Long-Run Production Costs - Microeconomics

SECTION 1

Historical Context & Motivation

The study of production costs has been central to economic thought since the classical economists first attempted to explain how goods are priced and why some firms thrive while others fail. In the short run, firms are constrained by fixed inputs—a factory of a given size, a set number of machines—so their cost structure is partly predetermined. The long run, however, represents a planning horizon in which all inputs are variable, allowing firms to select the combination of labor, capital, and technology that minimizes cost for any desired output level. Understanding this distinction is essential for business strategists, because it explains why firms build larger plants, merge with competitors, or outsource operations—decisions that reshape entire industries.

1776 • Adam Smith & Division of Labor

In The Wealth of Nations, Smith observed that specialization within a pin factory dramatically lowered per-unit costs, planting the seed for the concept of economies of scale.

1890 • Alfred Marshall's Cost Curves

Marshall formalized the distinction between short-run and long-run costs in his Principles of Economics, introducing the concept of the representative firm and the envelope relationship between short-run and long-run average cost curves.

1931 • Jacob Viner's Envelope Curve

Viner published his celebrated article on cost curves, demonstrating geometrically that the long-run average cost curve is the lower envelope of all possible short-run average cost curves—a result that became a staple of microeconomic pedagogy.

1970s • Minimum Efficient Scale in Industrial Organization

Empirical researchers measured minimum efficient scale across industries, connecting long-run cost theory to real-world market structure, entry barriers, and antitrust policy.

The central question that long-run cost analysis addresses is deceptively simple: How does a firm's cost per unit of output change as it scales its operations when no input is fixed? The answer shapes whether an industry gravitates toward many small competitors, a few large oligopolists, or a single natural monopoly—making it foundational to the study of competitive equilibrium.

SECTION 2

Core Principles & Definitions

Before diving into cost curves and mathematical optimization, it is important to establish the foundational ideas that govern long-run production costs. Unlike the short run, where at least one factor of production is fixed, the long run is defined not by a calendar duration but by the condition that every input can be adjusted. A restaurant choosing between leasing a small storefront or constructing a large dining hall is making a long-run decision; once the lease is signed and the kitchen is built, the firm moves into the short run with those inputs fixed.

Long-Run Total Cost (LRTC)

The minimum total cost of producing each output level when all inputs—labor, capital, land, and technology—are fully adjustable. Because there are no fixed costs in the long run, LRTC starts at the origin.

Long-Run Average Cost (LRAC)

Defined as LRTC ÷ Q, this curve traces the lowest attainable per-unit cost at every output level. Its characteristic U-shape reflects economies of scale on the left, constant returns in the middle, and diseconomies on the right.

Long-Run Marginal Cost (LRMC)

The additional cost of producing one more unit when all inputs are optimally adjusted. LRMC intersects LRAC at the latter's minimum point—mirroring the relationship between any marginal and average function.

Economies & Diseconomies of Scale

Economies of scale occur when doubling all inputs more than doubles output, causing LRAC to decline. Diseconomies of scale arise when managerial complexity or coordination costs cause LRAC to rise.

Minimum Efficient Scale (MES)

The smallest output level at which LRAC reaches its minimum. Industries with a high MES relative to market demand tend toward fewer, larger firms.

✦

KEY TAKEAWAY

Think of the long run as an architect's drafting table: you can redesign the entire factory from scratch—choosing any combination of floor space, equipment, and workforce—to produce at the lowest possible cost. The LRAC curve is the set of blueprints that shows the cheapest per-unit cost for every conceivable production volume. Once you break ground on a specific blueprint, you have committed to a short-run plant, and your cost flexibility narrows.

SECTION 3

The Envelope Curve — Visual Explanation

The most important visual in long-run cost theory is the envelope curve. Imagine a firm that can choose among three plant sizes—small, medium, and large—each with its own short-run average total cost (SRATC) curve. In the long run, the firm selects whichever plant size yields the lowest average cost for its target output. The LRAC curve is formed by tracing the lowest portions of all possible SRATC curves, creating a smooth envelope that just "kisses" each short-run curve without crossing above it.

The gold LRAC curve traces the lowest portions of each dashed SRATC curve (small plant in cyan, medium in violet, large in pink). The green dashed line marks the minimum efficient scale (MES) at Q*, where LRAC reaches its minimum. To the left of Q*, the firm enjoys economies of scale; to the right, diseconomies set in.

Several features of this diagram merit attention. First, note that the LRAC curve does not pass through the minimum of every SRATC curve. For plant sizes below the MES, the tangency point lies on the downward-sloping portion of the SRATC, because a slightly larger plant could produce the same output at even lower average cost. Symmetrically, for plant sizes above the MES, the tangency occurs on the upward-sloping portion of the SRATC. Only at the minimum efficient scale does the LRAC touch the SRATC at the latter's own minimum—the point where both short-run and long-run average costs are simultaneously minimized.

SECTION 4

Mathematical Framework

The mathematical derivation of long-run cost functions begins with the firm's cost-minimization problem. A firm with a production function Q = f(L, K) seeks the combination of labor (L) and capital (K) that minimizes total cost C = wL + rK, subject to producing a given output level Q₀. Here w is the wage rate and r is the rental rate of capital. The solution traces out the expansion path, and substituting the optimal input demands back into the cost equation yields the long-run total cost function LRTC(Q).

COST-MINIMIZATION CONDITION

MPₗ / w = MPₖ / r

At the cost-minimizing input mix, the marginal product per dollar must be equalized across all inputs. MPₗ = marginal product of labor; MPₖ = marginal product of capital; w = wage; r = rental rate of capital.

LONG-RUN AVERAGE COST

LRAC(Q) = LRTC(Q) / Q

Long-run average cost equals total cost divided by quantity. The LRAC curve is U-shaped when the production function exhibits first increasing returns to scale (LRAC falling), then constant returns (LRAC flat), then decreasing returns to scale (LRAC rising).

LONG-RUN MARGINAL COST

LRMC(Q) = dLRTC / dQ

LRMC is the derivative of long-run total cost with respect to output. When LRMC < LRAC, average cost is falling; when LRMC > LRAC, average cost is rising. The two curves intersect at the minimum of LRAC.

💡

Cobb-Douglas Example

For a Cobb-Douglas production function Q = ALᵅKᵝ, if α + β > 1 the firm has increasing returns to scale and LRAC declines; if α + β = 1 it has constant returns; if α + β < 1 it has decreasing returns. In the constant-returns case (α + β = 1), the LRTC function is linear in Q, meaning LRAC equals LRMC at every output level—a perfectly flat long-run average cost curve.

SECTION 5

Economies, Diseconomies & Returns to Scale

The shape of the LRAC curve is driven by returns to scale—the relationship between proportional increases in all inputs and the resulting change in output. When a firm doubles all inputs and output more than doubles, it experiences increasing returns to scale, which translates into economies of scale (falling LRAC). Conversely, when output less than doubles, the firm faces decreasing returns to scale and diseconomies of scale (rising LRAC). The sources of these phenomena differ markedly, and understanding them is critical for business planning.

The left column lists five common sources of economies of scale (falling LRAC), while the right column lists five typical sources of diseconomies of scale (rising LRAC). Real firms often experience both simultaneously, and the net effect determines the slope of the LRAC at a given output.

It is worth noting that some industries exhibit a long, flat bottom on the LRAC curve—a range of output over which average cost is essentially constant. This constant-returns-to-scale region is empirically common in manufacturing and services, and it implies that firms of varying sizes can coexist competitively. In contrast, industries like electric power generation have steeply declining LRAC curves over a wide range, producing natural monopoly conditions where a single firm can serve the entire market at lower cost than two or more firms.

LRAC Behavior Across Output Range

Low Q High Q

MES begins (35%)MES ends (65%)

SECTION 6

Worked Example — Deriving Long-Run Costs

Consider a firm with a Cobb-Douglas production function Q = L^0.5K^0.5. The wage rate is w = $20 per unit of labor and the rental rate of capital is r = $5 per unit of capital. We want to find the LRTC, LRAC, and LRMC functions, and determine whether this firm exhibits economies, diseconomies, or constant returns to scale.

Deriving Long-Run Cost Functions

1
Step 1 — Set Up the Cost-Minimization Condition

The cost-minimization condition requires MPₗ/w = MPₖ/r. For Q = L^0.5K^0.5, we compute MPₗ = 0.5L^−0.5K^0.5 and MPₖ = 0.5L^0.5K^−0.5. Substituting: (0.5L^−0.5K^0.5) / 20 = (0.5L^0.5K^−0.5) / 5.

Simplifying: K/L = 20/5 = 4, so K = 4L
2
Step 2 — Solve for Optimal Input Demands

Substitute K = 4L into the production function: Q = L^0.5(4L)^0.5 = L^0.5 × 2L^0.5 = 2L. Therefore L* = Q/2 and K* = 4(Q/2) = 2Q.

L* = Q/2, K* = 2Q
3
Step 3 — Derive the LRTC Function

Substitute into the cost equation C = wL + rK: LRTC = 20(Q/2) + 5(2Q) = 10Q + 10Q.

LRTC(Q) = 20Q
4
Step 4 — Compute LRAC and LRMC

LRAC = LRTC/Q = 20Q/Q = 20. LRMC = dLRTC/dQ = d(20Q)/dQ = 20. Both are constant and equal, confirming that this production function exhibits constant returns to scale (since α + β = 0.5 + 0.5 = 1).

LRAC = LRMC = $20 per unit (constant returns to scale)

ℹ

What if α + β ≠ 1?

If we changed the production function to Q = L^0.6K^0.6 (α + β = 1.2 > 1), the LRTC would be concave in Q, meaning LRAC declines as output rises—consistent with economies of scale. Conversely, α + β < 1 produces a convex LRTC and rising LRAC.

SECTION 7

Short-Run vs. Long-Run Costs — Comparisons & Limitations

One of the most common mistakes in applied cost analysis is conflating short-run and long-run perspectives. The table below highlights the critical differences and reminds us that the long-run framework, while powerful, rests on assumptions that may not always hold in practice.

Feature	Short Run	Long Run
Fixed Inputs	At least one input (e.g., capital) is fixed	All inputs are variable
Fixed Costs	Present (FC > 0)	Zero — all costs are variable
ATC Shape	U-shaped due to spreading FC then diminishing returns	U-shaped due to economies then diseconomies of scale
LRAC vs. SRATC	SRATC ≥ LRAC at every Q	LRAC is the lower envelope of all SRATCs
Decision Flexibility	Can adjust only variable inputs (labor, materials)	Can redesign plant, technology, and organization
Practical Limitation	Diminishing marginal returns dominate	Assumes perfect information about future demand and technology

Comparison of short-run and long-run cost frameworks

✦

KEY TAKEAWAY

The long-run cost framework is a planning tool, not a description of day-to-day operations. It answers "what should we build?" rather than "how do we operate what we've built?" In practice, firms are always operating in some short run, but they make long-run decisions whenever they consider expanding capacity, entering a new market, or adopting new technology—essentially any decision that changes their fixed inputs.

SECTION 8

Connection to Competitive Equilibrium & Industry Structure

Long-run production costs are not merely a firm-level concept; they determine the structure of entire industries and the nature of long-run competitive equilibrium. In a perfectly competitive market with free entry and exit, economic profits attract new entrants. As supply expands and price falls, firms continue entering until each surviving firm earns zero economic profit—a condition that requires price to equal the minimum of the LRAC curve. This is one of the most elegant results in microeconomics: competitive pressure drives each firm to produce at its most efficient scale.

Concept	Firm-Level Implication	Industry-Level Implication
P = min LRAC	Each firm produces at MES; zero economic profit	Maximum productive efficiency across the industry
High MES / Market Size	Few firms can survive; each must be large to compete	Oligopoly or natural monopoly tends to emerge
Low MES / Market Size	Firms can be small and still achieve low per-unit costs	Many firms coexist; competitive market structure
Flat LRAC Bottom	Firms of varying sizes are equally efficient	Wide range of firm sizes observed in the market

How LRAC shape determines market structure

Looking ahead, long-run cost analysis connects directly to advanced topics in industrial organization, including contestable markets theory (where potential entry constrains pricing even with few incumbents), multi-plant economies (where operating several small plants may be cheaper than one giant plant due to transportation costs), and economies of scope (where producing multiple products jointly is cheaper than producing each separately). These extensions enrich the basic LRAC framework and make it applicable to the complex, multi-product firms that populate modern economies.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

Explain why the long-run average cost curve is described as the "envelope" of the short-run average total cost curves. Why must LRAC ≤ SRATC at every output level?

Show answer

In the long run, the firm is free to choose whatever plant size minimizes cost for a given output level, whereas in the short run it is locked into a particular plant. Because the long run includes all short-run options (and more), the firm can always do at least as well as any particular short-run configuration. Therefore, LRAC cannot exceed any SRATC, and it equals the SRATC only at the output level for which that particular plant size happens to be optimal. Graphically, the LRAC curve traces the lowest points across all possible SRATC curves, just touching—or 'enveloping'—each one from below without crossing above it.

Question 2: BASIC CALCULATION

A firm has the long-run total cost function LRTC(Q) = 5Q² + 200Q. Compute LRAC(Q) and LRMC(Q). At what output does LRAC reach its minimum? What type of returns to scale does this firm exhibit?

Show answer

LRAC(Q) = LRTC/Q = 5Q + 200. LRMC(Q) = dLRTC/dQ = 10Q + 200. Setting LRMC = LRAC: 10Q + 200 = 5Q + 200, which gives 5Q = 0, so Q = 0. Since LRAC = 5Q + 200 is a linearly increasing function of Q, the minimum occurs at Q → 0 (i.e., at the intercept of $200). This firm exhibits decreasing returns to scale throughout, because LRAC rises continuously as output increases. In practical terms, the firm's cost per unit grows with every additional unit produced.

Question 3: INTERMEDIATE

A firm has a production function Q = L^0.4K^0.8 with w = $10 and r = $40. (a) Does this firm exhibit increasing, constant, or decreasing returns to scale? (b) Derive the optimal capital-to-labor ratio. (c) What is the qualitative shape of the LRAC curve?

Show answer

The sum of exponents is α + β = 0.4 + 0.8 = 1.2 > 1, so the firm exhibits increasing returns to scale. To find the cost-minimizing ratio, set MPₗ/w = MPₖ/r. MPₗ = 0.4(K/L)^0.8 and MPₖ = 0.8(L/K)^0.4. Equating MPₗ/10 = MPₖ/40 yields K/L = (0.4 × 40)/(0.8 × 10) = 2, so K = 2L. Because returns to scale are increasing, doubling inputs more than doubles output, meaning LRTC increases less than proportionally with Q. Consequently, LRAC is a downward-sloping function—the firm continually benefits from scale, and there is no interior minimum efficient scale.

Question 4: APPLIED

A startup brewery estimates three possible plant sizes with the following per-barrel costs: Small plant — SRATC minimized at $12/barrel (at 5,000 barrels); Medium plant — SRATC minimized at $8/barrel (at 20,000 barrels); Large plant — SRATC minimized at $10/barrel (at 50,000 barrels). The expected market demand for the brewery is 18,000 barrels per year. Which plant should the brewery choose, and why? What does the rising minimum SRATC for the large plant suggest about returns to scale?

Show answer

At 18,000 barrels, the brewery should choose the medium plant. Although the medium plant's SRATC is minimized at 20,000 barrels, it will still achieve per-barrel costs close to its minimum of $8 at 18,000 barrels. The small plant would be operating far beyond its efficient range, driving costs well above $12, and the large plant would be severely underutilized. The fact that the large plant's minimum SRATC ($10) is higher than the medium plant's ($8) indicates that the firm enters diseconomies of scale between 20,000 and 50,000 barrels. This means the LRAC curve reaches its minimum near 20,000 barrels and begins to rise, making the medium plant the long-run cost-minimizing choice for demand levels in the 15,000–25,000 range.

Question 5: CRITICAL THINKING

In long-run competitive equilibrium, price equals the minimum of LRAC and each firm earns zero economic profit. If a technological innovation shifts the LRAC curve downward for all firms in an industry, trace through the adjustment process: What happens to (a) individual firm profits in the short run, (b) market entry and supply, (c) market price, and (d) the new long-run equilibrium? Does the number of firms increase, decrease, or remain ambiguous?

Show answer

When a technology shift lowers every firm's LRAC, firms that adopt the technology immediately see their costs fall below the current market price, earning positive economic profit in the short run. These profits attract new entrants (and incentivize existing firms to expand), increasing market supply and pushing the market price downward. Entry continues until price once again equals the new, lower minimum of LRAC, restoring zero economic profit. The new long-run equilibrium features a lower price and greater total industry output. However, the effect on the number of firms is ambiguous: if the new MES is proportionally larger (the technology favors bigger plants), fewer but larger firms may serve the market. If MES is unchanged or smaller, more firms may coexist. The answer depends on whether the technology shifts the LRAC curve downward uniformly or also changes its shape—particularly the location of the minimum efficient scale relative to market demand.

SUMMARY

Summary — Long-Run Production Costs

Summary

In the long run, all inputs are variable, allowing firms to choose the cost-minimizing combination of labor, capital, and technology for any output level. The long-run average cost (LRAC) curve is the lower envelope of all possible short-run average total cost curves. Its typically U-shaped profile reflects economies of scale (falling LRAC from specialization, bulk purchasing, and technology) at low output, a flat region of constant returns to scale around the minimum efficient scale (MES), and diseconomies of scale (rising LRAC from coordination complexity and principal-agent problems) at high output.

Mathematically, long-run costs are derived by solving the firm's cost-minimization problem, equating the marginal product per dollar across inputs (MPₗ/w = MPₖ/r). The LRMC curve intersects LRAC at its minimum. In long-run competitive equilibrium, free entry and exit drive price to the minimum of LRAC, ensuring zero economic profit and maximum productive efficiency. The ratio of MES to market demand shapes industry structure—determining whether an industry gravitates toward perfect competition, oligopoly, or natural monopoly.