Why Futures Mark-to-Market Uses the Contract Price Change
Summary
The document explains why a futures position’s mark-to-market adjustment is the change in the futures contract’s quoted market price, rather than that change discounted back to the contract’s expiration. The key point is that the contract’s observable exchange price represents what the position could be sold for at that time; the day-to-day change in that price determines the cash settlement.
It contrasts this direct valuation with a theoretical fair-price model that derives futures prices from spot, interest rates, storage costs, or convenience yield. Such a model can help describe or infer market relationships, but it does not replace the observed contract price when valuing the position. The explanation is conceptual and does not develop details such as margin mechanics, financing effects, or differences between futures and forwards.
Key ideas
- A futures position’s daily mark-to-market is based on the change in its quoted contract price.
- The exchange price represents the amount for which the position could be sold at that time.
- Discounting the quoted price change to expiration is not needed to determine the current cash adjustment.
- Spot-based cost-of-carry models can estimate fair futures prices or infer inputs such as convenience yield.
Tags
Full text
# Why aren't the prices discounted when futures are marked-to-market?
# Why aren't the prices discounted when futures are marked-to-market?
I am a complete novice with a background in physics, currently self-studying derivatives. My primary reading resources are John Hull's book and "Introduction to the Economics and Mathematics of Financial Markets" by Jvaska Cvitanic.
I have a confusion regarding marking-to-market. Suppose $A$ takes a long position on a futures contract $t = t_0$ at time with expiry at $T$. The futures price at $t_0$ is given by $F_T(t_0) = S_0 e^{r(T-t_0)}$ where $S_0$ is the price of the underlying at $t_0$ and $r$ is the rate of risk-free return which we assume remains fixed.
If at $t_1$ the price of the underlying becomes $S_1$, the futures price at time $t_1$ with the same expiry is $F_T(t_1) = S_1 e^{r(T-t_0)}$. So, at $t_1$, the account of $A$ has to be adjusted with an amount $F_T(t_1) - F_T(t_0) $
My confusion is since the $F_T(t_0)$ and $F_T(t_1)$ are the futures prices and they are being adjusted at time $t_1$, why isn't the adjustment calculated as $F_T(t_1)e^{-r(T-t_1)} - F_T(t_0)e^{-r(T-t_1)}$?
Pardon me if this is very obvious, I am a complete novice.
## Answer by yoggi-yalla (score 2, accepted)
https://quant.stackexchange.com/a/73409
When considering the value of a futures contract it often helps to take a step back to fundamentals. What is the market value of your contract? It's almost a trick question because you can observe the price directly on the exchange (ignoring any bid/ask spreads).
The day-on-day difference in market value of your contract is exactly the difference in the market price of said contract, because this is the price at which you could sell it.
That being said, it's common to use a heuristic similar to yours, where one assumes that the futures price is a function of some spot price (which may not be well-defined to begin with, oftentimes the front month contracts are used as a starting point) together with a combination of risk free rates, storage costs, convenience yield, etc., to arrive at some fair futures price. It could also be done in reverse, i.e. finding some implied convenience yield for a given futures price, to be used in some model where your risk factors are defined in terms of spot prices + rates, rather than futures prices directly.
I hope this helps!Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.