Maybe I can give some perspective as I am both a “classically trained graphic designer” (who happens to work on iOS App interfaces on a daily basis) and a UX designer. I often found myself filling both roles in projects.
First of all, aesthetics, as a designer, matter to me. Not in the sense of veneer or clutter – but scaling down Retina-optimized graphics to older displays can often lead to blurring problems, as Guy mentioned. It can look shoddy. Since it is my responsibility to deliver in an adequate quality it just irks me to go the easy way because it reflects badly on the quality of my work – even if the “average” users will never notice it because they either have a device with a Retina display or they don’t.
I still feel that it is important to treat users with older devices with the same amount of respect as users with newer devices. If I see a gap in quality and I am able to fix it, I will do it. Within reason. There is a saying here: “It’s possible to die in beauty” – you don’t want to do that. But if you can identify a problem and are able to improve on the work you do, why not. I get the economic argument, but I don’t think it really applies here. Designers are cheap compared to developers and the developers I worked with didn’t care if they have to swap one asset for another – actually they insisted that we aim for the highest possible quality for the product we wanted to deliver. Development timelines are not affected by these small design tweaks.
Another thing that is very important to keep in mind: there is a big difference in how graphics are displayed on a computer screen and an iPhone display. It’s hard to judge how big the difference in Jack’s example is without looking at it on the devices it is intended for.
That said, I understand your argument. I wouldn’t consider the scaling problem a showstopper, but since it’s easily fixable with a few tweaks… why not just do it.