Competition and Efficiency of Coalitions in Cournot Games With Uncertainty

Baosen Zhang, Ramesh Johari, Ram RajagopalIEEE Trans. Control. Netw. Syst 2019 · OpenReview

TL;DR

This paper explores how firms forming coalitions (groups) can improve market efficiency in Cournot competition when facing uncertain production capacity. They find that while individual firms tend to underproduce to hedge against risk, strategically sized coalitions can achieve near-optimal welfare and production levels by balancing market power and uncertainty reduction.

What Problem Does This Paper Address?

Standard Cournot models often assume perfect information and deterministic production. However, many real-world markets, like electricity generation from renewables (wind, solar), involve significant uncertainty in production capacity. Firms must commit to output levels before knowing their exact capacity, leading them to act conservatively and underproduce to avoid penalties for shortfalls. This conservative behavior reduces overall market efficiency. The paper investigates how allowing firms to form coalitions can mitigate this inefficiency and under what conditions these coalitions can lead to socially optimal outcomes.

Methods

The authors model a two-stage Cournot game. In the first stage, firms commit to production quantities, which determines the market price. In the second stage, actual production capacity is realized, and firms are penalized for any shortfall from their commitment. They analyze this model under various scenarios: individual firms, a single "grand coalition" of all firms, and intermediate coalition structures. A key analytical tool is the characterization of "efficiency ratios" (both for welfare and output) that compare the outcome of the Cournot game with a centralized system operator's optimal outcome. They derive asymptotic results for large numbers of firms and analyze the optimal coalition size by balancing the trade-off between reduced uncertainty (larger coalitions) and increased market power (larger coalitions). They also establish an equivalence between Cournot oligopoly (firms selling) and Cournot oligopsony (firms buying) under demand uncertainty.

Datasets

The paper uses wind data from the National Renewable Energy Laboratory (NREL) for the Eastern United States, specifically for the PJM control area. This data includes wind forecast values and forecast errors, which are used to model uncertain production capacity.

Key Insights and Ideas

  • Uncertainty Breeds Conservatism: When firms face production uncertainty, they tend to underproduce to avoid penalties, leading to inefficient market outcomes. This is contrary to standard Cournot models where more firms generally lead to higher efficiency.
  • Coalitions as a Solution: Forming coalitions can significantly improve efficiency by pooling risks. Larger coalitions can better average out individual production uncertainties.
  • The Trade-off: Coalitions introduce a trade-off: while they reduce uncertainty, they also increase market power, which can lead to higher prices and reduced output.
  • Optimal Coalition Size: There's a "sweet spot" for coalition size. Too small, and uncertainty isn't sufficiently reduced. Too large, and market power becomes dominant. The paper identifies optimal sizes for achieving efficiency in terms of welfare and output.
  • Oligopsony Equivalence: The insights from Cournot oligopoly with supply uncertainty can be directly applied to Cournot oligopsony with demand uncertainty, broadening the applicability of the findings.

Key Results

  • For a large number of firms ($N$), competition among groups of size $\Omega(\sqrt{N})$ leads to welfare efficiency approaching 1.
  • For output efficiency, groups of size $\Omega(N^{2/3})$ are found to be optimal.
  • The efficiency loss due to market power scales as $1/K^2$ (where $K$ is the number of coalitions), while the loss due to uncertainty scales as $K/N$.
  • The paper provides an empirical validation using PJM electricity market data, showing that wind power integration can significantly reduce prices and that coalition formation can be beneficial.

Limitations and Future Work

  • Exogenous Penalties: The model assumes penalties for production shortfalls are exogenous. Real markets often have endogenous penalties through spot or secondary markets, which is a significant area for future research.
  • Asymptotic Results: The primary results are asymptotic, focusing on the behavior as the number of firms approaches infinity. Extending these to finite, potentially heterogeneous firms is a challenging open problem.
  • Internal Coalition Dynamics: The paper focuses on the coalition's aggregate payoff and doesn't model internal profit-sharing mechanisms within coalitions.
  • Correlated Uncertainty: While the paper addresses both weakly and strongly correlated production uncertainty, the complexity of modeling arbitrary joint distributions remains a challenge.

Summary updated

Abstract

We investigate the impact of coalition formation on the efficiency of Cournot games where producers face uncertainties. In particular, we study a market model where firms must determine their output before an uncertain production capacity is realized. In contrast to standard Cournot models, we show that the game is not efficient when there are many small firms. Instead, producers tend to act conservatively to hedge against their risks. We show that in the presence of uncertainty, the game becomes efficient when firms are allowed to take advantage of diversity to form groups of certain sizes. We characterize the tradeoff between market power and uncertainty reduction as a function of group size. In particular, we compare the welfare and output obtained with coalitional competition, with the same benchmarks when output is controlled by a single system operator. We show that when there are N firms present, competition between groups of size Ω (√N) results in equilibria that are socially optimal in terms of welfare and groups of size Ω (N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2/3</sup> ) are socially optimal in terms of production. We also extend our results to the case of uncertain demand by establishing an equivalency between the Cournot oligopoly and Cournot Oligopsony. We demonstrate our results with real data from electricity markets with significant wind power penetration.
BibTeX
@article{zhang2019competition,
  author = {Baosen Zhang and Ramesh Johari and Ram Rajagopal},
  title = {{Competition and Efficiency of Coalitions in Cournot Games With Uncertainty}},
  journal = {IEEE Trans. Control. Netw. Syst.},
  year = {2019},
  url = {https://openreview.net/forum?id=maZc-6d9DB5D},
}
Figure