I'm no political scientist of course, but a lot of the points he raises here seem to make good sense. One thing it made me think of is the issue of centralization and decentralization of power in institutions of governance. We generally think decentralization is good. But what, exactly, is to be decentralized? It probably makes more sense to decentralize jurisdictions and to decentralize tasks, but to more heavily centralize the execution of those tasks.
So I've advocated on here a lot that more of the heavy lifting of health reform needs to be done at the state level (just like welfare reform), and not at the federal level. That's a decentralization of the task itself, and I think that's a good idea for all the traditional reasons we think decentralizing power is a good idea. But given the distribution of tasks, I think Frum has a point that you want to be careful with how much you decentralize the decision making within a governance institution. Of course you want an ultimate decentralization of authority through democratic and constitutional measures (i.e. - voting decision makers in and out, and checking and balancing their authority), but you can still have that in a system where decision making is centralized and there aren't a million different individuals that have the authority to hold things up.
Arguably this is what was wrong with the Articles of Confederation. The Articles actually empowered the central government to do a lot of the things that it was empowered to do under the Constitution. There was some change to the centralization of tasks in the Constitution, but there was much more change to the centralization of power within the institution performing the tasks.
So next time someone talks about supporting "decentralization of power", probe a little on exactly what they mean by that.
More Macro Modeling Meta
22 minutes ago