The following question was given as a part of a task in microeconomic theory course. It is not from some textbook and since I still haven't figured a way to solve it I will leave it here. Thank you in advance.

Let $f:\mathbb{R}_{+}^2\to \mathbb{R}$ be a single output production function such that

$$f(z_1, z_2) = (z_1^\delta+z_2^\delta)^{\frac{1}{\delta}},\quad \text{with $0<\delta<1$}$$

Let $X = \mathbb{R}^3$ be the consumption set. The vector of commodities $(z_1, z_2, q)$ is a typical element of $X$. There are $N=\{1, 2, \dots, n\}$ individuals (with $n\geq 2$) where the typical individual is denoted by $i$. Each $i$ has a preference relation $\succsim_i$ defined over $X$ and the utility representation of $\succsim_i$ is given by $u_i(z_1, z_2, q) = i(z_1+z_2) +q$. Individual $1$ owns the technology, but has an endowment $\mathcal{E}_1 =(0, 0, 0)$. For every $i\in N-\{1\}$ and some pair of non-negative real numbers $a_1$, $a_2$, $\mathcal{E}_i =(\frac{a_1}{n-1}, \frac{a_2}{n-1}, 0)$.

Which is the competitive (Arrow-Debreu) equilibrium for this economy?

I believe it will be easier if we set $n=2$ to solve the problem.

  • $\begingroup$ Should it be $X=\mathbb{R}^3_+$, or is negative consumption really allowed here? $\endgroup$ Commented Jan 8 at 22:37
  • $\begingroup$ As it is given, it is $X = \mathbb{R}^3$ $\endgroup$ Commented Jan 8 at 22:39

2 Answers 2


There will be no equilibrium. Both technology and endowment play no role in the argument.

Assume there would be an equilibrium with price system $(p_1,p_2,p_3)$.

Notice that all prices must be strictly positive since all goods are desirable.

The optimality condition of agent $1$ implies that $p_1=p_3$. If, say, $p_1<p_3$, agent $1$ could always improve by consuming one unit less of good $3$ and consuming $p_3/p_1$ units more of good $1$. Since negative consumption is allowed, this would always be feasible and an improvement.

Similarly, we must have $p_1=2 p_3$. Otherwise, agent $2$ could improve.

All in all, we get $0<p_1=2p_3=2p_1$, which is clearly absurd.

  • $\begingroup$ Sorry.... I don't follow...you write price system $(p_1, p_2, p_2)$ and you say optimality condition of agent $1$ is $p_1=p_3$.... and what do you mean by optimality condition? Please, could you explain a bit more? $\endgroup$ Commented Jan 8 at 23:20
  • $\begingroup$ If a bundle is preferred by an agent to what they consume, it must cost more. That is all I am using here. If it helps you, you can also argue that equality of marginal rates of substitution is necessary here since there are boundary constraints. $\endgroup$ Commented Jan 8 at 23:24
  • 1
    $\begingroup$ Yes, that would change things a lot. That's why I asked. $\endgroup$ Commented Jan 8 at 23:38
  • 2
    $\begingroup$ The firm maximizes profit. Who owns it is irrelevant at that stage. This is just tedious undergraduate microeconomics. $\endgroup$ Commented Jan 8 at 23:55
  • 2
    $\begingroup$ Profit maximization has nothing to do with ownership structure. The firm maximizes profit. Agent 1 consumes. Their income is the firm's profit, but the decisions are made separately. And yes, it should be $p_3$. $\endgroup$ Commented Jan 9 at 0:04

Assuming that the economy has two agents, $n=2$ and since individual $1$ owns the production, this means that the firm belongs to agent $1$ who has no endowment.

Suppose that the firm produces the commodity good $q$ by using as inputs the commodities $z_1$ and $z_2$. Then the problem of agent $1$ is

$$\text{max}_{(z_1,z_2,q)}\{z_1+z_2+q\},\quad\text{s.t. $p_1z_1+p_2z_2+p_3q\leq \pi_f$}$$

Where $\pi_f = p_3f(z_1,z_2)-(p_1z_2+p_2z_2)$ is the profit function and $f(z_1,z_2)=q$

Also the problem of agent $2$ is

$$\text{max}_{(z_1,z_2,q)}\{2(z_1+z_2)+q\},\quad\text{s.t. $p_1z_1+p_2z_2+p_3q\leq p_1\alpha_1+p_2\alpha_2$}$$

By solving the lagrangian for agent $1$ and agent $2$, we have that $p_1=p_2=p_3$ and $p_1=p_2=2p_3$ respectively. The only vector of prices that satisfies both of the latter conditions about the prices is $(p_1, p_2, p_3) = (0,0,0)$

Then (either) there is not an equilibrium for this economy, because prices must be non-negative and at least some of them strictly positive.

Or else the only equilibrium that satisfy the given economy is the one that implies a zero profit condition (which I am not sure if this is just equivalent to the fact that $(p_1, p_2, p_3) = (0,0,0)$)

  • $\begingroup$ Using the same variables for both agents' consumptions and the firm's inputs is a bit confusing. $\endgroup$ Commented Jan 21 at 12:34
  • $\begingroup$ Yes, but it saves time to the paper in order to solve the problem :P Do you want me to change the notation? $\endgroup$ Commented Jan 21 at 12:35
  • $\begingroup$ It does ignore the firm's problem. If you want to try the case in which consumptions have to be nonnegative too, this needs to be done. $\endgroup$ Commented Jan 21 at 12:37
  • $\begingroup$ yes, the firms problem does not play much of a role. Do I miss a case? $\endgroup$ Commented Jan 21 at 12:40
  • 1
    $\begingroup$ The demand of the consumers will look different if nonnegativity constraints bind and then the actual amount of profit can matter. $\endgroup$ Commented Jan 21 at 12:53

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.