How can generators get the best price for their PPA?

Traditionally, renewable generators sell their power to suppliers through a marketplace. In this method, potential purchasers submit bids for the tendered PPA at a fixed point in time, the auction date.

One of the key variables that affects the price that generators achieve is the underlying market power price which fluctuates continuously. In this series, we investigate what the best selling strategy might be for generators.

Part 1

If we look back at a sample period from 2020, we see large fluctuations in the seasonal power price for Winter 2021.

During this sample 80 day period, the power price starts at £53.13 and ranges from as low as £42.25 and finishes at a high of £62.07. This shows how key the timing of an auction might be.

Get the right day and suppliers could potentially have 30% more to offer us than we would get on a bad day.

However, hindsight isn’t so useful as we go forward. We cannot predict with certainty whether prices will go up or down in future: indeed, if we decide to wait until the end of a period in search of the best price, we might find that the power price continues to decline.

Strategy 1 – immediate sale

At this stage we know very little. Strategy 1 will be just to hold the auction right away. The current price is £53.13 and this is our expected price. Note that we will not definitely get this price. The nature of markets is that they constantly change. Actually we only have a 50% chance of achieving this, as 50% of the time the price will go down before we can close our auction.

Strategy 2 – informed estimation

Here, we are going to have to do a bit of educated guessing. We will make an assumption:

The power price will continue to behave in the same random way as it has done before.

We first need to extract information out of our existing data, that describes how power prices behave, so we can then use this to model what will happen in the future. The traditional way of doing this in financial markets is to calculate the volatility and drift of our power price, and to also note the way it changes and decide how it is distributed.

For our purposes however, we only need the drift.

The drift is how the power price tends to change overtime: does it tend to go up or down, and by how much? In order to calculate drift, our first job is to convert our series of daily power prices into a series of the changes in the daily power price.

For example, if we had a power price on Day 1 of £10, on Day 2 of £11 and on Day 3 of £12, what are the changes in power price as a fraction? To find these, we simply divide the current day’s power price by the previous day’s price and subtract 1, giving us the following series of changes:

[£10, £11, £12] → [11/10, 12/11] = [1.1, 1.09]

We then average these values to get the drift. For our data we get a drift of: 0.00059 or 0.059%, which means that the power price tends to increase by 0.059% every day. If we have a power price of £10 on Day 1, it will most likely be £10.0059 on Day 2, and £10.60 on Day 100.

What does all this tell us? It suggests that if we are given a choice over what date we hold our auction, then as prices tend to go up over time, we should wait until the end of the period to do so. If we do this over the next 80 days we will most likely achieve a power price of:

£53.13 * (1.00059)^80 = £55.70

Conclusion

This is as far as our current approach can take us. It results in a strategy of setting our auction date as far into the future as possible and leaving it up to chance. But could there be a better way? We will investigate this in Part 2…