Skip to main content

Futures and Forwards

  • Chapter
  • First Online:
Capital Market Finance

Part of the book series: Springer Texts in Business and Economics ((STBE))

  • 1417 Accesses

Abstract

A forward or a futures contract fixes today the terms of a transaction to be carried out on a future date. In organized exchanges, futures contracts are standardized in their amounts, maturities, and the quality of the underlying asset. They are subject to daily margin calls and cleared by official clearing houses. As a result, they are practically void of counterparty risk. By contrast, forward contracts are traded over the counter, tailor-made, and flexible as the parties design the terms of the contract at their own convenience, but are in general less liquid and fraught by counterparty risk. After a general presentation of futures and forwards, we examine the relationship between the forward/futures price and the spot price (cash and carry). We then study hedging with these instruments and finally present the main contracts according to their underlying assets.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

eBook
USD 16.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Hardcover Book
USD 119.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    In practice, the delivery date is several days after the contract’s maturity date. The latter is defined as the last date when the contract may be traded (after the maturity date one may neither buy nor sell the contract). In the following analyses, these two dates will be treated as the same for simplicity.

  2. 2.

    For the sake of simplicity, we do not distinguish in our notations here the forward and the futures prices (both denoted by F), which, in general, are in fact slightly different (see Proposition 1).

  3. 3.

    However, as we will see in the sequel, on organized markets a guarantee (deposit) is required from both the buyer and the seller.

  4. 4.

    Note, however, that the parties tend increasingly to proceed to margin calls, as in the case of futures (see Sect. 9.1.2.2), to decrease the counterparty risk for each of them. The frequency of these margin call payments is determined over-the-counter at the origination of the forward contract. The financial crisis of 2007–2008 has amplified this tendency.

  5. 5.

    Typically, it ranges from 1% to 2%, and is always less than or equal to 5%.

  6. 6.

    There are discrepancies between forward and futures prices, which can be ascribed to differences in counterparty risk (which is virtually inexistent for futures), in liquidity (often much higher for futures), in transaction costs (often smaller for futures), in taxes, etc.

  7. 7.

    On organized markets, where most futures transactions are handled, this unwinding amounts to a pure and simple cancellation of the agent’s position by the clearinghouse.

  8. 8.

    Rigorously speaking, this result holds exactly only in the evening after the margin call, or for continuous time under the hypothesis of continuous margin calls.

  9. 9.

    This only applies to futures in a world with deterministic rates.

  10. 10.

    However, these transactions are impossible for some commodities, as we will see in Sect. 9.4 and 9.4.1.

  11. 11.

    One can also remark that (9.3-b) implies Ft = St /BT(t) + D BT’(t) /BT(t) = St /BT(t) + D(1 + fT-T’(t))T-T’ where fT-T’(t) is the forward rate in force at t for the future period (T’, T); D(1 + fT-T’(t))T-T’ is thus the yield R capitalized at T.

  12. 12.

    If n(t) denotes the number of shares in the portfolio at instant t, the dividends received and immediately reinvested in the interval (t, t + dt) amount to nr*St dt which allows acquiring dn = r*ndt new shares so that dn/n = r*dt, and n(t) = n(0) er*t.

  13. 13.

    This assertion, although less useful in the (numerous) cases where the cash-and-carry relation holds, remains true since the equilibrium price S0 does itself depend on market expectations.

  14. 14.

    Such a situation does occur for agricultural products awaiting harvest, and in some other cases because of constraints resulting from regulations, customers or senior management, or of transaction or information costs.

  15. 15.

    See, for example, the book by A. Lioui and P. Poncet: Dynamic Asset Allocation with Forwards and Futures, Springer, 2005.

  16. 16.

    It would indeed be necessary that ρCFσC equals σF, which would happen for example if the prices of C and F were perfectly correlated and had identical variances: if, in particular, the contract’s underlying is the same as the asset to be hedged, the hedge is perfect and it suffices to sell forward the existing asset with maturity H to be sure to receive on that date the price F0.

  17. 17.

    BH, equal to the carrying cost for the period (H, T), therefore depends on the short-term rate rT–H(H) which is usually imperfectly correlated (or even uncorrelated) with the price CH of the asset to be hedged.

  18. 18.

    The agent uses the mean/variance criterion whose foundations are discussed in Chap. 21.

  19. 19.

    The attentive reader will notice that here we have in fact a variable used to adjust the cost of carry so as to make it equal to the observed basis, and that we are trying to rationalize it ex post in an ad hoc way. There are, however, models that attempt, with more or less success, to forecast it ex ante on the basis of the behavior of supply of and demand for the commodity. These models are outside the scope of this book.

  20. 20.

    Those are inventories essentially held for transaction and/or precautionary motives. In fact, there is often also a component, even if a small one, of the inventory which can be attributed to the expectation of an increase in the price of the asset held.

  21. 21.

    Brokers are compensated by fees.

  22. 22.

    The CME Group was born as a result of the takeover of the CBOT (Chicago Board of Trade) by the CME (Chicago Mercantile Exchange). It combines four large markets: the CME, the CBOT, the NYMEX and the COMEX and offers a large variety of futures contracts and options. It is the largest market of its kind in the world.

  23. 23.

    The NYSE-EURONEXT-LIFFE has been renamed ICE Futures Europe after several mergers and acquisitions that left it under the ownership of the ICE (Intercontinental Exchange).

  24. 24.

    An interesting case is the official French system SRD (Service de Règlement Différé) created in 2000 where traders can take forward positions on some listed French stocks obeying minimum requirements.

  25. 25.

    The deposit requirements are as follows: (i) 20% in cash, Treasury Bonds or monetary mutual funds; (ii) 25% in listed bonds, negotiable debt securities or bond mutual funds; and (iii) 40% in listed stocks or equity mutual funds.

  26. 26.

    When the rates are set up on an actual or 365 basis, these durations are calculated on an actual/actual or actual/365 basis.

  27. 27.

    Initiated on date 0 and not on T; this is therefore a forward swap (see Chap. 7, Sect. 2 and 4.3.5).

  28. 28.

    In certain markets (outside the Euro Zone), the rate rD(T) used could be a discounted one and not post-counted as here, and the same for the forward rate. The cash flow would then be ± M (rD(T) – f) D.

  29. 29.

    The rate r* depends on the period (0, T): indeed dividend distributions are often concentrated at certain times of the year (May and June, for example).

  30. 30.

    This result is proved in Chap. 19; it is a simple consequence of the fact that taking a position in a futures contract which requires no disbursement of funds on date t leads to a margin dF on t + dt: as a result, EtQdF = 0, otherwise the expected return would be +/− ∞, which is incompatible with the risk-neutrality that is supposed to hold under Q. Ft is therefore a Q-martingale.

Suggestions for Further Reading

Books

  • Hull, J. (2018). Options, futures and other derivatives (10th ed.). Prentice Hall Pearson Education.

    Google Scholar 

  • Kolb, R. (2007). Futures, options and swaps (5th ed.). Blackwell.

    Google Scholar 

  • *Lioui, A., & Poncet, P. (2005). Dynamic asset allocation with forwards and futures. Springer.

    Google Scholar 

Articles

  • Black, F. (1976). The pricing of commodity contracts. Journal of Financial Economics, 3, 167–179.

    Article  Google Scholar 

  • Cox, J., Ingersoll, J., & Ross, S. (1981). The relation between forward and futures prices. Journal of Financial Economics, 9, 321–346.

    Article  Google Scholar 

  • Jarrow, R. A., & Oldfield, G. S. (1981). Forward contracts and futures contracts. Journal of Financial Economics, 9, 373–382.

    Article  Google Scholar 

  • *Richard, S., & Sundaresan, M. (1981). A continuous time model of forward and futures prices in a multi-good economy. Journal of Financial Economics, 9, 347–372.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Patrice Poncet .

Appendix

Appendix

1.1 The Relationship Between Forward and Futures Prices

The prices of forward and futures contracts written on the same underlying and with the same maturity T do, in general, differ slightly. However:

Proposition 1

Under AAO and with deterministic interest rates, the futures price equals the forward price.

Proof

Let us use the notation:

  • St: the spot price on t of an asset x .

  • Ft: the price at t of a futures contract with maturity T written on x (FT = ST) .

  • Φt: the price at t of a forward contract on x with the same maturity (ΦT = ST).

  • rT–t(t): the continuous rate prevailing at the instant t for transactions with duration T–t and assumed to be deterministic.

  • Consider the period (0, T), just T days long, the jth day being bracketed by the instants (j–1, j) for j = 1,…, T.

First, let us consider strategy A consisting in holding during day j (from j–1 to j) n(j–1) futures contracts with maturity T, with

$$ n\left(j-1\right)=e{{{}^{-\left(T-j\right)r}}_{T-j}}^{(j)}. $$

This strategy provides each day j a margin = n(j–1)[FjFj–1] which one loans or borrows between the dates j and T. The sum of these margins capitalized at T can be written

\( \sum \limits_{j=1}^T{e}^{\left(T-j\right){r}_{T-j}(j)}n\left(j-1\right)\left[{F}_j-{F}_{j-1}\right]=\sum \limits_{j=1}^T\left[{F}_j-{F}_{j-1}\right]={F}_T-{F}_0={S}_T-{F}_0. \)

Remark

  • Strategy A only involves a single cash flow (random, taking place at T and equal to STF0) .

  • Strategy A is only possible if rates are deterministic since in the contrary case n(j–1) cannot be determined at time j–1 as a function of the rate rT–j(j) which will only become known at time j.

Now, let us consider strategy B consisting of buying on date 0 a forward contract at the price Φ0 and held to maturity; B only leads to a single (random) cash flow on date T, which writes:

$$ {\Phi}_T-{\Phi}_0={S}_T-{\Phi}_0. $$

If Φ0 > F0 the strategy “A–B” consisting in adopting strategy A and selling a forward contract would generate a single, certain, cash flow on date T equal to

ST – F0 – [ST – Φ0] = Φ0 – F0 > 0.

This would therefore constitute an arbitrage. Strategy “B–A” would be an arbitrage if F0 > Φ0. Consequently, at any instant denoted by 0, under AAO the equality F0 = Φ0 must hold.

Proposition 2

When interest rates are stochastic, the futures price exceeds the forward price if the interest rates are positively correlated with the underlying asset price, and falls below in the reverse case.

This result has an intuitive explanation. If rate and price are positively correlated, for a price hike the positive margin received by the holder of the futures contract will be invested at a rate presumably increasing. In the reverse case the negative margin will be borrowed at a falling rate. In this way, for a buyer of futures, the positive effect of an appreciation is amplified and the negative effect of a depreciation is damped: the existence of margin calls benefits the buyer, which explains why the futures price is (slightly) higher than the forward price if the interest rate and the underlying asset price are positively correlated. The opposite is true if rate and price are negatively correlated.

*This intuitive result can be proved with a continuous-time stochastic model. Consider the risk-neutral probability Q (different from the true probability P) under which the futures price Ft is a martingale.Footnote 30 Since FT = ST, we obtain

$$ {F}_t={E_t}^Q\ \left[{S}_T\right] $$
(9.A-1)

where EtQ is the conditional expectation under Q conditioned on the information available at t.

Applying Itô’s Lemma to the forward price, that is to the ratio \( {\Phi}_t\left(S,B\right)=\frac{S_t}{B_T(t)} \), where BT(t) denotes the price of the maturity-T zero-coupon bond at t, we have

$$ \frac{d{\Phi}_t}{\Phi_t}=\frac{d{S}_t}{S_t}-\frac{d{B}_T(t)}{B_T(t)}-\mathit{\operatorname{cov}}\left(\frac{d{S}_t}{S_t},\frac{d{B}_T(t)}{B_T(t)}\right)+\mathit{\operatorname{var}}\left(\frac{d{B}_T(t)}{B_T(t)}\right) $$
$$ \frac{d{S}_t}{S_t}-\frac{d{B}_T(t)}{B_T(t)}-\mathit{\operatorname{cov}}\left(\frac{d{\Phi}_t}{\Phi_t},\frac{d{B}_T(t)}{B_T(t)}\right). $$

And since, under Q, the derivatives of δS/S and δB/B are both equal to the risk-free rate r(t), the derivative of \( \frac{d{\Phi}_t}{\Phi_t} \)equals – cov (\( \frac{d{\Phi}_t}{\Phi_t},\frac{d{B}_T(t)}{B_T(t)}\Big) \)) which is denoted by –σΦB(t)δt. Using the notation σΦ for the volatility of the forward price and W for a Brownian process under Q, we can then write

$$ \frac{d{\Phi}_t}{\Phi_t}=-{\sigma}_{\Phi B}(t) dt+{\sigma}_{\Phi}(t) dW, $$

which implies that \( {\Phi}_t{e}^{\underset{0}{\int^t}{\sigma}_{\Phi B}(u) du} \) is a martingale with respect to Q; from this follows:

\( {\Phi}_t{e}^{\underset{0}{\int^t}{\sigma}_{\Phi B}(u) du}={E}_tQ\left[{\Phi}_T{e}^{\underset{0}{\int^T}{\sigma}_{\Phi B}(u) du}\right] \), or else

$$ {\Phi}_t={E}_t^Q\left[{S}_T{e}^{\underset{t}{\int^T}{\sigma}_{\Phi B}(u) du}\right]. $$
(9.A-2)

Assuming that \( {\sigma}_{\Phi B}(t) dt\equiv \operatorname{cov}\kern-0.025em \left(\frac{d{\Phi}_t}{\Phi_t},\frac{d{B}_T(t)}{B_T(t)}\right) \) is deterministic, (9.A-2) simplifies to:

\( {\Phi}_t={e}^{\underset{t}{\int^T}{\sigma}_{\Phi B}(u) du}{E}_t^Q\left[{S}_T\right] \), which, by (9.A-1), implies the following relation between forward and futures prices:

$$ {\Phi}_t={e}^{\underset{t}{\int^T}{\sigma}_{\Phi B}(u) du}{F}_t. $$
(9.A-3)

(9.A-3) shows that the futures price exceeds the forward price if and only if \( {\int}_t^T{\sigma}_{\Phi B}(u) du \) < 0, i.e., if and only if the covariance between the relative variations of the forward price and the zero-coupon price, which will hold on average from now until the end of the contract, is negative, therefore if and only if the covariance between the variations of the forward price and those of the zero-coupon rate is positive.

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this chapter

Check for updates. Verify currency and authenticity via CrossMark

Cite this chapter

Poncet, P., Portait, R. (2022). Futures and Forwards. In: Capital Market Finance. Springer Texts in Business and Economics. Springer, Cham. https://doi.org/10.1007/978-3-030-84600-8_9

Download citation

Publish with us

Policies and ethics