What’s a CfD worth to you?

Auction theory can help participants in the forthcoming CfD auctions formulate winning bidder strategies. Jostein Kristensen outlines three different approaches.

One of the key aims of Electricity Market Reform (EMR) is to bring forward investment in low-carbon generation through the use of long-term contracts for difference (CfDs). These CfDs will be allocated to generators via ‘auctions’, and the first CfD allocation round is rapidly approaching – the closing date for bids to the first CfD auction is expected to be in December 2014.

A set budget sets the overall funding cap and may include various ‘maxima’ and ‘minima’ capacity thresholds or other constraints for given low-carbon generation technologies targeted by the Department of Energy and Climate Change (Decc). Eligible generators can submit multiple applications for various projects to the delivery body (National Grid), and an auction will be called if the applications collectively exceed the budget in any delivery year, or if any minima or maxima thresholds are breached. If an auction is called, bidders compete on the basis of the strike prices offered for specific projects, with the CfD establishing payments that are equal to the difference between the strike price and a reference price.

From the perspective of auction theory, two other features of the CfD allocation process are especially important when devising a bidding strategy: eligible generators must submit sealed bids; and all projects planned for a given delivery year are awarded a clearing price established by the highest winning bid. The CfD auction is therefore a multi-unit, sealed-bid, uniform price auction. In addition, bidders in the CfD auction would also be expected to have different, but related, valuations of the strike prices for the projects they are putting forward, and would not be expected to have perfect knowledge of the true valuation of their own project. As a result, bidders would be expected to change their own valuation if they knew a rival’s bid, and private information can give an advantage.

Importantly, these features of the CfD auction mean that the optimal bidding strategy (in the sense of it being a ‘strictly dominant’ or ‘equilibrium’ strategy) is not necessarily to simply bid on the basis of bidders’ estimated own ‘true’ valuations. In particular, bidders have the incentive to reduce (or ‘shade’) their bids below their central estimates in order to increase the pay-off from winning the auction, albeit this also marginally reduces the probability of winning.

The ‘winner’s curse’ refers to the regret experienced by successful bidders when they believe that the very fact they have won implies that they “overbid” or bid naively. The winner’s curse can arise whenever bidders have interdependent (but not purely private) valuations and where there is uncertainty about those valuations, both of which apply to CfD auctions. Moreover, the winner’s curse is generally more prevalent in sealed-bid auctions (compared with multi-round, open-bid mechanisms), since these make price discovery more difficult. Indeed, the desire to avoid the winner’s curse is often a motivation for bid-shading.

Of course, it is also possible for the auctioneer – in this case, the government or the ultimate consumers funding the CfD levy – to experience disappointment at certain outcomes. For example, granting CfDs that are later perceived as being too high.

Given that the opportunities and risks afforded by the CfD auction rules are potentially significant, it is important for prospective auction participants to hone their bidding strategies. Three different approaches are possible.

The first would be to harness the insights from auction theory and to learn the lessons from other auction settings that share features with the CfD auction process. Alongside the growth in academic literature on multi-unit auctions, auctions in some form are increasingly common in procurement processes and there is now considerable scope for the insights from theoretical models to be tested empirically. In Oxera’s experience, such a qualitative approach, based on a small number of relevant case studies, can quickly and effectively help bidders gain an appreciation of the strategic intuition necessary to improve their bidding strategies.

Extending the case study approach by formulating a range of bidding scenarios is the second approach. These scenarios can be used to test the impact of alternative bidder configurations and bidding strategies on auction outcomes. In the case of CfD auctions, the critical inputs would be the number and variety of developers and project configurations. This would provide a basis for establishing a ‘merit order’ of CfD auction bids, which can then be used to test the impacts of alternative bidding strategies in the context of prospective budget allocations.

The aim of such scenarios would be to assess the conditions under which different strategies would be expected to increase the probability of submitting a winning bid or increasing the clearing strike price, and the potential gains from strategic bidding. For example, CfD auction bidding strategies could include strategically deviating from a bidder’s estimate of the true valuation by: increasing it to achieve a higher clearing strike price; reducing the capacity of the project; or shifting the project to a later allocation round expected to be less competitive.

The third approach is to develop a simulation model to test a large number of individual scenarios and bidding strategies, thereby allowing the benefits, costs and risks of alternative bidding strategies to be quantified. A simulation approach also allows different bidders to be modelled as autonomous agents, thereby directly taking account of their strategic interactions subject to the auction rules. This approach can then be used to compute the optimal strategies of different bidders under a variety of circumstances.

As with most modelling or simulation-based analysis, there is frequently a trade-off between the degree of realism and detail achieved by the model and the effort required to develop it. Fortunately, there is considerable flexibility in how auctions can be modelled, depending on the precise aims and requirements.

The CfD allocation process offers a number of opportunities for bidders to ensure a successful outcome. Developing a robust bidding strategy is vital, and harnessing the insights from auction theory can enable this.