The language of economics is designed so that this is tautologically true and all empirical evidence that this is false is euphemized away with the term 'externality' or 'market failure'.
Market failure is a subjective term that assumes a problem can be always be solved fairly (no one is made worse off). That euphemism tends to be used by those who believe "free market" (no such thing) solutions are inferior to state-imposed solutions, which often involve force and/or benefit the ruling class. There's plenty of empirical evidence to support that.
Market failure is defined as a situation in which markets fail to achieve pareto efficiency; pareto inefficiency is definitionally the condition where someone's situation can be improved fairly, as you call it. A few heterodox schools simply choose a different definition of efficiency to wave away the problem of market failure, but this is circular logic: They define markets as the best system of allocating resources and therefore it is impossible for markets to fail.
If that's the position you want to take, please be forthright about your assumptions. I no longer think we're using the same basic definitions and I am fairly certain I am talking about well established mainstream concepts in economics.
P.S.
> state-imposed solutions, which often involve force
Private property as it exists is definitionally a state-imposed by force concept. I don't think your moral claim about people who use the term market failure is as biting an indictment as you think.
And while state-enforced property rights have enabled many other technological advancements and objectively moral outcomes, this does not give the state the moral high ground to use force for its own arbitrary purposes. On the micro scale, just because my authority allows my children to claim their own space, clothes, etc. doesn't necessarily mean that authority can always lead to good outcomes in every interaction they have with each other. It's just a means to an end, like the state.
https://en.wikipedia.org/wiki/Tragedy_of_the_commons#Critici...