Can capacity markets be designed by democracy?

Abstract

In the United States, Regional Transmission Organizations (RTOs) operate the power grid serving nearly 70% of electricity customers and are critical organizations for ensuring reliable system operations and facilitating the integration of new technologies and market participants. RTOs are designed to be stakeholder-driven organizations, with rules and policies crafted through a highly participatory process. While the decisions that RTOs make have implications for industry, society and the environment, their decision processes have not been modeled in any systematic way. In this paper, we develop a modeling framework for the stakeholder process of PJM, an RTO serving thirteen states plus the District of Columbia, adapting some of the seminal literature from political science and political economy on the theory of voting systems. This modeling framework can generate predictions of stakeholder process outcomes, identify strong coalitions among stakeholders and identify shifts in political power in the formulation of RTO market rules. We illustrate this analysis framework using a detailed data set from stakeholder deliberations of capacity market reform in PJM. Our model predicts that the current structure of the stakeholder process in PJM makes the passage of capacity market reforms through the stakeholder process virtually impossible because it creates strong coalitions that would favor or oppose changes to capacity market rules. In the capacity market case, we also identify a small subset of voters that act as swing voters and confirm that political power is shifted to these voters by deviations from otherwise strong coalitions and abstentions from the voting process altogether. Our framework represents the first attempt to model the decision-making behavior of RTOs in any systematic way, and points towards emerging research needs in evaluating the governance structure of RTOs.

Appendix

Consider abstention from sector B (a _B ). Then, the threshold condition considering a _B is Eq. (A-1).

$$ \frac{{X_{B} }}{{\textit{TN}_{B} - a_{B} }} + \frac{{X_{A} }}{{\textit{TN}_{A} }} > 1.665 - \mathop \sum \limits_{i \in C\backslash A,B} w_{i, j} $$

(A-1)

Y-intercept (I _B ) and X-intercept (I _A ) of this inequality condition with the abstention (a _B ) are:

$$ I_{B} = \left( {\textit{TN}_{B} - a_{B} } \right)\left( {1.665 - \mathop \sum \limits_{i \in C\backslash A,B} w_{i, j} } \right)\quad I_{A} = \textit{TN}_{A} \left( {1.665 - \mathop \sum \limits_{i \in C\backslash A,B} w_{i, j} } \right) $$

(A-2)

From Eq. (5), we know that Y-intercept ($I_{B}^{{\prime }}$) and X-intercept ($I_{A}^{{\prime }}$) of the inequality condition without abstention are:

$$ I_{B}^{{\prime }} = \textit{TN}_{B} \left( {1.665 - \mathop \sum \limits_{i \in C\backslash A,B} w_{i, j} } \right)\quad I_{A}^{{\prime }} = \textit{TN}_{A} \left( {1.665 - \mathop \sum \limits_{i \in C\backslash A,B} w_{i, j} } \right) $$

(A-3)

Since $I_{A} = I_{A}^{{\prime }}$, we check that abstention (a _B ) rotates the threshold line anchored to an intercept (I _A ).