A friend of mine came to me with a dilemma the other night. He, along with a group of 9 other friends, decided to celebrate a bachelor party with an omakase dinner at a restaurant which cost $100 a ticket normally. But this restaurant had a deal where a party of 10 or more would get 20% off each ticket -- $80 for an omakase dinner. Each would pay for their own meal, and so everyone agreed to meet for dinner next week. The groom paid a $50 reservation fee for the group, which was non-refundable.
However, that morning, he had received a group text from Tom saying that he couldn't make it anymore.
Without Tom, and without a replacement, the group of 10 would, in total, spend 9 x $20 = $180 more for the dinner. If the group were to instead cancel, they would lose the $50 reservation fee.
Does Tom "owe" anyone money, from an ethical-or-otherwise standpoint?
On one anti-Tom exrteme, you might say that if the party decides to go ahead, Tom should pay for everyone's 20%. And if the party cancels, Tom owes the groom $50 to cover the reservation fee.
On the other end, you might say that Tom doesn't owe anyone anything, because there was no "terms and conditions", and therefore no expectation that Tom broke.
Arguably, though, you could talk about a set of reasonable expectations. You could also argue about a standard of fairness. For instance, landlords expect to maintain some rate of tenancy and rental income on their properties, so if you break contract and leave early it seems fair enough that you cover some portion of the cost while they look for a new tenant.
At the same time, in general, you cannot reasonably be held responsible for every single consequence that occurs after a decision. Suppose a 10yo kid decides to DDOS a streamer on Twitch, and that streamer comes at him with a lawsuit for his typical earnings multiplied by the amount of time he's been unable to stream due to the DDOS. The kid maybe should have known better, but he's still a kid. And does the math really check out to sue for that entire amount?
This really reduces to the "fault assignment" problem, which deals with calculations involving hypotheticals, causality, responsibility by based on knowledge, so on and so on. But at the end of the day, I think that even most people who look to assign fault agree that the end goal of such a mechanism should be to maximize well-being of the whole, and that assignment of fault plays an "dis-incentivizing" by scaling the fault to the causal responsibility, and arguably, also estimates the amount of reparational responsibility. Depending on who you speak to there are other theories of fault and punishment, of course...