General shoutout to anyone who's less lazy than me: has there been any good research on whether or not advertising increases discretionary spending, and if so by how much?

The Tide example here makes me feel a little bit weird, because Tide pods in particular are a pretty bad deal for laundry detergent. If people are buying Tide pods instead of generic laundry detergent, it seems to me that the advertising isn't just shifting money around, it's actually increasing spending. But it could just be bad instincts, I don't know of any actual real research on the topic.

Another way of phrasing this -- if Apple stopped advertising iPhones, some Apple users would shift over to Android, which (ignoring cost differences) theoretically would just mean they gave the same money to someone else. But would there be another contingent that stayed on iPhone and just stopped buying new hardware every year? How large would a contingent like that be?