Earlier, I wrote about Lemon Laws. These laws exist to solve the problem of cars which require "too many" repairs. The problem is not that the seller does not honor a warranty. The assumption is that the existance of problems -- even repaired problems -- is evidence of the existance of more problems. You can see this in the computer software field. Some programs are written securely and never have any security problems. Other programs were written without security as a goal, and trying to bolt on security later proves difficult.
So, the problem exists. The question for an economist is what to do about it. I can answer that question (and I may be right or wrong), but the process of answering it is more interesting than the answer itself. Leo writes to tell me that car dealers often try to cheat their customers. That fact (which, really, surprises nobody) is almost completely besides the point.
First, interesting economic things happen in the middle, not the edges. It's not likely that any businessman is completely honest or completely dishonest. Instead, some businessmen will be more honest than other businessmen. So, saying that car dealers often try to cheat their customers says nothing. The real question is whether more honest car dealers make more money than dishonest car dealers. If they do, then the tendency will be for honest car dealers to out-compete dishonest ones. If they don't, then there's clearly a business opportunity for someone.
Every trade in a free market generates a surplus in value for both parties. You could argue that a "fair trade" is one in which the surplus is equal. Most often the surplus isn't equal. Sometimes it's a seller's market, where the seller sets the price, for example in an emergency. Any time the surplus (on either party's side) is large enough, it will attract more entrants into the market. This results in competition and tends to reduce the surplus.
Markets work best when people buy things often, and when there are multiple suppliers. This isn't the case when you're buying a car, which people may do only five times in their life. Automobile companies try to reduce competition by only establishing a limited number of dealerships in a region. As a general rule, competition works well because experts are competing against experts. You may not know how to bottle up soda, but the experts are coke, pepsi, and many other cola suppliers do. By competing against each other for your business, they keep each other honest. That competition is lacking in the automobile market.
There oughtta be a law!
You've no doubt hear the clarion cry "There oughtta be a law!" uttered in response to some inequity or another. You shouldn't be surprised to hear me question that request. Laws and markets work in very different manners. For any arbitrary problem that a market has trouble solving, it may be that a legal solution works better or worse. Far too many people assume that a law can solve any problem which is not solved to their satisfaction by markets.
For example, you could attempt to solve this problem by passing various laws related to automobile quality representation and guarantees. Or you buy insurance from a company which will buy your car if it needs too many repairs. Since the company doesn't want to lose money, it will charge you more to insure a car from a company which often makes lemons. You could use that as a clue not to buy that company's cars, or if the cars are special, just knuckle down and pay the higher premiums. Such a company doesn't exist right now. That's because the lemon laws have legislated such a company out of existance. Or, at least, they ensure that no such company could make money.
The biggest problem with passing a law is that you lose information. Since everybody has to comply with the law, there is no way to find out if a more or less strict law would work better. With a market solution, you have competition to provide the best solution. People are free to experiment with different solutions.