Entry Barriers and The Power of the Permanent Bureaucracy

One of the things that Brad found when researching our recent public choice paper is that there has been some past economic work on switching costs, though not barriers of entry:

The first rigorous treatment of competitive government from an evolutionary perspective comes from Vanberg and Kerber (1994). As in the neoclassical understanding of Tiebout competition, governments have an incentive to attract mobile resources. To achieve this, they will engage in rule innovation and successful conjectures will be imitated, leading to an improvement in rules over time. Variation is generated by existing jurisdictions coming up with new rules, and the selection process is consumer choice.  Wohlgemuth (1995, 2008) considers institutional competition as an evolutionary learning process and argues that the number of parallel experiments and the strength of the selection mechanism (that is, the ease of citizen exit) are crucial to the rate of institutional evolution.

However, we disagree with this past focus.  For example, at The Seasteading Institute, we have been shifting to emphasize the idea of “startup countries” and the main benefit being the startup dynamic, rather than the modular rearrangement possibilities.  This is because Barney Pell, Jon Rauch, Tim Lee and others have pointed out the weakness with modular rearrangement, and while I think it still provides some benefits, I agree that I overestimated them initially.

Fortunately, lowering the barriers to entry and allowing startup countries can still over great improvements.  After all, the neoclassical assumption that existing firms can innovate as well as new ones is awfully questionable when applied to innovation in organizational structure, which is very difficult to change “in place”.  And in the specific case of democracy, we know that special interests and unelected bureaucrats add significant rigidity to the system.  When the very failure of democracy is the way that certain special interests entrench themselves at the expense of others, expecting a democracy to fix this “in place” seems naively optimistic.

As evidence, Moldbug offered me this great link on how the bureaucracy actually works, and how unresponsive it is to politics:

The first thing to realize about government service is that even in the most momentous election, a tiny fraction of government workers actually change jobs. A few hundred politicians and their staff and perhaps the heads of most of the agencies would lose their job after a significant electoral shift. At most, a few hundred people lose their job out of a total staff out of 1.8 million. In no other bureaucracy would anyone consider the removal of 0.01% of the bureaucracy a major change.

The day in the life of the head of an agency is very busy. He or she is shuttled from one activity to another. Generally the bureaucracy tells him or her what to say in all these situations. I get to write little statements for the head of my agency all the time. This busy schedule also serves to keep the head of the agency from actually accomplishing anything.

The fifth thing to realize is best realized by observing the recent “monumental” pieces of legislation that have passed. Where did Congressmen get 2,000 pages of laws on healthcare reform or financial reform? . . . If you answered that they slaved over a computer to write 2,000 pages, you’re wrong. They asked the bureaucracy for it. Your humble blogger wrote the first draft of an amendment in one of the recent mega-bills, for example.

The sixth thing to realize is an addendum to the fifth. If you read the recent mega-bills you’ll see a strange thing. Congress isn’t actually legislating most of the changes. In most cases, Congress lays some vague ground rules and mandates that agencies actually write the rules. Thus, the bills serve, first and foremost, to give power to the agencies.

This is not a situation than can be effectively fixed through mere politics.  If one wishes to truly, deeply, radically change how government works, starting from scratch with a new set of initial conditions and rules for the evolution of that government, and then scaling it up over time, seems far more likely to be effective.

Read more at Let A Thousand Nations Bloom