This article appeared in a journal published by Elsevier.

The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elseviers archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Author's personal copy

Available online at www.sciencedirect.com

International Journal of Project Management 29 (2011) 108116 www.elsevier.com/locate/ijproman

Impacts of multi-layer chain subcontracting on project management performance

Vivian W.Y. Tam a,*, L.Y. Shen b, Joseph S.Y. Kong b
a

School of Engineering, University of Western Sydney, Locked Bag 1797, Penrith South DC, NSW 1797, Australia b Dept. of Building and Real Estate, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong Received 11 June 2009; received in revised form 13 January 2010; accepted 26 January 2010

Abstract This paper investigates the impacts of applying the multi-layer chain subcontracting system on project management performance with reference to Hong Kong construction industry. Multi-layer chain subcontracting system is widely used within construction industry as it is considered advantageous in many aspects such as better eciency of subcontractors operation due to their unique skills. However, the fact of poor quality products in construction practice raises the doubt about the eectiveness of the chain system. Accordingly, the reasons why the applications of the system contribute to poor project performance are examined. A survey conducted in the Hong Kong construction industry demonstrates that the multi-layer chain subcontracting system, while widely adopted, is largely awed. Based on the survey results, application of multi-layer chain subcontracting system contributes largely to the poor performance across the all major aspects including quality and time management, cost control, and communication and coordination performance. The association exists between poor project management performance and the increase of the number of layers in the chain of the subcontracting arrangement. The long communication chain because of the increasing layers of subcontractors results in various problems such as communication errors, poor supervision on the bottom-layer contractors. Consequently overruns in cost and time, and abortive and remedial works are common. Recommendations for improving the practice are suggested and explored, including change the practice of the lowest bid to an approach which incorporate both price and technical performance, limit the number of subcontracting layers, restrain the use of supply-and-x subcontracting arrangement, and enforce the implementation of government regulations. The ndings of this study provide useful references in examining the practice of subcontracting system in other construction industries and identifying the areas where the improvements can be made for gaining the benets of using the system. 2010 Elsevier Ltd and IPMA. All rights reserved.
Keywords: Multi-layer chain subcontracting; Project management; Project performance; Construction; Hong Kong

1. Introduction The subcontracting system is usually described as the contractual process in which a main contractor subcontracts parts of the job to another contractor, who may also subcontract to another rm or further subcontract (Chiang, 2009). For example, a contractor can subcontract work to a concreter, a steelworker, a mechanic and an electronic installation rm, or a plumber. The concrete subcontractor or the steelworker further subcontracts to a third
*

Corresponding author. Tel.: +61 02 4736 0105. E-mail address: vivianwytam@gmail.com (V.W.Y. Tam).

party and this party can also further subcontract work. This multi-layer supply chain arrangement can graphically be presented in Fig. 1. A single main contractor cannot possibly handle all related project tasks. The delivery of a construction project involves dierent skills at dierent construction stages and these skills are usually managed in dierent organizations (Construction Industry Review Committee, 2001). Coupled with the uctuating amounts of construction work in the Hong Kong construction industry, in particular over last decade, using the multi-tier supply chain subcontracting system is considered as an eective approach for avoiding a changeable demand on the main contractors own in-

0263-7863/$36.00 2010 Elsevier Ltd and IPMA. All rights reserved. doi:10.1016/j.ijproman.2010.01.005

Author's personal copy

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

109

Concreting Project Client Steelwork

Formwork

M Main Contractor

&

Plumbing

Tiling

Cleaning

1st layer

2nd layer

3rd layer

4th layer nth layer

Fig. 1. A multi-layer supply chain subcontracting system in implementing a construction project.

house resources (Ng et al., 2009). Subcontractors could bring unique skills and talents for specialized work such as steel work, installation and other sophisticated facility systems (Hinze and Tracey, 1994). It had also shown benets in using information systems for supply chain management practices (Bayraktar et al., 2009; Soroor et al., 2009). Under multi-layer supply chain subcontracting arrangement, a main contractor will rarely need to invest on training programs or sponsor training to site workers whom they do not directly employ (Ofori and Debrah, 1998). It is said that the risk for main contractors to pay overtime is also reduced as subcontractors are often contracted for individual tasks and not based on a period of time. In the United Kingdom, subcontractors are used to relieve main contractors from providing training, pension rights, redundancy payments and sick leave. This can signicantly reduce main contractors expensise to the project. Main contractors can thus reduce general expenses and yet, still accept and complete the same amount of work (Wong and Chan, 1997). Subcontracting in construction business is applied worldwide, and generally larger projects will involve more layers in the subcontracting arrangement. If the use of the system is properly monitored and all subcontractors are adequately instructed, a project team can eectively and eciently work towards common goals (Elazouni and Metwally, 2000). Multi-layer chain subcontracting practice is widely used in the Hong Kong construction industry. The system is a long established practice and will continue to function in the local construction market (Construction Industry Review Committee, 2001). The subcontracting practice dramatically developed in the local industry during the

construction boom in mid-1990s when construction rms could not manage too many work oered in the industry. During this boom period, contractors at various levels had to subcontract work to fulll market demand. However, there are criticisms that in the practice, the benets of using the subcontracting system have not eectively been gained. It is reported that there are notable aws in the application of the system in Hong Kong (Construction Industry Review Committee, 2001). Construction quality in the local construction industry suers in the absence of adequate control and supervision by main contractors over subcontractors work. The cutting corner practice in subcontractors work is common. The responsibility is diused along the hierarchy of subcontractors, and construction quality accordingly deteriorates (Tam et al., 2000). It is often dicult to identify who does what and who is responsible for the quality of the work in multi-layer chains. Furthermore, the bottom-layer subcontractors are often contracted for unreasonably low bids. In order to make a prot, these bottom tiers have to use cheap or deteriorated materials, employ unskilled workers and engage poor workmanship, thus deliver inferior products (Kam and Tang, 1998; Low and Sua, 2000). This normally results in inevitable sacrices of quality, as bottom-layers have to survive on cutting corners by applying poor quality resources and ignoring construction regulations such as safety, labor ordinances, insurance policies and environmental issues (Elazouni and Metwally, 2000; Yik and Lai, 2008). Poor site management and supervision are also one of the major common causes of delay (Chan and Kumaraswamy, 1997). There is a growing consensus that connection exists between the multi-layer supply chain subcontracting practice and poor project management performance in the Hong Kong construction. Through examining the eects of using the subcontracting system on the key aspects of project management, this paper examines the reasons why the application of the subcontracting system contributes to the poor project performance. The data used for analysis is from a survey to the local construction industry. Following the analysis on the factors contributing to the poor management performance from using multi-layer subcontracting system, recommendations to improve the existing practice are suggested and explored. 2. Research methodologies This study is focused to investigate reasons why the application of the multi-layer supply chain subcontracting contributes to poor project management performance in the Hong Kong construction. Data used for analysis is from a questionnaire survey in the Hong Kong construction industry. Twenty-one factors contributing to the ineectiveness of multi-layer chain subcontracting are identied based on the review of the existing studies (see Table 1). The ineectiveness is evidenced by poor quality performance, poor time

Author's personal copy

110

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

Table 1 Response results from the survey on the impact reasons from the practice of applying multi-layer supply chain subcontracting to project management performance. Reasons for the impacts to project performance For poor quality performance Q1. Employing inferior materials by bottom-layer subcontractors Q2. Employing poor labor force by bottom-layer subcontractors Q3. Subcontractors irresponsible for quality because of limited prot Q4. Subcontractors improper work practice Q5. Subcontractors technical incompetence in performing quality work Q6. Subcontractors non-compliance to quality specications For poor time management performance T1. Setting unrealistic contract time in subcontracting arrangements T2. Subcontractors low eciency as expecting limited gaining in the end T3. Subcontractors late response to instructions because of long chain of communication T4. Time consuming on remedial work T5. Time consuming on solving disputes among various layers of subcontractors For poor cost control performance C1. More overheads for managerial sta involved along subcontracting chains C2. Extra cost as intermediate subcontractors charging fees without adding value C3. Increasing construction cost due to more abortive and remedial work C4. Increasing construction cost due to more claims and disputes For poor communication and coordination CC1. Delay in communicating decision to bottom-layers due to long chain CC2. Increasing communication errors when increasing layers of subcontractors CC3. Poor communication channel between main contractor and subcontractors CC4. Lack of communications between multi-layers CC5. Dicult in sharing timely information among multi-layer supply chain subcontractors CC6. Lack of main contractors mediation on disputes among subcontractors I 2 0 1 1 2 1 1 0 0 0 2 1 1 1 4 1 0 0 2 0 0 II 36 28 19 11 26 27 28 31 6 21 32 41 12 7 28 2 5 25 4 10 13 III 47 31 34 28 28 27 57 43 20 40 52 44 46 32 58 41 15 37 27 32 40 IV 46 68 76 86 72 73 50 57 94 67 50 51 58 82 46 78 96 66 91 86 79 V 8 12 9 13 11 11 3 8 19 11 3 2 22 17 3 7 23 11 15 11 7 Total 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 139 RII 0.632 0.688 0.706 0.742 0.692 0.694 0.638 0.66 0.782 0.68 0.628 0.618 0.726 0.754 0.624 0.712 0.798 0.69 0.762 0.742 0.716

management performance, poor cost control, and poor communication and coordination. In the process of collecting research data, the 21 factors are sent to 300 contractors listed in the Hong Kong Governmental List of Approved Contractors in the format of a questionnaire. Respondents are requested to indicate a particular grade of truthfulness of each factor. Grade I indicates strongly disagree and Grade V indicates strongly agree. Other grades are between the two extreme ends. As a result, 143 responses are received, with the response rate of about 47.7%. However, four returned responses are incomplete, thus only 139 are used for the analysis. To determine the relative ranking between the factors, the grade received from the questionnaire for each factor is transformed to indices based on (Tam et al., 2000): P w RII 1 AN where w is the weighting given to each factor by the respondent, ranging from 1 to 5, in which 1 denotes strongly disagree; 5 strongly agree; A the highest weight (A = 5); N the total number of samples; and RII the relative importance index, 0 6 RII 6 1 . After collecting the survey data, fourteen contractors indicate that they are agreed to conduct individual in-depth interviews to gather further comments, elaboration and interpretation on the results obtained from the survey. A

list of open-ended questions is used in the interview discussions (see Appendix A). The in-depth interviews are used to validate the results obtained from the survey to ensure the consistency between the questionnaire survey and in-depth interviews. Based on the interviewees experience, in-depth understanding of the existing problems in the system and recommendations are then explored. 3. Results and discussion Table 1 summarizes the survey results for the impact factors to the ineectiveness of multi-layer supply chain subcontracting, particularly under the headings of project quality performance, project time management performance, project cost control performance, and project communication and coordination. 3.1. Impacts of multi-layer chain subcontracting to project quality performance The survey shows that improper use of subcontracting system potentially exacerbates poor construction quality in the Hong Kong construction industry. The major reasons include: Q1 employing inferior materials by bottomlayer subcontractors; Q2 employing poor labor force by bottom-layer subcontractors; Q3 subcontractors irresponsible for quality because of limited prot; Q4 subcontractors improper work practice; Q5 subcontractors

Author's personal copy

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

111

technical incompetence in performing quality work; and Q6 subcontractors non-compliance to quality specications. In fact, in a construction market where there is intensive competition, such as in Hong Kong construction market, bottom-layer contractors usually adopt cutting corners methods to save cost at the expense of quality. Q4 subcontractors improper work practice received the highest value of RII. This was exemplied in the piling scandals in three housing estate projects in Hong Kong, where site supervisory sta were bribed by bottom-layer subcontractors to sign false reports indicating that the driven piles had reached the design depths (Kong, 2001). In these projects, it was later revealed that only 67% of the driven piles reached the design specication and 33% of piles were shortened to various extents. As the pile shortages would substantially weaken the structural stability, demolishment and rebuilding of these aected residential buildings have to be carried out. From Table 1, it should also be noted that factors Q3 subcontractors irresponsible for quality because of limited prot and Q5 subcontractors technical incompetence in performing quality work both are identied as important factors. The Hong Kong construction industry is largely prot driven often at the expense of quality. In adopting subcontracting system, a main contractor usually subcontract work to other contractors who will do the work for less than what the client pays to the main contractor. This payment dierence results in that subcontractor adopts a cheap and poor quality work practice to generate certain prots. An example is highlighted during the discussion with a surveyed practitioner. In an oce building project, a client oered $500,000 for tiling work to the nominated subcontractor A. Contractor A subcontracted the work to Subcontractor B for $450,000, thus gaining a $50,000 prot. Subcontractor B further subcontracted the work to Subcontractor C for $405,000, thus Subcontractor B could make a $45,000 prot for doing limited but only passing the job along. Along the chain, Subcontractor C also subcontracted to Subcontractor D the work for $370,000. When the works reached the Subcontractor D who actually would do the work, the payment from the client for the work was so signicantly reduced to the level that could not cover the necessary materials and labor cost. To make a living, the Subcontractor D employed cheap laborers and poor materials, resulting in the poor quality of the work. Upon the completion of the work, the client found that over 45% of the work was not acceptable. Another signicant factor is Q6 subcontractors noncompliance to quality specications which receives the value of RII with 0.69. It is appreciated during the survey that subcontractors non-compliance with specication is a major reason contributing to poor quality performance. In a project surveyed, the main contractor employed two subcontractors for the tiling work. Subcontractor A was responsible for blocks 610 while Subcontractor B was responsible for blocks 15, plus a car park and a kindergarten. Prior to the commencement of the construction, both

subcontractors visited sample ats and agreed to the workmanship standard set by the client. However, upon the completion of the work, the clients engineers inspected the site and found damaged wall tiles, misaligned tiling joint, uneven surfaces and poor cement grouting in the work by Subcontractor A. Investigation revealed that subcontractor A had subcontracted the work to a third-layer subcontractor who further subcontracted to a fourth-layer subcontractor. The bottom-layer subcontractor was oblivious to the clients expectations, did not visit the sample ats, and employed unskilled workers, resulting in poor quality performance. A further point can be noted that the main contractor did not provide adequate supervision on its subcontractors work as the shoddy work was not detected until work had been completed. On the other hand, the work by Subcontractor B was acceptable, contributed by the fact that the Subcontractor B did not subcontract the work and its directly hired workmen with regularly visited the sample ats for ensuring that the clients specications were met. This example indicates that the possibility of committing poor quality performance increases when there are more layers in the subcontracting arrangement. Considering the factors Q2 employing poor labor force by bottom-layer subcontractors and Q1 employing inferior materials by bottom-layer subcontractors, both are identied important with the values of RIIs with 0.688 and 0.632 respectively. The discussion with site engineers reveals an example to show the importance of the factors. Formwork is normally subcontracted to supply-and-x carpenters in the Hong Kong construction practice. Subcontractors normally reuse the timber for up to ten times before replacing it. The strength of the timber will decrease after used for many times. This often results in problems such as concrete bulging and honeycombed concrete. In fact, in a proper practice, timber for formwork should be replaced after ve/six times of use. However, to save the cost of timber materials, bottom-layer formwork contractors normally increase the number of usage times. Saving materials cost is considered an eective way to cut cost as materials generally account for the majority of the total construction cost in Hong Kong practice. On the other hand, in the local practice, the supply-and-x subcontracting arrangement is widely adopted, which allows the bottom-layer subcontractor to provide both construction materials and labor force. This practice is dierent to the labor-only contractual arrangement where subcontractors use the materials supplied by the main contractor. The supply-and-x practice presents opportunities to the bottom-layer subcontractors for cutting materials cost by procuring and using poor materials. The above discussions highlight that project quality can signicantly be downgraded by the use of multi-layer subcontracting system. The subcontractor who actually does the work normally does not have formal contractual relationship with the client. Therefore there is no direct communication and supervision from client to the bot-

Author's personal copy

112

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

tom-layer contractors. There is also lack of supervision from the main contractor to dierent levels of subcontractors, which is the main cause for the non-compliance in performing quality work and specication. As the number of subcontracting layer increases, limited prot can be gained for the subcontractor who actually does the work. Inferior materials therefore have to be adopted to reduce construction cost for fullling the budget. 3.2. Impacts of multi-layer chain subcontracting to project time management performance There are ve major factors identied, which cause the ineectiveness of using the multi-layer chain subcontracting system in controlling construction time. These include: T1 setting unrealistic contract time in subcontracting arrangements; T2 subcontractors low eciency as expecting limited gaining in the end; T3 subcontractors late response to instruction because of long chain of communication; T4 time consuming on remedial work; and T5 time consuming on solving disputes among various layers of subcontractors. From the survey results, it is found that the factor T3 subcontractors late response to instruction because of long chain of communication is the most signicant factor, receiving the value of RII of 0.782. It is appreciated during survey discussions that that multi-layer supply chain subcontracting arrangement contributes to spend extra time on the long communication chain. When a project client or a main contractor makes decision on matters such as issuing variation orders or site instructions, it will need to go through all the layers before it reaches the contractor that actually does the work. This long communication chain is costly and time consuming. Likewise, the bottom-layer subcontractor needs to go through the same lengthy channel to obtain instructions or approvals for responding to any problems occurred on site. However, in the construction practice, immediate action often needs to be taken in responding to problems, but the long chain communication delays this action, thus delays project progress. Furthermore, bottom-layer subcontractors are less ecient as they are less motivated. This can also explain the factor T2 subcontractors low eciency as expecting limited gaining in the end as a major factor with receiving the value of RII with 0.66. The interview discussion indicates that in the local practice, bottom-layer subcontractors are not only under-paid but often receive delayed interim or nal payments. In particular, nal payments are usually delayed, often exceeding 1 year after project completion. In return, subcontractors will delay to correct any defects identied, and they normally will hold a wait-and-see attitude, hoping that the main contractor or the upper layer subcontractors will correct the problems. The wait-and-see practice is common in the local construction, which signicantly delaying handover time.

From the survey results, it is found that T1 setting unrealistic contract time in subcontracting arrangements is not signicant to aect time performance with the RII of 0.638. However, interviewees argued that it is depended on the numbers of layers of subcontracting arrangements for the project. As the main contractor subcontracts the work to the rst layer of subcontractor, the main contractor will try to shorten the contract period for the rst layer of subcontractor to ensure the work can be nished within their timeframe to the client. The rst layer of subcontractor will also shorten the contract period if they subcontract the work to the second-layer of subcontractor. Therefore, the numbers of subcontracting arrangement will directly aect the real contract time for the activities. If the layer of subcontracting arrangement is high, it is common that remedial work and disputes may be happened. This can also explain T4 time consuming on remedial work and T5 time consuming on solving disputes among various layers of subcontractors. The above analysis demonstrates that there is a close association between the application of multi-layer chain subcontracting system and the poor project time management performance. Long communication chain for receiving instructions from the client or main contractor is the main contributor to the time delay performance. Solving disputes and conducting remedial work are also the contributors to the time delay. This can be further deteriorated if the number of the subcontracting layers increases. 3.3. Impacts of multi-layer chain subcontracting to project cost control performance The empirical survey in this study indicates that the improper application of multi-layer chain subcontracting arrangement can increase construction cost. The major reasons for this are identied as: C1 more overheads for managerial sta involved along subcontracting chains; C2 extra cost as intermediate subcontractors charging fees without adding value; C3 increasing construction cost due to more abortive and remedial work; and C4 increasing construction cost due to more claims and disputes. The data in Table 1 demonstrates that C3 increasing construction cost due to more abortive and remedial work is the most signicant factor and receives the value of RII with 0.754. It is suggested that subcontractors at the lower level in a multi-layer chain arrangement can easily give up their commitments once they expect that they have negative gain and suer loss. In particular, in the case that supply-and-x work is subcontracted to bottom-layers, when a bottom-layer suers loss or is in the instance of bankruptcy, he/she can easily either surrender the unnished work back to the preceding subcontractor or simply disappear by leaving the work unnished and workers unpaid. This often happens during rectication stage. Consequently, other contractors, main contractor or client has to bear the cost. As the number of subcontracting layers increases, the problems for increasing cost from remedial

Author's personal copy

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

113

work will become more serious. Argument between dierent layers of subcontractors is more likely to occur. This can also explain C4 increasing construction cost due to more claims and disputes. Another major factor contributing to extra cost in using multi-layer contracting is C2 extra cost as intermediate subcontractors charging fees without adding value, which receives the value of RII with 0.726. The discussions during the survey reveal that in the local practice, considerable part of construction cost goes to middle-layer subcontractors who do not actually add values to the project but simply charge a management fee. Under such circumstance, a middle-layer subcontractor may contract a job from its preceding contractor, and then subcontract the work to a further layer without adding any value but retaining a prot margin for itself. In other words, these middle-layer subcontractors are essentially being paid to do limited other than passing the job along. The payment to the middle-layer subcontractors who may not add any value which can be called nonvalue cost, and this cost will increase when the number of middle-layer subcontractors increases. Therefore, the bottom-layer subcontractors who carry out the actual works will receive insucient payments for the work when nonvalue cost becomes signicant, accordingly these do-thejob contractors are forced to cut corners in order to survive. The overheads cost is also the non-value cost explained in here. This can explain C1 more overheads for managerial sta involved along subcontracting chains when the layer of subcontracting arrangement increases. It is clear that cost overruns can be induced from the practice of multi-layer supply chain subcontracting. This is largely due to the increase of overhead cost, intermediate subcontractor charging fees, and non-value cost. These costs are not necessary and cannot help improve the project performance. As the job is not directly contracted to the bottom-layer subcontractor, he/she will have less sense of responsibilities to client, and abortive and remedial work, claims and disputes can easily occurs. 3.4. Impacts of multi-layer chain subcontracting to project communication and coordination Communication among project parties is critical to the success of implementing a construction project. However, communication is poor when the number of subcontracting layers increases as misinterpretation happens during information transfer between subcontracting layers (Tam and Tong, 1996). The main problems in project communication and coordination include: CC1 delay in communicating decision to bottom-layers due to long chain; CC2 increasing communication errors when increasing layers of subcontractors; CC3 poor communication channel between main contractor and subcontractors; CC4 lack of communications between multi-layers; CC5 dicult in sharing timely information among multi-layer supply chain subcontractors; and CC6 lack of main contractors medication on disputes among subcontractors.

Information transfer carries the possibility of communication errors. When main contractor conveys decision information to the bottom-layer subcontractors, communication errors often occur as middle-layer subcontractors misunderstand the information and/or miss pertinent details. Thus the information becomes ambiguous when it approaches to the bottom-layer subcontractors. The survey results in Table 1 suggest that the factors CC2 increasing communication errors when increasing layers of subcontractors and CC4 lack of communications between multi-layers are important, receiving the value of RIIs with 0.798 and 0.762 respectively. An interview discussion reveals a case where a main contractor provided comprehensive drawings and written instructions to the secondlayer subcontractor. When the information approached to the bottom-layers, detailed instructions became brief notes and simple verbal instructions. In general, the main contractor often relies on middle-layer subcontractors for passing information and coordinate work, thus the bottom-layer subcontractors will not be able to receive correct instructions if middle-layer subcontractors misinterpret the information. Furthermore, the use of various languages in the local construction industry, typically English and Chinese, is also considered contributable to the miscommunication along subcontracting chains. The interpretation among dierent languages carries the possibility of misunderstanding. The factor CC3 poor communication channel between main contractor and subcontractors is also important, receiving the value of RII with 0.69. In general, there are ineective communication and coordination channels between a main contractor and its multi-layer supply chain subcontractors in the local construction industry. As a main contractor does not have contractual relationship with other subcontractors apart from its immediate subcontractors, the main contractor normally does not communicate and exchange with further layers of subcontractors. This is considered the major reason that a main contractor usually cannot detect the disputes happening amongst other multi-layer supply chain subcontractors. On the other hand, the weak communication results in those subcontractors at dierent layers are unable to timely share project information; therefore they will not be able to respond to any new changes or instructions in a consistent way. This can also explain why CC5 dicult in sharing timely information among multi-layer supply chain subcontractors; and CC6 lack of main contractors medication on disputes among subcontractors, with both being identied as major factors. CC1 delay in communicating decision to bottom-layers due to long chain is another signicant factor receiving the value of RII with 0.712. Typical consequences of poor communication and coordination among various layers of contractors are delaying project, overrunning cost and degrading quality. This problem is said particularly true when project comes to the rectication period. Because of the poor communication amongst various layers, contrac-

Author's personal copy

114

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

tors at bottom-layer are given the opportunity of delaying to address defects or even hide defects. As a result, the main contractor cannot promptly respond to the defects as their subcontractors often disappear as soon as payments are settled. In many cases, even if the subcontractors committing poor quality work are caught, they cannot aord for the high repair cost. It can be highlighted from the above discussion that lack of communication and communication errors are obvious from increasing layers of subcontractors. As the number of subcontracting layer is high, it will take longer time for transferring and sharing information, which often aected in transferring timely and important information to the subcontractor who actually does the work. Claims and disputes can thus be happened. 4. Improvement methods The above analysis demonstrates that there are many problems in Hong Kong local construction practice in practicing the multi-layer supply chain subcontracting system. Based on the discussions with the professionals during an empirical survey, four improvement methods are recommended. 4.1. Changing the practice of the lowest bid to an approach which incorporate both price and technical performance In the current lowest bid practice in the local construction industry, a main contractor usually invites subcontractors to submit a tender including price and work duration. The main contractor often oers a subcontract to the bidder who submits the lowest price for maximizing prots (Low and Tan, 1996; Kam and Tang, 1998). However, this practice leaves subcontractors particularly the bottomlayer subcontractors with very low prot margins, and subcontractors usually nd lack of motivation. This is identied as the main reason contributing to the cut corner practice in the local construction industry. It is proposed that both price and technical performance should be considered in awarding subcontracts. In fact, the Works Bureau of the Hong Kong Government has introduced two mechanisms in tender evaluation for various public work contracts for selecting main contractors. These two methods are called the Marking Scheme eective from June 2002 and the Formula Approach eective from November 2002 (Works Bureau, 2002a,b). The new approaches evaluate a contractors tender by collectively scoring its tender price and performance attributes, with the latter including contractor experience, past performance, technical resources and technical content of the contractors proposal. In applying these methods, the tender with the highest combined score (combined price and performance score) will be normally recommended for acceptance. It is suggested that the application of these principles should be extended to subcontract arrangements.

Another alternative approach for mitigating the negative consequence of the existing lowest bid practice is to adopt a benchmark cost which can be formulated by independent consultants. By using the benchmark, only those tender prices falling within a specied range to the benchmark are considered as valid bids. Benchmark bidding approach has been eectively used in China mainland construction (Shen and Song, 1998). One important principle in using the approach is to ensure that benchmark cost is kept in strict condentiality thus the principle of fairness among bidders can be uphold. In fact, the principle of benchmarking cost can be extended to benchmark construction time. In the existing practice in the Hong Kong construction, unrealistic construction time is often set up in subcontracting arrangement. A tight construction schedule will only impose time pressure to bottom-layer subcontractors, and this has been identied in this survey study as one of the reasons contributing poor construction performance. The main contractor can benchmark the time for completing a specic subcontract by considering the normal practice in the local industry. The rule can be made so that only these tenders in which construction time is within a specied range to the benchmark time are valid. 4.2. Limit the number of subcontracting layers Disadvantages of allowing increasing number of layers in subcontracting chains are multiple, which have been addressed in the previous sections. If the number of the layers is few, for example, two to three, communication and coordination would be easier and more ecient. When there is a concern about progress or requires rectication work, the concerned subcontractors can easily be identied, thus the problems can be solved within limited time frame. Fewer subcontracting layers also allow a higher possibility for the bottom-layer subcontractors to share a healthy prot margin as the results of that the possible extra overheads for the middle-layer subcontractors are reduced. Therefore, to limit the number of subcontracting layers can mitigate the temptation of cutting corners in the bottom-layer subcontractors operations. 4.3. Restrain the use of supply-and-x subcontracting arrangement Experience gained in other area such as China mainland is that main contractors are requested to employ laboronly subcontractors and main contractors provide all building materials (Chen et al., 2008; Yung and Yip, 2010). This approach allocates main contractors autonomy and responsibilities of controlling the quality of building materials. Therefore, when non-conformance materials are detected, the main contractors can easily take measures for correcting the defects.

Author's personal copy

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116

115

4.4. Enforce the implementation of governmental regulations Subcontractors particularly those specialist contractors are expected to register through a screening process. Requirements on subcontractors capability and creditability should be imposed, such as past nancial records, years of relevant experience, past job performance reference, number of skilled workers employed and training schemes for new employees. These aspects should carefully be assessed in implementing the screening process for considering subcontractors registration. The screening process should be authoritative, by which representatives from government departments, professional institutions and construction practitioners should be included on the board of registration for monitoring the screening process. Proper screening on subcontractors at registration stage can ensure better quality of subcontractors in the whole industry, thus improvement of workmanship can be throughout gained for the industry. 5. Conclusion Poor construction performance has been commented in many countries and regions. Whilst there are many factors contributing to this traditional phenomenon, the improper practice of multi-layer subcontracting system is one of the major contributors. The data collected from Hong Kong construction industry in this study demonstrated that curbing malpractice within the multi-layer chain subcontracting system in the local construction industry is in pressing needs to restore industry reputation and public condence. The application of multi-layer chain subcontracting system is common in the Hong Kong construction, and yet the application is largely awed, resulting in signicant poor project performance. The study did not suggest that the subcontracting approach itself in construction is improper, instead it revealed the importance of the system to the industry, and also highlighted the fact of the practice under improper control. Major reasons or factors were identied for ineectively use of the multi-layer supply chain subcontracting system for poor quality, time management, cost control, and communication and coordination performance. Improper work practice from subcontractors contributes to poor quality performance. Long communication chain for increasing layers of subcontractors contributes to poor time management performance, increased abortive and remedial work from subcontractors to cost overruns, and communication errors for increasing layers of subcontractors to the consequence of poor communication and coordination. Improvement measures practicing subcontracting system included changing the practice of the lowest bid to an approach which incorporate both price and technical performance; limiting the number of subcontracting layers; restraining the use of supply-and-x subcontracting arrangements; and enforcing the implementation of governmental regulations. The research results provided useful

reference for comparing subcontracting practice from different construction industries thus improving the eectiveness of the practice in construction industry. Appendix A. Interview questions

(1) Do you agree with the questionnaire survey results? (2) Do you have any examples support the questionnaire survey results? (3) Do you have any examples that are against the questionnaire survey results? (4) What is your experience in the ineectiveness of the multi-layer chain subcontracting? (5) Which factor you personally think that is the major factor aecting poor project management performance from the multi-layer supply chain subcontracting? References
Bayraktar, E., Demirbag, M., Koh, S.C.L., Tatoglu, E., Zaim, H., 2009. A causal analysis of the impact of information systems and supply chain management practices on operational performance: evidence from manufacturing SMEs in Turkey. International Journal of Production Economics 122 (1), 133149. Chan, D.W.M., Kumaraswamy, M.M., 1997. A comparative study of causes of time overruns in Hong Kong construction projects. International Journal of Project Management 15 (1), 5563. Chen, P., Partington, D., Wang, J.N., 2008. Conceptual determinants of construction project management competence: a Chinese perspective. International Journal of Project Management 26 (6), 655664. Chiang, Y.H., 2009. Subcontracting and its ramications: a survey of the building industry in Hong Kong. International Journal of Project Management 27 (1), 8088. Construction Industry Review Committee, 2001. Construct for Excellence: Report of the Construction Industry Review Committee. Hong Kong Government. Elazouni, A., Metwally, F., 2000. D-Sub: decision support system for subcontracting construction works. Journal of Construction Engineering and Management 126 (3), 191200. Hinze, J., Tracey, A., 1994. The contractorsubcontractor relationship: the subcontractors point of view. Journal of Construction Engineering and Management 120 (2), 274287. Kam, C.W., Tang, S.L., 1998. ISO 9000 for building and civil engineering. Journal of Hong Kong Institution of Engineers 5 (2), 610. Kong, S.Y. 2001. Impacts of multi-layered sub-contracting system on quality in Hong Kong construction industry. The Hong Kong Polytechnic University, Hong Kong Special Administrative Region. Low, S.P., Sua, C.S., 2000. The maintenance of construction safety: riding on ISO 9000 quality management systems. Journal of Quality in Maintenance Engineering 6 (1), 2844. Low, S.P., Tan, W., 1996. Public policies for managing construction quality: the grand strategy in Singapore. Construction Management and Economics 14 (4), 295309. Ng, S.T., Tang, Z., Palaneeswaran, E., 2009. Factors contributing to the success of equipment-intensive subcontractors in construction. International Journal of Project Management 27 (7), 736744. Ofori, G., Debrah, Y.A., 1998. Flexible management of workers: review of employment practices in the construction industry in Singapore. Construction Management and Economics 16 (4), 397408. Shen, L.Y., Song, W.G., 1998. Competitive tendering practice in Chinese construction. Journal of Construction Engineering and Management 124 (2), 155161.

Author's personal copy

116

V.W.Y. Tam et al. / International Journal of Project Management 29 (2011) 108116 Works Bureau, 2002a. A formula approach to tender evaluation for works contracts. Works Bureau Technical Circular No. 23/2002, Ref. WB(W) 272/31/19, Works Bureau. Works Bureau, 2002b. Marking scheme in tender evaluation for works contracts. Works Bureau Technical Circular No. 22/2002, Ref. WB(W) 272/31/19, Works Bureau. Yik, F.W.H., Lai, J.H.K., 2008. Multilayer subcontracting of specialist works in buildings in Hong Kong. International Journal of Project Management 26 (4), 399407. Yung, P., Yip, B., 2010. Construction quality in China during transition: a review of literature and empirical examination. International Journal of Project Management 28 (1), 7991.

Soroor, J., Tarokh, M.J., Shemshadi, A., 2009. Initiating a state of the art system for real-time supply chain coordination. European Journal of Operational Research 196 (2), 635650. Tam, C.M., Deng, Z.M., Zeng, S.X., Ho, C.S., 2000. Quest for continuous quality improvement for public housing construction in Hong Kong. Journal of Construction Management and Economics 18 (4), 437446. Tam, C.M., Tong, T.K.L., 1996. A quality management system in Hong Kong: a lesson for all people in the building industry. The Australian Institute of Building Papers 7, 121131. Wong, W.S., Chan, H.W., 1997. Professional Practice for Architects in Hong Kong. Pace Publishing Limited.