摘要：2020年11月，G20 Insights平台发布《建立智能化多方位PPP市场，应对PPP长期困境》（An intelligent multisided PPP marketplace to resolve chronic impediments to PPP，以下简称“报告”）。报告显示，平台经济、多方参与、金融技术、人工智能等新技术和新模式正在快速发展，利用上述新技术和新模式开发设计一个智能PPP市场，并严格监管，可以有效解决PPP发展障碍。报告简要回顾了PPP面临的交易成本高、合同复杂、信息不对称、参与方激励措施不足、财政政策紧缩等突出挑战，并提出构建开放性的智能PPP市场平台（i3PM），促进PPP投资、建设、运营等各阶段高效运转、激励相关方参与的解决办法。最后，报告建议，可以通过引导全球PPP相关方参与共建、制定标准化PPP合同等，推动建立应对PPP长期发展困境的智能PPP市场。
An intelligent multisided PPP marketplace to resolve chronic impediments to PPP
Khaled Albulaihed Hamid Adem
ABSTRACT: The advancement in marketplace platform economies, collective intelligence, financial technology, and artificial intelligence models is radically resolving formidable business and economic impediments. Likewise, these models could be adopted for public–private partnership (PPP) infrastructure projects through the development of rigorously designed and regulated intelligent marketplaces. This policy brief reviews the prominent challenges facing PPP projects, particularly, contractual complexity, information asymmetry, austerity policy inefficiencies, mismatches in stakeholder incentivization, and high transaction costs. It proposes the adoption of an extensible intelligent PPP marketplace architecture and the development of international standards for PPP contracts.
KEYWORDS：Infrastructure Investment and Financing
Most governments and experts propound that the private sector has adequate agility and competencies to efficiently implement infrastructure and public services (Boardman et al. 2015; Engel et al. 2014; Forrer et al. 2010; Levitt and Erikson 2016). Besides, risk transfer from the public to the private sector is the most prominent objective of public–private partnership (PPP). In contrast, empirical evidence reveals that most PPP projects are inefficient and induce new types of risks to the public sector (Chung and Hensher 2016; Chan, Osei-Kyei, et al. 2018; Iossa and Martimort 2012; Lewis 2002; OCED 2018). These are mainly attributed to austerity policy anomalies, information asymmetry, PPP complexity and high transaction costs, inefficient stakeholder incentives, and contractual anomalies (Engel et al. 2014; Aizawa 2015; Hall 2015; Tang et al. 2013; US Department of Treasury 2015; Iossa and Martimort 2012; Tang et al. 2013; Berner et al. 2014).
1. Austerity Policy Anomalies
The main goal of most governments for PPP engagement is to decrease government spending and debt. This austerity goal incurs more expensive private borrowing with inefficient implementation of infrastructure and public services. Most governments use PPP as a vehicle to conceal borrowing and this is referred to as sovereign-debt phobia (Engel et al. 2014; Gonzalez et al. 2015).
2. Information Asymmetry
Due to the involvement of parties with conflicting interests, different sets of information are shared between the parties, where the private sector especially conceals much of this information to maximize their return. Furthermore, PPP projects are executed over comparatively longer time horizons and become more prone to unexpected events such as economic, political, environmental, and public health crises, among others. Information asymmetry coupled with unexpected events impose great uncertainty in PPP projects (Aizawa 2015; Hall 2015).
3. PPP Complexity and Higher Transaction Cost
PPPs are inherently complex compared to public sector projects. This complexity is induced mainly via complex negotiations, contracting, coordination, and management requirements of PPP projects. Furthermore, uncertainty, information asymmetry, and lack of efficient contextual knowledge elevate this complexity. Such complexity inevitably imposes higher transaction cost that debilitate the feasibility of PPP projects (Hall 2015; Iossa and Martimort 2015; Love 2015; Roberts 2015).
4. Inefficient Mutual Incentives
To optimize PPP projects, we must design efficient and balanced incentives for the diverse stakeholders. These incentives should be elicited by strongly emphasizing the needs and challenges of each stakeholder. Nonetheless, most PPP incentives are neither rigorously nor holistically designed to collectively satisfy stakeholders (Hall 2015; Cruz et al. 2015; Athias and Saussier 2007).
5. Contractual and Financial Guarantee Anomalies
Due to the complexity of PPP engagements, information asymmetry, and natural language ambiguity, PPP contracts are fraught with immense ambiguity, incompleteness, and inconsistencies (Berner et al. 2014; Chohra et al. 2011; Cruz et al. 2015; Umar et al. 2018). Furthermore, most of these contracts do not address performance efficiency guarantees, financial and revenue guarantees, specification of nonrecourse loans, and holistic risk management (Albalate et al. 2015; Athias and Saussier 2007; Burger and Hawkesworth 2011).
Most PPP engagements are based on the consensus of the private sector to invest based on revenue guarantees pledged by the public authority. Governments also provide subsidized loans, and financial guarantees to the private sector. Empirical studies show that these guarantees fail due to contractual anomalies or unexpected events such as economic crises (Ashuri et al. 2011; Feng et al. 2015; Hemming et al. 2006). Due to these contractual complexities, fewer partners would be willing to bid, which decreases the number of competing parties and significantly compromises the bargaining power of the public authority to optimize public interests (Iossa and Martimort 2015; Hall 2015).
To resolve these challenges, we propose the adoption of an open and intelligent PPP marketplace (i3PM) platform by public and private sector stakeholders. The proposed marketplace employs the hybrid of multisided platform economies, artificial intelligence, financial technology, and collective intelligence. These factors foster efficient PPP business, investment, contractual, and operational models that create optimal mutual value and incentives for diverse stakeholders to efficiently catalyze the adoption of PPP projects. This marketplace has a dual functionality; it promotes the optimal execution of PPP projects and provides an exchange marketplace platform where PPP securities are seamlessly traded. These securities are called tradable PPP partitions (TPPs).
Figure 1 depicts the interactions of diverse stakeholders with i3PM marketplaces.
This efficient solution to resolve the complex problems currently associated with PPPs provides a holistic approach that concurrently addresses the multifaceted root causes. The proposed marketplace comprises the 10 pillars depicted in Figure 2, while Table D1 (Appendix D) maps these pillars to the impediments that they address.
i3PM employs hybrid platform economies to efficiently resolve the current formidable challenges facing PPP projects. The following are the prominent features of i3PM:
i3PM enables and incentivizes diverse stakeholders to embrace PPP initiatives and symbiotically interact to create mutual value.
It uses high information fidelity models to efficiently manage information asymmetry.
i3PM standardizes PPP contracts using rigorous contract and computational models to promote the efficient design and validation of contracts. These are wrapped within a smart contract in a blockchain infrastructure, thereby enforcing reliability, optimal arbitration, regulatory compliance, fraud management, and exception handling. These features inevitably resolve contractual complexity, ambiguity, and inconsistency.
It standardizes PPP engagement processes using rigorously business, operations, innovation, and financial models. These standard engagement models are developed by diverse competent stakeholders interacting via collective intelligence and marketplace platform economies. This resolves PPP complexity and ambiguity and optimizes PPP efficiency while minimizing transaction costs.
This marketplace seamlessly connects PPP stakeholders with globally competent innovators and multidisciplinary domain knowledge experts, thereby promoting the execution of PPP projects with high competency.
i3PM can easily attract the required critical mass of stakeholders due to high information symmetry, flexible capital guarantee models, and expected high returns.
TPP via i3PM significantly stimulates both supply and demand, thereby enabling the liquidity and scalability of this marketplace.
i3PM liquidity, transparency, reliability, and efficiency inevitably attract diverse domestic and foreign investors into PPP projects.
1. Hybrid Platform Economies
The proposed i3PM employs a hybrid of multisided platform economies, artificial intelligence, financial technology, and collective intelligence to foster efficient PPP business, investment, contractual, and operational models. It creates optimal mutual value and incentives for diverse stakeholders to efficiently catalyze the adoption of PPP projects. Figure C1 (Appendix C) depicts a higher level i3PM topology. This marketplace seamlessly connects PPP stakeholders with globally competent innovators and multidisciplinary domain knowledge experts. Thus, it significantly leverages the competencies of stakeholders and promotes the implementation of efficient and innovative PPP projects. Hence, i3PM encourages the initiation of PPP projects with relatively higher maturity and efficiently satisfies the ultimate PPP goals. Appendix C explains how each model contributes to the implementation of hybrid platform economies.
2. PPP Engagement Models
Most PPP engagement models such as Build–Operate–Transfer (BOT) are generically and non-rigorously defined. They are complex in nature and are characterized by extreme rigidity, fuzziness, and inefficiency (Cruz et al. 2015; Umar et al. 2018). The implementation of intelligent marketplaces necessitates the standardization and computational representation of engagement models. This cultivates collective intelligence to develop agile, adoptable, and extendable standard models by engaging regulators, domain experts, and other competent stakeholders. The entire interaction is seamlessly facilitated and mediated by the market economies and collective intelligence to foster mutual collaboration. These standard models can be represented using international standards such as W3C standard ontology web language (OWL) (Trang 2010). This computational representation inevitably promotes PPP engagement automation, optimization, rigorous validation, and competency development via the embedded knowledge within the standards. These models can easily be tailored and extended based on the constraints imposed by specific PPP projects to stimulate innovation and efficiently achieve the ultimate goals. Table A1 (Appendix A) lists some of the proposed PPP engagement models.
3. Tradable PPP Partitions (TPPs)
PPP projects in i3PM are tradable, and the atomic tradable unit is called a TPP. These are identical to tokenized securities that define the ownership and rights of a shareholder in a PPP project or in an entity created via a PPP initiative. TPP certificates are represented digitally and are stored within a tamper-free blockchain infrastructure.
These partitions are traded on the hybrid marketplace (Smith et al. 2019; Kaal and Evans 2019; Mass 2019; Le Gear 2020). TPPs are well regulated and are distinct from Initial Coin Offerings (ICO) (Mass 2019). Their function is to digitally represent PPP securities certificates in a flexible, secure, and efficient format to promote seamless trading of these assets via the i3PM environment.
4. Mutual Value Development
The main objective of this marketplace is to create mutual and balanced value for all stakeholders. This is accomplished through the hybrid platform economies that automate the coordination and negotiation processes to promote the rigorous specification and optimization of stakeholder benefits, which are codified using computation ontology. The entire mutual benefit development lifecycle is managed by a negotiation engine that utilizes computational negotiation models, an intelligent bargaining engine, distributed multiagents, computational Nash equilibrium models, operations research and artificial intelligence (AI) based optimization models. Robo-advisors are used to simplify the interaction and optimize the experience of stakeholders by abstracting technical and deep knowledge required to codify and optimize their returns (Dong and Li 2011; Rangaswamy and Shell 1994).
5. PPP Process Map
i3PM projects are required to generate unique returns for stakeholders to symbiotically attract investors, public and private sector entities, consumers, regulators, and other stakeholders. As such, this intelligent marketplace manages the entire project lifecycle, from pre-inception to operation, thereby assuring the development of mutual benefits, and the execution of PPP projects with high transparency, efficiency, and optimal ROI.
In this policy brief, we propose the development of a standard process map through global collaboration. This map encompasses the best practices developed by groups of multidisciplinary domain experts, regulators, and public and private sector stakeholders. These processes are expected to be represented in a digital format to promote seamless tailoring and automation (Trang 2010).
6. High Fidelity Information Models
To handle information asymmetry, we propose the adoption of a high-fidelity information model (HFIM) that captures relevant structural, and behavioral information about an enterprise (Peri and Campana 2003; Lee et al. 2019). This includes in-depth knowledge of its supply chain, revenue streams, and product and consumer models, among others. Figure A3 (Appendix A) depicts the topology of the proposed 32-dimensional information model. These models are represented in machine comprehensible format as a computational ontology (Trang 2010). Each sector has its own HFIM schema developed by its domain experts, and other relevant stakeholders. This schema is used to elicit high-fidelity information about a project or an enterprise.
The purpose of HFIMs can be summarized as follows:
Promotes the elicitation of high-quality PPP engagement information with minimum time, cost, and effort.
Resolves daunting information asymmetry impediments to promote high transparency and improves market efficiency.
Serves as domain knowledge represented in digital format.
Promotes the development of sector or industry specific benchmarking, rating, and scoring models.
7. Smart PPP Contracts (S3PCs)
The purpose of Smart PPP Contracts (S3PCs) is to resolve critical PPP contractual pitfalls and anomalies such as information asymmetry, contractual ambiguity, incompleteness, and inconsistency (Berner et al. 2014; Chohra et al. 2011; Cruz et al. 2015; Umar et al. 2018). S3PCs resolve these by employing an efficient PPP contract designing methodology based on hybrid and artificial intelligence models (Pourshahid et al. 2009; Liiva et al. 2018; Wei 2019; Cath 2018; Pennings and Leuthold 2000). Most of these anomalies are induced during the inception of the contract (Viljanen 2020; Chohra et al. 2011; Cruz et al. 2015). Consequently, this methodology resolves ambiguity via rigorous computational representations, debiasing behavioral anomalies via plausible reasoning, and enforces contractual completeness and consistency via model-based validation (Pourshahid et al. 2009; Liiva et al. 2018; Wei 2019). Furthermore, this facilitates the design of resilient contracts by stress testing them under extreme conditions such as during a financial crisis, with their contents represented in a digital format using computational ontologies (Kruijff and Weigand 2017a, 2017b). Once the contract is validated and approved by the stakeholders, it is deployed as a smart contract into a blockchain infrastructure. Smart contracts enforce contractual reliability, regulatory compliance, security, optimal arbitration, fraud management, and exception handling (Tai 2017; Caso et al. 2012).
Since the development of such contracts from scratch can be time consuming, we propose the origination of reusable contracts called PPP contract templates (P3CTs) by subject matter experts and relevant stakeholders. These templates can be created using the collective intelligence environment supported by i3PM. P3CTs are extendable by nature. Hence, they can easily be tailored and adapted according to the constraints imposed by a specific PPP engagement.
8. Capital Guarantee Models
Financial guarantees in PPP projects play a pivotal role in promoting their attractiveness to diverse private sector stakeholders. However, most are inflexible and misaligned with performance, thereby creating information asymmetry and offering inadequate incentives to private sector stakeholders (Albalate et al. 2015; Athias and Saussier 2007; Burger and Hawkesworth 2011).
To efficiently address these issues and appeal to diverse stakeholders, we propose the development of rigorous capital guarantee models designed using an engineering tradeoff approach that holistically addresses the mutual benefits, constraints, and risks to stakeholders (Daniels et al. 2001). Stakeholders employ a workshop analysis approach to elicit the spatiotemporal constraints and risks of the contract, while designing the capital guarantee model to efficiently resolve these issues (AntolínDíaz et al. 2020; Barbacci et al. 2020). In the context of the i3PM, this is referred to as the optimization of the return on capital guarantee (RoCG). This process also employs mathematical and AI models to help in traversing the solution space and generates optimal solutions and RoCG (Peri and Campana 2003). Table A2 (Appendix A) lists sample capital guarantee models developed using this framework.
This marketplace can easily attract the required critical mass because of relatively high information symmetry, flexible capital guarantee models, and expected high returns. The TPPs are incrementally offered through the marketplace, and at each stage, different guarantee models are used to handle risks that conform to the risk appetite of investors. This is expected to incrementally increase the value of the asset, thereby stimulating both supply and demand, which inevitably increases the liquidity of the market. Due to its high liquidity and scalability characteristics, this marketplace efficiently attracts sufficient domestic and foreign investments.
9. Holistic Risk Management
PPP projects are complex in nature as are the risks associated with them. This complexity arises from numerous factors and their causal interactions (Chan, Osei-Kyei, et al. 2018; Chan, Yeung, et al. 2011). Most PPP projects do not efficiently enumerate these factors due to information asymmetry. In an i3PM, an HFIM spanning 32 dimensions is employed. These provide high-resolution information about the supply chains, revenue streams, product, microeconomics, and project models, among others (Peri and Campana 2003; Lee et al. 2019). Figure A3 (Appendix A) illustrates the proposed model. By ultimately optimizing information symmetry, the risk parameters can be efficiently extracted and defined with high levels of accuracy, credibility, and fidelity (Waring 2003; Soliwoda et al. 2018). Furthermore, the causal interaction between these risk parameters can be defined using computational models such as system dynamics, and Bayesian models (Wang et al. 2013; Milling 2006; Struben et al. 2015). The rationale of using these computational models is to overcome the limitations of human cognition in dealing with complex causal relationships (Yao 2017; Tóbiás 2020; Rescher 1997), which otherwise result in cognitive limitations and inefficiencies in addressing the salient interactions within PPP projects.
Once the risk attributes are efficiently captured, a scenario-based analysis is used to identify, communicate, evaluate, prioritize, and define risk mitigation plans (AntolínDíaz et al. 2020; Barbacci et al. 2020). The entire risk management process is derived from a continuous risk management framework (Alberts et al. 1996).
10. Crisis Resilience Models
It is clear that PPP projects are vulnerable to exceptional events such as financial or public health crises (including the coronavirus pandemic), which inevitably have serious impacts that compromise the benefits of PPP projects, and in some extreme cases, lead to their termination (Coelho et al. 2009; Hertati et al. 2002). The latest economic crisis precipitated by the coronavirus pandemic has taught us a severe lesson: it is impracticable to react to these crises when they transpire because their impacts are rapidly and chaotically propagated. Short-term interventions generally fail to absorb or deter this chaotic propagation. Yet, we need not reinvent the wheel, as we can adopt the proactive style that is currently employed by the financial services industry. An example is the BASEL III accord, which requires financial institutions to proactively deal with unexpected exceptional events by developing a multitude of proactive measures (McNamara et al. 2014; Demirguc-Kunt and Detragiache 2005). Therefore, the aim of embedded resilience models (ERM) is to analyze diverse exceptional scenarios during the inception of PPP projects and proactively formulate interventions to deter their occurrence or absorb their adverse impacts.
The purpose of these models is to proactively deal with PPP engagement exceptions and risks during crises using multidisciplinary resilience and reliability management models (Soroka et al. 2019; Bristow and Healy 2017; Hermansen and Roehn 2017; Sensier et al. 2016; Sunley and Martin 2014). To efficiently implement this, we propose an iterative approach called Proactive Resilience Lifecycle Management (PRLM). This constitutes Exception Event Scenario Analysis, Potential Event Enumeration, Resilience Model Development, Resilience Model Validation, Continuous Exception Surveillance, and Resilience Model Optimization. Appendix B presents a more detailed explanation of this iterative lifecycle.
To promote the adoption of an intelligent PPP marketplace that can resolve formidable PPP impediments, we recommend the following main actions:
Extendable PPP marketplace architecture
We call on G20 policy makers and multilateral financial institutions to facilitate the development of an open and extendable PPP marketplace by incentivizing global stakeholders to collaborate in extending the i3PM pillars. The proposed incentives include:
The adoption of a PPP marketplace by governments and multilateral institutions in the implementation of their PPP projects.
The formal standardization of the architecture of the PPP platform economies by government and other credible standard bodies.
Funding multidisciplinary academic research aimed at extending and optimizing PPP platform economies.
Standardization of PPP contracts
We recommend that G20 governments, multilateral financial institutions, and standard bodies standardize the process, structure, and representation of PPP contracts to handle contractual complexity, ambiguity, risk, and high transaction costs. These standards should include:
The development of an optimal and rigorous contract designing and validation methodology by efficiently extending the smart PPP contract methods employed in i3PM.
The implementation of PPP contracts using smart contracts.
The implementation of reusable domain-specific PPP contract templates that can easily be tailored and extended. This will minimize the time, cost, effort, and risk to develop PPP contracts from scratch.