented to respondents. The main pur (2)Legal Breach of contracts by other participants 0.222 pose of the survey is not to develop a new list of risks but to Breach of contracts by project partner 0.2336 Mo alyze the relative significance among the risks identified and 0.3345 Loss due to insufficient law for joint ventures 0.3505 The respondents were requested to judge the significance or ncertainty and unfairness of court justic 0.4319 expected loss"of each risk. There are many criteria that re 0.1641 spondents may need to consider. One alternative approach Change of organization within local partner mproper project feasibility study adopted by previous researchers is to consider two attributes 459 for each risk: the probability level of the risk occurrence, de mproper selection of project location noted by a; and the degree of impact or the level of loss if Improper selection of project type 0.5523 the risk occurs, denoted by B(He 1995; Li 1997). Therefore Inadequate choice of project partner risk significance, denoted as RS. can be described as the func- Inadequate project organization structure tion of the two attributes ncompetence of project management team Incomplete contract terms with partner oject management overhead 0.334 RS=f(α,阝) Poor relation and disputes with partner By applying this approach, the respondents were asked to re- Poor relation with govemment departments 0.3935 Problems associated with culture difference 0.1791 pond to the two attributes for each risk. For considering ox Project delay 0.6364 the respondents were required to judge the probability level of (4)Market risk occurrence by selecting one from among three levels, namely, low possibility, normal possibility, and high possibil- Fall short of expected income from project ity. For considering B, the respondents were required to judge Increase of accessory facilities price the degree of impact if the risk concerned occurs, by selecting f labor costs one from among three grades, namely, small impact, neutral Increase of materials pric rease of resettlement costs 0.5500 mpact, and large impact. Local protectionism ANALYSIS OF SURVEY RESULTS 0.1708 (5)Policy and political risk To assess the relative significance among risks, this study Cost increase due to changes of 0.6409 suggests to establish a risk significance index by calculating a Loss incurred due to corruption and briber 0.4595 0.2509 core for each risk. An alternative for calculating Loss incurred due to political changes Loss due to bureaucracy for late approvals 0.5141 he significance score is to multiply the probability of occur-(6)Technical risk rence by the degree of impact. Thus, the significance score for Accidents on site each risk assessed by each respondent can be obtained through the model Errors in design 0.2518 Hazards of en where S,=significance score assessed by respondent for risk 0.2714 i;af= probability of occurrence of risk i, assessed by respon- dent j; and B,=degree of impact of risk i, assessed by re Materials short 0.0518 Obsoleteness of building equipment 0.1305 y averaging scores from all 54 responses, it is possible to get an average sl cance e for each risk. and this av Poor quality of procured materials erage score is called the risk index score and is used to rank among all risks. The model he calculation of risk index Shortage in accessory facilities score can be written hortage in skillful workers 0.1527 Shortage in supply of water, gas, and electricity Unknown site physical conditions 0.2345 Unusual weather and force majeure 0.1145 where RS= index score for risk i; and S= significance score "neutral"take a value of 0.5, and that"high"and"large assessed by respondent j for risk i take a value of 1. By adopting these numerical scales, when To calculate S/, the three-point scales for a and p(low- the 54 responses are applied to model 2 and 3, the index scores normal-high and small-neutral-large)need to be converted int for all risks are calculated. as shown in Table 1. Based on the numerical scales. As an alternative method, it is suggested that index scores, risk ranking is established, as shown in Fig. 1 low"and"small"take a value of 0. 1, that"normal"and The risk ranking provides a useful reference to help joint ven JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT JANUARY/FEBRUARY 2001/77JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / JANUARY/FEBRUARY 2001 / 77 TABLE 1. Risks Associated with Sino-Foreign Construction Joint Ventures Rick classification (alphabetical order) (1) Index score (2) (1) Financial risk Bankruptcy of project partner 0.2595 Difficult convertibility of RMB 0.3091 Loss due to fluctuation of inflation rate 0.3236 Loss due to fluctuation of interest rate 0.3650 Loss due to fluctuation of RMB exchange rate 0.2500 Low credibility of shareholders and lenders 0.4555 (2) Legal risk Breach of contracts by other participants 0.2223 Breach of contracts by project partner 0.2336 Lack of enforcement of legal judgment 0.3345 Loss due to insufficient law for joint ventures 0.3505 Uncertainty and unfairness of court justice 0.4319 (3) Management risk Change of organization within local partner 0.1641 Improper project feasibility study 0.6386 Improper project planning and budgeting 0.4591 Improper selection of project location 0.5796 Improper selection of project type 0.5523 Inadequate choice of project partner 0.5229 Inadequate project organization structure 0.3163 Incompetence of project management team 0.4023 Incomplete contract terms with partner 0.4933 Increase in project management overheads 0.3341 Poor relation and disputes with partner 0.1932 Poor relation with government departments 0.3935 Problems associated with culture difference 0.1791 Project delay 0.6364 (4) Market risk Competition from other similar projects 0.4614 Fall short of expected income from project use 0.2827 Increase of accessory facilities price 0.1500 Increase of labor costs 0.0391 Increase of materials price 0.1786 Increase of resettlement costs 0.5500 Inadequate forecast about market demand 0.5819 Local protectionism 0.4941 Unfairness in tendering 0.1708 (5) Policy and political risk Cost increase due to changes of policies 0.6409 Loss incurred due to corruption and bribery 0.4595 Loss incurred due to political changes 0.2509 Loss due to bureaucracy for late approvals 0.5141 (6) Technical risk Accidents on site 0.1214 Design changes 0.4978 Equipment failure 0.1141 Errors in design drawings 0.2518 Hazards of environmental regulations 0.1563 Incompetence of transportation facilities 0.0377 Increase in site overheads 0.2714 Industrial disputes 0.0722 Local firm’s incompetence and low credibility 0.4664 Materials shortage 0.0518 Obsoleteness of building equipment 0.1305 Poor quality of procured accessory facilities 0.2234 Poor quality of procured materials 0.2459 Problems due to partners’ different practice 0.3205 Shortage in accessory facilities 0.1450 Shortage in skillful workers 0.1527 Shortage in supply of water, gas, and electricity 0.2450 Subcontractor’s low credibility 0.3909 Unknown site physical conditions 0.2345 Unusual weather and force majeure 0.1145 Sun Hung Kai Properties Ltd., etc. Five in-depth interviews were conducted in December 1998 and February 1999 that supported the survey analysis. It is widely accepted that construction activity is particularly subject to more risks than other business activities because of its complexity, and a wide range of risks associated with con￾struction businesses have been previously identified. A typical classification of risks includes technical risks, management risks, market risks, legal risks, financial risks, and political risks (Shen 1997). He (1995) presented some examples of risks involved in foreign investment in China. Based on pre￾vious works, a list of 58 risk factors, as shown in Table 1, were constructed and presented to respondents. The main pur￾pose of the survey is not to develop a new list of risks but to analyze the relative significance among the risks identified and to highlight the major risks. The respondents were requested to judge the significance or ‘‘expected loss’’ of each risk. There are many criteria that re￾spondents may need to consider. One alternative approach adopted by previous researchers is to consider two attributes for each risk: the probability level of the risk occurrence, de￾noted by a; and the degree of impact or the level of loss if the risk occurs, denoted by b (He 1995; Li 1997). Therefore, risk significance, denoted as RS, can be described as the func￾tion of the two attributes RS = f(a, b) (1) By applying this approach, the respondents were asked to re￾spond to the two attributes for each risk. For considering a, the respondents were required to judge the probability level of risk occurrence by selecting one from among three levels, namely, low possibility, normal possibility, and high possibil￾ity. For considering b, the respondents were required to judge the degree of impact if the risk concerned occurs, by selecting one from among three grades, namely, small impact, neutral impact, and large impact. ANALYSIS OF SURVEY RESULTS To assess the relative significance among risks, this study suggests to establish a risk significance index by calculating a significance score for each risk. An alternative for calculating the significance score is to multiply the probability of occur￾rence by the degree of impact. Thus, the significance score for each risk assessed by each respondent can be obtained through the model i ii Sj jj = a b (2) where = significance score assessed by respondent j for risk i Sj i; = probability of occurrence of risk i, assessed by respon- i aj dent j; and = degree of impact of risk i, assessed by re- i bj spondent j. By averaging scores from all 54 responses, it is possible to get an average significance score for each risk, and this av￾erage score is called the risk index score and is used to rank among all risks. The model for the calculation of risk index score can be written 54 i O Sj i j=1 RS = (3) 54 where RSi = index score for risk i; and = significance score i Sj assessed by respondent j for risk i. To calculate , the three-point scales for a and b (low- i Sj normal-high and small-neutral-large) need to be converted into numerical scales. As an alternative method, it is suggested that ‘‘low’’ and ‘‘small’’ take a value of 0.1, that ‘‘normal’’ and ‘‘neutral’’ take a value of 0.5, and that ‘‘high’’ and ‘‘large’’ take a value of 1. By adopting these numerical scales, when the 54 responses are applied to model 2 and 3, the index scores for all risks are calculated, as shown in Table 1. Based on the index scores, risk ranking is established, as shown in Fig. 1. The risk ranking provides a useful reference to help joint ven-