electricity forward markets

Vargön Alloys A.B. (a subsidiary of Yildirim Holding) has an outstanding electricity contract with the energy company Vattenfall A.B. The contract stipulates a base delivery amount of 60 Mega Watts per Hour. Unfortunately, due to planned production cuts, Vargön Alloys will never be consuming this much electricity next year. Hence the undelivered electricity will have to be sold into the spot market. In order to lock in prices for these future sales, Yildirim Holding decided to engage in derivatives trading in Nordpool which is the common electricity forwarding market of Nordic countries.

When should these forwarding contracts be sold? Does the forwarding market correctly estimate the 2009 spot electricity prices in Sweden? Yildirim Holding is involved in coal trading, but this does not put the company in a more advantageous position than the average energy trader when it comes to making accurate guesses about the future electricity prices in Sweden. Or does it?

The initial reasoning of Yildirim went as follows:

1) Half of the electricity consumption in Sweden is industrial.

2) It is clear that the industry heading for a severe contraction.

3) Along with the electricity consumption, the electricity prices should go down as well.

A few afterthoughts immediately revealed how naïve this reasoning was…

Who was going to be on the other side of the trade when Yildirim sells the electricity forward? It is very likely to be Vattenfall and there is plenty of relevant information that Vattenfall has access to but Yildirim does not. However this should not be a big worry for Yildirim. Since Vattenfall is the market maker in Nordpool, whatever information Vattenfall has should already be reflected in the forward prices. The essential question is whether Yildirim has access to any information that Vattenfall does not.

As an active participant in the Swedish industry, Yildirim has a more accurate picture of the magnitude of the industry’s plight. Hence Yildirim may rightly conclude that the future Swedish electricity consumption will be less than the amount anticipated by the market. But does this necessarily imply that future electricity prices will be lower as well? In other words, is step 3 of the initial reasoning correct?

In the coal market the supply side can not immediately react to drops in consumption. New coal mining projects can be shelved, but existing mining operations cannot immediately be halted. Hence when there are less many buyers out there, the coal stocks start piling up. That is essentially why the prices go down. The dynamics of electricity markets is different. The amount of electricity in the market is always equal to the amount demanded. There may be electricity producers on stand-by, but there is never any excess electricity sitting around. A producer will immediately stop its operations if the best bid in the market does not generate a return above its operational and financial costs.

In Sweden, nuclear reactors and hydropower stations are each responsible for half of the total production. However the electricity price is set in a market that encompasses the whole Nordic region which contains types of producers other than hydro and nuclear:

In above graphics we assume that the consumption is not price-sensitive. (i.e. Regardless the price, the factories and households demand the same quantity of electricity.) This may sound like a weird assumption but it can be justified through the microstructure of the market: The system operator acts as a single buyer, ranks the offers of suppliers from the least expensive to the most, computes the expected demand and pays the marginal supplier's offer to every supplier whose offer was equal to or lower than that of the marginal supplier.

In the Nordic market electricity price is the offer made by the marginal coal power plant. Therefore if production costs stay constant, a fall in demand will cause the current marginal producer to go offline and be replaced by a more efficient one. This will in turn pull the price down since the new marginal producer has a lower cost of production due to greater efficiency. The fall in price will be negligible though since the variation among the production costs of coal power plants is probably not that great. However, if demand falls drastically and the production capacity of dams shoots up due an exceptionally wet weather, all the coal power plants will go offline and the electricity price will tank. Since latter scenario is quite unlikely, step 3 of our initial reasoning can only be thoroughly justified if the production costs do not stay constant.

Regardless the weather, hydropower stations will always stay on due to their extremely low variable costs. Therefore we can safely ignore them for the rest of the analysis. This does not mean we can ignore the future weather conditions completely. (e.g. If the winter is cold, households will consume more electricity for heating purposes.) Nevertheless we will do so because forecasting future weather conditions is extremely hard and whatever is known in a probabilistically significant way is already incorporated into the electricity forward prices.

Variable costs of nuclear reactors do not change much. Hence we will simply ignore them.

The cost of gas turbine generators will never go below that of coal power plants. Since the demand for electricity is projected to contract, not expand, we will ignore possible future changes in gas prices.

World coal market is very dynamic. It contains lots of different suppliers and the trade does not require a pre-established framework as in the gas market where construction of expensive pipelines is a necessity. If everything else stays constant and the coal price goes up (down), the Nordic electricity price will go (down) up as well. (This causation will happen with a lag since the coal power plants have to first go through their inventories which were bought at older prices.) Since the magnitude of possible future price fluctuations in the coal market is greater than the possible future variation among the production costs of coal power plants, Yildirim should focus on world coal prices and not worry about the possibility of a greater-than-anticipated fall in the Swedish electricity consumption.

This assertion can be verified by a glance at the correlation between the 2009 coal and Nordpool electricity forward prices:

(The red dots indicate 2009 coal forward prices since February 2008.)

For some (possibly psychological) reason, today the coal forward prices follow the crude oil futures extremely closely. Therefore Yildirim’s decision to sell electricity forwarding contracts now or later is essentially a bet against the market’s estimation of 2009 oil prices. Hence Yildirim should pay more attention to the US unemployment statistics (which constitute a pretty good indicator of how deep the US recession is) than to the Swedish industrial contraction. (Changes in US GDP affect the oil prices more than changes in Swedish GDP.)

Moreover Yildirim also needs to decide whether the forwarding markets correctly estimate the average 2009 future spot SEK/€ and $/€ exchange rates. (Nordpool trades in Euros, but Yildirim’s contract price with Vattenfall is fixed in Swedish Kronas. / Coal is traded in Dollars, but European coal power plants' income is in Euros.)

P.S. : Two days ago, the European Central Bank announced a 0.75% cut in its main policy interest rate. The Swedish Central Bank countered this move by a much sharper cut of 1.75% and consequently SEK/€ exchange rate jumped up. That is good news for Yildirim! Unfortunately on the same day, U.S. Department of Labor declared an unanticipated drop of 533,000 in nonfarm payroll employment. In response the market pushed Brent crude oil prices below $40 level. That is definitely very bad news for Yildirim!