Energy trading companies need to know the current market price for all forward deals that were made on their trading desks so that they can properly evaluate every deal and generate a profit and loss (P&L) report and evaluate their risk exposure.
This current market price comes from the forward curve, which is normally owned and generated by the risk team, controlling or other middle office departments, and not by the front office, to prevent potential P&L manipulation by the traders.
In general, the forward curve is generated from the available forward prices.
On very liquid markets, like WTI or Brent crude oil, forwards are traded constantly for many months in the future. To build a forward curve for those products, one could simply extract the currently traded forward prices from a market screen.
But it’s not always that easy. In many markets, you don’t have the luxury of publicly available real-time data for the next 12 monthly forward prices that you would need to easily create a one-year forward curve.
This is where our product CurveBuilder comes into play.
These are some examples where a forward curve generation is more complicated and where we would need the help of math and formulas to generate a forward curve.
- Trading for a certain product might be done OTC (“over the counter”) under bilateral agreements and not via a trading platform. Price data for those markets is provided by data providers, like Platts, who call market participants to get the latest price information of those markets. That data might have gaps or does not get published regularly.
In this case, one would need to work with interpolation to fill the gaps. And on days where the data was not published, the previous forward curve could be used.
- A company has been trading crude oil futures on a market with much lower liquidity than the two major crude oil products Brent or WTI. This could be, for example, Nigerian crude oil. Most trading activity is happening on the first two forward months, but not beyond. How would we generate a forward curve for this market?
One solution could be to use a forward curve from a similar market (e.g. Brent) and extrapolate the forwards beyond the two available front months from the Nigerian crude oil market based on the shape of the Brent forward curve.
- A company has a gas delivery contract that is priced based on a Rhine River oil price index, which is published monthly, but there is no forward market for those prices that could be used to create a forward curve.
In this specific case, the oil gets shipped down the Rhine River from Amsterdam, one would identify a strong price correlation between Amsterdam oil prices and the monthly Rhine River price index. A forward curve could be built by applying a mathematical correlation formula onto the existing Amsterdam oil forward curve to generate the forward curve for the Rhine River price index.
In our next blog on forward curves, we will discuss how CurveBuilder can manage these challenges.