DERIVATIVES AND RISK MANAGEMENT CASE STUDY DUE ON Tuesday 9 June 2020 at 17h00 Learning Objectives By completing this case you will: – Learn to work with futures and spot data – Estimate hedge ratios – Compute hedged and unhedged valuation changes with historical data – Back-test and discuss possible hedging strategies – Consider how hedge ratio estimates can be employed in dynamic settings – Communicate complex analysis to a non-technical audience Learning Activities By completing this case you will: – Estimate and/or compute hedge ratios – Compute cash flows to hedging strategies using energy futures contracts – Comment on the consequences of changing market conditions on possible hedging strategies – Write a brief report for a non-technical audience Task You are an associate at a commercial bank. One of your colleagues has sent the following email: “Thanks again for sending Hull’s chapter on hedging with futures. As you know we do a lot of work with energy-intensive companies, and one of his examples seemed especially relevant. His cross-hedging strategy (jet fuel and heating oil) is interesting, but we were wondering what would happen in turbulent markets? For example, in October 2018 crude oil reached $76.50 to drop down to $44 by December 2018 and more recently it crashed and fall to a negative territory (-$37.63) on 20 April 2020. What would happen to someone using Hull’s proposed strategy over this time? Would it make sense to consider other energy futures as part of a hedging strategy (i.e. we have been wondering about including crude oil futures in addition to heating oil)? Any thoughts you have would be greatly appreciated.” / 2 Your colleague has limited quantitative skills and you feel that you can best demonstrate these concepts through a clear example, based on data taken from the period she noted. You consider what might happen over this time to someone who wanted to hedge the price risk of 5,000,000 gallons of jet fuel. You note that the futures data is taken from NYMEX where heating oil contracts are quoted in USD per gallon (contract size of 42,000 gallons) and crude oil is quoted in USD per barrel (contract size of 1,000 barrels). Your assistant has downloaded spot jet fuel, crude oil futures and heating oil futures daily prices from June 01, 2016 through April 21, 2020 (see spreadsheet “Combined” in the file “Daily Energy Data – 25762 DRM – AUT 2020.xlsx”). You double check that their use of the “lookup” function in Excel was done properly so that the combined data is correctly lined up through time. Your aim is to assess two factors of hedging jet fuel price changes with futures contracts; a) the stability of the estimated hedge ratios over time and b) the role of different hedging instruments in the performance of the hedge. To draw your conclusions, you perform the following investigations: i) You compute and compare the hedge ratios and the optimal number of futures contracts by using two data samples; June 2016-June 2018 and June 2018-April 2020. For the hedging strategy, you consider three type of hedging applications in terms of hedging instruments: hedge with heating oil futures only, hedge with crude oil futures only, and, hedge with both heating oil and crude oil futures. How stable the hedge ratios have been over the time? ii) You assess the out-of-sample hedge performance of these hedge ratios. By using the hedge ratio estimated from the June 2016-June 2018 dataset, you hedge jet fuel price changes in two periods; June 2018-June 2019 and June 2019-April 2020 and then compare the hedging performance over these two periods (by comparing standard deviations of unhedged and hedged positions). iii) What are the main issues you have identified with this hedging application and what are your suggestions to deal with those? iv) First time in the history, oil markets experienced negative prices. Explain what happened to the crude oil spot and futures markets on 20 April 2020. Explain why not all oil prices (e.g. Brent, WTI) went to negative territories. News announcements and online articles should be used to support your discussion. / 3 Submission Details Please submit both your team’s report AND the spreadsheet you used for your quantitative analysis. Both documents will be reviewed in assessing your work. However, please write your report so that the reader does not have to refer to the spreadsheet to understand your key points. Please submit your team’s cover page for the submission, the written report and the spreadsheet you used for your quantitative analysis by an online submission under Case Study on Canvas by Tuesday 9 June 2020 17h00. The file names of these documents should include your team surnames / number. Late submissions attract a penalty of 5 marks per day. Your submission cannot be more than 6 pages in total (including cover sheet) with at least one-inch margins and double-spaced 12-point type. You are welcome to discuss your basic approach with current DRM students but all the analysis and the writing up must be from your team. Note that reports will be checked for plagiarism via Turnitin. Canvas will not accept submissions after the due date and time. Please be sure to cite the work of others where appropriate. The quality of your writing is as important as the technical accuracy of your report. / 4 25762 DERIVATIVES AND RISK MANAGEMENT Case Study GRADE BREAKDOWN

Item

Points

Computations/Spreadsheet

Compute price changes across relevant time periods

/2

Compute hedge ratios and optimal number of contracts (with each futures singly across periods)

/4

Compute hedge ratios and optimal number of contracts (with both futures across periods)

/2

Compute hedging outcome / st. dev. of cash flows between scenarios

/4

Compute hedging outcome / st. dev. of cash flows with both futures

/2

Write-up

Executive summary (to start the report)

/1

Describing the basic approach to your analysis

/1

Discuss choice of futures contracts for the hedge

/1

Discuss and explain hedging performance under different scenarios

/2

Discuss issues (see factors a) and b) in case study) and how to deal with

/3

Explain negative oil prices in Apr 2020

/3

BONUS: Report lodged as requested

/1

Total

/25

/ 5 Technical Note: Hedge Ratios and Why Hull’s Formulae are Used This section is designed to give you a bit more detail on where the formulae in Hull are used for constructing cross hedges. It is not part of the assignment, rather designed to give you a bit more insight into how these procedures are meant to work. Using the jet fuel example an airline would face the uncertainty of the spot jet fuel price at some future date; we denote this value as ST. They use an associated futures contract to hedge (there is no actively traded jet fuel futures available). Let h be the number of gallons of heating oil gone long through a futures contract per gallon of jet fuel to be purchased. In this case the net cost of a gallon of jet fuel would be (F0 and FT are the per gallon heating oil futures prices at the start and end of the hedging period, respectively): C S h F F T T T = – – ( ) 0 The variance of this net cost is: ( ) ( ) ( ) ( ) ( ) 0 2 ( ) 2 , T T T T T T T Var C Var S h F F Var S h Var F hCov S F = – – = + – Taking the derivative (with respect to h) to minimize the variance of the cost of acquiring the jet fuel, while longing futures, we have: ( ) Var CT 2 2 , 0 hVar F Cov S F ( T T T ) ( ) dh = – = Which yields: ( ) ( ) * , T T T T S T F Cov S F h Var F = = This matches Hull equation (3.1). When we adjust for the size of the position to be hedged (in gallons) and the size of a futures contract (in gallons) we have: * * A F Q N h Q = This matches Hull’s expression (3.2). / 6 Hull also notes the following (page 82): “The parameters , F, and S in equation (3.1) are usually estimated from historical data on S and F. (The implicit assumption is that the future will in some sense be like the past.) A number of equal nonoverlapping time intervals are chosen, and the values for each of S and F are observed. Ideally, the length of the time intervals is the same as the length of the time interval for which the hedge is in effect. In practice, this sometime severely limits the number of observations that are available, and a shorter time interval is used.” When estimating the variances and covariances (or regression coefficients) we often use price changes taken with frequencies much shorter than the hedging horizon (e.g. daily data used for estimates when the hedge might be planned for a year). Consider computing a variance over T days. From the equations above we would be interested in Var S ( T ). We can write this as follows: