Consensual trades are mutually beneficial
This is a general heuristic that is accepted (both tacitly and explicitly) in most microeconomic theories: if two parties consent to a particular trade (or other activity) then the trade is beneficial to both parties relative to the null status of the trade not happening. (In the limiting case, one party is indifferent to the trade -- neither benefited nor harmed).
The heuristic isn't a precise statement of anything because its interpretation depends on the meanings of the word consent and benefit.
Explanation in a price theory framework
Consider a simple situation where there is a buyer and a seller of one unit of a good. The buyer buys the good if and only if the price of purchase is less than or equal to the buyer's reservation price. The buyer's gain from the trade, called the consumer surplus, is the difference between the reservation price (the maximum the buyer would have paid for the good, a measure of the extent to which the buyer values the good) and the actual price.
The seller sells the good if and only if the price is less than or equal to the seller's reservation price. The seller's gain from the trade, called the producer surplus, is the difference between the actual price and the reservation price (the minimum the seller would have accepted for the good)
If the seller's reservation price is less than the buyer's reservation price, then they may consent to trade at any intermediate price. At such an intermediate price, both the consumer surplus and the producer surplus are positive, indicating that the trade is mutually beneficial.
Here, one or both parties lacks information about the trade, or has misleading information about the trade, which may be the result of deceit by the other party or may be the result of other false information.
IMPORTANT NOTE: It is true that imperfect information can lead to consensual trades that are not mutually beneficial. This does not mean that failure to meet a theoretical standard of perfect information automatically implies that the trade is not mutually beneficial. Even without perfect information, it is still a general heuristic that consensual trades are mutually beneficial.
Perverse incentives: blackmail
We consider trades of a blackmail form. Here, one party threatens to perform a certain activity with the intent to extract money from the other party for the promise of not performing that activity. If the other party agrees to pay the money, then that trade is, in the strict sense, mutually beneficial, because the party paying the money gets the benefit of not having the threatening activity being done.
However, if the trade were forbidden or impossible, then the threatening party would not have made the threat, so overall, the possibility of a trade is not mutually beneficial to both parties. In fact, we can show that under these conditions, the possibility of a trade is not socially beneficial either.
However, in practice, it may be impossible to sort out blackmail trades from other trades where the threatening party does derive benefit from the activity being threatened, and is not solely threatening to do the activity to extort money.