I don't know in what type of environments you've worked, but "rejecting a feature" and "managers blindly deferring to engineering estimates" are very alien concepts to me, and I'm pretty sure to most people out there.

What usually happens in my experience is that management comes up with some new requirements that they may or may not have already sold to a customer on a more or less specific timeline. They then ask for an estimate on when that will be implemented in the product.

If the estimate doesn't conform to their expected timeline, they typically challenge it vehemently. If you can convince them that it's a realistic estimate for what they're asking, they then start discussing ways to reduce the scope while still delivering more or less the same feature. If you can't, then it will end up at "you're being pessimistic, let's start and we'll pull through", which will later turn into "we'll have to work this Saturday because we're way behind on what we committed on that feature".

Sometimes, a very well respected engineer will occasionally be able to convince someone in management that a feature is just not worth it. But that is the rarest case I've seen, and it will definitely not happen if some junior or middle engineer tries to invent some generic excuse.