Bush displays his commitment to free markets again. Wednesday, March 29, 2006

I don't mind Republicans. Many of my friends and relatives are republicans. In youthful indiscretion, I've even voted for a few Republicans. But I wish Republicans would quit borrowing pro free market talking points if they don't really believe in them.

Apparently, Bush seems to think that some policy is needed to reduce the use of fuel as prices continue to rise. The Washington Post reports on the latest federal political intervention into the auto industry. From the article:
Under the proposal small light vehicles would have to achieve 28.4 mpg by 2011 and bigger models, 21.3 mpg by 2011. But those numbers will likely go up in the final rule because current fuel prices are running at $2.50 per gallon, about $1 higher than what the administration projected when it devised the new standard last August.
Why is this necessary or desirable? There is already a well functioning mechanism for reducing fuel consumption. It's called a price. When the price of something goes up, people buy less of it. For car owners, that means less driving. For those looking to buy cars, that means seeking more fuel efficient automobiles. Perhaps the supporters of these kinds of policies have forgotten how the Honda automobile originally got a foothold in the American market during previous experiences of rising energy costs. In fact, the Hondas then were more efficient that the laws of that time required. No politician made a law requiring that Honda's imports be as fuel efficient as they were. Markets did.

Equally absurd are the environmentalist claims that this policy does not go far enough. If the goal is emissions reduction, then why not make exactly the opposite kind of law. Note, I'm not actually suggesting this, only pointing out how means-ends incompatible the environmentalist vision has become. If the environmentalists want a law that results in decreased emissions, that law should require that no vehicle get more than half a mile per gallon. Driving would become so costly that people would be more inclined to carpool, to take public transit, to walk, to bike, to telecommute, to stay home, etc. Actually driving somewhere would become a rare event. Again, I'm not actually in favor of such a measure, but it would certainly be more effective for accomplishing reduced emissions that the environmentalists claim to want. (Frankly, I suspect most of them just favor anything that they see as harmful to capitalism.) Well what comes closer to the half a mile per gallon maximum: a legal minimum standard, or a price?

[ posted by James @ 8:18 PM | | Digg this! | del.icio.us | Permanent link | links to this post ]

Types of libertarianism Saturday, March 25, 2006

Lately, I've been spending a fair amount of time reading Jeff Miron's fairly new weblog. Like me, he's a libertarian. Unlike me, he's a consequentialist libertarian. I'm not unsympathetic to consequentialist arguments for libertarianism. In fact, I find them very appealing. That's because I already agree with the consequences that consequentialist libertarians have in mind. The problem I have with the consequentialist approach is the belief that appeals to consequences can constitute a freestanding case for libertarianism, or any other ideological position.

I suspect that most people of any stripe would claim that they are concerned with the consequences of various policy options. Sometimes the disagreement is about what means best accomplish given ends, in which case consequentialist arguments are persuasive enough. But even to speak of costs and benefits presupposes normative categories, so I'm not too keen on the idea that cost-benefit analysis is all you need to justify any policy stance, libertarian or otherwise. If all parties to a discussion share the same conception of what constitutes a cost and what constitutes a benefit, then cost-benefit analysis might be persuasive, but if they disagree as to what constitutes a benefit or cost, I'm not sure where the resolution might come from aside from moral philosophy. Consequentialism doesn't even address that issue.

Consequentialism retains its appeal for many libertarians largely because so many libertarians see only two options: consequences based libertarianism or rights based libertarianism. The rights based approach can seem unappealing for a handful of reasons. Rights based arguments are unpersuasive to those who embrace popular philosophical positions such as social constructivism or moral skepticism or any other view that denies the existence of real objective rights. Rights based arguments are equally unpersuasive to those who believe in rights but have a different conception of what they are.

For what it's worth, I don't think that libertarians are stuck choosing between consequences and rights. There are at least two other options.

First, there is skeptical libertarianism. I'm not a skeptical libertarian because I think that there is a strong enough positive case to be made for libertarianism. However, I think it's a position that any non-libertarian should have a response for. The skeptical libertarian's argument is that the believer in each kind of statism agrees with the libertarian view about every other kind of statism, so the skeptical libertarian says, "Look, statist. We already agree that all of these other kinds of governments have no right to tax, conscript, regulate, incarcerate, indoctrinate, subjugate, etc. I just believe in one less government than you, statist. So the burden of proof falls squarely on you, but you can't give me a single reason to believe in your form of statism that you wouldn't call a fallacy if any other kind of statist appealed to that same reason to justify their belief in their preferred flavor of statism."

My own favorite is consistency based libertarianism. Here's the argument.

If I wanted to fund my own activities through the collection of involuntary payments, I doubt that anyone, libertarian or otherwise, would approve. If Ford Motor Company wanted to fund its activities through the collection of involuntary payments, libertarians and non-libertarians alike would object. If the local ice cream parlor wanted to fund its activities with involuntary contributions, libertarians and non-libertarians would generally be of one mind on this issue as well.

People differ in the exact reasons why they would object to letting any of these agencies fund their activities this way. Some believe that such a practice violates the natural rights of the folks that are forced to supply these involuntary contributions. Others believe that there would be bad consequences to letting any of the above agencies fund their activities this way. But non-libertarians make an exception when it's the government collecting involuntary payments to fund its operations.

The same idea generalizes over the range of activities that libertarians object to. Like our friends on the left and right, we would object if any other agency in society were to print money to obscure its operating costs, or to force people to participate in retirement plans against their will, or to regulate what kinds of material people could publish, or to regulate what kinds of goods people could buy and sell, or to regulate the ownership of firearms, etc. But non-libertarians make an exception when the party engaging in these activities is the government.

Whatever the basis might be for the statist's objection to these activities when other agencies do them, consider it correct, at least temporarily. For example, a statist might oppose funding one's activies through involuntary payments because the financial burden of the payment has the consequence of making the payer poorer. Maybe the statist opposes funding one's activities through involuntary payments because the payee violates the rights of the payer. Maybe there is some other reason. But whatever the reason might be, there is nothing about that reason to indicate that it doesn't also pertain to such acts when committed by a government. The consequences of the act, the impact on people's rights, etc., are the same no matter who commits it.

So whatever the statist's basis for normative judgements, there are really only three options: (1) Just assume from the start that there is some special exception which pertains only to the state. This amounts to question begging on the part of the statist. (2) Deny that one's bases for normative judgements are suitable for applying consistently. This amounts to a streetcar fallacy, pursuing assumptions only to the desired destination and ignoring their other implications. The only sensible option is to (3) be a libertarian.

[ posted by James @ 1:57 AM | | Digg this! | del.icio.us | Permanent link | links to this post ]

Prediction Markets Again Wednesday, March 15, 2006

Some years ago, people were taking seriously the idea of using prediction markets to forecast international events. That program didn't work out, mostly due to political obstacles, but the theory still appeals to many people. Tyler Cowen asks "Why don't more businesses use prediction markets?" I can't comment about some of the other possibilites he recommends, but there is one problem with prediction markets that really mitigates their usefulness, especially in a business setting. Below is a letter I once wrote to someone regarding election futures markets.
Rather than view election (or terrorism, any event with real world consequences) betting as equivalent to gambling on events where people can choose to be affected or not by the outcome (such as horse races where nonparticipants are unaffected) election betting has far more in common with options trading or insurance pricing, where users place a bet in one market in order to hedge risk in another market in which they choose to participate. Election outcome differs, of course, in that the results affect everyone whether they choose to participate or not.

This difference matters because it implies that the price in an election futures market doesn't simply represent an estimate of expected outcome. Rather, it is an expectation estimate with a bias in favor or the outcome that most users believe will harm them most. For example, suppose I believe that candidate A will erode my purchasing power by more than candidate B and I also believe that both candidates have equal chances of winning. In this case, I'll place a bet on candidate A sufficient to bring the risk of lost purchasing power into neutrality.

A good real world example of this problem comes from the prices that people will pay for car insurance. When I buy insurance, I'm essentially betting that a crash will occur. The maximum price I'll willingly pay is equal to the financial loss I expect from an accident multiplied by what I perceive to be the odds of a crash. Just knowing the price I'm willing to pay for insurance without knowing how much I expect the crash to cost is not enough to determine what I perceive the odds of a crash to be.

At best this would imply that election futures, rather than providing an estimate outcome, provides instead a product of expected election results and expected harm or benefit associated with each candidate's potential victory. This implies that some way of determining expected harm or benefit is necessary in order to determine election outcome using the price in an election outcome market.

Regards,
James
The same logic applies for business prediction markets.

[ posted by James @ 10:33 PM | | Digg this! | del.icio.us | Permanent link | links to this post ]

Capitalism is a dirty word 

Some would suggest that the reason capitalism is so unpopular with those on the left is that the word has come to mean something to people that free-market types would never actually support. One solution: change the word to free-market and express opposition to corporatism. That's all well and good, but I'm still skeptical about any linguistic quick fix. There are those on the left who are already denying that there is any such thing as a free market. Trying to beat seasoned political hacks (left or right) at Orwellian language tricks is a fool's errand. So what to do?

The approach that worked with every free market, capitalism supporting, laissez faire libertarian I ever knew was to make plain the libertarian view and exactly what it entailed. No language games, no memetics pseudoscience, none of that. Sure, the tie-dye and Birkenstock folks won't have the patience for that, and they will get excited about using the word corporatism as their new pejorative. And then so what? Support for "living" wages, a "right" to health care, government "aid" to the poor and so on will adopt "anti-corporatism" as yet another tool in the language games toolbox. No thanks.

When it comes to capitalism vs. mercantilism vs. corporatism and so on, nothing is more convincing than an impression that you know a touch of history, perhaps more than whoever it is that you are dealing with. Of course for those who still think that there is something to this whole game of language manipulation, there is always the one word that those too impatient for terminological clarifications can't possibly resist: anti-government.

[ posted by James @ 1:31 AM | | Digg this! | del.icio.us | Permanent link | links to this post ]

Is economics faith-based? Tuesday, March 07, 2006

I've encountered more than one person making the claim that economics is "faith-based." I wish the people who engage in this misuse of religious language would instead say whatever it is that they actually mean. Maybe this is too charitable of me, but I'm guessing that no one who says that economics is faith-based actually believes that economists believe demand curves slope downward as a part of their religious outlook.

If the intended meaning of the expression is that economics uses deductive methods, then I have to wonder why people don't just say that. Is there some benefit to misusing a religious term as though it were a smear word? If people have a problem with deduction, they should just say, "Economics uses deductive methods, and from that it somehow follows that economics is spurious."

Of course even the folks who believe this would probably be ashamed to actually say it so plainly. Reliance on deductive methods doesn't invalidate a discipline or make it unscientific. To the best of my knowledge, every real science relies on deductive methods. It's the pseudosciences that don't. Astronomers and astrologers alike rely on observations, but while astronomers interpret their observations in light of deductive methods such as logic, mathematics, statistics, etc., astrologers avoid these deductive methods entirely and concentrate solely on what they observe in the sky. When asked to provide some logical explanation for their claims, astronomers will eagerly oblige and astrologers will become openly combative.

If there is something wrong with an approach to economics that relies on deductive methods, then it seems that the solution is either an approach that uses no deductive methods, or an approach which relies on some alternate set of deductive methods. I don't think the first of these makes a lot of sense. The second might, depending on what that alternate set of deductive methods consists of. But anyone who has some alternate set of deductive methods in mind probably wouldn't waste time accusing the economics of being faith-based. They'd be busy talking about the advantages of their new set of deductive methods.

[ posted by James @ 8:50 AM | | Digg this! | del.icio.us | Permanent link | links to this post ]

But what about...? Wednesday, March 01, 2006

Very few people have a favorable opinion of anarchism, if they've given it any thought at all. Some of the objections to anarchism are better than others, but some just make no sense.

I really don't understand the critiques of anarchism that begin "But someone might X ..." For one, I don't know of any X that governments are uniquely and inherently more competent to deter than nongovernments.

Since I've been wrong once or twice before, I'll suppose that perhaps there is something that governments, just by being governments, can provide more competently than non-governments. So what? To impeach anarchism via these sorts of consequentialist objections, it should be shown that anarchy is worse than government on balance, not that some hand picked attribute of anarchy is worse than the corresponding attribute of government. I'm not so keen on consequentialism myself, but the consequentialist minarchists don't even know their own position well enough to set the problem up this way.

Besides all that, it's cherry picking to analyse anarchism in terms of its worst case outcomes but not to analyze states in the same way. Frankly, I'd rather the worst case version of anarchy than the worst case version of government.

[ posted by James @ 3:57 AM | | Digg this! | del.icio.us | Permanent link | links to this post ]