Macroeconomics,
the whole thing at once.

Macroeconomics is the study of the economy as a whole: not the price of one coffee, but the average level of all prices; not one person's job, but the fraction of everyone who has one; not one firm's output, but the total output of a country in a year. It asks four questions over and over: how much are we producing, are prices rising, is everyone who wants work finding it, and what can policy do about any of it.

The circular flow: one economy, three ways to count it

Picture the economy as a loop. Households own the labor and capital; they sell those to firms and get paid wages and profit. Firms use those inputs to make goods, which households then buy. Money flows one way around the loop, goods and services the other.

That loop hides a deep fact you will use constantly: every dollar of spending is someone's income, and pays for something produced. So you can measure the size of the economy three different ways and get the same number: add up all spending, add up all income, or add up all value produced. One economy, three counts, one answer. The next chapter makes that concrete.

The short run and the long run

Macro splits cleanly into two problems that use different tools. The long run asks why some countries are rich and others poor, and why living standards grow: this is the theory of growth (Part II). The short run asks why the economy booms and busts around that trend, and what policy can do to smooth it: this is the theory of fluctuations (Parts IV and V). Money and prices (Part III) bridge the two. Keep the distinction in mind: a lot of confusion in economics comes from applying a short-run tool to a long-run question.

Every year someone asks the same question: how much richer did we get? One way to answer it is brute force: add up the market value of every haircut, every car, and every loaf of bread sold in the country that year. That sum is GDP.

Gross Domestic Product (GDP) is the market value of all final goods and services produced within a country in a given period. It is the single most-watched number in macroeconomics, and the closest thing we have to a scoreboard for an entire economy.

The expenditure approach: C + I + G + NX

The most common way to count GDP is to add up everything that gets bought, sorted into four buckets:

• C onsumption: households buying goods and services (most of GDP in rich economies).
• I nvestment: firms buying capital (machines, buildings), plus new housing and inventory. Note: buying stocks is not investment here; only newly produced capital counts.
• G overnment purchases: spending on goods and services (roads, salaries, defense). Transfers like pensions are excluded, since they buy nothing produced.
• NX net exports: exports minus imports (X − M). We subtract imports because C, I, and G already include spending on foreign goods, which were not produced here.

Build one below. Set each component and watch the total recompute. Push imports up and watch GDP fall: those dollars were spent, but on output made somewhere else.

Two other ways to the same number

Because spending equals income equals output, two other methods must give the same total. The income approach adds up all wages, rent, interest, and profit earned in production. The value-added approach adds up, at each firm, the value of its output minus the cost of intermediate inputs it bought, which automatically strips out double-counting. Farmer sells wheat to miller, miller to baker, baker to you: only the value each one adds is counted, and it sums to the price of the final loaf.

GDP can rise for two very different reasons: we made more stuff, or the same stuff got more expensive. Only the first makes us better off, so we have to separate them. That is the difference between nominal and real.

Below is a two-good economy. Base-year prices are fixed (that is the ruler we measure with). Move current prices and quantities and watch the split: raising a price lifts nominal GDP but leaves real GDP untouched; raising a quantity lifts both.

The ratio of the two is a price index in its own right, the GDP deflator: 100 × nominal / real. It answers "how much of this year's dollar-growth was just prices?" A deflator of 120 means the basket of everything the economy produces costs 20% more than in the base year.

Inflation is a sustained rise in the general price level: not one good getting pricier, but the average of everything creeping up, which is the same as each dollar buying a little less. The headline gauge is the Consumer Price Index (CPI), which tracks the cost of a fixed basket of what a typical household buys.

Fix the basket, move the prices. The index below is the cost of the same basket relative to the base year. Notice that a big price jump in a small-weight item barely moves the index, while a modest rise in a heavily weighted item (housing) moves it a lot: the CPI is a weighted average.

The CPI and the deflator usually tell a similar story but differ in construction: the CPI holds the basket fixed and lets prices vary (so it misses that people substitute away from what got expensive), while the deflator uses the current mix of all output. Both are imperfect. Real-world inflation measurement wrestles with substitution, quality improvements, and new goods, which is why measured inflation is an estimate, not a fact.

The second headline gauge is the unemployment rate. It sounds simple but hides a careful definition: you are only "unemployed" if you are without a job and actively looking. Give up looking and you leave the labor force entirely, which, as you will see, can make the number lie.

Move the three groups below. The trap to watch: press Discouraged workers and some searchers stop looking. They are still jobless, but the measured unemployment rate falls, because they left the denominator. A falling rate is not always good news.

Not all unemployment is the same

• Frictional: the normal churn of people between jobs. Healthy and unavoidable.
• Structural: a mismatch between the skills or locations workers have and the jobs on offer. Automation and trade create it.
• Cyclical: the extra unemployment that appears in a recession when demand falls. This is the part policy tries to fight.

Even a healthy economy has some unemployment (frictional plus structural). That baseline is the natural rate, and the economy is at "full employment" when cyclical unemployment is zero, not when the rate is literally zero.

Output and unemployment are two windows on the same thing: how hard the economy is working. When firms produce less, they need fewer workers, so the two move together in a stable empirical way. Arthur Okun measured it: each percentage point of unemployment above the natural rate goes with output roughly 2% below potential.

Here u* is the natural rate. Slide unemployment away from it and watch the output gap open up. It is not an exact law, more a rule of thumb, but a reliable one: it lets you translate a labor-market number into a production number and back.

Picture one worker on a farm with bare hands: output is small. Hand that worker a tractor and output jumps. A second tractor helps too, but by less than the first did. By the tenth tractor, most of them sit idling in the barn and output barely moves at all. Each extra unit of equipment adds less than the one before it: that pattern is diminishing returns, and it shapes almost everything in this chapter.

Growth is the most important thing in macroeconomics: a country's living standard a century from now depends almost entirely on it. To study growth we need a recipe that turns inputs into output. That recipe is the production function.

Since we care about living standards, divide by the number of workers and work per person. Output per worker y = Y/L depends on capital per worker k = K/L:

The crucial feature, as the tractor story already showed, is diminishing returns to capital. Below, raise capital per worker and watch output rise but bend over: each extra unit of capital adds less than the last. The slope of the curve is the marginal product of capital, and it falls the whole way.

Two ways to grow, then: accumulate capital (move rightward along the curve) or raise productivity A (lift the whole curve up). Diminishing returns means the first runs out of steam; the second never does. That tension is the engine of the next chapter.

This is the central model of long-run macro, and the centerpiece of this guide. Robert Solow asked a simple question: if a country saves and invests a fixed share of its output, and capital wears out at a fixed rate, where does it end up? The answer is beautiful, and it explains one of the deepest facts in economics: capital accumulation alone cannot grow you forever.

The mechanism, in one line

Each period, savings add new capital and depreciation eats it away:

• s·f(k) is investment: the share s of output that gets saved and turned into new capital. Because f is concave, this curve bends over.
• δ·k is depreciation: a fixed fraction of existing capital wears out each period. This is a straight line.

When investment exceeds depreciation, capital grows. When depreciation exceeds investment, capital shrinks. They cross at exactly one point, the steady state k*, where new capital exactly replaces what wore out and k stops changing. Setting s·A·k*α = δ·k* and solving:

Now watch it happen. The left chart is the phase diagram: investment (teal) against depreciation (plum), crossing at the amber steady state. Press Step to run the recurrence one period at a time, or Run to animate. The right chart traces kt over time. Then hit Start low and Start high: from either end, the economy converges to the same k*. That convergence is the model's signature prediction.

What the sliders tell you

• Save more (s up): the investment curve lifts, k* moves right. Richer in the long run, but still a fixed destination.
• Depreciate faster (δ up): the depreciation line steepens, k* moves left. Capital-hungry economies stay poorer.
• Convergence: poor countries (low k) grow faster than rich ones, because the return to capital is higher when capital is scarce. This is why the model predicts catch-up.
• The productivity residual: sustained growth in living standards must come from A. Everything else runs into diminishing returns.

The Solow model hands us a stark conclusion: over the long haul, living standards are a story about productivity growth, and productivity growth is a story about compounding. Small differences in the annual rate become enormous differences over a lifetime, because growth multiplies on itself.

Set two growth rates below and watch them diverge. A single extra percentage point looks trivial year to year, but run it out over decades and the faster economy pulls away dramatically. This is why economists obsess over productivity: it is the only lever that compounds.

Where does productivity growth come from? Research and new ideas, better education, deeper capital markets, institutions that protect property and enforce contracts, and openness to trade and competition. Growth theory beyond Solow (endogenous growth) tries to explain A from inside the model rather than treating it as a gift. But the headline is settled: productivity is destiny.

Money is one of humanity's great inventions, and it is easy to take for granted. Strip it away and every trade needs a double coincidence of wants: to swap your bread for shoes, you must find a shoemaker who happens to want bread right now. Money dissolves that problem. It is defined not by what it is (shells, gold, paper, digits) but by what it does.

Modern money is fiat money: it has value because a government declares it legal tender and, crucially, because everyone expects everyone else to accept it. That shared expectation is the whole game. Most money is not cash at all but balances in bank accounts, created when banks lend. Central banks sit on top, controlling the quantity of base money and the interest rate at which banks fund themselves.

Why the quantity of money matters

Here is the pivot into the next two chapters. Money is a claim on goods, not goods themselves. If the amount of money doubles but the pile of goods does not, each unit of money must chase more dollars per good: prices rise. Money is neutral in the long run, meaning it changes the price level, not the real quantity of stuff. Understanding exactly how much prices rise is the job of the quantity theory, next.

Picture a small island with a fixed pile of coconuts already harvested for the year: no more will grow. If $1,000 of money changes hands 10 times over the year, that is $10,000 of total spending chasing that one fixed pile, and the ratio of the two pins the price of a coconut. Now suppose the island's ruler prints another $1,000 with not one extra coconut grown. The same coconuts now face twice the dollars, and the price roughly doubles. That is the whole idea below, written as an equation.

The quantity theory pins down the link between money and prices with a single accounting identity, the equation of exchange:

• M: the money supply.
• V: velocity, how many times each unit of money is spent per year.
• P: the price level.
• Y: real output (real GDP).

The left side is total spending (money times how often it turns over). The right side is the value of everything sold (prices times quantities). They are equal by definition. It becomes a theory once you add two assumptions: velocity V is roughly stable, and real output Y is set by the real economy (Part II), not by money. Then P = M·V / Y: the price level moves one-for-one with the money supply.

Test it below. With V and Y held fixed, double the money supply and watch the price level double: pure inflation, no extra output. Then raise real output Y and watch prices fall for a given M: more goods to spread the money across.

Inflation quietly rewrites every interest rate. If your savings account pays 5% but prices rose 3%, your purchasing power only grew about 2%. The 5% is the nominal rate you see; the 2% is the real rate that actually matters. The link is the Fisher equation:

Rearranged, the real return is r ≈ i − π. Move the two sliders below. Hold the nominal rate fixed and crank up inflation: the real return sinks, and once inflation passes the nominal rate it goes negative, meaning a saver quietly loses purchasing power despite earning "interest."

When it runs away: hyperinflation

Inflation is usually a slow leak. Hyperinflation is a burst pipe: prices rising so fast that money becomes hot to hold, so people spend it instantly, which raises velocity, which raises prices further, a self-feeding spiral. It happens when a government finances spending by printing money faster and faster (Weimar Germany 1923, Zimbabwe 2008, Venezuela 2018). The arithmetic is brutal: a "modest-sounding" 20% per month compounds to roughly (1.20)12 − 1 ≈ 790% per year. The cure is always the same, and always hard: stop printing.

Growth explains the long-run trend. But real economies do not glide along it; they boom and bust around it. To model that, we scale supply and demand up to the whole economy and put total output Y on the horizontal axis and the price level P on the vertical. These are the aggregate curves.

Aggregate demand slopes down

Imagine every price in the economy fell 10% overnight. Your savings would suddenly buy more (the wealth effect), borrowing would get easier as rates ease (the interest-rate effect), and your exports would look like a bargain to foreigners (the exchange-rate effect). All three push spending up when the price level falls, which is why aggregate demand (AD), the total output demanded at each price level, slopes down, for different reasons than a single market's demand curve. AD shifts when C, I, G, or NX change for reasons other than the price level: a confidence collapse, a tax cut, a spending program, a central-bank rate move.

Aggregate supply has two time frames

This is where the short run and long run split apart, and it is the single most important idea in short-run macro:

• Long-run aggregate supply (LRAS) is a vertical line at potential output Y*. In the long run, output is set by the real economy of Part II (capital, labor, productivity), not by prices. Doubling all prices doubles all wages too and changes nothing real.
• Short-run aggregate supply (SRAS) slopes up. Think of a restaurant that printed today's menu this morning: if ingredient costs jump this afternoon, the prices on the menu stay put until the next printing. In the short run, some prices and wages are sticky the same way (fixed by contracts, menus, habit), so a higher price level lets firms sell at better margins and they produce more. Output can temporarily run above or below potential.

Picture two very different news days. On one, a viral new product craze sweeps the country and everyone rushes out to spend: total demand lifts nationwide. On the other, an oil pipeline explodes and fuel costs spike overnight: a key input gets suddenly expensive everywhere at once. Both jolt the economy, but as you will see, in opposite ways. That difference is the whole chapter.

Put the curves together. The short-run equilibrium sits where AD crosses SRAS: that fixes both output and the price level. When it lands to the right of LRAS, the economy is overheating (an inflationary boom); to the left, it is in recession with an output gap. The interesting part is what happens when a shock hits.

Two kinds of shock, two very different signatures. Fire a demand shock, like that viral craze, and AD shifts: output and prices move the same direction (a boom raises both; a slump lowers both). Fire a supply shock, like the oil-pipeline explosion, and SRAS shifts: output and prices move in opposite directions. An adverse supply shock is the nightmare case, stagflation: falling output and rising prices at once, with no easy policy fix.

How the long run reasserts itself

Suppose a demand boom pushes output above potential. The economy is running hot; labor and materials get scarce; wages and prices start to rise. That rise shifts SRAS up and left over time, sliding the economy back along AD until it returns to Y* at a higher price level. The boom was temporary; the inflation stuck. This self-correction is slow and painful, which is the entire motivation for stabilization policy in Part V.

The AD-AS model has a twin that plots the same fluctuations in the variables people actually feel: inflation and unemployment. In the short run they trade off. When demand runs hot, unemployment falls below its natural rate and inflation rises; when demand is weak, unemployment rises and inflation falls. That downward-sloping tradeoff is the Phillips curve.

Inflation equals expected inflation πe minus a term for how far unemployment sits above its natural rate u*. Move the policy slider below to ride along the short-run curve: push unemployment down and inflation climbs.

But there is a catch that broke a generation of policymakers. The tradeoff is only temporary. Hold unemployment below u* and people come to expect the higher inflation; expectations πe ratchet up; the whole curve shifts upward. Press Expectations adjust and watch: unemployment slides back to u*, but at permanently higher inflation. In the long run the Phillips curve is vertical at u*. There is no permanent tradeoff, only a temporary one you pay for later.

If the economy self-corrects only slowly and painfully, maybe policy can speed it up. The first lever is monetary policy: the central bank moving the short-term interest rate to steer aggregate demand. Cut rates and borrowing gets cheaper, investment and spending rise, AD shifts right. Raise rates and demand cools. The bank leans against the cycle: stimulate in slumps, restrain in booms.

The critical detail is the Taylor principle: the coefficient on inflation must be greater than one, so that when inflation rises, the bank raises the nominal rate by more, pushing the real rate up and actually cooling things. Below, set inflation and the output gap and read off the prescribed rate. Then press Run policy: the feedback loop follows the rule period by period and drives inflation back to its 2% target.

Central banks act through more than the overnight rate: reserve policy, asset purchases (quantitative easing) when rates hit zero, and above all communication, since expectations of future rates drive spending today. But the Taylor rule captures the core reflex: watch inflation and the output gap, and lean.

The second lever is fiscal policy: the government changing its own spending G or taxes T to move aggregate demand directly. Its signature feature is the multiplier. A dollar of government spending does not stop at one dollar of demand: the recipient spends part of it, that recipient spends part of that, and the ripples add up to more than the original dollar.

The total is a geometric series: ΔG + MPC·ΔG + MPC2·ΔG + ..., which sums to ΔG/(1−MPC). Press Next round below to watch the rounds of spending stack up and converge on that total. Raise the MPC and the multiplier grows because less leaks out of the spending stream at each round.

Crowding out and the catch

The multiplier is not the whole story. When the government borrows to spend, it competes for savings and pushes interest rates up, which discourages private investment. That offset is crowding out, and it shrinks the effective multiplier, sometimes a lot. Fiscal policy also acts slowly (budgets take time) and adds to public debt. It is most powerful exactly when monetary policy is stuck, which is the subject of the next chapter.

Here is the chain reaction before the labels. The central bank cuts interest rates, loans get cheaper, and firms and households build more houses and buy more equipment: that response of output to falling rates is the IS curve moving. As output rises, people need more cash on hand for everyday spending, and if the money supply does not grow to match, rates get pushed back up: that reaction is the LM curve.

The IS-LM model puts monetary and fiscal policy in one picture, in (output, interest rate) space. Two curves, two markets:

• The IS curve is the goods market. It slopes down: a higher interest rate depresses investment, which lowers output. Fiscal expansion (more G) shifts IS right.
• The LM curve is the money market. It slopes up: higher output raises money demand, which pushes the interest rate up for a fixed money supply. Monetary expansion (more M) shifts LM right, lowering rates.

Where they cross fixes output and the interest rate together. Below, use fiscal and monetary policy and read the results. Note the tell-tale of crowding out: fiscal expansion raises output and the interest rate, and that higher rate is exactly the private investment being crowded out.

Now assemble the pieces into the thing you see on the news: the business cycle, the economy's repeated swing above and below its long-run trend. Growth (Part II) sets the rising trend line; fluctuations (Part IV) are the wobble around it. The cycle has four phases that repeat, though never on a fixed schedule.

Press Run and watch output (blue) oscillate around potential (teal, dashed) while unemployment (plum) moves the opposite way, exactly as Okun's law demands. The label tracks the current phase. This single animation ties together output, potential, unemployment, and the countercyclical relationship between them.

Step back and the whole subject fits on one page. You now hold every piece.

We measure the economy with GDP, splitting nominal from real to see true output, and track its health with inflation (CPI, deflator) and unemployment, linked by Okun's law. Over the long run, the production function and the Solow model explain the level of prosperity: capital accumulates to a steady state, and only productivity growth raises living standards forever. Money sits underneath prices: the quantity theory ties the money supply to the price level, and the Fisher equation separates real from nominal interest.

In the short run, aggregate demand and supply explain booms, busts, and stagflation, with the Phillips curve translating them into the inflation-unemployment tradeoff that fades once expectations adjust. Policy leans against the cycle: monetary policy through the interest rate (the Taylor rule), fiscal policy through spending and the multiplier, and IS-LM showing how they interact, including the liquidity trap where only fiscal policy bites. All of it plays out as the business cycle around a growing trend.

You have the core of an intermediate macro course. To go deeper:

• Microfoundations: how individual optimization builds up to these aggregates. The companion microeconomics guide is the natural pairing.
• Growth theory: endogenous growth (Romer, Aghion-Howitt), where productivity A is explained from inside the model rather than assumed.
• New Keynesian macro: the modern workhorse, with explicit sticky-price microfoundations and a forward-looking Phillips curve.
• Open-economy macro: exchange rates, capital flows, and the Mundell-Fleming extension of IS-LM.
• Data: pull real series from FRED (GDP, CPI, unemployment, the fed funds rate) and re-derive Okun's law and the Phillips curve yourself. Nothing cements a model like fitting it to real numbers.

The models in this guide are deliberate simplifications, chosen because they carry the most intuition per equation. Real research adds frictions, expectations, and heterogeneity, but the skeleton is the same one you just built by hand.