Unit 4 · Competition, Firms and Market Structure

A business takes in money when it sells things and pays out money to make them. The money coming in is revenue. Total revenue is simply price times quantity sold. Marisol sold 6,000 tacos at $3 each, so her total revenue was 6,000 × $3 = $18,000. If a lemonade stand sells 250 cups at $2, its revenue is $500. Revenue is the first number an owner sees, and it is the one that fools people.

The money going out is total cost: everything the business spends to produce and sell its product. Marisol’s total cost in April was $13,000. Profit is what is left when you subtract total cost from total revenue. Her profit was $18,000 − $13,000 = $5,000. When total cost is larger than total revenue, the difference is a loss. A truck that spent $13,000 but took in only $11,000 would have a loss of $2,000.

The most common mistake is to treat revenue as if it were profit. A store can double its revenue and still end the year worse off if its costs rise faster. Suppose a shop’s revenue climbs from $20,000 to $22,000 in a month, but its costs climb from $15,000 to $19,000. Profit did not rise. It fell from $5,000 to $3,000. Owners who watch only the register miss this.

A higher price does not always mean higher revenue, either. If Marisol raised her price to $4 and customers responded by buying only 4,000 tacos, her revenue would be 4,000 × $4 = $16,000, which is $2,000 less than before. Price and quantity work together, and the demand curve you studied earlier decides how they trade off. Every decision in this chapter starts by asking what happens to revenue and what happens to cost.

Look again at Marisol’s napkin and sort the costs into two piles. The truck loan ($900), insurance ($300) and the commissary kitchen ($1,200) add up to $2,400. She owes that $2,400 in a month when she sells 6,000 tacos, and she owes it in a month when a blizzard keeps her home and she sells none. A cost that does not change with the amount produced is a fixed cost.

Tortillas, meat, propane and paper napkins are different. Every extra taco uses about $1.10 more of ingredients. If she sells nothing, she buys nothing. A cost that rises and falls with output is a variable cost. Her helper’s wages count as variable too, because she schedules more hours on busy days and fewer on slow ones. Some costs can be either, depending on how the business is set up. The test is always the same: does this bill change when output changes?

Total cost is fixed cost plus variable cost. A small table shows how it grows. At 0 tacos, total cost is $2,400 + $0 = $2,400. At 1,000 tacos, variable cost is 1,000 × $1.10 = $1,100, so total cost is $3,500. At 3,000 tacos it is $2,400 + $3,300 = $5,700. Notice that each extra thousand tacos adds the same $1,100. The fixed part never moves; the variable part climbs in a straight line.

Two mistakes are common. The first is thinking fixed means small or unimportant. Fixed costs are often the biggest bills a young business faces, and they come due even in a bad month. The second is thinking fixed means forever. Economists call the period when at least one cost is fixed the short run. In the long run, everything can change: Marisol can sell the truck, drop the kitchen lease or buy a second truck. In the long run, every cost is variable.

Marisol’s napkin listed every bill she paid: the loan, the kitchen, the meat, the wages. Costs that are paid out in money are explicit costs. But her sister spotted a cost with no receipt. To run the truck, Marisol gave up a job that paid $4,200 a month. That paycheck is the opportunity cost of her own time. A cost that is the value of something the owner already has and uses, rather than a bill, is an implicit cost.

This gives two ways to measure profit. Accounting profit is revenue minus explicit costs: $18,000 − $13,000 = $5,000. Economic profit also subtracts implicit costs: $5,000 − $4,200 = $800. Suppose Marisol had also pulled $10,000 out of a savings account that paid $30 a month in interest. That lost interest is another implicit cost, and economic profit falls to $770. Same truck, same month, three different numbers, each answering a different question.

Economic profit answers the question owners most need: am I doing better than my next-best option? A positive economic profit means yes. A negative one means the owner would be better off, in money terms, doing something else. Zero economic profit is not failure. It means the business pays the owner exactly what she could earn elsewhere, which economists call a normal profit. Many healthy small businesses live near zero economic profit for years.

Remember the rule from the opportunity cost chapter: the implicit cost is the single next-best alternative, not a list. Marisol gave up the restaurant job. She did not also give up a job at a hotel kitchen and a job at a bakery, because she could only have worked one of them. Counting three lost salaries would make the truck look far worse than it is. Count one, the best one.

A print shop in Lakeview makes custom T-shirts in batches of ten. Its fixed costs come to $60 a day. Here is its total cost at each level of output: 0 batches, $60; 1 batch, $80; 2 batches, $95; 3 batches, $105; 4 batches, $120; 5 batches, $140. Two more numbers can be pulled out of this list, and they answer different questions.

Average total cost answers how much each batch costs on average. Divide total cost by the number of batches. At 4 batches, average total cost is $120 ÷ 4 = $30 per batch. At 2 batches it is $95 ÷ 2 = $47.50. Average cost usually falls at first because the $60 of fixed cost is spread over more batches. Then it stops falling and turns up, for a reason the next lesson explains.

Marginal cost answers what one more batch would add to total cost. Subtract the total cost before from the total cost after. The third batch raises total cost from $95 to $105, so its marginal cost is $10. The fifth batch raises it from $120 to $140, so its marginal cost is $20. The word marginal always means the next one, the extra one, the one at the edge.

People mix these up because both are stated in dollars per batch. Average cost looks backward at everything made so far. Marginal cost looks forward at the next unit only. When the shop decides whether to take one more order, the number that matters is marginal cost, because the fixed $60 is already spent either way. Keep the two questions separate: what does each batch cost on average, and what would the next batch cost?

Marisol’s truck has one grill, one prep counter and one window. Watch what happens as she adds workers during a lunch rush. Alone, she can turn out 40 tacos an hour. With a second worker taking orders and money, output jumps to 90 an hour. A third worker, chopping and wrapping, brings it to 120. A fourth brings 135. A fifth, squeezed in by the door, brings 140. The truck still makes more tacos with each hire, but by less and less.

The extra output from one more worker is the marginal product of that worker. Here it is 40, then 50, then 30, then 15, then 5. It rises at first because two people can split the work sensibly. Then it falls, not because anyone is lazy, but because the grill and counter are fixed. Five people cannot all reach one grill. This pattern is the law of diminishing returns: when you add more of one input to a fixed amount of another, the marginal product eventually falls.

Diminishing returns explains why marginal cost eventually rises. Suppose each worker costs $20 an hour. The third worker adds 30 tacos for $20, about 67 cents of labor per extra taco. The fifth worker adds 5 tacos for $20, or $4 of labor per extra taco. Each extra taco gets more expensive to make, so the marginal cost curve turns upward. In the print shop, the same crowding around one press explains why marginal cost began climbing after the third batch.

Do not confuse diminishing returns with falling output. Total output kept rising in the truck all the way to five workers. Diminishing means the additions shrink, not that the total drops. Only if Marisol jammed in a sixth or seventh worker, so that people bumped into each other and dropped orders, would output actually fall. That is called negative returns, and a sensible owner never gets there.

Diminishing returns is a short-run story: the truck is fixed. In the long run Marisol can buy a bigger truck, or two, or a restaurant. Now a new question appears. When a business grows and every input grows with it, what happens to the average cost of each unit? Often it falls. When average total cost falls as a firm gets larger, the firm has economies of scale.

Take two breweries as a labeled example. A small one makes 10,000 barrels a year at a total cost of $2,000,000, so each barrel costs $200 on average. A large one makes 100,000 barrels at a total cost of $12,000,000, or $120 per barrel. The big brewery is not more careful. It is cheaper per barrel for three reasons. Its large fixed costs, such as tanks and a bottling line, are spread over more barrels. Its workers can specialize in one task each. And it buys grain and bottles in bulk at lower prices.

Economies of scale explain a lot of what you see. Illinois farms have grown larger over the decades partly because a combine costs about the same to own whether it harvests 500 acres or 2,000. Big-box stores, airlines and car makers all run on the same logic. It is also why a lone food truck cannot match a chain’s prices on paper goods and meat.

Bigger is not always cheaper, though. Past some size, a firm becomes hard to run. Layers of managers slow decisions, workers feel unseen, and the head office loses touch with customers. When average cost starts rising as a firm grows, it faces diseconomies of scale. Between the two lies a range where size makes no difference, called constant returns to scale. The best size for a business is wherever its long-run average cost is lowest, and that differs by industry: small for barbers, huge for airplane makers.

How many tacos should Marisol make in a day? Not as many as possible, and not the number that makes the average cost lowest. The answer comes from comparing two margins, one unit at a time. Marginal revenue is the extra revenue from selling one more unit. Marisol is a small seller in a crowded downtown lunch market, so she can sell as many tacos as she makes at the going price of $3. Each extra taco brings in $3, so her marginal revenue is $3.

Now line that up against marginal cost. Because of diminishing returns, marginal cost rises as the day gets busier. Suppose the 100th taco of the day costs $1.50 to make, the 200th costs $2.00, the 300th costs $2.60, the 400th costs $3.00 and the 500th costs $3.60. The 300th taco earns $3 and costs $2.60, so it adds 40 cents of profit. Make it. The 500th earns $3 but costs $3.60, so it loses 60 cents. Do not make it.

This is the profit-maximizing rule: keep producing as long as marginal revenue is above marginal cost, and stop where they are equal. Marisol should make about 400 tacos a day. Making fewer leaves profitable tacos unmade. Making more adds tacos that cost more than they bring in. The rule works for a farm deciding how many acres to plant and a factory deciding how many shifts to run. It is the heart of what economists call marginal analysis: decide at the edge, one unit at a time.

Two mistakes are common. Some owners try to maximize profit per unit, which would mean making very few tacos and missing all the profit on the rest. Others try to maximize revenue by selling as many as possible, which means making tacos that lose money. If the going price rose to $3.60, the rule would tell Marisol to move out to 500 tacos, because the 500th now breaks even. If the price fell to $2.60, she should pull back to 300. The rule does not change; the answer does.

Lupe asked Marisol a fair question: how many tacos do you have to sell just to cover your costs? That number is the break-even point, where total revenue equals total cost. Each taco sells for $3 and uses $1.10 of variable cost, so each taco leaves $1.90 to put toward the $2,400 of fixed costs. Divide: $2,400 ÷ $1.90 is about 1,263. Marisol must sell roughly 1,264 tacos a month before she earns her first dollar of profit. Below that she loses; above it she gains $1.90 per taco.

Now picture a slow winter month. Marisol sells only 1,000 tacos, for $3,000 in revenue. Her variable costs are 1,000 × $1.10 = $1,100, and her fixed costs are still $2,400. Total cost is $3,500, so she loses $500. Should she close for the month? Compare the loss from staying open to the loss from closing. If she closes, she still owes the $2,400 of fixed costs and takes in nothing, so she loses $2,400. Open, she loses only $500. Open is better.

The reason is that revenue of $3,000 more than covers the variable cost of $1,100, leaving $1,900 to chip away at the fixed bills. This is the shutdown rule. In the short run, stay open as long as price is at least average variable cost, the variable cost per unit. Here average variable cost is $1.10 and the price is $3. Only if the price fell below $1.10, so that each taco lost money on ingredients alone, would closing be the smaller loss.

The break-even point and the shutdown point answer different questions. Break-even asks when the business earns a profit; it includes fixed costs. Shutdown asks whether to operate at all this month; it ignores fixed costs because they are owed either way. In the long run, a business that never reaches break-even should leave the industry, since the lease will eventually end and the truck can be sold. But in the short run, many businesses correctly stay open while losing money.

When Marisol started, she did not fill out any special form to become a business. She just was one. A business owned and run by one person is a sole proprietorship. It is the most common form of business in the United States by count: most businesses are one-person operations, from barbers to freelance coders to farm stands. Yet all of them together bring in a small share of total business sales, because each one is small.

The advantages are real. A sole proprietor decides everything, keeps every dollar of profit, and pays tax on that profit once, as personal income. The disadvantages are just as real. The biggest is unlimited liability: the business and the owner are legally the same, so if the truck is in an accident and a court awards $200,000, Marisol’s savings, car and home can be taken to pay it. A sole proprietor also finds it hard to raise money, and the business ends when the owner quits or dies.

A partnership is a business owned by two or more people who share profits and decisions under an agreement. If Lupe joins Marisol, the truck gains Lupe’s bookkeeping skill and her savings, and both sisters share the workload. Law firms, medical practices and many family businesses are partnerships. The cost is shared control: if one sister wants a second truck and the other does not, someone loses. And in an ordinary partnership, each general partner has unlimited liability for the debts of the whole business, even debts the other partner ran up.

The choice between these forms is a trade-off, like every choice in this course. Going alone buys speed and control at the price of risk and limited money. Partnering buys money and skills at the price of shared control and shared exposure. The next lesson shows a third form that changes the liability problem entirely.

A corporation is a business that the law treats as a person of its own, separate from the people who own it. It can own property, sign contracts, borrow, sue and be sued in its own name. Its owners hold shares of stock, and each share is a slice of ownership. A small corporation might have three shareholders; a giant one might have millions. In Illinois, a corporation comes into being when its founders file papers with the Secretary of State.

The great advantage is limited liability. If a corporation is sued or goes broke, its shareholders can lose what they paid for their shares, but no more. Their homes and savings are safe. That protection makes people willing to invest in businesses they do not run, which is the second advantage: a corporation can raise large sums by selling stock to many investors or by borrowing through bonds. A corporation also has unlimited life. Shareholders come and go, but the company goes on.

The disadvantages are cost and taxes. Setting up and running a corporation means more paperwork, more rules and more reporting. And profits can be taxed twice: the corporation pays corporate income tax on its earnings, and when it pays some of those earnings to shareholders as a dividend, the shareholders pay income tax on that too. Owners also share control. Shareholders elect a board of directors, and the board hires managers, so a founder can be outvoted or even fired.

Many small businesses now choose a middle path, the limited liability company, or LLC. An LLC gives its owners limited liability like a corporation, but its profits are normally taxed only once, on the owners’ personal returns, like a proprietorship or partnership. It is simpler to run than a corporation. For Marisol, forming an LLC would mean that a lawsuit over the truck could reach the truck and the business’s bank account, but not her home. Most economists would say that protection is worth the filing fee.

Land, labor and capital do not organize themselves into a food truck. Someone has to spot the chance, borrow the money, sign the lease, hire the helper and decide what goes on the menu. That person is the entrepreneur, the fourth factor of production. The entrepreneur’s special job is to bear risk. The helper gets paid every Friday whether April was good or bad. Marisol gets paid only if there is something left after everyone else.

Profit is the entrepreneur’s reward for taking that risk and for getting the combination right. It is also a signal to everyone else. When food trucks in a neighborhood earn high profits, other cooks notice and open trucks of their own. The new supply pushes prices and profits down toward normal. When trucks lose money, some close, supply falls and prices recover. Profit and loss steer resources toward what people want and away from what they do not, without anyone in charge.

The risk is not a figure of speech. Government data show that roughly half of new businesses in the United States are still operating five years after they open. The other half close, often taking the founder’s savings with them. Chicago’s history is full of both outcomes. Cyrus McCormick moved his reaper factory to the city in 1847 and built one of the largest firms in the world; countless other Chicago ventures vanished without a trace.

Entrepreneurs also drive innovation, the introduction of new products and better ways of making old ones. A new app, a cheaper solar panel or a faster way to wrap a taco all begin as somebody’s risky bet. Economies with many entrepreneurs tend to grow faster because more bets are placed and the winners spread. Not everyone should start a business; the numbers say most should not. But an economy needs the ones who do.

Put the whole chapter to work on one business, a labeled example. Nadia runs a small screen-printing shop in Chicago’s Lakeview neighborhood that prints custom T-shirts for schools, teams and bars. Her fixed costs are $4,000 a month: rent, the lease on her press, insurance and a website. Each shirt uses about $6 of blank shirt and ink, her variable cost. She charges $15 a shirt. Her startup costs two years ago, mostly the press and a deposit, were $22,000, which she has already paid; they are sunk and do not enter today’s decisions.

First, the margin. Each shirt brings in $15 and costs $6 to make, leaving a margin of $9 to cover fixed costs and then become profit. Break-even is $4,000 ÷ $9, about 445 shirts a month. In a typical month Nadia sells 900 shirts. Revenue is 900 × $15 = $13,500. Variable cost is 900 × $6 = $5,400, so total cost is $9,400 and accounting profit is $4,100. She gave up a $3,500-a-month job to do this, so her economic profit is about $600. She is doing a little better than her next-best option, and she gets to be her own boss.

Now the growth question. A second press would raise her fixed costs to $5,500 a month, but it would let her print faster and buy blanks in larger lots, cutting variable cost to $5 a shirt. Her margin would rise to $10 and her new break-even would be $5,500 ÷ $10 = 550 shirts. If sales stayed at 900, profit would be 900 × $10 − $5,500 = $3,500, which is worse than today. The second press only pays if it brings more orders. At 1,200 shirts a month, profit would be 1,200 × $10 − $5,500 = $6,500, a clear gain.

So the decision turns on one estimate: can she sell more than 960 shirts a month with the second press? At 960 shirts, 960 × $10 − $5,500 = $4,100, exactly what she earns today. Notice how she got there. She separated fixed from variable cost, ignored the sunk startup cost, computed the margin and the break-even, subtracted her opportunity cost, and compared the extra revenue of growing with its extra cost. She also has to watch cash flow, since the new press must be paid for before the new orders arrive. That is the economic way of thinking applied to a Tuesday afternoon.

A market structure is the way a market is organized: how many firms compete, how alike their products are, and how easy it is to join. Economists use three questions to sort any market. First, count the sellers. Is it thousands of wheat farmers, a few airlines, or one water company? Second, check how hard it is to enter. Can a new seller open next month with a few thousand dollars, or does it need a billion dollars and a government license? Third, ask whether the buyer can leave. If the price goes up, is there a close substitute?

The answers tell you how much market power sellers have. Market power is the ability of a firm to raise its price above what competition would allow without losing all its customers. Jordan’s pizza shops had very little. If one charged $17 when the rest charged $14, it lost most of its customers. The Carvel phone company had a lot. It charged $80 a month, and families kept paying because the other choice was no phone that worked at home.

The second question points to barriers to entry: anything that makes it hard for a new firm to start competing. A cell tower that costs a fortune, a patent, a license that the city limits, or a brand that customers trust for decades can all keep rivals out. Where barriers are low, high profits attract newcomers, and prices get pushed back down. Where barriers are high, profits can last for years.

Put the three answers together and most markets fall into one of four types, from the most competitive to the least: perfect competition, monopolistic competition, oligopoly and monopoly. These are models, not labels printed on a store. Real markets sit somewhere along the line, and the same product can land in different spots in different places. Phone service in Chicago is one kind of market; in Carvel it is another.

Picture a corn farmer near Champaign with 50,000 bushels to sell. On the day she sells, the market price is $4.50 a bushel, a labeled example. Thousands of other farmers are selling the same grade of corn, and buyers cannot tell one farm’s corn from another’s. If she asks $4.60, buyers simply buy from someone else, and she sells nothing. If she asks $4.40, she sells the same 50,000 bushels but gives away 50,000 × $0.10 = $5,000 for no reason. So she sells at $4.50 and earns 50,000 × $4.50 = $225,000.

This is perfect competition. It has many buyers and sellers, an identical product, and easy entry and exit, and everyone knows the price. Each seller is a price taker, a firm too small to affect the market price, so it accepts the price the whole market sets. Grain prices are a good example. The Chicago Board of Trade was founded in 1848. Trading there has long set prices for corn, wheat and soybeans that Midwest farmers watch every day.

The key result shows up in the long run. Suppose corn prices rise and farmers earn an economic profit. Other farmers switch fields from soybeans to corn, and new growers enter. Supply shifts right and the price falls until economic profit is gone. If prices fall and farmers lose money, some switch crops or leave, supply shifts left and the price recovers. Firms in perfect competition end up earning normal profit, and the price ends up near the lowest possible average cost.

A common mistake is to think perfect means good, or that such markets are everywhere. Very few markets meet every condition. Farm crops, some metals and stock shares come close. The model is useful anyway, because it is the benchmark. It shows what prices and profits would look like with the most competition possible, so we can measure how far other markets fall from it.

Why did no second phone company come to Carvel? Run the numbers, using labeled examples. Suppose a new tower and its equipment cost $1,500,000. The town has 2,400 people and perhaps 1,500 phone lines. If a newcomer won half of them, 750 lines, and earned $25 a month above its operating costs on each, it would take in 750 × $25 × 12 = $225,000 a year toward the tower. Paying back $1,500,000 would take between six and seven years, and the first company could cut its price the day the newcomer arrived. Few investors take that bet.

That is a barrier based on economies of scale: one firm serving the whole town has lower average cost than two firms splitting it. Other barriers are legal. A patent gives an inventor the sole right to make and sell an invention, usually for 20 years from the date the patent application was filed. Drug companies depend on patents to earn back research costs. Governments also limit entry with licenses. For decades Chicago capped the number of taxi medallions, the permits that let a cab pick up riders on the street.

Some barriers come from owning something no one else can get. For much of the 20th century, one company, De Beers, controlled most of the world’s supply of rough diamonds. Others come from the product itself. A network effect exists when a product becomes more valuable as more people use it. A social media app with a billion users is hard to challenge, because a new app with no users is worth little to anyone.

Barriers are not always bad. Patents reward invention, licenses can protect safety, and economies of scale can make a product cheaper than it would be if many small firms made it. The cost is that firms protected by barriers face less pressure to cut prices or improve. Economists weigh these trade-offs case by case. The first question to ask about any profitable business is always the same: what is stopping someone else from doing this too?

Jordan’s pizza shops were not price takers like the corn farmer. Each pizza was a little different: one had a crispier crust, another stayed open until 3 a.m., another was on the corner by the train. Suppose Tony’s sells 300 large pizzas a week at $14, a labeled example. When Tony raises his price to $15, some customers leave, but his regulars stay, and he sells 270. His revenue goes from 300 × $14 = $4,200 to 270 × $15 = $4,050. He lost some sales but not all of them. That is a small amount of market power.

A market like this is called monopolistic competition: many sellers, easy entry, and products that are similar but not identical. The name sounds like a contradiction. It means each firm has a tiny monopoly on its own version of the product, while it still competes with many close substitutes. Restaurants, hair salons, clothing stores, gas stations on different corners and dentists usually fit this model.

The key idea is product differentiation: making a product seem different from rivals’ products so that some buyers prefer it. Firms differentiate by quality, style, location, service, hours and brand. Because buyers have favorites, each firm faces a demand curve that slopes down. It can raise its price a little without losing everyone, which the corn farmer could never do.

The long run looks like perfect competition, though. Entry is easy, so when pizza shops earn big profits, new shops open and take some customers from each old one. Each shop’s demand shrinks until economic profit is about zero. That is why Jordan’s twenty pizza shops did not seem rich. The cost to society is that prices sit a bit above the lowest average cost. The benefit is variety: twenty kinds of pizza instead of one.

Tony’s pizza spends $2,000 a month on online ads, a labeled example. Is it worth it? Use the margin from the last chapter. Each large pizza sells for $15 and uses $6 of ingredients, so each extra pizza adds $9 toward fixed costs and profit. If the ads bring in 200 extra pizzas a month, they add 200 × $9 = $1,800, which is less than the $2,000 they cost. If they bring in 300 extra pizzas, they add $2,700 and pay for themselves with $700 to spare. An ad is a business decision like any other.

Advertising is paid communication meant to persuade buyers to purchase a product. Firms in monopolistic competition and oligopoly advertise heavily, because they sell products that can be told apart. The corn farmer never advertises, since her corn is identical to everyone else’s. Much advertising builds a brand, a name or symbol that buyers link with a certain product and level of quality.

Economists disagree about whether advertising helps buyers. On one side, ads give information: prices, sales, new products and store hours. They lower the cost of shopping around, and a trusted brand signals that the maker has a reputation to protect. On the other side, some ads mainly persuade, creating a difference in buyers’ minds that is not in the product. The cost of those ads is added to the price, and a strong brand can become a barrier to new rivals.

Store brands show the trade-off. A store-brand pain reliever often contains the same active ingredient in the same amount as the famous name brand, and it usually costs less. Buyers who choose the name brand are paying partly for the reassurance of the brand. Whether that is a waste or a fair price for trust depends on the buyer. Reading labels is how a consumer checks.

An oligopoly is a market dominated by a few large firms, often protected by high barriers to entry. National wireless service is one example: since two carriers merged in 2020, three large companies have served most U.S. phone customers. Airlines, soft drinks and car manufacturing are others. One way to measure it is the concentration ratio: the share of total sales made by the four largest firms. If four firms have 35%, 25%, 20% and 10% of sales, the four-firm concentration ratio is 35 + 25 + 20 + 10 = 90%.

What makes oligopoly different is interdependence: each firm’s best move depends on what its rivals do. Picture two gas stations at a rural crossroads, a labeled example. If both charge $3.80 a gallon, each earns $900 a day. If both charge $3.60, each earns $600. If one charges $3.60 while the other stays at $3.80, the cheaper one takes most of the traffic and earns $1,100, while the other earns $300. Each owner reasons: whatever my rival does, I earn more by cutting. So both cut, and each ends up with $600 instead of $900.

The two stations would both be better off agreeing to stay at $3.80. An agreement among firms to raise prices or split markets is collusion, and in the United States it is illegal under antitrust law. A formal group of firms or countries that sets output and prices together is a cartel. OPEC, a group of oil-exporting countries, is the best-known example. Its members are nations, so U.S. law does not reach it. Cartels are hard to hold together, since each member earns more by quietly cheating.

Oligopolies often avoid price wars without talking to each other. They compete through ads, loyalty programs and features instead of price, and prices can stay fairly steady for long stretches. Whether an oligopoly behaves more like competition or more like a monopoly depends on how easy it is to watch rivals, how much the firms differ, and how closely the government enforces antitrust law.

A monopoly is a market with one seller of a product that has no close substitutes, protected by barriers that keep rivals out. The Carvel phone company is a small monopoly. Unlike the corn farmer, it is a price maker: it can set its price. But it cannot charge any price it likes and still sell to everyone. Demand still slopes down. Suppose, as a labeled example, that at $80 a month 1,000 lines sign up, at $70 there are 2,000, at $60 there are 3,000, at $50 there are 4,000 and at $40 there are 5,000.

Now find total revenue at each price: $80,000, $140,000, $180,000, $200,000 and $200,000. Marginal revenue is the extra revenue from each added block of 1,000 lines. Going from 1,000 to 2,000 lines adds $60,000, or $60 per line, even though the price is $70. Marginal revenue is below price because the monopoly must lower its price for every customer, not just the new ones. The next blocks add $40, $20 and $0 per line.

Suppose each line costs the company $30 a month to serve. It follows the same profit rule as any firm: keep adding customers while marginal revenue is above marginal cost. The third block adds $40 per line, more than $30, so take it. The fourth adds only $20, less than $30, so stop. The monopoly serves 3,000 lines at $60. Competing firms would keep adding customers until price fell toward the $30 cost, serving more people at a lower price.

That comparison is the case against monopoly. It charges more, sells less, and some people who would gladly pay more than the $30 cost go without. Economists call the lost value deadweight loss: gains from trade that never happen. A common mistake is to think a monopoly always earns huge profits. It still has costs, and if demand is weak it can lose money. Another mistake is to think it charges the highest possible price. At $80, it would earn less than at $60.

Think about the water pipes under your street. Suppose, as a labeled example, a town’s water system costs $2,000,000 a year to own and maintain, and delivering water to each home costs another $100 a year. If one company serves all 10,000 homes, the average cost per home is $2,000,000 ÷ 10,000 + $100 = $300 a year. If two companies each lay their own pipes and split the town, each has the same $2,000,000 cost but only 5,000 homes: $400 + $100 = $500 a year. Two companies make water more expensive.

This is a natural monopoly: an industry where one firm can supply the whole market at a lower average cost than two or more firms could. It happens when fixed costs are huge and the cost of serving one more customer is small. Water, sewer, natural gas pipes and the wires that carry electricity to homes are the classic cases. Competition here would waste money on duplicate pipes and wires.

The problem is that a single unregulated seller of water could charge a monopoly price. So governments step in, in one of two main ways. They can own the system; many cities, Chicago among them, run their own water systems. Or they can let a private company run it under regulation, rules set by government. In Illinois, the Illinois Commerce Commission reviews and approves the rates that investor-owned utilities such as ComEd, which delivers electricity in northern Illinois, can charge.

Regulators usually aim for a price that covers the utility’s costs plus a fair return for its investors, close to the $300 average cost in the example. That protects customers from monopoly prices. It has drawbacks too. A firm that is sure to cover its costs has less reason to cut them, and regulators may not know the true costs. Economists debate how to reward utilities for efficiency, and in some states the power plants that generate electricity now compete even though the wires remain a monopoly.

In the 1870s and 1880s John D. Rockefeller’s Standard Oil bought up rival refineries, won secret discounts from railroads, and came to control roughly 90% of the oil refined in the United States. It organized its holdings as a trust, a legal device that put many companies under one board. Railroads, sugar and other industries followed. Farmers and small businesses complained that trusts crushed rivals and raised prices, and the word became a political one.

Congress answered with the Sherman Antitrust Act of 1890. Antitrust law is the set of laws that promote competition by banning collusion and unfair attempts to monopolize. Section 1 of the Sherman Act forbids agreements in restraint of trade, such as price fixing. Section 2 forbids monopolizing or attempting to monopolize. In 1911 the Supreme Court ruled that Standard Oil had broken the law and ordered it split into 34 separate companies. In 1914 Congress added the Clayton Act and created the Federal Trade Commission.

Today the Justice Department’s Antitrust Division and the Federal Trade Commission enforce these laws. Price fixing is treated most harshly. In 1996 Archer Daniels Midland, then based in Decatur, Illinois, pleaded guilty to fixing the price of a feed additive and paid a $100 million fine, a record at the time. Three of its executives were later sentenced to prison.

A common mistake is to think that being a monopoly is itself illegal. It is not. A firm that wins nearly the whole market by making a better or cheaper product has broken no law. What the law forbids is gaining or keeping a monopoly by unfair means, such as agreeing with rivals, threatening suppliers who deal with competitors, or buying up every rival to shut out competition. Courts often have to judge which is which.

A merger is the joining of two companies into one. Suppose, as a labeled example, a region has four grocery chains with 30%, 25%, 25% and 20% of sales. The second and third chains want to merge. The new company would have 50% of the market, and shoppers would have three chains to choose from instead of four. Large mergers must be reported to the federal antitrust agencies before they close, and the agencies can go to court to block one they think would harm competition.

How should they decide? Starting in the 1950s, economists and law professors at the University of Chicago, among them George Stigler and Richard Posner, argued that antitrust should focus on one question: will this make consumers better or worse off, through prices, output and quality? This approach, often called the Chicago School, stressed that big firms are often big because they are efficient. A merger that cuts costs could lower prices. By the late 1970s and 1980s, courts and agencies had largely adopted this consumer welfare standard.

Apply it to the grocery merger. If combining warehouses and trucks saves the merged chain 3% of its costs and competition from the other two chains forces it to pass on the savings, shoppers gain. If instead the merged chain, with half the market, can raise prices 5% because shoppers have fewer places to go, shoppers lose. The agencies study past price data and store locations to guess which story is closer to the truth.

The approach is debated. Supporters say it gives courts a clear test and keeps them from punishing firms just for success. Critics argue that it underrates the harm of growing concentration, including lower wages when few employers compete for workers, less new business formation, and the political power of very large firms. In recent years some enforcers and scholars have pushed for tougher rules. Both sides agree that the questions are about evidence: what will this deal actually do?

A job is a market too. The wage is the price of labor, and it is shaped by supply and demand like any price. Employers demand workers because workers help produce things to sell. Suppose a warehouse worker packs 30 boxes an hour, and each box earns the company $1 above its other costs, a labeled example. That worker adds $30 an hour. The company will hire at any wage up to about $30 an hour, but not above it. Skills and tools that raise what a worker produces tend to raise what employers will pay.

Market structure matters in labor markets as well. When many employers compete for workers, a worker who is paid too little can leave for a better offer. When one employer dominates a town’s jobs, workers have fewer places to go. A market with a single main buyer of labor is a monopsony, and it can pay lower wages than competition would. A small town with one large factory or hospital can come close to this. Economists use it as one argument in the debate over the minimum wage.

A labor union is an organization of workers that bargains with employers as a group. That process is collective bargaining: the union negotiates wages, benefits and working conditions for all its members at once. Chicago has a long labor history. In 1894 the Pullman Strike, which began among workers at the Pullman railroad car works on the city’s South Side, spread across the country and was ended by federal troops. The National Labor Relations Act of 1935 gave most private-sector workers the legal right to organize and bargain.

Economists debate unions’ effects. Supporters say unions raise pay and improve safety, give workers a voice, and balance the power of large employers. Critics say union wages above the market level can raise costs, reduce the number of jobs, and make firms less flexible. Union membership has fallen over the decades. According to the Bureau of Labor Statistics, about 20% of U.S. wage and salary workers belonged to unions in 1983, and about 10% did in recent years.