For many participants involved in power trading an accurate valuation of generation assets and their associated risks is crucial. Not only does rigorous valuation provide an accurate view of a portfolio’s current value, it also enhances the ability of the owner to quantify and hedge the risk associated with the asset. Given the ability to run (or not run), it has become common place to view generation units as “real options” when attempting to value them; however, this can run into strong headwinds attempting to evaluate all the associated physical variables.
At the most basic level the inherent value of an asset can be thought of as a series of spread options, whilst more complex models attempt to layer in the subtleties of the asset in question. This, and a following post, will look in more depth at the most widely methods used. Before jumping in too far, it’s useful to consider with properties of generations assets need to be considered.
Properties of generation assets
Unlike financial options, generation assets have many different interacting physical properties that add flavour and complexity to their valuation. It is important to consider a number of these as they are crucial in accurately capturing the assets value and risk. The most important of these are as follows:
|Maximum capacity||The maximum amount of power that can be produced in an hour. It can change from month to month depending – typically – upon the thermal gradient between the assets and the ambient air temperature. The asset is most efficient at this level|
|Minimum stable generation||The minimum the asset can generate but still remain stable. In models that capture the operation of the generation unit the unit is dispatched at a level between the maximum and minimum generation loads|
|Heat Rate||This can be thought of as the efficiency of the unit. A larger the gradient in temperature between the unit and the ambient air will lead to more efficient operation|
|Variable operational and maintenance cost||The non-fuel costs associated with running the unit|
|Start-up cost||The costs associated with starting the unit. These can range from start-up fuel costs to wear and tear on the unit. These are the hardest to model since they are themselves variables of other properties; for example consideration must be given to how long the unit has been off for|
|Ramp-up and Ramp-down rate||These rates determine how quickly the unit can increase or decrease generation. They apply both from off to stable generation and then from stable to maximum. They are similar to start-up costs in that they are a function of how long the unit has been on/off for|
|Minimum up/down time||These state that if a unit is off that it must be off for a certain amount of time. Conversely if a unit if brought on then it must stay of for a certain length of time. The constraints are in place to minimise excess wear and tear brought about my constant switching on and off|
|Emissions||Costs associated with CO2/NOx/SO2. These are important considerations since most emissions markets are still relatively immature and it's difficult to establish reliable models for their price. When parameters can be established it is common to use a stochastic model|
|Outages/Maintenance||Every unit will require periods of maintenance and thus planned outages – typically over the summer for thermal assets – needs to be considered. Since these are also physical assets the potential for random outages due to breakages must also be factored in as this can have a significant impact upon|
|Fuel/Power transportation costs||Generation assets are typically not located as gas terminals/coal mines or a the load centre. We thus have to also consider the costs with transporting the fuel and of getting the power onto the grid|
Financial options versus real options
One of the benefits of treating an asset as a real option is that we can use many of the valuation techniques that have been developed for the valuation of financial options. As is clear from the table above, there are a number of parameters we would also like to consider when valuing a generation asset. In short, our strike price will need to be a complex weighting of many considerations. Consequently, it is worth understanding the difference between financial and real options so that we understand the limitations these techniques impose on us when they are used to value generation assets.
Firstly, typical financial options are paid up front and there is no significant cost to exercising the option. As we have seen, there is usually a startup cost associated with the generation assets (motors need to be turned, water pumped, heaters initiated etc). Since the start-up accounted per start and not per hour run, it is more complicated to implement in a closed form solution than in a Monte Carlo simulation.
The second major difference is that once the financial option matures we can immediately exercise it. Generation assets, however, have a ramp rate which implies that we can instantly go from off to maximum capacity. In other words, we need to exercise the real option before its ‘expiry’.
Thirdly, most financial options can have the payoff described in a single payoff function that can be easily written down. This is even true for some path dependent options like Asian options. The operational constraints of a generation asset – start up costs, ramp rates, min off times – require us to keep track of prior unit states. This requirement makes it difficult to write a single simple payoff function. To simplify and to make use of the standard library of options models, many of these more esoteric factors of discounted out of our models. However, since significant costs can be incurred through these, this can be less than ideal.