If a government enacts a policy that messes with the workings of an otherwise efficiently working market, then it might be necessary to enact another policy to, at least partially, counteract some of the inefficiencies created by the first policy. This is known in economic circles as the Theory of Second Best. Three things should be noted about the Theory of Second Best.
First, second best is never first best. Once a policy creates a market inefficiency, the only way to return to full efficiency is to remove the policy that created the inefficiency. Creating a second policy to fix the problems caused by the first may bring us closer to efficiency than leaving the first policy to act alone, but we will never return to full efficiency.
Second, second best is NOT an argument against all government intervention in markets. Policies designed to correct market failures (like environmental externalities) can lead us to full efficiency by removing a problem with a market allocation. Fixing a market failure that wasn't created by inefficient policy, is the first best solution*.
Third, the theory of second best creates the possibility of a vicious policy cycle, where policies are enacted to counteract the negative effects of other inefficient policies which were initially enacted to counteract the effects of other inefficient policies.
Mexico is in the grip of the worst tortilla crisis in its modern history. Dramatically rising international corn prices, spurred by demand for the grain-based fuel ethanol, have led to expensive tortillas.
Why is this a problem?
Poor Mexicans get more than 40 percent of their protein from tortillas, according to Amanda Gálvez, a nutrition expert at the National Autonomous University of Mexico.
The typical Mexican family of four consumes about one kilo -- 2.2 pounds -- of tortillas each day. In some areas of Mexico, the price per kilo has risen from 63 cents a year ago to between $1.36 and $1.81 earlier this month.
Mexico's minimum wage is $4.60 PER DAY!
So how does the Mexican government respond--keeping in mind that Felipe Calderon, Mexico's president is "an avowed supporter of free trade"?
On Jan. 18, Calderón announced an agreement with business leaders capping tortilla prices at 78 cents per kilogram, or 2.2 pounds, less than half the highest reported prices. The president's move was a throwback to a previous era when Mexico controlled prices -- the government subsidized tortillas until 1999, at which point cheap corn imports were rising under the NAFTA trade agreement.
All together now: "NOOOOOOOO!!!! Price caps are bad policy. Price caps create inefficiencies."
But what if the Mexican Cap Stance (come-on, that's funny) is a response to inefficient policies enacted elsewhere. Say, policies that encourage the use of corn-based ethanol as a gas additive in the U.S.? or say, a policy that pegs yellow corn prices in Mexico (the corn used for tortillas) to international white corn prices (the corn used for ethanol)?
If the price cap is a response to another inefficient policy, then the price cap may actually improve efficiency. The first best solution would be to remove the policies creating the inefficiently high corn prices. The second best solution might be to create a new policy to counteract the effects of bad policy. That's the Theory of Second Best.
*As a side note, Wikipedia's definition of second-best is wrong on this count. Here is the Wiki definition:
The Canadian economist Richard Lipsey (1928-) and the Australian economist Kelvin Lancaster (1924-1999) showed that if one optimal condition in an economic model is not satisfied, it is possible that the next-best solution will involve changing other variables away from their previous optimal ones. This means that in an economy with some unavoidable market failure, it may be optimal for the government to intervene.
The last sentence is incorrect. If the market failure is unavoidable, then intervention is the first-best solution. If the market failure were AVOIDABLE, then the second-best solution would be intervention. The first-best solution would be avoidance.