Marginal Analysis and Decision-Making

Marginal analysis and decision-making
Marginal analysis — one of the main tools of economic thinking. It is based on a simple idea: optimal decisions are made "at the margins," comparing additional benefits with additional costs of each next step.

What does "marginal" mean
Marginal means "additional," "incremental," "one more unit." Marginal analysis does not ask "how much in total," but rather "is one more worth it?"

Marginal benefit (MB) — the additional benefit from consuming or producing one more unit.
Marginal cost (MC) — the additional cost from producing or consuming one more unit.

Optimality principle: an action should be continued as long as MB ≥ MC. Stop when MB = MC. This is the point where net benefit (total benefit − total cost) is maximized.

Examples of marginal thinking

Consumer: Should I eat yet another slice of pizza?
MB — pleasure from the slice.
MC — money, calories, feeling of overeating.
First slice: MB is high (hungry!), MC is low.
Fifth slice: MB is low (already full), MC increases (discomfort).
Somewhere between the first and fifth, MB = MC — the optimum.

Firm: Should I hire one more worker?
MB — additional product produced by him (marginal revenue product).
MC — his salary and related expenses.
Hire as long as MB ≥ MC.

Investor: Should I add one more stock to the portfolio?
MB — improved diversification, potential yield.
MC — transaction costs, time for analysis, increased portfolio complexity.
At some point, the costs of complexity outweigh the benefits of diversification.

Diminishing marginal utility

Law of diminishing marginal utility: each additional unit of a good yields less satisfaction than the previous one.
The first sip of water on a hot day — bliss.
The tenth — already normal.
The twentieth — unpleasant.
MB falls with each unit.

Consequence: the more you already consume, the less you are willing to pay for the next unit. This explains the downward slope of the demand curve.

Diminishing marginal returns

Law of diminishing marginal returns: with some inputs fixed, adding a unit of a variable input delivers ever-smaller increases in output.

You have a cafe with 10 tables.
One waiter — overloaded.
Second — sharply increases efficiency.
Third — helps some more.
The tenth — now interferes, waiters push each other.
Marginal product of labor decreases.

Consequence: marginal cost of production rises. To produce one more unit, more and more resources are needed. This explains the upward slope of the supply curve.

Marginal analysis vs average values

It is important to distinguish between marginal and average indicators:

Average cost (AC): TC / Q — how much each unit of output costs on average.
Marginal cost (MC): ΔTC / ΔQ — the cost of the additional unit.

Decisions are made based on marginal values, not average ones.

Example: Airline sold 190 tickets for a flight with a capacity of 200 seats.
Average cost per passenger — $200.
Should the last 10 tickets be sold for $50 each?

Incorrect answer (based on average): "No, $50 is less than $200, we incur losses."
Correct answer (marginal): "Yes, if the marginal cost for the extra passenger (food, fuel, cleaning) is less than $50. The plane flies anyway, most costs are fixed."

Sunk costs and marginal analysis

Marginal analysis ignores sunk costs — already incurred and irreversible expenses. They do not affect the optimal current decision.

Example:
You paid $10 for a movie ticket. The film is awful. Should you finish watching?

Incorrect: "I paid, so I must finish watching" (sunk cost fallacy).
Correct: $10 is already spent and will not return. The question: what to do with the next two hours? If leaving is more pleasant than staying, leave.

For the investor: loss-making positions — sunk cost. The question is not "how much I lost," but "what to do with the remaining money now."
If the stock's prospects are poor, sell, regardless of purchase price.

Marginal analysis in investments

Project evaluation:
Marginal return on investment vs marginal cost of capital. Invest as long as the return on each additional dollar invested exceeds the cost of raising that dollar.

Position size:
The first percent of the portfolio in a new idea — high marginal diversification benefit.
The tenth percent — now concentration risk.
Optimal size is where marginal benefit = marginal risk.

Rebalancing: Should you rebalance?
MB — return to target weights, risk control.
MC — transaction costs, taxes, time.
Rebalance when deviation is large enough that MB > MC.

Applications in business

Pricing: Optimal price is where marginal revenue = marginal cost.
Dropping price increases sales volume (MR), but reduces margin.
Balance — maximum profit.

Production: Produce more as long as MR ≥ MC.
Stop when the additional unit no longer covers its costs.

Hiring: Hire as long as the marginal revenue product of the worker ≥ his salary.

Limitations

Marginal analysis assumes:

Divisibility: one can add "one more unit." Not always so (you cannot build half a factory).

Measurability: it is possible to assess MB and MC. This is not always obvious.

Rationality: agents are able to compare and optimize. Behavioral economics shows deviations.

Nevertheless, marginal thinking is a powerful tool for structuring decisions: not "all or nothing," but "is one more step worth it."