I don't have a problem with macro bureaucracy system. I think you can make very good macro systems and very good micro systems. My problem is in their interaction, and how micro and macro scale actions impact each other. This, I feel, is where all 2e attempts at macro systems failed in my eyes.
@Shyft mentioned earlier what happens when five PCs slaughter a hundred city guards. This kind of encapsulates my issues with macro management. I have yet to find a system that satisfyingly handles the fundamental
difference of scales.
Like, you're a city leader. You have influence stats, and military stats. You can take macro actions in order to wage an extended war with another city. You have prized advisors (likely developed as minor characters in their own right) determining specific bonuses or stats in certain areas.
But a character can still engage all that on a micro scale. An Exalted warrior can cut through half your garrison as an individual character fighting a mass combat unit or battle group or what have you, using her personal stats. A seductor can wrap your minister of finance and your high priest around his little finger and get them to hand out valuable information and take decisions which advantage him. Your ambassador at the neighboring prince's court might have been eaten and replaced by a Lunar. An Outcaste who has sworn loyalty to you may be giving you a potent daiklave as a gift.
It's easy to say, in a vacuum, "well you run the fight scene as a micro scene and then you factor in how many soldiers died and affect the city's Military Rating appropriately" and so on. The problem is that Exalted is a game where there is going to be constant, fast-paced back-and-forth. Can any combinatorial system handle the constant zoom in/zoom out process of heroic characters performing individual actions with massive consequences over and over? I have yet to see a system that does.
And there's the problem of the level of abstractions. With two different resolution scales you'll end up with two different resolution processes and thus, resuls. What this means is that when you undertake a Raise Manse project, you're going to have the choice to resolve it as a macro scale action where your Twilight mason is an Asset granting a +X bonus or as a micro scale action of the Twilight actually rolling out a full crafting action with the 10 Charms she's invested in it, and the two will have massively different outcomes and rules.
Or you define crafting as a "macro scale action" and the Twilight
can't act on the micro scale, so her 10 Charms actually only apply to the macro-level Raise Manse project. But then you end up with that weird dissociated game with two levels; your Dawn is the Doomguy in one scene and a RTS "hero unit" in the next. I don't really like it, especially because in these situations the macro scale always ends up
mattering more. This was the Creation-Ruling Mandate's problem, in which you wanted to push actions to the macro scale because you wanted the ability to throw as many modifiers on long-time rolls as possible to win without having to go through individually challenging scenes in short-time that might actually press you for motes, willpower and screen time.
Even video games have failed to resolve that issue - and not for lack of trying. Look at the many attempts at "stronghold system" in party-based RPGs. There've been many interesting experiments in the genre, but it's never worked out great.