(Taken on Abovian Street in Yerevan, Spring 2007, by E. Claar.)
George Akerlof won the 2001 in the area of asymmetric information: when two parties do not equally share all of the relevant information available to their decision-making situation.
Most of the time, in a market of any kind, both potential trading parties know exactly what they will be getting from the exchange. When you buy a cup of coffee, especially from a vendor you know and trust, you know what you will be getting (the coffee), and the barista knows what she will be getting (your money). Moreover, once the money and coffee have been handed over, both of you are the better for it. You obviously wanted (needed?) the coffee more than your money, and the barista was much more interested in your money than in keeping the coffee for herself.
Economists love these kinds of interactions, by the way. Two people are now happier (you, with your hot caffeine jolt) and the barista (with your money, and hopefully a nice tip besides). As evidence, the two of you probably both say, "Thank you." And life goes on for the rest of us, in our ignorant bliss.
But there are some markets where potential buyers and potential sellers do not both have access to all of the information relevant to a potential transaction. That's a case of "asymmetric information" (see above). And where it is found, some potentially mutually-beneficial trades now prove impossible (and ones that would certainly have transpired in a world where everyone knew everything).
Think about that cup of coffee again. Suppose you are not buying it at your favorite coffee shop in your hometown, but instead from a stranger in another city. And not at some mega-corporate-branded place like Starbucks. How would you know what you'd be getting? Or that it really would be worth the $4.50 being asked of you?
Used car markets work (or, rather, don't work) like this. Buyers want reliable transportation because they need to go places. On the selling side of the market, though, there are always two kinds of cars: good ones (I'll call them "cherries") and bad ones ("lemons"). And only the sellers know which variety they are attempting to unload.
So all of the sellers try to price their cars as though they are cherries--whether they are or not. Now, consumers don't know which specific cars are which, but they are certainly aware that there are lemons out there to be avoided. Knowing this, there is no way that they will be willing to pay the full asking price for any of the cars, because any car they buy could be a lemon.
Now think of the unfortunate, honest, cherry owners. They might be asking for $9000 for a car that really is worth $9000. But nobody out there is willing to pay $9000 for any car, even a cherry, for fear of ending up with a lemon. So the cherry owners give up, and pull their cars out of the market.
For economists, that's a terrible shame. A win-win trade will not take place because of the uncertainty the potential-buyers face, given they have less info than the sellers do.
Akerlof won the Nobel largely for his work in exactly this situation, called the "lemons problem." You can read more about it--and its inevitable conclusion--at the Nobel site.