Political Influence over Bureaucracy
Agency theory has no doubt been extensively debated. In this study, this theory is extended to investigate the under-researched topic of project participant relationships in the construction sector.
Existing studies imply that there are no agreed or exhaustive list of factors that can fully explain the handling of construction projects, project participant relationship, and accountability issues of the public sector. Different researchers have emphasized on different sets of risk factors. For instance, Aritua et al. (2011) mention that public sector projects are exposed to risks associated with changes in government policy, diverse stakeholder aspirations, and challenges of multiple project procurement. Al-Sedairy (1994) found that project conflicts stem from timing, project concepts, specification issues, as well as costs, claiming that conflicts not only occur at later stages but remain even in the postconstruction period. Poor communication can be another most common project risk (Ceric 2003). In addition, political interference can affect project outcomes where accountability reports may be just reduced to bureaucratic communications and some insignificant stories to avoid public criticism (Christensen and Skærbæk 2007). It may be challenging to hold public officials accountable where monitors themselves are public servants (Cabral and Lazzarini 2015). What is still under debate is whether political pressure benefits or harms the public agency performance and outcomes (Rojas 2015). Project participant relationship is no doubt complex. Despite its rapid growth, this subject matter has received very little critical attention (Hodgson 2002).
There is little doubt that the agency theory has been extensively debated. This study, however, extends the theory to the under-researched subject matter of project participant relationships in the construction sector. The research is empirically based on a developing South Pacific country, Fiji.
Literature confirms the extensive application of the agency or principal-agent theory in construction, covering risks of adverse selection, moral hazard, and hold-up (Ceric 2013). Interestingly, there is no mention of the relationship between project managers in construction projects (Ceric 2013). As per Ceric (2013: 773), exploring the more complex project participant relationships deserve greater attention. This study is an attempt in this direction.
This study extends the agency theory by drawing factors from this theory and construction, public sector, project management, and accountability literature to include all project participants and risk factors. Ceric’s (2013: 773) work is the first found in literature to have extended the agency theory to construction including other participants like project managers and contractors. To a lesser extent, another study by Hadi and McBride’s (2000), also mentioned the role of the agent as more of a subcontractor as they applied the theory to study the role of marketing agents in the public sector information trading market.
The concept of principal and agent can be confusing in the contractual situations of public administration given involvement of multiple parties (Farazmand 2002). This research is thus a buildup of the pioneering work of Ceric (2013) in the construction sector that separates the role of such multiple actors. Like Ceric’s (2013) work, this entry describes participants as agent, principal, contractor, and project manager to avoid the confusion of multiple agents/principals.
Literature has well established the relevance of the agency theory since it has been applied to a variety of settings and has been used by scholars of various disciplines (Eisenhardt 1989). The key actors in the agency theory are the principal and the agent. The principal pays the agent to carry out an activity. Interestingly, evidence also suggests reflexive principal-agent relationship, where an agent may enact the role of the principal (Wickramasinghe and Lamb 2002). Central to this theory is the assumption that an agent may optimize its profits at the expense of the principal (agent opportunism). Such a situation is possible because the agents do have more knowledge or information than the principal – they perform the activity with a certain degree of empowerment (Hadi and McBride 2000). However, in construction projects, the principal may also exhibit opportunism (Ceric 2013). Given self-interest, agents do not readily share information with the principal. It is this hidden information/hidden action that generates moral hazard risk (Ceric 2013). The principal will thus try to transfer its risk to the agent by either investing in monitoring systems to obtain enough information about the agency activities or offering incentives to motivate the agent to efficiently perform its interests (Hadi and McBride 2000). Studies such as that of Kauppi and Raaij (2015) argue that agency problems may not only arise with agents withholding information from principals but principals may also hold information from agents.
Ceric’s (2013) research was based on an exploratory survey and Delphi methods. This study differs as it employs a qualitative institutional case-based approach, employing convenience and snowballing nonrandom sampling techniques. Individuals who were easiest to reach out to were interviewed first (convenience) and they recommended others who could be interviewed (snowballing).
Data was gathered through in-depth face-to-face interviews and relevant document search. The enterprise annual reports, cabinet decisions, legislation and parliamentary reports, and reputable magazines were examined to formulate questions and to cross-check interviewee responses. The secondary sources were journal articles, textbooks, conference proceedings, theses and dissertations, newspapers, and the internet. In collecting data, information was gathered from many sources, enabling triangulation among different data sources (Eisenhardt 1989).
Overall 22 in-depth interviews were conducted only with those who could provide relevant information, those who were part of the project, and those who were still employed by GSPS at the time of research. Interviewee names and descriptions are withheld given political and job security reasons. Anonymity encouraged provision of substantial information without the fear of being questioned later. During interviews, the researcher used probes to gather more elaborate information. After the interviews, the researcher rephrased the interviewee responses in a summarized manner to correct inaccuracies and to ensure that the receiving, understanding, and interpretation of information were the same as expressed by the interviewees (Wilkinson and Birmingham 2003).
The collected data was analyzed on the basis of findings from the case study in forms of interviewee responses and relevant documents. This research used the logic of analytic rather than enumerative induction. The research used Gillham’s (2001) transcription and analytical framework for the recording, verification, and analysis of data as follows. Substantive statements in each interview note and extra details (furnished during the interviews) were highlighted while repetitions, digressions, and irrelevant materials were put aside. Similar statements made by interviewees were noted as similarities and dissimilarities were marked. Following this, the researcher went through the collected primary documents to highlight noteworthy information. The researcher then went back to the entire interview notes and documents to note highlighted statements and categorized these as main events. The researcher first created major headings (preconstruction, construction, and postconstruction stage) to note the main events. For validity and reliability reasons, statements were cross-referenced between interviewee responses’ and with documents such as the company and ministry documents, published interviews in reputable magazines, and newspaper articles. Dissimilar statements were marked as queries and after an interval of a few days were clarified through quick repeat interviews, emails, or telephone inquiries. The chronological listing was revised after sorting out queries.
This research is premised on a single-case study; as such, findings have limited comparability and generalizability (Blenker et al. 2014). Nonetheless, it can offer insights for later comparative studies of construction public enterprises in other countries, particularly developing countries. While the empirical study is premised on one small developing Pacific Island economy, the study does have potentially wider application for construction public enterprises elsewhere.
On 1 January 1969, the boatshed section and slipways were transferred from the Public Works Department (PWD) to the Marine Department, which then became Government Shipyard and Public Slipways (GSPS) (Marine Department 1969). GSPS is now a subsidiary of another public enterprise – Fiji Ports Corporation Limited (FPCL).
The most controversial of all projects ever handled by GSPS was the “Reef Endeavour” shipbuilding project.
Reef Endeavour: The Shipbuilding Project
Government did initiate a legal action for SDMA’s misrepresentation and fraud but dropped it later because of SDMA’s liquidation. $F5.5 million was written off in 1991.
The Fiji Government, despite its higher cost estimates signed the initial contract for $A7.7 million upon reassurances by SDMA. Such assurances were supported by estimates provided by SDMA to the Shipyard Manager, only to be found grossly inaccurate later.
It was then expected that work on “Reef Endeavour” would complete by late 1994 but the same did not eventuate. In 1995, the Manager of the “Reef Endeavour” advised that final fittings, testing, and sea trials would conclude by 31 October 1995. But, once again, this did not eventuate because of yet another dispute. A maximum late delivery penalty of $F650,000, later lowered to $F225,000, was slapped on the shipyard by the clients (The Fiji Times 1995a: 4). GSPS disputed the penalty since clients failed to submit the required designs and specifications which were required to build the ship, leading to additional costs.
On 30 November 1995, the ship stood complete except for a few minor tasks. Government refused to hand over the “Reef Endeavour” to its Australian owners because payments and claims needed sorting out. Government blamed the owners, accusing them of “grossly inequitable contractual arrangements and misrepresentation,” while the owners blamed the shipyard for lack of planning and supervision which they said, resulted in high absenteeism and low productivity, causing delays in delivery (AAP 1995d: 5). When the issue could not be resolved amicably, both the parties opted for legal arbitration in Australia (A Special Correspondent 1995: 1). The project was held up for a year in a lengthy and very costly court battle. During the court hearing, the clients/owners argued that the Fijian Government failed to live up to its promise of delivery. A predelivery inspection by the owners in November 1995 revealed a 40-page defect report which according to the owners, the shipyard refused to correct. In December 1995, the government won an official pardon in the prolonged dispute. The Supreme Court ruled that the dispute be resolved through arbitration (The Fiji Times 1995b: 4). On 19 December 1995, the Fiji cabinet decided to rescind the contract because of the “immense” financial losses (AAP 1995d: 5).
Later, the government and Captain Cook Cruises realized that there was no point in prolonging arguments. It was decided that government be paid an additional $F4.5 million upon delivery of the ship (The Fiji Times 1996: 3). The parliamentary debate of 22 February 1996 (p. 2302) revealed that the cost of the project as at February 1996 was some $F20.5 million, while the amount received was much lower – $F11.8 million (Hansard Report 1996). Additional charges (legal fees of aborted arbitration and the cost of furnishing and furbishing the ship) kept flowing which mounted the total expenditure close to $F22 million. Fiji taxpayers hence subsidized the Australian company, a whopping $F10 million.
Under-Cost Signup and Accountability
It was the Director of Marine who signed the contract. PSC was requested for surcharge actions but nothing was done (The Hansard Report 1995: 2425) because the Director had retired and the Shipyard Manager left the shipyard (Hansard Report 1996). There was a discrepancy between the costing done by the shipyard and the actual signup amount (The Hansard Report 1996; two long-serving interviewees, union interviewees). The Fiji Labor Party (FLP) leader’s question below confirms the issue raised by most interviewees:
The blame was at one time pushed onto the then Manager of shipyard who was the first local Manager of GSPS between the years 1977 and 1992. He refused to sign the initial contract of the “Reef Endeavour” because it did not give much control to their division and was too loose a contract since it was between three parties – the architect (SDMA), the owners, and the builders. Since he was not the one who signed the contract, the blame was considered baseless. Resultantly, he was offered options to join other government departments but he opted for further studies. He was also one of the Directors to the Board of this enterprise.
The shipyard was telling them all along that they cannot build the ship in the design which was then proposed for less than $[F]15.5 million. It was not the civil servants and it was not the workers. They gave an accurate estimate of the cost of the ship in the form and design and that was $[F]15.5 million. But eventually their arms were twisted to sign the deal for $[F]7.5 million… But this was no fault of those who are managing the shipyard. It was the fault of the politicians who interfered (Hansard Report 1996: 2301–2302).
The theoretical model prepared for this research is supportive of the findings. In the case study, the “principal” was the client/owner – Captain Cook Cruises (Aust.) Ltd. and Qantas Holiday Tours Ltd. The principal’s project manager was SDMA. The “agent” was the Fijian Government while GSPS acted on behalf of the agent as the contractor that built the cruise liner. The government can also be seen as the project manager of GSPS because it was the key decision maker. GSPS cannot thus be noted as the independent agent. The point that Wickramasinghe and Lamb (2002) made in terms of reflexive principal-agent relationship does make some sense here since, when a government acts as the project manager and hands over the project to a public enterprise, the government may appear as the principal, with the public enterprise as the agent.
The contractor, GSPS (public enterprise) assumed the risks associated with the building of the “Reef Endeavour” at a cost higher than the contracted amount, leading to losses from the project. The politicians/government (agent) controlled the contractor by interfering with cost estimates and making major decisions. The goals of the contractor, the agents, and the principals were poorly aligned giving rise to disagreements and even litigation. For instance, the government blamed the owners, for “… inequitable contractual arrangements and misrepresentation” while the owners blamed GSPS for lack of planning and supervision. The agent, the then government, was interested in calling the shots and making the final decision irrespective of the higher cost estimate of the GSPS professionals which maximized the risk for GSPS. The lack of power of the contractors and the control exerted by the agent (government) through the bureaucratic systems and tendering boards cost GSPS. This worked to the disadvantage of both the agent and the contractor in terms of project loss. The arrangement was also rather entangled with the involvement of politicians and SDMA. This finding concurs with Ceric’s (2013) explanation of: different parties being involved in projects even before the execution and the existence of potential conflict of interests because of each participant’s self-interest. However, unlike the earlier government interference of deciding on the contract amount, the later government actions (seeking additional payments) reduced the loss of GSPS to some extent but this would not have been the case if the contractor’s (GSPS) estimates were favorably considered in the first place. This concurs with Roja’s (2015) statement of the effects of political pressures being agency specific. The new finding here is – political interference is negative given mistakes of lower cost estimates but the negative impact lessens with later government actions of seeking higher payments, postconstruction. This new finding can be particularly useful in contract formulation in terms of stipulating the rights of all parties concerned, in situations where the project costs may not remain within the contracted amount. The final decision on the contract amount by the politicians/government (agent) placed limits on the contractor (GSPS), reducing the level of its freedom, increasing uncertainty about what GSPS should or can do. Control was further exerted by the clients (principals). This finding is parallel to Ceric’s research (2013) in that – both the principal and the agent may exhibit opportunism. The agent – the government exhibited opportunism when asking for an additional $F3.06 million for the ship’s release while the principal exhibited opportunism by slapping a penalty of $F650,000, later reduced to $F225,000 for late delivery. This discerns from the common argument of the traditional agency theory that portrays only the agent as the opportunist. The contractor, GSPS assumed majority of the commercial risk and yet did not have the freedom to name its price to the client (principal). The total expenditure amounted to some $22 million while revenue was much lower – $F11.8 million. The divergent interests of all parties resulted in problems related to timeliness and delivery, drawn out legal battles, higher costs, and project loss. The client (principal) was also, no doubt, adversely affected in terms of having to pay more than the contracted amount to obtain the completed project, delayed project completion, and paying for litigation in order to resolve the tiff between itself and the government. This finding is in agreement with Aritua’s et al. (2011) explanation of diverse stakeholder aspirations and challenges of multiple project procurement.
And as Kauppi and Raaij (2015) argue, agency problems may not only arise with agents withholding information; principals may also withhold information. While SDMA (principal’s project manager) did not withhold information, it provided inaccurate estimates. The government, despite the higher cost estimates of its own marine experts signed the initial contract for $A7.7 million upon reassurances by SDMA. The principal however did withhold information as it failed to provide specifications of complete designs and construction plans as and when required, leading to delays. This finding in in line with Al-Sedairy’s (1994) findings – project conflicts stemming from timing, project concepts, costs, and specifications. The agent (government) was disadvantaged because of its own mistakes or control over the contractors (GSPS) who were more effective in preparing cost estimates. SDMA also went bankrupt and the government was unable to take any legal action for the inaccurate estimates. GSPS ended up completing the project at a higher cost than the contracted amount. The government decided to rescind the contract when it realized that its own taxpayers were subsidizing the foreign tourist vessel. The government chose to believe the SDMA cost estimates, not its own marine experts. This entry accordingly suggests that such contracts be deliberated in rigorous parliamentary debates. In addition, all partnerships or deals should go through proper due diligence by independent committees to ensure that partners are genuine and have the financial and management capabilities. There should also be ongoing periodic checks on new partnerships. To this end, a critic committee comprising the media, NGOs, private firms, social welfare groups, politicians, academics, and unions could be setup to objectively question, carry out background research on probable partners, help in the preparation of prudent deals, and monitor state actions.
Ceric’s (2003) finding of poor communication being a common project risk is also worth noting. When the government and the client could not reach amicable solutions, the matter ended up in courts. However, when both parties were open to dialogue, they realized that there was no point in prolonging arguments which helped to sort out the differences.
Furthermore, the complementary view to agency theory presents man as an honest but not fully competent actor (Kauppi and Raaij 2015). In the case study, the contractor was more competent than the politicians in the area of cost estimates, but later, upon realizing the costing mistake, the government did take necessary actions to ensure that it be paid a higher amount which reduced overall losses. Surprisingly, it was the principal’s project manager (SDMA) which proved out to be incompetent as it provided with inaccurate estimates. Here government’s (agent) incompetency was with respect to having more faith in SDMA than its own marine experts who estimated higher costs – a new finding. This entry suggests that when dealing with clients, thorough checks be done prior to making any commitments. An objective third party should attest (Burritt and Welch 1997) not only the background information on clients but also confirm the estimates. More research needs to be done in the area of governments having more faith in external companies than their own experts.
The theoretical contribution of this study is threefold. First, the study extended the agency theory to include the contractor (a public enterprise), the government as the agent, client as the principal, and the designer/architect as the client’s project manager. This extended model was empirically tested and is supportive of the empirical findings. Second, this model helped to better understand how the involvement of four parties can complicate a construction project in the public sector – this paves the way for the development of even more effective models. Third, this study sheds light on political interference, an issue that has not been given due consideration in many public management studies (Walker et al. 2013). In addition, unlike the earlier study of Ceric (2013), this study reported new findings of agent’s (government) interferences and mistakes leading to contractor (public enterprise owned by the agent) losing out, the outcome being negative for the contractor, the government, and even the principal (client). The government also chose to put more faith in the principal’s project manager’s estimates than its own contractor’s estimates. This matter requires further investigation and findings of future study may bring about a more effective holistic theory. The findings are applicable to countries, construction companies, and public sectors where a government signs the deal and is the key decision maker for public enterprise projects.
- A Special Correspondent (1995, December 14) Endeavour dispute to go before Supreme Court. The Fiji Times, p 1Google Scholar
- Australian Associated Press (AAP) (1995, December 22) Government seizes reef endeavor. The Fiji Times, p 5Google Scholar
- Ceric A (2003) A framework for process-driven risk management in construction projects. PhD diss., Research Institute for the Built & Human Environment, School of Construction and Property Management, University of SalfordGoogle Scholar
- Ceric A (2013) Application of the principal-agent theory to construction management: literature review. In: Smith SD, Ahiaga-Dagbui DD (eds) Proceedings of the 29th Annual ARCOM Conference Reading. Association of Researchers in Construction Management, UK, pp 1071–1081Google Scholar
- Eisenhardt KM (1989) Agency theory: an assessment and review. Acad Manag Rev 14(1):57–74Google Scholar
- Farazmand A (2002) Modern organization: theory and practice, 2nd edn. Praeger Publishers, WestportGoogle Scholar
- Gillham B (2001) Case study research methods. Continuum, LondonGoogle Scholar
- Hansard Report (1995, February 13) Parliamentary debates, House of Representatives, Parliament of Fiji. Daily Hansard, pp 2423–2427Google Scholar
- Hansard Report (1996, February 22) Parliamentary debates, House of Representatives, Parliament of Fiji. Daily Hansard, pp 2287–2332Google Scholar
- Kauppi K, Raaij EM (2015) Opportunism and honest incompetence – seeking explanations for noncompliance in public procurement. J Public Adm Res Theory 25(3)Google Scholar
- Klein JH (2009) Principal agent problems in the financial crisis of 2007–2009. A paper from Master Thesis, Faculty of Sciences, Business Mathematics and Informatics, AmsterdamGoogle Scholar
- Marine Department (1969) Annual report. Government of Fiji, SuvaGoogle Scholar
- Rojas P (2015) Political pressures: too large, too small, or just right? In: Gnan L, Alessandro H, Monteduro F (eds) Contingency, Behavioral and evolutionary perspectives on public and nonprofit governance, Studies in public and non-profit governance, vol 4. Emerald Group Publishing Ltd, London, pp 155–187CrossRefGoogle Scholar
- The Fiji Times (1995a, August 11) Ship’s late delivery could cost $225,000. The Fiji Times, p 4Google Scholar
- The Fiji Times (1995b, December 8) Shipyard dispute to be settled in Aust. The Fiji Times, p 4Google Scholar
- The Fiji Times (1996) Endeavour bids Suva farewell. The Fiji Times, 13 January, 3Google Scholar