π Lecture 35 β Concept of Optimization
π Overview: This lecture introduces the fundamental concept of optimization in economics, distinguishing between maximizing favorable and minimizing unfavorable variables. It progresses from the first-order necessary condition using calculus to the second-order sufficient condition, and finally introduces the Hessian matrix approach for multivariable optimization. Understanding these methods is critical for finding optimal economic outcomes like cost minimization and profit maximization.
ποΈ Topics Covered
The lecture covers five topics: the definition and etymology of optimization, the first-order test using derivatives (necessary condition), an average cost analysis example, the second-order test using the second derivative (sufficient condition), and the matrix approach to the second-order test using the Hessian determinant. The instance of a cubic function is used throughout to illustrate these concepts.
π Lecture Summary
Topic 171: Concept of Optimization
The word optimization comes from the Latin 'Optimus', meaning "best." It is the process of maximizing favorable variables (like profit) and minimizing unfavorable variables (like cost). The lecture assumes a function where a cubic specification creates a curve with "two wiggles"βone upward and one downward. Key terminology includes Extremum (singular) vs. Extrema (plural), Maximum vs. Maxima, and Minimum vs. Minima.
Topic 172: Calculus Approach to Optimization: 1st Order Test
This involves considering a continuous function ( y = f(x) ) that can have an extremum. The goal is to find the critical value at which the function is optimized. Because the slope of the function at a maximum or minimum is zero, the first derivative is set to zero. This, ( f'(x) = 0 ), is known as the First Order Condition (F.o.C) or the Necessary condition for optimization.
Instance: ( y = f(x) = 2x^3 - 3x^2 - 12x + 10 ) ( f'(x) = 6x^2 - 6x - 12 = 0 )
π Definition β Critical Value: The value of ( x ) where the first derivative of the function equals zero (( f'(x) = 0 )). π Formula: First Order Condition: ( f'(x) = 0 ). π Example: To find the critical values of ( y = 2x^3 - 3x^2 - 12x + 10 ), we set the first derivative equal to zero: ( f'(x) = 6x^2 - 6x - 12 = 0 ). Dividing by 6 gives ( x^2 - x - 2 = 0 ), which factors to ( (x - 2)(x + 1) = 0 ). Therefore, the critical values are ( x = 2 ) and ( x = -1 ).
Topic 173: Average Cost Analysis
This section analyzes average cost, likely connecting the first-order condition to a real economic problem. The specific example or formula for average cost is not fully detailed in the text, but it serves as an application of optimization, where minimizing average cost is the objective. π‘ Why this matters: This links the abstract concept of finding a minimum of a function to the concrete economic goal of minimizing a firm's average cost of production.
Topic 174: Calculus Approach to Optimization: 2nd Order Test
After finding critical values, one must confirm if the function is maximized, minimized, or just has an inflection point. This is done by examining the rate of change of the slope, which is the second derivative, ( f''(x) ). This is known as the Second Order Condition (S.o.C) or the Sufficient condition for optimization. The rule is:
- If ( f''(x) < 0 ), the function is at a maximum.
- If ( f''(x) > 0 ), the function is at a minimum.
- If ( f''(x) = 0 ), the point may be an inflection point.
π Definition β Second Order Condition (S.o.C): The test using the second derivative to determine whether a critical point is a maximum, minimum, or inflection point. π Formula: Sufficient Condition: ( f''(x) ). If ( f''(x) < 0 ), max; if ( f''(x) > 0 ), min; if ( f''(x) = 0 ), test is inconclusive (possibly inflection). π Example: For the function ( y = 2x^3 - 3x^2 - 12x + 10 ), the second derivative is ( f''(x) = 12x - 6 ).
- For the critical value ( x = -1 ): ( f''(-1) = 12(-1) - 6 = -18 ). Since ( -18 < 0 ), the function is at a maximum at ( x = -1 ).
- For the critical value ( x = 2 ): ( f''(2) = 12(2) - 6 = 18 ). Since ( 18 > 0 ), the function is at a minimum at ( x = 2 ).
- For a critical value of ( x = 0.5 ) (where ( f'(x) = 0 )), ( f''(0.5) = 12(0.5) - 6 = 0 ), which indicates an inflection point.
Topic 175: Matrix Approach to Optimization: 2nd Order Test β Hessian
For functions with multiple variables, the second-order condition uses the Hessian determinant. The Hessian ( |H| ) is a determinant of all second-order partial derivatives. The second-order direct partials are on the principal diagonal, and the second-order cross partials are off the diagonal. By Young's theorem, the cross partials are equal (( f_{xy} = f_{yx} )).
The standard ( 2 \times 2 ) Hessian is: ( |H| = \begin{vmatrix} f_{xx} & f_{xy} \ f_{yx} & f_{yy} \end{vmatrix} ) The principal minors are ( |H_1| = f_{xx} ) and ( |H_2| = |H| ).
- If both ( |H_1| > 0 ) and ( |H_2| > 0 ), there is a minimum.
- If ( |H_1| < 0 ) and ( |H_2| > 0 ), there is a maximum.
- For N variables, the signs of the principal minors alternate for a maximum and are all positive for a minimum.
π Definition β Hessian Determinant (|H|): A determinant containing second-order partial derivatives, used to determine the nature of a critical point for a multivariable function. π Formula: For a function of two variables, ( |H| = f_{xx}f_{yy} - (f_{xy})^2 ). A minimum requires ( f_{xx} > 0 ) and ( |H| > 0 ). A maximum requires ( f_{xx} < 0 ) and ( |H| > 0 ). π Example: For a function ( f(x,y) ) where ( f_{xx} = 2, f_{yy} = 8, f_{xy} = -2 ), the Hessian is ( |H| = \begin{vmatrix} 2 & -2 \ -2 & 8 \end{vmatrix} = (2)(8) - (-2)(-2) = 16 - 4 = 12 ). Since ( |H_1| = 2 > 0 ) and ( |H_2| = 12 > 0 ), the function has a minimum at that critical point. Another example with ( |H| = -30 ) indicates a saddle point (neither max nor min).
β Key Takeaways
The lecture establishes a three-step process for optimization. First, the First Order Condition (F.o.C), ( f'(x) = 0 ), is a necessary condition to find critical values (potential optima). Second, the Second Order Condition (S.o.C) using the second derivative ( f''(x) ) is the sufficient condition to distinguish between a maximum (( f''(x) < 0 )) and a minimum (( f''(x) > 0 )). For multivariable functions, this test is performed using the Hessian determinant of second-order partial derivatives. The core skill is applying these conditions to economic functions like cost and profit to find optimal values.
π§ Quick Revision Questions
- What is the Latin root of the word "optimization" and what does it mean?
- State the First Order Condition (F.o.C.) for a single-variable function ( y = f(x) ).
- How does the Second Order Condition (S.o.C.) distinguish between a maximum and a minimum?
- For a function of two variables, what are the sign conditions on the principal minors of the Hessian matrix for a minimum to exist?
- In the matrix approach, what is Young's theorem and how does it simplify the Hessian determinant?
π Lecture 36 β Profit Maximization Analysis
π Overview: This lecture examines how firms maximize profit under different market structures, including perfect competition and monopoly. It covers both single-product firms and firms producing multiple technically related goods, with particular emphasis on substitute and complementary products.
ποΈ Topics Covered
The lecture begins with the fundamental concept of profit maximization analysis, followed by a numerical example demonstrating profit maximization calculations. It then explores profit maximization for technically related goods, including monopolistic firms producing related goods, and concludes with the case of firms producing substitute goods.
π Lecture Summary
TOPIC 176: PROFIT MAXIMIZATION ANALYSIS
Profit maximization is the primary goal of any firm. For a firm producing a single good with a single variable input (labor), the production function is given by ( Q = f(L) ), where ( Q ) is output and ( L ) is labor input. The firm's revenue is ( R = P \cdot Q = P \cdot f(L) ), and the cost is ( C = w \cdot L ), where ( w ) is the wage rate. The profit function is therefore ( \pi = R - C = P \cdot f(L) - w \cdot L ).
To maximize profit, the first-order condition requires setting the derivative of profit with respect to labor equal to zero: [ \frac{d\pi}{dL} = P \cdot f'(L) - w = 0 ] This yields ( P \cdot MP_L = w ), where ( MP_L = f'(L) ) is the marginal product of labor. This condition says that the value of marginal product must equal the wage rate. The second-order condition for a maximum is ( P \cdot f''(L) < 0 ), which requires diminishing marginal returns to labor.
π Definition β Profit Function: ( \pi = P \cdot f(L) - w \L ) β The difference between total revenue and total cost, expressed as a function of labor input. π Formula: ( P \cdot MP_L = w ) β The firm hires labor until the value of the marginal product equals the wage rate. π‘ Why this matters: This condition determines the optimal level of labor employment for a profit-maximizing firm in a perfectly competitive market.
TOPIC 177: NUMERICAL EXAMPLE OF PROFIT MAXIMIZATION
Consider a firm with the production function ( Q = 10L - 0.5L^2 ). The price of output is ( P = 10 ) and the wage rate is ( w = 40 ). The marginal product of labor is derived as: [ MP_L = f'(L) = 10 - L ] Setting ( P \cdot MP_L = w ): [ 10(10 - L) = 40 ] [ 100 - 10L = 40 ] [ 10L = 60 ] [ L = 6 ] The optimal labor input is 6 units. The corresponding output is: [ Q = 10(6) - 0.5(36) = 60 - 18 = 42 ] Revenue = ( 10 \times 42 = 420 ), Cost = ( 40 \times 6 = 240 ), and Profit = ( 420 - 240 = 180 ).
The second-order condition requires ( P \cdot f''(L) < 0 ). Here, ( f''(L) = -1 ), so ( 10 \times (-1) = -10 < 0 ), confirming a maximum.
π Definition β Second-Order Condition: ( P \cdot f''(L) < 0 ) β Ensures the critical point yields a maximum rather than a minimum. π Formula: ( \pi = 180 ) at ( L = 6 ) β The maximum profit achievable given the production function and market prices. π Example: With ( Q = 10L - 0.5L^2 ), ( P = 10 ), and ( w = 40 ), the firm hires 6 workers, produces 42 units, and earns a profit of 180.
TOPIC 178: PROFIT MAXIMIZATION OF TECHNICALLY RELATED GOODS
When a firm produces two goods that are technically related in production, they may share inputs or be produced using linked technologies. The profit function for two goods ( Q_1 ) and ( Q_2 ) is: [ \pi = P_1Q_1 + P_2Q_2 - C(Q_1, Q_2) ] where ( C(Q_1, Q_2) ) is the joint cost function. The first-order conditions for profit maximization are: [ \frac{\partial\pi}{\partial Q_1} = P_1 - \frac{\partial C}{\partial Q_1} = 0 \quad \Rightarrow \quad P_1 = MC_1 ] [ \frac{\partial\pi}{\partial Q_2} = P_2 - \frac{\partial C}{\partial Q_2} = 0 \quad \Rightarrow \quad P_2 = MC_2 ] This shows that each good should be produced up to the point where its price equals its marginal cost. The second-order conditions require that the Hessian matrix of second-order partial derivatives be negative definite.
π Definition β Technical Relationship: When two goods share production inputs or technologies, their cost functions are interdependent. π Formula: ( P_1 = MC_1 ) and ( P_2 = MC_2 ) β Optimal output for each good is where price equals its marginal cost.
TOPIC 179: PROFIT MAXIMIZATION OF MONOPOLISTIC FIRM PRODUCING RELATED GOODS
A monopolistic firm producing related goods faces downward-sloping demand curves for each product. The demand functions are ( P_1 = f_1(Q_1, Q_2) ) and ( P_2 = f_2(Q_1, Q_2) ), where the price of one good depends on the quantity of both goods (substitutes or complements). The profit function is: [ \pi = P_1(Q_1, Q_2) \cdot Q_1 + P_2(Q_1, Q_2) \cdot Q_2 - C(Q_1, Q_2) ] The first-order conditions involve partial derivatives with respect to both quantities: [ \frac{\partial\pi}{\partial Q_1} = \frac{\partial P_1}{\partial Q_1}Q_1 + P_1 + \frac{\partial P_2}{\partial Q_1}Q_2 - MC_1 = 0 ] [ \frac{\partial\pi}{\partial Q_2} = \frac{\partial P_2}{\partial Q_2}Q_2 + P_2 + \frac{\partial P_1}{\partial Q_2}Q_1 - MC_2 = 0 ] The terms ( \frac{\partial P_2}{\partial Q_1}Q_2 ) and ( \frac{\partial P_1}{\partial Q_2}Q_1 ) capture the cross-price effects between the two goods.
π Definition β Cross-Price Effect: The change in revenue of one good due to a change in the quantity of another good. π‘ Why this matters: For a monopolist producing related goods, pricing decisions must account for how quantities affect prices of both products.
TOPIC 180: PROFIT MAXIMIZATION OF FIRM PRODUCING SUBSTITUTE GOODS
When a firm produces substitute goods, an increase in the quantity of one good reduces the demand (and price) of the other. This creates a negative cross-price effect. The profit function for substitute goods is maximized when: [ \frac{\partial\pi}{\partial Q_1} = MR_1 + \text{(negative cross effect)} - MC_1 = 0 ] [ \frac{\partial\pi}{\partial Q_2} = MR_2 + \text{(negative cross effect)} - MC_2 = 0 ] Because the cross effects are negative, the firm must produce less of each good compared to if they were independent. This is because increasing output of one good cannibalizes sales of the other.
For complementary goods, the cross effects are positive, meaning the firm can produce more of both goods as they boost each other's demand. The second-order conditions for a maximum require that the Hessian matrix be negative definite: [ H = \begin{bmatrix} \frac{\partial^2\pi}{\partial Q_1^2} & \frac{\partial^2\pi}{\partial Q_1\partial Q_2} \ \frac{\partial^2\pi}{\partial Q_2\partial Q_1} & \frac{\partial^2\pi}{\partial Q_2^2} \end{bmatrix} ] For a maximum, the principal minors must alternate in sign: ( |H_1| < 0 ) and ( |H_2| > 0 ).
π Definition β Substitute Goods: Goods where an increase in quantity of one decreases demand for the other, creating negative cross-price effects. π Formula: ( \frac{\partial\pi}{\partial Q_i} = MR_i + \text{cross effect} - MC_i = 0 ) β The first-order condition for each good, accounting for interdependence.
β Key Takeaways
Profit maximization for a single-product firm requires setting the value of marginal product equal to the input price (wage). For multiple technically related goods, the first-order conditions require each good's price to equal its marginal cost when goods are independent. However, a monopolistic firm producing related goods must account for cross-price effects: substitute goods have negative cross effects that constrain output, while complementary goods have positive cross effects that encourage higher production. The second-order conditions for multi-product firms involve a Hessian matrix that must be negative definite to ensure a maximum.
π§ Quick Revision Questions
- What is the first-order condition for profit maximization for a firm using a single variable input?
- In the numerical example with ( Q = 10L - 0.5L^2 ), ( P = 10 ), and ( w = 40 ), what is the optimal labor input and maximum profit?
- What are the first-order conditions for profit maximization when a firm produces two technically related goods?
- How do the first-order conditions differ for a monopolistic firm producing substitute goods compared to producing independent goods?
- What mathematical condition must the Hessian matrix satisfy for a multi-product profit maximum?
π Lecture 37 β Profit Maximization Analysis (Continued 1)
π Overview: This lecture continues the analysis of profit maximization by examining the relationship between marginal and average revenue. It then explores short-run production function analysis and concludes by integrating total cost, total revenue, and profit maximization into a unified framework. This lecture is crucial for understanding how firms make optimal output and input decisions.
ποΈ Topics Covered
The lecture covers three main topics: first, Marginal and Average Revenue Analysis including the relationship between price, marginal revenue, and average revenue under different market structures; second, Short Run Production Function Analysis focusing on marginal and average product concepts; and third, Total Cost, Total Revenue, and Profit Maximization showing how these three functions interact to determine the optimal output level.
π Lecture Summary
Topic 181: Marginal and Average Revenue Analysis
Marginal Revenue (MR) is the additional revenue generated from selling one more unit of output. Average Revenue (AR) is the revenue per unit sold, which equals the price of the good. The relationship between MR and AR depends on the market structure. Under perfect competition, the firm is a price taker, so Price (P) equals Marginal Revenue (MR) equals Average Revenue (AR). Under imperfect competition (monopoly, oligopoly, monopolistic competition), the firm faces a downward-sloping demand curve, so Price (P) is greater than Marginal Revenue (MR), and the MR curve lies below the AR curve.
The relationship between MR and the price elasticity of demand (Ξ΅) is given by the formula: MR = P (1 + 1/Ξ΅) where Ξ΅ is the price elasticity of demand (always negative). If demand is elastic (|Ξ΅| > 1), MR is positive. If demand is unitary elastic (|Ξ΅| = 1), MR is zero. If demand is inelastic (|Ξ΅| < 1), MR is negative.
π Definition β Marginal Revenue (MR): The change in total revenue resulting from a one-unit change in output. π Formula: MR = ΞTR/ΞQ or dTR/dQ β This measures the slope of the total revenue function. π Formula: MR = P (1 + 1/Ξ΅) β This shows how marginal revenue depends on price and demand elasticity. π Example: If a firm charges a price of $10 and the price elasticity of demand is -2, then MR = 10 (1 + 1/(-2)) = 10 (1 - 0.5) = $5. This means the additional revenue from selling one more unit is $5, which is less than the price of $10.
π‘ Why this matters: Understanding the MR-AR relationship is fundamental for pricing strategy. A monopolist will never operate in the inelastic portion of the demand curve because MR would be negative, meaning selling more reduces total revenue.
Topic 182: Short Run Production Function Analysis
In the short run, at least one input is fixed (typically capital), while labor is the variable input. The Short Run Production Function shows the relationship between the variable input (labor, L) and the total output (Q). From this, we derive the Marginal Product of Labor (MPL), which is the additional output produced by hiring one more unit of labor. The Average Product of Labor (APL) is the total output divided by the number of workers.
The Law of Diminishing Marginal Returns states that as more and more units of a variable input (labor) are added to a fixed input (capital), the marginal product of the variable input will eventually decrease. This gives the MPL curve an inverted U-shape: initially increasing, reaching a maximum, and then decreasing. The APL curve also has an inverted U-shape. When MPL > APL, APL is rising. When MPL < APL, APL is falling. MPL intersects APL at the maximum point of APL.
π Definition β Marginal Product of Labor (MPL): The change in total output resulting from a one-unit change in labor input, holding all other inputs constant. π Formula: MPL = ΞQ/ΞL or dQ/dL β This measures the slope of the total product function. π Definition β Average Product of Labor (APL): Total output divided by the quantity of labor used. π Formula: APL = Q/L β This measures the output per worker. π Example: Suppose a factory has a fixed amount of capital and hires workers. With 10 workers, total output (Q) is 100 units, so APL = 100/10 = 10 units per worker. When an 11th worker is hired, total output increases to 108 units. The MPL of the 11th worker is 108 - 100 = 8 units. Since MPL (8) < APL (10), the APL will fall.
π‘ Why this matters: The production function and the law of diminishing returns explain the shape of the firm's cost curves. When MPL is rising, marginal cost is falling; when MPL is falling, marginal cost is rising.
Topic 183: Total Cost, Total Revenue, and Profit Maximization
Profit maximization occurs where the difference between Total Revenue (TR) and Total Cost (TC) is greatest. Graphically, this is the point where the vertical distance between the TR curve and the TC curve is at a maximum. Mathematically, the profit function (Ο) is: Ο(Q) = TR(Q) - TC(Q).
To find the profit-maximizing output level, we set the first derivative of the profit function equal to zero: dΟ/dQ = dTR/dQ - dTC/dQ = 0, which implies MR = MC. This is the first-order condition for profit maximization. The second-order condition requires that the second derivative of the profit function be negative: dΒ²Ο/dQΒ² < 0, which means the slope of MR < slope of MC.
If MR > MC, the firm can increase profit by increasing output because the additional revenue from selling one more unit exceeds the additional cost. If MR < MC, the firm should decrease output because the additional cost of producing one more unit exceeds the additional revenue. The profit-maximizing output level is where MR = MC, provided that the firm is covering its variable costs in the short run (shutdown condition).
π Definition β Profit Function: The mathematical expression showing profit as a function of output: Ο(Q) = TR(Q) - TC(Q). π Formula (First-Order Condition): dΟ/dQ = 0 β MR = MC β The profit-maximizing output occurs where marginal revenue equals marginal cost. π Formula (Second-Order Condition): dΒ²Ο/dQΒ² < 0 β The slope of the MR curve must be less than the slope of the MC curve at the equilibrium point. π Example: A firm has TR = 50Q - 2QΒ² and TC = 20 + 10Q. MR = dTR/dQ = 50 - 4Q. MC = dTC/dQ = 10. Setting MR = MC: 50 - 4Q = 10 β 4Q = 40 β Q = 10. The second derivative: dΒ²Ο/dQΒ² = -4 < 0, confirming Q=10 is the profit-maximizing output. Profit at Q=10: Ο = TR(10) - TC(10) = (500 - 200) - (20 + 100) = 300 - 120 = 180.
π‘ Why this matters: The MR = MC rule is the most fundamental condition in microeconomics. It applies to all firms regardless of market structure. Any deviation from this output level would reduce profit.
β Key Takeaways
You must remember that marginal revenue is the additional revenue from one more unit, and its relationship with price depends on market structure (P = MR under perfect competition, P > MR under monopoly). The short-run production function shows that due to the law of diminishing marginal returns, both MPL and APL eventually decline. Profit is maximized where MR = MC, and this is found by setting the first derivative of the profit function to zero and checking that the second derivative is negative. The MR = MC rule is universal, but the specific values of MR depend on the demand curve facing the firm. Finally, the shapes of all these curves (TR, TC, MR, MC, MPL, APL) are interconnected and determined by the underlying production technology.
π§ Quick Revision Questions
- Under imperfect competition, why is marginal revenue always less than price? Explain using the formula MR = P(1 + 1/Ξ΅).
- A firm hires a 5th worker and total output increases from 45 to 56 units. Calculate the MPL of the 5th worker.
- If a firm is producing where MR = $15 and MC = $12, should it increase or decrease output to maximize profit? Why?
- At what point on the APL curve does the MPL curve intersect it? What is the significance of this point?
- A monopolist faces demand P = 100 - 2Q and has MC = 4Q. Find the profit-maximizing output and price using the MR = MC rule.
π Lecture 38 β PROFIT MAXIMIZATION ANALYSIS (CONTINUED 2)
π Overview: This lecture extends profit maximization to more complex scenarios, including quadratic profit functions, exponential revenue functions, and optimization problems with multiple choice variables. It also provides the mathematical conditions for identifying maxima, minima, saddle points, and inflection points for functions of two variables, with practical economic applications to multi-product and multi-plant firms.
ποΈ Topics Covered
The lecture covers five main topics: analysis of a quadratic profit function with a numerical example, optimization of an exponential revenue function, optimization of more than one choice variable including first and second-order conditions for relative extrema, economic application on a multi-product firm, and economic application on a multi-plant firm.
π Lecture Summary
TOPIC 184: QUADRATIC PROFIT FUNCTION ANALYSIS
This topic analyzes a quadratic profit function to find the profit-maximizing output level. The general form is given as ( \pi = aQ^2 + bQ + c ). To maximize profit, we take the first derivative and set it equal to zero: ( \frac{d\pi}{dQ} = 2aQ + b = 0 ). The optimal output is ( Q = -\frac{b}{2a} ). The second derivative test confirms a maximum if ( \frac{d^2\pi}{dQ^2} = 2a < 0 ), meaning ( a ) must be negative for a maximum.
π Definition β Quadratic Profit Function: A profit function of the form ( \pi = aQ^2 + bQ + c ), where ( a < 0 ) for a maximum. π Formula: ( Q = -\frac{b}{2a} ) β The output level that maximizes profit for a quadratic function. π Example: Given ( \pi = -Q^2 + 8Q - 10 ). First derivative: ( \frac{d\pi}{dQ} = -2Q + 8 = 0 ). Optimal output ( Q = 4 ). Second derivative: ( \frac{d^2\pi}{dQ^2} = -2 < 0 ), confirming a maximum. Maximum profit: ( \pi = -(4)^2 + 8(4) - 10 = -16 + 32 - 10 = 6 ).
TOPIC 185: OPTIMIZATION OF EXPONENTIAL REVENUE FUNCTION
This topic covers finding the maximum of an exponential revenue function, such as ( R(Q) = Qe^{-0.5Q} ). The optimization process uses the product rule of differentiation. The first-order condition involves setting the derivative equal to zero, which often simplifies using the property that an exponential term cannot be zero. The second derivative is then checked to confirm it is a maximum.
TOPIC 186: OPTIMIZATION OF MORE THAN ONE CHOICE VARIABLE
This topic introduces optimization for functions with two or more independent variables, e.g., ( z = f(x, y) ), creating a surface (or hypersurface). The objective is to find peaks of domes (maxima) and bottoms of bowls (minima). The Conditions for Relative Extremum are presented in a table:
| Condition | Maximum | Minimum |
|---|---|---|
| 1st Order (Necessary) | ( f_x = 0, f_y = 0 ) | ( f_x = 0, f_y = 0 ) |
| 2nd Order (Necessary) | ( f_{xx} < 0, f_{yy} < 0 ) | ( f_{xx} > 0, f_{yy} > 0 ) |
| 2nd Order (Sufficient) | ( f_{xx}f_{yy} - (f_{xy})^2 > 0 ) | ( f_{xx}f_{yy} - (f_{xy})^2 > 0 ) |
The Conditions for Inflection and Saddle Points are:
- Inflection Point: Necessary condition is ( f_{xx} = 0 ) OR ( f_{yy} = 0 ). Sufficient condition is ( f_{xx}f_{yy} - (f_{xy})^2 = 0 ).
- Saddle Point: Necessary condition is ( f_{xx} < 0, f_{yy} > 0 ) OR ( f_{xx} > 0, f_{yy} < 0 ). Sufficient condition is ( f_{xx}f_{yy} - (f_{xy})^2 < 0 ).
The Conditions for Inconclusiveness of the Test occur when the sufficient condition ( f_{xx}f_{yy} - (f_{xy})^2 = 0 ).
By Young's Theorem: ( f_{xy} = f_{yx} ), which implies ( f_{xx}f_{yy} - (f_{xy})^2 = f_{xx}f_{yy} - (f_{yx})^2 ).
π Definition β Saddle Point: A critical point where the function increases in one direction and decreases in another, not a local extremum.
π Numerical Example: Find the relative extrema of ( f(x, y) = x^3 + 3xy - y^3 + 15 ). 1st order conditions: ( f_x = 3x^2 + 3y = 0 ) and ( f_y = 3x - 3y^2 = 0 ). Solving yields critical points: ( (0, 0), (1, -1), (-1, 1), (0, 0) ). Notably, (1, -1) and (-1, 1) are unique.
Second-order partial derivatives: ( f_{xx} = 6x, f_{yy} = -6y, f_{xy} = 3 ).
- POINT I: At ( (0,0) ): ( f_{xx} = 0, f_{yy} = 0, f_{xx}f_{yy} - (f_{xy})^2 = 0 - 9 = -9 < 0 ). Different signs of ( f_{xx} & f_{yy} ) conditions not strictly met but the negative discriminant β Saddle point.
- POINT II: At ( (1, -1) ): ( f_{xx} = 6 > 0, f_{yy} = 6 > 0, f_{xx}f_{yy} - (f_{xy})^2 = (6)(6) - 9 = 27 > 0 ). ( f_{xx} & f_{yy} > 0 ) β Minimum/inflection. Since the sufficient condition is ( > 0 ), a Minimum is confirmed.
- POINT III: At ( (-1, 1) ): ( f_{xx} = -6 < 0, f_{yy} = -6 < 0, f_{xx}f_{yy} - (f_{xy})^2 = (-6)(-6) - 9 = 27 > 0 ). ( f_{xx} & f_{yy} < 0 ) β Maximum/inflection. Since the sufficient condition is ( > 0 ), a Maximum is confirmed.
- POINT IV: At ( (0, 0) ): Already analyzed as Saddle point.
D.I.Y (Do It Yourself):
- ( f(x,y) = x^3 + y^3 - 3xy )
- ( z = 5x^2 + 6y^2 - 10x + 8y )
TOPIC 187: ECONOMIC APPLICATION ON MULTI-PRODUCT FIRM
This topic applies the optimization of multiple variables to a multi-product firm, which produces more than one good. The firm's profit function, ( \pi(Q_1, Q_2) ), depends on the quantities of each product. The first-order conditions require setting the partial derivative of profit with respect to each quantity to zero: ( \frac{\partial \pi}{\partial Q_1} = 0 ) and ( \frac{\partial \pi}{\partial Q_2} = 0 ). The second-order conditions are then checked to ensure a maximum.
TOPIC 188: ECONOMIC APPLICATION ON MULTI-PLANT FIRM
This topic covers the optimization problem for a multi-plant firm, which produces the same good in different plants (e.g., Plant A and Plant B). The firm's total profit is the sum of revenues and costs from all plants. The objective is to allocate production between plants to maximize total profit. The first-order conditions involve setting the marginal profit from each plant equal to zero.
β Key Takeaways
For the exam, you must remember the first and second-order conditions for a maximum and minimum of a function with two independent variables. The sufficient condition is the discriminant ( f_{xx}f_{yy} - (f_{xy})^2 ). A positive discriminant, combined with negative ( f_{xx} ), confirms a maximum. Understanding how to apply these conditions to multi-product and multi-plant firm profit functions is critical. Also, remember that a quadratic profit function is maximized at ( Q = -b/2a ) only if ( a < 0 ).
π§ Quick Revision Questions
- What is the first-order condition for a relative maximum of a function ( z = f(x, y) )?
- What is the second-order sufficient condition for a relative maximum of ( z = f(x, y) )?
- What does the discriminant ( f_{xx}f_{yy} - (f_{xy})^2 < 0 ) indicate about a critical point?
- For a quadratic profit function ( \pi = aQ^2 + bQ + c ), what condition on the parameter ( a ) must hold for the function to have a maximum?
- How does the optimization problem for a multi-plant firm differ from that of a single-plant firm?
π Lecture 39 β Profit Maximization Analysis (continued 3)
π Overview: This lecture extends profit maximization analysis to market structures with market power, covering price discrimination by both monopoly (seller-side) and monopsony (buyer-side). It also examines the firm's input decision under profit maximization and concludes with the profit maximization problem for a two-product firm facing interdependent demands.
ποΈ Topics Covered
The lecture covers price discrimination by monopoly, price discrimination by monopsony, the input decision of a firm (including the use of the first-order condition for profit maximization with respect to the input), and profit maximization of a two-product firm where the products are related in demand (substitutes or complements), solved using partial derivatives and Cramer's rule.
π Lecture Summary
TOPIC 189: PRICE DISCRIMINATION BY MONOPOLY
Price discrimination occurs when a monopoly charges different prices for the same product to different groups of buyers, based on their different price elasticities of demand, not on cost differences. The goal is to extract more consumer surplus.
For a monopolist practicing price discrimination in two separate markets (Market 1 and Market 2), the total profit function is: Ο = Rβ(Qβ) + Rβ(Qβ) β C(Qβ + Qβ)
The first-order conditions for profit maximization are: βΟ/βQβ = MRβ β MC = 0 β MRβ = MC βΟ/βQβ = MRβ β MC = 0 β MRβ = MC
Thus, the monopolist will allocate output so that marginal revenue in each market equals the common marginal cost. Since MR = P(1 β 1/|E|), where E is price elasticity of demand, the monopolist will charge a higher price in the market with the lower elasticity (less elastic demand).
π Definition β Price Discrimination: A pricing strategy where a firm sells the same product at different prices to different buyers for reasons unrelated to cost.
π Formula: MRβ = MC = MRβ β Pβ(1 β 1/|Eβ|) = Pβ(1 β 1/|Eβ|) β The higher price is charged in the market with the lower absolute elasticity.
π‘ Why this matters: This shows the monopolist's incentive to segment markets based on elasticity. The more inelastic the demand, the higher the price the monopolist can charge.
TOPIC 190: PRICE DISCRIMINATION BY MONOPSONY
Monopsony is a market structure with a single buyer. Just as a monopolist is a price maker on the selling side, a monopsonist is a price maker on the buying side. Price discrimination by a monopsonist means paying different prices to different sellers for the same good or input, even when the cost of supply is the same.
The monopsonist's profit function when buying from two separate markets is: Ο = R(Qβ + Qβ) β Cβ(Qβ) β Cβ(Qβ)
The first-order conditions are: βΟ/βQβ = MR β MCβ = 0 β MR = MCβ βΟ/βQβ = MR β MCβ = 0 β MR = MCβ
Thus, the monopsonist will allocate purchases so that marginal revenue equals each seller's marginal cost. The monopsonist will pay a lower price to sellers with lower marginal cost of supply.
π‘ Why this matters: This is the buyer-side mirror image of monopoly price discrimination. It explains how a dominant buyer can exploit market power to pay different prices for the same input.
TOPIC 191: INPUT DECISION OF A FIRM
This topic extends profit maximization to the input decisionβchoosing the optimal amount of an input to hire. For a firm producing output Q using input L (labor), the profit function is: Ο = R(Q) β wL β FC, where Q = Q(L), w is the wage rate, and FC is fixed cost.
The firm chooses L to maximize Ο: βΟ/βL = (dR/dQ)(dQ/dL) β w = MR Γ MPL β w = 0 β MRP_L = w
The term MRP_L (Marginal Revenue Product of Labor) is MR Γ MPL.
π Definition β Marginal Revenue Product (MRP): The additional revenue the firm earns from employing one more unit of an input. MRP_L = MR Γ MPL.
π First-Order Condition for Profit-Maximizing Input Use: MRP_L = w β The firm hires labor until the marginal revenue product of labor equals the wage rate.
TOPIC 192: PROFIT MAXIMIZATION OF TWO-PRODUCT FIRM
This topic considers a firm producing two related products, say X and Y, where demands are interdependent (cross-price effects exist). The profit function is: Ο = P_X(X, Y) Γ X + P_Y(X, Y) Γ Y β C(X, Y)
The firm chooses X and Y simultaneously to maximize profit. The first-order conditions are obtained by setting partial derivatives to zero:
βΟ/βX = (βP_X/βX)X + P_X + (βP_Y/βX)Y β βC/βX = 0 βΟ/βY = (βP_X/βY)X + (βP_Y/βY)Y + P_Y β βC/βY = 0
These conditions incorporate the cross-effects: a change in X affects its own price (βP_X/βX), the price of Y (βP_Y/βX), and the marginal cost.
These two equations can be written as a system of linear equations and solved using Cramer's rule to find the optimal X* and Y*. The second-order conditions require that the Hessian matrix of second partial derivatives be negative definite.
π First-Order Conditions (General Form): (βP_X/βX)X + P_X + (βP_Y/βX)Y = MC_X (βP_X/βY)X + (βP_Y/βY)Y + P_Y = MC_Y
π Example: Given: P_X = 100 β 2X + Y P_Y = 120 β 3Y + 2X C = XΒ² + YΒ² + XY
Step 1: Write the total revenue function. R = P_X Γ X + P_Y Γ Y = (100 β 2X + Y)X + (120 β 3Y + 2X)Y R = 100X β 2XΒ² + XY + 120Y β 3YΒ² + 2XY R = 100X β 2XΒ² + 120Y β 3YΒ² + 3XY
Step 2: Write the profit function. Ο = R β C = (100X β 2XΒ² + 120Y β 3YΒ² + 3XY) β (XΒ² + YΒ² + XY) Ο = 100X β 3XΒ² + 120Y β 4YΒ² + 2XY
Step 3: First-order conditions. βΟ/βX = 100 β 6X + 2Y = 0 β 6X β 2Y = 100 (Equation 1) βΟ/βY = 120 β 8Y + 2X = 0 β β2X + 8Y = 120 β Multiply by -1: 2X β 8Y = β120 (Equation 2)
Step 4: Solve using Cramer's rule. Write in matrix form: [6 -2] [X] = [100] [-2 8] [Y] = [120]
Determinant of A: |A| = (6Γ8) β (β2Γβ2) = 48 β 4 = 44
Determinant of A_X: |A_X| = (100Γ8) β (β2Γ120) = 800 + 240 = 1040 X* = |A_X| / |A| = 1040 / 44 = 23.64
Determinant of A_Y: |A_Y| = (6Γ120) β (100Γβ2) = 720 + 200 = 920 Y* = |A_Y| / |A| = 920 / 44 = 20.91
Result: The profit-maximizing outputs are X* β 23.64 and Y* β 20.91.
β Key Takeaways
The lecture demonstrates that market powerβwhether on the selling side (monopoly) or buying side (monopsony)βenables firms to price discriminate by segmenting markets, charging higher prices to less elastic demanders (monopoly) or paying lower prices to lower-cost suppliers (monopsony). The optimal input decision requires hiring until the marginal revenue product equals the input price. For a two-product firm with interdependent demands, profit maximization requires solving a system of first-order partial derivative equations, using Cramer's rule to find optimal outputs that account for cross-price effects. The core analytical tool across all topics is setting marginal revenue (or MRP) equal to marginal cost (or input price) for each decision variable.
π§ Quick Revision Questions
- What are the two first-order conditions for a price-discriminating monopolist selling in two separate markets?
- How does the concept of marginal revenue product (MRP) determine the profit-maximizing input level?
- For a monopsonist, what is the condition for allocating purchases across two different supply markets?
- In the two-product firm example, what is the economic meaning of the cross-partial derivative βP_X/βY?
- Using the two-product example, if the price functions changed to P_X = 100 β 3X + 2Y, what would be the new first-order condition for βΟ/βX?