1c6n4g0og 217796

qwertyuiopasdfghjklzxcvbnmqwertyui
opasdfghjklzxcvbnmqwertyuiopasdfgh
jklzxcvbnmqwertyuiopasdfghjklzxcvb
nmqwertyuiopasdfghjklzxcvbnmqwer
Financial Trading Instruments
tyuiopasdfghjklzxcvbnmqwertyuiopas
ETRM Academy
dfghjklzxcvbnmqwertyuiopasdfghjklzx
www.etrmacademy.com
cvbnmqwertyuiopasdfghjklzxcvbnmq
wertyuiopasdfghjklzxcvbnmqwertyuio
pasdfghjklzxcvbnmqwertyuiopasdfghj
klzxcvbnmqwertyuiopasdfghjklzxcvbn
mqwertyuiopasdfghjklzxcvbnmqwerty
uiopasdfghjklzxcvbnmqwertyuiopasdf
ghjklzxcvbnmqwertyuiopasdfghjklzxc
vbnmqwertyuiopasdfghjklzxcvbnmrty
uiopasdfghjklzxcvbnmqwertyuiopasdf
ghjklzxcvbnmqwertyuiopasdfghjklzxc
vbnmqwertyuiopasdfghjklzxcvbnmqw
ertyuiopasdfghjklzxcvbnmqwertyuiop
Describe a virtual power plant.
______
A type of power plant modeled with options, locking in the processing spread (crack spread,
spark spread, or dark spread) with futures and/or options. Called a "paper refinery" in the case
of a crack spread.
List the types of energy derivatives (table).

______
Describe a forward contract.

______
A forward contract is a bilateral agreement to purchase or sell a certain amount of a

commodity at a fixed future date (delivery date) at a predetermined price. The seller of the
forward contract has the obligation to deliver the commodity. The payment date is usually at
or near the delivery date so that no cash changes hand until delivery.
This is your personal copy.

PLEASE DO NOT REPRODUCE, REPUBLISH, UPLOAD, POST, TRANSMIT, OR DISTRIBUTE.
3
Forwards are OTC contracts. They can be tailored but are more difficult to re-sell. The main
risk is credit risk. Forwards are used to:
1. Hedge the obligation to deliver or purchase a commodity in the future.

2. Secure a sales profit from commodity production.
3. Speculate on rising or falling commodity prices in case there is no liquid futures market.
Calculate the fair value of a forward contract.

______
The fair value of a forward contract is:
1. (one delivery date)

U: fair value of forward, price a neutral investor would pay to enter into the forward as a buyer
r: risk-free rate
T: duration of the forward, t: time expired since initiation
F: current forward price, K: contract price
-r(T-t)
e : discount factor
2. (several delivery dates, i.e. swap)

4
The fair price of a forward contract is the price where the value is zero:
= sum of discounted prices / sum of discount factors
Solution:
= sum of discounted prices / sum of discount factors
Describe a futures contract. Which risk is inherent?

______
Standardized forward contracts traded at commodity exchanges with a clearing house as the
central counterparty for all transactions. Obligation to buy or sell underlying in the future. No
counterparty risk (as in forwards). To protect the clearinghouse from losses, traders must pay an
initial margin as guarantee. Each day, futures contracts are marked-to-market with gains and
losses immediately realized, so that losses cannot escalate. Futures contracts are often settled
financially, but they carry basis risk. Futures and forward prices can be seen as equivalent.

5
What is the change of fair value of a forward contract
following the change of the contract price (formula)? Describe
the difference between hedging with forwards and ...
______
The change of fair value of a forward following the change of the contract price ΔF is:
From a futures contract, we immediately receive ΔF as a margin payment, which can be

invested in a risk-free asset.
Sensitivities to the underlying forward price are different in forwards and futures, even
though their prices coincide.

6
Describe a commodity swap, and calculate the fair value of a
swap and the par rate (formulae).
______
The fair value of a swap is equivalent to a forward with several delivery dates:
F: forward price
K: contract price (strike)
-r(T-t)
e : discount factor
The fair swap rate (par rate) is also equivalent to forward with several delivery dates:
We assume U (fair value) = 0.
Swaps are the newest derivatives. They are used to lock in a specific commodity price over a
period. Mostly traded OTC and settled financially based on a commodity index. In a plain
vanilla swap, the swap payer pays a fixed price (he holds a payer swap), while the receiver pays
a floating price at specific dates (he holds a receiver swap), netted against each other. The fixed
payments are called the fixed leg of the swap, the floating payments the floating leg of the swap.
The term of a swap can be up to 30 years.

7
Describe payoff profiles of plain vanilla call and put options.
______
As a consequence of put-call parity, an energy call option

can be replicated by ...
______
A call option can be replicated by a put option and a forward contract:
F: current underlying forward price

K: strike price

8
Describe the payoff profiles of a protective put and a covered
call.
______
Describe Asian options, calculate their payoff (formula) and

list their advantages (2).
______
For Asian options the payoff is determined by the average underlying price over some pre-set
period of time. This is different to the case of the usual European option and American option,
where the payoff of the option contract depends on the price of the underlying instrument at
maturity. Asian options are common in energy markets. They can be priced with a Monte Carlo
simulation model.
Call payoff = max [avg(F(T)) - K, 0]

Put payoff = max [K - avg(F(T)), 0]
avg(F(T)): arithmetic average of spot prices. Very rarely, geometric mean is used.
Advantages of Asian options are:
1. Averaged price is useful to hedge for example fuel over a period of time. More weight
can be given to closer periods. Budget targets and planning periods rely on price
averages, so Asian options are very important for energy.
2. Reduced risk of market manipulation of the underlying instrument at maturity.

9
3. Because of the averaging feature, Asian options reduce the volatility inherent in the
option; therefore, Asian options are typically cheaper than European or American
options.
4. Easier to hedge for the option writer, because they have low gamma.
Calculate the payoff for energy call options and put options,
and describe their inner value.
______
Energy options are usually on futures contracts.
Call option:
Put option:
The inner value of options describes their payoff if they are exercised before maturity, i.e. at
time t instead of T. It is a martingale.
Call option:
Put option:
Explain a martingale.
______
A martingale is a stochastic process (i.e., a sequence of random variables) such that the
conditional expected value of an observation at some time t, given all the observations up to
some earlier time s, is equal to the observation at that earlier time s. A martingale is a model of a
fair game.
designates the expected value under risk-neutral conditions.

10
Describe how the inner value and the price of options are
related.
______
Inner value of options describes their payoff if exercised before maturity.
Inner value call option =

Inner value put option =
Relationship between inner value and price of European options:

11
Describe the standard methods (2) used to price American
options.
______
1. Binomial/trinomial trees. Discrete time models with backwards iteration.
2. Finite-difference methods. Numerical approximations of the partial differential equation

satisfied by the options price.
American options (and Asian options) are path-dependent. The Monte Carlo model does not
work well to price American options.
Describe the main processing spreads (3) that market

participants in energy markets are exposed to. What is the
payoff of a spread (formula)? How is related to correlation?
______
1. Dark spread. Between power and coal, modeling a coal fired power plant.
2. Spark spread. Between power and gas, modeling a gas fired power plant.
3. Crack spread. Between different refinement levels of oil, modeling a refinery.
Such spreads can be hedged by selling/buying futures or by selling a spread option to directly
offset the risk. The payoff of a spread is:
The higher the correlation between the two underlyings, the lower the value of the spread,
because there is less risk (volatility) involved when correlations are high.

12
Describe how to price spot options.
______
The true spot option value lies between the lower bound (IV) and the upper bound (Monte
Carlo simulation):
The intrinsic value of the option gives a lower bound on the option value:
hiF: payoffs, Φ: exercise strategy with Φ=1 (exercise) and Φ=0 (not exercise)
The upper value is calculated using Monte Carlo simulation:
If S(t) are simulation paths, the Monte Carlo approximation of the upper value is:
where
For each path, a linear optimization problem must be solved.
Describe how to use the Black model to price energy call and
put options.
______
Energy options are usually on futures contracts, so they are priced with the Black model:
Call option price =
Put option price =

with:
and and

13
Describe the five types of financial risks faced by energy
companies.
______
1. Market risk, price changes of financial assets and liabilities.

2. Credit risk, defaults on contractual obligations.
3. Liquidity risk, lack of market activity, failure to meet a cash flow obligation.
4. Operational risk, technical problems with financial trading systems and fraud.
5. Legal risk, losses from failures to comply with the law or adverse regulatory changes.
Another risk is downside risk related to volatility. The desire to transfer this risk has led to the
creation of a financial derivatives market in 1973. Most energy derivatives are built on futures,
forwards, and options.
Energy options are usually written on ... .

______
Energy options are usually written on futures contracts. One option controls one futures
contract.
Which is the largest exchange for trading energy futures and

options in the world?
______
NYMEX, accounts for 80% of all energy futures and options traded.
Describe the main benefits (2) of marking-to-market futures

contracts.
______
1. Reduction of default risk for derivatives.

14
2. Potential to hold contracts for small margin.
Example of MTM:
One crude oil futures contract is for ... bbl of crude oil.
______
One futures contract is for 1,000 bbl of crude oil (contract size).
If someone buys a contract for $30, he will have to take delivery of 1,000 bbl at expiration
and has to pay the spot bbl price x 1,000. If the price rises, the long gains and the short loses.
Describe a clearinghouse.
______
Futures exchanges are clearinghouses that act as intermediaries. They buy contracts from the
sellers and sell them to the buyers, thereby ensuring performance. Sellers and buyers are kept
anonymous to each other.
If one party fails to perform, the clearinghouse acts as insurance. It also marks-to-market
contracts at the end of each trading day to minimize default.
Futures contracts take delivery on only about 5% of contracts, so they are mainly
financial contracts used to hedge price risk, and are referred to as "paper barrels".

15
Explain the meaning of daily price quotes (5) on NYMEX.
______
1. Open: First price of the day.

2. High: Highest price of the day.
3. Low: Lowest price of the day.
4. Settle: Last price of the day.
5. Change: Change in the settle price between yesterday and today. $0.009 is the maximum
for months beyond the first two.
Open interest is the number of contracts outstanding at the close of the day.

16
Describe how the Brent forward market differs from other
forward markets (3).
______
1. Brent forward contracts are standardized (as opposed to other forward contracts, which
are not).
2. There is no clearinghouse involved, so default risk prevails.
3. The Brent forward market is also called the Brent 15-day market, as 15 days notice is
required to schedule tankers to pick up the crude at the Solum Voe terminal in the North
Sea. The contract can only be traded until the 15th day of the month before delivery.
Cargoes scheduled for delivery can still be traded on the spot market and are called
"dated Brent".
Describe how energy futures and forwards differ (3).

______
1. Exchange/OTC. Forwards are traded OTC, are not standardized and have no
clearinghouse (default risk). Futures are standardized and have a clearinghouse.
2. Price transparency. Futures are more price transparent.
3. Hedging sensitivity. Different hedging sensitivity, as futures contracts are marked-to-

market daily.
Calculate the profit/loss on a hedged position (formula).

______
Profit/loss = Gain/loss spot + gain/loss future = [ S(T) - S(t) ] + [ F(T) - S(t) ]

17
Describe what sets prices of energy futures.
______
The futures price formula for commodities is:
with δ: convenience yield in %
with U: nominal storage cost
with μ: storage cost in %
If the futures prices are lower than the value calculated with the price formula, arbitrageurs will
buy the future and sell short spot, thereby bidding up the futures price until equilibrium is
reached.
In equilibrium:
Describe the conditions of (r + µ - d) for contango and

backwardated markets. How is the term structure of the
markets related to convenience yield?
______
If (r + μ - δ) > 0: Contango (normal). (Low) convenience yield < carrying cost.

Traders bid up the price of the futures above the spot price.
If (r + μ - δ) < 0: Backwardation (inverted). (High) convenience yield > carry cost.
Traders bid up the price of spot prices above the futures price.

18
μ: storage cost in %
δ: convenience yield in %
r+μ: carrying costs
FT

19
Calculate convenience yield.
______
1.
2.
By solving: or
Explain whether energy markets follow the efficient market

hypothesis. Is the natural gas market efficient?
______
According to the efficient market hypothesis, futures prices plus a risk premium would be a good
predictor of expected spot prices.
If the market were to be efficient, there would be no risk premium and hedgers on both
sides would dominate the market (no speculators). This could be tested with:
, test for: and

20
Herbert (1993) performed this test for the U.S. natural gas futures market and rejected the null
hypothesis, therefore found markets to be inefficient. There exists a risk premium and profits can
be made systematically.
Describe and calculate a crack spread (formula).

______
Buying crude futures and simultaneously selling product futures.
Crack spread = weighted average value for products - price of crude
Weighted avg. value products = 0.59*35.57 + 0.1*35 + 0.26*35 + .05*24.88 =

$34.84 Crack spread = $34.84 - $28.1 = $6.74.
In reality, the crack spread is slightly different, as it ignores some parts of the barrel. A typical
spread will be 3-2-1, with 3 bbl of crude for 2 bbl of gasoline and 1 bbl of heating oil
(distillate). A refinery seeking to hedge would sell a crack spread to buy crude oil and sell
products, with the spread sold closest to the product slate.

21
Describe a spark spread.
______
Shorting electric power and buying fuel used for generation (fuel oil, natural gas, coal). Since
2002, spark spreads became OTC (before NYMEX).

22
A call option contract on natural gas futures gives the holder the
right to buy ... . Total option cost equals ... .
______
10,000 MMBtu of natural gas at the strike price any time up to four business days prior to the
delivery month. The contract size is 10,000 MMBtu, the contract is quoted in $/MMBtu.
Total option cost = option price ($/MMBtu) x contract size (10,000 MMBtu).
In an option transaction, which party must post margin?

______
The seller of the option must post margin if the option is written on an exchange.
The buyer has to pay the full option cost (option price x contract size) and cannot use margin.
Most OTC options are ...

______
European options (= plain vanilla options).

23
Describe the payoffs of long and short call and put options
(European).
______
Long call payoff = max [ST - K, 0] → infinite possible gain

Long put payoff = max [K - ST, 0]
Short call payoff = - max [ST - K, 0] → infinite possible loss
Short put payoff = - max [K - ST, 0]
European options are only exercised at expiration, so the payoff is their maximum or minimum
value and not discounted with the risk-free rate (which would be their value). Net payoffs
(profits), taking into account options premium, are:
Long call payoff = max [ST - K - CT, -CT] → infinite possible gain
Long put payoff = max [K - ST - PT, - PT]
Short call payoff = - max [ST - K - CT, - CT] → infinite possible loss
Short put payoff = - max [K - ST - PT, - PT]
Describe how to calculate option values with the binomial

model (2 approaches).
______
Assuming risk-neutrality (π = 0.5), option values can be calculated by modeling a

portfolio consisting of a long position in stock and a short position in a bond.
= value of option.
The number of stock to buy is given by:

cu: call value up move, cd: call value down move
U: up movement multiplier (1+n), D: down movement multiplier (1-n), St: beginning stock price
The value of the bonds to sell is given by:

cu: call value up move, cd: call value down move
U: up movement multiplier (1+n), D: down movement multiplier (1-n), R: risk-free rate multiplier (1+rf)
If we take into account the probability of an up or down movement, the call option value would be calculated in
the following steps:
+ -
1. Calculate u, d: u = S /S and d = S /S = 1/u

24
2. Calculate risk-neutral probability π:
+ - + + - -
3. Calculate c , c : c = max(0, S -X) and c = max(0, S -X)
4. Calculate call value:
Both approaches will give the same result.
Calculate the value of a put option with the binomial model,

example.
______
Stock values are given by the following lattice:
Assuming a strike price of 101 and a risk-free rate of 5%, calculate European and American
put option prices.
Answer:
Assuming a strike price of 101, European put option values are:

25
Assuming a risk-free rate of 5%, we are calculating probabilities, p = (1+r-d) / (u-d), with u =
1.1 and d = 0.9:
2 2 2
D: 0.75 , E: 1-0.75 , F: 0.25
Inferring backward current values of put options weighted by probabilities:

26
For American options, we compare the discounted value with the value of exercising and
choose the higher. We would get:
General steps:

27
Describe the two ways to calculate up and down moves (U,
D) when valuing options with the binomial model.
______
1. If we know by how much the price goes up or down, we use the formulae:
o U = (St+1, up / St) = 1 + n
o D = (St+1, down / St) = 1 - n
2. If we do not know by how much the price goes up or down, we use the formulae:
σ√T
o U=e , with σ being the volatility of past returns (daily or other period)
o D=1/U
p = (1 + r - d) / (u - d)
Explain which underlying variables (5) affect option prices

and the direction of the effect on calls and puts.
______
1. Asset price
2. Exercise price
3. Asset risk (volatility)
4. Time until expiration

28
5. Interest rate
Describe the terms outlined in an energy swap (4).

______
1. Amount of the commodity covered in the swap.

2. Length of the agreement. < 3 months is short-term, long-term is 6 months - 30 yrs.
3. Price that serves as basis for floating rate payments.
4. Frequency of payments.
The agreement does not require upfront payments.
Swaps can provide a much longer protection (up to 30 years) than futures.
Describe parallelism as it pertains to futures markets.

______
The same factors that impact spot prices (cash prices) also impact futures prices, they have a
high correlation. The close relationship is forced by arbitrage. If cash prices change, future
prices change by about the same amount. The basis (F-S) is therefore relatively constant.
The principle of parallelism of cash and futures prices allows hedging in the futures market to be
effective (but not perfect). Parallelism reduces variability.

29
Mostly, future > spot (trades at a
premium).
Describe convergence as it pertains to the futures market.

What happens to the basis?
______
Approaching expiration, cash and futures prices converge and eventually become the same. At
expiration, owning the futures contract is the same as owning the underlying commodity, thus it
must have the same price. The basis (F-S) converges towards zero as expiration nears.
Describe contango markets.

______
The futures trade at a premium over spot. Contango markets are also called premium or carrying
charge markets. Futures prices are related to the cost of carrying physical commodities as inventory,
such as storage, insurance, and interest cost on funds borrowed to purchase inventory.

30
The price of futures is bid up (F < S) above the spot price as holders of inventory prefer futures
to buying inventory. Holders also may sell some of the cash product, causing the spot price to
fall. Equilibrium is reached then the futures price adequately reflects the spot price plus cost of
carry, a state called "full carry". The premium equals the cost of carry.
Describe full carry. What is the normal situation in futures

markets?
______
In a contango market, the price of futures is bid up above the spot price as holders of inventory
prefer futures to buying inventory. Holders also may sell some of the cash product, causing the
spot price to fall. Equilibrium is reached then the futures price adequately reflects the spot
price plus cost of carry, a state called "full carry". Premium = r + μ = cost of carry.
Futures markets are seldom at full carry, as it is usually advantageous to hold at least some of
the physical product, reflected in the convenience yield (δ). The normal situation for futures is to
sell at a premium but not at full carry.

31
Describe the components of carrying costs of commodities
(3).
______
1. Storage cost. Relatively stable.

2. Insurance. Minimal cost, relatively stable.
3. Interest cost for borrowing funds to buy inventory over the period. The most volatile
component. Usually at prime interest rate + 1 or 2 %.
Cost of carry = r + μ.
Describe backwardation.
______
Backwardated markets are also called inverted markets, because they are downward sloping
and F < S. They are caused by a shortage of supply relative to demand in cash markets, spot
prices are above futures prices, which encourages sales in the present rather than in the future.
The market discourages storage of goods. Crude oil futures markets are usually in
backwardation, because refineries constantly need to purchase oil for immediate consumption
in the spot market, putting upward pressure on prices.
Markets can change from a carrying charge market (contango) to a backwardated market based
on seasonal demand.

32
Describe cash market arbitrage.
______
Riskless operation that allows traders to profit from price differences in different markets for the
same commodity. Markets are usually separated geographically. Arbitrage opportunities occur
when prices differ by more than transportation costs between the markets. The trades must be
executed simultaneously.
Doing this trade actually diminishes the price differential between the markets, making them
more efficient, but eliminating the profit opportunity. Market forces will cause prices to
converge.
Describe cash/futures arbitrage.

______
In a contango market, if futures prices rise above cash prices by more than the carrying charge (r
+ μ), there will be an opportunity to arbitrage cash and futures markets, purchasing the cash
commodity and simultaneously selling futures. The commodity is then financed and stored until
expiration when it is delivered, yielding a profit (risk premium - carrying charge). This trade
also causes prices to converge. ~ Cash and carry arbitrage.

33
Futures markets are rarely at full carry, and even more rarely does the premium exceed the
carrying charge. Price differentials between nearby and deferred contracts are seldom more than
carrying costs.
Describe the basis (2 formulae) in futures markets and its

components (3).
______
Basis = F - S = futures price - local spot price = storage basis + location basis.
Usually, the first nearby contract is used to calculate basis. Because cash prices vary depending
on location (and transport fees), there is a unique basis for each location. Basis may be split into
three components:
1. Storage basis = F - delivery point spot price. Because carrying charges diminish as
expiration nears, storage basis diminishes and is zero at expiration. In contango, it
diminishes at the rate of the storage cost, in backwardation storage basis is unrelated to
carrying charges.
2. Location basis = delivery point spot price - local spot price. Also called transportation
basis. Different local cash prices related to transportation costs and supply/demand in
each location. Location basis is relatively stable.
3. Product basis = F - spot of a similar commodity. Other products with a consistent
product basis (stable and strong correlation) can be used to hedge exposure, e.g.,
Heating oil and gasoline futures can be used to hedge kerosene.

34
Describe how basis impacts hedging.
______
There is a very close relationship between basis and the effectiveness of hedging when the
basis is stable, as spot gains/losses equal futures gains/losses.
Describe how changes in basis can be split (2).

______
Changes in basis can be split into:
1. Systematic changes, caused by the diminution of carrying charges over time. In

contango markets at full carry, the cash and futures price will converge (basis will
narrow) at about the rate of the carrying cost per unit time. If not at full carry, the basis
will narrow at < rate of carrying cost. In backwardated markets, the basis will narrow at
unpredictable changes.
2. Unsystematic changes. Random, caused by differential changes in spot and futures
prices.
At which rate will the basis narrow in contango (2) and

backwardated markets?
______
In contango markets at full carry, the cash and futures price will converge (i.e., the basis will
narrow) at about the rate of the carrying cost per unit time (r + μ).
In contango markets not at full carry, basis will narrow at a rate < carrying cost.

35
In backwardated markets, the basis will narrow at unpredictable changes caused by the
relative price changes of futures and spot prices.
Compare futures spread trading to arbitrage.

______
Spread trading takes both a long position and a short position in different futures contracts of the
same or related commodities. It is more risky than arbitrage (which is risk-free), but still less
risky than outright long or short positions in futures (position trading). Even if the relative price
changes, it may fluctuate in the mean time, resulting in margin calls for the spread-trader.
25% of all commodity futures volume is estimated to be spread volume.
Spread trading is not concerned with absolute price changes (as they will be offset), but only
with relative price changes, like arbitrage. Only the price difference and its change between
contracts matters. Margin requirements on qualified spreads are much lower than for long or
short futures contracts.
A scalper provides ... to the market.

______
A scalper provides liquidity to the market.
Scalpers are inside speculators (locals on the exchange), who profit on very small price
differentials on buying up futures on small price concessions on large orders and selling
them back into the market over time.

36
Why are inside speculators necessary for the smooth
operation of a market (3)?
______
Inside speculators are locals (on the exchange) who trade different strategies (spread-trader,
day-trader, position-trader), bear risk, provide liquidity, increase efficiency.
1. If speculation is banned from futures markets, liquidity and open interest necessary
for effective hedging disappear.
2. Hedgers are usually net short (want to sell futures), so they need counterparties for those
short trades.
3. Speculators also help translate new information into fair prices by acting fast in their own
self interest.
Explain the three types of futures spread positions.

______
Futures spread positions may be initiated in:
1. Futures contracts for different but related commodities (intermarket spread).
2. Futures contracts with different delivery months (intramarket spread).
3. Futures contracts for the same commodity traded on different exchanges (intermarket
spread).
Explain "spread rule 1" and "spread rule 2" in the futures
markets.
______
Spread rule 1: If the spread between two contracts narrows, a profit occurs if the lower-priced
contract has been purchased and the higher-priced contract has been sold.
If spreads are expected to narrow, buy low, sell high.
Spread rule 2: If the spread between two contracts widens, a profit will occur if the lower-
priced contract has been sold and the higher-priced contract has been purchased. If spreads are
expected to widen, buy high, sell low.

37
The spread rules generally work, but sometimes involve changing of the sign if
directions reverse.
Describe intramarket spreads, and two types.

______
Intramarket spread are also called intracommodity or time spreads. They involve the
simultaneous purchase and sale of futures contracts on the same commodity for different
delivery months. Spreads between futures contracts for different delivery months are related to
carrying charges. Two types:
1. Bull spread: The trader expects the nearby contract to gain on the deferred contract. The
general rule on a bull spread is to buy the nearby and sell the deferred contract, which
holds true for a contango and backwardated market.
2. Bear spread: The nearby contract is expected to decrease relative to the deferred
contract, so the rule is to sell the nearby and buy the deferred contract in a contango
market when spreads are expected to widen, and to buy the nearby and sell the deferred
contact in backwardation when spreads are expected to narrow.
In the case of cash/futures arbitrage, if traders think the spread between two delivery months is
too large, there will be an incentive to purchase nearby contracts (buy low), take delivery and
store the goods in order to deliver them on deferred contract (sell high). They apply spread rule
1. If spreads between months are too narrow, traders will prefer to purchase futures contracts
(buy high, sell low)) to lock up their future supply rather than paying carrying charges to
store the physical commodity. They apply spread rule 2.

38
Describe intermarket spreads (7 types).
______
Intermarket spreads involve simultaneous purchase and sale of different but related
commodities that have a stable relationship between them, with same or different delivery
months. Opportunities arise when commodities are substitutes for each other or prices are
correlated for some other reason, and when futures prices diverge because of some random
supply/demand imbalances in cash and futures markets. With the 13 current energy futures,
different delivery dates and exchanges, thousands of spreads are possible, but not all have
sufficient depth and liquidity. Types of intermarket spreads are:
1. Crack spread (= paper refinery). Intercommodity spreads, involving crude, gasoline,

heating oil futures. Used to lock in the refining margin. About 2/3 of a crude oil barrel
produce gasoline products, about 1/3 heating oil: 3-2-1 = +3 crude, -2 gasoline, -1 heating
oil.

39
If the netback (= product prices - crude price) significantly rises above $4/bbl, it is
expected that the spread will eventually narrow and a crack spread is in order. If it
falls below $3/bbl, a reverse crack spread should be initiated as a hedge.
o A crack spread (buying 3 crude, selling 2 gasoline and 1 heating oil) is initiated
when prices of refined products rise, demand for crude increases and the spread
is expected to decrease. Product futures are sold one month after delivery of the
crude, imitating refinery time frames. Common ratios traded are 3-2-1, 10-7-3, 5-
3-2, and 2-1-1.
o A reverse crack spread (selling crude, buying products) is initiated when refined
product prices are low relative to crude and expected to rise.
2. Spark spread. Newest intermarket spread. Purchase natural gas futures and sell electric
futures. Used by utilities to lock in margin on generation. Common ratios are 3 natural
gas - 4 electric and 3 natural gas - 5 electric.
3. Heating oil vs. gasoline spread. Seasonal spread. Peak demand for heating oil is winter,
for gasoline summer. Spread consists of:
o Buying heating oil futures in the winter and selling gasoline futures for the same
months, hoping that heating oil demand will increase in winter and gasoline
demand decrease or remain unchanged.
o Buying gasoline futures in spring and sell heating oil futures, hoping that
gasoline demand will increase in summer. The gasoline leg should move up
faster than the heating oil leg.
4. Heating oil vs. gas oil spread. Contract specifications between exchanges (NYMEX,
IPE) vary and must be considered. Ratio usually 5 NYMEX - 7 IPE or 3 NYMEX - 4
IPE.
5. NYMEX vs. IPE crude oil spread. Between NYMEX light, sweet crude and IPE Brent
crude.
6. Frac spread. Between natural gas futures and propane futures. 50% of the world's
propane is extracted from natural gas.

40
7. Henry Hub natural gas vs. Permian/WAHA Hub natural gas spread. Henry Hub
trades on NYMEX, Permian/WAHA on KCBOT. NYMEX-KCBOT spread is also
known as the inter-exchange spread. KCBOT became inactive in 1999.
Describe a frac spread.

______
A frac spread is an intermarket spread between natural gas futures and propane futures. 50% of
the world's propane is extracted from natural gas.
1 barrel of oil = ... gallons of oil

______
1 bbl = 42 gallons
Often applied to 1,000 bbl = 42,000 gallons of oil in futures markets, since this is the
contract size for futures and options.
Crude is quoted in $/bbl and a contract size of 1,000 bbl, liquid products in $/gallon and
a contract size of 42,000 gallons.

41
Define netback in refining and how it can be locked in (2).
______
Netback = price of refined products - cost of crude oil.
The profit to the refinery, also called per barrel premium. Most refineries are profitable at a
netback over $4/bbl and lose money below $3/bbl. Crack spreads or crack spread options are
used to lock in the netback (refining margin):
If the netback significantly rises above $4/bbl, it is expected that the spread
will eventually narrow and a crack spread is in order.
If the netback falls below $3/bbl, a reverse crack spread should be initiated as a hedge.
Most refineries are profitable at a netback over ... and lose

money below ... .
______
Most refineries are profitable at a netback > $4/bbl and lose money < $3/bbl. Crack spreads or
crack spread options are used to lock in the netback (refining margin):
If the netback significantly rises above $4/bbl, it is expected that the spread
will eventually narrow and a crack spread is in order.
If the netback falls below $3/bbl, a reverse crack spread should be initiated as a hedge.

42
On NYMEX, one futures or option contract on heating oil is for
... . One futures or option contract for crude is for ... .
______
One futures or options contract on heating oil is for 42,000 gallons of heating oil. This is the
same as 1,000 bbl, but the prices are quoted in $/gallon.
Crude oil, on the other hand, is quoted in $/bbl, with a contract size of 1,000 bbl.
When calculating crack spreads, a per bbl or per gallon price must be established.
Futures on European exchanges (ICE) for crude oil are quoted per tonne, so a conversion factor
must be used to compare them to U.S. prices.
Calculate the breakeven price for call and put options.

______
BEP call option = strike + premium = X + c.
BEP put option = strike - premium = X - p.

43
Compare spread trading in options to spread trading in
futures. What do they profit from?
______
Option spreads are a means of speculating on the direction of commodity prices while limiting
risk. Option spreads attempt to profit from absolute changes in prices but with loss limitations.
Futures spreads attempt to profit from relative price changes, not absolute changes.
Describe an energy options bull spread. When is maximum

profit reached? How is payoff calculated?
______
In a bull spread, profit occurs when the underlying futures rise in price. Created by buying an
ATM (or OTM) call and selling an OTM call with a higher strike price ("supercap") and the
same expiration date. Upside potential and downside risk are both limited. Maximum profit is
reached when the futures price is at or above the OTM strike price. Payoff is calculated as:
Payoff = (spread x contract size) - spread cost. The spread cost is the maximum loss. In bull
spreads, usually calls are used, a bull put spread is also possible (requiring margin deposit).

44
Example:
Assume the following options prices for heating oil:
Calculate the profit from a bull call spread with strike prices $0.43 and $0.46.
Soultion:
Payoff = (spread x contract size) - cost = (0.03 x 42,000) - (0.019 x 42,000) = $462.

45
Describe an energy options bear spread. When is maximum
profit reached? How is payoff calculated?
______
In a bear spread, the trader hopes for a price decline. Buying an ATM (or OTM) put and
selling an OTM put with a lower strike and the same expiration date. The net cost is the
maximum loss and the maximum gain occurs when the futures price is at or below the strike
of the short OTM option. Payoff = (spread x contract size) - spread cost. The spread cost is the
maximum loss. A bear call spread is possible, but it requires a margin deposit.
Example:
Assume the following options prices for heating oil:
Calculate the profit from a bear put spread with strike prices $0.43 and $0.40.
Solution:
Payoff = (spread x contract size) - spread cost

46
Describe a butterfly spread and calculate its payoff. How
many options are involved?
______
A butterfly spread involves 1+1-2 call options (4 total) with different strike prices. Usually
entered by buying a call with a low strike and one with a high strike price while selling two
calls with strike prices in between. Generates a profit if the underlying futures contract does
not change much and leads to a small loss if a significant price move occurs.
Similar to a straddle.

47
Describe a straddle, and differentiate between long and short
straddles. They are ... plays. Contrast a straddle with a strangle.
______
A straddle involves a put and a call at the same strike price and expiration date. In order for a
long straddle to generate a profit, the futures price must change. Straddles are volatility plays,
expecting a large move with unknown direction.
Short straddles generate income and are subject to a margin deposit. They involve selling a
call and a put with the same strike and expiration. Risky because losses from a move in either
direction are unlimited.
A strangle is a straddle where the options have different strikes.
Describe a strangle. Contrast a strangle with a straddle.

______
A long (short) strangle involves the purchase (sale) of an OTM call and an OTM put with the
same expiration date but different strike prices.

48
A short strangle generates income but is subject to a margin deposit and unlimited
potential losses. A straddle is a strangle where options have the same strikes.
Differentiate between hedging with futures and hedging with

options.
______
Hedging with futures involves (perfectly) offsetting spot positions with futures contracts.
There should be zero gain and all movements (up and down) should offset.
Hedging with options is more like buying insurance. Their payoffs are asymmetrical and they
involve an options premium, which can be likened to an insurance premium. If options are used
on a consistent basis, they are expensive, especially if the risk of the underlying futures contracts
is high. Two cost-reducing strategies are:
1. Hedging cash market risk selectively, when there is fear of adverse price moves.
2. Hedging cash market risk partially, and looking for ways to reduce hedging cost
(cap/floor, bull/bear spreads).
Explain the use of a cap in hedging.

______
A cap is the purchase of a call option to limit the cost of acquiring product in the future. A
refiner is vulnerable to increases in crude oil prices as it reduces his profit. By buying a call,
he caps the price of crude to the strike price + call premium.

49
Cap prices can be reduced by:
1. Collar or fence. Selling a floor.

2. Bull spread. Selling an OTM call.
Cap prices are sometimes prohibitively expensive and can be mitigated by selling a floor (put),
i.e., giving up any gains resulting from a fall in prices. The refiner locks in a minimum price at
which he is required to buy crude. The combination of buying a cap and selling a floor is called
a collar or fence. In some cases, the net cost is zero. A collar will not be a true price insurance,
because price declines in crude will put the refiner at a disadvantage against his competitors who
may be able to lower prices.

50
Cap prices can also be lowered by entering into a bull spread: Selling an OTM call with a
higher strike than the cap and the same expiration.
Using a bull spread to reduce the cost of a cap leaves the refiner with a window of profit
potential, should product prices rise.

51
Explain the use of a floor in hedging. How can the cost of a
floor be reduced (2)?
______
A floor establishes a minimum price at which product may be bought or sold. It is a cap in
reverse.

52
Floor prices can be reduced by:
1. Collar or fence. Selling a cap.

2. Bear spread. Selling a further OTM put.
To mitigate the (sometimes prohibitively high) cost of a floor, a cap (call) can be sold at the
same time, which is called a collar or fence. The refiner thus limits the price at which he will sell
product to the upside. This will reduce the cost of the floor, but put a refiner at a disadvantage as
he forfeits any price gains from increasing market prices.
Another way to reduce the cost of a floor is a bear spread by selling an OTM put option with
a strike below the floor and the same expiration as the floor. This will leave the refiner with a
window where he can still profit from decreasing crude oil market prices.
Describe the use of bull and bear spreads to reduce the cost of
caps and floors.
______
Bull and bear spreads are used to reduce the high price of caps and floors by selling further OTM
calls or puts.
They lower the prices less than selling a floor (cap) or selling a cap (floor), but give the refiner
a window where he can still profit from decreases in crude oil or increases in product prices.
Those prices would be locked in otherwise, leaving the refiner at a disadvantage.

53
Describe crack spread options, and calculate their payoff
(formula).
______
Crack spread options are also designed to lock in the refining margin (just as crack spreads
with futures), but give the refiner the opportunity to still profit from favorable price changes. A
put option that protects against a decreasing crack spread.
A crack spread option is a combination of two options: one on crude oil and one on either
gasoline or heating oil. They trade with a one-to-one ratio of crude to product and differ from
conventional options in that a single option position results in two futures positions if the option
is exercised.
To protect a refiner against a narrowing of the crack spread (refining margin), a put on the crack
spread could be bought. The strike price of the put would be the margin the refiner wants to
lock in.
Put option payoff = [(strike - crack spread at expiration - put premium), 0].

54
Which are the main global energy exchanges (3 groups)?
______
International energy futures markets:
1. TOCOM Tokyo/SGX
2. Dubai Mercantile Exchange
3. ICE London
4. SGX Singapore
5. NYMEX New York
6. SFE Australia
The main natural gas futures markets are:
1. NYMEX
2. ICE
Power futures markets are:
1. Nord Pool Scandinavia

2. EEX European Energy Exchange Germany
3. ICE
4. NYMEX
5. APX

55
Describe the specific terms of energy futures contracts (6).
______
1. Underlying instrument.
2. Size. Amount of underlying item covered by each contract (contract size).
3. Delivery cycle. Specific months for which contracts can be traded.
4. Expiry date. Date on which contract ceases and all obligations end.
5. Grade or quality specification and delivery location.
6. Settlement mechanism. Terms of physical delivery or terminal cash payment.
Describe initial margin in futures trading (%). Who must

post initial margin?
______
Both buyers and sellers of a contract deposit the initial margin with a brokerage firm who
serves as a clearing member of the exchange. The initial margin is typically around 10% of the
total notional contract value.
If the account value drops below the maintenance margin because of adverse price movements, the
trader must deposit a variation margin which will restore the initial margin in his account.
Explain the futures trading process (diagram).

______

56
Describe examples of risk management with futures (7).
______
1. Stabilizing cash flows. Contracting in the forward market instead of spot market.
2. Setting purchase or sale price of commodities or securities by taking a futures position
opposite to cash-market position.
3. More closely matching balance sheet assets and liabilities.
4. Reducing transaction costs.
5. Decreasing storage costs.
6. Locking in "cost of carry" forward profits (profit from holding inventory).
7. Minimizing capital needed to carry inventory and the size of security-of-supply
inventories required by locking in guaranteed physical delivery in the future.

57
Explain the options available (3) for physical delivery in
futures contracts.
______
Only NYMEX and ICE offer exchange for physicals. No more than 2-5% of futures contracts
go to physical delivery at expiration via the exchange. Companies which do choose to deliver
have several options:
1. Standard delivery. As laid out in the futures contract specifications laid down in the rule
book of the exchange.
2. Alternative delivery procedure (ADP). If energy does not conform with contract
specifications (and thus cannot be settled via the clearing house) or if parties wish to
negotiate directly with each other about the settlement. Market participants release the
clearing house and their clearing broker from liabilities.
3. Exchange for physicals (EFP). Allow companies to choose their trading partner,
delivery site, grade of product, and delivery timing. Physical energy-market
transactions can be executed on the basis of a directly negotiated price between parties.
Contract size must be approximately equal to specified size.
Explain exchange for physicals (EFP).

______
Making or taking delivery at a location different from the one specified in the futures contract.
Companies who deliver also often do not want to be matched to a trading partner by the
exchange. EFPs allow companies to choose their trading partner, delivery site, grade of product,
and delivery timing. Physical energy-market transactions can be executed on the basis of a
directly negotiated price between parties. Contract size must be approximately equal to specified
size, so exchanges sometimes investigate EFPs to ensure the equivalent volume of physical to
futures.
After both parties of an EFP have agreed to a transaction, the price is submitted to the exchange
via the futures broker. The nominal price of an EFP can be outside of the daily trading range of
the futures market, and is usually chosen to be the exact basis of the futures contract. Once
registered by the exchange, the futures are offset and the hedge is taken off. The actual physical
exchange is transacted at the price negotiated by the parties (which may differ from the nominal
price).
An EFP market is quoted amongst futures brokers. The NYMEX requires written confirmation
of all EFPs, the ICE does not but reserves to make inquiries.
An EFP transaction has a number of uses and benefits for energy market participants:

58
1. Counterparty credit exposure may be reduced when an existing swap position is
reversed and replaced with a futures position. Thus, an EFP allows counterparties to
release “credit”, clearing the way for further OTC trading.
2. Reduced balance sheet and margin requirements .By netting OTC positions against
opposing futures positions margin and credit support requirements can be reduced.
Note that an EFP transaction will result in a large negative variation margin call for an
entity which transacts the futures leg of an EFP if the EFP futures price struck at an
"out of market" price.
3. Off-Market, EFPs are negotiated and traded between counterparties off-market with
price and volume details not being reflected “on the screen”. EFP futures prices are
not considered in the end of day price settlement process.
4. 24-Hour Trading, EFP transactions can be negotiated around the clock but must be
registered between 7 am to 5 pm during Exchange business days.
Describe the specifications of the ICE Brent crude futures

contract.
______
Date of launch: 23 June 1988. Money settlement, no delivery.

Trading hours: 08:00 - 09:45 (electronic), 10:02 - 19:30 (open outcry)
Unit of trading: 1,000 net bbl (42,000 gallons) of Brent crude oil
Specification: Current pICEline export quality Brent blend as supplied at Sullom Voe.
Quotation: $/bbl.
Minimum price fluctuation: 1 cent per bbl, tick value is $10.
Maximum daily price fluctuation: unlimited.
Daily margin: All open contracts marked-to-marked daily.

59
Trading period: 12 consecutive months, then quarterly out to max. 24 months, then half-
yearly out to max. 36 months.
Position limits: Unlimited.
Describe the specifications of the NYMEX light sweet crude

oil futures contract.
______
Unit of trading: 1,000 U.S. bbl (42,000 gallons) of WTI light sweet crude oil.
Trading hours: 10:00 - 14:30 the following day (open outcry). After-hours trading via
NYMEX ACCESS on Monday through Thursday from 15:15 - 09:00 the following
day. On Sunday, the session begins at 19:00.
Trading months: 30 consecutive months plus long-dated futures 36, 48, 60, 72, 84
months prior to delivery.
Price quotation: $/bbl.
Minimum daily price fluctuation: One cent / bbl ($10 per contract).
Maximum daily price fluctuation: $3/bbl in all but the first two months. $6/bbl if the
previous day's settlement price is at the $3 limit. In the event of a move of $7.50/bbl in
either of the first two contract months, limits on all months become $7.50/bbl following a
one-hour trading halt.
Last trading day: 3rd business day prior to the 25th calendar day of the month
preceding the delivery month, or the last business day of that month.
Delivery: FOB seller's facility, Cushing, Oklahoma, at any pICEline or storage facility
with pICEline access to TEPPCO, Cushing storage, or Equilon pICEline co., by in-
tank transfer, in-line transfer, book-out, or inter-facility transfer (pumpover).
Delivery period: Initiated on or after the first calendar day of the deliver month
and completed by the last calendar day of the delivery month.
Alternative Delivery Procedure (ADP): Matching of buyers and sellers by the exchange.
Exchange of futures for physicals (EFP): May be used to initiate or liquidate a futures
position.
Deliverable grades:
o For domestic crude streams: 37-42 °API, < 0.42% S. Different domestic
crude streams are available.
o For foreign crude streams: 34-42 °API. Different international crude streams
are available.
Inspection: In accordance with pICEline practices. The buyer or seller may request
an inspection, but must pay for it.
Position limits: 20,000 contracts for all delivery months combined, but not to
exceed 1,000 in the last three days of trading or 10,000 in any month.
Margin requirements: Required for long and short positions.

60
Describe the specifications of the NYMEX heating oil futures
contract.
______
Trading unit: 42,000 gallons (1,000 bbl).

Trading hours: 10:05 - 14:30 (open outcry). After-hours trading via NYMEX
ACCESS on Monday through Thursday from 15:15 - 09:00 the following day. On
Sunday, the session begins at 19:00.
Trading months: 18 consecutive months commencing with next calendar month.
Price quotation: $/gallon.
Minimum daily price fluctuation: One cent / gallon ($4.20 per contract).
Maximum daily price fluctuation: 6 cents / gallon in all but the first two months. 9 cents
/ gallon if the previous day's settlement price is at the 6 cent limit. In the event of a move
of 20 cent / gallon in either of the first two contract months, limits on all months become
20 cents / gallon following a one-hour trading halt.
Last trading day: At the close of business on the last business day of the month
preceding the delivery month.
Delivery: FOB seller's facility in New York Harbor, ex-shore. All duties, entitlement,
fees, charges, taxes paid. Requirements for seller's shore facility: capability to deliver into
barges. Buyer may request delivery by truck but must pay a surcharge. Delivery may also
be completed by pICEline, tanker, book transfer, or inter- or intra-facility transfer.
Delivery must be made in accordance with applicable federal, state, and local licensing
and tax laws.
Delivery period: Initiated after the fifth business day of the deliver month and
completed by the last calendar day of the delivery month.
position.
Grade and quality: Fungible No. 2 heating oil
Inspection: The buyer will require a quantity inspection for barge, tanker, or inter-
facility transfer. Cost shared equally between buyer and seller. If the product fails
quality inspection, seller bear cost, otherwise it is split equally.
Position limits: 7,000 contracts for all delivery months combined, but not to exceed 1,000
in the last three days of trading or 5,000 in any month.
Margin requirements: Required for long and short positions.

61
Describe the specifications of the NYMEX Henry Hub
natural gas futures and options contract.
______
Trading unit:
o Futures: 10,000 MMBtu.
o Options: One NYMEX Division natural gas futures contract.
Trading hours: 10:00 - 14:30 (open outcry). Close is 14:45 on any futures termination day
that falls on a Wednesday. After-hours futures trading via NYMEX ACCESS on Monday
through Thursday from 15:15 - 09:00 the following day. On Sunday, the session begins at
19:00.
Trading months:
o Futures: 72 consecutive months commencing with the next calendar month
o Options: 12 consecutive months plus contracts initially listed 15, 18 , 21 ..., 72
months out on a March/June/Sept/Dec cycle.
Price quotation: $/MMBtu for futures and options.
Minimum daily price fluctuation: $0.001/MMBtu ($10 per contract) for futures and
options.
Maximum daily price fluctuation:
o Futures: $1/MMBtu ($10,000 per contract) for all months. If any contract is traded,
bid or offered at the limit for 5 minutes, trading halts 15 minutes. When trading
resumes, the limit is $2 in either direction of the previous day's settlement
price. No limits on the last three days of the spot month.
o Options: Unlimited.
Last trading day:
o Futures: 3 business days prior to the first calendar day of the delivery month.
o Options: At the close of business on the last business day immediately preceding
the expiry of the underlying futures contract.
Exercise of options: No later than 17:30 or 45 minutes after the underlying futures
settlement price is posted, which ever is later, on any day up to and including the option's
expiry.
Option strike prices: 20 strike prices in increments of 5c per MMBtu above and below the
ATM strike price in all months. Additional strikes for nearby months and lowest and
highest prices. At least 81 strikes in first three nearby months and 61 strike prices for four
months and beyond.
Delivery location: Sabine pICEline Co.'s Henry Hub in Louisiana. Seller is responsible
for moving the gas through the hub, the buyer from the hub. Seller pays hub fee.
Delivery period: Initiated no later than the first calendar day of the deliver month and
completed by the last calendar day of the delivery month.
position.
Inspection: pICEline specifications in effect at time of delivery.

62
Position limits: 7,000 contracts for all delivery months combined, but not to exceed 1,000
in the last three days of trading or 5,000 in any month.
Margin requirements: Required for long and short positions in futures and short
positions in options.
Describe the specifications of the NYMEX Central

Appalachian coal futures contract.
______
Trading symbol: QL
Trading unit: 1,550 t of coal.
Trading hours: 10:30 - 14:30 (open outcry).
Trading months: 24-26 consecutive months based on a quarterly schedule. As
contracts expire, the 26th month will roll forward until it becomes the 23rd month. At
that point, new 24th,25th and 26th month contracts are added.
Price quotation: $/ton.
Minimum daily price fluctuation: One cent / t ($15.50 per contract).
Maximum daily price fluctuation: $12 per ton ($18,600 per contract) for all month. If any
contract is traded, bid, or offered at the limit for 5 minutes, trading will halt for 10
minutes. When trading resumes, limit is $24 / ton in either direction of the previous day's
settlement price. No limits during the last 3 days of spot month.
Last trading day: At the fourth-to-last business day of the month preceding the delivery
month.
Contract delivery unit: 1,550 t per contract, with a loading tolerance of 2% or 60t,
whichever is greater, over the total number of contracts.
Delivery location: FOB buyer's barge at seller's delivery facility on the Ohio River
between mileposts 306 and 317, or on the Big Sandy River, with all duties, taxes, fees,
entitlements and charges paid by the seller. Discount of 10 cents per ton for delivery to
a terminal on the Big Sandy River.
Heat content: Min. 12,000 Btu per pound gross calorific value with analysis tolerance
of 250 Btu per pound below.
Ash content: Max. 13.5% by weight with no analysis tolerance.
Sulfur content: Max. 1% with analysis tolerance of 0.05% above.
Moisture content: Max. 10% with no analysis tolerance.
Volatile matter: Max. 30% with no analysis tolerance.
Hardness/grindability: Min. 41 Hardgrove index with 3 points analysis tolerance.
Size: 3 inches topsize, nominal, with a max. of 55% passing one-quarter-inch square wire
cloth sieve or smaller.
position. Deadline is 10:00 NY time on the first business day following termination of
trading.

63
Position limits: 5,000 contracts for all delivery months combined, but not to exceed
200 in the last three days of trading.
State the key steps (10) in developing a risk-policy statement

for energy markets.
______
1. Is the organization a consumer, a producer, or a trader?

o What products does it have in price risk?
o Are risks fixed or floating?
o Is the company trading (needs liquidity) or simply hedging (let contracts run full
term)?
2. Assess credit risk from derivatives. Too much exposure to single counterparty?
3. Does the company allow its traders to speculate along with hedging? What is the total
volume available to hedge and/or trade?
4. Are traders allowed to buy and sell options? The sale of options can be used to cover
up losses with cash flow from the options premium.
5. Which derivatives should be used? Futures, options, OTC, on-exchange, swaps.
6. Does the required skill set, IT infrastructure exist within the company?
7. How and how often are derivatives positions valued?
8. What are position limits (by tenure or by products)?
9. Who is responsible for monitoring, and ensuring compliance with statement?
10. Line managers should fine tune the statement for traders, operations staff and managers.
A hedger will create a risk-management policy, a speculator a derivatives-usage document.
Understand general levels of hedging positions.

______
When hedging as a producer or consumer, it is wise to look at up to 50% exposure for general day-
to-day hedging requirements. Amounts larger than 50% of consumption or production volume are
speculative and should only be considered as a rare pre-emptive measure ahead of a disaster scenario
(war, weather) or to protect against near term volatility in energy-market prices.
End-users will hedge about 10-30% of volumes around budget levels up to four years forward,
leaving an additional 50-70% for opportunistic hedging if levels come below budget levels
closer to prompt. End-users should have another policy about hedging that allows for hedging in
times of extreme price movements, such as in the California power markets.

64
Discuss a sound framework for energy-market risks (4
points), key points in recording risk (6) including trading
controls and position limits (5).
______
A sound risk management process includes:
1. Comprehensive approach to risk management.

2. Detailed structure of derivatives position limits.
3. Clear guidelines for governing risk taken by the company's officers.
4. Strong information system for controlling, monitoring, reporting and aggregating risk,
including that from energy and emissions-trading positions.
Key points in recording risk:
1. Risk should be measured and aggregated across all derivatives activities.

2. Management should be able to assess exposures on a consolidated basis.
3. Risk-measurement system used should accurately reflect the multiple types of risks faced
by the organization.
4. Relevant personnel should understand risk management procedures at all levels of
the organization, and there should be a common framework.
5. Marking derivatives to market is fundamental to measuring and reporting
exposures accurately.
6. An organization speculating in OTC energy and other derivatives should have the ability
to monitor credit exposures, physical and derivatives trading positions and market-price
movements on a daily basis.
Simulation and stress testing are paramount. Worst-case analysis should be conducted across the
entire enterprise (plus subsidiaries). Quantitative and qualitative disaster scenarios should be
tested.
Trading controls, in form of a trading/position-limit system, should set boundaries

for organizational risk-taking and risk-reduction through hedging. Market-risk limits must be
set. Position-control breaches should trigger an "exception report", which should be reviewed on
a daily basis.
Position limits imposed on specific positions may be by:
1. Energy product/type.
2. Derivative type.
3. Tenure (size and forward reach of a position).
4. Division or office or group of such.

65
5. Individual trader.
Risk management systems should be the more real-time the better. Best is a computerized system
that allows board members to see derivatives and underlying energy positions along with risk
assessment instantly. The board should also be updated about derivatives risk at least quarterly.
Formulate a risk-management policy for emissions markets (7

factors).
______
1. Financial objectives of the company (may differ in emissions and energy markets).
2. Predictability of (GHG) production.
3. Extent to which movements in allowance prices can be absorbed within profit
margins and how much risk the board of directors is willing to accept.
4. Extent to which the company can pass on effects of movements in allowance prices to
customers.
5. Extent to which competitors are able to absorb price fluctuations or pass them on
to customers.
6. Volatility of allowance prices and subsequent price changes that may have an
effect on future cash flows.
7. Ability of the company to hedge relevant exposures. Availability of hedging
instruments (derivatives).
What are the advantages of energy OTC options (1+2)?

______
Energy OTC options and swaps are generally cash-settled. Their value at settlement is normally
based on the average price over a period (calendar month). Advantages of cash settlement are:
1. Physical asset need not be resold to capture increase in value.

2. Settlement against average prices is generally preferred, as it provides a better hedge for
cash flow type exposures. Averages are difficult to replicate in the futures market, so
OTC cash-settlement is more cost effective.

66
Explain how options and energy swaps are related.
______
A forward (swap) can be decomposed into two options: The swap buyer is long the
upside, a strip of calls, and short the downside, a strip of puts.
A swap can be seen as an exchange of upside for downside, as payments are netted.
Upside: Amount by which floating exceeds fixed payment.

Downside: Amount by which floating is less than fixed payment.
If floating > fixed, then the upside is paid by the swap seller to the buyer, if floating < fixed,
then the downside is paid by the swap buyer to the seller.

67
Explain the two main components of an option's value.
______
1. Intrinsic value. Amount by which the option is ITM.

2. Time value. Value related to the possibility of an increase in intrinsic value. Highest
when underlying trades ATM.

68
Most exchange-traded options on NYMEX are
[American/Asian/European] options, most OTC options are
[American/Asian/European] options.
______
Most exchange-traded options on NYMEX are American options. Can be exercised any time.
Most OTC options are Asian options. Exercised automatically if ITM. A few OTC options are
European or American style.
Discuss the impact of price specificity on options prices.

______
Short-term relationships are heavily exposed to supply-demand imbalances, while longer-

term relationships are less volatile. Price volatility of further-out options decreases
(downward sloping volatility curve), which decreases their price.

69
Describe the four main option greeks.
______
Delta: Change in option value with respect to underlying price. Measure of probability
that the option finishes ITM at maturity.
Gamma: Second-order derivative, describes change of delta with respect to underlying
price. Highest when option is ATM and when expiry is approaching.
Theta: Change in option value with respect to time. Option value changes with square
root of time (√t). Starts out small and becomes large for OTM options, very small for
deep ITM options.
Vega: Change in option value with respect to volatility. Volatility only affects time value,
not intrinsic value.

70
Explain delta hedging (3 steps).
______
1. Find the delta equivalent value of the option position (e.g., delta 0.5).
Hedge ratio:
2. Enter opposite position of the delta equivalent value (e.g., sell short half [n] the
futures controlled by the option). Price moves offset each other.
3. Re-hedge as price moves.

71
72
73
Describe the inputs to an energy options pricing model and
their effects on ATM options.
______
1. Price of the underlying commodity. Nearby option prices on futures become "whippy"
(more volatile) than longer-dated options. OTC options are less volatile because they
settle against the average price. Swaptions are less volatile.
2. Strike price. Historical spot prices have a fat tail (non-normal distribution). The higher
probability of a move upwards creates a volatility smile.
3. Time to maturity. An options value varies with the square root of time (√t) as a rule of
thumb. A 6-month option can be expected to cost 41.4% more than a 3-month option, as
√2 = 1.414. A one-year option costs 100% more than a 3-month option, as √4 = 2.
Interest rates slightly reduces option values though.
4. Volatility estimate. Direct correlation to price. An increase in volatility from 20% to
30% (increase in half), will increase the option price by 50%.
5. Interest rates. Small impact. Use of rho is more common in financial markets.
Through the interaction of several effects we have an unusually slow time decay in energy
options than for financial options.
A 6-month option can be expected to cost ... % more than a 3-

month option.
A 6-month option can be expected to cost ... % more than a 3-month option.
______
A 6-month option can be expected to cost 41.4% more than a 3-month option, as √2 = 1.414.
Option costs are related by √t.

74
An increase in volatility from 20% to 30% will increase the
option price by ... %.
An increase in volatility from 20% to 30% will increase the option price by ... %.
______
An increase in volatility from 20% to 30% (increase in half), will increase the option price by
50%.
Direct correlation between volatility estimate and price. %Δσ = %ΔP
Describe a volatility smile.

______
A volatility smile is a long-observed pattern in which ATM options tend to have lower implied
volatilities than ITM or OTM options. The pattern displays different characteristics for different
markets and results from the probability of extreme moves.
Describe the impact of the lognormal distribution of prices on

options.
______
Commodity prices are bound by zero (cannot be negative). Therefore, a put and a call are not
symmetrically distributed around the strike price, because the put has limited upside (bound by
zero price of the underlying), while the call can be worth a potentially unlimited amount.

75
The probability of prices are distributed in an upwardly skewed distribution around the
forward price.
Describe a participating collar and a participating swap.

______
The most swap-like options-related instruments.
In a participating collar, a company sells a quantity of ATM options and buys a higher
quantity of OTM options, so that their values offset each other (no upfront premium).
In a participating swap, the difference between strike prices is eliminated. A company also sells
a quantity of ATM options and buys a higher quantity of OTM options, but the strike price of
the ATM options is moved more ITM (upward) in order to give a larger quantity to the
purchased options. Zero participation means the quantities on both sides are the same. The
instrument is equivalent to a swap then. Participating swaps are used if there is a high
probability that prices will move so that the embedded swap will be ITM.

76
Example participating collar:
Example participating swap:

77
Describe a swaption. Swaptions may be written on ... (4).
How are they valued?
______
A swaption is an option to buy or sell a swap. It is considered a compound option. Cheaper than
a cap, since it still has downside risk. Buyers need assurance of a maximum fixed price but feel
the price may fall before the expiry of the swaption. If prices fall, they let the swaption expire
and buy a swap at lower cost in the market. Typically held until maturity.
Swaptions may be written on:
1. Plain-vanilla swaps.
2. Differential swaps.

78
3. Participation swaps.
4. Extendible swaps (callable and putable). Swap price can be lowered by selling a
swaption.
Swaptions are complex compound options whose valuation requires numerical techniques,
most often Monte Carlo simulation or tree-based techniques as an alternative.
Compare hedging with speculation (6 conditions of hedging).

______
A transaction can be classified as hedging if:
1. The direction of the hedging position is the opposite of the underlying position and the
overall exposure of the company.
2. The size of a hedging transaction relates to the underlying position size.
3. Risk/reward should make sense.
4. The hedging transaction should be understandable and measureable to keep exposure in
check.
5. There should be a clear hedging policy.
6. There should be close supervision from the senior levels of the corporation.
If these conditions are violated, speculation is likely to be mistaken for hedging.
What are exotic options? List the major differences (4) to

plain vanilla options.
______
Exotic options diverge from early options models (BSM, etc.) in their definition of payoff
and the stochastic process used to describe the dynamics of the underlying asset's price.
Major differences between exotic and plain vanilla options:
1. Path-dependency. Option payoff depends not only on price of the underlying at

expiration but also on price trajectory during the life of the option. Examples are:
o Asian options.
 Call payoff = max [avg[F(T) - K, 0]
 Put payoff = max [K - avg(F(T), 0]
o Average strike options.
 Call payoff = max [F(T) - avg(F), 0]
 Put payoff = max [avg(F) - F(T), 0]

79
o Lookback options.
 Call payoff = F(T) - min(F)
 Put payoff = max(F) - F(T)
o Barrier options. Extinguished or activated based on a certain event defined in
terms of the underlying (e.g., residual fuel oil price).
2. Multiple-commodity options with payoffs that depend on the price of two or more
commodities. Examples are:
o Spread options.
 Call payoff = max [F1 - F2 - K, 0]
 Put payoff = max [ K - (F1 - F2), 0]
o Basket options. One or more of the underlying assets that determine payoff are
comprised of a basket of commodities.
o Options to exchange one asset for another. Get better of two assets.
3. Compound options. The underlying is itself an option.
4. Digital or binary options. Predetermined payoffs that depend on the occurrence of

events usually defined as one or more prices.
Calculate the payoff of Asian options (formula).

______
Asian options.
Call payoff = max [avg(F(T)) - K, 0]

Put payoff = max [K - avg(F(T)), 0]
Average strike options.
Call payoff = max [F(T) - avg(F), 0]

Put payoff = max [avg(F) - F(T), 0]
Lookback options.
Call payoff = F(T) - min(F)

Put payoff = max(F) - F(T)
Barrier options. Extinguished or activated based on a certain event defined in terms of the
underlying (e.g., residual fuel oil price).

80
Calculate the payoff of average strike options.
______

Asian options.
Call payoff = max [avg[F(T) - K, 0]

Put payoff = max [K - avg(F(T), 0]
Lookback options.

Calculate the payoff of lookback options.

______
Lookback options.

Asian options.



81
Calculate the payoff of barrier options.
______
Barrier options are activated (knocked in) or extinguished (knocked out) based on a certain
event defined in terms of the underlying (e.g., residual fuel oil price).
Asian options.


Lookback options.

Calculate the payoff of spread options.

______
Spread options.
Call payoff = max [F1 - F2 - K, 0]

Put payoff = max [K - (F1 - F2), 0]
Asian options.



82
Lookback options.

Contrast the Black-Scholes model for pricing stock options

and forward contracts.
______
For stock options, the BSM used for a call is:
with
For forward contracts, the drift coefficient is assumed to be zero since the zero cost associated
with entering into a forward is incompatible with an assumption of positive expected return
(Black). The Black model for a call on a forward or future is:
with
Describe how to price exotic options (4 approaches).

______
1. Replication approach. To price an exotic option it should be replicated with a package

of standard options or simpler exotics.
2. Closed-form expression of option price comparable to the BSM.
3. Approximation of a closed-form expression.
4. Numerical methods. Monte Carlo simulation, tree methods, finite-difference or

numerical integration models.

83
Explain how stock prices differ from energy commodity
prices (5) and implications.
______
Stock prices are modeled with general Brownian motion (GBM) equations. However, the GBM
assumptions used do not apply to commodity prices. Differences are:
1. Investment assets vs. consumption goods. Energy commodities cannot be treated as

purely financial assets. They are inputs into production processes and/or consumption
goods. GBM does not allow negative prices, which is often violated in the case of
electricity (when output cannot be stored).
2. Seasonality. Prices of energy commodities display seasonality. Intra-year to intra-day
resulting from demand fluctuations depending on weather. Requires forward-looking
modeling.
3. Gapping or jump behavior. Driven by demand fluctuations and low supply elasticity.
Jump-diffusion process can be used.
4. Prices gravitate to the cost of production. GBM may produce unrealistic spreads
between related commodities. Mean reverting behavior can be (somewhat
inaccurately) modeled with the Ornstein-Uhlenbeck process (OU).
5. Price behavior changes closer to maturity. Prices of energy commodities behave
differently during different periods of their life. According to the Samuelson-
hypothesis, volatility of forward prices increases as they get closer to maturity. More
information becomes available, which results in more trading.

84
Explain how Monte Carlo methods are used to price energy
options. What is the main drawback?
______
The Monte Carlo approach models pricing of the underlying, from which we can derive options
prices for each scenario of price movements. Option payoffs are discounted to present time and
then averaged, which represents the estimate of the option value.
Monte Carlo simulation can use GBM or any other type of price evolution. Because of its
simplicity, Monte Carlo is often used to validate prices found with other approaches. A variant
is American Monte Carlo (AMC), used to model American options using path-dependent lattice
methods. Computationally intensive, but promising.
The main drawback of Monte Carlo is relatively slow speed, which is why variance-reduction
methods are often used to make it compatible for trading.
Explain how binomial trees are used to price energy options. Do

they address seasonality?
______
Binomial, trinomial, n-trees are (lattice methods) are one of the most widely used approaches to
pricing options. They occupy a middle ground between Monte Carlo and finite-difference
methods. The life of the option is subdivided into different time intervals, where the price of the
underlying can move into a smaller number of states. Via backward recursion (induction), the
option value is calculated. The more time steps are used, the more accurate the model becomes.
Binomial trees do not address seasonality, but trinomial trees can.

85
at time 3.
Explain finite-difference and numerical methods to price

exotic options.
______
Finite differencing reduces option pricing to an integral equation:
The life of the option is broken down into small intervals.

The range of possible prices broken down into a number of intermediate discrete prices,
actually similar to the binomial tree method.
Describe how to find out volatility of options (3).

______
1. Inferring from market prices via an options pricing model (e.g., BSM). Volatility is
varied until the calculated price matches the market price.

86
2. Historical volatility information. Standard deviation of natural logarithms of ratios of
historical prices. . Correlation (for dual commodity options) is also
calculated this way (NOT standard deviation of historical prices).
3. GARCH (p,q). Generalized autoregressive conditional heteroskedasticity.
: Error term, conditional on information is normally distributed.
α and δ:vectors of unknown parameters
How are Asian options priced (2)? Which model cannot be

used to price them?
______
1. Numerical methods and approximations.

2. Monte Carlo simulation. Often with variance-reducing techniques.
Asian options cannot be priced with the BSM because the average price is not lognormally
distributed (a condition for using the BSM). IT can be postulated though that the distribution is
near lognormal for Asian options with short tenors (<1 yr) and low volatility (<40%). Tree
models are getting too complicated for Asian options.
If the average period shrinks, the price of an Asian option converges with that of a
European option. Deep ITM Asian options also converge with European options, because
they are dominated by intrinsic value.

87
Describe lookback options.
______
Lookback options are exotic options. A standard lookback call (put) grants the holder the
right to purchase an underlying at the lowest (highest) price reached during the period. The
option always expires ITM or ATM.

Lookback options are much more expensive than corresponding European or American options.
Energy markets have not embraced them.

88
Describe barrier options.
______
Barrier options are exotic options, invented to reduce the initial cost of hedging and allowing the
buyer to readjust the hedge if conditions change. Barrier options either come to live (knocked
in) or are extinguished (knocked out) when underlying price passes a certain level. The price
level (barrier, knock-in, or knock-out price) may be reached at any time during the option's life.
The following combinations are possible:
Barrier options may be combined with a rebate, paid when the option is extinguished as
compensation to the holder. An example is the "up-and-out put" purchased by an energy
producer to hedge its natural long position. It is cheaper than a vanilla option but affords same
level of downward price protection. If prices move up significantly, the option is
extinguished, and the holder may enter a new option with a higher strike.

89
Barrier options are more popular in FOREX markets than in energy markets, where prices
could be easier manipulated to push through the barrier and extinguish the option because of
lower liquidity.
Barrier options impose hedging difficulties on the option writer, as gamma can be infinite when
the price is knocked out.

90
Barrier options either ... or ... when underlying price passes a
certain level.
______
Barrier options either come to live (knocked in) or are extinguished (knocked out) when
underlying price passes a certain level.
Barrier options are exotic options, invented to reduce the initial cost of hedging and allowing the
buyer to readjust the hedge if conditions change. The price level (barrier, knock-in, or knock-out
price) may be reached at any time during the option's life.
Describe multi-commodity options (3 main groups and

subgroups).
______
Multiple-commodity options have payoffs that depend on the price of two or more commodities.
Examples are:
1. Spread options.
Call payoff = max [F1 - F2 - K, 0], put payoff = max [ K - (F1 - F2), 0].
o Location spread. Based on price differences of the same commodity in different
locations.
o Calendar spread. Price on same commodity in different points in time. o
Processing spread. Inputs and outputs of a process.
o Quality spread. Differences in quality, e.g. sulfur content of crude or heating oil
(grades).
2. Basket options. One or more of the underlying assets that determine payoff are
comprised of a basket of commodities. The average price determines the payoff. E.g.,
hedges for gas fractionation plants that produce a basket of NGL. The basket is treated as

91
the underlying commodity.
3. Options to exchange one asset for another. Get better of two assets. The short will
often choose cheapest to deliver. In energy market, an application would be a contract
which allows the customer to buy natural gas at prices related to heating oil, which is
essentially an option to exchange the two.

92
Although there is no numerical relationship for spread options, options to exchange one
commodity for another can be valued like this:
with

93
Describe the four types of energy spread options.
______
1. Location spread. Based on price differences of the same commodity in different

locations.
2. Calendar spread. Price on same commodity in different points in time.
3. Processing spread. Inputs and outputs of a process.
4. Quality spread. Differences in quality, e.g. sulfur content of crude or heating oil
(grades).
Payoffs of spread options are:
Call payoff = max [F1 - F2 - K, 0]

Put payoff = max [ K - (F1 - F2), 0]
Calculate a spark spread.

______
Spread between electrical power prices and natural gas prices. A problem is the
unstable correlation between the two.
Spark spread = output price ($/MWh) - input price ($/MMBtu).
Suppose the price of natural gas is $2/MMBtu and the price of power is $25/MWh.

94
In more detail:

95
96
How are spread options priced (2)?
______
1. The spread between two products can be treated as a good. Pricing of such a "good" is
very complicated.
2. Assume the prices defining the spread follow a joint lognormal distribution. The option
is priced as the discounted expected value, which boils down to a double integral.
Describe compound options.

______
A compound option is an option on a derivative, e.g., an option that allows the holder to buy or
sell another option for a fixed price. The compound option is referred to as the "overlaying
option", the "underlying option" is the option that can be called or put by the holder of the
compound option. Typical cases are European/European options, classified as call on call, call on
put, put on call, and put on put.
Compound options allow the holder to lock in commodity price protection at an

initial cost that is lower than a cap or floor.
The also allow price protection contingent on a future event, often used in financing
and satisfying loan covenants.
An option on power capacity is also a compound option.
Another example is an oil warrant, an American option on an Asian option, which are
often priced by binomial procedures.
For valuation of a compound option, the distribution of the underlying of the underlying is
used, as it is also used in hedging.

97
Describe digital options. How are they valued?
______
Digital (binary) options have discontinuous payoffs and are not widely used in energy markets
but are sometimes embedded in swap and option structures. Used when the customer has a strong
view about the relation of future energy prices to certain price levels. Digital options pay either a
constant value X or nothing, depending on whether they finish ITM, i.e. F(T) - K > 0. Binary
options are difficult to hedge.
The valuation is straightforward: The expected payoff multiplied by the risk-neutral probability
that the option will finish ITM. Equal to XN(d2), discounted to the valuation date. COD (cash
on delivery) options require no payment at inception and cash-settle only if the option is ITM
(payoff F(T) - K with a premium deducted). If OTM, there is no payment. COD options can
have negative payoffs, painful to the option buyer.

98
Describe the characteristics of natural gas options.
______
1. Take-or-pay call options. The buyer agrees to purchase between a minimum and a
maximum of natural gas every day for the duration of the contract. If the total volume
falls below a certain amount, a penalty is due.
2. A swing ratchet option is a monthly, seasonal, or annual option. The daily volume can
be increased (swing up) or decreased (swing down) by a fixed amount only. The
number of swings is limited to a maximum.
These options are swing options, i.e., options that require delivery of physical gas or electricity.
The quantity purchased can be modified on a daily basis. Characteristics of natural gas options
are similar to American options with path-dependence.
Valuation of swing options is done with binomial trees extended to a connected set of binomial
trees (a binomial forest). A node in the tree is represented by a triplet (i, j, k), with i representing
the time step, j the price level, and k the number of days the maximum daily volume has been
taken up to the current time. k identifies the appropriate tree in the forest.

99
100
Contrast time series analysis and distribution analysis.
______
Time series analysis analyzes changes in the price from day to day. The process of analyzing
daily price returns. Used for parameter calibration and event and seasonal calibration. Actual
data = Model predictions + residuals. Regression models. Used commonly in business.
Distribution analysis explores price behavior over a period of time. Critical in comparing
suitability of models and seeing effect of mean reversion and changes in model factors.
Testing, benchmarking, selecting and comparing models. Getting insights in options pricing.
Should be used more.

101
What is time series analysis used for (2)? What is
distribution analysis used for (2)?
______
Time series analysis:
1. Parameter calibration
2. Event and seasonal calibration
Distribution analysis:
1. Testing, benchmarking, selecting models

2. Getting insights about option pricing
Explain why moments are important in energy risk

management (3).
______
1. Relatively easy to calculate.

2. Can be used for correcting modeling errors.
3. Can be very important during option valuation.
The distribution moments are used to relate the actual to the model-generated distributions.

102
Explain and calculate the moments of a distribution.
______
The n-th moment for a distribution of variable x is the expected value of the variable raised to the
n-th power. It is calculated as:
The most important moments are the first four moments:
1. Mean.
2. Standard deviation. . Measurement of the distribution's width.
and
3. Skewness. . The third moment adjusted for the distribution center.
4. Kurtosis. ,
In the special case where the distribution is centered around zero, the first moment will be zero.
The second moment will equal the variance, the third moment the skewness and the fourth
moment the kurtosis.
What is meant by distribution width?

______
The second moment of a distribution, related to the standard deviation.

103
Describe the two distributions generally used in valuation
and risk management of derivatives.
______
2
1. Normal distributions. Perfectly symmetric, zero skew. Kurtosis = 3 x var .
2. Lognormal distributions. Generally used in finance. Skewed to the right, values are
always positive.
Moments must be related to the historical distribution by the following formula:
Describe the Q-Q plot.

______
The quantile-to-quantile (Q-Q) plot (or test) is used to test for normal distribution. It compares
the actual probabilities of the random variable to the expected probabilities if this variable were
normally distributed. If the variable is normally distributed, the Q-Q plot looks like

104
a diagonal line, as the expected = actual distribution. If the distribution is not normal, we get an
S-shaped Q-Q plot.
Normal distribution (diagonal line): Non-normal distribution (S-shaped line):
Explain the autocorrelation test.

______
The autocorrelation test is used for normality. If a random variable follows a normal
distribution, variables will take values that are uncorrelated, because they are truly
independent. If the distribution is normal, all the correlations (steps) = 0.

105
Normally distributed price returns: Price returns with mean reversion:
To determine whether a time series is mean-reverting, autocorrelation should be < 0.
Contrast mean squared error and R-squared.

______
Mean squared error (MSE) is the standard deviation of model residuals. The MSE should be as
small as possible for an accurate model.
2
R-squared tells us how much of the uncertainty in the actual data is captured by the model. If R
= 1, then the model has 100% predictive power.
and if residual mean = 0:

106
Describe the six-step model selection process.
______
1. Informally look at actual market data. Descriptive, nonquantitative analysis of the

price time series. Explain spikes and market turbulence and relate them to fundamental
drivers. Get a feeling for price behavior.
2. Create a shortlist of possible models. E.g., lognormal, and mean-reversion.
3. Calibrate parameters through time series analysis. A close look at historical time series of
price returns. Generates estimates for model parameters and seasonality. The
autocorrelation of price returns may point out the appropriateness of a mean-reverting
model: negative autocorrelation is a strong indicator of mean-reversion. Information
about market volatility. Run normality tests (Q-Q, autocorrelation of residuals).
4. Generate distributions from models. Calculate price distribution moments from the
model(s).
5. Perform distribution analysis. Deciding how well the pricing model fits market reality.
The ultimate benchmarking tool between models.
6. Compare results and select the most appropriate model. Based on evidence.
A good model should capture most of the market characteristics defined by different types of
analysis (time series analysis of price returns, distribution analysis of price levels, other
statistical tests).
Which models are the most appropriate for energy

commodity spot prices?
______
Mean-reverting models.

107
In time-series analysis of spot prices, what can be done when
certain historical spot prices are unavailable?
______
The most nearby forwards can be used as proxies for spot prices.
What are the main advantages (2) of the lognormal model?

______
1. The lognormal model guarantees that prices will never be negative.
2. Black-equivalent volatility of the price-process = spot price volatility, and not zero
like in the mean-reverting model with log of prices.
Describe mean-reverting models (2 kinds).

______
Energy market require mean-reverting models, as they capture the distribution of energy
prices best. Changes in spot prices have a negative autocorrelation. Variants:

108
1. Mean reversion in the log of the price. Captures distribution width but not the tails,
the higher moments (skewness, kurtosis). Often used in electricity markets.
2. Mean reversion in the price. Black-equivalent volatility = spot price volatility.
Mean reversion in the log of the price. Uses the following formulae:
Spot price
Expected value of spot price
The stronger the mean reversion, the quicker the drop-off in volatility. A drawback of single-
factor mean-reversion models is that volatility will approach zero with longer terms. When
jumps are incorporated, the model becomes more accurate.
Mean reversion in price. Uses the following formulae:

109
Spot price
Expected value of spot price
The width of the distribution (standard deviation) continues to grow at the same rate as the
equilibrium price volatility. The Black-equivalent volatility in this case becomes the
equilibrium price volatility in the long-run. The effective volatility of the spot price
process never goes to zero.
Calculate the Black-equivalent volatility over time (3 models,

formula).
______
The Black-equivalent volatility is the implied volatility from market prices.
1. Lognormal model: Black-equivalent volatility = spot price volatility.
2. Mean-reverting model for price: Black-equivalent volatility = spot price volatility.

3. Mean-reverting model for log of price Black-equivalent volatility will be zero over the
long term.
Black-equivalent (implied) volatility is calculated as:
α: rate of mean reversion, σ: volatility, t: time of observation

110
Explain the drawback of the single-factor mean-reverting
model.
______
The drawback of the single-factor mean-reverting model is that it forces the implied Black-
equivalent average volatility of the price distribution to go to zero over a longer period of time
(as the spot approaches the immobile long-term mean level). Caution must be used whenever
using a single factor mean-reverting model in valuing longer-term options. In reality, volatility
does not approach zero in the long term.
In the case of Black-Scholes, we can correct for the constant volatility assumption by allowing
each option price of different maturity to have a different volatility value. Thus we somewhat
capture market reality of the marketplace (at least allowing for marked-to-market option
prices), not in the valuation model, but rather in its implementation.
The Black-equivalent (implied) volatility of a mean-reverting model with the price and a
lognormal model is equal to the spot price volatility. For a mean-reverting model with the log
of the price the Black-equivalent volatility is:
α: rate of mean reversion

σ: volatility
t: time of observation
Explain cost-based models for electric utilities and their

disadvantages (3).
______
Modeling the price of electricity based on its cost. Useful for one particular utility, but not
the market. Disadvantages are:
1. Cost ≠ market price of electricity.

2. Cost-based distribution models cannot satisfy arbitrage-free requirements of forward
prices.
3. Do not allow for mark-to-market valuation, only for an internal mark-to-cost
valuation.

111
Describe how to calibrate parameters through time series
analysis (4 tests). What are the outputs of time series
analysis for the main three models?
______
With the outputs of time series analysis we calculate:
1. Seasonality.
2
2. R .
3. "3 test": The ratio of the 4th moment and the 2nd moment should equal 3 for a normal
distribution.
4. Autocorrelation. Negative for mean reversion.
Time series analysis of spot prices will result in:
Lognormal model:
1. An expected value (μ).
2. Price volatility (σ).
Mean-reverting model with the log of price:

1. Rate of mean reversion (α).
2. Log of the price around which mean-reversion occurs (b).
3. Price volatility (σ).
Mean-reverting model with price:

1. Rate of mean reversion (α).
2. Value of equilibrium price at start of historical data set (L0).
3. Rate of return on the equilibrium (μ).
4. Spot price volatility (σ).
When conducting time series analysis, unfortunately, we do not have the forward prices to help
us in a two-factor model, so we need to use a single-factor model with just the spot prices. We
have to assume that such prices do have a rate of return but zero volatility, i.e., they are
perfectly stable.

112
Incorporating seasonality effects. We assume:
and
We allow three seasonality factors: summer, winter, and an annual event factor (like a
seasonal hump in the fall, for example). The change in price is:
Differences are significant between the lognormal and the mean-reverting models, but not
between the mean-reverting models. Model-specific parameter vary significantly from market
to market. In all models and markets, the drift terms are much less significant than the
2
stochastic terms, so the type of drift incorporated does not matter. R values are far better for
the mean-reverting models than the lognormal model, which tells us that mean reversion is
extremely important in predicting next-day prices.
We also test for normality with the "3 test": The ratio of the 4th moment and the 2nd
moment should equal 3 for a normal distribution.

113
Autocorrelation should also be tested. Mean-reversion has negative autocorrelation.
Autocorrelation is [positive / negative] for mean reversion.

______
Autocorrelation is negative for mean reversion.

114
Describe the "3 test".
______
The ratio of the 4th moment and the 2nd moment should equal 3 for a normal distribution.
4th moment / 2nd moment = 3
Describe how to perform distribution analysis and its aim.

Conclusions about power and natural gas markets (2)?
______
In the distribution analysis we test how well models derived from time series analysis act over a
longer period of time compared with the actual market. With this step we will be able to
conclude which model ultimately is consistent with both short-term and long-term price
behavior. Historical market spot price distribution are compared to distributions implied by each
of the models we are testing, with the width of the distribution of primary importance. Tails
(skewness, kurtosis) are still important.
Model-implied distributions can be found by using Monte Carlo simulation (gives us visual
representation and moments) or mathematics and probability distributions of random variables
(no visuals but quicker and more precise moments). For the lognormal distribution, moments
are calculated as:
Conclusions are:
1. Power and natural gas markets are mean-reverting.

2. Power and natural gas markets are at least two-factor markets.

115
Discuss how locational marginal pricing works in electricity
markets. What if the system is congested/not congested? How
can LMP be modeled?
______
LMP was introduced by PJM in 1998 to clear the spot electricity markets. The LMP is the cost
of serving the next MW load at a specific location using the lowest cost generation and
observing all transmission constraints.
When system is congested, it produces nodal prices.

When system is not congested, it produces a single-market clearing price.
LMPs are published by the independent system operator (ISO) who has all relevant
information to calculate them. They can be modeled by using a backcasting process based on:
1. Historical fuel prices

2. Generation/transmission outages
3. Transmission constraints
4. Emission prices
5. Control variables

116
Why can energy forwards and futures be valued in a similar
fashion?
______
Because the correlation between interest rates and energy futures prices is null.
The energy futures and forward price can be used interchangeably if contracts are identical.
In the bond and interest rates markets, interest rates are directly related to futures prices, so
futures and forward prices are not equal.
Explain the role of seasonality in the three main forward

markets. Which market is the most complex?
______
Seasonality complicates the forward price curve. Depending on the use of the commodity,
seasonality peaks are stronger or weaker. Whether first- or second-generation fuel is used also
matters. In general, the greater the volume of trading of cross-commodity spread contracts, the
more unified would the energy market price behavior become across commodities. Seasonal
behavior can be broken down into three main markets: Heating oil, natural gas, and power.
1. Heating oil is mostly used in winter for heating. Backwardation and contango in the
near term, contango in the long-term of forward price curves. Requires a two-factor
mean-reverting model with an annual seasonality component.
The forward curve for heating oil has evolved over time and became more
complicated, mandating different models for newer curves.

117
2. Natural gas is also mostly used for heating in winter (strong winter seasonality). The
summer peak is very weak, as natural gas is rarely used for cooling in summer. Very
complex behavior in recent times, mandating many degrees of freedom. The natural gas
market behaves more like the power market.
The natural gas forward curve used to be simpler.
3. Electricity (power) has a strong winter and an even stronger summer peak, as it is mostly
used for cooling in summer. Most complex behavior. Lack of immediate storage causes
extreme volatility. Double-level seasonality: Seasonality of seasonality magnitudes.

118
Power markets used to be simpler. Such behavior can still be seen if there are no
medium- to long-term event expectations or slow dissipations of a very strong currency
event.
Examples of models are:

119
Explain "first-generation fuel".
______
First-generation fuel delivers the cheapest-to-deliver power to the customer. For electricity
generation, coal is the first-generation fuel and natural gas the second-generation fuel.

120
How are the energy forward price curve and spot prices
related?
______
Just as spot prices tells us about how forward prices act, the energy forward price curve tells us
about spot price behavior. Price behavior can be forecast. If the correlation between forward
prices and spot prices is very high, a single-factor model will be sufficient for a market.
However, as a rule, forward spot prices are NOT equal to expected spot prices.
Explain the arbitrage-free condition in energy commodity

forward markets. Complications in energy markets (3).
______
Arbitrage-free means that we should be able to construct a portfolio with the forward, spot,
bonds that will yield the risk-free return. The same type of ultimate benefit of owning an
energy commodity can be restructured in at least one other way, so we are indifferent about the
two scenarios.
In energy markets this is difficult for the following reasons:
1. Supply side limitations in volume, storage, transportation (transmission).

2. Market manipulation through few suppliers or producer organizations.
3. Consumers may not have access to markets with limited participation possibilities to
exploit the market.
Market characteristics could be captured in the model or in the implementation of the model.
The best approach is modeling the market with seasonality and incorporate characteristics in
implementation.
Explain convenience yield in energy commodity forward

markets.
______
Modeling energy commodity forward prices by solving a differential equation is made difficult by
defining the value of the convenience yield. The convenience yield reflects the value an

121
energy commodity brings to the user of having the commodity at hand minus cost for storing
and maintaining the commodity.
The convenience yield does not appear directly in the modeling of spot prices but during
supply shortages (surpluses), the spot prices in effect ride up (down) the convenience yield
curve, as spot prices capture the premium (discount) that users are willing to pay to have the
commodity at hand (get rid of it).
Convenience yield is always present in forward market prices, but difficult to model, as it is
only quantifiable when the commodity is actually delivered. It can be positive or negative
(parts of Europe who are in constant electricity surplus from too many functioning nuclear
plants).
Model an energy commodity forward price (formula).

______
with and
S: spot price
L: long-term equilibrium price
α': risk-adjusted rate of mean reversion
μ: expected return
μ': risk-adjusted growth rate
λξ: risk adjustment term, a function of the equilibrium price volatility, not spot price volatility
In more detail (Pilipovic):
In the stock market, assuming no dividends, we have:
(arbitrage-free condition with risk-free portfolio)
If we assume the stock will earn an expected return μ, then:
and . Introducing the market cost of risk λ:

and , it follows that: .
Translating into the energy commodity market, we assume the stock follows a lognormal model
with a rate of return μ and a volatility σ:

122
The risk-free portfolio can be constructed with a forward and a number of shares of the
underlying stock. Forwards require no upfront cash payment, so this is not the initial cash investment. Only the
stocks require cash at origination. Portfolio value:
The change in the value of the portfolio over time dt is:
The number of shares we need to hold is:
(derived from Ito's Lemma)
Solving the differential equation for the forward price we arrive at:
This is the same as the asset forward, assuming no dividends.
If we assume dividends, we must model the price as a non-holder of the stock, as the forward holder
does not own the stock until settlement. We can set dividends equal to convenience yield.
Solving the differential equation (Ito's Lemma), we get if we assume the convenience yield to
be constant:
Cy: convenience yield
Convenience yield can be calculated as:
or

123
Seasonality is incorporated in the implementation stage. We get:
Solving the differential equation we finally arrive (after some simplifications) at:
with and
S: spot price
L: long-term equilibrium price
α': risk-adjusted rate of mean reversion
μ: expected return
μ': risk-adjusted growth rate
λξ: risk adjustment term, a function of the equilibrium price volatility, not spot price volatility
λ: market cost of risk
Describe the market cost of risk (formula).

______
The market cost of risk (= Sharpe ratio) is defined as:
μ: expected return, r: risk-free return, σ: stock price volatility
In the stock market, assuming no dividends, we have:
(arbitrage-free condition with risk-free portfolio)
If we assume the stock will earn an expected return μ, then:
and . Incorporating , it follows that:
Describe production cost models and fundamental equilibrium

models in energy markets. Contrast them with hybrid models.
______

124
The goal of production cost models is to minimize cost of generating power to meet a certain
demand while satisfying operational and environmental constraints. Used by regulated utilities
for short-term dispatch and long-term planning. Production cost models do not capture price
evolution, and historical price information is scarce, opening up room for error.
Fundamental equilibrium models are not limited to cost minimization and incorporate market
data and trading activity. They model supply/demand relationships. They still do not model
price dynamics though.
Hybrid models seek to fuse the benefits of different modeling methodologies. They combine
fundamental equilibrium models (supply/demand) and pure stochastic models (evolution of
underlying drivers). Motivation for employing the hybrid model is:
1. Natural representation of the dynamics of power prices.

2. Consistent and explicit use of non-price forward looking information.
3. Ability to extend available data set from historical (and forward looking) power prices to
information on fuel prices, historical load, temperature, outage, etc.
Explain the hybrid model for power prices. Motivation (3).

Two types. What is the principal advantage of the hybrid
model?
______
Hybrid models seek to fuse the benefits of different modeling methodologies. They combine
fundamental equilibrium models (supply/demand) and pure stochastic models (evolution of
underlying drivers). Motivation for employing hybrid models:
1. Natural representation of the dynamics of power prices.

2. Consistent and explicit use of non-price forward looking information.
3. Ability to extend available data set from historical (and forward looking) power prices
to information on fuel prices, historical load, temperature, outage, etc.
The advantage of hybrid models is efficient use of disparate sources of information, which
leads to more efficient use of scarce price data. Hybrid models can be thought of as not a model
of specific volatilities, correlations, etc. but as a model of the relationship among all parameters
of the joint distribution. Two types of hybrid models are reduced-form and fundamental
models:
1. Reduced-form hybrid models. Transformation implied from historical prices (limited

data).
2. Fundamental hybrid models. Transformation implied from the structure of the market
and only tested on historical prices. Data is readily available.

125
The principal advantage of the hybrid model is that it generates a distribution of power prices
only using available market data and technical characteristics of the generation units, and does
not require use of historical data for calibration.
Explain the electric price jump diffusion model (2 types) and its
parameters.
______
1. Reduced form model depending on price data only (Merton, Heston).

2. Hybrid model using price data and temperature data.
The standard Merton jump-diffusion model combined with Heston stochastic volatility is
a reduced form model depending on price data only:
The jump rate λt on the variance is nonstandard, but corresponds to the economics of the
power generation. The probability of price spikes depends on market conditions, i.e., they only
occur when the system is constrained and the reserve margin is < 10%. Under hot weather,
expensive quick-start generation will be forced to keep the system stable, which results in a
significant price spike. The probability of a jump therefore depends on temperature ( ). This
is incorporated in the following hybrid model using price data and temperature data:

126
This equation significantly expands our data sample, as there are at least 50 years of temperature
data available in addition to the price data, which is very limited.
The approach to an effective separation of the estimation problem in two steps:
1. Estimation of the conditional response function conditioned on the dynamic factors

(e.g., fuel prices, loads, etc), usually on price data.
2. Estimation of the unconditional distribution, usually on nonprice or other data (e.g.,
fuel).
In general, jump diffusion models do not work well for electricity markets, since historical price
information is used to calibrate them which is scarce. We cannot have confidence the model will
accurately predict future prices based on historical prices.
Calculate the bid curve and market clearing price (MCP).

______
Steps to determine the MCP: Bid curve → bid stack → MCP.
1. Determining the bid curve of each generator. E.g., the ISO received the following bids
from a generator:
The generator is willing to produce the first 50 MWh for $20/MWh. If more power is
needed, the generator will produce from 0 to 100 MWh for $25/MWh etc. Any amount of
power between 200 and 400 MWh will be produced at a constant of $35/MWh.
2. ISO collects bids from all generators in the system and sorts them by price to obtain the
bid stack. Demand is determined from demand bids or other means. The system bid
stack is derived by combining all bids and plotting it.

127
Combining all bids.
3. MCP is determined as the highest priced on the system bid stack on which
total generation will be matched by demand. The bid from the marginal
generator determines MCP.

128
After the MCP is set, other markets (transmission, constraints) will open, which employ
similar bidding methods to determined the price. Extensions to multistack markets are
possible, but require cumbersome notation.
What are underlying drivers for power prices?

______
1. Price of underlying fuel(s).

2. Variable cost of running the generation unit (simplified into emission costs and block of
operating costs). Variable operational and emissions cost.
3. Outages, availability of units. Planned and forced outages.
4. Demand or temperature. The most important factor affecting prices in most markets.
Cost of generating 1 MWh can be written as:
HR is usually not a constant but varies.

129
The generation stack (supply stack) can be used as an approximation of the bid stack with
satisfactory results.
Construct a generation stack for electricity bid pricing.

______
A generation stack is a particular representation of a bid stack where generators bid into the
market precisely the generation costs of their units. Units are then sorted by generation cost
(from least to most expensive). Bid stacks are the result of rather simple transformations of the
generation stack.

130
Explain how the generation stack and the bid stack are
related in power markets.
______
Bid stacks are the result of rather simple transformations (multiplication and scaling) of the
generation stack. The higher-demand segments of the generation stack are responsible for
setting the MCP and price distribution, so higher normalized moments (skewness and kurtosis)
of the distribution should not be affected by multiplication and scaling transformations and are
about the same for generation and bid stack.
and ,
Linear transformation (stack linearity) is of course an approximation, so we expect skewness

and kurtosis to be preserved only approximately.
Prices (marginal generation costs) created using the stack can be represented as:

131
We know that, in the bid stack:
So we see that the higher moments of both distributions are roughly the same, which is
actually confirmed by historical data.
What are the cornerstones of successful modeling of power

prices (2)?
______
1. With the help of the generation stack and the distribution of demand, determine the
distribution of the marginal generation costs using .

2. Find a transformation with Pt being the MCP of power, such that:
o The distribution of power prices Pt is such that the prices of market instruments
used for calibration (i.e. forward and option contracts) are matched.
o The normalized higher moments (skewness and kurtosis) are preserved (at
least approximately).
If we know that our model matches all higher normalized moments plus two market constraints
(the forward price and an ATM option), then we can be sure we captured the whole
distribution.
The class of transformations Φ is extensive. A possible choice can be the family of

transformations consisting of multiplication and stack scaling operators. Historical prices are
never used to calibrate the model, only to test its soundness. This is the main weakness of the
jump diffusion model, which is based on historical prices.
Matching lower moments is not relevant, as they are functions of prices of traded
instruments and their impact on values of power derivatives can be hedged.

132
When can we be sure a model captures the entire
distribution (2)?
______
A model captures the whole distribution if we know that it:
1. Matches all higher normalized moments
2. Matches two market constraints

o Forward price.
o ATM option.
What precludes generators from always bidding their

generation costs (6)?
______
1. Outages. Random outages make the generation stack a random function. A probability-
weighted sum of all generation stacks could be bid, i.e. the expected generation stack
over all possible outages outcomes. A bid stack constructed this way will differ
significantly from a simple generation stack. Uncertainty in available generating capacity
when bidding one-day forward should be taken into account by the generator.
2. Operational optionality. Upper-bound on total run-time from technical or emissions
constraints. Running a unit at a today may preclude it from running at a more opportune
time later. A bid should therefore be higher than the unconstrained case to account for
opportunity cost. Applies especially for peaking plants which have limited run-time.
The option value of running the plant must be priced in. "Losing money today to win
more tomorrow".
3. Ancillary service markets. Established to stabilize and manage system reliability.
Generators can bid into energy market or ancillary service markets.
4. Real-time market. A generator may decide not to bid all the units in the day-ahead
market to preserve some capacity.
5. Environmental constraints. May prevent generators from bidding total capacity. Air
quality, temperature, adjacent water resources, etc.
6. Self-insurance premium. Added to price to provide protection against losses from
forced outages.

133
Describe the hybrid model for power prices (6 general
phases).
______
1. Fuel price model. Two main groups (group 1: tradable commodities, group 2:
alternative). Tradable commodities allow for forecasting fuel price evolution.
o Fuel cost of group 1 fuels is modeled by:
o Price evolution process with .

o Market data and model calibration. Market constraints:
 Matching the current forward curve.
  Matching options prices (ATM).
 Matching total fuel correlation structure.

2. Outage model. Outages modeled independently for each unit in a given regional
generation stack. We want a model to generate random outages in agreement with
technical and empirical data. The outage model generates an indicator vector. Steps are:
1. Model random time intervals between forced outages (i.e., availability periods).

134
2. Model the length of each outage.
3. Create an indicator vector, which takes on
values between 0 and 1.
1. Temperature model. Modeling daily average, maximum or minimum etc., the "spot
temperature" . Deviation from the historical mean and modeling this deviation can
be achieved with standard Brownian motion:
The forward temperature term structure could also be modeled using the HJM-
type approach:
Modeling temperature is not that different from modeling fuel:
2. Bid stack function.

o Group 1 fuels: . If fuel switching is possible:
.
o Group 2 fuels are not traded and prices must be estimated as a constant value.
o Emission costs. Quoted in $ per ton (SOx, NOx, etc.). These quotes must be
translated into $/MWh.
o Other variable costs (VOM). Variable operation and management costs (VOM).
All variable costs directly connected to generating output. Labor costs,
heating/cooling costs, a fraction of fixed costs.
o Generation stack. Sorting all available units according to costs. The value of the
generation stack function is: . The generation stack is entirely
described as: .
o Bid stack. .
Historical data is not used to calibrate the model, only to test it. Alpha parameters
are constants. We must find the triplet that best matches market data
(calibration).

135
1. Power demand. Models for demand (or load). Factor approach preferred to direct
modeling of load. Factors are temperature, humidity, wind-chill factor, economic,
demographic data, load composition. Resulting joint distribution of load should be tested
against historical data. Often, only temperature is needed to accurately describe
power demand. .
2. Power prices. Complication is wide range of power prices.
General notation convention:
Calibration of the model must match model output (simulated data) with market data.

136
Alpha parameters should not fluctuate more than 10% from month to month. Stability of
model parameters is an indication for a robust model.
Calculate emission costs for NOx and SOx for a power plant
(formulae).
______
Emission costs are quoted in $ per ton (SOx, NOx, etc.). These quotes must be translated into
$/MWh with the following equations:
NOx price = (heat rate / 1000) x (NOx rate / 2000) x NOx market price
Heat rate: Amount of Btus needed to generate 1 kWh.
NOx and SOx rate are plant-specific emission rates per MMBtu of fuel.
Example:

137
Describe the input data into the hybrid model for power
prices (6).
______
1. Power. Market prices used for calibration of the model.

o Monthly forward prices.
o Monthly option prices of options exercised into monthly forwards.
o Daily option prices.
2. Fuel.
o Monthly forward prices.
o Monthly option prices.
o Daily option prices.
3. Regional generation of stack data.
o Heat rate curve of group 1 fuels (as a function of generation levels).
o Variable cost of generation per 1 MWh for group 2 fuels.
o Emission costs ($/MMBtu).
o Additional information (start-up costs, ramp-up/down rates, min. uptime
and downtime, total emission constraints, etc.).
4. Outage data. Equivalent forced outage rates (EFOR) supplied for each generating unit in
the regional stack.
5. Temperature.
o Historical average daily temperatures or forecast.
o Historical standard deviation of daily temperatures.

138
6. Modeling goal. To describe evolution of power and fuel prices, and daily
temperatures such that:
o Constraints imposed by market data are satisfied.
o Distribution of power prices has required properties.
o Joint distribution of power and fuel process has required properties.
Summarize the steps in producing the hybrid model for

power prices.
______
1. Simulating fuel prices.
2. Simulating temperatures.
3. Simulating power demand (load).
4. Simulating outages.
o Corresponding stack unit is available:
o Corresponding stack unit is unavailable:
5. Building the generation stack.
6. Process of power prices.

139
Discuss the justification of the hybrid model for power
prices (9 points).
______
We need to demonstrate that the model accurately depicts market data and empirical
properties of power prices. Does the model account for:
1. Spikes.
2. Mean reversion in both high and low demand periods.
3. Fat tails of the price distribution because prices are modeled as an increasing function of
temperature.
4. Seasonality.
5. Volatility structure.
6. Correlation structure between different forward contracts for same commodity.
7. Test of higher normalized moments of the distribution (should be equal).
8. Test for correlation between power and natural gas prices (especially out-of-
sample). Implied heat rate = power price / natural gas price.
9. Test of the inverse-leverage effect. Specify increase of power price volatility at
higher levels of the implied heat rate.

140
Explain the implied heat rate and what a low and a high rate
means.
______
Implied heat rate = power price / natural gas price.

Low implied heat rate: Gas is not a marginal fuel and lower-cost units, such as coal plants, are
setting the power price. Low correlation between power and gas price is expected.
High implied heat rate: Demand is high and power prices are set by high-cost units located on
the steep segment at the end of the generation stack. Power prices are mostly sensitive to
changes in demand.
The implied heat rate metric is used to assess the hybrid model of power prices,
comparing model simulations with empirical market data.

141
Describe a costless collar. How is it constructed?
______
If single calls or puts are too expensive and the customer is willing to give up part of a potential
profit from low prices. Long OTM call and short OTM put, reducing the cost of the call.
Protection levels can be chosen for the collar to be "costless".
Describe a swing, recall, and nomination.

______
Volumetric options that give the holder the right to adjust the volume of the received or delivered
commodity.
Swing: On N dates, the holder receives a set amount of a commodity. The option gives
the holder the right to vary the amount delivered within a certain range. This right can
be exercised K≤N times. Common in natural gas. Used to manage demand.
Recall: On N dates, the seller must deliver a set amount of a commodity. A recall gives
him the right to not deliver (or recall) nominated volume a couple of times during the
period. Used to interrupt delivery under stressful circumstances.
Nomination: Offers the holder the right to change the volume received, but the level of
volume is adjusted upwards or downwards to the last chosen volume for remainder of
the contract until next nomination right is exercised. Allow a very limited number of
nomination rights. No limitation on individual swing size, but for cumulative minimum
or maximum value taken (during a year).

142
Explain how a swing option is valued.
______
A swing gives the holder the right to vary the quantity received within a certain range for a
certain number of times during the period. This right can be exercised K≤N times. Swings
are used to manage demand. For valuation, swing options can be treated as multiple
American options with mutually exclusive exercises approximated by EC < swing option <
AC, depending on the number of swing rights.
The value of a swing option comes from two sources:
1. End-of-period premium. Ability to exercise the option at the end of the period,
significant in contango markets. At term end, volatility contribution is highest.
2. Early exercise premium.
The value of a swing option is homogeneous to degree one. Valuation involves building a two-
dimensional binomial tree and backward induction.

143
144
Explain the impacts of volatility, convenience yield, and
transmission constraints on energy options pricing.
______
The lower the volatility, the higher the early exercise premium.
Negative convenience yield has a zero early exercise premium. Early exercise only
makes sense when the convenience yield > 0. Convenience yield is very difficult to
estimate.
Distinguish between economic and uneconomic exercise of

an option. What are causes (2) of uneconomic exercise?
______
Parties usually follow mixed policy, sometimes optimizing the exercise to the spot price and
responding to demand changes at other times.
Uneconomic exercise may be due to:
1. It may be physically impossible to vary required volume with the prices.
2. A city gate specified for delivery may be illiquid, so the excess gas cannot be re-sold.

145
Describe the two flavors of Asian options, and how they are
related.
______
1. Average price. , .
Valued best with Monte Carlo simulation (slow), but the more practical (accuracy and
speed) solution is numerical based on geometric averaging (Vorst or Curran, as the log or
the geometric average is normally distributed which can be solved with numerical
methods).
2. Average strike. ,
. Also valued with the Curran method.
The following symmetry exists:
with
Often, the contracts do not have a fixed strike price but are struck at the first-of-the-month index
price. A forward start Asian option with a floating strike can be seen as a forward start call
option on a swap, with the strike set to the swap price.
Describe a caption and floortion.

______
A caption is a call on a cap (series of calls).

A floortion is a call on a floor (series of puts).

146
Value a spread option with the Macgrabe formula.
______
with
←
Almost identical to the Black formula for a call but for the volatility. The spread option price
decreases with increasing correlation. Sensitivity to correlation is highest for OTM options.

147
As the dependence structure is somewhat of a thorny subject, spread options can also be modeled
directly, omitting the lognormality assumption (which is not very realistic in the first place).
Explain and calculate the forward value of storage in energy

markets.
______
Storage is one of the most complicated options structures in energy markets, even more than
power generation. The major functions of storage in fuel markets are:
1. Seasonal cycling. Shifting cheap idle summer production of natural gas into winter
months at a relatively low cost.
2. Peaking service/peaking deliverability. A fast storage facility costs much less than a
fast production facility.
3. Balancing service. To respond to pipeline disruptions.
Main operational characteristics of storage are:
1. Capacity. Measured in Bcf (1 Bcf = 1,000,000 MMBtu).

2. Withdrawal rate (withdrawal daily quantity, WDQ). Expressed in MMBtu/day or
percentage of total capacity per month.
3. Injection rate (injection daily quantity, IDQ). Expressed in MMBtu/day or percentage of
total capacity per month.
4. Injection/withdrawal costs.
5. Injection/withdrawal losses.
6. Ratchets (as WDQ and IDQ are not constant but depend on inventory level).

148
Explain WDQ and IDQ.
______
WDQ (withdrawal daily quantity), indicates the withdrawal rate.
IDQ (injection daily quantity), indicates the injection rate.
Expressed in MMBtu/day or percentage of total capacity per month. WDQ and IDQ are
not constant but depend on inventory levels.
Explain possible storage contracts in energy markets (4).

______
1. Peaking services.
2. Summer fill. Lessee receives an empty facility at the beginning of the term and returns it
full at the end of summer (i.e., October).
3. Seasonal cycling. Lessee receives an empty facility and returns it empty.
4. Storage carry. Lessee receives facility full and returns it full.

149
Explain the valuation approaches to storage in the energy
market (4).
______
1. Forward optimization.
2. Forward dynamic optimization. (Spread) option portfolio (linear and nonlinear
optimization).
3. Stochastic dynamic programming. Backward induction on trees, Lagrangian relaxation,
etc. Relies on direct modeling of the spot process (calibrated to forward prices) along
with the development of the optimal exercise rule for storage flows given the dynamics.
4. Combinations.
Calculate the forward value of storage in energy markets

(formula).
______
Storage value = withdrawal revenue - injection cost.
Injection cost = arithmetic average of injection points.

Withdrawal value = weighted average of withdrawal points.
The value of storage options is essentially a portfolio of complex spread options.

150
from:
Example:

151
If the storage facility is turned more than once:
Explain how the value of a storage facility can be increased

(2). The value of storage options is essentially ... .
______
1. Enter into the forward positions suggested by the optimal injection / withdrawal
schedule for this forward curve.

152
2. If the forward changes favorably, readjust positions to capture the positive difference. If
the curve moves in an unfavorable way, do nothing.
The strategy always has positive value, and is therefore an option position with a value
dependent on the variability of the forward curve (which depends on the variability of the
contracts and their correlation). The higher the volatility, and the lower the correlation, the
higher the option value. The value of storage options is essentially a portfolio of complex
spread options.
with
Describe transportation and transmission transactions in the gas

market (2).
______
1. Firm transmission capacity. The receiver gains control over the flow of the fuel through
the pipe. Valuation as a strip of spread options.
2. Interruptible capacity. Seller has the discretion to flow the gas through the pipe on a
best-effort basis.
The value of the spread depends on topography of the terrain (e.g., CA, where no other pipelines
exist).

153
Describe transmission issues in the electricity market.
Transmission can be seen as ... .
______
With wheeling power from one location to another we cannot be sure which way the
electricity exactly takes to reach destination, depending on system-wide conditions.
Transmission can be seen as a spread option, but to manage the uncertainty of the path,
transmission contracts specify a path that is used for financial reconciliation of all
transmission charges incurred along the way. The strike price of the spread option therefore
does not have a straightforward structure.
Explain load-serving contracts (5).

______
1. Full requirement deal. The delivering party takes on all obligations associated with
serving the load (paying for/supplying installed capacity, ancillary services).
2. Interruptible contract. Variable energy contract with provisions for interruption of

service for a given amount of load a set number of times during the term. Can be
decomposed into a standard load serving-deal and swing/recall options. Exercise,
whenever a spike occurs.
3. Banded hourly shaped energy. Specifies an hourly load range within which the energy
delivered will be sold at a set price, possibly indexed to the forward curve at expiration.
Outside the range, the differential is priced at the current spot price.
4. Block power. Involves a set amount of power throughout the tenor of the
contract. Examples are standard on-peak forwards and their variations (round-the-clock
[RTC], off-peak, peak, super-peak, etc.).

154
5. Fixed shape. A block contract but the everyday shape within the contract can vary.
Similar to the banded hourly shaped contract, except the delivered amount is known with
certainty at inception of the contract. Essentially a banded product with min. and max.
bands equal.
Explain the load-factor.

______
In unbanded variable energy products: A = average load / peak load

In banded variable energy products: A = average load / max. contractual load
Variable load contracts often contain look-back provisions on realized load factors, which
adjusts pricing terms depending on the actual load factor.
where:
Discuss the potential for using weather derivatives to hedge

retail contracts.
______
Weather derivatives are largely unfeasible for hedging, even though they promise significant
potential for hedging load exposure, as power load is largely driven by weather factors
(temperature). But residual weather risk may be insignificant: If the correlation between load
and price is high, the optimal price hedges (forwards, options) may already hedge most of the
volumetric risk.
The relationship between load and price tends to be strong for low to medium prices but
breaks down for high prices. The volatility of power prices increases and drives most cash
flow variability and associated risks. There is usually no way to decide what the optimal
weather hedge should be, making it largely unfeasible. Price products also offer more
liquidity, making for a better hedge.

155
Explain the characteristics of the power market that set it
apart from other commodity markets (3).
______
1. Non-storability of electricity and hence lack of inventories, requires the development of

new approaches to power markets from a financial and economical standpoint.
2. Geographical distinct US markets. Moving power is either impossible or

uneconomical between some markets. Geographical refinement of contracts is
taking place.
3. Non-existing or incomplete market for power options. Hedging portfolios do not exist or
are difficult to identify, especially for daily options. Nonexistence of a unique options
price widens the bid-ask spread.

156
Discuss the implication of non-storability of electricity on
pricing spot price power options (3).
______
Non-storability of electricity imposes obstacles for extending the notion of convenience yield
to power options:
1. Convenience yield undefinable. By definition, convenience yield = difference between

the positive return from owning the commodity for delivery and storage cost. Both
quantities cannot be specified because of non-storability.
2. Breakdown of the no-arbitrage argument, which requires the commodity to be bought
and stored until expiration of the futures contract.
3. Delta hedging cannot be implemented (hedges involving the underlying asset)
because it requires storability.
By using futures and forwards the problems of non-storability of power can be avoided, as they
do not involve convenience yield.
Describe why the use of futures and forwards provides a

solution for valuing power options.
______
By using futures and forwards the problems of non-storability of power can be avoided, as
their formulae do not involve convenience yield.
The power stack function is:
The evolution of F(t,T) can be modeled using the standard Black-Scholes framework with
an appropriately chosen volatility term structure, generating adequate hedges for monthly
and calendar yearly options, but not daily options.

157
Explain why daily forwards should not be used to hedge daily
positions. What is the safest way to hedge daily power
options?
______
Most markets do not have liquid daily forward or futures contracts. We are therefore forced to
use an inadequate surrogate for the daily futures contract called the "balance-of-the-month"
(BOM) contract (price of power delivered every day until the end of the month). This will
become a surrogate for the daily spot price, leaving us with an incomplete market.
When modeling spot price dynamics, main issues are matching fat tails of marginal and
conditional distributions and the spikes of spot prices. Today, the safest way to hedge daily
power options is to lease or own a power plant. Operating a merchant power plant is
financially equivalent to owning a portfolio of daily options between fuel and electricity
(spark spreads). Max. profit from running the plant only if the market price is higher than the
cost of fuel and operating cost is:

158
Describe the major sources (6) of greenhouse gases in the
EU. Which is the largest emitter?
______
1. Energy industries
2. Transport
3. Industry
4. Households and SMEs
5. Agriculture
6. Others

159
Explain how emissions trading saves money for participants,
example.
______
Consider two market participants who must reduce remissions:
A's cost of reduction is larger than B's, so A could ask B to reduce emissions for A while
being paid a price that is lower than A's price for the reduction. If B agrees to reduce
emissions by an additional 10 tons for $75/t, then:
The overall reduction has stayed the same, while emissions cost has been reduced. In a large
market, the market will set the price for emissions. Also, emission reduction costs will increase
with increasing emission reduction.

160
Discuss the history of worldwide emission trading.
______
The EU did not invent emissions trading. The emissions markets covered in the Kyoto protocol
are largely based on American CO2 and SO2 markets, which have many years of experience. The
SO2 market started in 1995 in America to reduce acid rain, with much success. The carbon
market (80% GHGs) is the largest. However, multi-jurisdictional registries in the U.S. are a
disincentive for companies to engage. Harmonized standards would encourage more companies
to participate and emissions to trade globally.
In 2006, the estimated total notional value of emissions trading (green trading) was $10B
(including outstanding trades). The carbon market was about $2B, the SO2 market about $8B,
and remaining markets below $1B. These markets are anticipated to be over $100B in 2010.
SO2 markets started in 1995 and are the most sophisticated and liquid. Annual trading volume is
estimated at $3B. NOx trading started in 1999.
Discuss thoroughly cap-and-trade.

______
Under cap-and-trade, allowances ("rights to pollute") equal in number to the volume of

allowed emissions are allocated to market participants. One permit allows a ton of the
specified emission. If taken up, the permit must be retired. If emissions exceed permits,
they must be bought in the open market. Non-compliance is subject to government
sanctions and fines, but the market sets the actual price.
The goal is to reduce pollution in a cost-effective manner while causing minimal economic
disruption. Emissions are reduced over time and the industry has time to comply with the
law and invest in new, environmentally benign, technology.
The business case can be summed up as "pay less now or more later".
Define renewable energy credits (RECs) and explain how

RECs are used in the United States. Market drivers (5).
______
RECs, also called green certificates, are the currency of trade in RPS (renewable portfolio
standards) programs, which require electricity wholesellers and retailers to include a

161
proportion of electricity generated from renewable energy in their sales portfolio. They must
demonstrate annually that they have sold the minimum amount of renewable energy specified
in the RPS. Shortcomings can be covered with RECs in 12 out of 19 U.S. programs and in
Europe.
The market drivers for RECs are essentially the same as for renewable energy or green
power, including:
1. Green-pricing programs for utilities.

2. Competitive green power marketing.
3. State renewable portfolio standards (RPS) and purchase mandates.
4. Electricity information disclosure.
5. Requirement that RECs be used for compliance with certain programs.
The market for RECs is expanding fast. But the application of RECs to emissions markets is in
its infancy. The value of RECs is currently higher in energy markets than in GHG-emissions
markets.
Discuss SO[sub]2[/sub], NOx, and CO[sub]2[/sub] markets.

2,NOx, and CO2 markets.
______
SO2 markets started in 1995 and are the most sophisticated and liquid. Annual trading volume is
estimated at $3B. Markets were establish to combat acid rain, which they reduced very
successfully. Units must be registered and verified by the EPA. A new EPA program, the Clean
Air Interstate Rule (CAIR, 2005) is set to reduce SO2 emissions by a further 70% in the 28
affected eastern states. The SO2 market is not a true commodity market but a creation of the
government, imposing a cap on the market. Declining banked emissions created excessive
volatility and high prices lately, about $700/t. The Chicago Climate Exchange (CCX, which
created its SFI, sulfur financial instrument) and the NYMEX have therefore created SO2 futures
to increase liquidity and market efficiency.
NOx markets have been in effect since 1999 as a federal program. Markets are regional, as
smog is a regional problem. The EPA's NOx Budget Trading (NBT) program is a cap-and-
trade program similar to the SO2 program, where units must be verified with the EPA.
CO2 and GHG markets are expected to see a large boost from ratification of the Kyoto protocol
in 2005 with JI and CDM projects. The EU ETS market is expected to be worth $15B per year.

162
Define negawatts. What are benefits (7)?
______
The negawatt (or energy-efficiency) market measures energy reduced rather than energy
produced. The key to today's negawatt markets is demand response (curtailing power
consumption when load is high). Demand response benefits include:
1. Cost savings.
2. Systems reliability.
3. Market efficiency.
4. Risk management.
5. Environmental.
6. Customer service.
7. Market power mitigation. Can reduce or defer new plant requirements.
8. Customer service. Greater control and choice.
9. Market power mitigation. Less vulnerability through choice.
Describe the term hockey stick as it pertains to energy and

emissions markets.
______
Flat growth at the beginning and sudden exponential growth later.
List the world’s major climate exchanges (9), and their main
participants (6).
______
The main exchanges are:
1. Chicago Climate Exchange (CCX).

2. Chicago Climate Futures Exchange, a CCX subsidiary.
3. NYMEX.
4. European Union Emissions Trading Scheme (EU ETS).
5. European Climate Exchange (ECX), a CCX subsidiary in London with ICE.
6. Nord Pool, trading EUAs.
7. Powernext (power and carbon exchange in France)
8. EEX (energy exchange in Germany)

163
9. EXAA (Austrian energy
exchange) The main participants are:
1. Companies seeking to comply with emissions targets by buying allowances.

2. Companies selling their surplus allowances for profit.
3. Municipalities.
4. Financial institutions.
5. Private individual investors.
6. Projects that have been created under the Kyoto Protocol.
Explain how the Chicago Climate Exchange (CCX) operates, its

classes of membership (4), and major products (2).
______
The CCX is a voluntary but legally binding GHG emissions-trading program to develop
institutions and skills needed to manage GHG emissions in absence of government regulation
and to demonstrate that companies will reduce emissions voluntarily. Started based on a
feasibility study of Northwestern's Kellogg Graduate School of Management and several grants
in 2000/1. Classes of membership include:
1. Members who emit GHGs in facilities in North America.

2. Associate Members who have small or no direct GHG emissions but offset business-
related emissions.
3. Participant Members who are offset providers (project owners or implementers), or
liquidity providers (40%).
4. Exchange Participants, non-US parties with an account for acquiring traded
allowances.
Only CO2 is monitored and offset. To standardize CO2, CCX developed the Carbon Financial
Instrument (CFI), equivalent to 100 metric tons of CO2. The instrument comes in two forms:
1. Exchange Allowances (XAs). Issued to exchange and associate members according to

their baseline emissions for carbon sequestration in forests and reductions in electricity
use.
2. Exchange Offsets (XOs). Issued by other mitigation projects registered with CCX by
participant members. XOs are always issued after CO2 reduction has occurred.
Eligible projects include:
o Landfill-methane destruction in the U.S.
o Agricultural-methane destruction in the U.S. o
Carbon sequestration in U.S. forestry projects. o
Carbon sequestration in U.S. agricultural soils.

164
o Fuel switching, landfill-methane destruction, renewable energy, forestry projects
in Brazil.
Data is available in Bloomberg to non-members.
Explain how the Chicago Climate Futures Exchange

operates, and its major product.
______
The Chicago Climate Futures Exchange is a subsidiary of the CCX, started in late 2004. It
launched Sulfur Financial Instruments (SFI) in December 2004, which standardized and cleared
futures contracts for SO2 emissions allowances (which were traded OTC since 1990) for
management of price risk and fluctuations of SO2 allowances. Contracts are for 25t of emission
allowances (contended to be 100t).
Explain how NYMEX is involved with emissions trading.

______
The New York Mercantile Exchange (NYMEX) has the opportunity to become the premier
green exchange as it is already the market leader in energy commodity trading. The exchange
does not like to be the first mover though, and will wait for emissions trading to become more
established.
NYMEX has an established OTC clearing operation through Clearport which since
2005 includes SO2 and NOx futures contracts, but no emissions certificates.

165
Explain the European Emissions Trading Scheme (EU ETS).
What is the major shortcoming?
______
EU ETS is a cap-and-trade program designed to reduce GHGs in accordance with the Kyoto
protocol targets (8% CO2 reduction below 1990 levels by 2008-2012). The ETS permits both JI
(joint implementation) and CDM (clean development mechanism) projects. Larger, more-
developed EU countries like Germany and Italy will share the bulk of reductions in Europe (8%
bulk).
Several exchanges have emerged under EU ETS. The scheme itself is not a trading platform
which is supplied by private ventures.
ECX
Nord Pool, trading EUAs.
Powernext (power and carbon exchange in
France) EEX (energy exchange in Germany)
EXAA (Austrian energy exchange)
The failure lies in only creating a scheme, but not a unified European trading platform for
GHGs, renewable energy and emissions trading. All the exchanges that have sprung up will
either fail or have to consolidate in the long run.
Explain how the European Climate Exchange (ECX)

operates, and its major products.
______
The ECX is a subsidiary of the CCX, launched on April 22, 2005. It uses the electronic
technology and clearing mechanism of the Intercontinental Exchange's ICE Futures (formerly
International Petroleum Exchange), the second-largest energy futures exchange in the world.
The ICE Futures exchange is all-electronic.
The ECX offers carbon financial instruments (CFIs) worldwide, which are advanced, low-cost
and financially guaranteed (London Clearing House LCH) tools for trading emissions
allowances issued under EU ETS. Over 50 leading global banks and companies trade ECX
products. The ECX works with other exchanges and EU countries with the largest emissions to
develop smooth reporting and trade clearing.
CFI Futures are traded on ICE Futures, with EU ETS being the underlying market. EU ETS
accepts CERs, but does not allow their trading.

166
Explain how Nord Pool operates, and its major products.
______
The Nordic Power Exchange (Nord Pool) also participates in the EU ETS. Nord Pool is the
most successful electricity-futures exchange in the world and also clears OTC, but has struggled
to capture as much volume as ECX or Powernext.
It started trading and clearing European Union Allowances (EUAs) in 2005. EUA forwards
traded on Nord Pool have physical delivery. Uses its power Click trading platform.
Explain how Powernext Carbon operates. Principles (4).

______
Powernext Carbon is a joint venture of Caisse des Depots, Powernext, and Euronext. It offers a
spot market enabling exchange of allowances and non-compliance risk management since the
end of 2004. Based in France. Principles:
1. Straight-through process from transaction to delivery.

2. Price transparency.
3. Non-discriminatory access to market and complete anonymity of participants.
4. Guarantee of transaction performance based on a settlement mechanism which secures
payment and quota delivery.

167
Explain how the European Energy Exchange (EEX)
______
EEX launched futures trading in EU emissions allowances (EUA futures) in October 2005,
paving the way for a CO2 derivatives market. Based in Germany. Trades European Carbon
Futures, denominated in 1,000 EU emission allowances. Maturity up to and including
2012. Prices of emission allowances can be hedged, with the EEX clearinghouse
assuming counterparty risk. OTC futures are also cleared.
Highest number of members (128) and trading turnover in Europe. EEX already
traded electricity futures and options before.

168
169
Explain how the Energy Exchange Austria (EXAA)
______
Austria has been the front-runner of liberalizing energy market in Europe in 2001. All-electronic
platform. Electricity spot trading was introduced in 2002, CO2 emissions certificates in 2005.
Per Austrian law (Emissionszertifikatsgesetz), carbon allowances are considered goods (not assets,
like almost everywhere else), which enables EXAA to trade EUAs as a commodity exchange.
EXAA guarantees fulfillment of open positions and physical delivery of certificates.

170
Discuss the structure of a risk matrix (how many types of
risk?).
______
9 risk types.
Discuss each type of major risk (9) in energy markets.

______
1. Regulatory risk. The risk that regulators, governments, international authorities, etc.
impose new controlling practices, law, specifications of materials, and market
design/structure or change existing rules.
2. Price risk (= market risk). Risk of losing money when markets move against us.
3. Credit risk. Risk of financial losses from default of a counterparty.
4. Liquidity risk. Strong adverse market moves cause traders to not even bid, making it
difficult to liquidate certain derivatives positions. Derivatives positions, even when
entered in a text book way, may be come too large and move the market. If liquidity is
bad, a proxy hedge can be considered (Brent futures instead of WTI futures, "Cushing
Cushion").
5. Cash flow risk. Risk of not having enough currency (in the right denomination) to meet
derivatives obligations. If hedges are held to maturity, there is usually no liquidity or
market risk, but supporting the hedges imposes cash flow risk. Often also caused by
currency risk.
6. Basis risk. Risk of loss due to an adverse move or breakdown of an expected differential
between prices. Risk that the value of a hedge may not move in sync. In OTC swaps and
options, basis risk can be zero at times because they can have the same price reference as
the underlying, contrary to futures.

171
7. Legal risk. Risk that derivatives contracts may not be enforceable, especially netting and
settlement. Consult ISDA and netting agreements.
8. Tax risk.
9. Operational risk. Through errors and omissions in processing and settlement of derivatives.
Internal controls and an appropriate back office system can mitigate this risk.
Understand and list the major types of basis risk (4).

______
Basis risk is the risk of loss due to an adverse move or breakdown of an expected differential
between prices. Risk that the value of a hedge may not move in sync. In OTC swaps and
options, basis risk can be zero at times because they can have the same price reference as the
underlying, contrary to futures. Major types of basis risk are:
1. Product basis. Most important basis risk in energy markets. Mismatch in quality. A large
number of products exists in energy markets, but only a limited number of liquid hedging
tools. Energy price risk is therefore hedged only with a limited number of liquid OTC and
futures markets. In times of high volatility (Gulf War), historical relationships can break
down.
2. Time basis. Common exposure in many markets. Occurs when there is a sudden shift in
demand or transportation problems.
3. Locational basis. When energies are located in different regions.
4. Mixed basis risk. More than one type of mismatch (locational and time basis).

172
Assess basis risk (5 steps).
______
Between hedging tool and underlying exposure:
1. Same oil specification?

2. Same physical location?
3. Same timing?
4. Price correlation? Best over three years.
5. Price causation?
Which factors can introduce major energy market moves

(4)?
______
1. Weather.
2. Political events.
3. Physical events.
4. Regulation.
Calculate the payments on a plain vanilla energy swap.

______
The swap buyer pays a fixed price and receives a floating price. The swap seller pays a floating
price and receives a fixed price. The floating price is a reference price derived from Platt's or
Argus or a futures exchange. The fixed price is agreed upon by the parties. A producer of energy
(refiner) will sell a swap to lock in the sales price of his product. He will receive fixed and pay
floating.
If fixed > floating price, the swap buyer pays the difference to the swap seller. If
fixed < floating price, the swap seller pays the difference to the swap buyer.
Difference = contracted monthly volume x (fixed - floating price)
Example:

173
A consumer of energy (airline) wants to lock in the buying price. He will pay fixed and
receive floating.
If fixed > floating price, the buyer pays the difference to the seller. If
fixed < floating price, the seller pays the difference to the buyer.
Example:

174
175
Explain a differential swap. Who would use it and why?
______
A differential swap is based on the difference between the fixed differential between two
products and the actual or floating differential over time. Examples are jet fuel vs. gasoil,
physical gasoil vs. futures, Brent vs. WTI, etc.
Differential swaps are used by refiners to hedge changing margins of refined product by selling
a swap. The refiner receives a payment if the differential narrows (i.e., the margin has fallen)
and must pay if it widens. An airline may buy a differential swap to hedge potential basis risk.
Example:

176
Explain a margin swap (= ...).
______

177
Margin swaps (= crack swaps) can guarantee the profitability of refinery for a few years
forward. Refiners sell a margin swap and receive fixed, pay floating. Product output and
feedstock input are simultaneously hedged. The deal is usually expressed as USD x per bbl
margin. This hedge is often used in development projects when the financiers want to ensure the
viability of the project by underwriting a minimum revenue stream.

178
179
Explain a participation swap.
______
Participation swaps are similar to regular swaps in that the fixed rate payer is protected against
input price hikes, but "participates" if prices fall. E.g., a participation swap at a level of $80 per
ton of oil with a 50% participation will protect the buyer against price rises above $80/t and will
allow him to keep 50% of profits if prices fell below $80/t.
P: participation percentage
Example:

180
181
Explain an extendable swap.
______
The provider has the right to extend the swap at the end of the period for a further determined
period. The swap price will be adjusted by the option premium.
Example:

182
Describe the major benefits (9) of energy swaps.
______
1. Producers and processors can offer fixed-priced products to their customers.

2. Refiners can lock in refining margins.
3. Production margins can be guaranteed in development projects.
4. Banks can offer more attractive financing when price exposure is controlled.
5. Pre-export financing can be secured on NPV of swap cash flows.
6. Exposure to one oil product can be switched to another (an airline fixing futures prices in
gasoil can eliminate the inherent basis risk by entering in a jet-gasoil differential swap).
7. Competitive advantage can be secured by locking in high or low prices.
8. Limitations of the exchanges can be overcome (liquidity, duration, margin adjustments,
limited product range).
9. Regulation and accounting policies can be overcome.
Most users only hedge the current fiscal year, with an exposure of 40-60%. More companies now
also protect their 3 to 5 year budgets with swaps.
Why do users of swaps not hedge their entire exposure?

Main reason and 4 other reasons.
______
They want to profit from favorable price developments. If all prices were locked in, the
company may find itself at a price disadvantage compared to competitors.
Most users only hedge the current fiscal year, with an exposure of 40-60%. More companies now
also protect their 3 to 5 year budgets with swaps. Other reasons for only partial hedging:
1. Lack of knowledge of derivatives.

2. Inefficiency. Energy swaps have been inefficient due to lack of liquidity in the
derivatives market.
3. Debate on accounting treatment of derivatives.
4. True cost of a swap may be difficult to resolve.

183
Describe the major users (9) of energy swaps.
______
1. Airlines. Hedging jet fuel (fuel is 20% of airline operating costs). Deteriorating balance
sheets have allowed airline to only hedge about 11 months forward due to concerns
over default. They must use their house banks for access to derivatives markets. Market
bid-offers have widened.
2. Shipping companies. Variable cost determined by bunker-fuel prices. Using long-term
derivatives, as shipping contracts often span 10+ years. Less efficient "bunker adjustment
clauses" in contracts are replaced by OTC derivatives.
3. Transport companies. Exposed to diesel fuel prices. Fragmented and diverse client base
(often small companies), poses airline-like risks.
4. Power companies. Oil, gas, coal are a large portion of variable cost. Prices to consumer
can not quickly be varied. No adequate reference price mechanism.
5. Industrial groups. Firms with high energy consumption, such as metal smelting, cement,
glass manufacturers increase use of swaps for fuel oil.
6. Chemical companies. Petrochemical companies hedge naphtha prices.
7. Financing organizations. Banks and institutions that finance oil development projects.
Repayment often linked to oil output which is linked to oil prices. Oil swaps may be
linked to bonds, warrants or other securities to create investment products as an
alternative to conventional assets.
8. Hedge funds. Specialist fund management market uses commodity swaps.
9. Investors. While bonds and equities have performed poorly between 2000 and 2003,
commodity returns have averaged 7-45% via complex investment vehicles such as
"range accrual notes", "enhanced yield", or "commodity tracker".
Describe financing problems (4) that can be solved with

energy swaps.
______
1. Project finance. Limited-recourse projects. Removing market risk with swap structure.
Pre-selling output of a development project at a fixed price.
2. Pre-export financing. Producing countries can hedge price risk to ensure they can finance
(and roll over) debt accrued because of development spending. Future production (if
properly hedged against downturns in prices) can be pledged as collateral against
immediate cash.
3. Asset finance. Cost of financing can be linked to fuel exposure. Aircraft or ship finance
can be indexed to jet or bunker fuel prices, refinery equipment cost can be indexed to
precise refinery margins.
4. Bond and equity issues, placing. Coupon or interest can be indexed to oil.

184
Describe how an energy swap is priced (4 steps).
______
1. Assess spot prices of input for the term of the swap.

2. Add crack spread for product 1 to prices and convert to tonnes (EU) or gallons
(U.S.) with a conversion factor. Mark up or down to reflect volatility, liquidity, and
basis risk.

185
3. Add differential between product 1 and 2 to prices from step 2. Mark up or down to
reflect volatility, liquidity, and basis risk.
4. Calculate average to find offer price.

186
Describe the impact of illiquidity (2 risks) on energy swaps.
______
Not only liquidity of the swap market itself (which influences bid-offer spreads), but liquidity of
wider energy market must be taken into account in risk management. Risks associated with
illiquidity are:
1. Basis risk. Further-out liquidity is important.

2. Hedge risk in the near term, where any trade will move the price.
Describe the use of conversion factors in energy swaps.

______
Conversion factors are used to take into account the different densities of various commodities.
1t gasoline = 8.33
bbl 1t jet fuel = 7.88
bbl 1t fuel = 6.35 bbl
1bbl = 42 gallons

187
Explain a Btu swap.
______
Fixed-for-floating swap in which the receiving party pays a fixed amount and receives the value
of a set amount of energy. The source of the energy can by any amount of fuel, of which
usually the cheapest is chosen for delivery by the supplier.
A Btu swap can be decomposed into a natural gas swap and a strip of spread options:
State the key assumptions with regard to cap-and-trade.

______
Cap-and-trade, simplified, is the choice to "make or buy". The market will decide the price of
emissions, ensuring least expensive and most effective emissions reduction through markets.
Discuss the key differences between command and control

and cap-and-trade.
______
Under command and control, companies are fined for noncompliance of certain regulations. A
company must achieve emissions reduction targets on its own, not taking into account its
reduction cost (which may be very high).
Under cap-and-trade, emissions credits can be traded, so that the maximum of emissions
reduction is achieved at the lowest possible cost for all.
Example:

188
State the differences (2) between SO[sub]2[/sub] and
CO[sub]2[/sub] emission controls.
2 and CO2 emission controls.
______
1. The key difference between SO2 and CO2 is the residence time in the atmosphere: SO2
only stays in the atmosphere for 30-60 days, while GHGs and CO2 stay there for
decades. SO2 is therefore deposited closer to its source.
2. Technical options for reduction of SO2 were well known when legislation came into
force (using low-sulphur coal, install scrubbers). CO2 can use similar technologies,
but they are more involved and less developed.
Similarities were wide-spread skepticism at inception of the program (political issues) and
the use of markets to reduce emission reduction costs.

189
Discuss the history of the Kyoto Protocol.
______
The Kyoto Protocol grew out of the United Nations Framework Climate
Change Convention (UNFCC), adopted at the Rio Earth Summit in 1992.
UNFCC signatories met as a conference of the parties (COP) in Berlin in 1995.
Participating parties agreed to the Kyoto Protocol in 1997, and COPs continued to meet
regularly to hammer out details.
The "Kyoto Rulebook" (Marrakech Accord) laid out the terms of emissions trading in
2001. Useful lessons for the formation of the EU ETS were gained from U.S. markets in
SO2 and NOx, which operated since 1995.
The USA (2001) and Australia (2002) withdrew.
The Protocol came into force February 16, 2005.
The first commitment period runs from 2008-2012.
Highlight the history of the European Union Emission

Trading Scheme (EU ETS).
______
The ETS was the EU's plan to meets its Kyoto commitment (8% bubble target CO2 reduction
from 1990 levels in the period of 2008-2012). About 11,500 installations (industries, producers,
etc. > 20MW) among the then 25 member states were assigned carbon caps, within the national
allocation plan (NAP) of the country, which is the signatory of the Kyoto Protocol. A new
entrant reserve provided flexibility for new installations and companies.
The notion of trading emissions under a flexible mechanism was first propagated by the U.S.
After its withdrawal after a change of president (2001), U.S. ideas were further developed into
the EU ETS. Emissions trading under the ETS began in 2005.
There are three flexible mechanisms in the protocol:
1. International Emissions Trading (IET).

2. Clean Development Mechanism (CDM).
3. Joint Implementation (JI).
The principal operating entity in the protocol is the signatory country, who has to enforce
emissions reductions. Non-EU states have their own trading schemes to adhere to the Kyoto
targets, but they are nowhere close to the details of the EU ETS. It is important for these states
to be connected to the same markets as the EU in order to exploit market mechanisms and low
prices.

190
191
Discuss the history of carbon markets in Australia and the
United States. What is the difference to the EU ETS?
______
The U.S. and Australia have left the Kyoto Protocol in 2001 and 2002, respectively. They have
established their own Asia-Pacific partnership on Clean Development and Climate with India,
China, Japan, and South Korea to discuss voluntary approaches to emissions reduction relying
on technology, but without targets and caps.
The federal government in the U.S. (2006) and Australia do not support initiatives to cap
emissions, which is in stark contrast to state (most U.S. states have GHG reduction
programs, RPS) and private sector activism. Regional trading schemes have taken off in both
countries, without the support of the federal government.
Explain the importance of the Clean Development

Mechanism (CDM) and contrast it with Joint
Implementation (JI).
______
The CDM was designed to encourage Kyoto-capped countries to invest in new GHG
reduction projects in developing non-Kyoto countries and thereby earning CERs (certified
emissions reductions), which could be used against their own reduction targets. This should
bridge the gap between rich and poor nations.
Each project would need its own protocol for project design, operation and monitoring, which
needed to be approved by the Executive Board of the CDM. CERs would only be issued after
approval of the finished project. Sponsoring companies would still need to meet a certain
reduction target in their own country and could not simply outsource all reductions.

192
Joint Implementation (JI) can be seen as a CDM with fellow Kyoto-members. JI has been very
slow in implementation, and is still a cumbersome process. The JI Supervisory Committee was
established in 2006.
Example:

193
List the major carbon exchanges established at the opening of
the EU ETS and related contracts that are traded.
______
The main contract is the EU Allowance (EUA), the basis of the EU ETS.

194
Discuss the key issues associated with climate control efforts
(6).
______
1. Verification. Protocols for measuring emissions reductions. Predictable for the HFC
destruction process, but not biomass. Many countries were late in setting up verification
procedures and technologies, slowing CDM implementation.
2. Controlling sale of "hot air". The baseline year is 1990, which coincides with the fall of
communism. All formerly communist countries are far below their baseline targets, so
Kyoto designers fear downward pressure on emissions credits. The 1990 figure is
meaningless for those countries, so their baseline under the NAP was set close to their
current emissions targets.
3. Quality and price of carbon credits. Large spread between CERs and EUAs (€ 11-30),
largely due to uncertainty with CERs and associated transaction cost. Convergence is
expected, as both credits can be applied to the same goal.

195
4. Enforcing compliance. Penalty for noncompliance set at €40 per metric ton CO2
above the cap in phase one (2005-07) and €100 in phase two (2008-12).
5. Integrating various trading platforms.
6. CDM bottleneck. Lack of personnel, piecemeal approach. Number of new
protocols delaying projects in the pipeline.
List the key risks associated with climate change

control (1+6).
______
Performance risk associated with carbon finance can be broken down into:
1. Counterparty risk.
2. Carbon regulation risk.
3. Country investment risk.
4. Technology performance risk.
5. Business interruption.
6. Directors' and officers' liability.
Insuring carbon finance risk is difficult, as there is no market and each project is different.
Some risks are not insurable with standard instruments, such as catastrophes and weather.
Do emissions caps put a country at a

competitive disadvantage?
______
No, as BP, Alcan, Alcoa, DuPont, IBM and other major companies have shown. It is
possible to reduce emissions drastically and make a profit from energy savings.

1c6n4g0og 217796

Uploaded by

Copyright:

Available Formats

1c6n4g0og 217796

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1c6n4g0og 217796

Uploaded by

Copyright:

Available Formats

qwertyuiopasdfghjklzxcvbnmqwertyui

List the types of energy derivatives (table).

Describe a forward contract.

A forward contract is a bilateral agreement to purchase or sell a certain amount of a

This is your personal copy.

1. Hedge the obligation to deliver or purchase a commodity in the future.

Calculate the fair value of a forward contract.

The fair value of a forward contract is:

1. (one delivery date)

2. (several delivery dates, i.e. swap)

This is your personal copy.

= sum of discounted prices / sum of discount factors

= sum of discounted prices / sum of discount factors

Describe a futures contract. Which risk is inherent?

This is your personal copy.

From a futures contract, we immediately receive ΔF as a margin payment, which can be

This is your personal copy.

We assume U (fair value) = 0.

This is your personal copy.

As a consequence of put-call parity, an energy call option

A call option can be replicated by a put option and a forward contract:

F: current underlying forward price

This is your personal copy.

Describe Asian options, calculate their payoff (formula) and

Call payoff = max [avg(F(T)) - K, 0]

Advantages of Asian options are:

This is your personal copy.

Energy options are usually on futures contracts.

designates the expected value under risk-neutral conditions.

This is your personal copy.

Inner value of options describes their payoff if exercised before maturity.

Inner value call option =

Relationship between inner value and price of European options:

This is your personal copy.

1. Binomial/trinomial trees. Discrete time models with backwards iteration.

2. Finite-difference methods. Numerical approximations of the partial differential equation

Describe the main processing spreads (3) that market

This is your personal copy.

The upper value is calculated using Monte Carlo simulation:

For each path, a linear optimization problem must be solved.

Call option price =

Put option price =

This is your personal copy.

1. Market risk, price changes of financial assets and liabilities.

Energy options are usually written on ... .

Which is the largest exchange for trading energy futures and

Describe the main benefits (2) of marking-to-market futures

1. Reduction of default risk for derivatives.

This is your personal copy.

This is your personal copy.

1. Open: First price of the day.

This is your personal copy.

2. There is no clearinghouse involved, so default risk prevails.

Describe how energy futures and forwards differ (3).

2. Price transparency. Futures are more price transparent.

3. Hedging sensitivity. Different hedging sensitivity, as futures contracts are marked-to-

Calculate the profit/loss on a hedged position (formula).

Profit/loss = Gain/loss spot + gain/loss future = [ S(T) - S(t) ] + [ F(T) - S(t) ]

Weighted avg. value products = 0.5935.57 + 0.135 + 0.2635 + .0524.88 =