Production, Costs, and Industry Structure

This is Unit Six of Principles of Microeconomics, part two of two: the long run, where every input can change, and the cost curve that decides how many firms an industry holds. Carry the same idea as part one: cost curves are the production function in dollars. In the long run the firm chooses its scale and its input mix, and the shape of its long-run cost curve, read against the market's size, sets up the four market structures. By the end you will be able to tell a fixed cost from a sunk one and read the profit margin. To say what the long run changes and choose a least-cost technology. To state and apply the cost-minimizing input rule. To read a long-run average cost curve for economies of scale and minimum efficient scale. And to connect scale, technology and market size to the number of firms.

Three entries on costs that are gone, cost patterns that differ, and the margin that shows profit per unit.

Fixed costs that are already gone.

MIC 0525 Sunk Fixed Costs

Many fixed costs are sunk: already paid and unrecoverable, such as last year's brand campaign, and a sunk cost should not affect any decision still open. Fixed is about output and sunk is about recoverability; a fixed cost the firm could still cancel is not sunk.

How cost patterns differ.

MIC 0526 Cost Patterns Differ Across Industries

Some industries have high fixed and low marginal costs, others the reverse, and some hit diminishing returns sharply; the curves differ, and so do the decisions they drive. The same tools read every pattern, and where the curves sit decides the firm count later in this part.

Profit per unit.

MIC 0527 Profit Margin: Price Against Average Cost

Profit per unit is price minus average total cost, the profit margin. Price above average total cost means profit; price below means a loss on every unit. The margin says whether there is profit, not how much to produce; that needs marginal cost, next unit.

Four entries on the long run: every input moves, more capital delays diminishing returns, and choosing among ways to produce.

Where every input moves.

MIC 0528 Production in the Long Run

In the long run every input is variable, including the short run's fixed ones. The firm chooses its scale and its input mix, and there are no fixed costs. The long run is a planning horizon, not a date, and a firm lives in a short run while planning the long.

What more capital does to diminishing returns.

MIC 0529 More Capital Delays Diminishing Returns

Diminishing returns set in because workers share a fixed input. Give them more capital and the fall in marginal product starts later, at a larger workforce: with two ovens the fourth baker still adds plenty, with one, the third already crowds. More capital shifts the whole marginal product curve out; it does not repeal the law, it moves where it bites.

Picking among ways to produce.

MIC 0530 Choosing a Technology

The same output can be made with more labor and less capital or the reverse, and the firm picks the cheapest mix at current prices, using less of whatever gets dearer. This is why labor demand slopes down: a higher wage makes every labor-heavy method less attractive.

The comparison with numbers.

MIC 0531 Least-Cost Technology from Three Bundles

Given several input bundles that each produce the same output, cost each one, workers times the wage plus machines times the rental, and pick the cheapest; then change one input price and cost them again, and the choice moves away from the input that got dearer. A park can be cleaned by twelve workers with two machines, seven with four, or three with seven. At fifty dollars a worker and one hundred a machine the bundles cost eight hundred, seven hundred fifty and eight hundred fifty, so the middle method wins. Raise the wage to eighty: one thousand one hundred sixty, nine hundred sixty and nine hundred forty, so the machine-heavy method wins. Cut it to thirty and the labor-heavy method wins. Cost every bundle at the same prices, and remember the cheapest bundle is cheapest for this output.

Two entries on the rule behind the choice: the cost-minimizing input rule, and applying it.

The rule behind the choice.

MIC 0532 The Cost-Minimizing Input Rule

Cost is minimized when marginal product per dollar is equal across inputs: labor's marginal product over the wage equals capital's over its price, and if one ratio is higher, shift a dollar there. It is the utility-per-dollar rule with output for satisfaction. The anchor works only fixed bundles; the rule comes from the witness, and the exam asks for it.

Applying it.

MIC 0533 Applying the Input Rule

Divide each input's marginal product by its price. If the ratios are equal, the firm is minimizing cost for that output and should change nothing; if they differ, use more of the input with the higher marginal product per dollar and less of the other, and as that happens the higher ratio falls and the lower rises until they meet. A workshop's last worker adds sixty units at twelve dollars, five per dollar, and its last machine forty units at four dollars, ten per dollar: capital gives more per dollar, so use more capital and less labor. Compare per dollar, never raw marginal products, which would have favored labor's sixty; and the answer is a direction, never use only the cheaper input.

Five entries on scale: the long-run curve, economies and diseconomies, the confusion that recurs, and reading the curve.

The long-run cost curve.

MIC 0534 Long-Run Average Cost

The long-run average cost curve gives the lowest average cost of each output when every input, plant size included, is free: the lower envelope of the short-run curves. A firm planning an output picks the plant whose curve is lowest there. Short-run minima lie off the envelope except at its own minimum, so a plant rarely runs at its cheapest point.

The falling part.

MIC 0535 Economies of Scale

Economies of scale exist when a larger scale lowers average cost: doubling every input more than doubles output, or spreads a big fixed investment over more units. Economies of scale are a long-run idea, all inputs rising, not the spreading of a short-run fixed cost.

The flat and rising parts.

MIC 0536 Constant Returns and Diseconomies of Scale

Constant returns to scale hold where expanding every input leaves average cost unchanged, the flat stretch. Diseconomies of scale set in where a bigger firm costs more per unit, usually from coordination failures. The flat stretch is often wide, which is why one industry holds firms of many sizes.

The confusion that recurs.

MIC 0537 Diminishing Returns Are Not Diseconomies

Diminishing returns happen in the short run as one input rises against a fixed one; diseconomies of scale happen in the long run as every input rises together. A firm can have diseconomies without diminishing returns, and the reverse; never explain one by the other.

Reading the long-run curve.

MIC 0538 Reading a Long-Run Average Cost Curve

Given a long-run average cost curve and an output, classify the region. Falling: economies of scale, and the firm would lower its unit cost by growing. Flat: constant returns, where size does not matter. Rising: diseconomies, where a smaller firm would beat it. Then find the smallest output at which average cost reaches its minimum, the minimum efficient scale, and compare it with the market's quantity. A bottling plant's average cost falls to five dollars a case at ten thousand cases a week, stays there to thirty thousand, and rises after. At six thousand cases the firm has economies of scale; at twenty thousand, constant returns; at forty thousand, diseconomies. Minimum efficient scale is ten thousand, and a market buying three hundred thousand cases at five dollars holds between ten and thirty efficient firms, a range because of the flat stretch.

Two entries on firms and industries: what the count implies, and technology moving the scale.

What the count implies.

MIC 0539 Minimum Efficient Scale and the Number of Firms

Minimum efficient scale is the smallest output at which average cost is lowest, and market demand at that cost divided by it gives the number of efficient firms: a million units and a scale of ten thousand hold about a hundred firms. A scale as large as the market leaves room for one firm, the natural monopoly of units eight and ten.

Technology moving the scale.

MIC 0540 Technology Shifts the Scale

A new technology can enlarge minimum efficient scale, favoring big firms, or shrink it, letting small firms compete; the shape of the industry follows the shape of its cost curve. Which way information technology pushes is contested, and the course reports it as an open question.

Two capstones: the whole cost family built from one table, and why industries differ in firm size.

The whole cost family from one table.

MIC 0541 From a Production Function to Every Cost Curve

Every short-run cost curve descends from one production table and two prices. Take marginal product between the rows; it rises while workers specialize and falls once they crowd the fixed input. Attach the wage: variable cost is workers times the wage, and total cost adds the fixed cost. Marginal cost between the rows is the change in variable cost over the change in output, which is the wage over marginal product, so its low point is marginal product's high point. Average variable cost bottoms where marginal cost crosses it; average total cost bottoms further right, where marginal cost crosses it too. The bakery in full: loaves zero, twenty, fifty, seventy, eighty, eighty-five; marginal product twenty, thirty, twenty, ten, five; wage sixty and fixed cost one hundred; marginal cost three, two, three, six, twelve, each sixty over the marginal product beside it; average variable cost lowest at fifty loaves, where marginal cost passes from two to three; average total cost lowest at seventy, where it passes from three to six; and at every row the two smaller averages add to the larger. The first loaves cost three dollars each and not eight, the averages bottom at different outputs, and the whole family is a short-run family.

Why industries look the way they do.

MIC 0542 Why Industries Differ in Firm Size

Why do some industries hold thousands of small firms and others three giants, and why do some hold both? The long-run average cost curve answers all three. If it reaches its minimum at a small output and turns up soon after, minimum efficient scale is small and the market fills with many small firms: restaurants and barbers. If it keeps falling out to a large share of the whole market, a single big firm undercuts every smaller rival and the market holds a few firms or one: refrigerator makers, or the water utility. If it is flat over a wide range, the industry mixes firms of every size in it. Market demand sets the other half: the count is the market's quantity at the minimum cost divided by the minimum efficient scale, and technology moves the curve over time. A city's restaurant trade of ten million dollars a year, with costs lowest at three hundred thousand of sales, holds twenty to thirty restaurants. A regional cement market of a million tons, with plant costs still falling to four hundred thousand, holds two or three plants. A town needing two million gallons a day, with plant costs still falling at five million, is served cheapest by one plant. Read the curve against market size, and expect the structure to change when a technology halves the scale, which unit ten returns to.

Look at what you can now do. You can tell a fixed cost, the same at any output, from a sunk one, unrecoverable, and read the profit margin as price minus average total cost. You can say that the long run frees every input and that the cheapest technology depends on prices as much as on engineering. You can state the input rule, marginal product per dollar equal across inputs, and say which input to use more of. You can read a long-run average cost curve for economies of scale, constant returns and diseconomies, and find minimum efficient scale. And you can divide market quantity by that scale to count the efficient firms, and see the four market structures as four answers to how big a firm must be to be cheap. Unit seven takes the first of them, the firm that takes its price as given.