Pareto efficiency (or Pareto optimality) is an important efficiency concept in welfare economics used to evaluate or compare different allocations of resources. It is named after Italian economist Vilfredo Pareto (1848–1923).

An allocation of resources is Pareto efficient if it cannot be modified to increase the wellbeing of one individual without diminishing the wellbeing of any other individual. If it is possible to reallocate resources and improve the welfare of one person without harming anyone else, this reallocation is called a Pareto improvement, and consequently the initial allocation was not Pareto efficient.

Sometimes we distinguish between Pareto efficiency in the allocation of resources and pareto efficiency in the production of goods and services. The concepts are very similar.

Pareto efficiency in consumption (i.e., in the allocation of resources) means that no one can be made better off without making someone else worse off. Pareto efficiency in production implies that we cannot increase the production of one good (or service) without decreasing the production of another good or service.

Let’s look at an example to better understand what a Pareto efficient allocation is: suppose you’re walking together with a friend along a beach when you discover two beautiful seashells laying there. You decide to keep one for yourself and give the other to your friend, so that each of you has one seashell. Is this allocation Pareto efficient?

To answer this question let's consider whether we can reallocate resources to improve your welfare without harming your friend. If we want to increase your welfare, we have to give you your friend's seashell. But this means that we have to take away the seashell from them and their wellbeing will decrease.

So, taking away your friend’s seashell makes them worse off and is not a Pareto improvement. Similarly, it is not a Pareto improvement if we take your seashell to give it to your friend. Given that we can not reallocate the resources in this example without making someone worse off, the initial allocation was Pareto efficient.

What if instead of picking up both shells, you only pick up one and leave the other on the beach? Is this Pareto efficient?

For simplicity, let's assume that no one else would come to pick up the seashell and also no other living being would use it. Can we reallocate the resources and improve efficiency?

In this case, the answer is clearly yes. If instead of leaving the seashell on the beach, your friend picks it up, his or her wellbeing will increase without reducing the wellbeing of anyone else. This means that the initial allocation was Pareto inefficient - and leaving a seashell on the beach is not Pareto efficient.

Importantly, Pareto efficiency does not tell us anything about the fairness of the distribution of resources. For example, suppose that the entire wealth of a country is given to just one single citizen, while all the other citizens do not obtain anything. Is this allocation of resources Pareto efficient? Yes, it is. Why?

In order to see the answer, think about whether it is possible to reassign the resources in a way that the wellbeing of one person increases without making anyone else worse off. Is this possible?

Obviously not. If we want to give something to the other citizens, we have to take resources away from the rich person. That is, in order to increase the welfare of another individual, we have to reallocate it from the rich person, and this means that their utility will decrease. Since we can’t make anyone better off without harming the rich person, the initial distribution was Pareto efficient. Coming back to our seashell example, an allocation in which you decide to keep both seashells that you found is Pareto efficient as well.

Sometimes, the analogy of a cake is used to explain the difference between equality (or equity) and efficiency. The size of the cake can be measured by efficiency. If we can increase the size of the cake and people can have more cake, this clearly is an efficiency improvement.

The size of the cake however does not tell us anything about how the cake is being distributed. To decide who should get what is difficult as this usually involves some judgment about what is fair or just. What would be a fair way to split the cake? Should everyone receive a piece of equal size? Should the person who made the cake get a larger piece? Should we give a smaller person a smaller piece?

As you can see, there are many considerations that have to be taken into account when deciding on how to distribute the cake and asking three different people might get three different opinions. Yet, if someone invents a new technology that can convert the same amount of ingredients into a larger and equally tasty cake, then no one would object to doing so.

Evidently, Pareto efficiency is a concept that is less controversial than fairness. What seems fair or just to one person does not necessarily seem fair to another person. For these reasons, economists usually focus their analyses on efficiency.  

Further reading

In the seashell example used above we said that an allocation in which you keep both seashells and your friend has none is Pareto efficient. What happens if your friend gets jealous of you having two seashells while they don’t have any?

Beckman, Formby, Smith and Zheng (“Envy, malice and Pareto efficiency: An experimental examination”, Social choice and welfare, 2000) used an experimental approach to analyze the role that envy and malice play in economic decisions. The authors found that envy and malice are strong motivations that can lead to reduced support for Pareto efficient outcomes.

Good to know

What would Vilfredo Pareto have to say about Robin Hood, who stole from the rich to give to the poor? He clearly would disapprove of such a reallocation of resources based on efficiency considerations. To increase the welfare of the poor, Robin Hood is willing to decrease the welfare of the rich, which means that the reallocation is not a Pareto improvement. Of course, Robin Hood would probably not use Pareto efficiency as an argument to justify his deeds.