A university lays a new quad. The architect draws crisp concrete walkways between the buildings — straight lines, right angles, exactly where the plan says people should walk. Grass everywhere else. By October there is a brown stripe cutting diagonally across one corner of that grass, worn down to dirt by ten thousand footsteps, going precisely where no walkway was drawn. The groundskeeper is furious. He puts up a small fence. Within a week there is a new dirt stripe curving around the fence.

Planners have a name for that brown stripe, half-affectionate and half-defeated: a desire path . It is the route people actually want, written into the ground by the only authority that could possibly know it — their own feet. The smartest campuses have stopped fighting it: they wait a year, see where the dirt stripes form, and pave those.

That stripe is the cleanest example you will ever see of a spontaneous order — a pattern that looks designed but isn't, emerging from many people each making their own small choices, with no one in charge of the whole. A route that fits human need exactly, produced by no architect, authored only by feet.

Once you have the idea, you see it everywhere. Language is a spontaneous order — no committee invented English, and its words and rules keep shifting without anyone's permission. (When France set up an academy to command its language by decree, the language mostly ignored it.) A footpath through a forest, a financial market's price for a stock, the rough consensus on which side of the sidewalk to pass — all are orders that no one ordered. Even an illegal market is one: no board runs the drug trade, no office assigns its routes, no company advertises the jobs — yet remove the boss at the top and the trade reorganizes and keeps supplying the city anyway, an order no one ordered and no one can easily un-order.

So if no one is in charge of bread, what coordinates it? What tells the farmer to plant, the miller to grind, the baker to rise at four, and routes the loaves to exactly the shops where they'll be wanted — with no commander, no master list? The answer is a single, humble thing: a price. The rest of this chapter is the machinery of how a price does that job. We build it from its two halves — what buyers are both willing and able to purchase (demand) and what sellers are willing and able to offer (supply).

A brief note: We use algebra to model things in economics, and we're going to start building one of the most important models in all of economics shortly. If you want to understand the interaction of supply and demand, you need to be able to use algebra to manipulate them.

We use linear equations to model simple linear relationships. Any linear relationship can be modeled in four ways (we call it the "rule of four" and it's a key thing to be able to do if you want to be successful in economics). The first is the story: the form you always see in word problems. Next, we often will see a linear equation. Then, we have a table of values. Finally, we have one of the economist's favorite tools, the graph. Get used to using all four of these, and translating between them in your head. You should be able to manipulate them fluently.

Imagine for a moment that I could survey the population with a perfect truth serum: I ask them "how much would you be willing and able to buy of a certain good at a price of X?" That number is the quantity demanded of a good. Notice that it is a combination of two things: a price, and a quantity that consumers would want to purchase at that price. Note, too, that it only tells you what consumers want at one precise point.

Now, let's imagine that you use the same truth serum and somehow ask consumers hundreds of times and didn't just get one pair of prices and quantities, but all of them. What you would have is a measure of how willing and able consumers are to buy a given good at every price. This is called demand.

Now, let's do a little bit of an intuition pump: What is the shape of a given quantity demanded?

A single pair of coordinates is a just a single point. In algebra, we represent those with coordinate pairs. For example, if 10 people want to buy my garden-fresh tomatoes at a price of $2, then the coordinate pair would be (10, 2). If 1,000 people want to buy my cucumbers at a price of $0.01, then the quantity demanded would be 1000, and the combination would be (1000, 0.01). If 0 people want to buy my definitely-not-poison magic potion at a price of $100, the coordinate would be (0, 100).

But what about all the various combinations at once? That's demand , and it is shaped like an algebraic curve. You can represent it in a table (we call them "demand schedules" in economics) or a set of points, and you can also use a story. In real life, such curves are complicated; in this class, we'll use linear demand curves to keep things simple. They are the same kind you learned about in 8th grade algebra: y = mx + b. (If you need a refresher, now is a good time to study up. Chapter 0 might help!) Of course, since we're using quantity and price, we'll put price on the y-axis, and quantity on the x-axis. Our simple demand equation will look something like this:

Qd = a - bP

Now let's ask about the shape of the demand curve.

Economists assume people respond to incentives, and the first systematized example is called the law of demand: all else equal, the lower the price, the more people buy . It makes plain sense — you buy more of something when it's cheaper, and you hold back when it's dear. This is your cue: call your mother right now and tell her that you have learned that demand curves are downward-sloping! How exciting! And then, when she asks you what on earth that means, you can tell her that it means that people are rational!

Notice: Demand is the relationship. Quantity demanded is a point along the demand curve.

Plotted with price on the vertical axis and quantity on the horizontal, that relationship slopes down. This is just a visual form of the intuition of the law of demand: demand curves slope down because people buy more when the price is lower. Raise the price of a thing, and people buy less.

Let's imagine the market for one-bedroom apartments in a mid-sized city. Let's build a demand schedule. Where does a schedule like this come from ? Constructing a real demand schedule can be a bit complicated, but a simple version is to use a survey to figure out how many people would buy at a given price. Imagine surveying renters: "What's the most you'd pay each month for a one-bedroom before you'd walk away?" That number — a person's reservation price — is the height of their personal demand. Someone whose reservation price is $1,500 is a "yes" at any rent up to $1,500 and a "no" above it. To get the market quantity demanded at a given rent, you add up anyone who is willing to buy at that price.

The same schedule as an equation: Q _d = 900 − 0.5P. Plug in any rent and it returns the quantity demanded; at $1,400, for example, 900 − 700 = 200. Story, table, equation, and (next) graph: four forms of one relationship.

Now flip to the landlords. Same question, other side of the table: at each possible rent, how many one-bedroom apartments are owners willing and able to put on the market?

Here the schedule tilts the other way. At a rock-bottom rent, many owners would rather leave a unit empty, convert it, or sell — it isn't worth the hassle. As the rent climbs, more units come onto the market: owners renovate basements, split large units, build. This is the law of supply: all else equal, the higher the price, the more sellers offer . It makes plain sense too — a better price is worth more effort, so people supply more of what pays well. The supply curve slopes up .

Now, a little hint that will help you economics majors in the upper-level classes: behind all that effort is the question of cost. The first units to reach the market are the cheap ones, like the empty unit that just needs a listing. Each extra unit costs more to bring to market (the basement that needs renovating, the building that takes years to put up), and only a higher rent covers it. For those of you who know some calculus, the derivative of the supply curve is marginal cost; not needed for this course, but important for later on.

Let's get back to a simple rental apartment market. Notice that quantity supplied is still just a single point; supply is the full set of combinations, which makes up a curve.

As an equation: Q _s = −300 + 0.5P. At $1,400, that's −300 + 700 = 400 apartments.

So far only one thing has been allowed to change: the price. Slide the rent up or down and you read a different row of the same schedule. But the price isn't the only thing that affects how much buyers are willing and able to buy, or how much sellers are willing and able to offer. When one of those other things changes, the whole schedule changes, every row at once, and the curve itself moves. Economists call these factors the determinants of demand and supply. You'll also hear them called "shifters."

Demand shifts when any of these change:

Supply shifts when any of these change:

There is one thing missing from both lists, and it's worth noticing: the price of the good itself. A change in the good's own price moves you along the curve; a change in anything on these lists moves the curve itself. That distinction trips up more students than anything else in this course, so it gets its own concept, next. (Note: There is one more determinant of demand, the prices of related goods, which we will cover in Chapter 4.)

Perhaps the most common mistake at this stage of microeconomics is mixing up a single point on a curve with the curve itself. For example, Qd is not the same as D. Qs is not the same as S. Many students get tripped up by it, and I want to get ahead of it with you. The two phrases sound interchangeable, but they aren't. If we can get this right, you're going to do a lot better moving forward.

In everyday speech, and in nearly every news story, "demand went up" just means people bought more. Economists use the phrase more precisely: demand went up only if buyers are willing and able to buy more at the same price. When people buy more simply because the price fell, economists say the quantity demanded rose, and demand itself never moved. You'll hear the loose version for the rest of your life. In this course, use the precise one. Especially on exams!

"Quantity demanded" (just like 'quantity supplied') is a single point — the amount at one particular price. "Demand" (and "supply") is the whole curve — the entire schedule, the whole line. The difference matters because the two change for different reasons and look different on a graph:

On a graph the tell is simple. If you've moved to a different spot on the same line, that's a change in quantity demanded. If the entire line has picked up and moved, that's a change in demand. The same-price test is easy to read off the picture, too: run your finger across at any one price. If the quantity where your finger meets the curve has changed, the curve moved.

The Rule of Four gives you one more way to see it. Our apartment demand is Q _d = 900 − 0.5P. Change P and the same equation hands you a new quantity: that's a movement along the curve. Now suppose a large employer moves to town, and at every rent, 100 more households want a one-bedroom. The equation itself changes, to Q _d = 1000 − 0.5P. The 900 became 1000.

That is the whole distinction in algebra. A change in P is a movement along the curve. A change in the constant is a shift of the curve. In a table, a movement means reading to a different row; a shift means every row changes at once.

Look at the highlighted row. The rent is $1,200 in both columns, yet the quantity demanded went from 300 to 400. Same price, different quantity: that's the same-price test, and it says demand shifted.

Each sentence below uses "demand" or "supply" the loose, everyday way. Rewrite it the precise way in your head, then reveal.

Neither side sets the price alone; it comes from buyers and sellers pulling against each other. Put both schedules on the same axes. The demand curve slopes down, the supply curve slopes up, and being two lines that lean opposite ways, they cross — once . At that crossing, and nowhere else, the quantity renters want to lease exactly equals the quantity landlords want to offer. That price is the equilibrium , the market-clearing price .

From our two schedules, the lines cross at a rent of $1,200, where quantity demanded (300) equals quantity supplied (300). And what's cool? You can get there four different ways: as an equation, a table, a graph, and a story. That isn't decoration. Each representation catches what the others hide, and a thing you can move fluently between all four is a thing you actually understand.

The magic of equilibrium is that the market will get there all on its own, even if the original set price is wrong. Suppose rent sits at $1,000, below equilibrium. Renters want 400 (Qd); landlords offer only 200 (Qs). That gap is a shortage of 200 apartments. Two hundred renters who want a place at $1,000 can't get one. What do they do? They offer a little more. The rent gets bid up — and as it rises, quantity demanded slides down the demand curve while quantity supplied climbs up the supply curve, closing the gap from both ends, until at $1,200 the pressure is gone.

Now let's flip the script. Suppose rent sits at $1,400, above equilibrium. Landlords offer 400; renters want only 200. That gap is a surplus of 200 empty units. Empty apartments earn nothing, so landlords shade the rent down to fill them — and again the two quantities slide toward each other until, at $1,200, the market clears.

This is the desire path forming. No one announces $1,200. It is the dirt stripe left behind by thousands of renters and landlords each pursuing their own plans. Set the price wrong and the market doesn't sit there broken — it moves , the way water finds its level, until plans become compatible. A shortage or a surplus never lasts long in a competitive market, because competition puts pressure on sellers to lower their prices to clear inventory, or on buyers to pay more lest they not get something they want. This is the invisible hand in action: the bread-commissar's job from Concept 1 — finding the number that makes a city's housing appear — done by no one, by everyone, all at once.

When inflation occurs, many politicians love to blame the greed of corporations and businesses. Is that fair?

Blaming high prices on corporate greed has become a bipartisan habit. On the left, Senator Elizabeth Warren blamed grocery prices on "good old-fashioned corporate price gouging" and popularized the word "greedflation." On the right, Treasury Secretary Scott Bessent has promised "to keep the retail gas stations honest" and to go after "bad actors" who don't cut prices when crude gets cheaper. The emphasis differs by party, but the story is the same: a price went up, so someone must have gotten greedier.

What is the response by economists?

In the first place, most economists appeal to the universality of self-interest. Sellers are not always saints. Self-interest is real. A man selling peaches wants every dollar he can get. So does the oil company. So does your landlord.

So do you.

Are sellers the only ones that are greedy? Aren't you a seller when you offer your wages? Is your greed the problem, or is it only theirs?

Those are theological, philosophical, and ethical problems, and most economists either shrug or simply argue that greed is merely a constant: we all have it. To the extent that greed is a problem, it is just as prevalent among sellers and buyers.

That is why the oldest question in this chapter (does supply set the price, or does demand ?) has no winner. Alfred Marshall settled it in his Principles of Economics (1890) with an image worth remembering forever:

"We might as reasonably dispute whether it is the upper or the under blade of a pair of scissors that cuts a piece of paper, as whether value is governed by utility or cost of production."

Neither blade cuts alone. Demand alone is a wish; supply alone is a warehouse. It is not the greed of buyers nor the greed of sellers, but the interaction of the two that sets prices. The price is the line where the two blades close — where the marginal buyer's willingness to pay meets the marginal seller's willingness to accept, and not a penny to either side.

Secondly, economists point out that there is good in self-interest as well as bad. The genius of markets is that they channel it to be useful. Self-interest is the fuel of the whole machine. The baker rises at four not to feed the city but to feed his family; the city gets bread as a by-product. Take the self-interest out and the shelves go empty.

In 1776 Adam Smith watched self-interested butchers, brewers, and bakers feed a city none of them meant to serve, and wrote that such a person, "intending only his own gain," is "led by an invisible hand to promote an end which was no part of his intention." That phrase is the whole idea of this chapter in brief. The baker rises at four to earn a living, and a city gets bread. Self-interest isn't evil greed, it is the fuel of an effective market.

But finally, economists will ask this: if greed causes prices to spike, is charity responsible when prices recede?

Warren called a costly turkey plain-old greed, yet the same growers were no less greedy in the years it was cheap. The gas stations Bessent wants to keep honest were just as greedy in the years gas was cheap. Every one of these explanations needs a greed that varies — that switches on when prices climb and off when they fall. There is no such thing. Greed is roughly a constant, and a constant cannot explain a change. The economists Brian Albrecht and Joshua Hendrickson put the point to the greed theorists directly: if greed drove prices, then "inflation has been low for 40 years because greed has been low."

No one believes that.

Nobody decides a price — no person, no office, no board. And also everybody decides it — every buyer who walks away, every seller who marks the sign down at four o'clock, every choice folded into the crossing. The price is the running tally of thousands of decisions no one coordinated. That is the answer to the question this chapter opened with, and it is also the foundation of the case for the political idea we call laissez-faire: leave the price alone.

The strongest version of that case is not the one you usually hear. It is not "markets are always fair." The economist Friedrich Hayek gave the real one in 1945, and it is an argument about knowledge . The information needed to set a good price — who needs the thing most, who can make it most cheaply, what is growing scarce and where — is never gathered in one place. It exists only in scattered fragments, a little in each of thousands of separate heads, most of it never written down and some of it not even conscious. No central authority can collect it, because much of it disappears the moment you try to report it upward. The price is the one device that uses it anyway. When a metal grows scarce, its price rises, and every user in the world quietly economizes — without being told why, without anyone issuing an order, most of them never learning what happened upstream. The rising number carried the news. Hayek called the system that does this, flatly, a marvel.

This book is a blend of normative and positive economics. We will now give a first, widely-agreed upon stance: let prices work. Before reaching for a law, a cap, or a planner, ask first whether the market can do the job itself — because when you override a price, you are not just changing a number, you are throwing away the knowledge that number was carrying. The burden of proof sits with whoever wants to override it, and in future chapters we will explore the kinds of problems that arise when the wisdom of prices is dismissed.

A high price is not, by itself, a market failing; usually it is a market working — telling the truth about scarcity that no one wanted to hear. Some high prices are simple reflections of reality. The price of immortality is infinite; the price of the meaningful life is priceless; the price of peace is without number. To use goods in a real economy, if oranges have a high price, it could be that an orange seller has found a way to capture profits; it could also be the result of an orange famine.

What does meet the burden is a specific, nameable defect: the price is set by a monopolist with no rival to discipline it; it leaves out a real cost paid by third parties who never agreed to it; one side is trading on knowledge the other can't see; or it cleanly prices out people a decent society has decided to protect. Each of those is a genuine reason, and each gets its own chapter later in this book. The point of the rule is not that intervention is forbidden. It is that intervention owes an argument, and the argument has to be better than a feeling.

This chapter has shown you that the market coordinates — that a free price knows things no planner could. It has not shown you that the market is kind, or fair, or that the price it lands on is one you can live with. Those are separate questions, and "the market decided it" is never their answer. Even the coordination itself runs on something the market did not build: the rules of property and honest exchange that someone had to establish and defend, and that can rot or fail. So the case for laissez-faire is a presumption , not a verdict — a starting point that the rest of this course spends its time rebutting, case by earned case. We begin overriding prices in the very next chapter, and we measure what the overriding costs a few chapters after that.

Let markets set prices — and reach for the planner only after asking whether the market could do the job itself. A price knows things no single person does. Before you correct it, be sure you know something it doesn't.

When a price stings, the economist's first instinct is not to hunt the villain who chose it — usually no one did — but to ask what moved: did the good get scarcer, the buyers hungrier, the production dearer? That question is where the next chapter begins, on a market someone has spent decades trying to shut down by force. When enforcement finally cuts its supply, what does the price do — and why do new sellers keep arriving to replace the ones it removes? We now hold the machine that answers it.