International Journal of Project Management 21 (2003) 419424 www.elsevier.

com/locate/ijproman

Construction contracts: the cost of mistrust

Ramy Zaghloul*, Francis Hartman
Project Management Specialization, Department of Civil Engineering, University of Calgary, 2500 University Dr. NW, Calgary, Alberta, Canada, T2N 1N4 Received 12 February 2002; received in revised form 20 September 2002; accepted 19 November 2002

Abstract Current contractual relationships are mainly based on confrontational situations that reect the level of trust (or mistrust) in the contract documents. This can be the driver to increase the total cost of a specic project and aect the overall relationship between the contracting parties. This has been tested in the construction industry in Canada, and appears to be generalizable across North America. Based on two independent surveys (including the one presented in this paper) of Owners, Consultants and Contractors across Canada, the assessed premium associated with the ve most commonly used exculpatory clauses in construction is between 8 and 20% in a sellers market. It should be obvious that trust and contracting methods are related and that this relationship is of vital importance to eective project management and contract administration. To date, little work has been done to explore the advantages of this relationship. This paper presents some of the results of a survey conducted across the Canadian construction industry that identies some opportunities for better risk allocation mechanism and contracting strategies that are based on a trust relationship between the contracting parties. These opportunities are based on a trust relationship that can be the root cause for a signicant saving in the annual bill for construction. # 2003 Elsevier Ltd and IPMA. All rights reserved.
Keywords: Risk allocation; Trust relationships; Transaction costs; Contracting strategies

1. Introduction The construction industry in both Canada and the United States is the single largest non-governmental employer. In 1997, the industry was estimated in Canada to have a value of about $90 billion, representing 15% of the gross domestic product and employing more than 872,386 workers directly [1]. However, within the last 20 years considerable cost wastage has been identied by the Construction Industry Institute [2]. A signicant portion of this cost wastage may be attributed to inappropriate risk allocation in contracts, as cited in various examples analysing risk allocation in the construction industry and the underlying causes of disputes conducted in Canada and the United States [3,4,30]. Meanwhile, construction risks are a major element that can signicantly aects the nal cost of any project.
* Corresponding author. Tel.: +1-403-220-6185; fax: +1-403-2827026. E-mail addresses: rszmoham@ucalgary.ca (R. Zaghloul), fhartman@ucalgary.ca (F. Hartman). 0263-7863/03/$30.00 # 2003 Elsevier Ltd and IPMA. All rights reserved. doi:10.1016/S0263-7863(02)00082-0

Specically, how these risks are allocated has a direct bearing on the nal total cost. The cost of these risks is carried by the owner, contractor, or both, as determined by the type of the construction contract [5]. Risk allocation always occurs in any situation where more than one party (owner, contractor, consultant, etc.) is responsible for the execution of a project. Making sure that as many risks as possible are recognised and managed is good practice in any project. This activity is an important step in that this allocation can signicantly inuence the behaviour of the project participants and hence impact both project performance and nal cost.

2. Construction contracts and risk allocation Construction contracts are the written agreements signed by the contracting parties (mainly an owner and a contractor), which bind them, dening relationships and obligations [6]. In any certain project, the owners goal can best be achieved by selecting the contract type that will most eectively motivate the contractor to the desired end. This step is also dependent on completeness of information for

420

R. Zaghloul, F. Hartman / International Journal of Project Management 21 (2003) 419424

the bidder(s) at tender time and the extent that the owner wishes to take specic risk. In this context, contract risk can be divided into performance risks and cost risks [5]. Regarding risk allocation, the concept of limitation of liability dates back more than three hundred years, when the British Parliament declared, as part of Maritime Law, that a ships owner should not bear greater liability than the value of the ships hull [7]. In this context, every contract allocates risk. Not all contracts allocate risk equitably or such that the power and authority to manage the risk is allocated along with the risk itself. Given the opportunity, an owner should favour ecient allocation of risk between parties to a project that simultaneously reduces risk and improves project performance. This appropriate risk allocation is a signicant contributor to low transaction cost of any specic project and vital issue in the success of the contracting process. However, in an ownercontractor relationship at least, a common aim of owners appears to be to avoid risk as far as possible by allocating as many risks as it can to the contractor [8].

3. Risk allocation through disclaimer clauses Any construction project involves risk and there is no possibility to eliminate all the risks associated with a specic project. All that can be done is to regulate the risk allocated to dierent parties and then to properly manage the risk. This can be done through the language of the construction contract. The decisions regarding risk sharing or risk shifting are made within the context of an owners contracting policy [9]. One of the objectives of the contract is to serve as a framework between the parties to establish which one has assumed which risk. One way in which the contracting parties attempt to address the right and responsibilities for risk is through dealing directly with the issue of legal liability by including provisions that exclude liability arising from certain causes. Disclaimer (exculpatory) clauses are examples of such provisions. Such clauses attempt to transfer one partys risk (which may be a legal liability) to another by contractual terms [5]. In other words, these clauses are intended to exclude an owners liability in contract and often in tort for cost incurred by a contractor [10]. Using disclaimer clauses to allocate risk has been identied by recent studies and industrypractice as a main reason to increase the overall cost of a project [2,4,11,13,14,31]. When a risk is shifted to the contractor and the contractor has no means by which to control the occurrence or outcome of the risk, the contractor must either insure against it or add a contingency to the bid price [16]. Two recent studies (including the one presented in this paper) indicate that using disclaimer clauses in Canadian contracts carries a

premium of between 8 and 20%, depending on whether business conditions were favourable (low need for work, low technical complexity, fair contract administration, negotiated and suitable contract type, and complete design work) or adverse (high need for work, high technical complexity, unfair contract administration, un negotiated and un suitable contract type, and un complete design work) [11,12,17,18]. Contractors add these risk premiums to each disclaimer clause in their contract with the owner based on some main criteria. The most important of these criteria are their business relationship with the owner, work conditions, and contract type and fairness. On multimillion-dollar projects, such an increase to the project cost can obviously be very signicant. An additional but less visible, cost of shifting risk to the contractor through disclaimer clauses presents a number of hidden costs including restricted bid competition, increased potential for claims and disputes and above all, more adversial ownercontractor relationships [1921]. Exhibit 1 illustrates the general outcomes for using exculpatory clauses in contracts. There is a signicant amount of literature that discusses possible options for improved contracting methods and better risk allocation processes. These studies present new ways for doing business such as partnering/ alliances, risk sharing/reward systems, incentive-based contracts and others. However, these new contracting strategies between owners and contractors are still founded largely on the self-interest of each participant. Motivation for performance under these strategies has focused largely on retribution for non-performance. In most cases, the relationships of the contracting parties are still defensive and, in some cases, adversarial. The atmosphere spawned by these relationships has not been conducive to innovation or cooperation between parties [5,22]. With all of these solutions, owners are generally unwilling to carry project risks and where possible, transfer them contractually to contractors through dis-

Exhibit 1. General outcomes of risk allocation through disclaimer clauses.

R. Zaghloul, F. Hartman / International Journal of Project Management 21 (2003) 419424

421

claimer clauses in order to have a control system over these risks in what we call contract documents [17,23,24].

situation, or an artefact. It simply answers the question Does this relationship feel right? In this paper, the authors present some of the ndings of a study conducted across the Canadian construction industry including owners, contractors, and consultants and appear to be generalizable to the United States construction industry. These ndings identify the relationship between trust and risk allocation practices in construction contracts and how a strong trust relationship through the Colours of Trust Model can reduce the nal cost of any specic project by improving the risk allocation method between the contracting parties.

4. Why trust With the absence of trust in business relationships, there is a signicant need for good and powerful control system to manage and administrate the contracting process. However, even with the existence of this powerful control system (the contract documents), with the absence of trust, the success of any project or business relationship is always questionable. Trust should be at the heart of how people do and think about risks. Meanwhile, these risks vary as the form of a relationship between the contracting parties varies. Risks that the trusting contracting parties assume and the mechanisms for mitigating those risks emerge as a function of the form of the relationship between those parties. Trust inuences almost every aspect of the project management process, which aect the execution, and the cost of any specic project. However, The concept of trust is very complex and multidimensional and there has been much debate within academic circles regarding a common denition or model [2527]. In an attempt to advance the conceptual understanding of the topic of trust, Hartman [5,27] developed a model of trust that enables a more simplied understanding of the concept. This model has been adapted for the purpose of this research.

6. The industry survey In order to collect the data for this study, an industry survey was conducted in the Canadian and the United States construction industries. A self-administered mailout was considered as the data-gathering method in order to overlap the problems of bias, cost, and geographic barriers [28]. The questionnaire included openended questions for identifying the respondent and identifying suitable information for the research, and scales, assumption with yes/no answers, and multiplechoice questions to identify respondents professional judgment and experience on certain aspects concerning risk allocation practice in the industry. The survey also solicited qualitative and quantitative information on individuals perception of the most common disclaimer clauses, the premiums associated with these clauses, and the rationale behind this behaviour of allocating risk in the North American contracts identied in recent research and studies. These clauses include:      Uncertainty of work conditions Delaying events Indemnication Liquidated damages Suciency of contract documents

5. The colour of trust model The colour of trust model species three colours (or types) of trust: Blue, Yellow, and Red trust. These types are identied as follows:  Blue (or competence) trust is all about ability and competence, which is based on the perception of the others capacity to perform what is required. It is simply the answer for the question Can you do the job?.  Yellow (or Integrity) trust is based on integrity, which is founded upon the perception of the others attitude to act ethically, to adhere to values that we hold important, and to be motivated to not take advantage of the other party. It is simply answer the question Will you consistently take care of my interests?.  Red (or intuitive) trust is based on intuition, which is the result of a combination of emotional response and rapid processing of information and may be described as the instincts or gut feelings that one person has about the other, a

The results of the study are based on more than 300 respondents to the survey. The study sample includes owners, contractors, and consultants from both private and public industry sectors, working in dierent types of projects; civil, industrial, commercial, residential, and others such as oil and gas pipelines.

7. Results and discussion Based on the survey results, it is noted that most of the contracts used in the construction industry are prepared

422

R. Zaghloul, F. Hartman / International Journal of Project Management 21 (2003) 419424

Table 1 Percentage distribution of contract type; negotiated vs. prepared Contract type Negotiated Prepared Other Percentage 25.2 74.3 0.20

contracts that have been written by one of the contracting parties, normally the owner. In other words, more than 74% of construction contracts are prepared by the owner and not negotiated. Table 1 indicates these results. The study shows that these contracts, whether negotiated or prepared, typically include one or more of the ve disclaimer clauses mentioned before. Actually these clauses exist in more than 75% of the survey respondents contracts. Table 2 illustrates the frequency of disclaimer clauses usage in construction contracts. The study reports that under all circumstances, with the existence of disclaimer clauses, contractors attached risk premiums to the total cost of a project in order to protect themselves against these clauses. To a great extent, the ndings validate the work of Hartman and Khan reported by Hartman [12]. These premiums fall into a range of 820% of the total cost of a project based on whether the business situation is favourable (low need for work, low technical complexity, fair contract administration, negotiated and suitable contract type, and complete design work) or adverse (high need for work, high technical complexity, unfair contract administration, un negotiated and un suitable contract type, and un complete design work). What interesting is that these clauses continue to be used in some of the newer contractual agreements between owners and contractors such as partnering/ alliances.

8. Trust, risk behaviour, and cost reduction In the construction industry, risk behaviour has been grouped into three general classications: risk averse, risk neutral, and risk taker [15]. Each of these behaviours will aect how an individual or organization would react to a specic contractual agreement. The
Table 2 Frequency of disclaimer clauses usage in construction contracts Disclaimer clause types Uncertainty of work conditions Delaying events Indemnication Liquidated damages Suciency of contract documents Percentage 92.0 72.0 77.0 61.0 67.0

results of this research report that in most cases, industry owners are risk averse or risk neutral based on project complexity and size. Industry contractors however, are typically risk takers, neutral or risk averse based on project complexity, size, and market conditions. Results form this study show that the amount of the premium is based on the contractors perception of the disclaimer clause risk. Which means that if the contractors perception of the disclaimer clause risk is high, the premium will be large. One of the most important nding of this study is that a signicant relationship exists between the amount of the premiums associated with the disclaimer clauses and the level of trust between the contracting parties, specically between owners and contractors. In other words, under a high trust relationship with owners, contractors tend to lower the risk premiums associated with disclaimer clauses because their perception of the disclaimer clauses risk is low. Meanwhile, under low trust relationship with owners, contractors tend to associate what we can call a normal level of risk premiums (820%) with disclaimer clauses risks because their perception of the risk is high. The perception of disclaimer clauses risk under low trust relationships is very high (average of 4.4 out of 5 point scale). The perception of disclaimer clauses risk under high trust relationships is very low (average of 2.3 out of 5 point scale). Table 3 shows these averages (based on 5 point scale). Another interesting nding of this study is that the trust level that generally exists in the construction industry contracts between contracting parties is low (average 2.3 out of 5 point scale), which reects the level of mistrust in the industry contracting practice. As the results report, owners and contractors risk allocation contracting practice is mainly a function of their trust (or mistrust) relationship between each other. If the ownercontractor contract is based on a strong trust relationship, the amount of the premiums associated with disclaimer clauses is very low, or even better; the disclaimer clauses would not exist on the contract from the outset. The results also show that owners and contractors are willing to change their risk allocation practice regarding
Table 3 Perception of disclaimer clauses risks under low and high trust relationships (averages out of 5 point scale) Disclaimer clause Uncertainty of work conditions Delaying events Indemnication Liquidated damages Suciency of contract documents Low trust relationship 4.5 4.5 4.4 4.3 4.5 High trust relationship 2.4 2.4 2.2 2.1 2.1

R. Zaghloul, F. Hartman / International Journal of Project Management 21 (2003) 419424

423

these clauses (use a dierent mechanism for risk allocation) when contracting parties have previous working experience with each other, which can be related to trust relationship. As a result, contracting parties are willing to have another method of risk allocation when:  A contracting party has evidence that other party will protect his interests;  A contracting party has an industry or practical evidence that other party is knowledgeable enough to manage the risk; and  A contracting party has a good industry reputation. It is the authors belief that most of these criteria can be met through a trust-based relationship between the contracting parties. Such a relationship can be built at the front-end of a project and can be a signicant contributor to cost reduction. To a great extent, these ndings are consistent with the CII CostTrust relationship [29]. The participants report that an ownercontractor trust-based relationship enables them to command structure and authority systems, incentive systems, administrated pricing systems (cost, qualities, and prices), good communication, and team working environment in order to reduce the total cost of a project. In other words, owners and contractors under a trustbased relationship are likely try to manage and mitigate risk primarily for their own benet and not to the disadvantage of the other party. The converse is more typical in most of todays contractual relationship in the construction industry.

 more time and eort in the front-end of a project and sucient experience to manage or mitigate the risks and administrate the contract;  a negotiation phase prior to the start of the contract should exist, this phase is required to built a trust relationship between the contracting parties, then this negotiation phase can be part of the contract itself; and  adequate risk-sharing or risk-reward system should exist to share the benets if the risk does not occur during the project lifecycle. The rationale for better risk allocation between owners and contractors ought to be based on meeting these conditions as far as possible. Missing one of these criteria is very likely to trigger inappropriate risk allocation process for any given project and hence bring additional cost for the contracting parties.

References
[1] Statistic Canada. Construction in Canada. Cat. No. 64-201, 1997, 2001. [2] Construction Industry Institute (CII). Impact of various construction contract types and clauses on project performance. Publication # 5-1, 1986. pp. 114. [3] ACE & AGC. Owners guide to saving money by risk allocation. Washington: American Consulting Engineers Council and Associated General Contractors of America; 1991. [4] Construction Industry Institute (CII), Impact of risk allocation and equity in construction contracts. Publication SD-44, 1988. pp. 17. [5] Hartman F. Dont park your brain outsidea practical guide to improving shareholder value with SMART management. NC: PMI Publications; 2000. [6] OReilly M. Civil engineering construction contracts. London, England: Thomas Telford Publication; 1996. [7] Zoino W. Cautious risk taking. Civil Engineering, October 1989; 59(10):658. [8] Gransberg D, Ellicot M. Best-value contracting criteria. Cost Engineering, Morgantown, June 1997;39(6):314. [9] Kozek J, Hebberd C. Contracts: share the risk. Journal of Construction Engineering and Management 1998;111(2):35661. [10] Goldsmith I, Heintzman T. Goldsmith on Canadian building contracts. 4th ed. Canada: Carswell, Ontario; 1995. [11] Khan Z. Risk premiums associated with exculpatory clauses. Masters thesis, University of Calgary, Calgary, Alberta, Canada; 1998. [12] Hartman F. The real cost of weasel clauses in your contracts. Proceeding of the 29th Annual Project Management Institute Seminars and Symposium, 915 October 1998. [13] Jergeas G, Hartman F. A contract clause for allocating risk. Transaction of the American Association of Cost Engineers, 1996. [14] Zack J. Risk sharinggood concept, bad name. Cost Engineering, Transaction of the American Association of Cost Engineers, July 1996;38(7):2631. [15] Erikson C, OConnor M, Boyer L. Construction process risk allocation. Transaction of the American Association of Cost Engineers, July 1978:912.

9. Conclusion and recommendations Recent research and industry experts have indicated that inappropriate risk allocation through disclaimer clauses in contracts is a signicant reason for increasing the total cost of a project. Any improvement in the process would result in signicant savings for the construction industry. This study indicates that using disclaimer clauses to shift project risks to the other contracting party is still the general traditional practice in the construction industry. It also indicates that there is a signicant relationship between trust and risk allocation through disclaimer clauses that can result in cost saving in the construction industry. The survey respondents report that to reach a better risk allocation process, a trust relationship between the contracting parties should exist rst. This can be done through certain stages as follow:  a clear understanding of the risks being born by each party and who owns or can manage the risk;

424

R. Zaghloul, F. Hartman / International Journal of Project Management 21 (2003) 419424 [24] Zaghloul R, Hartman F. How to reduce your project cost. AACE (American Association for Cost Engineers) Annual Conference, Calgary, Alberta, Canada; 1999. [25] Hosmer LT. Trust: the connecting link between organizational theory and philosophical ethics. Academy of Management Review 1995;20(2):125. [26] Mayer R, Davis C, Schoorman F. An integrative model of organizational trust. Academy of Management Review 1995;20:70934. [27] Hartman F. The role of trust in project management. Proceeding of the PMI research Conference 1999. [28] Morse A, Katton G. Survey method in social investigation. UK: Heinemann Education; 1971. [29] Construction Industry Institute (CII), Costtrust relationship, contracting phase II. Task Force. Publication 24-1.1993. pp. 3 10. [30] Enger T. Beyond NORSOK and CRINE. Transaction of the Annual EPCI Conference, Stanvanger, Norway, 1113 June, 1997. [31] Hartman F. Construction dispute resolution through an improved contracting process in the Canadian context. PhD. thesis. Loughborough University of Technology, UK; 1993.

[16] Jergeas G, Hartman F. Contractors protection against construction claims. American Association of Cost Engineers, AACE Transactions, 1994. [17] Zaghloul R. Contracts hidden costs: can it be saved? AACE (American Association for Cost Engineers), Unpublished, won the AACE Graduate Students Papers Scholarship; 1998. [18] Zaghloul R. Contracts hidden costs: a trust/risk allocation approach. Unpublished PhD dissertation, Project Management Specialization, University of Calgary, Canada; 2001. [19] Robert C. Managing change orders and claims. Journal of Management in Engineering 1997;2(1):2731. [20] Becker D. The cost of general conditions. Transaction of the American Association of Cost Engineers, September 1993:710. [21] Tenah K. The design-build approach: an overview. Cost Engineering 2000;42(3):317. [22] Drexler J, Larson E. Partnering: why project ownercontractor relationship change. Journal of Construction Engineering and Management, July/August 2000:2937. [23] Ward S, Chapman C, Curtis B. On the allocation of risk in construction projects. International Journal of Project Management 1991;9(3):1407.