Developing Renewable Energy
Investigating the scenarios arising from the aggressive renewable built out and how the bank hedge can turn against you.
In the rather dry previous post we went through the simple mechanics of a Bank hedge. The purpose was to show each component in the settlement formula.
This post will dive into the Bank Hedge and investigate how well it hedges the revenue of a Wind Project.
Horror scenarios can arise with this product, which was teased in the previous post. Therefore, we will focus on “letting the cat out of the bag” and graphically show how these scenarios unfold and explain why.
Let’s start with a short introduction to what hedging is. The most common hedging product in people’s everyday life is insurance. In principle we pay for an insurance on an ongoing basis to cover us against unfortunate scenarios (layoffs, house fires, cars crashes), which could have a severe financial impact. When the unfortunate scenario occurs, we will partially or fully be covered by our insurance (hedge). On the contrary we must remember that by paying an ongoing premium for the insurance (hedge) we will also give away some of the upside of our income. The insurance provider will also have to earn a living, so “Ceteris paribus” we will pay more to the insurance provider than we will get from the insurance provider, in the long run.
This gives rise to a few important points about a hedge:
With this introduction completed, let’s turn the focus back on hedging wind projects revenue streams.
As the Financial hedge has the purpose of mitigating the impact of one or more risks of a Wind Project, it is relevant with a brief introduction to the risks faced by a merchant (no subsidies considered) Wind Project.
The fuel used for a wind turbine generator is the wind, which is free (contrary to coal and gas) but unpredictable, especially mid and long term. Therefore, we can only estimate based on years of historical wind, what the produced power could look like on an average basis in the future.
Operational risk simply deals with how efficient the wind is converted into energy going into the grid. The actual operational efficiency will always be less than 100% as various operational risk and losses are unavoidable. Examples being the performance of the turbine (power Curve, sub-optimal performance), wake effects, curtailment, availability, electrical losses etc. Some of the losses are out of the projects control, while other are impacted by how efficient the Wind Farm is operated e.g. service efficiency and responsiveness to component exchanges. The operational efficiency generally lies between 80-90 %.
The price received for the produced electricity dependents on various factors e.g. the mix of generation capacity in the specific market, which generation assets are setting prices and the seasonal demand. Pricing in energy markets is a topic of its own, but the important point here is that prices in some markets can fluctuate significantly. It is not unseen that the price increases from 20-30 USD/MWh to + 1000 USD/MWh within 5-15 minutes. Especially market with high reliance on renewable generation from wind and solar experiences volatile prices.
As mentioned in the previous post the Bank hedge is also referred to as the “Fixed Price Volume Hedge” as there is a fixed volume, which the hedge provider has committed to pay a known price for. This could give one the feeling that the fuel risk is hedged, which was also the case if the Hedge Volume was equal to the actual production during all trading intervals. Unfortunately, this is not the case and the Bank Hedge should be seen mainly as a hedging of the Price Risk. This will eventually also play out in some scenarios where months of heavy wind will keep the prices low enough for the Bank Hedge to settle in favor of the Wind Project. Therefore, the bank Hedge to some extent hedges the Price Risk.
A Swap can simply be explained as the exchange, between two parties, of a fixed cashflow for a floating cashflow. When looking at the simple settlement of the Bank Hedge we are exchanging a floating price for a fixed price, which is a swap of the Price Risk, but the other element of the equation is not swapped. We exchange a fixed production for a fixed production, whereby we do not get rid of the Fuel (Wind) Risk. Lets once again remember that the aim is to hedge a Wind Projects volatile revenue streams, where days without wind will eventually mean no revenue at all while we must deliver fixed amount to meet the Hedge Obligation.
Let’s get into the nitty gritty details where the danger lies. As teased in the intro to the previous post there will be scenarios with a Bank Hedge where the Wind Project will be watching in horror. This eventually comes down to the risks you get under your roof when you decide to let a bank hedge into your house.
The volume risk concerns the overall deviation between the MWh merchant production for the wind farm and the MWh Hedge Volume. The Hedge Volume is the estimated P991P-level of 99: means that there is 99% probability of the production being equal or higher than the calculated P99 production value. production, which is approximately 80% of the expected (P50) yearly production. Due to the uncertainty of the wind and operational performance, there is a risk that the Project will not produce the P99 level in every time interval (hours, days, months), which will often lead to the Hedge Volume being higher than the merchant production or vice versa. If we look at this in the perspective of 80% of the merchant production needs to be sold to the hedge provider to meet the Hedge Volume we are committed to, then we will constantly be short of production to sell or having too much production, which will be sold unhedged at the spot price.
The shape risk refers to the risk that the shape of the Hedge Volume does not fit the actual MWh merchant production of the wind farm in the hourly time intervals. This is easier understood by looking at a daily Hedge Volume together with the actual production from the Wind Project:
As shown in above graph the MWh Hedge Volume changes two times every day (“7×16;7×8” Hedge Shape as presented in the previous post). The shape risk is clearly shown by the deviation between the MWh Production (green area) and the MWh Hedge volume (blue area), where the example above shows a big gap from 15:00 to 18:00. This gap represents a deficit in Merchant Production relative to the amount we are hedging. Let’s set it up in a table:
|Time Interval (Hour Ending)||Merchant Production (MWh)||Hedge Volume (MWh)||Long/Short (MWh)|
The intervals, in the above table, shows an example where we have a deviation where the Merchant Production is bigger than the Hedge Volume and vice versa. This could make one think that the shape risk, and the fact that we are either over hedged or under hedged I some intervals, seems like a harmless zero sum gave where “you win some, and you lose some”. The real danger arises when we couple these Long / Short MWh amount with the pricing dynamics.
It is a general known fact that energy is cheapest at nighttime, where demand for energy is low, and higher during the day when energy demand is higher. Furthermore, the hours when people come home and starts cooking during a hot summer’s day can create high peak demands and therefore high prices. Especially when the wind and the contribution from other intermittency energy sources are non-existing.
Let’s now add the price to our graph and turn our attention to the price of energy during the intervals in the table:
It is easily observable that the decrease in production from the Wind Project is negatively correlated with the power price (Negative covariance), which translated into plain English, means that an increase in windspeed leads to a decrease in Power Price to some extent2This is not a 1 on 1 relationship (not 100% correlated). This is a general issue with energy markets, where there is a high penetration of intermittent generators, where a lot of the “old school” baseload has simply been shot down as they cannot cover their costs of operation due to the competition from wind and solar. This requires some very expensive energy sources to cover the peak loads that can occur doing a hot summer (where there is no wind).
Now let’s reveal the month and year of the day shown in the graph above. The shown data is for July 2018 from the ERCOT market (Texas). This was a month where the temperatures in Texas surpassed 40 degrees Celsius, which lead to a high usage of air conditioning. This high demand for energy coupled with almost no contribution from wind, during peak hours, lead to price intervals of plus 1.000 USD/MWh in July.
Now let’s reintroduce our table – with prices included and hedge obligations:
|Time Interval (Hour Ending)||Merchant Production (MWh)||Hedge Volume (MWh)||Long / Short (MWh)||Settlement Power Price (USD/MWh)||Merchant Revenue (USD)||Hedge Obligation (USD)||Mismatch (USD)|
The “Long/Short” column shows the MWh produced relative to the Hedge Volume, which was previously mentioned as the Shape risk. The green box on the table shows a significant excess production relative to our hedge volume at a price, which is relatively low. The red box on the contrary shows a significant shortage production relative to our hedge volume at a price, which is very high. This is further evident in the mismatch (8), where a short hedge volume position shows a much higher mismatch relative to a long hedge volume position despite being almost the same MWh quantity difference.
Let’s try to put this into a graph:
The relationship shown in the graph is a product of the shape risk, which we get with a hedging product, which does not take care of the fuel (wind) risk. It is fair to say that the Hedge Shape used for this example has a poor shape (relative to the production in the example), but no shape will ever be close to a good match for the randomness of the wind. Additionally, we need to acknowledge that this is not all due to the risks faced with a bank hedge, as it is the application of the Hedge Product in a market where negative covariance exist. Meaning, the fact that we get low prices for our energy when we produce a lot, and prices often are very high when production is close to nothing.
The above examples do not mention the Hedge Revenue, which is the fixed cash flow from the Hedge Provider at settlement. This makes the cash flows look like a bond if the Tracking Account limit has not been reached. Therefore, if the developer’s business case has not taken the increasing covariance risk, resulting in a higher draw on the tracking account, into consideration – then the shortened duration of the bond-like cash flow will come as an unpleasant surprise. This is illustrated in the following graph:
As per the example in the graph, the cash cashflow could become variable again before anticipated as the tracking account is maxed out before originally assumed. This will probably disappoint investors who thought they signed up for a steady cashflow stream for the next 10 years, while it ends up only being 6 years.
Therefore, the impact of having a bank hedge signed for plus 10 years in a market, where the negative covariance is increasing due to the wind built out and retirement of baseload, can be damaging to the stable long-term returns investors initially signed up for.
The best way to sum up the investigations above is to bring the attention back to the few points about a well performing hedge:
|Hedge Consideration Points||Bank Hedge Performance|
|A Hedge should be an offsetting mechanism, which will cover you when unfortunate scenarios occur.||The bank hedge can in some scenarios be an offsetting mechanism but will turn some unprofitable days (due to lack of wind) even more unprofitable (due to a volume commitment).|
|It costs money to sleep well at night. In the long run, we will give a fraction of our income for that comfortable sleep.||The hedge product will cost money. Therefore, an examination of the impact to the overall business case must be done. E.g. what will the effect of a more certain cashflow have on the financing options for a project.|
|The better you want to sleep – the more you will have to pay.||A better sleep for a wind project owner could be a hedge that would not only cover the Price risk, but also the fuel risk and potentially the operational risk, but eventually the "cost of hedging" will increase.|
|Before you pay for the full package, you should examine, which exact risks will keep you up at night.||The risks that an owner can live with often dependent on the owner background. Many utilities and dedicated wind players are comfortable with the operational risk, while financial players might want to transfer this risk to more competent owners of the risk.|
The examination of the Bank Hedge showed us the ups and downs of hedging with this standard hedging product for wind projects. The Bank Hedge only addresses the price risk and the fact that the fuel risk is not addresses, will give the wind project other risks as a bi-product of this. One of these bi-products was the shape risk, which will be impossible to shape directly after the future wind profile of the wind project, which leads to the instances of high-volume commitment towards your hedge provider, while your actual production from the wind farm is much lower, which leads to a mismatch. The next post will examine the new opportunities that strive to cover more bases and take on the fuel risk. Especially the Proxy Revenue Swap will be covered extensively over the next few posts.
Blue Power Partners is developing and building renewable power with a dedicated focus on wind, solar and storage projects worldwide. Our core business is centered around Development, Construction and Operation, and we are partnering with globally leading developers and asset owners within the Industry.
Together we aim at pioneering a greener future.