As always, I lost my thread in my post on the left and the right. Here is another maddeningly imprecise, highly abstract notion that may or may not pass muster.
To the extent that the goal of the state is to supplant or even merely to supplement kin-based social networks (i.e., extended families, super-extended families or tribal formations, thick ethnoreligious communities, all of which have historically been complexly interrelated) as the vehicle for nurturing and disciplining individual behavior, bureaucratic rationalization is a mixed bag. Bureaucratic rationalization notionally represents an alternative to the so-called patrimonial state, in which the state is essentially an extension of a tribal formation, e.g., a ruling family that secures the allegiance of the landed aristocracy and rules the “state” (or protection racket) in accordance with the interests of said coalition of families. The great advantage of a law-governed state and routinized bureaucratic procedure is that arbitrariness, and thus uncertainty, to use a polarizing term, is markedly reduced. This is true for citizens, or the subjects of state authority, and also those employed by the state apparatus, who have a series of legal-institutional protections against the elite(s) that control the state.
Classic instruments of the patrimonial state, like tax-farming, in which individuals purchase or are assigned the right to collect revenue on behalf of the state, are abandoned in favor of routinized bureaucratic procedures. There are, however, theoretical advantages to a well-designed tax-farming regime, e.g., it might be more flexible, and profit-seeking entrepreneurs might introduce “disruptive innovations” that create economies in the revenue raising process. Consider a scenario in which the state puts out an RFP and says, “Hey, we need to raise $500 to defeat the Gauls. Whoever raises it with the least hassle for us gets this particular contract, but there will be others for other purposes.” A fervent believer in the power of the Laffer Curve might pursue a different strategy from someone who loves unimproved land taxes, etc. And over time, we might find better and better ways to raise revenue in ways that are conducive to the interests of the citizenry. Yet it is also powerful, in a regime in which citizens are radically disempowered, that this arrangement will yield brutal tactics that trample the rights of individuals, etc.
It is easy to imagine that a decentralized discovery process is going to yield better results in some domains than in others. Predictability, consistency, etc., might be the watchword in others. Much depends on the nature of the “problem” being solved — and of course “problems” are identified or constructed by rival constituencies, elites, etc.
If the “problem” we aim to solve is to supplant or to supplement the nurturing and disciplining function of kin-based social networks, or rather to lower the cost of disembeddedness, routinized bureaucratic procedures — highly centralized, resistant to change, likely to create powerful political and economic constituencies in favor of maintaining the status quo — might be an unusually bad way to go. For all of the downsides of kin-based social networks (they can be oppressive, stultifying, narrow, controlling, etc.), they tend to rely on a high level of local or intimate knowledge, shame-based punishment tailored to individual circumstances, and, in the best case scenario, a richer understanding of the broader circumstances that shape individual decisions. Think of the juvenile court judge deciding what to do with the kid found with a dime bag. That judge is likely to rely on all kinds of heuristics to reach a verdict — yet she actually has very limited information, and he heuristics, though based on long experience, might actually be extremely misleading. Regardless, using this heuristics might be preferable to, say, a regime defined by mandatory minimum sentences. And even better (all of this is in theory) would be a situation in which someone who knows the child reasonably well and has an interest in seeing the child flourish long-term, but not so strong an interest that this outweighs the interests of the broader group, is responsible for determining a suitable punishment. Note that this is an imperfect description of a tribal leader.
Because we need access to a richer portrait of individual circumstances when making the kind of decisions that were once made in the context of kin-based social networks, one could argue that we need a different kind of state — a state that is more nimble and responsive to change, and that can thus evolve in response to rapidly changing social circumstances as traditional formations dissolve or change shape.
All of this is to say that if one embraces the idea that the state should take on more of the role that had been played by traditional formations, the case for things like the reform of work rules that tie the hands of public sector managers looking to restructure state agencies is actually strengthened. The state necessarily has many disadvantages relative to more intimate associations, the biggest one being that states can only “see” or discern a limited set of facts that rarely provide a very thorough or accurate portrait of reality, and preventing state agencies from learning and evolving makes its task vastly harder than it might otherwise be.