WHEN: Thursday (March 7), 1PM-2PM

WHERE: Chem & Physics 107

SPEAKER: Prof. Agostino Capponi (Professor and Director of the Center for Digital Finance and Technologies, Columbia University)

**********************************************************************************

Title:  Virtual Trading in Multi-Settlement Electricity Markets  

Abstract:   We study the role of virtual trading in electricity markets that consist of day-ahead (DA) markets, bids adjustment (BA) periods, and real time (RT) dispatches. Suppliers and load-serving entities engage in contracts for electricity in the DA market. They utilize the BA period to reconcile any discrepancies between the contracted quantities and the actual deliveries that occur during the real-time dispatch period. This multi-settlement market structure potentially gives rise to market inefficiencies. Virtual trading, designed as a purely financial transaction, was introduced to address these issues by allowing purely financial entities to arbitrage between the day-ahead market and the real-time dispatch without commitment to physical participation. While virtual trading has been widely adopted in practice, theoretical studies of its effects, particularly in markets with renewable energy sources, are limited in the literature. In this paper, we narrow this gap by characterizing the equilibrium behaviors of renewable and conventional suppliers, load serving entities, and virtual traders in a supply function equilibrium model. Our analysis shows that without virtual trading, load serving entities exert market power by underbidding their true demand estimates in the DA market, leading to a DA price lower than the expected RT price. Introducing virtual trading reduces the price gap between the two markets and eventually eliminates it as trading volume increases. Nevertheless, virtual trading additionally incentivizes load serving entities to underbid rather than present truthful demand estimates. Moreover, the inclusion of renewable suppliers also reduces load serving entities’ bidding amounts in the day-ahead market equilibrium. We numerically confirm our findings using data from both the California Independent System Operator Open Access Same-time Information System and the New York Independent System Operator. (joint work with Dan Bienstock, Garud Iyengar, and Bo Yang).

*********************************************************************************