Social science and the news usually treat organizations as blind and mute. But if we look at each person in a job in a bureaucracy we see people under great stress trying to maneuver the winds of rumor and change. I have at places like Bell Labs and HP, watched the staff and their managers work hard to neutralize change programs proposed and funded by top management. A bureaucracy is not a dead thing just following policies and order but a vibrant organization dedicated to its survival.
Perhaps the major factor in bureaucracies being able to change is having really attractive managers in top positions working with each other. If lower-down employees see their managers as able to cooperate with peers and cooperate up a level or two in the hierarchical structure, change is likely. But the reality is most managers don't cooperate across lines.
Once at the World Bank I initiated a project on trying to understand the ways and whys people worked. The VP of the West Africa Division (this was in about 1975) asked me to work with his directors. Then the VP of the Economics Department also asked me.to work with him. The then-president of the bank mandated more cooperation across units. The VP of the economic division said in his weekly staff meeting “We need to engage with this. I want a volunteer to go to the West Africa staff meeting in a few weeks and ask them what they would like us to work on with them.” I was able to attend this series of West Africa weekly staff meetings in which the VP described the situation to his directors and said “We need to come up with something for them to do that will have no impact on what we do" The group, cooperatively, quickly came up with an issue for the Economics Division to work on and described this in the next meeting to the representative from Economics who then reported to the economics senior staff the topic. The economics team was self-congratulatory that they had something significant to do, just as the West Africa staff was self-congratulatory on coming up with a useless project to distract economics from seeing into the inner workings of the West Africa department. I got to watch this educational moment. No one ever caught on. There were serious issues of developmental economics that were not discussed in either department.
Imagine, as later was the case, one of those issues was climate change. Bureaucracy is needed to manage climate change but most bureaucracies are stuck in self-produced quicksand.