Risk assessment and allocation model for public-private partnership infrastructure projects in Pakistan

Abstract

To overcome the budgetary constraints in the provision, operation and maintenance of public infrastructure and in recognition of the superior private sector skills and expertise, governments worldwide, including the Pakistani government, are increasingly turning to public-private partnerships (PPPs) for infrastructure delivery (economic and social). The initial impression of the option of PPPs may seem like a panacea for all public infrastructure needs; however, international literature has reported mixed results regarding their performance and success. PPP infrastructure projects are risky in nature, and inadequate risk management on projects is a principal cause of project distress or failure. Adequate assessment of risk is essential to assist stakeholders in planning for efficient risk allocation and mitigation and ensure success in business and projects. Furthermore, appropriate risk allocation and sharing is a critical success factor. Although Pakistan has some experience in delivering infrastructure projects via PPPs, especially in power and transport infrastructure sectors, limited research is available to ascertain the situation in the local context. Moreover, the existing PPP body of knowledge and risk management literature can benefit from additional research in an effort to overcome certain limitations. Hence, the overall aim of this thesis is to develop an appropriate mechanism to enhance risk management outcomes in the context of PPP infrastructure projects in Pakistan. This aim was achieved with empirical investigations on the identification of risks and development of measures of effective risk management (ERM) to guarantee project success. The thesis also developed, demonstrated and validated risk assessment and allocation models to assist stakeholders in risk management decision making on projects. Data for the achievement of the objectives was primarily collected via a questionnaire survey of 90 experts in the local industry, who were selected based on purposive sampling and semi-snowballing approaches. Eight semi-structured interviews, seven case-based surveys and expert reviews were also conducted to ground the study in an industrial and professional context. Self- and investigator-administered surveys were conducted. Various statistical tests and analytical approaches for risk assessment and allocation modeling were adopted. Statistical tests included: mean score ranking, inter- and intra-group agreement analysis, tests for reliability and validity and factor analysis. Fuzzy set theory (FST) in conjunction with simple additive weighting and fuzzy measure based fuzzy integrals were the utilized multiple-criteria decision-making methods for the risk assessment and allocation models. Investigation to improve risk management outcomes on PPP projects resulted in identification and development of 30 ERM measures, all of which were rated at least moderately important on average. This outcome signifies the relevance of the proposed measures in terms of potentially influencing quality and outcomes of risk management efforts and guiding industry practitioners to deploy prevailing risk management guidelines, processes, tools and techniques effectively for achieving successful PPP projects. Factor analysis established six critical underlying dimensions for ERM as follows: (1) well-documented structured management approach; (2) comprehensive requirements and risk evaluation; (3) post-contract risk management; (4) knowledge-driven risk management; (5) risk assessment quality; and (6) public sector risk management. A conceptual framework for ERM on PPP infrastructure projects was also proposed to provide a systematic guideline to industry stakeholders and encourage implementation of the identified measures by clarifying their relationship with project parties, the project lifecycle and the risk management process. Application of FST in risk analysis revealed 22 critical risk factors out of the 45-factor risk register developed for this study, that were categorized into seven critical risk groups (CRGs) of correlated factors using factor analysis. Risk factors that achieved a linguistic assessment of high impact reflect issues related to institutional capacity and local economy, which tallied well with outcomes reported in research on developing countries. Further analysis based on fuzzy measure and non-additive fuzzy integral combined with arithmetic mean helped obtain an overall risk index that indicated a moderate risk outlook for power and transport infrastructure sectors. Whereas, 'public sector maturity' and 'project finance' were assessed as high-impact CRGs in the power sector, 'project planning and implementation' and 'project revenue' were additionally rated as high-impact CRGs in the transport infrastructure sector. Case-based surveys revealed relatively better performance of the proposed model in mimicking experts' holistic project risk evaluations compared with the additive aggregation approach. The developed framework could be used to assess a country's condition or overall project risk at the initial project stage with minimal input of time and resources, thereby facilitating an efficient and robust risk assessment. Aggregate assessments at the CRG level could facilitate in highlighting key risk areas and may thus enable targeted and effective risk response planning and execution. A comprehensive literature review augmented by industry experts' input identified 17 key risk factors that could exhibit diversity in risk allocation preferences (risk could be shared or allocated to a public or private party), which emanated from and could be attributed to contextual aspects (market, sector and project characteristics). A methodology in conjunction with non-additive fuzzy integral based multiple attribute risk allocation decision approach was proposed. Such methodology could effectively aggregate each stakeholder's risk management capability assessments on accepted risk allocation principles, which were derived from qualitative judgements and experience-based knowledge of experts. Data collected on the key risk factors from privately financed and developed power and transport infrastructure projects in Pakistan were used to demonstrate and validate the model. The model's output comprised the risk management capability index of each party for the key risks being considered, which could then be utilized to make an informed decision on allocation and sharing of risks. Comparison of results with an additive aggregation approach revealed the suitability of the adopted methodology as it performed better in modeling the risk allocation preferences of experts due to its capability to handle interdependencies in the risk allocation criteria. Analysis of the case studies advocated the need to investigate the allocation and sharing of key risks on a case-by-case basis to recognize the contextual factors and obtain an equitable and efficient risk apportionment for project stakeholders. Research outcomes from this thesis have contributed to the body of knowledge for the risk management of local and international PPP infrastructure projects.