Some Thoughts on Externalities

Particular irregularities in the world of economics cause the mainstream to throw up their hands, abandon hope for the marketplace, and cry out relentlessly for their government-savior. One such thing is externalities: when transactions between two parties have an unintended effect on a third. It couldn't possibly be that the people-- narrowly focused on their personal gain-- could account for such a misfortune more adequately than the all-knowing centralized planners of the state (or so they say).

Suppose that in the process of making toys for children, my factory exudes pollution that reigns over a city. Suppose my airport, while safely delivering its customers across the world, creates a bubble of disruptive noise for the surrounding neighborhoods. Or suppose that my school reduces the likelihood that its go-toers become criminals-- a "positive" externality that benefits many who will never pay for my good-doing. In such instances, do we leave it to the private sector to create its own solutions? Or do we call upon the power of government to try and set things straight?

I posit that neither does a particularly good job, but that the government-- as always-- does worse.

First, to outline some typical government-interventionist arguments:

I. A toy factory emits harmful pollutants-- a negative externality.

a) The government could ban the factory from polluting.

b) The government could impose a tax on the toy factory for each cubic meter of pollution equal to the harm caused by a cubic meter, thereby encouraging the factory to pollute only when the total costs of the pollutant are outweighed by the benefits. This is known as Pigouvian taxation.

c) The government could nationalize the toy factory (make it public property) and pollute only as much as is deemed reasonable.

II. A school educates students, making them less likely to commit crime against others-- a positive externality.

a) The government could give money to schools equal to the benefit granted to others, thereby encouraging schools to open when the total benefit outweighs the cost. This is known as subsidization.

b) The government could nationalize schooling (making it public) and educate as much as is deemed reasonable.

Second, the reasons I find government to be a worse solution than laissez-faire:

1) When a positive externality occurs, the producer has incentive to find ways to capture profit from the "free riders" (those who benefit without pay). Likewise, when a negative externality occurs, those affected have incentive to eliminate the harm.

2) Members of government have only the incentive to please the public-- which almost certainly has a poor understanding of a given externality's exact costs and benefits. A government has no incentive to determine the costs and benefits of an externality for itself, and ignores whichever half of the equation is ignored by the public.

3) Government powers, when extended to correcting for externalities, will be abused by interest groups whose resources are concentrated.

Now for some further examples, taking all arguments into account:

Example 1: A city contains a location where, if a park were built, the total benefit to all those who'd like to use the park plus those who'd get to view the park would outweigh the cost of the park's construction.

Suppose the following is true about the park:

Cost of constructing park: $1500

Benefit to those who'd like to play on the park: $1000

Benefit to those who'd get to view the park: $1000

Suppose that while a private park builder would be able to charge those who'd like to enter and play on his park, he couldn't reasonably charge those who'd have the pleasure of viewing it. Thus, the park-viewers would be free-riders, as the park would produce for them a positive externality. The government-interventionist would claim that the park should be built-- as total benefit outweighs total cost-- but won't be built, as the benefit received by the park-builder would not outweigh the cost he bears to construct it.

However, the park-builder has incentive to capture the benefit that would be received by viewers, if he can-- and as a matter of fact it's quite simple to do: first, he buys the surrounding houses-- perhaps under different names so nobody suspects what he's up to. Then he constructs the park in between them, and then he sells the houses. By constructing the park, the houses the park-builder owns have all appreciated in value, as their future residents will have the benefit of viewing the park. The benefit of viewing the park is in this way captured, and the procedure described should take place.

One can contend as to the added hassle of having to buy and sell a bunch of houses just to build a park, but we should be mindful that this will likely be a common occurrence, and park-building agencies would adapt to specialize at house purchase and sale.

It's not a perfect solution, but it seems quite alright, contrasting to those which involve the government. If we subsidize parks, how much should they be subsidized? No one knows. Do we subsidize different parks at different rates, depending on how much they receive from being viewed, rather than played upon? How could we-- who makes that calculation? Who has the incentive to? Further, how do we prevent park-buiding agencies from lobbying for higher subsidization? If we nationalize parks, how would we know where's best to build them? Or what the optimal size? How do we stop larger cities from voting in bigger and better parks based solely on the fact that they have more votes? Are we considering the negative externality of public parks being a convenient place for criminal activity? Who calculates the optimum rate for police to investigate the area?

Example 2: A school educates children, making them less likely to become criminals.

Example 3: A factory emits air pollution, making breathing a cumbersome chore in its city.

Example 4: A car emits carbon dioxide, furthering global warming.

But schools also turn people liberal.

But air pollution makes for prettier sunsets.

But global warming has perks.

Here's where "ignores whichever half of the equation is ignored by the public" comes into play.

In theory, to make for a sufficient Pigouvian tax/subsidization, or to justify government intervention in any form, the following calculation should take place:

Positive effects of externality - Negative effects of externality = x

Then use x, if positive, to determine the level of subsidization, or, if negative, the level of Pigouvian tax. Or whatever.

But really, the calculation is usually made one-sided (and badly). Whichever way conforms with public opinion, because that's the only measure that the government needs to stand up to. And public opinion sucks. No one has incentive to learn, in detail, what aspects of a given externality might contribute to the positive or negative welfare of people, or by how much. No one gets a raise for learning the full economic impact of raising the Earth's temperature, nor does anyone have even the capability to measure the disaster that strikes when a generation of kids in school fall in love with being subdued by their government, who fall in love with using its power to steal from whatever perfectly good people some crazed environmental nut decided have a ruinous effect on the non-existent Gaia, who fall in love with meaningless phrases and abstract intellectual nonsense that takes a mystical precedence over the welfare of human beings.

My position on pollution is to treat it as a tort liable for sue-- a legal measure perfectly capable of being enacted by private law in the absence of state, and by which government agencies have no need to take and waste perfectly good dollars on some detestably-useless firefight in Afghanistan.

I invite those with questions to comment.