Security of Generation Supply in Electricity Markets

Abstract

As indicated in Chap. 3, calling “liberalization” and, particularly, “deregulation” to the regulatory reforms that have taken place worldwide during the 1990s and early 2000s is somewhat misleading. The main reasons are that only some of the activities involved in the supply of electricity have been subject to an in-depth reform, most governments still have a heavy hand on their power sectors and the volume and complexity of the new regulation is frequently similar, or even greater, than the so-called traditional regulatory framework, see (Borenstein and Bushnell 2000) or (Ruff 2003). This is why the term “restructuring” has been preferred in some electric power systems (particularly in the United States). The subject matter of this chapter, i.e. the need for regulatory intervention to complement electricity markets in order to guarantee security of generation supply, is a good illustration of this point.

I had to abandon free market principles in order to save the free market system.

George W. Bush

This chapter draws heavily on [4], [5] and [42].

Annex: The Method of Reliability Options

The reliability option seen as a way to introduce a market-compatible price cap

One possible way of describing the motivation for the reliability options design could start from the idea of implicit insurance. When there is a security of supply problem, and prices for final consumers are high and shortages appear, the situation may become a very difficult political problem for the regulator. Consumers are also hurt but, as we have just seen, they do not react. So it is the regulator who should try to protect consumers (and also himself against the associated political risk), through a change in the market design. He would like to impose a price cap on the market, but this is a problem with the very old and inefficient plants, which may not produce even if needed if the price cap is low, and also with the new entrants, which may be discouraged with the perspective of not being able to see high prices. In financial markets, a buyer who wants to get a price cap on his future purchases can acquire a certain kind of derivatives, known as a call option, which gives him the right, but not the obligation, to buy the item at a predetermined price (the strike price) in exchange for a premium fee. This is a way to obtain market-compatible price caps. Accordingly, we propose that the regulator should buy call options from the generators, probably through a centralised auction, and therefore isolate consumers from the high prices.

At the same time, the generator that is selling a call option is giving up receiving the part of the spot price that is above the strike price in exchange for the premium fee or, in other words, he is exchanging some uncertain and very volatile income from the spot market (which only happens in periods of stress for the system) for a certain stable and predictable remuneration. This greatly reduces the risk for generators, which, as has been analysed in this chapter, is one of the major reasons obstructing that the “right” amount of new investment may happen at the “right” time. This is particularly true for peaking units, which are exposed to much larger income volatility.

Thus, two main objectives are achieved with this mechanism: on one hand, consumers do not have to bear the risk of having high energy prices reflected in their bills; on the other hand, efficient economic signals for new investment are being provided.

Two relevant time-related parameters associated to the reliability option

As with any long-term contract, there are two relevant time-related parameters associated to the reliability option contract: the lag period (also known as lead time, or planning period, as defined in the Colombian regulation) and the duration of the commitment. Defining a sufficiently large lag period is essential when the objective is to allow enough time to build the plant, and the latter has also to be long if we want to reduce risk exposure and thus facilitate project financing.

Further sophisticating the product definition: tying a physical obligation to the financial option

It is worth noting that the financial option without any associated physical delivery obligation is a tool that serves exclusively to hedge market agent’s risks. Whether the mere efficient allocation of risk, coupled with short-term signals, is sufficient to ensure that the security of supply is solved or not, is a matter that the regulator should evaluate.

If, for some reason, the regulator desires to add further incentives for being available when needed in the short term, one alternative would be to tie a physical delivery obligation to the call option. This means that an option-selling generator that, when the prices are high, fails to provide the committed power has to bear an extra penalty (i.e. on top of the payment because of the financial side of the call option) for each non-delivered megawatt.

The mechanism to trade the reliability option

This reliability product could fit in the context of either a price or a quantity-based mechanism. In the latter alternative, as with any other product, it could be traded either bilaterally or in a (centralised or not) auction mechanism. Although diverse alternative schemes could be devised, the fact is that in practice this product has only been defined in the context of centralised quantity-based auctions. Indeed, when making reference to the “reliability options mechanism” it is usually understood not only the reliability option product, but also these latter features (i.e. the physical obligation and the centralised auction). In the following we will restrict the discussion to this framework, which, in a nutshell, could be described as follows:

• An auction is organised where the auctioneer has to determine, in advance, at least the following parameters:

  • the strike price, $s$: it should not be too low, since it acts as a price cap for demand and somehow represents the frontier between the “normal” energy prices and the “near-rationing” energy prices,

  • the time horizon: typically 1 year for existing units and longer time horizons for new entrants. The seller can be required to generate the committed capacity at any time during that period,

  • the total amount of power to be bought, $Q$,

  • and the value of the explicit per unit penalty, $pen$.

• The generators submit one or several bids to the auction, expressing quantity (the capacity they want to sell) and price (the required premium). Note that a significant additional advantage of the reliability option approach is that it eliminates the need for the regulator to calculate the “firm capacity” of each unit. This is clearly an improvement over alternative methods, especially when there are energy-limited plants involved.

• The market is cleared as a simple auction and all of the accepted bids receive the premium that was solicited by the marginal bid.

• During the specified time horizon, any time the spot price $\rho$ exceeds the strike price $s$, the bids that were accepted in the capacity auction will have to refund the regulator—and, indirectly, consumers—for the difference $(\rho - s)$ for each megawatt sold in the capacity market. Henceforth, we will refer to this refund as the “implicit penalty”.

• In case the physical obligation is tied to the financial option, then if the spot price is above the strike price and the production $g$ of a certain generator is lower than the committed capacity $q$, then he would have to pay to the regulator an “explicit penalty”, computed as $pen \cdot (q - g)$. An example of this is illustrated in Fig. 12.1 where it is shown how much demand (D) has to pay during each time interval, as well as how much generators (G) who have signed a reliability option should receive.

  Fig. 12. 1

  

  The reliability product