Neoclassical economics (the mainstream of economics these days) would say there’s got to be some gain in net utility, because it presupposes that people are utility maximising. If they’re spending money as tourists that must create more utility for them than if they had stayed at home and spent the money there. It also presupposes, however, that people have perfect information and are perfectly rational, in which case advertising campaigns wouldn’t affect their decisions anyway.Now there are a number things wrong with this paragraph. First it can be argue that "neoclassical economics" isn't the mainstream of economics today, and hasn't been for a while now. See David Colander's 1999 presidential address to the History of Economics Society on the subject of "The Death of NeoClassical Economics" to get the point. Colander writes
The term, neoClassical economics, was born in 1900; in this paper I am proposing economist-assisted terminasia; by the powers vested in me as president of the History of Economics Society, I hereby declare the term, neoClassical economics, dead.What I ask is "net utility", utility net of what?
By and large economists do assume people maximise utility, but you don't have to, there are other models of individual activity out there. One example being the Nobel Prize winner Herbert Simon's idea of "satisficing", but as Oliver Williamson has put it
[...] a cumulative research tradition within economics did not develop. It is now generally agreed that the satisficing approach has not been broadly applicable.Also it does not have to assumed that people maximise utility in a world of certainty. Choice under uncertainty is a standard part of economics today. There is the extension of utility maximisation to expected utility maximisation as a way of dealing with risk. But there are also a number non-expected utility models of choice under uncertainty as well.
Perfect information does not have to assumed, and in much of economics today it is not. Asymmetric information is taught from first year on. No student can study economics today without learning, at the very least, the basic ideas of adverse selection and moral hazard. These ideas won Akerlof, Spence and Stiglitz the Nobel Prize in 2001, so its got to be mainstream. Not to mention the even older literature on information and knowledge by people like Hayek.
It is not always assumed that people are perfectly rational. Herbert Simon's idea of bounded rationality is commonplace in economics today. Oliver Williamson makes the point that one of the behavioural assumptions of transaction cost economics, for example, is bounded rationality.
Last, I'm not sure what is meant by "in which case advertising campaigns wouldn’t affect their decisions anyway." Insofar as advertising gives useful information, then this information could form part of the information that consumers use to make their decisions. Or is it meant that people will already know the information that the advertisement gives and so will have already factored this information into their decision making. But if this is the case, then the advertiser will be able to work this out, and will have not need to advertise in the first place.
To end I should point out that most of the issues discussed above, and a number of other recent development-eg game theory, behavioural economics-move economics beyond the realm of neo-classical economics and so in this sense neo-classical economics isn't the economics of today and thus Colander is right, neo-classical economics is dead. So we end where we began.