26 March 2011

Politics or Rules-Manipulation?

Aretae believes that politics is inevitable, and looks to reduce the damage that it can do.

I should clarify what I mean by politics, because I've perhaps warped the meaning of the word a little. I feel a bit like I'm a fish trying to invent a word for water.

The exercise of power is not, in itself, politics. Politics is the process of attempting to gain or retain power. I am concerned with state power, but other forms of power (such as in an organisation) also can produce politics. A company department manager trying to make the department more profitable is not politics; trying to make his department larger is politics.

The actions of a person with power, if he is rational, will be motivated in his actions by one or more of the following three concerns:

  1. To increase the value of those things he has power over ("improve")
  2. To increase the share of that value that comes to him ("appropriate")
  3. To increase or maintain the power that he has ("win")

Improving is positive-sum. The more a ruler acts succesfully towards aim 1, the more I would call what he is doing "good government".

Appropriating is nearly zero-sum. The ruler gains, but whoever would otherwise have received the value loses. Appropriating can be in conflict with Improving, because rearranging resources is likely to reduce efficiency.

Winning can be strongly negative-sum. Whatever resources are diverted to aim 3 are not available for other purposes. A policy of Winning at all costs can be so destructive as to appear insane.

Conventional attitudes to political systems are shaped primarily by fear of Appropriating. Mechanisms are intended to set Appropriating and Winning in opposition, so that rulers avoid taking a large share for themselves as that risks their power. They work.

However, the mechanisms that do this have to legitimise Winning: rulers acting under these mechanisms openly seek to extend their power, because that is "how the system works".

Worse, for the system to work, it has to also legitimise threats to the ruler's power ("Losing"). If the ruler's power is not threatened, Winning is not operative, and Appropriating is unchecked.

Introducing these mechanisms works, and improves government. Introducing a threat to a ruler's power that will become stronger the more he appropriates will discourage him from appropriating.

But for attempts to cause the ruler to Lose to be affect him, they must have a realistic chance of succeeding. A realistic chance of power is power in itself. It can be traded, borrowed against, threatened with. A "politician" is one who holds "Virtual Power", and tries to increase it, just as a fund manager tries to increase the assets he holds.

But since the ruler, by the design of the system, is held responsible for the condition of his realm, and gains power by making it successful, his opponents the politicians gain only by making the realm less successful.

Democracy is a method of producing a group of people with both the capability and the motivation to make government worse.

There is a way around the problem, which is to make authority clear and simple enough that it is obvious when problems are the fault of the opposition rather than of the government.

Aretae says, "One huge component of increasing the net welfare of the citizenry is to decrease the ROI on manipulating the rules of the game. How can you do that?"

Why put such emphasis on manipulating the rules? I would only worry about that if I thought the rules were any good to begin with. You can — and we do — have hugely destructive politics entirely within the rules of the game, as opposing parties quite legitimately divert resources to one favoured group or another in order to acquire and retain supporters. That isn't either a manipulation or a breach of the rules: it's democracy working as designed. Opposition politicans, with their virtual power, also make threats and promise favours, some openly and some in secret.

Notably, the ruler and the opposition have one area of shared interest — one direction in which the power and the virtual power can be united. And that is to keep out of the power system anyone who isn't already in. That needn't even require "manipulating the rules", though that is the obvious way. Threatening those who support outsiders is effective enough. Threats need not be direct. For any identifiable group, there are policies that harm it.

Of course, the outsiders can't be protected, because giving someone any kind of protection from reprisals by the combined forces of politicians means giving them yet another lump of unaccountable power.



Aretae's next solution is to limit the ability of the ruler to do anything — the less power he has, the less his power is worth fighting for.

There are several problems with that. The first is that limited government implies that someone is doing the limiting (the "political perpetual motion machine"). They must have power too. It is therefore not just the power of the nominal ruler that is being fought over, but also the power that belongs to whoever the limiter is. That goes also for any attempt to protect outsider groups from politicians — the result is they become independent power centres.

The track record of limitations on government is possibly even worse than that of monarchs' lack of rivals.

The second problem is that power is always worth fighting for, because power is status. It is worse messing up a country in order to keep power over it, even if the power is limited, and it is worth fighting to increase your power, even if there's nothing particularly useful you could do with more power.

The scale question is difficult too. Small states can be very effective, but they usually require some level of cooperation between each other, at least for defense. That means a division of power between national and supernational authorities, and that division is another variable which can be fought over. The EU is the prime example; for everyone in European-level government, the primary question is what the extent of EU power is. Any ideas as to what would be good or bad to do with that power are entirely secondary to retaining and extending it.

International politics are a problem for my vision too, of course. I mentioned the Sun King in my previous post, without mentioning the fact that his reign was a period of continual war. Will an absolute ruler always lean towards conquest? I need to address that.

Mencius like Aretae preferred the city-state scale, which he called "Patchwork". However, his explanation for why this would be peaceful rested on the rulers of each patch being rational, which itself rested on neocameralism, which rests on our old friends the cryptographic weapon locks in which I do not believe.

My Westphalian World of Monarchies is not going to be as peaceful as Moldbug's Patchwork, because my kings will not be as rational as his chief executives. Some kings are going to succumb to the lure of conquest as a source of excitement and challenge, even if it is clearly not optimal in return-on-investment terms.

Therefore external security will be a much larger consideration in the design of the countries themselves. City-states may turn out impractical on defensive grounds, while very large states have to devolve power for practical reasons, which tends to produce serious internal politics. (If there is a way of managing a very large operation without devolving power, the commercial world has not yet found it.) Maybe there is some sweet spot of size that is large enough to be defensible, but small enough to be managed without compromising centralism. Obviously doing as little active management as possible is a key technique.

2 comments:

Devin Finbarr said...

Very good post. I like this sentence, "The first is that limited government implies that someone is doing the limiting (the "political perpetual motion machine"). They must have power too."

When you say you are a fan of monarchy, are you thinking traditional hereditary monarchy, or are you open to the idea of some sort of elective monarchy?

sconzey said...

This is a fantastic post.