Reliability as a common-pool resource

Transmission grid reliability is best understood as a CPR rather than a pure public good, due to its physical and economic characteristics.Footnote ¹ Unlike a public good – non-excludable and non-rival – a CPR is non-excludable but rivalrous: congestion diminishes availability for others.

CPR dilemmas fall into two types: appropriation and timing/spacing. In appropriation, users deplete a finite stock (e.g., fisheries (Gordon, Reference Gordon1954), oil fields (Libecap and Wiggins, Reference Libecap and Wiggins1985; Wiggins and Libecap, Reference Wiggins and Libecap1985)). In timing/spacing dilemmas, value depends on coordinated use; uncoordinated actions impose congestion or interference costs. Examples include radio spectrum (Coase, Reference Coase1959), irrigation (Ostrom, Reference Ostrom1992), and electric grids (Blomkvist and Larsson, Reference Blomkvist and Larsson2013).

The electric grid is non-excludable, since connected users cannot easily be denied access, and rivalrous, since finite capacity makes it congestible. In an AC grid, Kirchhoff’s Laws and Ohm’s Law mean that electricity flows along all available paths, preventing allocation through contracts alone. Translating from physics to economics, power flow in an AC system necessarily has ill-defined property rights.

This environment of ill-defined property rights creates three governance challenges: shared vulnerability to congestion, non-excludability, and the need for collective management. RTOs address these challenges institutionally, through reliability standards, loop flow management, and congestion mitigation.Footnote ²

Reliability may appear public good-like under normal conditions; benefits are non-rival when the grid is stable. But under congestion, it becomes rivalrous: increased use constrains others’ access, revealing its CPR nature. The classification also depends on perspective. For RTO participants – transmission owners, generators, buyers – reliability is rival and interdependent. For end users, it resembles a public good, especially when grid stability is ensured through centralized coordination or private backup systems.Footnote ³

Viewing reliability as a CPR offers a more realistic governance framework than treating it as a pure public good. It highlights the externalities participants impose on each other and supports mechanisms to internalize them, such as transmission rights, capacity limits, and enforcement. RTOs function as self-organized institutions that govern access, coordinate investment, and preserve reliability in an increasingly complex and decentralized grid.

Ostrom’s design principles

Elinor Ostrom’s work on CPR governance highlights the importance of user-developed rules to mitigate the risk of resource depletion (or in this context, degraded reliability due to a lack of coordination).Footnote ⁴ Ostrom found that voluntary, self-organized institutions can govern CPRs effectively if they establish:

1. 1. Clearly-defined boundaries: Participants know what constitutes the resource system and who is part of the resource system, and access is controlled.

2. 2. Proportional equivalence between benefits and costs. Rules should ensure that those who benefit from resource use bear the costs or responsibilities commensurate with those benefits.

3. 3. Collective choice arrangements: Stakeholders directly participate in rulemaking that governs the resource.

4. 4. Monitoring: Users and/or third parties monitor one another’s behaviour.

5. 5. Graduated sanctions: Rule violations incur penalties calibrated to the seriousness of the offense.

6. 6. Conflict-resolution mechanisms: Accessible dispute-resolution processes manage conflicts

7. 7. Minimal recognition of rights: Government or other external authorities recognize the legitimacy of user-developed institutions.

8. 8. Nested enterprises: Multiple layers of governance structures build upon smaller units of local governance as appropriate to handle complexity. (Ostrom, Reference Ostrom2005: 259)

A key dimension of Ostrom’s framework is polycentric governance, where multiple decision-making centres operate autonomously yet interact at different levels – local, regional, and national. This structure emerges in complex resource systems where diverse stakeholders coordinate across overlapping jurisdictions. The concept originated in Vincent Ostrom’s work with Tiebout and Warren (Ostrom et al., Reference Ostrom, Tiebout and Warren1961) and was further developed by Elinor Ostrom (Ostrom, Reference Ostrom2005, Chapter 9). RTO governance reflects both external and internal polycentricity (to be analysed in Section Polycentricity in RTO governance).

For the transmission grid, these principles are embodied in rules for allocating scarce transmission capacity, ensuring real-time operational coordination for balancing supply and demand, managing congestion, procuring ancillary services, and assigning the costs of maintaining and enhancing reliability services. These rules are implemented through the RTO’s governance structure, which is typically approved by the Federal Energy Regulatory Commission (FERC). The RTO becomes the institutional ‘hub’ that integrates Ostrom’s CPR principles into operational practice by setting detailed membership and market participation requirements and credit policies, establishing sophisticated real-time monitoring of system flows and market operations, applying a schedule of sanctions for non-compliance with reliability standards or market rules, and managing conflicts among diverse members through established dispute resolution procedures.

Theory of clubs

Buchanan’s theory of clubs (Buchanan, Reference Buchanan1965) fills a gap in neoclassical public goods theory by addressing goods that fall between purely private and purely public. He defined clubs as groups that jointly consume goods or services, where costs vary with membership size. Unlike Samuelson’s model of pure public goods – with theoretically infinite optimal group size (Samuelson, Reference Samuelson1954) – Buchanan focused on identifying the optimal size that balances marginal costs and benefits.

In Buchanan’s formal setup, an individual’s utility depends on the consumption of a private numeraire good and on access to a shared facility whose service quality declines with congestion as membership rises. Let utility for a representative member be $U\left( {x,s\left( N \right)} \right)$ , where $x$ is private consumption and $s\left( N \right)$ is the service flow from the club good. $s\left( N \right)$ is decreasing in membership $N$ because crowding reduces each member’s enjoyment. Providing the facility entails a cost $C\left( N \right)$ , which can include fixed costs and member-dependent operating costs, and these costs are shared among members through dues or a fee schedule. The club’s problem is to choose membership $N$ (and implicitly the fee and facility scale) to maximize representative net benefit, trading off the marginal gain from spreading fixed costs over more members against the marginal congestion cost imposed by an additional member. The optimal club size N* occurs where the marginal benefit of adding a member (lower per-member cost or expanded provision) equals the marginal congestion damage, yielding Buchanan’s condition that the marginal rate of substitution in consumption equals the marginal rate of transformation in production for the club good.

This framework launched a broad literature (e.g., Anderson et al., Reference Anderson, Shughart, Tollison, Rowley and Schneider2004; Cornes and Sandler, Reference Cornes and Sandler1986; Marciano, Reference Marciano2021; Sandler and Tschirhart, Reference Cramer, Tschirhart and Sandler1980, Reference Sandler and Tschirhart1997; Sandler, Reference Sandler2013) that refines our understanding of collective consumption and partial publicness. Club goods are partially excludable and rivalrous, and the optimal club size arises when the marginal rate of substitution in consumption equals that in production (Buchanan, Reference Buchanan1965: 5). The literature explores how these properties shape provision, pricing, and incentives to join or form clubs.

Club theory has clear policy relevance where governments provide goods that could be managed more efficiently by voluntary groups, such as parks, highways, and other local public goods. It is especially applicable to goods with partial publicness, where sharing arrangements fall between purely private and purely public provision.

At its core, club theory is a theory of optimal exclusion. As Buchanan observed, few goods are inherently non-excludable due to physical constraints alone (Buchanan, Reference Buchanan1965: 13–14). Where exclusion is feasible, flexible property rights and enforcement mechanisms (e.g., prosecuting poachers in a hunting preserve) can reduce free riding and facilitate voluntary cooperation.

Buchanan’s model assumes replicability: any excluded individual can, in principle, form a new, identical club. This capability ensures competitive entry and prevents durable market power. When the population is an integer multiple of the optimal club size ${N^{\rm{*}}}$ , society naturally partitions into equal-sized clubs. Each club sets entry fees precisely high enough to cover operating costs, eliminating durable market power since dissatisfied members can always establish a new club. Buchanan thus characterizes this approach explicitly as a ‘theory of optimal exclusion’ suited specifically for goods where physical exclusion is practical.

However, this assumption breaks down in the context of electric power grids. The physical and economic characteristics of transmission networks make replication infeasible. Yet this lack of replicability does not invalidate club theory’s relevance for RTOs. Instead, it reshapes how exclusion operates, a topic explored further in Section RTOs as clubs.

Emergence and core function of RTOs

The evolution of RTOs and wholesale power markets traces back to early inter-utility agreements for interconnection and emergency backup, beginning in the late 1920s (Neufeld, Reference Neufeld2019). Investor-owned utilities formed power pools to coordinate bulk power operations, centralizing dispatch (the decision of when to use which generators), spinning reserves, emergency power supply, and maintenance scheduling. These arrangements improved generation efficiency through economies of scale, enhanced reliability, reduced energy prices, and provided resource adequacy while lowering environmental costs (Cramer and Tschirhart, Reference Cramer, Tschirhart and Sandler1980: 217–218). The first power pool, the Pennsylvania-New Jersey-Maryland Interconnection (PJM), was established in 1927, and additional power pools emerged after World War II that later evolved into today’s RTOs (Cramer and Tschirhart, Reference Cramer, Tschirhart and Sandler1980: 216).Footnote ⁵

The turbulent energy environment of the 1970s set the stage for power pools to evolve into competitive wholesale markets. The U.S. electricity industry embarked on a significant transformation in the late 20th century, moving from a structure dominated by vertically integrated utility monopolies towards the establishment of competitive wholesale power markets (Borenstein and Bushnell, Reference Borenstein and Bushnell2015). Several factors propelled this shift, including the aim for increased economic efficiency and enabling federal policy changes, such as the Public Utility Regulatory Policies Act of 1978 (PURPA) and the Energy Policy Act of 1992 (EPAct). PURPA unintentionally stimulated competition by requiring utilities to purchase power from independent qualifying facilities, while EPAct reduced barriers to wholesale competition by allowing non-utility wholesale generators (Hirsh, Reference Hirsh1999). Technological advancements, particularly in combined-cycle gas turbine technology, also played a crucial role by making smaller-scale generation economically viable and challenging the natural monopoly model (Joskow, Reference Joskow2008).

In the 1990s, independent power producers gained greater access to expanding wholesale markets, while regulated utilities were required to maintain operational and financial separation from their unregulated affiliates.Footnote ⁶ EPAct promoted wholesale competition, leaving retail restructuring to the states. Fifteen states and the District of Columbia enacted legislation requiring utilities to divest generation assets and support retail competition.Footnote ⁷ Some states retained regulated monopolies in transmission and distribution while introducing competition in generation; others adopted full retail choice. Restructuring, however, faced significant challenges, including market manipulation, price volatility, and weak regulatory oversight. The California electricity crisis of 2000–2001 revealed the dangers of poor market design and market power abuse (Joskow, Reference Joskow2008; Wolak, Reference Wolak, Griffin and Puller2005).

At the core of successful competition was non-discriminatory transmission access, which was difficult to achieve when incumbent utilities continued to operate most transmission assets. These conflicts of interest spurred the formation of RTOs, which aimed to separate transmission operations from generation ownership and ensure open access for all market participants. This move marked a structural shift from vertically integrated monopolies towards regionally coordinated, market-based oversight (Joskow, Reference Joskow2005: 106).

RTOs emerged as independent entities with two core responsibilities: maintaining transmission reliability and operating wholesale power markets. The FERC regulates most RTOs to ensure transmission rates and market outcomes remain ‘just and reasonable’.Footnote ⁸ The original intent of both RTO organizers and FERC was for RTOs to manage the regional grid independently and impartially, a critical function underpinning all market activity (Joskow, Reference Joskow2005: 103–104), although that independent impartiality is not always realized in practice (Peskoe, Reference Peskoe2023). FERC Order 888 (1996) laid the foundation for open access by mandating functional unbundling and non-discriminatory transmission tariffs (Lenhart and Fox, Reference Lenhart and Fox2022, Table 1).

Alongside reliability, RTOs administer competitive wholesale markets, including day-ahead and real-time energy markets and ancillary services. Participation is governed by defined rules and protocols. FERC Order 2000 (1999) encouraged voluntary RTO formation and outlined features such as independence from market participants and operational control over transmission.

An important link between reliability and market efficiency is security-constrained economic dispatch (SCED), which uses uniform-price auctions to optimize generation dispatch while maintaining system security. This auction mechanism promotes truthful bidding and efficient outcomes.

Transforming legacy power pools into RTOs required significant changes to organizational structure and oversight, guided by FERC Order 2000. Existing coordination efforts helped ease the transition but also created institutional path dependencies that continue to shape RTO governance. By formalizing both market operations and grid management, RTOs have enabled competitive wholesale markets while preserving reliability.

Since the mid-1990s, RTOs and ISOs have expanded and evolved. Figure 1 shows their geographic boundaries as of 2022. PJM, MISO, and SPP span partial-state territories due to differing utility participation. PJM and MISO expanded as additional utilities joined.Footnote ⁹ SPP’s members remain vertically integrated and state-regulated, while MISO includes restructured states like Illinois and Michigan. The other ISOs are composed of utilities in restructured states.Footnote ¹⁰

Figure 1.

ISO/RTO territories in the United States, 2022 (Federal Energy Regulatory Commission, 2020).

Each RTO developed governance structures defining membership, voting rights, and stakeholder participation, governance structures that have some common features across the seven RTOs in the US but also reflect regional variations in economic activity, fuel portfolios, geography, and climate. These institutions were designed for independence, transparency, and representation, but now face pressure from accelerating technological change.

The challenge of a shifting landscape: the pacing problem

The evolution of the electricity industry reveals a dynamic interplay between technological innovation and regulatory adaptation. The context that gave rise to RTOs in the 1990s has shifted dramatically in the 21st century, creating a ‘pacing problem’ in which institutional change struggles to keep up with rapid technological and policy developments (Hagemann et al. Reference Hagemann, Skees and Thierer2018; Kiesling, Reference Kiesling2024). RTOs emerged during a period shaped by centralized, fossil-fuelled generation, and their design reflects that era’s technological and policy assumptions.

Today’s power systems look markedly different. The rise of variable renewable resources, such as wind and solar PV, has introduced significant supply-side variability and reshaped market dynamics. These technologies, with near-zero marginal costs, alter price formation and dispatch. Wind capacity in the U.S. grew from 2.4 GW in 2000 to over 150 GW by April 2024 – a 62-fold increase – driven by technological progress, declining costs, and policy support (U.S. Energy Information Administration, 2024). Solar power followed, with nearly 50 GW installed in 2024 alone, up 21% from 2023, and continued growth projected in 2025 and 2026 (U.S. Energy Information Administration, 2025a). Large-scale storage, crucial for resilience and for integrating renewables, reached 62 GW in 2024 and is expected to expand by nearly 20 GW in 2025 (U.S. Energy Information Administration, 2025b).

At the distribution level, DERs are proliferating. Rooftop solar PV installations grew 11% annually between 2010 and 2020, reaching 2.7 million by 2022, especially in states with favourable climates and policies (Solar Energy Industries Association, 2023). U.S. EV sales rose from 115,000 in 2014 to over 755,000 in 2022, supported by battery improvements (International Energy Agency, 2023). Meanwhile, battery storage – both grid-scale and behind-the-metre – has expanded system flexibility and enabled greater resilience and renewable integration.

These decentralized resources reflect a shift away from centralized generation. Their aggregation for wholesale market participation raises new challenges in coordination, visibility, and data management. Advanced sensors, automation, and analytics offer tools for managing this complexity, but realizing their value often requires changes to market rules and data-sharing protocols. FERC Order 2222 (2020) mandates RTOs to create pathways for DERs to participate in wholesale markets, but implementation remains slow and uneven (Walton, Reference Walton2023).

These technological shifts undermine the economies of scale that once justified vertically-integrated monopolies and expose frictions in legacy governance institutions, particularly in the misalignment of market participation rules. Decentralized, modular resources with distinct cost and operational profiles do not fit neatly into existing interconnection, market participation, and cost allocation frameworks; they strain legacy governance institutions.

Policy goals have also evolved. Reliability and affordability remain core concerns, but decarbonization and resilience have become prominent policy drivers. Federal and state initiatives, such as renewable portfolio standards, tax incentives, and emissions reduction targets, are accelerating the deployment of clean energy (Barbose, Reference Barbose2019). At the same time, resilience – a system’s capacity to endure and recover from disruptions – is now a key focus amid rising risks from extreme weather, wildfires, and other threats. Together, these shifting priorities place new demands on RTO planning, operations, and governance.

The resulting strain on RTO governance

The convergence of rapid technological change and evolving policy goals is straining RTO governance structures, market rules, and operational practices – many of which were designed for a centralized, fossil-fuelled system. This mismatch creates a pacing problem, in which deliberative, stakeholder-driven institutions struggle to adapt complex rule sets at the speed of external change.

Technologies like DERs, utility-scale renewables, and storage challenge legacy governance by introducing variable, decentralized resources that do not align with conventional interconnection processes or resource adequacy metrics. Maintaining reliability under these conditions often requires new protocols for integration, monitoring, and dispatch – protocols that lag the pace of technological change. Lenhart and Fox (Reference Lenhart and Fox2022) showed how stakeholder processes are strained by these shifts, while Johnson et al. (Reference Johnson, Lenhart and Blumsack2023) highlighted how diverging state policies within RTO footprints amplify institutional stress. This widening gap between clean-energy mandates and outdated governance models underscores the need for institutional reform to ensure system reliability and market efficiency.

Lenhart and Fox (Reference Lenhart and Fox2022) argued further that greater transparency and public-interest accountability would better serve the growing diversity of stakeholders – DER owners, consumers, and environmental groups – now shaping electricity policy agendas. Johnson et al. (Reference Johnson, Lenhart and Blumsack2023) emphasized how expanding stakeholder participation and complex policy contexts strain consensus-based decision-making. Welton (Reference Welton2021) critiqued legacy, industry-dominated governance structures for lacking public accountability, potentially distorting decisions around renewable integration. Collectively, these studies reveal how legacy frameworks are increasingly misaligned with the demands of a more diverse, decentralized, and policy-driven electricity system.

These pressures manifest in rising stakeholder complexity, administrative burdens, and accountability challenges. Governance structures built in the 1990s assumed centralized base-load generation and minimal DER presence. Today’s policy landscape emphasizes distributed systems, demanding major reforms in interconnection, adequacy metrics, and cost allocation. The dual mandate of reliable operations and efficient markets is being tested by the volume and diversity of new market participants enabled by technology and motivated by policy.

One of the clearest indicators of this strain is the interconnection queue. All RTOs maintain queues to track generation and storage projects seeking grid access. Because interconnection involves extensive technical and economic studies to preserve grid reliability, the process can be slow and costly. With a surge in renewable and storage proposals, queues have grown dramatically – totalling 2,600 GW of capacity in 2024, more than twice the current U.S. generation fleet, as seen in Figure 2. Many of these requests are speculative, and aligning requests with realistic demand growth has become a complex exercise in managing uncertainty. Since 2014, new capacity entering queues has increased annually, with solar, wind, and storage now comprising 95% of requests. Projects often spend 14 to 43 months in the queue, and mean development timelines are nearing five years (Rand et al., Reference Rand, Manderlink, Gorman, Wiser, Seel, Kemp, Jeong and Kahrl2024).

Figure 2.

Active U.S. Interconnection queue capacity, 2014–2023 (Rand et al., Reference Rand, Manderlink, Gorman, Wiser, Seel, Kemp, Jeong and Kahrl2024).

FERC and RTOs recognize these governance challenges, but institutional reform in such complex systems is inherently slow. The growing mismatch between emerging technologies, policy drivers, and existing institutions points to an urgent need for reform in governance, market design, and grid operations – reform that can support decentralization while preserving core goals of reliability, cost recovery, and sustainability.

This challenge sets the stage for analysing RTO governance through the institutional lenses of Ostrom and Buchanan, with particular attention to the role of exclusion and its implications for adapting to innovation and a changing electricity system.

RTOs as clubs

Club theory offers a valuable framework for understanding how RTOs govern the congestible CPR of transmission reliability by layering structured membership and exclusion. Historically rooted in power pools, the first ‘reliability clubs’, RTOs transform grid access from a CPR into a club good by setting tariffs, interconnection standards, and participation rules. Transmission and market services are non-rival until congestion arises, but are rendered excludable by rules: only entities that meet technical, financial, and operational criteria may participate. This transformation mitigates free riding and manages reliability risks (Welton, Reference Welton2021).

At the heart of this governance model is a tension between physical reality and institutional discretion. AC power flows follow Kirchhoff’s and Ohm’s laws: electrons do not obey contracts, and every injection or withdrawal affects the entire network. These physics make reliability non-excludable in practice much of the time but rivalrous under stress.Footnote ¹¹

This observation implies a narrower and more precise institutional claim than the shorthand ‘RTOs turn reliability from a CPR into a club good’. The reliability commons exists in both RTO and non-RTO regions; what differs is the governance technology used to manage it and to define who may use the system and under what obligations. In traditionally regulated, non-RTO regions, vertically integrated utilities largely ‘govern the commons’ through internal hierarchy, backed by state regulation: the same organization plans and operates generation and transmission within its territory, bears residual balancing responsibility, and relies on relatively limited, often bilateral interchange and emergency support arrangements at the seams with other transmission systems. In RTO regions, open-access transmission and market restructuring expand the number and heterogeneity of actors interacting through the network, and those interactions become routine rather than exceptional. That shift increases the need for multilateral rules that specify participation conditions, performance obligations, and cost responsibility.

From this perspective, RTOs do not make electrons excludable. Instead, they formalize and scale exclusion in the institutional sense by bundling access to centralized dispatch, standardized market participation, and settlement systems with enforceable reliability obligations. These rule systems function as a club-like governance regime that allows a larger membership to share a congestible network while limiting free riding and managing congestion and reliability risk.Footnote ¹²

RTOs impose legal and economic order onto this system. While they cannot change the physics, they apply excludability strategically – not to the physical flow of power, but to market access, risk management, and governance. FERC’s open-access transmission tariff requirements ensure non-discriminatory physical access. Participation rights like scheduling, hedging, and governance are reserved for members and market participants who meet credit, telemetry, and performance standards.Footnote ¹³ Congestion pricing captures rivalry by assigning marginal costs at the point of transaction. Reliability standards, however, remain universally binding.

Yet exclusion carries risks. Membership rules designed for large, centralized generators can turn RTOs into gated communities, deterring participation by DER aggregators, storage, and flexible demand. These actors still affect system conditions as market participants but lack equivalent access to decision-making. The policy challenge is to lower unnecessary barriers, such as scale-insensitive collateral or telemetry requirements, while maintaining the institutional mechanisms that support shared infrastructure.

RTOs operate along a CPR–club continuum, using exclusion to enhance governance and efficiency while keeping the physical fabric of power flows and the duty to maintain reliability squarely in the commons.

RTO governance reflects Buchanan’s club model in its use of exclusion, cost-sharing, and collective decision-making to manage reliability and market access. Entry rules prevent free riding; cost-allocation mechanisms distribute the burden of infrastructure and market administration. Investment decisions mirror club logic, as members weigh the costs and benefits of upgrades. Larger RTOs gain economies of scale and risk pooling, but growing and diversifying membership raises coordination costs, reinforcing the need for adaptable governance.

Like clubs, RTOs must balance static efficiency considerations of minimizing operational costs with the dynamic efficiency of accommodating technological and policy change. While exclusion supports reliability, it can stifle innovation if not calibrated to evolving conditions. Polycentric governance, involving RTOs, states, and local actors, helps recalibrate rules, internalize externalities, and reduce transaction costs through centralized operations.

RTOs do differ, though, from Buchanan’s model in one important respect: they are not replicable. High fixed costs, network externalities, and economies of scale in dispatch make it infeasible for excluded parties to form competing clubs. Without the exit option, exclusion becomes more consequential as a governance mechanism. RTO tariff structures, queueing procedures, and governance rules determine who can inject or withdraw power, on what terms, and using which technologies. This structure creates economic rents and strong incentives for incumbents to shape rules to their advantage, while raising participation barriers for new entrants.

These conditions do not invalidate club theory but require its adaptation. RTOs remain organized platforms that structure access to a shared resource. The definition of market prices (locational marginal pricing) reflects Buchanan’s principle of charging users for marginal congestion costs, while cost allocation parallels his distinction between membership fees and tolls. Yet in the absence of replicability, fair access and innovation must be safeguarded through internal institutional design and external oversight. For those reasons, Ostrom’s design principles and model of polycentric governance are crucial to understanding RTO governance and its adaptation to technological change.

FERC’s open-access rules, state policy initiatives, and independent market monitoring function as Ostromian polycentric governance mechanisms, compensating further for the lack of club competition. Buchanan explains why exclusion is necessary; Ostrom shows how it can be managed. Together, they provide a robust framework for RTO governance.

Ultimately, the challenge in this setting is not optimizing club size but ensuring that exclusion does not entrench incumbency or obstruct technological evolution. RTOs must balance short-term reliability with long-term adaptability, aligning governance with emerging technologies and shifting policy goals to support both system stability and innovation.

Ostrom’s design principles applied to RTOs

Ostrom’s design principles provide a useful lens for characterizing how RTOs function as a club to govern the transmission reliability commons while also operating market platforms. In this paper, exclusion refers to the bundle of rule-defined participation rights and obligations that governs access to a congestible network and to the market platform built on top of it. Technological change – particularly the growth of renewables, storage, and DERs – stresses this exclusion regime by expanding the set of feasible participants and increasing heterogeneity in capabilities, observability, and performance risk. Rather than treating each design principle as a separate checklist item, this section organizes them around four governance functions that are central in RTOs: (i) boundary-setting and proportionality (Principles 1–2), (ii) collective choice and legitimacy (Principle 3), (iii) monitoring, enforcement, and dispute resolution (Principles 4–6), and (iv) external recognition and nesting across governance scales (Principles 7–8). This organization retains Ostrom’s analytical approach while highlighting the mechanisms through which RTOs implement exclusion and manage the exclusion–innovation trade-off under technological change.

Polycentricity in RTO Governance

RTO governance is usefully understood as polycentric because the exclusion regime that protects reliability and enables market exchange is not set by a single authority. Instead, multiple decision-making centres determine jointly who can participate, on what terms, and how quickly those boundary rules can change as technology evolves. This polycentric structure matters for the paper’s core argument because the exclusion–innovation trade-off is shaped not only by the content of rules, but also by the number of venues that can revise, contest, or block them. When the grid is a non-replicable network commons, this governance architecture becomes a central determinant of dynamic adaptation.

Polycentricity has vertical and horizontal dimensions. Vertically, authority is distributed across several levels. State institutions pursue procurement and policy objectives that affect the resource mix and the location of new investments, while RTOs administer market rules, interconnection processes, and centralized dispatch to maintain reliability and coordinate trade across a larger footprint. Federal regulatory oversight through FERC constrains and enables RTO rule changes through tariff approval and policy directives, and reliability standards operate as an additional constraint that limits the feasible set of market and operational designs. Internally, RTO decision-making is itself multi-centred: boards, stakeholder committees, and market monitoring functions occupy distinct roles in agenda setting, rule development, adjudication, and enforcement. This layered structure means that participation, representation, and formal authority do not always coincide, which can become especially salient when new technologies seek entry or when policy objectives diverge across states within an RTO.

Horizontally, polycentricity arises because RTOs and balancing authorities must coordinate across seams and across regional transmission interconnections (except ERCOT). Interregional transmission planning, congestion management across borders, and the compatibility of participation models for emerging resources all require coordination among adjacent governance ‘centers’ that are formally autonomous. This horizontal fragmentation matters for innovation because many of the benefits of new resources depend on scale – moving power across regions, sharing reserves, and expanding the effective scope of the market platform – yet the institutional ability to expand or harmonize rules often lags the technological and investment impetus.

This polycentric architecture has a dual effect on innovation. On one hand, it can facilitate experimentation and learning because different RTOs and states can pilot participation models, procurement approaches, and market products without mandating uniformity, and successful designs can diffuse through observation and emulation. Polycentricity also provides checks on unilateral decision-making through overlapping oversight and escalation pathways. On the other hand, polycentricity can raise the transaction costs of adaptation by creating multiple veto points. When entry by new technologies requires aligned changes in market rules, interconnection procedures, and compliance frameworks, disagreement at any one node can delay reform, amplifying the pacing problem and indirectly strengthening exclusion.

Two recurring patterns illustrate this mechanism in a way that is directly relevant to the exclusion–innovation trade-off. First, vertical conflict emerges when state policies accelerate renewable and DER deployment while existing RTO participation and performance rules remain calibrated to centralized resources; integrating new entrants then requires changes that pass through stakeholder processes, board approvals, and federal review, often under binding reliability constraints. Second, horizontal conflict emerges when scaling renewables and improving reliability depend on interregional transmission expansion and seam coordination. Where planning and cost allocation cannot be agreed across neighbouring regions, congestion and queue delays increase, and the system responds by tightening or effectively hardening boundary rules through procedural barriers and higher entry costs. In both cases, polycentricity clarifies why exclusion persists even when many actors endorse innovation in principle: the costs of revising the exclusion regime are distributed across venues, and the benefits are often unevenly realized across participants and jurisdictions.

Viewed through this lens, polycentricity adds explanatory power to the Ostrom–Buchanan framework by showing how the governance system that transforms a reliability commons into a club-like regime can simultaneously enable and impede dynamic adaptation. Ostrom’s CPR framework explains why multi-layered rule systems, monitoring, and enforcement are necessary for a shared, congestible network. The polycentricity lens reveals why adapting those rules under technological change is itself a collective-action problem across multiple, partially autonomous decision centres.

The trade-offs of exclusion

Synthesizing Ostrom’s institutional design principles with Buchanan’s theory of clubs suggests that exclusion is the linchpin of RTO governance, but it also clarifies why the exclusion–innovation trade-off is unusually sharp in transmission systems. Exclusion in RTOs performs an essential static-efficiency function: by conditioning access on enforceable obligations, it limits free riding, manages congestion, and protects reliability in a congestible network where individual actions impose system-wide externalities. Yet exclusion also creates dynamic-efficiency costs because it restricts entry and contestability, the central processes through which innovation typically occurs. This tension becomes deeper once the transmission system’s non-replicability is taken seriously. In Buchanan’s canonical club setting, inefficient or stagnant clubs can be disciplined by exit and replication: dissatisfied members can form or join alternative clubs, and competition among clubs limits the exercise of exclusionary power. Transmission networks are constrained away from this competitive margin. High fixed costs, physical network effects, siting and permitting constraints, and the institutional realities of cost recovery mean that ‘building another club’ is rarely a feasible form of discipline. As a result, governance rules and the broader polycentric institutional environment must supply much of the disciplining force that replication and exit would otherwise provide.

Ostrom’s framework helps explain why this substitution is necessary. In a non-excludable-in-physics but congestible-in-practice network, rule-defined boundaries and obligations are the institutional technology that makes the reliability commons governable. RTOs have implemented this principle by specifying participation criteria, performance requirements, and compliance processes that enable centralized dispatch and coordinated planning across many heterogeneous users. Buchanan’s club theory then illuminates the risk of static membership and rigid boundary rules in a setting where replication is not feasible: the benefits of exclusion in producing an ‘optimal’ quantity of reliability and congestion management can come at the cost of Schumpeterian dynamism. When the primary margins of adaptation are institutional rather than competitive, a club that becomes procedurally rigid can lose the evolutionary advantages of entry by new participants with innovative business models, advanced digital tools, and DER capabilities, even when those entrants could lower system costs or improve flexibility.

This trade-off is particularly salient as the technological and policy environment shifts. Many RTO participation and market rules were established in the 1990s to support restructuring, open-access transmission, and investment in then-dominant technologies such as combined-cycle natural gas generation. Those institutional templates are now being stress-tested by a more heterogeneous portfolio of resources and services – intermittent renewables, storage, DER aggregations, and digitally mediated grid services – that create new reliability capabilities but also require new modes of measurement, verification, and coordination. Because the network cannot be replicated to accommodate competing institutional designs, adaptation must occur through rule revision within the existing governance architecture and through interactions across polycentric layers (RTO stakeholder processes and boards, FERC oversight, state policy, and reliability standards). In this environment, the exclusion problem is less about whether to exclude – some exclusion is unavoidable for reliability – and more about how to recalibrate boundary rules at the margin so that they preserve reliability while lowering unnecessary barriers to technologically valuable entry.

RTOs enforce exclusion through institutional mechanisms originally designed for large, centralized generators that can now function as barriers for newer technologies such as DER aggregators, demand response providers, and storage operators. Market participation rules often require substantial financial collateral,Footnote ¹⁴ creditworthiness standards, and extensive metering and telemetryFootnote ¹⁵ capabilities that can be cost-prohibitive for smaller or aggregated resources, limiting their ability to compete on equal footing with traditional generators. Capacity accreditation methods can disadvantage non-traditional resources by relying on historical performance metrics and conservative adjustment factors that do not fully reflect the reliability contributions of fast-ramping storage or aggregated demand response. Interconnection queues and related governance hurdles further constrain entry, as the backlog of pending generation and storage projects (notably in PJM, MISO, and CAISO) delays participation for years, often reflecting study processes and cost-allocation rules that assign a disproportionate share of network upgrade costs to later-arriving projects (Johnston et al., Reference Johnston, Liu and Yang2023). More broadly, Lenhart and Fox (Reference Lenhart and Fox2021) find that participation structures designed around legacy technologies and the initial objectives of restructuring are misaligned with newer participants and interests.

These exclusionary structures were originally designed to support reliability and market integrity, but under technological change, they can become institutional bottlenecks that impede the integration of distributed, flexible resources. Because the network cannot be replicated to provide an ‘exit’ option for excluded entrants, the fixed costs embedded in participation, accreditation, and interconnection rules can translate directly into higher minimum efficient scale for market participation and slower competitive experimentation. In effect, the same governance machinery that stabilizes a congestible reliability commons can harden into a procedural form of exclusion that limits the main margin of dynamic adaptation – entry – precisely under conditions when entry is most valuable.

This dilemma is a concrete manifestation of the pacing problem: when technological possibilities change faster than the institutions that define participation rights and obligations, the resulting mismatch produces congestion in governance as well as congestion on the grid. The institutional task is therefore not to eliminate exclusion, but to design exclusion that is appropriately contingent, evidence-based, and revisable – so that, in a non-replicable network club, polycentric governance and rule revision can supply the adaptive discipline that competition among clubs would otherwise provide.

An Ostrom-Buchanan approach to reconciling reliability and innovation

Effective RTO governance requires balancing congestion costs and reliability risk against the benefits of broader participation and technological variety. Buchanan’s club framework helps clarify the logic of this balance: exclusion is valuable because it limits congestion and free riding, but in a setting where the club cannot be replicated, exclusion also becomes a powerful lever that can harden into a barrier to entry. Ostrom’s adaptive governance principles then help identify how a non-replicable, congestible network can remain open to innovation without abandoning the institutional structure that preserves reliability. In transmission systems, the key implication is that the margin of adjustment is institutional rather than competitive: because exit and replication provide limited discipline, RTO rules and the surrounding polycentric governance environment must supply the adaptive ‘pressure’ that would otherwise come from competition among clubs.

A first implication is that RTOs should treat boundary rules as deliberately revisable, rather than as fixed participation templates inherited from the restructuring era. One practical approach is to define evolving boundaries by recognizing new categories of participants and participation models – such as DER aggregators, microgrids, storage providers, and other service providers – while nesting those participation arrangements within the broader RTO framework. This approach is consistent with Ostrom’s ‘nested enterprises’ principle: specialized sub-arrangements can operate within a larger governance regime, lowering the fixed costs of entry by tailoring measurement, telemetry, and performance obligations to the operational characteristics of new resource types. A second implication is that tiered participation structures can substitute for the missing competitive discipline that replication would otherwise provide. Rather than a binary choice between full inclusion and exclusion, tiering can allow limited but well-specified access rights and obligations that scale with a resource’s system impact. In practice, tiering can mean differentiated telemetry requirements, phased performance obligations, and scaled financial assurances, with clear pathways for moving towards fuller participation as compliance and performance are demonstrated. A third implication is that RTOs should adopt explicit protocols for periodic rule revision – including network integration rules, capacity accreditation methods, and cost-allocation frameworks – so that the exclusion regime remains evidence-based and adaptive as technology changes. In a non-replicable club, such protocols operate as a constitutional substitute for exit: they create predictable opportunities to revisit boundary rules in light of observed performance and changing system conditions, rather than leaving reform to episodic conflict.

These institutional adjustments must be complemented by governance capacities that make inclusion credible. Stronger monitoring, performance verification, and graduated sanctions can reduce the reliability risks associated with new entrants and thereby lower the informational and political barriers to inclusion. Advanced monitoring and data systems can improve visibility into DER performance and load patterns, reducing uncertainty, and enabling more targeted incentives and penalties, including penalties that reflect real-time deviations where appropriate. Graduated sanctions can also be structured as a learning-oriented pathway for entry: heightened supervision and conservative performance assumptions can apply initially, with requirements relaxed as resources demonstrate reliable compliance. At the same time, collective-choice arrangements must evolve to prevent the exclusion regime from being locked in by legacy participation structures. If technological change expands the set of affected parties, then stakeholder processes that were engineered around a smaller number of incumbent categories can become an institutional bottleneck. RTOs can respond by creating specialized committees for emerging technologies, adjusting sector definitions, or otherwise ensuring that entrants with materially different capabilities and business models can participate meaningfully in rulemaking. In Ostromian terms, this is not political reform for its own sake; it is an institutional design requirement for maintaining legitimacy and adaptive capacity in a rapidly changing governance environment.

Several RTOs have begun experimenting with governance and market reforms to accommodate emerging technologies while maintaining reliability, illustrating pathways for reconciling exclusion with innovation. CAISO’s DER Provider model was one of the first attempts to integrate aggregated DERs into wholesale markets, allowing multiple small resources to participate as a single entity, although challenges remain in metering and market dispatch alignment, and demand-side resources are still redefined as supply to enable their market participation.Footnote ¹⁶ NYISO has piloted programmes for storage and demand response, including its market design for storage as a dispatchable resource, aiming to better reflect the fast-response capabilities of batteries and aggregated loads in wholesale price formation. Meanwhile, FERC has facilitated stakeholder information sharing around questions of integrating new technologies in bulk power systems, such as their Innovations and Efficiencies in Generator Interconnection Workshop in September, 2024.Footnote ¹⁷ FERC’s ongoing efforts to reform interregional transmission planning also highlight the structural barriers to integrating new, heterogeneous technologies and balancing supply across large geographic footprints, as legacy cost allocation rules and jurisdictional conflicts have slowed investment in needed transmission capacity (Macey et al., Reference Macey, Welton and Wiseman2024). These examples reinforce the central implication of the Ostrom–Buchanan lens: in a system constrained away from replication, innovation depends on the capacity of institutions to revise boundary rules and to coordinate reforms across nested and adjacent governance centres.

These reforms are difficult precisely because the incentives for institutional change are endogenous to the existing exclusion regime. Institutional path dependence gives incumbent stakeholders substantial influence over rulemaking, and incumbents may resist reforms that lower entry barriers or reallocate governance rights. In a replicable club environment, dissatisfied users could respond by forming alternative clubs; in transmission governance, that disciplinary mechanism is weak, which increases the importance of polycentric checks and externally recognized pathways for reform. FERC oversight, state policy pressures, market monitoring, and inter-RTO coordination can therefore play a constructive role as complementary sources of discipline – helping to keep the exclusion regime adaptable when internal governance would otherwise tilt towards stasis. The practical challenge is to design exclusion so that it remains reliability-preserving while also being contingently adjustable, allowing the transmission network club to capture the static gains from coordination without sacrificing the dynamic gains from entry and experimentation.