When evaluating proposed or existing policy measures, economists & government officials typically aim to improve the lives of others. The objective of any given tax, welfare, or law change is to improve the lives of many citizens while not worsening the lives of others. The most ideal policies improve the lives of others without making anyone else worse off. This is what we refer to as Pareto efficiency.
Something Pareto efficient makes one person better off without making anyone worse off. The state of Pareto efficiency is one in which it is impossible to make someone better off without making anyone else worse off. This would indicate that in this state resources are allocated perfectly.
So, where could we find Pareto efficiency?
Well, lets say you are babysitting 2 children, and have 4 candies. You can allocate these candies a number of ways.
Child A gets 0, Child B gets 4
Child A gets 1, Child B gets 3
Child A gets 2, Child B gets 2
Child A gets 3, Child B gets 1
Child A gets 4, Child B gets 0
or, you could keep them all for yourself.
Now, if you are me, you would want to keep these all for yourself. Good choice! I'm sure you can guess which allocation is Pareto efficient - both kids get 2 candies. This allocation of candy is Pareto efficient because if we moved up or down either way, one child would be worse off.
It is commonly accepted that Pareto efficiency should be a criteria for any given public policy or economic system. While not every public policy will end up being good for everyone, the idea is to find the lesser evil within your options. Pareto efficiency does not equal desirability, it just means that you have found the most efficient way to deal with something.
Where else can you hear about Pareto efficiency?
The field of welfare economics discusses Pareto efficiency quite often. A theory called the first welfare theorem says that free markets (that is, those with perfect information, no barriers to entry, etc) accomplish Pareto efficiency on their own.