What Are the Causes and Cures of Poor Megaproject Performance? A Systematic Literature Review and Research Agenda

1School of Construction and Project Management, University College London, UK 2Science Policy Research Unit, University of Sussex Business School, UK

Corresponding Author: Juliano Denicol, School of Construction and Project Management, University College London, Gower Street, London WC1E 6BT, UK. Email: juliano. [email protected] ucl. ac. uk

Project Management Journal Vol. 51(3) 328–345

© 2020 Project Management Institute, Inc. Article reuse guidelines:

sagepub. com/ journals- permissions DOI: 10. 1177/ 8756 9728 19896113

journals. sagepub. com/ home/ pmx

What Are the Causes and Cures of Poor Megaproject Performance? A Systematic Literature Review and Research Agenda

Juliano Denicol1, Andrew Davies2, and Ilias Krystallis1

Abstract This systematic literature review explores the megaproject management literature and contributes by improving our under- standing of the causes and cures of poor megaproject performance. The review analyzes 6,007 titles and abstracts and 86 full papers, identifying a total of 18 causes and 54 cures to address poor megaproject performance. We suggest five avenues for future research that should consider examining megaprojects as large- scale, inter- organizational production systems: (1) design- ing the system architecture; (2) bridging the gap with manufacturing; (3) building and leading collaborations; (4) engaging institu- tions and communities; and (5) decomposing and integrating the supply chain.

Keywords megaproject management, performance, problems, solutions, failure, success

Article

Introduction Megaprojects are the delivery model used to produce large- scale, complex, and one- off capital investments in a variety of public and private sectors. With a total capital cost of US$1 billion or more, megaprojects are extremely risky ven- tures, notoriously difficult to manage, and often fail to achieve their original objectives (Altshuler & Luberoff, 2003; Flyvbjerg, Bruzelius, et al., 2003; Merrow, 2011; Priemus & Van Wee, 2013). In 2013, McKinsey suggested that US$57 trillion would be spent on infrastructure invest- ment between 2013 and 2030 (McKinsey Global Institute, 2013). Similarly, Flyvbjerg (2014) estimated the global spending on megaprojects at US$6 to US$9 trillion annually, emphasizing a statement made by The Economist (2008): “The biggest investment boom in history is under way.”

Established as a standalone temporary organization, megaprojects can be led by a client team, prime contractor, or some form of temporary alliance, joint venture, or coalition of multiple parties (owners, sponsors, clients, contractors, suppliers, and other stakeholders) that work jointly on a shared activity for a limited period of time in an uncertain environment (Jones & Lichtenstein, 2008; Merrow, 2011). Each megaproject is usually decomposed into many smaller inter- related projects and organized as a program. A large organization—the client, prime contractor, and/or delivery partner—is established to coordinate and integrate the efforts of numerous subgroups and suppliers involved in project activities (Davies & Mackenzie, 2014; Davies et al., 2009;

Merrow, 2011). This organization manages the overall pro- gram and the interfaces between projects; deals with external suppliers through separate contracts; and is accountable for meeting time, cost, and quality performance goals.

Most of the extant research is concerned with understanding why megaprojects fail so frequently and seeks to identify some of the dimensions that make megaprojects so difficult to man- age, including their size (Flyvbjerg, Bruzelius, et al., 2003; Flyvbjerg, 2017; Merrow, 2011; Morris & Hough, 1987), uncertainty (Lenfle & Loch, 2010; Miller & Lessard, 2000; Stinchcombe & Heimer, 1985), complexity (Brady & Davies, 2014; Davies & Mackenzie, 2014), urgency (Morris & Hough, 1987; Shenhar & Dvir, 2007), and institutional structure (Scott et al., 2011).

This research aims to deepen and extend our understand- ing of the causes and cures of poor megaproject performance, a question raised by Flyvbjerg (2014). To achieve this aim, the objectives of our systematic literature review were to identify prior research on megaprojects, including adjacent

Denicol et al. 329

literature on large engineering projects, major projects, grand- scale projects and other related terms, and categorize the research according to how it identifies the main causes of poor megaproject performance. We then categorized the research according to the cures—the strategies and practices of megaprojects around the world—offered to resolve poor megaproject performance. We identified five research ave- nues with emerging topics that can offer novel insights into the causes and cures of poor megaproject performance.

A variety of concepts and theoretical frameworks have been developed and applied to understand the causes and cures of poor megaproject performance. Our research find- ings categorize the literature under six themes. Under each theme we identify the three predominant concepts and dis- cuss the causes and cures of poor megaproject performance. We found that each concept draws upon its own distinct the- oretical foundations and frameworks, although there is no space in this article to explore each in detail. For example, optimism bias applies cognitive psychology to understand how managers of megaprojects deal with uncertain outcomes, and systems integration draws upon the organizational capa- bilities and design literature to identify how megaprojects are decomposed and integrated. While significant efforts have been made to improve our understanding of megaproject per- formance, each contribution alone provides insights into a partial or isolated phenomenon. There is no overarching the- ory or framework that can connect the disparate contribu- tions into a complete picture identifying how performance depends on various components—such as decision making, integration, leadership, and teamwork—working together as an integrated whole. We conclude the article by suggesting that new research and theory building should adopt a sys- temic view, taking into account some of the different aspects impacting megaproject performance. We suggest the litera- ture could be enhanced by research that considers a megaproj- ect as a system of production and by studying their individual topics through a systems lens.

Research Methods Originally developed in the medical sciences to consolidate infor- mation from several sources, a systematic literature review is a transparent, rigorous, and detailed methodology used to support decision making (Tranfield et al., 2003). This research method is used to build theory by accumulating knowledge and evidence after analyzing a large number of studies and methods, thereby increasing the consistency of the results and the conclusions (Akobeng, 2005). Informed by Denyer and Tranfield (2009) and Tranfield et al. (2003), our systematic literature review was under- taken in three stages. First, a planning stage identifies the needs of the review and develops the protocol, which defined the overall strategy, the keywords, and its interactions in the search for arti- cles. Second, a development stage selects articles for data extraction, assessment, and data synthesis. Third, a dissemination stage connects the research findings with ongoing conversations

in the academic literature and with practice through accessible material for practitioners.

Planning Stage The rigid protocol of systematic reviews is a major limitation when analyzing the research field of management and organi- zation studies. In our study, the terminology is not conver- gent as in medical sciences, but rather divergent with many authors developing different conceptualizations and termi- nologies to refer and explain the same phenomenon. We identified keywords on the subject based on our prior experi- ence through the mechanism of brainstorming during two 1 hour meetings. The strategy was to include a wide range of words and synonyms, in which the keywords were grouped in three categories: Megaproject synonyms, Success syn- onyms, and Failure synonyms. The full list of synonyms for each category can be found in Appendix 1 at the end of the article. The keywords were organized into two search strings, which were used to search the papers on academic databases. The first search string included all Megaprojects synonyms associated with Success synonyms, such as (“large scale project*”) AND (“success*”); the second included all Megaproject synonyms associated with failure synonyms, such as (“grand scale project*”) AND (“failure*”).

Development Stage We carried out a systematic search for academic articles in two of the largest academic online databases: Web of Science and Scopus, from all years until September of 2017. The search for articles was conducted through the combination of keywords in three areas of interest: Synonyms for the term Megaproject most commonly used in the literature, Success synonyms, and Failure synonyms. The review process was conducted according to the following steps:

1. The academic databases Scopus and Web of Science were chosen to conduct the search for papers using the strings identified in the Development Stage.

2. The first string related to Megaprojects and Success re- turned Scopus (3,423) and Web of Science (2,498). The second string, related to Megaprojects and Failure, returned Scopus (1,659) and Web of Science (880). The papers from the two search engines were then consoli- dated on Mendeley aiming to exclude duplications (8,460), resulting in a final folder called Megaprojects AND Success AND Failure (6,007), as illustrated in Figure 1. Journal articles were included, whereas confer- ence papers, reports, and book chapters were excluded. On Scopus the papers were limited to the following sub- ject areas: Business, Management Accounting; Computer Science; Decision Sciences; Economics, Econometrics, and Finance; Energy; Engineering; Social Sciences; Environmental Science; Materials Science;

Project Management Journal 51(3)330

Multidisciplinary and Undefined. On Web of Science the papers were limited to the following subject areas: Architecture; Area Studies; Business Economics; Computer Science; Construction Building Technology; Energy Fuels; Engineering; Environmental Sciences Ecology; Geography; Government Law; International Relations; Materials Science; Metallurgy Metallurgical Engineering; Operations Research Management Science; Public Administration; Science Technology; Social Sciences; Telecommunications; Transportation; Urban Studies; Water Resources.

3. Titles and abstracts of articles were analyzed according to the inclusion and exclusion criteria (see Appendix 2 at the end of the article), reducing the number from 6,007 to 1,075.

4. We met to cross- check and discuss the results of the evaluation by title and abstract and given the remain- ing high number (1,075) decided to further categorize the papers into three categories. This strategy was ad- opted aiming to isolate and exclude the high number

of papers about deterministic models and algorithms and financial mechanisms (mainly organized around the public–private partnerships [PPPs] literature).

5. Using the inclusion and exclusion criteria, the articles were separated into categories A (248), B (216), and C (611). Category A represented articles of Size, catego- ry B represented articles where the focus was around Complexity, and the category C list represented arti- cles of quantitative models, contractual arrangement, funding, and financing and had little focus on the man- agerial aspects of megaprojects.

6. The papers included categories A (248) and B (216), which were consolidated again (464); in light of the high number we adopted the strategy of employing the impact factor as a measure to maintain quality and re- duce the number of papers entirely reviewed (Aliaga- Isla & Rialp, 2013; Crossan & Apaydin, 2010; Klang et al., 2014; Papamitsiou & Economides, 2014; Podsakoff et al., 2005; Pölkki et al., 2014; Richards et al., 2014; Tompkins & Arendt, 2015; Überbacher,

Figure 1. Steps of the systematic literature review.

Denicol et al. 331

2014; Wielenga‐Meijer et al., 2010; Zhang et al., 2011). We analyzed the list of 464 papers and concluded that the megaproject literature is still concentrated in project management journals; therefore, it was necessary to limit the impact factor to include only the main journals in the field. We clarify that by applying an impact factor filter, two categories of papers were excluded, namely: (1) papers from journals with an impact factor below the threshold, and (2) papers from journals without an im- pact factor found via Scopus. It was agreed that papers published in academic journals with an impact factor above (1.70) would be included for this review. By lim- iting the impact factor to above (1.70), it was possible to include the main journals of project management: Journal of Engineering Construction and Management (ASCE) (1.73), Journal of Management in Engineering (2.01), Project Management Journal® (PMI) (2.71), Automation in Construction (2.91), and the International Journal of Project Management (4.03).

7. The final list contained (145) articles that were filtered from the initial search (6,007) following steps (3 – 6). Although the abstract of the (145) articles fit the inclu- sion criteria, several of them did not meet the criteria after reading them from start to finish; therefore (86) were considered to inform the review. Using an ex- traction sheet on Microsoft Excel, relevant information related to descriptive data (title, authors, journal, year, and so forth), and information that answered the initial research questions (aims and objectives, causes, cures, and future research) was extracted in a structured fash- ion. The full description of the extraction spreadsheet can be found in Appendix 3 at the end of the article.

8. The articles were reviewed to extract the causes and cures of poor megaproject performance. The process of extracting the causes and cures through an in- depth anal- ysis of each paper followed the coding method presented by Saldaña (2016), where the reviewer used each cause or cure as a first order code, which represented one entry in the extraction Excel spreadsheet. The first order codes were clustered into categories, which were later orga- nized into themes. We provide an illustrative example of this process. Cure extracted from Brady and Davies (2014): Establish an integrated project team approach including the client, the system integrator, and first- tier contractors. Category: Integration. Theme: Strategy, Governance, and Procurement. Concept: Delivery mod- el strategy. Combined concept cure: Adopt integrated project teams to deliver the project, involving key deci- sion makers from institutional to supply chain levels (owner, sponsor, client, system integrator, delivery part- ners, first- tier contractors, second- tier suppliers, and operator).

9. After all papers were reviewed and the extraction fin- ished, we met again to discuss our independent analy- sis and consolidate the categories into themes. The

categories of each reviewer were analyzed over 1.50 hours and six themes emerged from its consolidation.

10. After the six themes were defined, we identified three predominant concepts in each theme that helped to ex- plain the causes and cures of poor megaproject perfor- mance. The division of the themes into smaller units (concepts) allowed us to increase the level of detail, aiming to contribute to theory and practice. Each con- cept was explored by the identification of its main cause and three potential cures, drawing upon material extracted from the analyzed 86 papers.

11. In an effort to connect the findings of the systematic lit- erature review with industrial debates and inform our two workshops, we identified industrial reports from the last five years where those concepts were discussed. Although those reports are not part of our dataset, the quotes extracted provided an extra contextualization lay- er aiming to stimulate lively discussions in the work- shops with academics and senior practitioners.

12. The six themes were validated by professor Peter Hansford from University College London, who served as chief construction adviser for the UK government and has considerable industrial experience, often providing strategic advice on megaprojects and infrastructure policy.

Our decision to exclude highly influential books on large- scale projects is recognized as one of the limitations of the systematic review methodology. However, the exclusion is common practice given their classification as gray literature (Adams et al., 2017), supported by the often- missing peer- review process, which is perceived as an indication of rigor and quality, and the inconsistency of searches in books when compared to articles included in academic databases. Many pioneering project management ideas first developed in books, such as the concept of a strong owner (Morris & Hough, 1987), the front- end definition in Morris (1994), and the owner–contractor interface in Merrow (2011), which appear as key references in the papers identified in our liter- ature review. Therefore, although these books are not identi- fied in our review, their profound influence on the research undertaken on megaprojects is evident in many of the articles appearing in our review.

Dissemination Stage The findings were presented, assessed, and verified in two workshops: the first with Professor Peter Morris and academ- ics in the School of Construction and Project Management at University College London; and the second with senior prac- titioners from some of the United Kingdom's largest infra- structure megaprojects (Crossrail, Thames Tideway Tunnel, High Speed Two, and Hinkley Point C). Both workshops generated a productive, lively, and hugely insightful discus- sion. Participants recognized the value of the categorization

Project Management Journal 51(3)332

but were critical of the existing literature exploration in silos. They encouraged us to think about more engaged and com- prehensive research to understand the variety of institutional, behavioral, organizational, and other factors affecting the performance of megaprojects—from front- end planning, through execution, to operational outcomes.

Results After executing the analysis as outlined in the Methods section, the literature dataset of 86 articles were clustered into six themes: (1) decision- making behavior; (2) strategy, governance, and pro- curement; (3) risk and uncertainty; (4) leadership and capable teams; (5) stakeholder engagement and management; and (6) sup- ply chain integration and coordination. These six themes make sense of the sample and reveal the main causes and cures of poor megaproject performance as found in the academic literature. Each theme is further subdivided by concepts that help to discuss the causes and cures of poor megaproject performance and con- tribute to the ongoing conversations in the literature. For each concept, the main cause of poor megaproject performance is iden- tified and a list of three associated cures (strategies and practices) is presented. The reason we deliberately selected to illustrate only the main cause for each concept is twofold: (1) physical limitation in the paper, given that the addition of another cause per concept would result in a significantly larger document, and (2) an attempt to move the academic conversation to a more positive (at least less pessimistic) discourse emphasizing solutions rather than problems, enabling the focus on an expanded number of cures. Therefore, this research presents six themes, 18 concepts, 18 main causes, and three cures associated with each concept (therefore, 54 cures in total). The papers categorized under each theme are documented along with a comprehensive list of extracted strate- gies and practices.

Theme 1: Decision-Making Behavior A significant body of literature on megaproject performance is related to decision- making behaviors. Theme 1 identifies how behaviors in the front- end and during execution are associated with poor performance in decision making. This theme rejects technical explanations as the main reason for inadequate forecast- ing and discusses poor performance as a result of psychological and behavioral reasons and how those affect decision making. The three most predominant concepts in this theme are: (1) opti- mism bias (delusion): executives are overly optimistic and thus overestimate benefits and underestimate costs; (2) strategic mis- representation (deception): executives strategically misrepresent the truth and seek to satisfy their own interests; and (3) escalating commitment: executives continue to follow the pattern of behav- ior leading to unsuccessful outcomes rather than follow an alter- native course of action.

The main cause of poor performance associated with opti- mism bias is biased judgment and advice provided by experts in their fields who tend to create an optimistic scenario and

circumvent known risks and unforeseeable uncertainties. This is an unconscious phenomenon that psychologists classify as the planning fallacy, leading executives to underestimate costs in several areas of complex projects. The leading cause related to strategic misrepresentation refers to diverse pres- sures (political, organizational, and individual) forcing the decision maker to manipulate the situation by usually under- estimating costs and ignoring risks. Early estimates and fore- casts are used deceptively to inform decision making and achieve the necessary alignment and support from stakehold- ers (including the taxpayer) to proceed with that preferred project. The primary cause connected to escalating commit- ment is the overall perception, which mostly works as a norm, that, once started, a megaproject is too big to fail and too costly to stop. Managers allocate resources in order to com- plete the project, even when subsequent assessments and audits indicate a decision in another direction, where the final benefits are no longer superior to the necessary investment. A list of strategies and practices to cure the causes of each con- cept of theme 1 is presented in Table 1.

Theme 2: Strategy, Governance, and Procurement The second theme encompasses the definitions of strategy, gov- ernance, and procurement, including the processes during the initiation and planning phases of a megaproject, which the lit- erature typically addresses as the front- end stage of projects. Decisions made at this stage may influence subsequent stages and the ability to achieve successful project outputs and out- comes. The three most predominant concepts in this theme are: (1) sponsor, client, owner, operator: associated with the roles and responsibilities of these entities throughout the project life cycle, with particular emphasis on the front- end stage; (2) gov- ernance: linked to the delegation of authority formally and informally, at the organizational and individual levels; and (3) delivery model strategy: related to the strategy adopted by firms to organize themselves in combination with partners and suppliers, and combining in- house and external capabilities to best organize and deliver the project.

The main cause of poor performance associated with the sponsor, client, owner, and operator relates to inadequate definitions of roles and responsibilities during the project life cycle, the need to clarify which entity is the sponsor, where the ownership resides, who the intermediary client is, and who is going to operate the asset. In the absence of a long- term vision and clear definitions of roles, the entity promot- ing the project often seeks to transfer the risk to the supply chain. As a result, client organizations are rarely willing to bear the risk. The leading cause related to governance is an inadequate attention to the design of the governance structure and its evolution over time, including the balance between formal (hard, rigid) and informal (soft, gut- feeling, emerg- ing) governance structures. The leading cause connected with delivery model strategy is the poor understanding and

Denicol et al. 333

definition of the balance between the in- house capabilities of the client organization and those outsourced to the market and allocated to partners and contractors. Often, the mecha- nisms used to procure capacity and capability from the mar- ket result in transactional and adversarial relationships with the supply chain, rather than integrative and collaborative ones. A list of strategies and practices to cure the causes of each concept of theme 2 is presented in Table 2.

Theme 3: Risk and Uncertainty This theme captures the literature that addresses risk and uncertainty, where articles covered technology development processes and analyzed strategic decisions to overcome risks in megaprojects across several industrial sectors. The three most predominant concepts are: (1) technological novelty: first- of- a- kind technologies have frequently being introduced in large innovative projects and are associated with risks; (2) flexibility: the ability to be adaptive and responsive to chang- ing and uncertain circumstances; and (3) complexity: the underlying factor of megaprojects that can be defined by the large number of parts and its relationships among each other and with the external environment.

The main cause of poor megaproject performance associated with technological novelty is the introduction of unproven technol- ogy leading to cost and time overruns. The uncertainty about how to deal with a new technology often requires longer design and development phases of the project. The leading cause related to flexibility refers to early decisions (formal and informal) that

constrain the necessary adaptability by mutual adjustments in a complex, dynamic, and uncertain environment. Many factors restrict project flexibility, including centralized decision making, financing, regulatory frameworks, design, commercial arrange- ments, contracts, and technology, among others. The primary cause connected with complexity is the uncertain interactions between a large number of moving and evolving parts within the megaproject system, as well as their relationships with the external environment. The system can be affected by many dimensions, such as regulations, information, technical, and organizational components. A list of strategies and practices to cure the causes associated with each concept of theme 3 is presented in Table 3.

Theme 4: Leadership and Capable Teams This theme refers to relationships among project team mem- bers, individual competencies, required skills, and organiza- tional capabilities that contribute to the performance of megaprojects. The three most predominant concepts are: (1) project leadership: the need for project champions, dedicated leaders who are committed to the success of the project; (2) competencies: competencies and skills that individuals forming project teams need to possess; and (3) capabilities: the ability that firms have to produce specific products or services relying upon collective organizational knowledge.

The main cause of poor performance associated with proj- ect leadership is an inappropriate definition of the project culture and sense of purpose, which lead to intra- and inter- organizational misalignments. It promotes dysfunctional

Table 1. Cures for the Causes Associated with Theme 1’s Concepts

Strategies and Practices

Cures for Optimism Bias Publications 1. Develop a strong benchmarking exercise looking extensively to

previous similar projects, assessing what has worked and what could be improved from those projects

Ansar et al. (2014), Flyvbjerg et al. (2009)

2. Develop plans for uncertainties, given that megaprojects are complex open systems under constant change

Barnes and Wearne (1993), De Bruin et al. (2014), Dimitriou et al. (2013), Doan and Menyah (2013)

3. Invest appropriate time at the project front- end to develop the tools and processes that would scrutinize and prevent biases

Cantarelli, Molin, et al. (2012), Flyvbjerg (2014)

Cures for Strategic Misrepresentation Publications 1. Develop lines of defense and checks to challenge whether the

information being produced is appropriate and correct Flyvbjerg et al. (2002)

2. Introduce mechanisms to identify actors employing opportunistic behavior

Flyvbjerg, Holm, et al. (2003)

3. Develop penalties for ignoring or providing misleading information Flyvbjerg, Holm, et al. (2003)

Cures for Escalating Commitment Publications 1. Introduce the option to defer to further assess risks, the economic

viability and avoid over commitment Baccarini and Love (2014), Doan and Menyah (2013), Flyvbjerg, Holm,

et al. (2003), Genus (1997) 2. Assess the political scenario, recognizing that governments seek to

balance control and flexibility for political maneuver and electoral reasons

Lopez del Puerto et al. (2014), Marshall and Cowell (2016), Naderpajouh et al. (2014)

3. Invest resources

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.