Deriving Opportunity Cost from a Frontier’s Slope

Why the Frontier’s Slope Measures Opportunity Cost

A production possibility frontier (PPF) shows combinations of two goods that can be produced with the available resources and technology. Moving along the frontier to produce more of one good means giving up some of the other. The slope records that trade-off: it tells us how much of the vertical-axis good must be forgone for an additional unit of the horizontal-axis good.

This tutorial uses \(X\) for the good on the horizontal axis and \(Y\) for the good on the vertical axis. Assume resources and technology are fixed and production is on the frontier. The frontier slopes downward: increasing \(X\) requires reducing \(Y\). Keep track of the units as well as the sign. A slope of \(-2\) means a sacrifice of 2 units of \(Y\) for each additional unit of \(X\), not a negative quantity of goods given up.

Definition: Over a move along a downward-sloping frontier that increases \(X\), the opportunity cost of \(X\) is the amount of \(Y\) forgone per additional unit of \(X\). For a finite change, it is \(-\Delta Y/\Delta X\). The opportunity cost of \(Y\) over that same move is \(-\Delta X/\Delta Y\), the amount of \(X\) forgone per additional unit of \(Y\) gained.

Opportunity Cost on a Linear Frontier

A straight-line frontier has the same slope everywhere. Consequently, the opportunity cost is constant: every additional unit of \(X\) costs the same amount of \(Y\), regardless of the starting production point. If the frontier is written \(Y=a-bX\), where \(b>0\), its slope is \(-b\). Producing one more unit of \(X\) requires giving up \(b\) units of \(Y\), so the opportunity cost of \(X\) is \(b\) units of \(Y\).

\text{Opportunity cost of }X=-\frac{\Delta Y}{\Delta X}=b, \qquad \text{Opportunity cost of }Y=-\frac{\Delta X}{\Delta Y}=\frac{1}{b}.

The second expression is the reciprocal because the two costs describe opposite directions of the same trade-off. It is important to retain the units: if the cost of \(X\) is \(b\) units of \(Y\) per unit of \(X\), the cost of \(Y\) is \(1/b\) units of \(X\) per unit of \(Y\).

Worked Example: Reading Both Costs from a Linear Frontier

A workshop’s frontier is \(4X+3Y=240\), where \(X\) is measured in batches of insulated panels and \(Y\) in batches of thermal covers. Find the opportunity cost of each good. Then check the cost of increasing panel production from 15 batches to 30 batches.

Method. Rearrange the frontier to express the vertical-axis good in terms of the horizontal-axis good:

4X+3Y=240 \quad\Longrightarrow\quad 3Y=240-4X \quad\Longrightarrow\quad Y=80-\frac{4}{3}X.

The slope is \(-4/3\). Thus the opportunity cost of \(X\) is \(4/3\) batches of \(Y\) for each additional batch of \(X\). The reciprocal gives the opportunity cost of \(Y\): \(3/4\) batch of \(X\) per additional batch of \(Y\).

Check using the specified production change. At \(X=15\), \(Y=80-(4/3)(15)=80-20=60\). At \(X=30\), \(Y=80-(4/3)(30)=80-40=40\). The change is \(\Delta X=30-15=15\) and \(\Delta Y=40-60=-20\). Therefore:

-\frac{\Delta Y}{\Delta X} =-\frac{-20}{15} =\frac{4}{3}.

The finite-change calculation agrees with the slope. The common mistake this method avoids is reporting \(-4/3\) batches as the cost: the negative sign describes the direction of the slope, while the opportunity cost is the positive amount of \(Y\) forgone.

Curved Frontiers and Changing Marginal Cost

A bowed-outward PPF is not a straight line, so its slope changes as production shifts along it. With \(X\) on the horizontal axis, the usual bowed-outward shape becomes steeper in magnitude as production of \(X\) increases. That means each additional unit of \(X\) requires giving up more \(Y\) than the previous unit did: the marginal opportunity cost of \(X\) rises.

For a smooth frontier \(Y=f(X)\), the slope at a particular production point is the derivative \(dY/dX\). This is a local rate of change. Because the frontier slopes downward, the marginal opportunity cost of \(X\) is the positive quantity \(-dY/dX\). For a finite move, use the changes between the two endpoints instead. The finite-change cost is an average over that interval; it need not equal the marginal cost at either endpoint.

Key result: On a differentiable downward-sloping frontier, marginal opportunity cost of \(X\) at a point is \(-dY/dX\). If the frontier is bowed outward with \(d^2Y/dX^2<0\), its slope becomes more negative as \(X\) rises, and the marginal opportunity cost of \(X\) rises. The tangent-slope formula applies at a point; a finite production change must be evaluated using the frontier’s endpoint values.

Worked Example: Calculating Average Costs Along a Bowed Frontier

A farm’s frontier is \(Y=100-0.01X^2\), with \(X\) measured in crates of one crop and \(Y\) in crates of another. Calculate the average opportunity cost of \(X\) over the intervals from 20 to 30 crates and from 60 to 70 crates. Then compare those averages with the marginal costs at the intervals’ midpoints.

Method. Evaluate the frontier at each endpoint and divide the reduction in \(Y\) by the increase in \(X\).

For the interval from \(X=20\) to \(X=30\):

Y(20)=100-0.01(20^2)=100-4=96, \qquad Y(30)=100-0.01(30^2)=100-9=91.

So \(\Delta X=30-20=10\), while \(\Delta Y=91-96=-5\). The average opportunity cost is \(-(-5)/10=0.5\) crate of \(Y\) per crate of \(X\).

For the interval from \(X=60\) to \(X=70\):

Y(60)=100-0.01(60^2)=100-36=64, \qquad Y(70)=100-0.01(70^2)=100-49=51.

Here \(\Delta X=70-60=10\) and \(\Delta Y=51-64=-13\). The average opportunity cost is \(-(-13)/10=1.3\) crates of \(Y\) per crate of \(X\).

Differentiate the frontier to find the marginal cost at a point:

\frac{dY}{dX}=-0.02X, \qquad -\frac{dY}{dX}=0.02X.

The first interval’s midpoint is \(X=25\), where marginal cost is \(0.02(25)=0.5\). The second interval’s midpoint is \(X=65\), where marginal cost is \(0.02(65)=1.3\). For this quadratic frontier, each interval’s average equals its midpoint’s marginal cost. The average costs rise from 0.5 to 1.3, consistent with the frontier becoming steeper. Do not assume that an interval average always equals a midpoint marginal cost; that agreement follows here from the particular quadratic function.

Why the Marginal Formula Holds

The derivative is the limit of the frontier’s slope over a small change in production. Suppose \(X\) increases by a small amount \(\Delta X>0\). Since production remains on the frontier, the associated change in \(Y\) is approximately \((dY/dX)\Delta X\). On a downward-sloping frontier this change is negative. The amount of \(Y\) given up per unit of \(X\) is therefore:

-\frac{\Delta Y}{\Delta X} \ \longrightarrow\ -\frac{dY}{dX} \quad\text{as }\Delta X\to 0^+.

This justifies interpreting the negative derivative as marginal opportunity cost. If the frontier is bowed outward, \(d^2Y/dX^2<0\): as \(X\) rises, \(dY/dX\) becomes more negative. Its negative, \(-dY/dX\), therefore becomes larger.

Worked Example: Using a Tangent Slope and Its Reciprocal

Use the frontier \(Y=100-0.01X^2\). At \(X=40\), find the marginal opportunity cost of \(X\), the marginal opportunity cost of \(Y\), and the exact change in \(X\) if \(Y\) falls by one crate from its value at that point. State why the exact change need not equal the local marginal estimate.

At \(X=40\), the vertical coordinate is:

$$ Y(40)=100-0.01(40^2)=100-16=84. $$

The frontier’s slope there is \(-0.02(40)=-0.8\). Thus the marginal opportunity cost of \(X\) is \(0.8\) crate of \(Y\) per crate of \(X\). The marginal opportunity cost of \(Y\) is the reciprocal, \(1/0.8=1.25\) crates of \(X\) per crate of \(Y\).

Now reduce \(Y\) by one crate, from 84 to 83. Solve the frontier for the new \(X\):

83=100-0.01X^2 \quad\Longrightarrow\quad 0.01X^2=17 \quad\Longrightarrow\quad X^2=1700 \quad\Longrightarrow\quad X=10\sqrt{17}.

The positive root is required because production \(X\) is nonnegative. The exact increase is \(10\sqrt{17}-40\), approximately \(1.231\) crates of \(X\) (rounded to three decimals). The tangent-based marginal estimate is \(1.25\) crates for a one-crate reduction in \(Y\). These differ because the exact change follows the curved frontier over a finite interval, whereas the reciprocal tangent slope is local. The calculation avoids treating a marginal rate as an exact finite-change quantity.

Common Slope and Reciprocal Errors

The sign and units provide quick checks on a calculation. If \(X\) is increased along a downward-sloping frontier, \(Y\) must fall. Thus \(\Delta X>0\), \(\Delta Y<0\), the slope \(\Delta Y/\Delta X\) is negative, and the opportunity cost \(-\Delta Y/\Delta X\) is positive.

A frequent error is to report the slope itself as the cost. For example, if the slope is \(-0.8\) units of \(Y\) per unit of \(X\), the opportunity cost of \(X\) is \(0.8\) units of \(Y\) per unit of \(X\), not \(-0.8\). Another error is to use the same numerical rate for both goods. If the cost of \(X\) is \(0.8\) units of \(Y\), the reciprocal marginal cost of \(Y\) is \(1/0.8=1.25\) units of \(X\). On a curved frontier, this reciprocal statement applies to marginal costs at the same point; finite-change average costs over a large interval are not generally reciprocals.

Opportunity cost is useful because it gives a common, quantitative description of the production trade-off. When interpreting a result, always identify which good is being increased, which is being sacrificed, and the units of the rate. On a line, that rate is constant; on a bowed frontier, it depends on the production point.

Check Your Understanding

For each problem, identify the relevant slope or endpoint changes and keep the opportunity-cost units explicit.

A linear frontier is \(5A+2B=100\), with \(A\) on the horizontal axis. Find the opportunity cost of one unit of \(A\) in units of \(B\), and of one unit of \(B\) in units of \(A\). (Answer: 2.5 units of \(B\) per unit of \(A\); 0.4 units of \(A\) per unit of \(B\).)
A straight frontier joins \((X,Y)=(0,90)\) and \((30,0)\). Find the opportunity cost of \(X\), and verify it by calculating how much \(Y\) is given up when \(X\) rises from 12 to 20. (Answer: 3 units of \(Y\) per unit of \(X\); \(Y\) falls from 54 to 30, so 24 units are given up over 8 units of \(X\).)
A frontier is \(Y=144-0.04X^2\). Calculate the marginal opportunity cost of \(X\) at \(X=20\) and at \(X=40\), and state how it changes. (Answer: 1.6 units of \(Y\) per unit of \(X\) at \(X=20\), and 3.2 at \(X=40\); it doubles.)
On the frontier \(Y=100-0.01X^2\), calculate the average opportunity cost of \(X\) as production rises from 40 to 50. (Answer: \(Y(40)=84\), \(Y(50)=75\), so 9 units of \(Y\) are given up for 10 units of \(X\), or 0.9 unit of \(Y\) per unit of \(X\).)
At a point on a smooth frontier, \(dY/dX=-2.4\). State both marginal opportunity costs with their units. (Answer: 2.4 units of \(Y\) per unit of \(X\), and \(1/2.4=0.4167\) units of \(X\) per unit of \(Y\), to four decimals.)

What You'll Learn