screws. Or wooden pegs or glue or dovetail joints or really good balance.

Point being there is nothing per se wrong with subsidies. While I agree they tend to have adverse consequences and should be used sparingly, that's true of unconstrained price controls, traditional customs, haruspication, or just about any other system of distribution. When market signals lead to conditions unfavorable a large enough or influential enough group of people, then government intervention is a perfectly legitimate tool for addressing the problem. Subsidies are no more or less efficient than price signals -- depending of course on what you consider efficient. If buggy whips are an important cultural product, then subsidies are a highly efficient way to preserve them against encroachment from louder, smoggier forms of travel.

The example of protectionist laws for traditional European regions/products/ways of life has been given. Palaces and castles make damn poor living quarters by comparison with structures built out of modern materials, but they embody principles & heritage the local culture feels is worth a few drafty rooms.

It's best to think of governments in economic terms as big purchasers. Wal Mart is a behemoth that makes purchasing decisions based on maximizing its own interests; so do governments at an even larger scale. If what a government wants to "buy" is preservation of parkland or traditional methods of farming (or farming altogether from being exported), so long as the government has an abiding interest (voter support) in doing so, it is a perfectly efficient and rational decision.

So much for the abstract. This entire issue quickly becomes entangled in rentseeking, and these efficiencies go out the window.

Especially at US Dept of Agriculture, I know more about subsidies than are strictly good for anyone.

The problem isn't the subsidy per-se. Others have noted the caveats in their comments below. Subsidies are naked quid pro quo. The Farm Bill of 2002 was a blatant payoff for those who voted for the Iraq War. Only the politically connected get subsidized. When we took the agribusiness complex off their subsidies, they whined to their Congresscritters, arguing they could not compete with other countries who subsidized their farmers. There's a note of truth to that, but it's the outfits like Conagra who benefit most from ag subsidies.

What is the practical difference between a subsidy and a tax loophole? I see none. Those who wax wroth, sputtering about subsidies should view the tax loophole as a subsidy turned inside out.

[Redacted until later.]

The theory of comparative advantage is excellent on the micro level and terrible on the macro level. Here is why:

1) Time. Comparative advantage is a discussion of a snapshot in time and how specialization can produce tremendous results during that period. However, it does not handle macro effects on an economy of this sort of specialization, such as growth, risk, externalities, morality, and transition costs.

1a) Growth. If one specializes into a low-growth set of industries, then one cannot take advantage of high-growth sets of industries. Furthermore, it may be that a certain mix of industries (financial, service, and manufacturing) is the optimal approach for engendering productivity improvements. We simply don't know.

1b) Risk. If you put all your eggs in one basket, you can often get a great basket, but woe betide you if it slips. This is a bigger deal for nations that aren't as enormous as us, but if you bet the farm on only a few industries, you are massively increasing the probability that you will lose it. I am consistently baffled by economists that recommend that entire nations take on levels of risk from failure to diversify that they would never recommend for individual investors.

1c) Externalities. What is the cost of the disappearance of middle class status for persons with high school diplomas and less? That's half of our adult population. There are enormous political costs created by that transition, including the destruction of a political center capable of muddling through between the radicals on both sides.

1d) Morality. It is, in a sense, immoral for us to support a system built on slave (or near-slave) labor. If specialization moves a tremendous amount of work over to a location which is particularly brutal, that would be an issue.

1e) Transition costs. Yes, once one is specialized, one enjoys tremendous gains. But the world economy is fantastically dynamic; what is your clear area of specialization today may not be what is your clear area of specialization in ten years. That's a lot of capital, education, and human expertise you just made useless. But eventually, GDP will go up again! Unless you transition again . . .

All of these, of course, can be mitigated by various policies. But they serve as counterweights to Ricardo's simple partial equilibrium analysis.

I see nothing at all wrong with the Government helping the market invest in places it normally wouldn't. Revitalizing an old mill building might be too much for the market to do on its own but with gov't help that revitalization might lead to the revitalization of a whole section of a city. I have seen it happen in Lowell, Massachusetts. Lowell is still a bit of a sh!thole but it is about 567% better than it was 20 years ago. I doubt that improvement would have happened w/o gov't help.

The problems with subsidies arise when they stop benefiting the community as a whole and start benefiting specific individuals/organizations. If politicians had the stones to end subsidies when they stopped producing the benefits they were intended for then I don't think we would be having this discussion. IRL however politicians don't have the political will to administer subsidies properly.

One of the things I find amusing about the global warming deniers, and those who claim starting to move to a post-carbon energy system will "wreck the economy" is that they sound exactly like those buggy whip manufacturers you describe.

I agree with dionysus & Wagster that there definitely are such things as "good subsidies". Some subsidies solve problems that are insoluble by the market: two examples would be investments that are too long-term and high-risk for the market to take on by itself (like medical research or space exploration), and payouts to defuse the tragedy of the commons (for example, incentivizing loggers not to cut old-growth forests, or consumers to buy fluorescent bulbs).

In general, though, I agree. "Subsidies", as the term is usually applied, are usually bad. Protectionism is pretty much always bad. (Though reciprocal protectionism can be justified.) Ag subsidies are downright awful, have distorted the American economy in very ugly directions, and will never go away.

I agree with you. But...

Sometimes the market does not recognize some values. In general, I am opposed to agricultural subsidies, but I have to have some sympathy for the French when they argue that they are protecting the nature of their countryside, their essence as a country. (The American version of this argument tends to just be a bid for sympathy for the family farmer, who proportionately isn't a big beneficiary of farm subsidies.) Anyone who's spent time in the French countryside knows there's a wealth there that isn't factored in land values. It does translate into $ in tourism, for instance. But even if it didn't, it still enhances French lives.

(And as I write this I'm planning a visit to Burgundy in the summer.)

Subsidies are a bad idea. PERIOD.

The debate isn't really on economics of it. They are bad. No way around it.

The problem is in the dynamics of government and how money is allocated and kept that way. By that I mean that vested parties are more likely to organize and entrench themselves while the public at large cannot and will not. Small groups have a insurmountable advantage to the semi-apathetic public.

Funny, not to plug the book again but, The Logic of Life actually covers the political economics of this in depressing detail. I have little hope.