That hypothesis doesn’t fit Romer’s prior career. Mine does.
Romer is an economist specializing in new growth theory — he’s one of the seminal figures in the movement — and his work says that roughly all economic growth is knowledge growth. Romer is about increasing knowledge to increase growth, which is the only big issue besides violence and war.
Romer also knows that knowledge only increases by trying new stuff — and then he read public choice and saw that government institutions basically don’t change their rules. Hence governments don’t increase knowledge and don’t get better.
Since his obsession is growth, by means of trying new stuff, it’s hardly fair to say he’s only echoing the boring Moldbug hypothesis. He’s far smarter than that, and besides, politically mostly liberal.
His line is that you don’t know what will work best, and neither does he. At best he’s pushing neocameralism for a first try, in order to garner some extra support for his actual radical position of trying lots of different things.Again with Aretae's 1st law:
The feedback system defines the system itselfRomer is arguing that the system of governance worldwide is stagnant. The only path out, according to all of us contrarians, is to have places that try new things. Some of the new things will work better, some won't. But until we have places to try new things, nothing will get better.
Moldbug argues (with Carlyle) that a specific system (USG-Corp), or almost any system in a specific direction (formally aristocratic) would be better.
This is a nice line, and there's some chance that it's true...but it's nowhere near as interesting TO ME as the Romer line that the evolution and improvement of legal systems has been interrupted and needs restarted.
For those who have forgotten...here's Romer's kickoff of the idea at TED...in which he explains things the way I did.
Posts
8 comments:
but it's nowhere near as interesting TO ME as the Romer line that the evolution and improvement of legal systems has been interrupted and needs restarted.
No, it is Romer's line that is banal and uninteresting. A lot of people agree that we need more experimentation in governance structure. I believe it, Moldbug does, Foseti, Kling, even progressives I chat with at cocktail parties are always at theoretically in favor of political experimentation.
The hard questions are 1) why is the US State Department so resistant to allowing political experimentation around the world? 2) Why has the state department actually been hostile to political systems that actually worked? 3) How do we break the State Department's grip on power? 4) If we finally get a chance to run a political experiment, what should the form of this experiment be? These are the questions that Moldbug discusses, and that I like to discuss. Romer has nothing to contribute to these debates.
The question of what form the first experiment should take is very, very hard problem and very, very important problem. Experiments in political systems are ridiculously expensive to enact. A bad experiment will do an enormous amount of harm and discredit the project. I know you want to make experiments cheaper, but designing a political system that makes political experiments cheap is itself a very, very hard engineering problem. In fact, making experiments cheap would require overthrowing the power of the existing U.S. State Department. How is this done? What replaces it? These are the hard questions.
That's why people like me and Moldbug spend about 5% of our time on the need for experimentation and 95% thinking about what the form of the experiment should take. In the circles in which we travel the former question is assumed. The latter question is the harder question by far.
Devin,
I was expecting you to visit.
Unless I'm missing a lot, plans like yours and Moldbug's seem to be a lot like project/battle plans. Obsolete the moment they start.
The fact that lots of people are theoretically in favor of political experimentation reminds me of Robin Hanson's near-far distinction...and I'd call it far. In favor in theory, but opposed in all specific cases is where the progressives I talk to line up. Unimpressive. Kling is one of the most carefully Hayekian thinkers on the web, so I'm not shocked that he's in favor.
1) Experimentation threatens existing powers.
2) Hostile to which working systems recently? Singapore? or are you talking South Africa?
3) That's hard --
4) I assume it will fail somehow...so I want to know (a) what type of system will fail most informatively.
Our usual disagreement again:
You talk about success & design...I talk about failure-management.
Moldbug has a long post on how "charter cities" would have been recognizable to anyone alive 100 years ago as colonies (http://unqualified-reservations.blogspot.com/2009/08/from-cromer-to-romer-and-back-again.html).
The British ran different colonies in very different ways - after determining what worked and what didn't work in each colony, i.e. after experimenting.
What are the classic examples of charter cities? Singapore and Hong Kong, which also happen to be . . . British Colonies!
You're speaking as if Moldbug, Isegoria, Devin or I disagree with Romer's ideas. We don't. We just wish he wouldn't try to pass them off as new ideas. They're really old ideas.
In favor in theory, but opposed in all specific cases is where the progressives I talk to line up. Unimpressive.
Exactly. That's why the project is not to convince progressives that experiments are needed in general. The project is to convince progressives to undertake a specific experiment. What should that experiment be? What kind of evidence can be arrayed to convince progressives that the experiment is worth doing? Those are the hard questions, and the world needs help from smart minds like yours in figuring out answers.
Unless I'm missing a lot, plans like yours and Moldbug's seem to be a lot like project/battle plans. Obsolete the moment they start.
That doesn't mean project plans are not worthwhile. And it's more analogous to a bridge design blue prints, or some other project where iteration is much more difficult and so you need to spend a lot of time reviewing the plans.
2) Hostile to which working systems recently? Singapore? or are you talking South Africa?
The most recent example is Honduras. But in general I'm talking about the entire democratization/decolonization projects which destroyed functional political structures across Europe, Africa, Asia, and Latin America.
3) That's hard --
Yes it is. That's why it's exactly the question worth discussing.
4) I assume it will fail somehow...so I want to know (a) what type of system will fail most informatively.
Ok, so let's think about it. What would be a good experiment that would minimize the worst case, and maximize learning? Let's say your trying to convince the president of Madasgascar to setup a charter city. How do you propose to the president that the charter city be run?
We are in unanimous agreement that the world needs more political experiments. But the problem is the world has a dearth of political engineers. So if we want experiments, the next step is developing the intellectual framework for designing the next experiment. We have no good rocket/political engineers, and we need them if our next political experiment is to have any chance of success. That means learning about governance structures ancient and modern and figuring out where they went right and what went wrong. It means brainstorming designs, and then ripping them apart.
Unfortunately virtually no one in existence if actually doing this. Which is tragic because the entire world suffers from a serious case of bad government. I'm trying to get you to join in this project of taking political experimentation to the next step; of building the intellectual corpus needed to make the next political experiment successful (even if "success" is just defined as learning enough to make the next, better experiment).
You talk about success & design...I talk about failure-management.
Actually, I think a key component of any political system design is guarding against catostrophic failure. Safeguards against tyranny are even more important than designing for runaway success.
I read Moldbug's post on Romer when it was posted. Didn't think it was fair then either.
Libertarians say: try this system.
Moldbug says: That's stupid...try this one.
Romer says: The only way to get better legal systems is to try lots of things...and see what works.
I am unaware that anyone had previously suggested that the purpose of a "colony" was to try new rules so as to see what worked best.
Every time I read Moldbug he seems to be saying...we already know what works best...and we knew it 200 years ago. But we broke it since.
Romer is much more impressive in saying: WE DON'T KNOW.
It's like markets...the EMH folks who say...look, you can't know...are more impressive than the folks who are making specific predictions...and are usually wrong.
Devin,
You are horribly effective when you use praise like that. And your final line disarms my big objection. I now need time to think.
Foseti-
What's your take - do you think Romer is consciously aware that he is rebranding colonialism/for-profit government? Or do you think he honestly thinks there is a difference?
If he is deliberately lying - is he actually wrong to lie?
I find the Julius Caesar versus Augustus example very interesting. Caesar failed to put an end to the republic because he openly and avertly declared himself dictator. The result was rejection and decades of more violence. But Augustus maintained the fiction of the old republic, and in doing so was very successfully able to hold power.
I wonder if Moldbug underestimates the necessity of lying about the neocamerelist project.
Should neo-camerelism ever be proposed, perhaps it could be called "Formalist Democracy", with the following tenants:
a) an "economic development authority" is created to "manage" the city's property. This "authority" has shareholder management with non-voting stock held by citizens. The "authority" is not actually ruling, you know, it's just managing the city's property for the benefit of the residents. The "authority" controls the budget, the civil service, and has veto power over all city ordinances.
b) the mayor and city council is stripped of real power, but still exists. Maybe it has a small budget to throw parties. It becomes the equivalent of a student council.
c) fix taxes, and contractually fix all the current rights of citizens and property holders.
d) preserve trial by jury (this is the democracy part)
One would also note that a for-profit city would not design zoning laws that create urban sprawl (since sprawl is an underutilization of the cities assets). Thus the government can promote itself as being green and sustainable. Feudal government dressed in the clothes of leftism :-)
Devin, good question.
I think Romer honestly thinks he's proposing something different.
I think this stems from the fact that he has no idea what good colonialism looked like.
I remember Moldbug once saying that he often wins arguments with Progressives because they've never even heard of his theories while he's always heard all of their points. I suppose the same logic applies here. Romer has no idea what Cromer was doing in Egypt.
Aretae goes on and on about encouraging growth. I wonder if he can name better examples of growth that should be emulated than certain British colonies. If we're looking for things to try to improve growth, why not start with things that worked in the most hopeless areas in the world. Think about Cromer's Egypt, which included modern Sudan. As Moldbug said,
"Lord Cromer, a grown man named "Evelyn," governed an Arab country much like Iraq for 25 years, at minimal cost and without significant violence, and did such a good job that Egypt [including Sudan!] became an international boho destination for the likes of Lawrence Durrell, a sort of Edwardian Prague."
Again, I'd point out that I don't disagree with Romer on anything, except what to call "his" ideas.
Post a Comment