Welfare Impact of Virtual Trading on Wholesale Electricity Markets
Virtual bidding has become a standard feature of multi-settlement wholesale electricity markets in the United States. Virtual bids are financial instruments that allow market participants to take financial positions in the Day-Ahead (DA) market that are automatically reversed/closed in the Real-Time (RT) market. Most U.S. wholesale electricity markets only have two types of virtual bids: a decrement bid (DEC), which is virtual load, and an increment offer (INC), which is virtual generation. In theory, financial participants create benefits by seeking out profitable bidding opportunities through arbitrage or speculation. Benefits have been argued to take the form of increased competition, price convergence, increased market liquidity, and a more efficient dispatch of generation resources. Studies have found that price convergence between the DA and RT markets improved following the introduction of virtual bidding into wholesale electricity markets. The improvement in price convergence was taken as evidence that market efficiency had increased and many of the theoretical benefits realized. Persistent price differences between the DA and RT markets have led to calls to further expand virtual bidding as a means to address remaining market inefficiencies. However, the argument that price convergence is beneficial is extrapolated from the study of commodity and financial markets and the role of futures for increasing market efficiency in that context. This viewpoint largely ignores details that differentiate wholesale electricity markets from other commodity markets. This dissertation advances the understanding of virtual bidding by evaluating the impact of virtual bidding based on the standard definition of economic efficiency which is social welfare. In addition, an examination of the impacts of another type of virtual bid, up-to-congestion (UTC) transactions is presented. This virtual product significantly increased virtual bidding activity in the PJM interconnection market since it became available to be used by financial traders in September 2010. Stylized models are used to determine the optimal bidding strategy for the different virtual bids under different scenarios. The welfare analysis shows that the main impact of virtual bidding is surplus reallocation and that the impact on market efficiency is small by comparison. The market structure is such that it is more likely to see surplus transfers from consumers to producers. The results also show that outcomes with greater price convergence as a result of virtual bidding activity were not necessarily more efficient, nor do they always correct surplus distribution distortions that result from bias in the DA expectation of RT load. Compared to INCs and DECs, the UTC analysis showed that UTCs do not have the same self-corrective incentives towards price convergence and are less likely to lead to nodal price convergence or correct for surplus distribution distortions caused by uncertainty and bias in the DA expectation of RT load. Additionally, the analysis showed that UTCs allow financial traders to engage in low risk high volume trading strategies that, while profitable, may have little to no impact on price convergence or market efficiency.
Preckel, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our