In Agile, it is a two-way street when It comes to the three-level management framework

At many of the places I used work at as a traditional project manager (PM), there was almost always a weekly projects status meeting attended by the company executives, directors, managers, and PMs. 

In this meeting, prepared status statements on active projects were quickly reviewed, and new projects were briefly discussed (and most of the time approved to start).

I used to ask why can’t there be a second meeting for the directors, managers and PMs to figure out the resources, and then confirm that newly approved projects can start (staffed) or not.  

Almost always, there was no such appetite to disappoint upper management or be willing to tell the Emperor he has no clothes. That is tomorrow’s problem or better yet, the PM’s problem.

Part of the reason for such practice is the concept of three-level management framework pyramid of portfolio/program/projects. You can say it is an enterprise waterfall approach.

Meanwhile in Agile, we do have something that resembles these three levels as vision and themes, epics and features, and user stories of various complexities.

That said, I will only talk now about one of the many differences between the two pyramids.

Both approaches start with the flow of information of the vision of the product or the initiative; however, in Agile, the vision trickles down as theme(s) which are then broken into epics and features and then into product backlogs of user stories, but once Agile teams start delivering potentially shippable increments of the product and reflect back on their work, some findings should bubble up to the leadership levels.

The leadership then should be able to pivot, if there is a need, and make changes which filter down to the backlogs of features and user stories with less cost of change than in the traditional waterfall approach.

The above is a framework or process, but where the Agile mindset comes in is when the leadership actually anticipates hearing back from the other levels and from their costumers and consider the findings that call for a change/pivot as learning opportunities.

No matter what the news is, it is seen as positive, because the alternative is not delivering business value to customers... which means, a product no one wants.