Source: Intermediate Microeconomics, Florida International University
Difficulty: Intermediate | Prerequisites: Consumer theory notes, basic calculus (derivatives).
Production theory is the firm's counterpart to consumer theory. Instead of maximising utility subject to a budget constraint, firms maximise output subject to costs, or minimise costs subject to a target output. This material introduces how inputs (labour, capital) combine to produce output, how productivity changes as you vary inputs, and what happens when you scale everything up. It is the bridge between consumer demand and the cost curves that drive firm behaviour in every market structure.
The production function maps inputs to output. In the short run at least one input is fixed, and the law of diminishing returns sets in. In the long run all inputs are adjustable, and returns to scale tell you whether doubling inputs more than doubles, exactly doubles, or less than doubles output. You test returns to scale by multiplying all inputs by a constant n and comparing the result to n times the original output.
Production function
Q = F(K, L, M; A). The maximum output achievable from a given set of inputs. A is a technology parameter that shifts the output level for the same inputs. In simpler form: Q = F(K, L).
Labour (L)
Human effort. Includes both skilled and unskilled work.
Capital (K)
Goods used to produce other goods: machines, buildings, technology, equipment.
Materials (M)
Raw goods such as oil, lumber, minerals.
Total product (TP)
The total quantity of output produced for a given set of inputs. Simply Q = F(K, L) for specific values of K and L.
Marginal product (MP)
The change in output from a one-unit increase in a particular input, holding other inputs fixed. MP = ΔQ / ΔInput. Think of it as: how many more units can you produce by adding one more worker?
Average product (AP)
Total output divided by total input. AP = Q / Input.
Short run
The time horizon over which at least one input cannot be adjusted. Capital is the classic fixed input (you cannot resize the factory by the weekend).
Long run
The time horizon over which all inputs can be adjusted. You can build a larger factory, relocate, or change your entire production process.
Law of diminishing (marginal) returns
As you add equal increments of one input while holding others fixed, the marginal product of that input eventually falls. More workers in the same kitchen produce less additional output per worker.
Increasing returns to scale (IRTS)
Doubling all inputs more than doubles output. Additional units cost less to produce than previous ones.
Constant returns to scale (CRTS)
Doubling all inputs exactly doubles output.
Decreasing returns to scale (DRTS)
Doubling all inputs less than doubles output. Additional units cost more to produce than previous ones.
Isoquant
A curve showing all input combinations that produce the same output level. The producer's equivalent of an indifference curve.
Isocost line
A line showing all input combinations that cost the same total amount. The producer's equivalent of a budget line.
Marginal rate of technical substitution (MRTS)
The rate at which a firm can substitute one input for another while keeping output constant. MRTS = MP(L) / MP(K).
The production function Q = F(K, L, M; A) gives the highest output Q that can be generated from given inputs. Producing less than this maximum with the same inputs would be inefficient.
The technology parameter A shifts the entire production function. Technology increases production for any given level of capital and labour. A level shift up can be beneficial for society (feeding more people with the same farmland), while a shape change that substitutes capital for labour can displace workers (automating jobs).
Capital investment also affects wages. When K increases (say from 10 to 20), each worker becomes more productive, potentially earning higher wages. Consider a small business with one worker who keeps profits: if a new computer helps her increase production, her wages effectively increase.
Total product (TP): Q = F(K, L) for specific K and L. How many frozen bananas can you make if you work L hours?
Marginal product (MP): the change in output from a one-unit increase in one input, holding others fixed.
MP = ΔQ / ΔInput
Average product (AP): total output divided by total input.
AP = Q / Input
Key relationship: MP intersects AP at the maximum of the AP curve. When MP > AP, the average is rising. When MP < AP, the average is falling. This is a mathematical fact, not an economic assumption.
Short run: at least one input is fixed. Capital is the classic example (you cannot resize the factory overnight), though labour contracts can also make labour sticky.
The law of diminishing (marginal) returns applies in the short run. As you add equal amounts of one input while holding others fixed, output increments eventually get smaller. The fifth worker in a small kitchen adds less output than the second.
In the short run, as MP rises, MC falls (and vice versa). This inverse relationship between marginal product and marginal cost is central to cost analysis.
Long run: all inputs can be adjusted. The firm can build a larger factory, relocate to another country, or restructure entirely.
In the long run, the question becomes: what combinations of inputs generate the same level of output? These combinations trace out isoquants.
Example: Q = 4KL. To produce Q = 4, solve for K in terms of L: K = 1/L. There are infinitely many combinations.
Returns to scale ask what happens when you scale all inputs by the same factor.
How to test mathematically:
Start with Q = F(K, L).
Multiply all inputs by a constant n: compute F(nK, nL).
Compare F(nK, nL) to nQ.
If F(nK, nL) > nQ: increasing returns to scale (IRTS). Doubling inputs more than doubles output.
If F(nK, nL) = nQ: constant returns to scale (CRTS). Doubling inputs exactly doubles output.
If F(nK, nL) < nQ: decreasing returns to scale (DRTS). Doubling inputs less than doubles output.
Why returns to scale matter for market structure:
DRTS: firms that grow too large become inefficient. This favours many small firms and competitive markets.
IRTS: bigger is more efficient. This favours fewer, larger firms and can lead to natural monopolies.
CRTS: firm size does not affect efficiency. The market can support firms of any size.
Returns to scale vs diminishing returns: these are different concepts. Diminishing returns apply to one input varying while others stay fixed (short run). Returns to scale apply when all inputs change proportionally (long run).
Production function: Q = F(K, L) or Q = F(K, L, M; A)
Marginal product: MP = ΔQ / ΔInput
Average product: AP = Q / Input
Returns to scale test: multiply inputs by n, compare F(nK, nL) to nQ
Cost-minimisation condition (long run): MRTS = w / r, equivalently MP(L) / w = MP(K) / r
Relationship between MP and MC: MC = w / MP(L)
Relationship between AP and AVC: AVC = w / AP(L)
Students often confuse diminishing returns with decreasing returns to scale. Diminishing returns is a short-run concept (one input varies, others fixed). Decreasing returns to scale is a long-run concept (all inputs scale proportionally).
Students sometimes think the production function describes what the firm will produce. It describes the maximum the firm can produce. Producing less is feasible but inefficient.
A common error is assuming that more capital always raises wages. Whether capital investment raises wages depends on the specific production function and how capital and labour interact.
⚠️ The returns-to-scale test (multiply inputs by n, compare to nQ) is a standard exam question. Be able to carry it out for any production function.
⚠️ Know that MP intersects AP at AP's maximum. This shows up in graphical questions.
⚠️ Understand the inverse relationship: when MP rises, MC falls. When MP falls, MC rises.
⚠️ Be able to distinguish short-run diminishing returns from long-run returns to scale in a written answer.
⚠️ The cost-minimisation condition MRTS = w/r (or equivalently MP(L)/w = MP(K)/r) is frequently tested.
True or False: The law of diminishing returns applies when all inputs are scaled up together. (False, it applies when one input varies and others are fixed)
Fill in the blank: If F(nK, nL) > nQ, the production function exhibits ______ . (increasing returns to scale)
True or False: MP intersects AP at AP's minimum. (False, at AP's maximum)
Fill in the blank: In the short run, at least one input is ______ . (fixed)
True or False: MC = w / MP(L). (True)
Q: Given Q = 4KL, what are the returns to scale?
A: Replace K with nK and L with nL: F(nK, nL) = 4(nK)(nL) = 4n²KL = n²Q. Since n² > n for n > 1, we have F(nK, nL) > nQ. This is increasing returns to scale.
Q: In the short run, a food truck owner expects a busy weekend. What can she adjust and what is fixed?
A: She can adjust variable inputs such as ingredients and workers. The size of the kitchen is fixed.
Q: What is the cost-minimisation condition for a firm choosing between capital and labour?
A: MRTS = w/r, which means MP(L)/MP(K) = w/r. Equivalently, MP(L)/w = MP(K)/r. The firm equalises the marginal product per pound spent on each input.
Q: Explain in one sentence why diminishing returns and decreasing returns to scale are different concepts.
A: Diminishing returns describe falling marginal product when one input increases while others are fixed; decreasing returns to scale describe less-than-proportional output growth when all inputs increase together.
Production theory feeds directly into cost theory: the shapes of the MC and AVC curves are determined by the behaviour of MP and AP. Returns to scale influence long-run average cost curves and, in turn, market structure. The firm's cost-minimisation problem (isoquant tangent to isocost) mirrors the consumer's utility-maximisation problem (indifference curve tangent to budget line).
Tags: production function, inputs, labour, capital, materials, total product, marginal product, average product, short run, long run, law of diminishing returns, diminishing marginal returns, returns to scale, increasing returns to scale, IRTS, constant returns to scale, CRTS, decreasing returns to scale, DRTS, isoquant, isocost, MRTS, marginal rate of technical substitution, cost minimisation, technology parameter, intermediate microeconomics