Government

It's a bit of an exaggeration to say that "government" is nothing but another name for "sedentary bandit". But really not that big of one.

For the vast majority of governments past and present, the primary activity has been the forcible taking of goods and services from some and giving it to others, the support of the recipients being necessary to keep the government in power. This includes modern democracies.

It seems likely to me that some form of government is unavoidable, and perhaps even desirable. "Government" essentially denotes an organization that is ultimately capable of effectively applying coercion in a region, and although the arbitrary initiation of coercion
is clearly undesirable, I don't think a complete absence of coercion is even possible. When two disputants are absolutely incapable or unwilling to come to an agreement, either one will enforce its will upon the other or some third party will enforce a decision upon both.

It might be desirable to have some organization entrusted with enforcing resolutions upon disputants, with more or less monopoly power over a region. Or perhaps not, this strikes me as a topic particularly open to debate.

The particular set of functions and policies executed by modern governments is a result of historical processes, and doesn't necessarily make any sense when viewed as if they had been recently designed with some purpose in mind. For example, in some countries governments run the liquor stores and in others the brothels. This is not because private industry is incapable of adequately serving the public need in these areas, but because positioning itself as a monopoly supplier seemed a convenient way of raising revenue. They could almost certainly serve their customers better and get higher total revenue by privatizing these industries and taxing them rather than continuing to operate them directly, but this is not obvious to most people, and there are vested interests in continuing the current policies, whereas the individuals who would significantly benefit from privatization are just theoretical people. That is, they must exist, but nobody (including themselves) can know who they are.

I think economists have by now shown that it's a highly reliable rule that market-oriented dispersed decisionmaking will consistently outperform centrally planned decisionmaking. The implication is that, from an efficiency standpoint, the only activities that should be performed by governments directly are those which are intrinsically coercive, and that regulatory law should wherever possible be goal rather than process oriented. For example, if the objective is to decrease pollution from coal-fired plants, it would be better to limit legal emissions rather than to mandate specific pollution reduction measures, and still better to apply a tax based on the amount of emissions. "Better" in the sense of resulting in lower total pollution at lower cost. So why aren't things usually done this way?

Several possibilities spring to mind. First off, a lot of it really is just disguised payoff to certain groups for their support.

Second some of it may be crappy economics. The idea that central planning will outperform the market if only you have smart enough central planners seems intuitive to many people. You have to spend more time looking at the historical record than most people are willing to to convince yourself that it is not true, and more time studying theory than most people are willing to to understand why it is not true. Certainly I couldn't convince anyone of the advantage of the market who did not believe in it already.

Another is the possibility that certain functions must be operated directly by the government in order to fulfill the government's objectives. For example, governments run the post office because they want to be assured of the ability to read people's mail. Governments run schools because they want to determine what children are taught. Governments run television and radio stations because they wish not merely to objectively inform and entertain, but to influence public opinion in particular ways.

Last on my list is the possibility that the purported objectives are actually much less important than the signal sent. For example, it's likely that most people would receive more direct benefit from appearing to care about polar bears than actually increasing polar bear populations, since most of us will never encounter a polar bear outside a zoo, and that's probably for the best.

Other suggestions?