3. Risk Management Principles
Teaching rewrite of risk management for commodity hedgers and traders. Original wording; worked numbers recomputed. Prices are teaching values. Not investment advice.
Introduction
“Risk management” means different things to a mine, an airline, and a bank’s trading desk. This lesson builds one vocabulary for all three and then follows the risk from a corporate hedger to the bank that takes the other side — and on to the traders who manage it.
The big lesson from history: the most catastrophic losses rarely come from price moves alone. They come from not understanding the business and weak controls — risks no derivative can hedge.
This lesson covers:
- The five risk types and market-risk subcategories
- Who trades at the short, medium, and long end of the curve
- Corporate hedging choices and a strategic → tactical framework
- How banks hedge forwards, swaps, options, and correlation — plus basis risk
- Trading: spot vs forward breakeven, spreads, and volatility trading with delta hedging
- Case studies: Japan Airlines, Amaranth, Metallgesellschaft
3.1 Defining risk
| Risk | Meaning | Hedgeable with derivatives? |
|---|---|---|
| Market | Value of what you own or owe changes with prices | ✅ |
| Credit | Money owed is never repaid | ✅ (credit derivatives, collateral) |
| Business | Engaging in a business you don’t fully understand | ❌ |
| Operational | Losses from weak controls, processes, people, systems | ❌ |
| Legal / documentary | Contract void (e.g. ultra vires) or a contractual commitment causes loss | ❌ |
The irony: the risks most often behind big losses — business and operational — are the ones you can’t buy protection against. They come from culture and discipline. When a loss is “blamed on the market”, ask why nobody independent stopped it sooner.
Market risk subcategories
| Subcategory | Example |
|---|---|
| Interest rate | Floating-rate borrowing costs rise — and interest earned on cash surpluses can fall |
| Inflation | Inflation-linked pension promises or costs |
| FX | Transactional (a foreign-currency payable) and translational (restating foreign assets). Commodities are priced in USD, so non-US producers and consumers carry FX risk |
| Equity | Defined-benefit pension assets, share buybacks, employee share options, stakes in listed companies |
| Liquidity | Can’t meet short-term cash needs — no credit line, or can’t sell assets |
| Commodity | Direct (a gold miner) or secondary (a haulier’s diesel). Integrated supply chains may have offsetting exposures |
Credit risk in physical hedging
An automaker hedging metal directly with a mine or smelter faces:
- Specification — miners produce intermediate product, not the exact grade/shape needed
- Counterparty — operational problems may stop delivery
- Opposite timing — consumers want to hedge at cycle lows, producers at highs
Fixed-price component supply shifts the problem to the component maker: hedge with poor credit (bad terms) or don’t hedge (exposed to rising metal).
Legal risk: force majeure
A force majeure (“Act of God”) clause excuses performance when specified events beyond a party’s control prevent it. In 2020, some Chinese LNG and copper buyers declared force majeure as demand collapsed during COVID-19 — sellers complained some were really trying to swap long-term cargoes for cheaper spot ones. Drafting and interpreting these clauses is a real risk.
Checkpoint 1: risk types
Checkpoint — section 3.1
1 / 5A trader hides losses for months because nobody reconciles his positions independently. Which risk failed?
Pick an answer to see if you are right.
3.2 Who trades where on the curve
| Maturity | Main participants | Activity |
|---|---|---|
| Short-dated | Banks and trading houses; index investors; hedge funds and CTAs | Managing trading books, tracking indices, short-term views |
| Medium-dated | Producers and consumers | Strategic hedging: refinery margins, airline jet fuel |
| Long-dated | Producers selling forward (often required by lenders); consumers buying into backwardation; investors in structured notes | Stabilising cash flows; buying cheap long-dated prices; notes often built as a zero-coupon bond + call option |
3.3 Corporate hedging strategies
Using an automaker buying metal as the example:
| Strategy | When it makes sense | Trade-off |
|---|---|---|
| Don’t hedge | Metal is a small share of cost; price rises can be passed on | Price falls may have to be passed on too |
| Lock in unit costs | Fix the cost of a production run and the selling price | Customers get stable prices; no benefit if metal falls |
| Hedge the timing mismatch | Long processes, irregular deliveries, large stocks — product price linked to commodity | Protects stock bought before sale |
| Active (view-driven) hedging | Firm knows it will always need the metal and acts on favourable prices | Requires market views — many firms are uneasy |
How much? Hedges on forecast purchases may outlive the exposure if production falls — so many firms hedge only a proportion. Hedges are often cash-settled, so the company can still buy the exact grade or shape from its preferred supplier.
3.4 A framework for corporate risk
Hedging should serve the company’s top-level goal — share price, profit, or cash flow — not just a single price.
flowchart TD
S[Strategic<br/>goals, stakeholders, risk universe,<br/>performance metric, current policy, competitors] --> T[Tactical<br/>exposure certainty, timing, % covered,<br/>target price, market view, allowed instruments, premium appetite]
T --> SC[Scenario analysis<br/>what if prices rise, fall, swing?]
SC --> S
Strategic questions: What are our goals? What matters to shareholders, lenders, staff? Which financial risks do we face, and which can be hedged? Against which metric do we judge success? What do we do today? What do competitors do?
Tactical questions:
- How certain is the exposure? When should the hedge start?
- What percentage to cover? Is there a budget/target price?
- What is our view on direction, timing, and magnitude?
- Which instruments are allowed? Will we pay premium?
Option structures are three-dimensional — they draw value from direction, time, and volatility. A client who can express even a modest view on these lets a bank build a more efficient hedge. Scenario analysis shows how each hedge affects the key metric and loops back to the strategic level.
Checkpoint 2: hedging strategy
Checkpoint — sections 3.2 to 3.4
1 / 5Who typically dominates medium-dated maturities on a forward curve?
Pick an answer to see if you are right.
3.5 How banks hedge client exposures
Derivatives transform risk: the bank takes on what the client gives up and lays it off. Its trading book mixes client-driven positions and discretionary views.
| Product | Bank’s risks | Typical hedge |
|---|---|---|
| Futures (client on exchange) | None for the bank — the clearing house is the counterparty and collects margin | — |
| Forwards | Market risk (spot + carry components) and credit risk | Equal and opposite position; hedge carry components |
| Swaps | Market and credit risk | Offsetting swap; or a proxy (e.g. jet fuel swap hedged with crude futures) → basis risk |
| Options | Underlying price, strike, time, carry, implied volatility | Manage each with its Greek (lesson 2) |
| Correlation | Spread and basket options; related commodities (platinum/palladium); commodities vs financial assets | Historical or implied correlation — unstable in volatile markets |
Basis risk: the hedge doesn’t move one-for-one with the exposure. A jet-fuel swap hedged with crude futures loses if the jet–crude spread (the “crack”) moves against you.
Case study: Japan Airlines (2010)
JAL collapsed with debts of about ¥2.3 trillion; outstanding fuel hedges were estimated at about USD 440 million — many struck near the 2008 price peaks.
flowchart LR
S[Jet fuel supplier] -- physical jet fuel at market index --> A[Airline]
A -- market index price --> S
B[Bank] -- pays market index --> A
A -- pays fixed USD 800/t --> B
B -- long crude futures hedge --> X[Exchange]
The airline pays fixed and receives floating, locking its fuel cost. The bank pays floating and receives fixed, so it hedges by going long crude futures (basis risk: crude ≠ jet).
When jet fuel fell below USD 800/t, the swap was in the money for the bank — the airline owed it money. On default:
- ITM for the bank → the bank becomes an unsecured creditor for the mark-to-market.
- OTM for the bank → the bank must pay the bankrupt airline what it owes; no credit loss.
- Either way, the bank’s hedge is now orphaned — it must sell its crude futures, which pushed prices down in the short term.
Worked example: credit exposure on a jet-fuel swap
Bank receives fixed USD 800/t on 60,000 t remaining; jet fuel now USD 650/t (ignore discounting).
Mark-to-market for the bank = (800 − 650) × 60,000 = USD 9,000,000 → the bank’s unsecured claim if the airline defaults. If jet fuel were USD 900/t, the bank would owe USD 6,000,000 and have no credit exposure.
Checkpoint 3: bank risk management
Checkpoint — section 3.5
1 / 5A client buys an exchange-traded future through a bank. Who bears the client’s default risk?
Pick an answer to see if you are right.
3.6 Trading risk management
Trading = buying and selling with profit as the main motive — view-driven, not driven by an operational need.
Spot trading: the forward is the breakeven
Gold spot USD 1,425.40, 6-month forward ≈ USD 1,449 (lesson 2). A spot purchase financed and carried for six months costs the forward price. So the decision depends on where you think spot will be relative to the forward, not just up or down:
| Your view of spot in 6 months | Action |
|---|---|
| Above 1,449 | Buy spot, borrow cash, lend gold to earn the lease |
| Below 1,449 | Sell spot, deposit cash, borrow gold to cover |
| Equal to 1,449 | Indifferent |
You can expect gold to rise to 1,440 and still sell — because 1,440 < 1,449.
Forward and relative-value trades
- Curve slope: expect steepening → sell short-dated, buy long-dated.
- Curve curvature: expect more concave → sell short and long, buy the middle (a “butterfly”).
- Seasonal relationships: e.g. winter vs spring gas (see Amaranth).
- Inter-commodity spreads:
| Spread | Legs |
|---|---|
| Crack | Gasoline or heating oil vs crude |
| Spark | Power vs natural gas |
| Base metals | Copper vs aluminium |
| Precious metals | Gold vs silver, platinum vs palladium |
Worked example — crack spread: gasoline USD 2.40/gal × 42 gal/bbl = USD 100.80/bbl; crude USD 78/bbl → crack = USD 22.80/bbl.
Also: single-period physically settled swaps (spot + forward together — a temporary exchange of metal for cash, to borrow cheaply or source metal temporarily) and cash-settled swaps (single-period = cash-settled forward).
Trading volatility
Implied volatility is a market guess — so traders “buy” and “sell” it like a commodity, while neutralising direction:
| View | Strategies |
|---|---|
| Sell volatility (expect vol to fall / market to be calm) | Short ATM straddle (sell call + put); sell a call and delta-hedge; sell a put and delta-hedge |
| Buy volatility (expect vol to rise / market to swing) | Long ATM straddle; buy a call or put and delta-hedge |
Worked example: selling a delta-hedged gold call
Bank sells a 3-month call on 10,000 oz, strike 1,450, spot 1,425.40, vol 15% → premium USD 31.66/oz, delta −0.4223 → buys 4,223 oz to hedge.
| Step | Option P&L | Hedge P&L | Net |
|---|---|---|---|
| Gold +5 to 1,430.40; option value 31.66 → 33.82 | −2.16 × 10,000 = −21,600 | +5 × 4,223 = +21,115 | −485 |
| Re-hedge: delta now −0.44 → buy 177 oz more at 1,430.40 | |||
| Gold falls back to 1,425.40 | +21,600 | −5 × 4,400 = −22,000 | −400 |
| Total for the day | 0 | −885 (= 177 × 5) |
The loss came from re-hedging: a short-option trader buys high and sells low as delta shifts. That’s gamma — here (0.44 − 0.4223) = 1.77% delta change for a USD 5 move. Short options = short gamma = hurt by realised volatility.
The compensation is theta: initial theta USD 0.2266/oz/day × 10,000 = USD 2,266/day — more than the USD 885 re-hedging loss. Traders call theta “gamma rent”:
| Buyer of volatility | Seller of volatility | |
|---|---|---|
| Hopes for | Implied vol ↑ and/or big realised moves | Implied vol ↓ and/or calm markets |
| Profits if | Delta-hedging gains > time decay | Time decay > delta-hedging losses |
Daily volatility rule of thumb
Daily vol ≈ annual vol ÷ √250. 15% → ≈ 0.95%/day. With gold at 1,425.40, a one-day move of about ±13.54 → expected range ≈ 1,411.86 – 1,438.94.
Case study: Amaranth Advisors (2006)
- Hedge fund of about USD 9.2 bn; lost about USD 6 bn in two to three weeks on US natural gas spreads.
- Gas forward curves are humped: winter > summer.
- View: winter prices would rise relative to summer (demand growth, shortages, bottlenecks, weather). Trades included short Nov-06 / long Jan-07 and long Mar-07 / short Apr-07 — the volatile “widow maker” (March = end of withdrawal season; April = start of injection).
- Positions reached ~100,000 contracts (≈ 1 trillion cubic feet, ~5% of annual US use) and at times ~40% of NYMEX open interest in winter months.
- Late August 2006: quiet hurricane season and full storage → the Mar/Apr spread collapsed from USD 1.50–2.50 to about USD 0.50/MMBtu. Margin calls reached ~USD 3 bn; the fund had to sell its book.
Lessons: concentration and liquidity risk (you can’t exit when you are the market), spread trades are not low-risk, and margin turns paper losses into cash crises.
Case study: Metallgesellschaft (1993)
- US arm MGRM sold fixed-price oil products to service stations for 5–10 years — about 160 million barrels of commitments.
- Hedged with short-dated (~3-month) long futures, rolled every month — a maturity mismatch (a “stack-and-roll” hedge).
- In backwardation, rolling earned money (sell expiring dearer futures, buy cheaper next ones).
- Late 1993: OPEC didn’t cut, crude fell from about USD 22 to under 14.50, the curve flipped to contango:
- Margin calls of about USD 200 m on long futures — cash out today, while gains on fixed-price contracts were years away.
- Roll costs of about USD 30 m per roll.
- Market participants who knew MG had to roll traded against it.
- Losses ≈ USD 1.5 bn.
Lesson: a hedge that is economically sound can still fail on cash flow (funding liquidity), curve shape (roll yield), and predictability (others front-running your rolls).
Checkpoint 4: trading and case studies
Checkpoint — section 3.6
1 / 6Gold spot 1,425, 6-month forward 1,449. You expect gold at 1,440 in six months. Best trade?
Pick an answer to see if you are right.
Rank: risk management practice
Risk management practice
1 / 2Rank these hedges of a jet-fuel exposure by basis risk (LOWEST first).
OrderDrag the handle to rearrange — top should be first.
- 1Crude oil futures
- 2Heating oil futures (a close distillate)
- 3Natural gas futures
- 4Jet fuel swap referencing the same index as the supply contract
Drag items into the correct order (or use ↑↓), then check.
Check: risk management principles
Risk management principles
1 / 6Which pair of risks most often causes catastrophic losses, yet can’t be hedged with derivatives?
Pick an answer to see if you are right.
Calculate: numeric practice
Type the number (no units; minus sign for losses).
Formula sheet
| Item | Formula |
|---|---|
| Delta hedge size | notional × delta |
| Re-hedge loss (round trip) | extra units bought × price move |
| Daily volatility | annual vol ÷ √250 |
| Swap MTM (fixed receiver) | (fixed − market) × remaining volume |
| Spread trade P&L | change in spread × contract size × contracts |
| Futures roll cost (contango, long) | (next price − expiring price) × volume |
| Crack spread | product price per bbl − crude price |
Calculate: risk management
1 / 12You sold calls on 25,000 oz with delta 0.38. How many oz do you buy to delta-hedge?
Enter a number (at most 0 decimal places).
Randomized practice
Practice: hedging and volatility drills
1 / 6Long 2,000 gas calendar spreads (10,000 MMBtu each). Spread moves from 1.95 to 2.05. P&L in USD?
Enter a number (at most 0 decimal places).
Takeaways
- Five risks: market, credit, business, operational, legal — the last three can’t be hedged with derivatives
- Market risk includes interest rate, inflation, FX, equity, liquidity, commodity
- Short end = traders and investors; middle = strategic hedgers; long end = producers, lenders, structured notes
- Hedge choices: don’t hedge, lock costs, fix timing mismatch, active hedging — usually a proportion, often cash-settled
- Go strategic → tactical → scenarios, judged against the company’s key metric
- Banks lay off risk; proxies create basis risk; credit risk exists only when in the money
- The forward is the breakeven for spot trades; spreads trade relationships, not levels
- Short options: earn theta, pay gamma through re-hedging — and vice versa for buyers
- Amaranth and Metallgesellschaft: concentration, liquidity, margin, and roll can sink a “sound” position
Further Reading
- 2. Derivative valuation — forwards, Greeks, and volatility
- 1. Fundamentals of commodities and derivatives
- Export–Import: international payments — FX risk in trade
- Next: 4. Gold
- US Senate Permanent Subcommittee on Investigations, Excessive Speculation in the Natural Gas Market (2007)
- Background: N. Schofield, Commodity Derivatives: Markets and Applications (Wiley) — used as a study outline; J. Gregory, The xVA Challenge (derivative credit risk)