Read 5 of 5

The competitive-market model

When is it reasonable to treat buyers and sellers as price-takers?

Demand tells us how buyers respond to a price. Supply tells us how firms respond to a price. To put the two together, we need a model of the market where they meet.

The model we use is perfect competition. In it, every buyer and seller is a price-taker: they treat the market price as given and decide only how much to buy or sell. No one gets to set the price.

Perfect competition is a benchmark, not a claim that real markets look like this. It leaves out the features that let a buyer or seller influence the price: products that differ from seller to seller, the time and effort it takes to compare prices, and obstacles that stop new firms from joining a market. With those removed, we can see what happens when no one controls the price.

5.1

Why use a model that isn't literally true?

No real market fits perfect competition exactly. Think of tomatoes at a farmers’ market. Tomatoes differ in quality from stall to stall, so the products aren’t identical. Buyers spend time walking between stalls to compare prices, so comparing isn’t free. Starting a farm takes land and equipment, so joining the market isn’t free either. And some growers are much bigger than others, so not every seller is small.

So why use the model? Because it lets us look at one thing at a time. The competitive model asks a narrow question: how do buyers and sellers behave when each one is too small to affect the price? The answer gives us a baseline. To study a real market, we compare it with that baseline. We ask which assumption fails and what that failure lets a buyer or seller do, such as charge more than its rivals.

Economists use models this way all the time. Think of a map. Nobody complains that a map leaves out every tree. What matters is whether it gets you where you’re going. In the same way, the competitive model is useful whenever treating the price as given is a good enough approximation for the decision you’re studying.

5.2

Define the market before judging competition

Whether a market looks competitive depends on how you define it. A café may be one of hundreds in Melbourne, but the only one inside an airport terminal. A hospital may compete with many others for routine care across a city, but have no nearby rival for a specialized emergency service. The same business can face lots of competition in one market and almost none in another.

A market definition has two parts: a product and a place. Products belong in the same market when buyers see them as close enough substitutes. Places belong in the same market when buyers or sellers can switch between them without it costing too much.

Getting the boundary right matters. Define the market too broadly, say as all food sold in Australia, and every firm looks tiny, so competition looks stronger than it really is. Define it too narrowly, say as coffee sold at a single airport gate, and almost every seller looks like a monopolist, the only seller in its market. To find a sensible boundary, ask a simple question: if this seller raised its price a little, would enough buyers switch to another product or another place? For the airport café, probably not. Travelers who are past security can’t walk to a café in town. That’s why the café’s market is the terminal, not the city.

The competitive model can’t draw this boundary for you. It only becomes useful once you know what buyers’ real alternatives are. A wheat farmer selling into a global market may well be a price-taker. The only mechanic in a remote town isn’t, even though both are small businesses. What matters isn’t how big a business is, but how big it is relative to its market.

5.3

Assumption 1: homogeneous goods

Goods are when buyers see the units sold by different sellers as identical. A kilogram of standard-grade wheat from one farm does exactly the same job as a kilogram from another, so the buyer doesn’t care who grew it.

This matters because of what it does to prices. Suppose two sellers offer the same good and buyers can see both prices. If one seller charges more, buyers simply go to the cheaper one. Nothing about the product can justify the higher price, because buyers see no difference worth paying for.

Notice that homogeneity is about how buyers compare products, not only about the products themselves. Electricity from two power stations is literally the same once it’s in the grid, so buyers have no reason to prefer one over the other. Wheat of a given grade is close to homogeneous. Restaurant meals aren’t, even when two restaurants use the same ingredients, because location, service and reputation all affect how much a diner values the meal. Products that buyers see as different are called differentiated.

When products are differentiated, a firm can raise its price and still keep some customers: the ones who prefer its version. That firm faces its own downward-sloping demand curve, and it has some market power, meaning some ability to set its price rather than just take it. We’ll study firms like this in later weeks.

5.4

Assumption 2: price-taking

is the heart of the model. Each buyer and seller treats the market price as given and decides only how much to buy or sell.

This is believable when each participant is small relative to the market. If one tomato grower keeps a crate off the market, total supply barely changes. If one shopper buys an extra kilogram, total quantity demanded barely changes. Neither action moves the market price by enough to matter. You saw the same idea with Auto Eco: with a thousand similar firms, one firm’s output is a tenth of one percent of the market, so changing it leaves the price essentially where it was.

Price-taking doesn’t mean participants are passive. Buyers decide how much to buy by comparing the price with how much they value one more unit. Firms decide how much to produce by comparing the price with their marginal cost. Both can respond strongly when the price changes. What they can’t do is choose the price itself.

In the model, price-taking isn’t assumed out of thin air. The other assumptions are what make it hold. Because there are many small participants, no single one matters to the price. Because goods are homogeneous and prices are easy to compare, a seller who charges more than the market price loses its customers. And over a longer period, free entry keeps profits in check: if sellers earn high profits, new firms join, add to supply and push the price down. We look at entry more closely in 5.6.

5.5

Assumption 3: zero search cost

Search is costless when buyers and sellers can find each other and compare prices without spending any real time or money. At an open farmers’ market, a buyer can walk past every stall and see every price. On an online exchange for a commodity, prices from all sellers may appear on the screen instantly.

This matters because it keeps prices in line. If goods are identical and buyers can see every price for free, a seller who charges more than the others gets no customers, so every seller ends up charging the same price. When search is costly, that pressure weakens. Two sellers can charge different prices for similar goods, because some buyers won’t find the cheaper option or won’t bother to travel to it. A seller in a convenient location can charge more, because some buyers would rather pay a little extra than go further. An online platform that decides which offers buyers see first can steer them toward some sellers and away from others. In each case, control over information gives someone market power.

Zero search cost is an extreme assumption, since real search always takes some time. But search can be cheap enough that treating the market as having a single price is still a good approximation. The question to ask is whether the cost of searching meaningfully changes the decision you’re studying.

5.6

Assumption 4: free entry and exit

Free entry means a new firm can join the market on the same terms as the firms already in it. Free exit means a firm can leave the market without being stuck with costs that keep it there, such as a long lease it can’t break.

Starting a business always takes resources, but ordinary costs aren’t barriers. If every newcomer can rent the same equipment, hire workers at the same wages and use the same technology as existing firms, those costs are just part of producing. A barrier is different. It gives existing firms an advantage newcomers can’t match, or it keeps newcomers out even when joining would be profitable.

Many things can act as barriers. A market may require a license that only a few firms hold. A patent gives one firm the legal right to be the only producer of an invention for a period. An existing firm may sign exclusive contracts that tie up suppliers or customers. Good locations may be scarce. Some products are worth more the more people use them, as with a social network, so a newcomer with few users struggles to attract more. And some industries require large investments that can’t be recovered if the firm later leaves, which makes entering risky. How much these matter depends on the market. Starting a food truck takes a modest commitment. Building an electricity network takes an enormous one.

Entry and exit matter most over time. If firms in a market earn more than their opportunity cost, that is, more than they could earn elsewhere, and newcomers can copy what they do, those profits attract new firms. If firms keep losing money, some leave, and the market loses their capacity. We’ll see how this changes market outcomes when we study the long run.

5.7

How the assumptions fit together

The four assumptions aren’t a checklist of unrelated boxes to tick. As we saw in 5.4, the other three are what make price-taking hold. Identical goods and free price comparison stop any seller from charging more. Many small sellers mean no single seller moves total supply. Free entry means high profits attract competitors instead of staying protected.

That also means price-taking can be a good approximation even when an assumption doesn’t hold perfectly. A wheat farmer faces transport costs, and wheat varies in quality, so neither homogeneity nor zero search cost holds exactly. But the farmer is still far too small to affect the price grain buyers offer, so treating that price as given works well. The reverse can happen too. A market can have many sellers and still not be competitive if a single platform controls how they reach customers.

So the useful question about a real market isn’t “Is this market perfectly competitive?” because almost none is. It’s “Who can affect the price, what gives them that ability, and for how long?” The competitive model tells you where to look for the answer.

5.8

Short run and long run

The is a period in which at least one commitment can’t be changed and the number of firms is fixed. A commitment is something a firm has already locked in, such as a factory, a machine or a lease. In the short run, firms can change how much they produce by changing things like labor hours or materials. But they can’t build a new plant overnight, and new firms can’t set up in time to join.

The is a period long enough for everything to change: every input and the number of firms. Firms can expand or shrink their capacity, leases can run out, new firms can enter and existing ones can leave.

These aren’t calendar periods. The long run might be a few weeks for a market stall, years for a mine and decades for an electricity network. What matters is which decisions can change within the time frame you’re studying.

Think of hotel rooms during a major event. This week, the number of rooms in the city is essentially fixed. A higher price might persuade a hotel to open a few rooms it was holding back for maintenance, but supply can’t grow much. If high room prices last for several years, hotels renovate and add rooms, and new hotels get built. The same city then responds to price very differently. Neither picture is the wrong one: each answers a different question. A manager planning next weekend needs the short-run view. An investor deciding whether to build a hotel needs the long-run view.

The time frame also changes which costs count. Recall from Week 1 that a firm’s decisions depend only on the costs it can avoid. This month, a firm’s lease is already signed, so it pays the rent whether it produces or not, and the rent doesn’t affect this month’s decision. Next year, the firm can choose not to renew, so the rent becomes a cost it can avoid. That’s why a firm’s shutdown threshold can differ between the short run and the long run. Short-run supply reflects the capacity and commitments firms already have. Long-run supply also includes building new capacity, new firms entering and existing firms leaving.

5.9

What the model does not yet determine

We now have a demand curve, which tells us how much buyers want at each price, and a supply curve, which tells us how much firms want to sell at each price. What we don’t have yet is the price itself.

At some prices, buyers want more than firms are willing to supply. At others, firms want to supply more than buyers want to buy. To finish the model, we need to explain how these mismatches push the price up or down, and where the market settles: at the price where the quantity buyers want matches the quantity firms want to sell.

That’s the subject of the next reading. These readings have built its ingredients. How much buyers value a good gives us individual demand. What it costs firms to produce gives us individual supply. Adding up horizontally takes us from individual curves to market curves. And the competitive assumptions let each buyer and seller treat the price as given, even though the market as a whole determines it.

Summary

Perfect competition is a benchmark built on homogeneous goods, price-taking, costless search and free entry and exit. The assumptions support the idea that each buyer and firm treats price as given. The short run fixes some commitments and the number of firms; the long run allows capacity, entry and exit to change.

Concept atlas