MILAN — As wind power’s share of electricity generation has increased, so have the financial consequences of risks associated with its inherently high variability – risks which don’t just affect wind power producers but reverberate vertically through the value chain and horizontally to affect a range of electricity market players. Large utilities can manage risk with a combination of all three basic risk management strategies: limiting exposure to specific energy sources, diversifying, and engaging in extensive trading activities to hedge several types of risks across their operations.
But what options are available specifically for hedging risk from wind variability, i.e. weather-related ‘volume’ risk? How much risk can realistically be hedged through financial risk transfer mechanisms? What instruments are used? Which counterparties perceive rewards in shouldering this risk?
A Lack of Awareness
Wind power now accounts for a high proportion of generation capacity in many regions. For example, in the Texan transmission region managed by ERCOT (85 percent of Texas’ load), wind’s share of the 84 GW electricity generation capacity is 13 percent. Wind accounted for 17 percent of Germany’s 167.8 GW of installed generation capacity in 2011 – but only 8 percent of total power generation (579.3 TWh).
A 2011 survey by the Economist Intelligence Unit and Swiss Re estimates that though 60 percent and 48 percent of respondents used insurance and financial derivatives to manage financial risk within the renewable energy sector, only 4 percent of wind power-focused respondents bought insurance to protect against volume risk. Yet 18 percent of wind sector survey respondents described volume risk as a “high” risk (in contrast to 7 percent of respondents from the solar sector) and 47 percent described it as a “medium” risk. Lack of awareness of weather-based hedging instruments was cited as one of the top three barriers to effective risk management with only 1 percent of all respondents using derivatives to manage volume risk.
CelsiusPro AG, a Zurich-based originator of over-the-counter (OTC) weather derivatives, offers Low Wind Day and Low Wind Season certificates in addition to products for temperature, rain, snow and solar radiation. Founder and CEO Mark Rueegg says that wind hasn’t been a significant growth sector. “Wind is the only production factor for a wind power company,” he says laconically, “so I’m a bit puzzled as to why there hasn’t been more of an uptake.” While these certificates, which offer a payout in the event of below average wind speeds, have found a handful of customers in Northern Europe, unlikely candidates such as crane or ski-lift operators have instead expressed interest in protection against high wind.
Wind speed variability has many financial consequences. Low wind speeds reduce generation and revenues for wind power generators and may adversely affect the ability to meet debt payments, creating credit risks for the financier. In the U.S., with incentives like the Production Tax Credit (a federal tax incentive of US2.2 cents/kWh of electricity produced, recently renewed until the end of 2013) and Renewable Energy Credits, reduced production reduces revenue.
Conversely, excessively high wind speeds may temporarily halt generation or delay wind farm construction. Long periods of high wind speeds increase costs for baseload generators, for example, when they coincide with off-peak demand or, as recently happened in Germany, when reserve power plants had to be activated to stabilize the grid. When wind has priority access to the grid, thermal power plants have to balance generation regardless of whether wind is above or below forecasted levels. Wind speed variability may also compound price risk for other market players through its influence on wholesale electricity market clearing prices in competitive day-ahead and intraday markets. In some U.S. regions, wind generators sometimes bid zero or negative power prices in wholesale power markets to ensure electricity dispatch.
The Strategies: Derivatives and (Re)insurance
Hedging solutions for wind volume risk can be executed in either derivative or (re)insurance formats; the underlying mechanism is similar. Derivatives (options, swaps, futures) derive value from underlying assets. With wind derivatives, the “underlying” used in pricing and structuring, is often an index which models a wind farm’s output based on turbine power curves and historical wind speed data. Contracts pay out if wind speed is above or below a pre-specified trigger or strike. Index guarantees can be structured as put or call options which protect against unexpected outcomes.
Puts pay out if the wind index falls below the agreed strike, whereas a call structure might be used during construction to mitigate risks from high winds that halt construction. In this case calls pay out if the wind speed exceeds the strike value. With index exchange products such as swaps and collars, either party could pay the other depending on the direction in which the index moves. In swaps, any move away from the strike requires a payout from one party to the other, depending on the direction. A collar is a form of a swap with a low and high strike at either end of a range in which no payments are made. It could be used to limit the downside risk at little or no cost, in exchange for loss of the upside.
Lars Deckert, managing partner at 4initia GmbH, which provides consulting and asset management services on all aspects of wind project development (including risk management and derivative structuring) explains that currently, wind derivatives are most likely to be structured as puts, in that they pay out if wind speed drops below a certain level. “With swaps, counterparty risk is too high given that wind farms are usually non-recourse financed with a limited amount of equity,” he adds.
CelsiusPro’s certificates are indexed to wind data from a weather station of the buyer’s choice. However, Rueegg speculates that potential clients may be deterred by the basis risk that actual wind speed at the generation site varies from the speed as measured at the met station, thereby affecting the potential payout.
The alternative is a customized product. 4initia is currently involved in structuring its first derivative deal for a client and hopes to make an announcement within a few months. Deckert says, “We’re looking for way to make the product more durable for investors, i.e. reducing or even taking away the basis risk. Currently the market is for bespoke or highly customized products since there are so many input factors specific to a wind micro-site.”
Pricing and Structuring
While the underlying logic is simple, the devil is in the details when it comes to pricing and structuring the derivative: identifying the strike, the payout, maximum limit, the calculation period, and so forth. Reducing basis risk requires site-specific long-term wind data – which isn’t usually available. Derivative pricing is all about the price volatility of the underlying assets: standard option pricing models make several simplifying assumptions about expected volatility like that of geometric Brownian motion.
But taking wind for a random walk isn’t quite that simple: the critical parameter is the underlying wind data which cannot be modeled accurately with just mathematics. Without reliable wind data, there’s little basis with which to structure or price a contract.
This places weather data companies such as Seattle-based 3TIER at a pivotal point in the fledgling derivatives space. A provider of wind, solar and hydro data, 3TIER has proactively positioned itself as leader in the renewable energy risk analysis space. It uses models using inputs from 50 years of weather archives; terrain, soil, and vegetation data; and on-site observations (if available).
“It all starts with our long-term understanding of the wind,” explains Michael Grundmeyer, Managing Director Risk Products. He describes how 3TIER’s synthetic datasets are combined with onsite observations to generate historical times series, providing a 30-year hourly history of wind speed for any point in the world. “We did a study with a large North American wind developer who has 300 met towers and four years of data. We used this data to calibrate our models for their projects, so called MOS-corrected data [MOS: Model Output Statistics] and it is as good as quality-controlled met data for understanding climate variability. We get really good with even just one year of data to calibrate our models.”
In May 2011, Galileo Weather Risk, a provider of weather risk protection products announced WindLock: “the first financial products designed to assist the wind power industry in hedging the financial risk of wind variability.” The contracts (transacted either as derivatives or (re)insurance) are based on an index of modeled output from a wind farm as driven by average or hourly wind speed. 3TIER’s time series were used to price and structure WindLock.
“When structuring the derivative, the key issue is the floor for the customer,” says Grundmeyer. “What is the pain point? For some it could start at P60 or P75, maybe going down to P90.” 3TIER isn’t usually directly involved in establishing the strike. “We only get involved at a high level in establishing the pain point. The pain point is almost always correlated to variability of power production, which is mostly a function of weather and climate variability for wind projects. It is in providing this understanding of variability that 3TIER’s data is used throughout the deal flow. In half of the deals we know what strike they settle on, but not always.”
Jürg Trüb, head of environmental and commodity markets at Swiss Re Corporate Solutions, explains that Swiss Re’s hedging solutions for wind volume risk are structured as derivatives. “Typically, strike levels are chosen to be P95 or P99 but could of course be closer to the money.” He concurs that the underlying hedge mechanism is quite similar for insurance contracts and derivative instruments, while emphasizing the difference with respect to tax (an insurance premium tax versus none for derivatives), accounting implications and payout function. “Insurance is similar to options but the payout functions of swaps or collars can’t be replicated through insurance structures which require fixed premiums and variable payouts,” he states.
Other than structuring, 3TIER could also be involved in tracking and settlement for its partners. “Since realised production in MWh can be affected by several factors other than wind speed, it’s the met tower which becomes the primary settlement mechanism,” says Grundmeyer. “But this too has the potential for failure and the theoretical potential for moral hazard (tampering with the anemometer). You need multiple layers of redundancy. We are constantly reconciling power production with wind speed, looking at turbine performance – all the different datastreams,” he elaborates.
Risk Transfer: A Tough Sell
Deckert points out that while insurance is usually just for hedging risk exposure, derivatives have traditionally supported a speculative function too in facilitating the risk transfer from the risk averse to those with an appetite for it. But other than insurance companies, which other counterparties could perceive some reward in taking this risk on? “Perhaps hedge funds could be interested in risk exposure outside their traditional areas,” he suggests.
Nephila Capital Ltd. is a Bermuda-based hedge fund which specialises in weather-linked risk transfer instruments and also provides customised products for hedging wind volume risks. Barney Schauble, managing partner at the fund’s affiliate in San Francisco, asks, “How can you justify not hedging the volume risk? If there were a typical list of items to execute for wind financing, why wouldn’t this be a standard item on the list?”
But wind derivatives have a checkered past and their case hasn’t been helped by the image of derivatives in general needing a little airbrushing following the financial crisis. Deckert recalls an initiative by major investment banks a few years ago to implement wind derivatives: “The basis risk was too high because the indices were based on met stations that were too far away from the wind generating sites and wind farm owners lost money.”
Standardized exchange-traded products have a similarly spotty story. In 2007, the U.S. Futures Exchange announced it would offer wind futures: seven contracts from wind farm regions in Texas and New York were to be traded with monthly expiries. The exchange itself closed down a year later and the futures became history.
So what’s changed in the intervening years to make derivatives more feasible? In a word: data. Richard Oduntan, a financial engineer and structured trader at Nephila’s Bermuda office, says that even a few years ago, data quality wasn’t good enough to support accurate risk assessment. “Now data companies are coming to the space so we can do risk analysis with synthetic data sets.” The bulk of Nephila’s contracts have been with developers at the project financing stage. “As time frames for financing have shortened to five to seven years, this brought buyers back to this market.”
Yet all agree that it’s still a tough sell. Grundmeyer, who also works in solar and hydro, agrees that among renewable energy sources, wind is special in terms of its capacity for extreme variability. “It should be an easier sales pitch to the developers and offtakers,” he says. “It’s a double-edged sword – everyone understands the variability but, on the other side, there’s a lack of follow-through in trying to understand the insurance or derivative product. Perhaps there is a fear of overpaying for something you may or may not need.”
Customization – presumably – comes at a high cost that limits market growth for these products. Oduntan says, “While most of our risk is transferred in the form of customized transactions, we’re looking for ways to scale the market, to get traction with something that several counterparties can use.”
Changing regulations are also shaping the market for wind volume risk protection. With Germany’s renewable energy law under review and the scaling back of FiTs across Europe in the wake of the financial crisis, demand for risk protection can only increase. “With FiTs, there is no price risk for the generator,” says Deckert, “but if generators were also exposed to price volatility, there would be more pressure to deal with the volume risk.”
In Trüb’s opinion, “developments such as reduced FiTs, etc shift more of the risk to equity invested in renewable energy. Investments are subject to stable return projections; hence de-risking projects becomes increasingly important. Additionally, in countries like Germany it’s now apparent that intermittent renewable energy production heavily affects the grid and the potential to make money out of thermal power plants. Therefore we also see a move towards the introduction of regulation that addresses these effects and believe that hedging volatility in wind power production will become more popular.”
Rachana Raizada is a freelance journalist focusing on the energy sector.