# Gold Futures Term Structure and Basis

Explain basis and calendar-spread exposure using matched contracts.

Use this workbook alongside the course. Write your answers before opening the solutions. Practical work is self-reviewed; scored knowledge checks are in the Academy.

## 1. Contract relationships

### Spot and futures prices

Spot and futures quotes must be matched by underlying reference, currency and timestamp before calculating a relationship. A stale spot quote against a fresh futures quote can create an apparent basis that is only a timing mismatch.

### Expiry-specific quotes

Different expiries are different contracts. Their prices can differ because delivery timing, financing, availability and market conditions differ. A sequence of expiry prices is a curve, not a forecast that every future spot price must equal today's contract price.

### Contango and backwardation

Contango commonly describes deferred prices above nearer prices under a stated comparison; backwardation describes the reverse. Specify which contracts are compared. A label alone does not determine a profitable spread trade.

### Basis conventions

Define basis explicitly, for example futures minus spot. Another report may use the opposite sign. Units and sign conventions must be aligned before comparing values or interpreting a widening relationship.

### Worked example

Spot is 100 and a matched futures quote is 103. Under futures-minus-spot convention, basis is +3. If another analyst defines spot-minus-futures, the same observations produce −3 without disagreement about the prices.

### Independent exercise

Calculate both conventions when spot is 102 and futures is 101. Explain why the convention belongs in the column heading.

My inputs and assumptions:

My calculation or decision:

Evidence that would change my conclusion:


## 2. Carry framework

### Financing

Financing is one component of carrying an asset across time. A simplified carry model relates current ownership costs and future delivery, but its assumptions must be stated rather than treated as a universally executable arbitrage.

### Storage and insurance

Storage and insurance affect the economics of holding physical material. Actual arrangements, location and eligibility matter. A generic percentage may not represent the costs accessible to a particular participant.

### Lease and convenience concepts

Lease-related and convenience concepts describe benefits or costs associated with access to the physical asset. They can vary with availability and market conditions. An unexplained residual in a simple formula should not automatically be assigned to one hidden variable.

### Limits of simplified carry models

Simplified models omit constraints such as credit, transaction costs, delivery eligibility and operational access. A theoretical price difference can persist when the required trades are not available on the assumed terms.

### Worked example

A worksheet estimates carry of 2 while observed matched basis is 3. The difference of 1 is not automatically a free profit; financing access, execution costs and delivery constraints may explain or consume it.

### Independent exercise

List four assumptions that must be checked before treating that difference as an executable opportunity.

My inputs and assumptions:

My calculation or decision:

Evidence that would change my conclusion:


## 3. Calendar spreads

### Leg sizing

A calendar spread combines different expiries. Define both leg quantities, multipliers and direction. Equal contract counts do not always mean equal economic exposure if specifications differ.

### Relative versus outright exposure

A spread reduces some common outright exposure but retains exposure to changes in the price relationship. It can lose even if both legs move in the same direction. Evaluate the net sensitivity rather than treating two legs as automatically neutral.

### Spread execution

Legging introduces execution risk when one side fills before the other. A quoted spread instrument may have different execution mechanics from two separate orders. Model the method actually used.

### Expiry and liquidity migration

Liquidity often migrates between expiries, and deadlines can force action. Include the roll or exit plan in the study. A profitable historical spread path is irrelevant if the assumed legs were not tradable at the modeled size and time.

### Worked example

Buy the near contract at 100 and sell the deferred at 103, equal multiplier one. Later they are 102 and 106. The near leg gains 2 and the deferred short loses 3: net loss is 1.

### Independent exercise

Repeat when the later prices are 102 and 104. Explain which relationship change benefited the position.

My inputs and assumptions:

My calculation or decision:

Evidence that would change my conclusion:


## 4. Research controls

### Continuous-series distortions

Continuous series join contracts using a roll rule and possibly adjustments. These transformations can create a useful research display while altering historical price levels or returns. Preserve the raw contracts for execution-sensitive analysis.

### Roll schedule

A roll schedule can be based on dates, volume or another criterion. It must be known at the relevant time. A retrospective rule using later information about liquidity can bias the series used in a strategy test.

### Data synchronization

Synchronize both legs and the spot reference where used. A relationship calculated from different observation times can reflect market movement between samples rather than a true contemporaneous spread.

### Stress widening basis

Stress widening basis and reduced liquidity, including the cost of closing both legs. Report assumptions that cannot be validated from the available data. A single smooth continuous chart is not enough to establish realistic spread execution.

### Worked example

A back-adjusted series removes a five-point contract gap visually. That adjustment does not mean an actual roll occurred at zero cost or that the historical absolute prices remained tradable.

### Independent exercise

Describe a reproducible dataset for a spread study, including raw contracts, roll decisions and synchronized observations.

My inputs and assumptions:

My calculation or decision:

Evidence that would change my conclusion:


## Course project

Build a two-expiry scenario sheet with carry assumptions and roll-cost sensitivity.

### Self-review rubric

- Concepts and reasoning: 25%
- Calculations, data and evidence: 30%
- Process and risk controls: 25%
- Limitations and communication: 20%

Record one correction and one next practice task. This rubric is not automatically graded.

## Worked solutions

### Exercise 1

Futures-minus-spot is −1 and spot-minus-futures is +1. Naming the convention prevents opposite signs from being mistaken for contradictory data.

### Exercise 2

Check matched instruments and dates, actual financing terms, storage/delivery eligibility and all transaction costs. Also assess credit and operational constraints. The residual is a research question until the trade can be specified realistically.

### Exercise 3

The near gains 2 and the deferred short loses 1, for net gain 1. The deferred-minus-near spread narrowed from 3 to 2, benefiting a long-near/short-deferred position under these assumptions.

### Exercise 4

Store expiry-specific quotes, timestamps, units and liquidity fields, plus the rule and decision date for each roll. Keep transformed series separate and document costs and missing observations.

## Further reading

- https://www.cmegroup.com/education/courses/introduction-to-futures
