📘 Lecture 37 — ALTERNATIVE THEORIES OF THE FIRM (CONTINUED 1)
📖 Overview: This lecture continues the exploration of alternative theories of the firm, moving beyond traditional profit maximization. It details Baumol's Sales Maximization Model with advertising as a key policy variable and introduces Marris's Model of Maximization of Growth Rate, which reconciles the conflicting goals of managers and owners. Understanding these models is crucial for analyzing real-world firm behavior in oligopolistic markets.
🗂️ Topics Covered
The lecture covers two major alternative theories: Baumol's static sales maximization model incorporating advertising as a policy variable, including its assumptions, equilibrium conditions, and criticisms. It then presents Marris's growth maximization model, detailing the utility functions of managers and owners, the balanced growth rate concept, the managerial and job security constraints (including the three financial ratios), the determination of demand and capital growth rates, and the model's equilibrium conditions, contributions, and criticisms.
📝 Lecture Summary
SALES MAXIMIZATION MODEL: W.J.BAUMOL
BAUMOL'S STATIC MODEL with Advertising The goal of the firm is sales revenue maximization subject to a minimum profit constraint, which is exogenously determined. A new element is the introduction of advertising as a major instrument (policy variable) of the firm. Baumol argues that in the real world, non-price competition is the typical form of competition in oligopolistic markets.
The total revenue function is now: TR = f(Q, A) TR is a function of quantity (Q) and advertising expenditure (A).
ADDED ASSUMPTION OF THE ADVERTISING MODEL The central assumption is that sales revenue increases with advertising expenditure (that is, ∂TR/∂A > 0, where A = advertising expenditure). This implies that advertising will always shift the demand curve of the firm to the right. The firm will sell a larger quantity and earn larger revenue. The price is assumed to remain constant. Another simplifying assumption is that production costs are independent of advertising. Baumol recognizes this is unrealistic but claims it can be relaxed without significantly changing the analysis.
Baumol argues that an oligopolistic firm will prefer advertising over price cuts. While a price cut may or may not increase sales revenue (depending on elasticity), an increase in advertising will always increase sales revenue because the marginal revenue of advertising (∂TR/∂A) is positive by assumption.
With advertising, it is no longer possible to have equilibrium where the profit constraint is not operative. Unlike price reduction, increased advertising always increases sales revenue. Consequently, the sales maximizer will increase advertising expenditure until stopped by the profit constraint. Thus, the minimum profit constraint is always operative when advertising is introduced. The sales maximizer will normally have higher advertising expenditures than a profit maximizer.
🔑 Definition — Sales Maximizer's Advertising: A sales-maximizing firm will have higher advertising expenditures than a profit-maximizing firm, as it continues to advertise until constrained by minimum profit requirements.
The model is shown in Figure 1. Advertising expenditure is on the horizontal axis, and the advertising function is a 45° line. Production costs (CC') are independent of advertising. Adding advertising costs to production costs gives the total-cost curve (TC). Subtracting total cost from total revenue gives the total-profit curve (Π). The positive marginal revenue of advertising means unconstrained sales maximization is not possible. The advertising expenditure of the sales maximizer (0Aₛ) is higher than that of the profit maximizer (0Aπ), and the profit constraint (Π) is operative at equilibrium.
Formal Model: R = f (Q, A) = TR function C = f (Q) = total production function Π’ = minimum acceptable profit A(a) = total cost of advertising function
Maximize: R = f (Q, A) Subject to: Π = R – C – A ≥ Π’ With assumptions: ∂R/∂Q > 0, ∂C/∂Q > 0, Q > 0
CRITICISM OF BAUMOL'S MODEL
- The hypothesis cannot be tested without measuring demand and cost functions of individual firms, but such data are not disclosed.
- In the long run, sales-maximization and profit-maximization may yield identical solutions, but this lacks empirical evidence.
- The theory does not show how equilibrium in an industry of sales maximizers will be attained; the firm-industry relationship is not established.
- The hypothesis assumes the firm has market power (control over price and expansion) without being affected by competitors' reactions.
- The assumption that ∂TR/∂A > 0 is not justified and casual observation suggests it may not always hold.
MARRIS'S MODEL OF MAXIMIZATION OF GROWTH RATE (1964)
The goal of the firm is the maximization of the balanced rate of growth of the firm, meaning the maximization of the rate of growth of demand for the firm's products and the growth of its capital supply: Maximize g = gᴰ = gᶜ
Where: g = balanced growth rate gᴰ = growth of demand for the products of the firm gᶜ = growth of the supply of capital
🔑 Definition — Balanced Growth Rate: The growth rate where the rate of growth of demand for a firm's products equals the rate of growth of its capital supply.
This goal has two constraints: a managerial constraint (available team and skills) and a financial constraint (set by managers' desire for job security). The rationale is that maximizing both growth rates maximizes the utility of both managers and owners.
Managerial theorists argue that the separation of ownership and management allows managers to set different goals. The utility function of managers (Uᴹ) includes salaries, status, power, and job security. The utility function of owners (Uᵒ) includes profits, output size, capital size, market share, and public image. Marris argues these goals are not as different as other theories claim, because most variables correlate with the size of the firm.
Marris suggests the owners' utility function can be written as: *Uowners = f(gᶜ)**. The managerial utility function can be written as Uᴹ = f(gᴰ, s) where s is job security. Marris treats s as an exogenously determined constraint, so Uᴹ = f(gᴰ, s').
CONSTRAINTS
The managerial constraint and R&D capacity limit both gᴰ and gᶜ. The job security constraint is managed by avoiding risky investments and choosing prudent financial policies. Prudent financial policy involves determining optimal levels for three crucial financial ratios:
- Leverage ratio (debt ratio)
- Liquidity ratio
- Retention ratio
🔑 Definition — Three Financial Ratios:
- Liquidity ratio (a₁) = L/A (Liquid assets / Total assets)
- Leverage ratio (a₂) = D/A (Debt / Total assets)
- Retention ratio (a₃) = Πᴿ/Π (Retained profits / Total profits)
These are combined into a single parameter, ā (the financial security constraint), which is exogenously determined by top management's risk attitude. The overall constraint ā is negatively related to a₁ (liquidity) and positively related to a₂ (leverage) and a₃ (retention). There is a negative relation between job security (s) and the financial constraint (ā). The financial security constraint sets a limit to gᶜ.
THE MODEL: EQUILIBRIUM OF THE FIRM Managers maximize Uᴹ = f(gᴰ). Owners maximize Uᵒ = f*(gᶜ). The firm is in equilibrium when the maximum balanced growth rate is attained: *gᴰ = gᶜ = g (maximum)**.
The factors determining gᴰ and gᶜ can be expressed in terms of two variables: the diversification rate (d) and the average profit margin (m). The firm first determines its financial policy (ā), then chooses the diversification rate and profit margin to maximize g*.
📌 Example — Model Equilibrium (Figure 1):
- As the rate of growth of demand increases, profitability increases until a certain point (A). After that, a higher growth rate requires a reduction in the profit rate due to managerial constraints.
- The optimal position for managers is point B (the growth/utility maximization point), where the demand-growth and supply-growth curves intersect.
- The supply-growth curve shows a direct relationship between profitability and the maximum rate of capital growth.
- The model suggests that "owner-controlled firms achieve lower growth and higher profits."
POLICY VARIABLES IN THE MARRIS MODEL:
- The firm can change its balanced growth rate (g*) by changing the three security ratios (a₁, a₂, a₃).
- The firm can choose its diversification rate (d) by changing the style of existing products or expanding the product range.
- Price is not a policy variable in Marris's model; it is given by the oligopolistic industry structure.
THE RATE OF GROWTH OF DEMAND: gᴰ The firm grows by diversification (not merger or take-over). gᴰ = f₁(d, k), where d = diversification rate (number of new products per time period) and k = proportion of successful new products.
THE RATE OF GROWTH OF CAPITAL SUPPLY: gᶜ Corporate capital includes fixed assets, inventories, short-term assets, and cash reserves. Growth is financed from internal sources (profits) and external sources (new bonds or bank loans). Under Marris's assumptions: gᶜ = ā(Π), where ā is the financial security coefficient (constant and exogenous) and Π is the total profit level.
MARRIS MODEL CONTRIBUTIONS
- Incorporates financial policies of the firm into the decision-making process by introducing the financial coefficient ā as an additional (though exogenous) policy variable.
- Provides reconciliation between the conflicting utility functions of managers and owners.
CRITICISM OF MARRIS MODEL
- Assumes cost structure and price are given, thus assuming profit is given. Price determination, a major issue in firm theory, is ignored.
- Ignores the interdependence among firms' decisions in oligopolistic markets, meaning product differentiation is overlooked.
- Fails to explain oligopolistic interdependence in non-collusive firms, limiting the theory's applicability.
⭐ Key Takeaways
The lecture presents two major alternative theories to profit maximization. Baumol's model posits that firms maximize sales revenue subject to a minimum profit constraint, and with advertising, this constraint is always binding, leading to higher advertising spending than under profit maximization. Marris's model argues that firms maximize a balanced rate of growth of demand and capital, reconciling managerial and owner goals which are both linked to firm size. A critical distinction is that Marris introduces financial policy (the three ratios and the ā coefficient) and diversification rate as key decision variables, while price is taken as given by market structure. Both models highlight the separation of ownership and control, but are criticized for simplifying assumptions like ignoring competitive interdependence and price determination.
🧠 Quick Revision Questions
- In Baumol's sales maximization model with advertising, what is the key assumption about the marginal revenue of advertising (∂TR/∂A)?
- What is the central goal of the firm in Marris's model, and what condition defines equilibrium?
- Name the three financial ratios that constitute the financial security constraint in Marris's model and state how each is related to the overall constraint (ā)?
- According to the lecture, what is a major criticism that applies to both Baumol's and Marris's models regarding the market structure?
- In Marris's model, what are the two primary policy variables (other than the financial security coefficient) that the firm chooses to maximize the balanced growth rate?
📘 Lecture 38 — ALTERNATIVE THEORIES OF THE FIRM (CONTINUED 2)
📖 Overview: This lecture presents Williamson's Theory of Managerial Utility Maximization, a Nobel Prize-winning alternative to traditional profit-maximization assumptions. It explains how managers pursue their own utility through staff expenditure, emoluments, and discretionary investment, rather than simply maximizing profits. The model provides a more realistic view of firm behavior in less competitive markets.
🗂️ Topics Covered
Williamson's model of managerial utility maximization is introduced as an alternative to profit maximization, examining what managers want (utility from salary, power, status, and professional excellence), the basic relations and definitions (demand, production cost, actual profit, minimum profit, discretionary investment, and discretionary profit), a simplified model and its graphical analysis showing equilibrium where managerial indifference curves are tangent to the profit-staffing curve, the general model incorporating all three utility components, implications comparing Williamson's model to profit maximization, and criticisms of the theory.
📝 Lecture Summary
WILLIAMSON’S THEORY OF MANAGERIAL UTILITY MAXIMIZATION
The assumptions of profit-maximisation have been criticised, leading to two schools of thought: the "Managerial School" and the "Behavioural School". Oliver Williamson's model of utility maximisation of managerial utility function is a finale of the managerial utility models. He was awarded the Nobel Prize in Economic Science for 2009 for proposing a theory clarifying why some transactions take place inside firms and not in markets. Managers have different motives, desires, and aspirations which they want to maximise rather than maximising profit—including perks like big company cars, lavish offices, and luxurious bungalows.
💡 Why this matters: This model recognizes that real-world managers are not purely profit-driven automatons but have personal goals that influence firm decisions.
WHAT DO MANAGERS WANT?
UTILITY = happiness, satisfaction. What gives managers utility? Utility = f(Salary, power, status, professional excellence) expressed as Utility = f(S, M, I_D).
ASSUMPTIONS OF UTILITY MAXIMIZATION MODEL:
- Managers can act independently
- The firm's market is not highly competitive (firm can make supernormal profits)
- Managerial utility (U) is obtained from a combination of additional expenditure on staffing (S), managers' salaries and fringe benefits (M), and discretionary investment (I_D)
Williamson's model focuses on three areas: U = f₁(S, M, I_D)
- S = staff expenditure, including managerial salaries (administrative and selling expenditure)
- M = managerial emoluments
- I_D = discretionary investment
BASIC RELATIONS AND DEFINITIONS
The Demand of the Firm: The firm has a known downward-sloping demand curve defined by:
- X = f*(P, S, ε) or P = f₂(X, S, ε)
- Where X = output, P = price, S = staff expenditure, ε = condition of the environment (demand-shift parameter)
Demand is negatively related to price, but positively related to staff expenditure and to the shift factor ε. An increase in staff expenditure shifts the demand curve upward, allowing charging of a higher price.
THE PRODUCTION COST: Total production cost (C) is an increasing function of output: C = f₃(X) where δC/δX > 0.
ACTUAL PROFIT (π): Actual profit is revenue from sales (R), less production costs (C), and less staff expenditure (S): π = R – C – S This is profit reported to tax authorities—actual profit less managerial emoluments (M) which are tax deductible: π_R = π – M = R – C – S – M
🔑 Definition — Minimum Profit (π_O): The amount of profits (after tax) required for an acceptable dividend policy by shareholders. If shareholders do not receive some profit, they will sell shares or vote for a change in top management, reducing managers' job security. So managers ensure a minimum profit π_O sufficient to keep shareholders satisfied. The reported profits must be at least as high as the minimum profit requirement plus tax: π_R ≥ π_O – T
Where T = tax, with tax function: T = T' + t·π_R (t = marginal tax rate, T' = lump sum tax).
🔑 Definition — Discretionary Investment (I_D): The amount left from reported profit after subtracting minimum profit (π_O) and tax (T): I_D = π_R – π_O – T
🔑 Definition — Discretionary Profit (π_D): The amount of profit left after subtracting from actual profit (π) the minimum profit requirement (π_O) and tax (T): π_D = π – π_O – T
THE MODEL: A SIMPLIFIED MODEL OF MANAGERIAL DISCRETION
First stage assumes no managerial emoluments (M = 0), so actual profit equals reported profit.
Maximize U = f(S, I_D) subject to π > π_O + T
Since no emoluments, discretionary investment absorbs all discretionary profit. The managerial utility function becomes: U = f[S, (π – π_O – T)]
Assuming no lump-sum tax (T = tπ): U = f[S, (1 - t)π – π_O)] where (1 – t)π – π_O = π_D is discretionary profit.
The graphical presentation requires:
- Indifference curves map of managers (on axes measuring S and π_D)
- Curve showing relationship between S and π_D
Each indifference curve shows combinations of S and π_D giving same satisfaction to managers. Indifference curves are convex to the origin implying diminishing marginal rate of substitution of staff expenditure and discretionary profit. They do not intersect the axes—firm chooses positive levels of both π_D and S.
Figures 1 & 2 show the managerial indifference map with typical shape. Figure 3 shows the relationship between π_D and S determined by the profit function: π = f(X) = f(P, S, ε). Since t and π_O are exogenously given, and assuming output is chosen optimally (MC = MR), the π_D-S curve is derived.
The overall reported profit must satisfy shareholders (continued support) and pay taxes. Once these prior commitments are paid, the rest is at managers' discretion.
There is substitutability between S and π_D—the S-π_D curve shows trade-off between profit and staffing expenditure. Initially moving left to right along the S-π_D curve, both profit and staffing increase. After point S₁π_D₁, further increase in expenditure decreases profit—this is the profit maximizing point. If all reported profits were taken by shareholders/taxes, this would be optimal.
The model assumes this is not the case—managers will increase staff spending beyond this point for increased utility. U₁, U₂, U₃ are managerial indifference curves (U₃ preferred to U₂ preferred to U₁). Managers' ability to obtain utility is constrained by what the firm can afford.
📌 Example: In Figure 4, U₃ is out of reach. Managers choose point S₂π_D₂ where the S-π_D curve is tangent to U₂—the highest attainable utility level. At this managerial utility maximization point, staffing expenditure is higher and profit lower than if the firm pursued profit maximization.
THE GENERAL MODEL OF MANAGERIAL DISCRETION
Maximize U = f(S, M, π_R – π_O – T) subject to π_R ≥ π_O + T
Assuming positive but diminishing marginal utility for each component, the constraint becomes redundant. Substituting:
- π_R = π – M = R – C – S – M
- T = T' + t(R – C – S – M)
We obtain: U = f[S, M, {(1 - t)(R – C – S – M) – π_O}]
Defining σ as the ratio of retained to actual profit: σ = π_R/π, so π_R = π·σ. Solving for M: M = (1 – σ)π = (1 – σ)(R – C – S) where (1 – σ) is proportion of profits absorbed by emoluments.
The managerial utility function becomes: U = f[S, {(1 – σ)(R – C – S)}, {σ(1 – t)(R – C – S) – π_O}]
IMPLICATIONS OF THE MODEL
Comparing Williamson's model with profit maximization:
For the profit maximizer: π = R – C – S and π_R = π (σ = 1). The profit maximizer chooses X and S to maximize π = R – C – S.
First-order conditions for profit maximizer:
- (a) ∂π/∂X = ∂R/∂X – ∂C/∂X or ∂R/∂X = ∂C/∂X
- (b) ∂π/∂S = ∂R/∂S or ∂R/∂S = 1
Comparison Table:
| Condition | Williamson | Profit Maximizer |
|---|---|---|
| ∂R/∂X = ∂C/∂X | Yes | Yes |
| ∂R/∂S | < 1 | = 1 |
| σ | < 1 | = 1 |
| M (Emoluments) | > 0 | = 0 |
| S (Staffing) | > 0 | = 0 |
| I_D (Discretionary Investment) | > 0 | = 0 |
Staffing expenditure, managerial slack, and discretionary investment spending will be larger for a utility-maximizing firm than for a profit-maximizing firm.
Williamson's case studies support his model explaining real-world phenomena:
- Increase in S and M in booms, drastic cuts in recessions
- Reaction of firms to taxation changes
- Changes in X, S, and M in response to fixed cost changes
- Drastic cuts in S by newly appointed top management without affecting productivity
CRITICISM OF WILLIAMSON'S MANAGERIAL UTILITY MAXIMIZATION MODEL
- Available evidence is insufficient for verification of the theory
- The model fails to deal with the core problem of oligopolistic interdependence and strong oligopolistic rivalry—it holds only where rivalry is not strong; when rivalry is strong, a profit-maximizing model is more appropriate
⭐ Key Takeaways
Williamson's model fundamentally rejects the profit-maximization assumption, proposing instead that managers maximize their own utility from staff expenditure (S), managerial emoluments (M), and discretionary investment (I_D). The firm's equilibrium occurs where the highest attainable managerial indifference curve is tangent to the S-π_D trade-off curve, resulting in higher staffing, higher emoluments, and lower profits than under profit maximization. A key result is that the marginal revenue product of staff expenditure (∂R/∂S) is less than 1 under Williamson's model, versus exactly 1 under profit maximization. The model is most applicable in less competitive markets where managers have discretion, but fails to address strong oligopolistic rivalry.
🧠 Quick Revision Questions
- What are the three components of the managerial utility function in Williamson's model, and what does each represent?
- How does the equilibrium condition for staff expenditure (∂R/∂S) differ between Williamson's utility-maximizing firm and a profit-maximizing firm?
- Define discretionary investment (I_D) and explain how it relates to reported profit, minimum profit, and taxes.
- In the graphical analysis, why does the S-π_D curve first rise and then fall, and what is the significance of the tangency point?
- List two criticisms of Williamson's managerial utility maximization model.
📘 Lecture 39 — Alternative Theories of the Firm (Continued 3)
📖 Overview: This lecture continues the exploration of alternative theories to profit maximization, focusing on the Behavioral School. It explains why firms cannot maximize but instead "satisfice" due to bounded rationality, and details the Carnegie School's contributions, the multi-goal nature of firms, and how conflicts among stakeholders are resolved. This matters because it presents a more realistic, descriptive model of how actual corporations operate, contrasting sharply with traditional neoclassical assumptions.
🗂️ Topics Covered
The lecture covers the Behavioral School's main features and the Carnegie School's origins. It details the firm's goals in the satisficing model, the behavioral model of Cyert and March, and the concept of satisficing behavior. It then explains means for resolving conflict (money, side, and slack payments), uncertainty in the firm's environment, and provides a detailed comparison with traditional theory. Finally, it assesses the contributions and serious shortcomings of the behavioral theory.
📝 Lecture Summary
Behavioral School Theories
The Behavioral School, mainly presented by Herbert Simon, Richard Cyert and James March, challenges traditional profit-maximization assumptions. Its main features are: firms are multi-goal, multi-decision, multi-product organizational coalitions; managers have imperfect knowledge and bounded rationality; managers cannot meet the aspiration levels of all stakeholders and can never really know if they are maximizing profits, sales or growth. Consequently, managers cannot maximize; instead they have to satisfice.
🔑 Definition — Satisficing: A decision-making strategy that aims for a satisfactory or adequate result, rather than the optimal one, given limited information and cognitive abilities. 💡 Why this matters: This concept directly contradicts the traditional assumption of optimizing agents and provides a more psychologically realistic view of managerial decision-making.
Carnegie School
The 'Carnegie School' is identified with pioneering work in Behavioral Economics done by Herbert Simon, James G. March and Richard Cyert in the 1950s and 1960s at Carnegie Mellon University. These behavioralists focus on understanding how individuals and organizations act and make decisions in the real world, challenging neoclassical theories of optimization. Concepts like bounded rationality and satisficing were developed to describe decision-making in the face of 'the uncertainties and ambiguities of life'.
Firm's Goals: The Satisficing Model
The firm has multiple goals, not just profit. These include:
- Production Goal: Output must lie within a certain satisfactory range to ensure smooth running.
- Inventory Goal: Maintaining a balanced inventory of both raw materials and finished goods to ensure continuity of production and supply.
- Sales Goal: Achieving a satisfactory level of sales.
- Market Share Goal: Achieving a satisfactory size of market share as a measure of comparative success and firm growth.
- Profit Goal: Still an important goal, but one amongst a number rather than necessarily the goal of overriding importance.
Goals of the Firm: Satisficing Behaviour
The firm is not a single-goal, single-decision unit but a multi-goal, multi-decision organizational coalition of different groups (managers, workers, shareholders, customers, suppliers). Each group has its own set of goals. The goals are set by top management and take the form of aspiration levels. If attained, performance is considered 'satisfactory'. The firm is a satisficing, not a maximizing, organization.
This behavior is characterized by Simon as 'limited' or 'bounded rationality', as opposed to 'global' rationality. Traditional theory assumes unlimited, costless information and unlimited computational ability. Behavioral theory recognizes that managers have limited time, imperfect information, and limited computational ability. Therefore, they cannot examine all alternatives; they examine a small number and choose the 'best' given their constraints.
Means for the Resolution of the Conflict
The firm uses several means to satisfy the demands of its various coalition groups:
- Money payments: The major source of satisfying demands, paying factors their market prices.
- Side payments - Policy commitments: Allocating funds for specific projects to retain key personnel (e.g., funding a scientist's research plan).
- 'Slack' payments: Payments above what is required to keep a factor in the firm. Examples include wages higher than necessary for workers, higher salaries or perquisites (luxurious offices, limousines) for managers, higher dividends for shareholders, or unnecessary discounts for customers. This 'slack' is equivalent to economic rent.
Uncertainty and the Environment of the Firm
Cyert and March distinguish two types of uncertainty:
- Market uncertainty: Refers to changes in customers' preferences or production techniques. It is inherent and can be partly avoided by search activity but not completely.
- Uncertainty of competitors' reactions.
Given market uncertainty, the behavioral firm avoids long-term planning, works within a short time-horizon, and chooses to ignore the long-run consequences of short-run decisions.
A Comparison with the Traditional Theory
| Feature | Traditional Theory | Behavioral Theory |
|---|---|---|
| Nature of Firm | A single entrepreneur. | A coalition of groups with conflicting interests. |
| Ownership | Ownership and management are unified. | There is a dichotomy between ownership and management. |
| Firm's Objective | People do not have objectives; the organization does (profit). | "Organisations do not have objectives, only people have objectives." |
| Goal | A single goal: profit maximization. | A multiplicity of goals (aspiration levels). |
| Behavior | Maximizing. | Satisficing. |
| Rationality | 'Global' rationality (unlimited info & ability). | 'Bounded' or 'limited' rationality. |
| Response to Environment | Chooses policy variables to maximize long-run profits. | An adaptive organization that learns from experience and seeks satisficing levels. |
Behavioral Theory: Contributions
The main contributions are:
- Insight into the process of goal-formation and internal resource allocation, an aspect neglected in traditional theory.
- The analysis of 'slack' and its stabilizing role on the activity of the firm.
The Behavioral Theory: Serious Shortcomings
The theory has several major criticisms:
- It provides a simulation approach that predicts behavior but does not explain it.
- It does not deal with industry equilibrium or explain the interaction of firms.
- It gives no account of conditions of entry or the threat of potential entrants.
- It cannot explain the dynamic aspects of invention and innovation.
- It implies a short-sighted behavior that ignores the need for long-term planning.
- It resolves oligopolistic interdependence by assuming tacit collusion, which is an unstable solution.
- It is based on too few case studies to be a generalized theory.
- It has no predictive power whatsoever.
Which Approach is Most Useful?
The behavioral approach is a more accurate description of what happens inside the firm, but it tells us almost nothing about how the firm will respond to changes in the environment. However, if shareholders are powerful and demand maximum profit, the firm will behave like a profit-maximizer. Overall, alternative theories are useful complements to, not substitutes for, the traditional theory. The traditional theory of profit maximization still holds the ground firmly.
⭐ Key Takeaways
The Behavioral School, led by Simon, Cyert, and March, argues that firms are coalitions of groups with conflicting goals, not single-minded profit maximizers. Due to bounded rationality—limited information, time, and computational ability—managers cannot maximize profits; instead, they satisfice, setting and meeting satisfactory aspiration levels for production, inventory, sales, market share, and profit. The key concept of "slack" is an important mechanism for stabilizing the firm by paying stakeholders more than necessary. While behaviorism offers a more realistic internal view of the firm, its major shortcoming is a lack of predictive power regarding external market reactions, meaning the profit-maximization model remains the standard for predicting firm behavior.
🧠 Quick Revision Questions
- What are the two main schools of thought that criticize the profit-maximizing assumption of the firm, as introduced in this lecture?
- Define "bounded rationality" and explain how it leads to "satisficing" behavior.
- List the five main goals of the firm according to Cyert and March's satisficing model.
- What is "organizational slack" and what is its stabilizing role within the firm?
- What is the most significant shortcoming of the behavioral theory according to the lecture?
📘 Lecture 40 — RISK ANALYSIS
📖 Overview: This lecture examines risk and uncertainty in managerial decision-making, explaining how managers can quantify and measure risk when outcomes are not known with certainty. It introduces essential concepts such as probability distributions, expected values, standard deviation, and the coefficient of variation, providing tools for comparing and evaluating risky investment projects.
🗂️ Topics Covered
The lecture begins by differentiating between certainty, risk, and uncertainty in decision-making, introducing core concepts like strategy, states of nature, and payoff matrices. It then explains how to measure risk using probability distributions, calculate expected values, and assess absolute risk through standard deviation. Finally, it covers relative risk measurement using the coefficient of variation and demonstrates how to use the normal distribution and standardized variables for risk analysis.
📝 Lecture Summary
RISK AND UNCERTAINTY IN MANAGERIAL DECISION MAKING
Managerial decisions are made under three conditions: certainty, risk, or uncertainty. Certainty exists when there is only one possible outcome that is known precisely. For example, investing in Defense Savings Certificates from the Government of Pakistan yields a guaranteed 12.60% p.a. return, with only one possible outcome known with certainty. When more than one outcome is possible, risk or uncertainty is present.
RISK refers to a situation where there is more than one possible outcome and the probability of each specific outcome is known or can be estimated. Risk requires the decision maker to know all possible outcomes and have some idea of each outcome's probability. For instance, tossing a balanced coin has two possible outcomes (head or tail), each with a 50% probability. The greater the variability of possible outcomes, the greater the risk.
UNCERTAINTY occurs when there is more than one possible outcome but the probability of each outcome is not known or even meaningful, often due to insufficient past information or structural instability. In extreme cases, even the outcomes themselves are unknown. Drilling for oil in an unproven field exemplifies uncertainty when neither possible oil outputs nor their probabilities are known.
In analyzing risk, we use three key concepts: strategy, states of nature, and payoff matrix. A strategy is one of several alternative courses of action a decision maker can take to achieve a goal, such as building a large or small plant. States of nature are future conditions that significantly affect the success or failure of any strategy but over which the decision maker has little or no control, such as economic boom, normal, or recession. A payoff matrix is a table showing the possible outcomes or results of each strategy under each state of nature.
🔑 Definition — Probability: The chance that an event will occur, expressed as a decimal or percentage between 0 and 1.
🔑 Definition — Strategy: One of several alternative courses of action a decision maker can take to achieve a goal.
🔑 Definition — States of Nature: Future conditions that significantly affect the success or failure of any strategy but are beyond the decision maker's control.
🔑 Definition — Payoff Matrix: A table that shows the possible outcomes or results of each strategy under each state of nature.
MEASURING RISK WITH PROBABILITY DISTRIBUTIONS
PROBABILITY DISTRIBUTIONS list all possible outcomes of an event and the probability attached to each. The sum of all probabilities equals 1.0 (or 100%), since one of the outcomes must occur with certainty. For example, Table 1 shows a probability distribution with three states of the economy: boom (0.25), normal (0.50), and recession (0.25), totaling 1.00.
Table 1: Probability Distribution of States of the Economy
| State of the Economy | Probability of Occurrence |
|---|---|
| Boom | 0.25 |
| Normal | 0.50 |
| Recession | 0.25 |
| Total | 1.00 |
The expected value or expected profit of an investment is calculated by multiplying each possible outcome by its probability and summing these products. This gives the weighted average of all possible profit levels.
📐 Formula: E(π) = π̄ = Σ(πᵢ · Pᵢ) for i = 1 to n Where πᵢ is the profit level associated with outcome i, Pᵢ is the probability that outcome i will occur, and n is the number of possible outcomes.
Table 2 presents the payoff matrix for two projects (A and B) with identical expected profits of $500 but different risk levels.
Table 2: Payoff Matrix and Expected Values
| Project | State of Economy | Probability (P) | Outcome (π) | Expected Value |
|---|---|---|---|---|
| A | Boom | 0.25 | $600 | $150 |
| A | Normal | 0.50 | $500 | $250 |
| A | Recession | 0.25 | $400 | $100 |
| Expected profit from Project A | $500 | |||
| B | Boom | 0.25 | $800 | $200 |
| B | Normal | 0.50 | $500 | $250 |
| B | Recession | 0.25 | $200 | $50 |
| Expected profit from Project B | $500 |
Project A has a smaller range of outcomes ($400 to $600) compared to Project B ($200 to $800), making Project A less risky and preferable for a risk-averse manager.
📌 Example: Using Table 2, the expected profit for Project A is calculated as ($600 × 0.25) + ($500 × 0.50) + ($400 × 0.25) = $150 + $250 + $100 = $500. For Project B: ($800 × 0.25) + ($500 × 0.50) + ($200 × 0.25) = $200 + $250 + $50 = $500.
A discrete probability distribution shows outcomes as separate bars (Figure 1), while a continuous probability distribution shows a smooth curve as more states of nature are specified (Figure 2). The tighter (less dispersed) the distribution, the lower the risk.
💡 Why this matters: The expected value alone is insufficient for decision-making; managers must also consider the dispersion or variability of outcomes to fully assess risk.
ABSOLUTE RISK MEASUREMENT
The standard deviation (σ) is a popular measure of absolute risk, which is the overall dispersion of possible payoffs. The smaller the standard deviation, the tighter the probability distribution and the lower the absolute risk. Standard deviation measures the dispersion of possible outcomes from the expected value.
📐 Formula: σ = √[Σ(X - X̄)² · P] Where X is each outcome, X̄ is the expected value (mean), P is the probability, and the sum is taken over all possible outcomes.
For Project A (from Table 2): σ = √[(600-500)²(0.25) + (500-500)²(0.50) + (400-500)²(0.25)] σ = √[2,500 + 0 + 2,500] = √5,000 = $70.71
For Project B (from Table 2): σ = √[(800-500)²(0.25) + (500-500)²(0.50) + (200-500)²(0.25)] σ = √[22,500 + 0 + 22,500] = √45,000 = $212.13
📌 Example: The standard deviation of Project A ($70.71) is much smaller than that of Project B ($212.13), numerically confirming that Project A has less dispersion and lower absolute risk, matching the graphical evidence from Figures 1 and 2.
A RELATIVE MEASURE OF RISK: THE COEFFICIENT OF VARIATION
To measure relative dispersion, we use the coefficient of variation (v) , which equals the standard deviation divided by the expected value (mean).
📐 Formula: v = σ / X̄ Where σ is the standard deviation and X̄ is the expected value or mean.
The coefficient of variation measures the standard deviation per dollar of expected value. It is a pure number (dimension-free) that can compare the relative risk of two or more projects. The project with the largest coefficient of variation is the most risky.
For Projects A and B (both with expected value $500): Project A: v = $70.71 / $500 = 0.1414 Project B: v = $212.13 / $500 = 0.4243
Project A has a smaller coefficient of variation (0.1414) than Project B (0.4243), confirming that Project A has less relative risk.
💡 Why this matters: The coefficient of variation is useful when comparing projects with different expected values, as it standardizes risk per unit of expected return.
MEASURING PROBABILITIES WITH THE NORMAL DISTRIBUTION
A normal distribution is a symmetrical distribution about the mean or expected value. The relationship between standard deviation and probability in a normal distribution follows specific rules:
- Actual outcomes lie within ±1σ of the mean: approximately 68% of the time
- Actual outcomes lie within ±2σ of the mean: approximately 95% of the time
- Actual outcomes lie within ±3σ of the mean: greater than 99% of the time
The smaller the standard deviation, the tighter the distribution about the expected value and the smaller the probability of an outcome very different from the expected value.
STANDARDIZED VARIABLES have a mean of zero and a standard deviation of one. Any distribution of revenue, cost, or profit data can be standardized using the following formula:
📐 Formula: Z = (x - μ) / σ Where z is a standardized variable, x is a point of interest, μ is the mean, and σ is the standard deviation.
When the point of interest is 1 standard deviation away from the mean, x - μ = σ, so z = σ/σ = 1.0. When z = 1.0, the value is 1σ from the mean; when z = 2.0, the value is 2σ from the mean; and so on. Although the standard normal distribution theoretically runs from minus infinity to plus infinity, the probability of occurrences beyond 3 standard deviations is very near zero.
📌 Example: If the mean profit is $500 and the standard deviation is $70.71, then a profit of $570.71 would have a z-score of ($570.71 - $500) / $70.71 = 1.0, meaning it is 1 standard deviation above the mean.
⭐ Key Takeaways
The most critical distinction to remember is the difference between risk (where probabilities are known) and uncertainty (where probabilities are unknown). Risk is measured using probability distributions, with the expected value representing the weighted average of all possible outcomes. The standard deviation measures absolute risk as the dispersion of outcomes from the mean, while the coefficient of variation standardizes this risk per unit of expected value, allowing comparison across projects with different means. In a normal distribution, approximately 68% of outcomes fall within ±1 standard deviation, 95% within ±2 standard deviations, and over 99% within ±3 standard deviations. Standardized variables (z-scores) transform any distribution to have a mean of zero and standard deviation of one, facilitating probability calculations.
🧠 Quick Revision Questions
- What is the key difference between risk and uncertainty in managerial decision-making?
- How do you calculate the expected value (expected profit) of an investment project?
- What does the standard deviation measure in the context of risk analysis, and how is it calculated?
- When would you use the coefficient of variation instead of the standard deviation to compare risk?
- In a normal distribution, what percentage of outcomes fall within ±2 standard deviations of the mean?