# Are futures/forward contracts tradable in the middle of its life? If yes, how?

I think I'm having some trouble understanding what trading futures/forward contracts means. Assuming a market over the period $[0,T]$, for a European contingent claim $X_T$, my naive understanding is that: Bob buying such a contingent claim at time $t\in [0,T)$ from Alice means Bob paying Alice $V(t,\omega)$, the value of this contingent claim at $(t,\omega)$, and Alice giving the contingent claim to Bob so that now Bob owns it.

For forwards I think they are essentially a very simple type of European contingent claims, so they should also be traded in the above manner: the buyer pays the seller the value of the forward contract and the seller becomes the new owner of the contract. Is my understanding correct so far?

For futures things become more complicated due to the margin account mechanism. Let's say Bob establishes one long futures contract to buy the underlying (whose price process is, say, $S_t$) at the arbitrage-free futures price $F_0=S_0e^{rT}$ in time $T$. Now, at some time $t\in (0, T)$, Bob wants to sell the contract to Alice. Then
1). is it possible to do so in the real world financial market?
2). If such a trade is legitimate, is there any conventional way to do this? For example, how much should Alice pay Bob? This is now ambiguous because unlike forwards the value of a futures is automatically set to zero in its daily settlement. So I'm essentially asking are futures contracts traded prior or posterior to the daily marking to market.

• Contracts are traded during the day at market determined prices. In the computer that records trades at the exchange the price for Bob will be recorded as $P$, which may be (usually is) different from the settlement or m2m price $F_{t-1}$ of the day before or the m2m price of later that day $F_t$. Of course the price for Alice will be the same price as for Bob, since they traded with each other at this agreed price. – noob2 Apr 4 '17 at 17:07
• At the end of the day (mark to market time) Alice will have a profit (or loss) $F_t-P$ on the contract that she is now long. Bob on the other hand will have a profit $P-F_{t-1}$ in his account, which reflects the closing of his position. – noob2 Apr 4 '17 at 17:22
• @noob2 Alice should have a profit of $P-F_{t-1}$, and Bob would have $F_{t-1}-P$. Also (for the OP), this is all under the assumption Bob sold to Alice, which doesn't have to be the case. – msitt Apr 4 '17 at 17:27
• @msitt I think noob2 is correct. At the time of buying, Alice should pay Bob $P-F_{t-1}$ for the contract, which is exactly the value of the contract. So Bob makes $P-F_{t-1}$ and stops owning the contract; for Alice, she doesn't have to care about the history before the moment of buying when she didn't own the contract after all, but needs to be liable for profits/losses on the contract once she becomes the new owner of it, and thus at the end of the day her profit should be $F_{t}-P$. – Vim Apr 5 '17 at 5:10
• @vim Actually I just realized I completely misread the question. noob2 has it completely right above. – msitt Apr 5 '17 at 5:17