Aram Grigoryan Econ 100B, Spring 2023
Part III
Profit Maximization
Once we derive a firm’s cost functions, we can simplify its profit maximization problem to a one-
variable optimization problem:
max ⇡(q) = max R(q) c(q),
q q
where ⇡(q) = R(q) c(q) is the profit function, and R(q) = pq is the revenue from selling q units
of final product at price p.
We will study the firm’s profit maximization problem in the short-run and the long-run. The
first (obvious) di↵erence between the short-run and the long-run profit maximization problems, is
that in the former we use the short-run cost function cSR (q), and in the latter we use the long-run
cost function cLR (q). The more subtle di↵erence between the short-run and the long-run is that in
the long-run the firm has the option to shut down the production entirely, and guarantee a profit
of zero. In the short-run, the firm produces even when profits are negative.
We will first study the firm’s profit maximization problem maxq ⇡(q), without specifying
whether we are in the short-run or long-run, and then, we will study the firm’s optimal sup-
ply in the two time horizons separately.
First- and Second-Order Conditions for Profit Maximization (IMVH
D3.a,b,c, Perlo↵ 8.2)
Consider the optimization problem
max pq c(q).
q
This is an optimization problem with one variable, for which there is a clear recipe for optimization.
Namely, we can use the first- and second-order conditions to find the local maximizers. In general,
we will then need to compare the value of the function at the local maximizers with each other, and
with corner solutions, to find the global maximizer. The functions that we study are well-behaved,
so that there will be a unique point that satisfies both the first- and second-order conditions. Then,
we will compare this candidate solution to the corner solution (producing q = 0, or shutting down)
to get the globally optimal solution.
Numerical Example: Profit Maximization
Let’s study profit maximization with a numerical example.
Suppose the price of the good is p = 4, and the cost function is c(q) = q 3 6q 2 + 12q. Then, the
profit function is
⇡(q) = 4q q 3 + 6q 2 12q = q 3 + 6q 2 8q.
1
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
We get the candidate solutions for maxq ⇡(q) by looking at the first-order condition (FOC).
⇡ 0 (q) = 0 () 3q 2 + 12q 8 = 0.
This quadratic equation has two solutions:
p p
6+ 36 24 6 36 24
q1 = ⇡ 0.85 and q2 = ⇡ 3.15.
3 3
Among these candidate solutions, the maximizer will be the one that satisfies the second-order
condition (SOC):
⇡ 00 (q) < 0 () 6q + 12 < 0.
We can see only q2 satisfies the SOC. Hence the profit maximizing solution is q ⇤ ⇡ 3.15.
Let’s analyze the graph of the profit function ⇡(q).
0.85 q
3.15
⇡(q)
We can see from the graph that the profit curve is flat at points 0.85 and 3.15. This is because
those point satisfy the FOC: the derivative of the function is zero at that point. To understand
whether the FOC identifies a (local) minimum or a (local) maximum point, one needs to check
the SOC. On the graph, we see that at point 0.85 the derivative starts increasing, meaning that
the function has attained a minimum. In contrast, at point 3.15 the derivative starts decreasing,
meaning that the function has attained a maximum.
In addition to the profit function, it would be interesting to analyze the graphs of the revenue
and cost functions. Below we draw them on the same graph.
R(q)
c(q)
q
3.15
2
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
On this graph, the di↵erent between the blue and the orange curves denotes the profit. As we
can see, this di↵erent is largest at the point 3.15. Another interesting observation is that the slope
of the cost function at the profit maximizing quantity 3.15 is equal to the slope of R(q), which is the
price p = 4. This is not a coincidence, but a general fact, which follows from the FOC. Moreover,
at the optimal point the cost function is convex. This follows from SOC.
Consider a solution q ⇤ that satisfies the FOCs and SOCs. Then,
FOCs:
⇡ 0 (q ⇤ ) = 0 () p c0 (q ⇤ ) = 0 () p = c0 (q ⇤ ) () p = M C(q ⇤ )
SOCs:
⇡ 00 (q ⇤ ) = 0 () c00 (q ⇤ ) < 0 () c00 (q ⇤ ) > 0 () c is convex at (around) q ⇤
Optimal Supply in the Short-Run (IMVH D3.e,f,g, Perlo↵ 8.3)
In the short run, the firm maximizes the profit ⇡(q) = pq cSR (q). The problem will either have
an interior solution (satisfying the FOCs and SOCs), or a corner solution of producing zero. To
simplify notation, let’s denote the cost function by c instead of cSR
Note: Producing zero units will generate a negative profit of ⇡(0) = p · 0 c(0) = c(0) = F C.
Despite this, the firm may find it optimal to do so in the short-run. In the long-run, the firm has
the option to exit the industry, avoid fixed costs, and earn zero profit.
Suppose q ⇤ is the (only) candidate solution of the profit maximization problem: that is q ⇤ satisfies
the FOC (M C(q ⇤ ) = p) and the SOC. Then, the solution to the profit maximization problem will
either be q ⇤ (an interior solution) or zero (a corner solution).
This would depend on how ⇡(q ⇤ ) = pq ⇤ c(q ⇤ ) compares to ⇡(0) = F C. This would in turn
only depend on whether p < min AV C or p min AV C.
Result: In the short-run:
• If p < min AV C, then the firm prefers to produce 0 instead of q ⇤ .
• If p min AV C, then the firm prefers to produce q ⇤ instead of 0.
Let’s prove the first bullet point in the result. If p < min AV C, then p < min AV C(q ⇤ ).
3
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
Multiplying both sides by q ⇤ , we get that
pq ⇤ < min AV C(q ⇤ )q ⇤ () pq ⇤ < V C(q ⇤ ) () pq ⇤ V C(q ⇤ ) < 0,
Subtracting F C from both sides of the last inequality, we get that
pq ⇤ V C(q ⇤ ) FC < F C () pq ⇤ c(q ⇤ ) < F C () ⇡(q ⇤ ) < ⇡(0).
The second bullet point in the result is proved analogously. We can observe that if p min AV C,
then MC will intersect p above the AVC curve, that is, M C(q ⇤ ) = p AV C(q ⇤ ). By the same
equivalence relations, p AV C(q ⇤ ) would imply that ⇡(q ⇤ ) ⇡(0), and the firm will prefer to
⇤
produce q rather than 0.
Given the result above, we can directly identify the firm’s short-run supply function s(p),
which tells how much the firm produces (optimally) given a price level p. In the short-run, the
supply function is 8
<0 if p < min AV C
s(p) =
:q ⇤ (p) if p min AV C
where q ⇤ (p) is the production level that satisfies the FOC M C(q ⇤ (p)) = p (and the SOC).
The short-run supply curve coincides with the MC everywhere above the AVC curve.
Numerical Example: Optimal Supply Short-Run Suppose a firm’s cost function is c(q) =
q 2 + 9. Then, AC(q) = q + 9q , AV C(q) = q, and M C(q) = 2q. Suppose the price of the good
is p = 10. Since 10 > 0 = min AV C, we will have that q ⇤ will be determined by the FOC
M C(q ⇤ ) = 10. Setting 2q = 10, we obtain that the optimal quantity is q ⇤ = 5. The intersection
point of the p = 10 line and the M C is illustrated on the graph. The blue curve is the MC, the
orange curve is the AVC, and the green curve is the AC.
p = 10
6.8
q
3 q⇤ = 5
4
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
The firm’s profit is
⇥ ⇤
⇡(q ⇤ ) = pq ⇤ c(q ⇤ ) = pq ⇤ AC(q ⇤ )q ⇤ = p AC(q ⇤ ) q ⇤ = (10 6.8)5 = 16.
⇥ ⇤
As we can express the profit by p AC(q ⇤ ) q ⇤ , it follows that the profit equals to the yellow area
on the graph.
q
q⇤ = 5
Now suppose the price is p = 4. This price is still larger than the min AV C, hence, the firm will
produce according to q ⇤ satisfying M C(q ⇤ ) = p. Setting 2q = 4, we get q ⇤ = 2. The intersection
point of the p = 4 line and the M C is illustrated on the graph.
p=4
q
q⇤ = 2
⇥ ⇤
The firm’s profit is ⇡(q ⇤ ) = p AC(q ⇤ ) q ⇤ = [4 6.5]2 = 5. The firm earns a negative profit
because the price is below the AC. This is optimal in the short-run, because the firm cannot avoid
the fixed costs: producing zero gives a profit of ⇡(0) = F C = 9.
5
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
p=4
q
q⇤ = 2
Optimal Supply in the Long-Run (IMVH D3.h, Perlo↵ 8.4)
In the long-run the firm has two options: (1) either to remain in the industry and maximize the
profit ⇡(q) = pq cLR (q), (2) or to exit the industry, avoid fixed costs, and earn a zero profit. To
simplify notation, let’s denote the cost function by c instead of cLR
Suppose q ⇤ is the candidate solution of the profit maximization problem: that is q ⇤ satisfies
the FOC (M C(q ⇤ ) = p) and the SOC. Then, the solution to the profit maximization problem will
either be q ⇤ or exiting the industry and earning zero profit. The optimality of either solution would
will on how ⇡(q ⇤ ) = pq ⇤ c(q ⇤ ) compares to 0 (this is di↵erent from the short-run!) This would in
turn only depend on whether p < min AC or p min AC.
Result: In the long-run:
• If p < min AC, then the firm prefers to exit the industry instead of producing q ⇤ .
• If p min AC, then the firm prefers to produce q ⇤ instead of exiting the industry.
Let’s prove the first bullet point in the result. If p < min AC, then p < min AC(q ⇤ ). Multiplying
both sides by q ⇤ , we get that
pq ⇤ < min AC(q ⇤ )q ⇤ () pq ⇤ < c(q ⇤ ) () pq ⇤ c(q ⇤ ) < 0.
This proves both bullet points in the result.
Given the result above, we can directly identify the firm’s long-run supply function s(p),
which tells how much the firm produces (optimally) given a price level p. In the long-run, the
supply function is 8
<0 (exit) if p < min AC
s(p) =
:q ⇤ (p) if p min AC
6
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Aram Grigoryan Econ 100B, Spring 2023
where q ⇤ (p) is the production level that satisfies the FOC M C(q ⇤ (p)) = p (and the SOC).
The long-run supply curve coincides with the MC everywhere above the AC curve.
The steps for solving numerical examples in the long-run is similar to that in the short run. A
major di↵erence is that in the long-run the firm will never earn negative profit: it can exit the
industry and avoid fixed costs whenever the price is below the average cost.
Price Change and Profit (IMVH D3.d)
Suppose the firm’s profit maximization problem has an interior solution, the solution q(p), satisfies
the M C(q(p)) = p condition (to simplify notation, we use q(p) instead of the q ⇤ (p)). Then, the
profit function can be written as a function of the price
⇡(q(p)) = pq(p) c(q(p)).
If we want to evaluate the e↵ect of the price change on the profit, we shall look at the derivative
of the profit function with respect to p. Taking this derivative (and using the chain rule), we get
d⇡(q(p))
= q(p) + q 0 (p) · p c0 (q(p)) · q 0 (p)
dp
⇥ ⇤
= q(p) + p M C(q(p)) · q 0 (p) = q(p) + 0 = q(p).
⇥ ⇤
The p M C(q(p)) · q 0 (p) = 0 follows from that q(p) satisfies the FOC M C(q(p)) = p.
Thus, the e↵ect of a price change on profit is simply d⇡(q(p))
dp = q(p).
Note that q(p) would have also been the partial derivative of ⇡ with respect to p (if we did
not treat q(p) as a function of p) itself and ignored the last two terms in the total derivative
d⇡(q(p))
dp = q(p). This is a general property of derivatives of objective functions with respect to
parameters in the optimization problem. The result is known as the Envelope theorem.
7
This study source was downloaded by 100000883704362 from CourseHero.com on 11-09-2024 05:07:05 GMT -06:00
https://www.coursehero.com/file/234809107/Lecture-Notes-Part-IIIpdf/
Powered by TCPDF (www.tcpdf.org)
