What does marginal mean in economics? Thinking at the margin

In economics, marginal means "one more": the extra cost, benefit or satisfaction that comes from one additional unit of something, not from the total. Marginal cost is what one more pizza adds to a shop's costs; marginal utility is how much more you enjoy one more slice. What the word does mean, in practice, is a way of deciding: keep doing something while the next unit adds more benefit than cost, and stop when it doesn't.

In short

• Marginal means the change from one more unit: the extra cost, revenue or enjoyment, not the average or the total.
• The rule of marginal thinking: keep going while marginal benefit is greater than marginal cost.
• Marginal value usually falls as you have more of something, which is why water is cheap and diamonds are dear.

What does marginal mean, exactly?

The margin is the edge of a decision: the next hour, the next worker, the next unit made or bought. A marginal quantity measures what changes when you move that edge by one. OpenStax's Principles of Economics defines marginal cost as "the cost of producing one more unit of output," and writes it as a formula:

MC = ΔTC ÷ ΔQ

Here MC is marginal cost, TC is total cost, Q is the quantity produced, and Δ (the Greek letter delta) means "the change in." If making 20 more pizzas raises total cost by $120, the marginal cost is $120 ÷ 20, or $6 a pizza. The same pattern gives marginal revenue (the extra money from one more sale), marginal benefit (the extra value to the buyer) and marginal utility (the extra satisfaction). In the limit, where the step shrinks to almost nothing, marginal cost is the slope of the total cost curve, the derivative of cost with respect to quantity.

Marginal doesn't mean small or unimportant, as it does in everyday speech. A government deciding whether to build one more hospital is making a marginal decision about a billion dollars.

How does a pizza shop use marginal cost?

Take a shop deciding how many pizzas to make on a Friday evening. It has $300 of fixed costs for the evening, rent and the oven, which it pays whatever it makes. Every pizza adds dough, toppings and staff time, and past a point the kitchen gets crowded and each pizza takes longer. Pizzas sell for $10.

Pizzas	Total cost	Marginal cost per extra pizza	Revenue at $10	Profit
0	$300	—	$0	−$300
20	$420	$6	$200	−$220
40	$520	$5	$400	−$120
60	$620	$5	$600	−$20
80	$740	$6	$800	$60
100	$900	$8	$1,000	$100
120	$1,140	$12	$1,200	$60

Read down the marginal cost column. Each pizza from the 81st to the 100th costs $8 to make and sells for $10, so each adds $2 of profit. Each pizza from the 101st to the 120th costs $12 to make and sells for $10, so each loses $2. The shop makes the most profit at 100 pizzas, exactly where the next pizza would cost more than it brings in.

Average cost would have misled it. At 120 pizzas, the average cost is $1,140 ÷ 120, or $9.50, below the $10 price, so every pizza looks profitable on average. Only the margin shows that the last 20 lose money. The fixed $300 doesn't change the answer either: it is paid whether the shop makes 80 pizzas or 120, so it plays no part in deciding whether to make one more. Letting a cost already paid steer the next choice is the sunk cost fallacy.

What is marginal utility? The third slice

Marginal utility is the extra satisfaction from one more unit of something, and for most things it falls as you have more. OpenStax calls this diminishing marginal utility. Suppose you'd pay up to these amounts for slices of pizza at lunch:

Slice	Most you'd pay for it	Price	Buy it?
1st	$6.00	$3	Yes: $3 of value to spare
2nd	$4.00	$3	Yes: $1 to spare
3rd	$2.00	$3	No: it's worth $1 less than it costs
4th	$0.50	$3	No

You buy two. The third slice is still pleasant, it's worth $2 to you, but it costs $3, and that $3 could buy something else you'd value more, its opportunity cost. Notice that the decision never compares lunch as a whole with the price of lunch as a whole. It is made one slice at a time, at the margin. Falling marginal utility is also why demand curves slope down: the lower the price, the further down the list of slices a buyer goes, the logic behind supply and demand.

Why is water cheap and diamonds expensive?

Adam Smith posed the puzzle in The Wealth of Nations in 1776: "Nothing is more useful than water," yet it buys almost nothing in exchange, while a diamond, of little practical use, buys a great deal. Smith separated "value in use" from "value in exchange" but didn't resolve the gap.

Marginal utility resolves it. The total value of water is enormous, since without it people die. But most people have so much water that the marginal unit, one more liter from the tap, goes to watering a lawn or a longer shower, and is worth very little. Diamonds are scarce, so even the marginal diamond is still wanted badly by someone. Prices follow the value of the last unit, not the total value of all units. A person lost in a desert, with no water at all, would swap a diamond for a liter without hesitating, because there the marginal liter is worth their life.

That answer came in the early 1870s, when three economists working separately, William Stanley Jevons in England, Carl Menger in Austria and Léon Walras, a Frenchman teaching in Switzerland, put marginal utility at the center of value. Historians call it the marginal revolution, as Wikipedia's article on marginalism describes, and it is the foundation of how economics has explained prices since.

What is the rule for thinking at the margin?

The rule is short: keep doing something while its marginal benefit is greater than its marginal cost, and stop where they meet. It works for a firm choosing output, a shopper choosing slices, and a student choosing how long to revise. The first hour of revision for an exam may raise the mark a lot; the fifth, much less, once the main material is covered and tiredness sets in. That falling return is the law of diminishing returns, and it is why the margin, not the average, decides when to stop.

Marginal thinking also explains why "all or nothing" questions are usually the wrong ones. A city doesn't choose between having police and having none; it chooses whether one more officer is worth more than what the same money would do elsewhere. Asking about the margin turns impossible questions into answerable ones.

Is a marginal tax rate the same idea?

Yes, and it clears up a common worry. A marginal tax rate is the rate on the next dollar of income, not on all of it. Under a system of brackets, such as US federal income tax, a raise that crosses into a higher bracket is taxed at the higher rate only on the part above the line, so the raise itself never leaves you with less pay after income tax. (Some benefits that stop at an income limit can create real cliffs, which is a separate issue.) The average rate, total tax divided by total income, is never higher than the top marginal rate a person pays, and for most taxpayers it is well below it. This explains the arithmetic; it isn't tax advice.

Check yourself

• In the pizza table, why does the shop stop at 100 pizzas even though the average cost at 120 is below the price?
• Why don't fixed costs affect the decision to make one more pizza?
• Explain, using marginal utility, why water is cheap even though it is more useful than diamonds.

Sources

• 7.3 Costs in the Short Run — OpenStax, Principles of Economics 3e, 2022
• 6.1 Consumption Choices — OpenStax, Principles of Economics 3e, 2022
• An Inquiry into the Nature and Causes of the Wealth of Nations — Adam Smith, 1776, Library of Economics and Liberty
• Marginalism — Wikipedia
• Paradox of value — Wikipedia