An individual's willingness to pay for a good not only depends on how much they value that good, but also on their income level (at least under the conventional non-economics definition of the word 'value').

Consider the case with a single indivisible unit of a good and 2 consumers, a very wealthy consumer and a very poor consumer. Even if the poor consumer would benefit more from receiving the good than the wealthy consumer (let's say the good is a drug that the poor consumer medically needs but the rich consumer wants to take recreationally), the 'efficiency-maximizing' free market will still give the good to the wealthy consumer under most conditions. The wealthier consumer has a higher willingness-to-pay not because the good is worth more to them, but because a single dollar of their wealth is worth less.

The normative idea that a good should always go to the individual with the highest willingness to pay is then surely flawed. The policy objective of maximizing surplus, which follows from this idea, must also be flawed. Given how glaring this problem is, I am surprised to see so many papers that still use WTP-derived consumer surplus in their normative analysis.

My questions are:

-What does the literature have to say about this issue?

-Have alternative measures of surplus been proposed?

-Is my reasoning flawed/is the measure of CS not as unfair as I am suggesting?

Partly inspired by some of the arguments in this thread: What does economics say about "price gouging" during an emergency?


2 Answers 2


Consumer surplus in Microeconomics 101 is indeed often used in the way you describe and to illustrate a point about the allocation efficiency of a free market.

However, it is important to keep in mind the assumptions and limitations of the framework you describe.

First, the case you describe is a partial equilibrium case. This framework does not allow for such general statements as you are making about all goods always going to richer people. In partial equilibrium we look at one small market and forget about everything else. To use such a framework we must accept that it is reasonable for our question at hand to analyze one purchase in isolation. That assumption is clearly not reasonable for the question you have on all goods and as such the framework you describe cannot be used as it will lead to false conclusions.

With partial equilibrium we can look at this purchase in isolation from other markets. If income effects and the purchases in question are important then we cannot use the consumer surplus idea as done in Micro 101. Certainly this is an issue for the point you raise about all goods always going to a wealthier consumer. All goods cannot be analyzed with partial equilibrium.

The good allocation properties that economist often talk and that you mention about are based on general equilibrium and the proofs do not really involve the concept of consumer surplus.

To fully ensure avoidance of income issues, we often work with quasilinear utility functions when talking about consumer surplus and allocation, as you are doing. It is important, however, to keep in mind that these utility functions are special cases. They imply that income effects do not matter so they are widely used in Micro 101 for illustrative purposes.

Consumer surplus has a nice interpretation for quasilinear utility functions: it’s equal to the gain in utility. This makes quasilinear utility functions especially instructive. If everyone’s utility function has this form then the Pareto efficient allocations are exactly the allocations that maximize total consumer surplus (in the absence of production).

So when using consumer surplus to talk about efficient allocation you need to accept the assumptions that go along with quasilinear utility. However, such utility function are only realistic for certain goods and certainly not for all goods.

Luckily, what you learned about free markets achieving allocation efficiency is still true even when considering more complex cases. However, these results are then true for different reasons and under different assumptions than what is taught through consumer surplus in introductory lessons.

  • $\begingroup$ You're correct that my original question is about partial equilibria, I'm not sure what you're referring to when you talk about "general statements" that I make. But could you elaborate on how the (normative) desirability of market outcomes is measured when studying GE? The only such measure I have been taught is whether an allocation is Pareto Optimal, which is obviously a very weak measure that is of next to no use in any real-world application. $\endgroup$
    – H Rogers
    Commented May 21, 2020 at 23:34
  • $\begingroup$ So that would go beyond the scope of a single question. But what I mean with your general statements are the statements that any good should always go to the individual with the highest willingness to pay. Any statement with the word "always" is general. Further, you say "The policy objective of maximizing consumer surplus, which follows from this idea, must also be flawed.". That statement is general & false. The policy objective does not follow from there. Simply the CS idea is sometimes in line with that policy objective under certain circumstances relevant to introductory economics. $\endgroup$
    – BB King
    Commented May 21, 2020 at 23:44
  • $\begingroup$ At the risk of descending into pedantry, I feel forced to say that the negation of a general statement is not itself a general statement. "Ben always eats cake" is a general statement, but "Ben does not always east cake" is not, because we must only observe Ben's behaviour in one specific instance to be justified in making it. Furthermore, I'd point out that the objective of maxmizing CS is not simply an instructional artifact but a very real idea present in many current papers as well as the broader econ-pundit rhetoric. Do you have a justification for this goal? $\endgroup$
    – H Rogers
    Commented May 21, 2020 at 23:51
  • $\begingroup$ My point is simply that CS maximization can imply allocative efficiency of free markets, but it is not a necessary condition. The reverse implication is not generally true. Hence finding a case that invalidates CS leading to an efficient allocation does not imply that markets do not achieve efficient allocation. I am not aware of any current research papers that rely on CS for these points. I am also not qualified to speak about econ pundits. CS as used in your example serves more illustrative purposes than cutting edge scientific purposes as I understand it. $\endgroup$
    – BB King
    Commented May 22, 2020 at 2:20
  • 2
    $\begingroup$ The moral grounding is as follows: The framework you discuss is only appropriate when income effects can be ignored. Your question discusses the (moral) failings of this particular framework as a result of income effects. I hope this helps. $\endgroup$
    – BB King
    Commented May 22, 2020 at 12:31

First, in all the equilibrium analyses, you should consider both consumers and producers (There are special cases like pure exchange economy, which is simplified to illustrate some points). So normally, the policy objective is not to maximize consumer surplus only but to maximize consumer surplus plus producer surplus.

Second, the equilibria described in Micro are usually specified by a price vector, a production vector, a consumption vector, and a transfer (equivalent as a wealth redistribution/assignment/allocation). The transfer here is specifically important to your questions. With a proper transfer among the rich, the poor, and the producer (of the drug), the problem can be solved together with the assumptions in the (price, Walrasian) competitive equilibrium.

As you have said that Pareto optimal is actually the minimum requirement for an equilibrium outcome. An economy can have extremely high inequality among people while still be Pareto optimal. The Second Theorem of Welfare Economics solves this problem by proving that any Pareto optimal outcome can be achieved by a price equilibrium with transfers. (The Frist Theorem says that every equilibrium is Pareto optimal just for your reference.) However, there are several problems with this theorem.

At that time, the assumptions used in achieving this result were considered mild and reasonable. However, the assumption about complete information seems to be very unrealistic right now. Incomplete information, as well as asymmetric information, is present in every corner of the economy. There are so many to say on this issue.

Also, the assumption of a competitive market is also unrealistic. Oligopoly is everywhere in the economy. As the example you mentioned, the pharmacy industry is definitely going to be an oligopoly market. In order to extract the most profit from the economy, they won't choose the production bundle that is best for all the people in the economy, so there will definitely be a shortage of new drugs.

Back to your question, first, it is not even a pure exchange economy in your example, and you miss the role of producers. The policy objective is not to maximize consumer surplus only. Second, under a set of assumptions, a proper transfer together with a proper price vector, a proper production bundle, would constitute the desired Pareto optimal outcome. However, how to get to that place? How to eliminate monopolistic market power while making the firms have enough incentive to innovate at the same time? Those are tough questions, and the role of information is also important in answering those questions.

  • $\begingroup$ For the sake of simplicity my question takes production levels as given, and examines the goal of maximizing consumer surplus. You could generalize to a model with production and look at overall surplus and the key point would still hold. You refer to a "desired Pareto optimal outcome". How would an economist evaluate which outcome is the desired one? $\endgroup$
    – H Rogers
    Commented May 22, 2020 at 12:26
  • $\begingroup$ Based on whatever principle one thinks appropriate, one can find the corresponding desired Pareto optimal outcome. The equilibrium is characterized by an allocation, not by whatever the economy is endowed with. So, in your example, a transfer from the rich to the poor would constitute an equilibrium which is also Pareto optimal. Taxation is serves this purpose of transfer/redistribution, but it has never been optimal nor perfect due to many reasons. $\endgroup$
    – Lin Jing
    Commented May 22, 2020 at 12:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.