Economic indicators and the business cycle

1. GROSS DOMESTIC PRODUCT: THE DEFINITION AND ITS BOUNDARY

GDP is the market value of all final goods and services produced within a country’s borders during a given period. Every word in that sentence is doing work. The first four gates below come directly from it. The fifth is a measurement boundary: national accounts include marketed output, government and nonprofit services valued by their production cost, and specified imputations such as the housing service homeowners provide to themselves. They exclude many household services for which the accounts do not construct a value. A literal sale is common; it is not a universal requirement.

Tests 1–4 sort the ordinary AP cases. Gate 5 is different in kind: an activity can be genuinely productive and still sit outside the chosen boundary. Unpaid family childcare is the case in Problem 1(c). But do not turn that example into “no sale, no GDP.” Public-school instruction and other government output are included even without a market sale, and the accounts impute values for a small set of services where no cash changes hands. The boundary is a limitation, not a claim that omitted work has no value.

The trap inside test 1. When ownership of an existing asset moves, the asset itself does not count — but any service someone performed to arrange the transfer does, because that service was produced this year. The used house does not count; the real estate agent’s fee does. The shares do not count; the broker’s commission does. This catches almost everyone the first time. One national-accounts detail matters in Problem 1: a household’s securities-broker service is consumption, while a residential broker’s ownership-transfer service is recorded with residential fixed investment.

2. THE CIRCULAR FLOW: WHY SPENDING EQUALS INCOME

GDP can be measured three different ways — by adding up what was spent, by adding up what was earned, or by adding up the value added at each stage of production — and all three land on the same number. That is not a coincidence to be memorized. It follows from the shape of the economy.

The two markets are where the exchanges happen. In the product market, firms sell finished goods and services to households. In the factor market, households sell the resources they own — their labor, land, and capital — to firms.

The outer loop is money; the inner loop, running the other way, is the real stuff the money pays for. The loop is closed, so the same flow measured at any point gives the same total: one person’s spending is another person’s income, which is why the expenditure approach and the income approach cannot disagree.

The third route cuts the same identity a third way. A firm’s value added is what its own work contributed — what it sold its output for, minus what it paid other firms for the inputs it built into that output:

Summing value added across every firm counts each firm’s contribution exactly once, and lands on the same total as counting final sales. Problem 2 proves that arithmetically rather than asserting it.

3. THE EXPENDITURE IDENTITY

Write it out before you substitute, every time. Four buckets, and the whole difficulty is sorting items into them.

• $C$ — consumption. Household spending on goods and services: food, rent, haircuts, a car. Not new houses — those are investment. A securities broker’s service purchased by a household is a current financial service and belongs here.

• $I$ — gross private domestic investment. Firms buying capital: machinery, tools, factories, software. Plus two items students routinely misfile:

  • New residential construction — a newly built house counts as $I$, not $C$, however ordinary the buyer. National accounts also place brokers’ commissions and other ownership-transfer costs on residential property in residential fixed investment, including when the underlying house is not new.
  • The change in inventories — output produced this year but not yet sold. The firm is treated as having bought its own unsold output. This keeps GDP a measure of production rather than of sales, and it is why the change can be negative: a year spent selling down the warehouse subtracts.

  “Investment” here never means buying financial assets. Shares and bonds are not in $I$; they are not in GDP at all.

• $G$ — government purchases of goods and services at every level: roads, salaries of public employees, defense equipment. Not transfer payments — pensions, unemployment benefits, subsidies. Money moves; nothing is produced. (When the recipient later spends it, that spending shows up in $C$, where it belongs.)
• $X-M$ — net exports. Exports minus imports.

Why imports are subtracted, since this is the component students get backwards. It is not a claim that buying from abroad shrinks the economy. It is a correction. The $C$, $I$, and $G$ figures are collected as total spending, and that total already includes spending on imported goods — which were not produced here. Subtracting $M$ removes what was double-counted in, so that what is left is domestic production. Problem 1(b) follows a single imported purchase all the way through.

4. THE CPI, AND THE FOUR WORDS FOR WHAT PRICES DO

The consumer price index (CPI) prices a fixed market basket of consumer goods and services in one year relative to a chosen base year. It is intended to measure the income change a consumer would need to maintain the same standard of living as the price level changes.

The base year’s index is 100 by construction — if you compute it and get anything else, you made an arithmetic error, not a discovery.

Reading an index aloud. $\mathrm{\text{CPI}}=114.71$ means the basket costs 14.71% more than in the base year. It does not mean prices are 114.71% higher, and it does not mean anything at all about one specific product. Practice saying the sentence; free-response questions ask for it.

The inflation rate is the percentage change in the index, and its denominator is the earlier index — not 100, unless the earlier year happens to be the base year:

Four words describe four different things, and three of them get confused:

• Inflation — the price level is rising. The index goes up. The rate is positive.
• Disinflation — the price level is still rising, but more slowly than before. The index still goes up; the rate is still positive, just smaller. The overall basket got more expensive; individual prices inside it may still fall.
• Deflation — the price level is falling. The index goes down. The rate is negative. This is a different event from disinflation and it is not a milder version of it.
• Real — any variable with the price effect removed. See toolbox 5.

Substitution bias. The basket is held fixed on purpose — that is what makes the comparison honest about prices rather than about changing tastes. But it means the index assumes households keep buying the same quantities even when one item’s price runs away. In reality they buy less of it and more of the cheaper substitute, so their true cost of living rises by less than the fixed basket implies. When that substitution occurs, a fixed-basket CPI overstates the increase needed to maintain the same standard of living. Note the direction.

5. REAL VS. NOMINAL — THE SPINE OF THIS UNIT

Nominal means measured at the prices actually prevailing at the time. Real means the price effect has been removed. This packet first does that with the AP classroom’s fixed-base method: value each year’s output at one base year’s prices. Converting nominal to real is called deflating.

The mechanism, in one sentence. A nominal figure is contaminated by two things moving at once — how much was produced and what it cost. Deflating divides the price movement out, so what is left moved only because production moved.

Two forms, same operation. Use whichever matches the numbers you were handed:

The second form is the useful one when you are asked to put two salaries, or two prices, “in the same year’s dollars.” Notice that it works in either direction, and that both directions give the same percentage change — Problem 5 does it both ways to prove that.

Deflator or CPI? They answer different questions and the exam distinguishes them.

If a question is about households’ purchasing power, reach for the CPI. If it is about converting the nation’s output to real terms, reach for the deflator. The imported phone is the tell: it raises the CPI and never touches the deflator.

In the base year, real equals nominal. By construction: you are valuing base-year output at base-year prices. The index is 100 and dividing by 1 changes nothing. If your real and nominal figures agree in the base year, that is a successful check, not a mistake.

The fixed-base limitation. Holding one year’s prices still is one way to isolate quantities, not the only way national statistical agencies measure real growth. Relative prices change, and households and firms tend to shift toward goods that became relatively cheaper. An old base year’s prices can then put too much weight on the goods whose quantities expanded and too little on the goods whose quantities contracted, so a fixed-base calculation can overstate real GDP growth. Agencies use methods that update the weights over time. The AP task is to know that limitation and its direction; Problem 9 makes two fixed bases disagree without asking you to construct an agency’s chain-type measure.

Problem 4 computes every figure on that diagram.

Extension: why growth rates do not quite add. Nominal growth is not real growth plus inflation. Exactly,

so $g_{\text{nominal}}=g_{\text{real}}+g_{\text{price}}+g_{\text{real}}{g}_{\text{price}}$. Subtracting is a good approximation when both rates are small, and it drifts as they grow — the leftover is that cross term. Problem 4 measures the drift at 2.85 percentage points and Problem 5 measures it at 0.79. Use the subtraction for a sanity check; use the division when the question wants a number.

6. WHAT INFLATION COSTS, AND WHO PAYS IT

This unit focuses on the AP-required redistribution channel: inflation is especially consequential when it is unexpected and a contract fixes a nominal amount years in advance. A thought experiment in which every price and every income changes together removes that redistribution; it does not prove surprise is inflation’s only possible cost. Even anticipated inflation can consume time and resources through price changes, cash management, tax interactions, and noisier dollar comparisons. The mechanism below is the one this unit asks you to trace, not an exhaustive theory of inflation.

The mechanism. A loan, a wage agreement, a pension: each fixes a number of dollars to be paid in the future. Inflation does not alter that number. It alters what the number buys. So when inflation comes in higher than the two parties expected when they signed:

• The party receiving the fixed payment gets dollars that buy less than planned — lenders lose.
• The party making the fixed payment discharges the same obligation with cheaper dollars — borrowers gain.

Unexpectedly low inflation, or deflation, runs the transfer the other way: from borrowers to lenders. That is why deflation is particularly hard on anyone carrying debt.

Unexpected inflation redistributes purchasing power from lenders to borrowers. A complete explanation names both parties and states why the fixed nominal repayment has lower purchasing power than expected.

Why “unexpected” is the load-bearing word. The transfer turns on inflation differing from a fixed nominal contract’s forecast, not on inflation being high. Expected inflation can be priced into the nominal rate; the forecast error is what moves purchasing power. Problem 6 measures both directions in dollars.

The same logic reaches beyond loans. Anyone holding a claim fixed in nominal terms — a pension with no cost-of-living adjustment, a multi-year fixed wage, cash itself — loses when inflation surprises upward. Anyone owing a fixed nominal amount gains. Contracts that are indexed to the CPI are insulated, which is precisely why indexation clauses exist.

7. THE LABOR FORCE: WHO IS IN THE DENOMINATOR

Every unemployment question is decided by which box a person is in. Three statuses, and the boundaries between them are where the points live.

• Employed — did any work for pay or profit in the survey week; worked at least 15 unpaid hours in a family business; or had a job but was temporarily absent from it. Part-time counts as employed. Someone working ten hours a week who wants forty is employed, fully, with no asterisk in the official rate.
• Unemployed — no job, available to work, and actively searched during the previous four weeks. There is one official exception: someone on temporary layoff who expects recall is counted as unemployed without an active search.
• Not in the labor force — everyone else in the adult population: retirees, full-time students not seeking work, full-time carers, and discouraged workers — people who want a job but have stopped searching because they believe none is available. A discouraged worker fails the search condition and is therefore not counted as unemployed.

The boundary that matters is between “Unemployed” and “Not in the labor force,” because a person crossing it changes the numerator and the denominator at the same time, in the same direction. That is what makes the unemployment rate capable of improving while the labor market worsens. Problem 7 works the full contrast.

8. THREE KINDS OF UNEMPLOYMENT, AND THE NATURAL RATE

The categories are defined by cause, not by how long the person has been out of work and not by how they feel about it. Ask what would have to change for this person to be working:

• Frictional — between jobs, searching, relocating, or newly entering the workforce. A healthy economy has this permanently, and some of it is productive: a good match found in six weeks beats a bad one taken in one.
• Structural — a lasting mismatch between the skills or locations workers have and the ones employers need: automation retiring a job category, an industry leaving a region, a credential the worker does not hold. Retraining or moving changes the outcome; waiting does not.
• Cyclical — caused by a downturn in the economy as a whole. It rises in recessions and falls in expansions, and it is the only one of the three that a recovery fixes on its own.

“Full employment” does not mean zero unemployment. It means cyclical unemployment is zero and the actual rate has settled at the natural rate. Frictional and structural unemployment remain, because people will always be between jobs and industries will always be changing. A country reporting 0% unemployment would be reporting a broken statistic, not a triumph.

The natural rate itself drifts over decades as the workforce changes — its age mix, how easily job-seekers and employers find each other, how fast technology retires occupations. It is not a constant of nature and the exam does not expect you to treat it as one.

9. THE BUSINESS CYCLE AND THE OUTPUT GAP

Business cycles are fluctuations in aggregate output and employment caused by changes in aggregate demand and/or aggregate supply. Real output does not grow smoothly; it moves around a long-run path, and the vocabulary for those movements is precise.

• Phases: expansion and recession. Expansion — output and employment rising. Recession — output and employment falling.
• Turning points: peak and trough. The peak is the last period of an expansion, where output stops rising and starts to fall. The trough is the bottom, where a recession ends and expansion begins.
• Potential output (also called full-employment output) is what the economy can produce when unemployment sits at the natural rate. It often follows a rising long-run trend, including through a demand-driven recession, but it is not guaranteed to rise every period. Changes in labor, capital, productivity, or supply conditions can change its level or growth rate.
• Output gap $=$ actual output $-$ potential output. Negative (a recessionary gap): unemployment is above its natural rate and cyclical unemployment is positive. Positive (an inflationary gap): unemployment is below its natural rate and the economy is running beyond its sustainable rate.

One thing the picture makes obvious that a sentence does not: the second peak is higher than the first — a peak is a turning point, not a record. Problem 8 reads all of this off a data table.