Let me just outline a couple of points:
what’s in it for the shareholders ?
how do you solve the problems that plague representative democracy ?

Coming soon…

The other notions you explore such as separation of powers, democracy, and federated lawmaking all have the potential to trigger a feeling of uncertainty in the unprepared manager. That would be ok, if not for the perfectly natural human aversion to uncertainty, for one thing. In addition, it might not be obvious how to separate accountability from power and decision/law-making. Also picture the sweat on the unprepared manager’s brow when confronted with a team who have self-organised to agree different objectives to what he or she committed to their clients. There are many factors that conspire against the picture of a project that you may have and that you were able to paint in my mind when reading through your post. The model you might describe with those notions would be very egalitarian. The closest I have been to that was one of the best experiences of my career and I would like to draw from that to offer some factors that could be considered when trying to move in that direction.

- All team members are created equal…
…but some are more equal than others. Perhaps democracy is at least as much (if not more) about representation than about equality. That means that some people will read that word and think of the vote that makes everyone equal. Others, however, will think of the politicians that are being elected to represent the voters, and the power bestowed upon them. The metaphor further breaks down because those on the second camp don’t always perceive the post in the spirit it was intended, i.e. that of a “servant of the people”. The reality is even worse because managers are typically not elected but appointed in the first place (by their senior managers). In that situation, the temptation to assume full decision-rights on behalf of their team is just too strong to resist. Should we suggest that the team should be able to elect their manager? Yes. Should his or her term be finite? Should new elections be held to reelect or replace the manager? Yes and Yes. These would be more accurate implementations of the equality that contributes to the high performance of agile teams. Who are the leaders in this structure? They would step up (and step aside) as the task requires it.

These are not easy concepts to implement in the medium-to-large organizations you refer to. However, any step in that direction would help influence the behavior of the manager towards decision-making “at the request”, rather than “on behalf” of the team. The difference would be visible on the manager’s brow. Decisions at the request of the team should display no sweat.

- Agile Enterprise
When talking about agile, I have observed that the conversation can seem somewhat constrained to the development team and its immediate management. That could be because I typically have those conversations with developers, but I would suggest that the literature also focuses at that level and doesn’t go far enough in explicitly addressing the “whole agile enterprise”. Bear with my daydreaming for a bit and help me think through this: What would a company that is entirely run on agile principles and methodologies look like? (I’ve never seen one but I’m told they exist) In my mind they would have a very different idea of what a manager is required to do. There would be no “project managers” in the sense that you and I understand the term. There would be people whose job it is to maintain the agile mechanisms in working order and to “oil the gears” frequently. The role of a manager in this context is more about enablement and facilitation. This is reasonably within the domain of mainstream agile methodologies and the literature but what is less obvious is what the senior layers of “facilitators” are expected to do. Are there perhaps no further layers above this? I doubt it, but I suspect the term “hierarchy” would have a very different implementation. I would imagine that small teams would send a representative or two to link up with other teams and learn about other work going on around them. In the model in my mind, these also happen in a stand-up somewhere on the open plan. Say up to ten representatives from different teams gather together to hold a cross-project/product/function stand-up. They would form an inner circle of discussion and go through their achievements, plans, obstacles etc. Everyone is aware of this meeting and can opt in by getting up and standing around that inner circle and observe. Some would simply learn and gather information that might shape how they do their own work. Some would be on the lookout for information on work they depend on to move their own project forwards. Others might be better placed to influence the “bigger picture” and will only participate to offer the company or industry context that the ten participants might not have.

This structure is repeated in groups of up to ten people until all the work taking place in the organization rolls up to the executive committee (CxOs or whatever the agile equivalent would be). During this process, the CxOs have had the chance to participate in any of the stand-ups along the way to either validate the communication methodology and information flow, or to provide context directly at the layer where it is required rather than through traditional Chinese whispers. Similarly, a new joiner would be able to observe and learn about all the work that is taking place in the organization and choose where his or her contribution will be of highest impact. Wherever there is leadership, it is obvious. Where leadership is entirely missing, it is good that this is also obvious because the group can seek out a leader before taking the next step.

- Uncertanity
If we take the above descriptions of management and the enterprise they operate in, you need to staff it with people who are very comfortable being surrounded by “uncertainty”. Chris Matts has been developing the arbitrage of uncertainty through the application of real-options theory to management. Here is a good article on that:
http://www.infoq.com/articles/real-options-enhance-agility
I was fortunate enough to work with Chris and observe these concepts in action. Observing him applying his theories made it obvious to me that many decisions are made as early as possible and unilaterally by senior managers. This is what they have been conditioned to do. It addresses their own aversion to uncertainty but the net effect is that they are very likely to be making that decision when the least information is available. They rather make a poorly-informed decision before it is needed than live with the uncertainty until there is enough information to make a better decision or until a trigger event requires that decision to be made.

- Insanity?
The manager I’m describing earns the voters’ trust and gets elected. He or she needs to give up decision powers to those that elected them and to make things happen at their request. He or she can be deposed or may be required to temporarily rotate out of the role. Their measure of success is to create a self-organising and self-sufficient team, effectively making themselves redundant. A really good one would also need to have invisibility powers. This does not appear to have the makings of a compelling job spec, but I know they are out there. The obstacles include a) how to create an organization that knows how to use them, and b) how to separate the valid candidates from all the “nutters” that will undoubtedly apply for such a job.

Leaders are easy. They are attracted by the opportunity to make a difference. They will find you.