Part 4 · Employment and business conditions · Lesson 22
Consumption and business investment: the engines of the economy
About two-thirds of the US economy is consumption, but the biggest swings come from business capital expenditure. Let's build up how to read the dashboards of these two engines.
QToday's question
The news says 'consumer sentiment at a record low', yet retail sales that same month rose more than 6%. Are people gloomy but spending more anyway?
In May 2026, the University of Michigan consumer sentiment index in the US was 44.8, the lowest since records began, yet retail sales were up 6.7% from a year earlier. In this lesson, we'll build up from what the engines of the economy, consumption and investment, are made of, to why the dashboard numbers disagree, to how all this connects to stocks.
MAPLearning map for this lesson
We build it up in three steps. At the end of each step, a 'Quick check' lets you test your understanding before moving on.
In Lesson 4, GDP, you learned that GDP = consumption + investment + government spending + net exports, and that consumption made up about 68% of the US. We'll also bring along the cyclical and defensive stocks from Lesson 5, the business cycle, the nominal vs. real distinction from Lesson 7, inflation, and the earnings dial from Lesson 21, when bad news becomes good news.
- STEP 1 · Basic conceptsThe two engines of the economy: consumption and investment
You'll be able to explain the three baskets of consumption (durable goods, nondurable goods, services) and the difference between and .
- STEP 2 · Understanding the principleThree dashboards and the accelerator principle
You'll be able to read , and capital expenditure indicators, and explain why investment swings more than consumption.
- STEP 3 · Real-world applicationReading US consumption, 2019–2026
You'll be able to interpret periods when sentiment and actual spending diverged, and connect consumption and investment signals to sectors.
STEP 1The two engines of the economy: consumption and investment
In Lesson 21, we said the way weak employment turns the 'earnings dial' is ultimately through lower consumption. In this lesson, we look directly at that consumption and at investment, the money businesses spend.
The size of the engines within GDP
In Lesson 4, you learned that GDP is the sum of consumption, investment, government spending and net exports. Of these, the two engines that most shape the course of the economy are household and business investment. This is because government spending is largely set by policy, and net exports depend on overseas conditions.
The size of the two engines differs from country to country. In 2025, consumption was 67.6% of GDP in the US and 48.1% in Korea. The US economy is essentially a 'consumption economy', while in Korea investment and exports matter as much as consumption. If US consumption wobbles, US company revenue is hit directly, and Korean companies feel it through export orders.
GDP expenditure breakdown, 2025
Source: US BEA national accounts (via FRED), Bank of Korea ECOS GDP by expenditure (200Y109), 2025 nominal basis. Same data as Lesson 4. Because US net exports are −3.0% (imports exceed exports), the bar shows 0 and the figure is given in the legend.
Consumption is divided into three baskets
Consumption can be further divided into three baskets: used for three years or more, such as cars, appliances and furniture; nondurable goods used up quickly, such as food, clothing and gasoline; and services such as healthcare, housing, dining out and travel. In US personal consumption expenditures in August 2026, durable goods were 11.0%, nondurable goods 20.2% and services 68.8%. When we say 'consumption' we think of shopping, but in reality services account for more than two-thirds.
Each basket reacts to the economy at a different speed. Durable goods are things you can 'buy later', so they are the first to be postponed when the economy looks shaky. They are often bought on installment, so they are also sensitive to interest rates. Groceries and daily necessities, on the other hand, can't be cut much even in a bad economy. This difference leads to the sector split in the stock market between and .
An easy analogy · Your household budget
If you heard that your salary might shrink, what would you cut first in your household budget? Rice, toilet paper and the electricity bill stay the same. Instead you put off a new TV, an overseas trip and weekend dining out. When millions of households do this at the same time, TV makers, airlines and restaurants see revenue drop sharply, while revenue at household-necessities makers barely changes. That's the difference between consumer discretionary and consumer staples.
See it in a diagram: consumption baskets and sectors
Tap a basket. It shows what spending it contains, how sensitive it is to the economy and interest rates, and which sector of the stock market it connects to.
Composition of US personal consumption expenditures (Aug 2026, nominal)
Source: US BEA personal consumption expenditures, FRED PCEDG, PCEND, PCES (Aug 2026, seasonally adjusted annual rate, calculated by Robin Invest). The Jan 2019 shares were 10.4%, 20.8% and 68.8%, so the broad picture is similar. Sector links are general tendencies.
Quick checkWhen the economy starts to cool, which is more likely to see its revenue swing sharply, a car company or a toothpaste company? Why?
STEP 2Three dashboards and the accelerator principle
Now that you know how the engines are built, let's read the dashboards. We'll build up two numbers that show consumption, one number that shows investment, and the principle behind why investment swings so much.
Dashboard 1. Retail sales: money actually spent
is a figure that adds up the amounts sold at stores, online, gas stations, restaurants and so on. In the US, the Census Bureau of the Commerce Department releases the previous month's figure in the middle of each month, and in Korea, Statistics Korea releases the retail sales index at the end of each month in its Industrial Activity Trends report. Because it is money that actually left wallets, it shows the flow of consumption fastest.
One caution is that retail sales are mostly nominal amounts. As you learned in Lesson 7, if prices rise 5%, sales amounts rise 5% even if you buy the same goods. So retail sales growth should be read alongside the inflation rate. Also, retail sales are goods-focused, so services, which make up two-thirds of consumption, are only partly included (dining out).
Dashboard 2. Consumer sentiment: the will to open the wallet
is the result of surveys that ask households, "Do you expect your situation to improve?" and "Is now a good time to buy big-ticket items?" In the US, the University of Michigan consumer sentiment index and the Conference Board consumer confidence index are the best known, and in Korea the Bank of Korea publishes the Consumer Composite Sentiment Index (CCSI) every month. A Korean reading above 100 means more optimistic than the long-term average.
Because sentiment indexes ask about 'the future', they sometimes move before actual spending. But a survey measures mood, so in times of high inflation or noisy political news it can diverge widely from actual spending. People say "life is hard" and yet keep spending if wages are rising and jobs are available. That's why sentiment indexes are not read alone but checked together with retail sales and employment.
Dashboard 3. Capital expenditure: business confidence
is the money companies spend on factories, machinery, equipment and software. It can be confirmed through the quarterly GDP release, but that comes late, so the market watches , which are US durable goods orders excluding aircraft and defense, as a leading signal for capital expenditure. In Korea, Statistics Korea's Industrial Activity Trends report publishes the facilities investment index every month.
Why does investment swing more?
Put consumption and investment side by side and a clear difference shows. Consumption is big but moves slowly, while investment is small but swings widely. The explains why. Companies expand their equipment in line with 'changes in demand', not the 'level of demand'.
Hypothetical example: a factory that produces with 10 machines and replaces 1 each year. Even if demand moves by just 10%, machine orders double or get cut in half.
The real numbers show the same thing. From 2008 to 2025, annual growth in US real consumption ranged from −2.5% to +8.8%, while growth in real capital expenditure ranged from −14.5% to +9.5%. The standard deviation, which measures how erratic the figures are year to year, was 2.2 percentage points for consumption and 5.5 percentage points for capital expenditure.
US real consumption vs. real capital expenditure, annual growth (2008–2025)
Tap a bar.
Source: US BEA real personal consumption expenditures FRED PCECC96, real private nonresidential fixed investment FRED PNFIC1 (year-over-year change of quarterly sums, calculated by Robin Invest). Checked 2026-10-05.
Common misconceptions
- "If retail sales rose, the economy must be doing well?"
- Not necessarily. If prices rise 8% and sales rise 8%, the quantity of goods actually bought is unchanged. You have to separate nominal from real.
- "If consumer sentiment is bad, consumption will soon fall?"
- Sentiment is just a signal. As long as jobs and income hold up, a period of 'gloomy but still spending' can last a long time. We'll look at real examples in STEP 3.
- "Investment is small in GDP, so it matters less?"
- It is small in size, but it swings widely, so investment often leads the change at the moment the economy turns down. In 2009, US real capital expenditure fell 14.5%, while real consumption fell 1.3%.
Quick checkIn a period when prices are rising 6% a year, retail sales rose 5% from a year earlier. Did the quantity of goods people actually bought rise or fall?
STEP 3Reading US consumption, 2019–2026
Now that you've built up how to read the dashboards, let's go to real numbers. Today's question, the puzzle of 'gloomy but spending more', gets solved here too.
The two dashboards side by side
The top shows retail sales growth versus a year earlier, and the bottom shows the University of Michigan consumer sentiment index. Tap the graph and both numbers for that month appear together.
US retail sales growth (year over year, nominal) and consumer sentiment index
Tap the graph to see that month's numbers.
Source: US Census Bureau retail sales FRED RSAFS (seasonally adjusted, year-over-year calculated by Robin Invest), University of Michigan consumer sentiment index FRED UMCSENT (final values). Jan 2019 – Aug 2026, checked 2026-10-05. The +51.5% in April 2021 largely reflects the base effect from the sharp drop in April 2020.
Reading three scenes
2020–2021: the plunge and the base effect. In April 2020, retail sales fell 19.5% from a year earlier. A year later, in April 2021, they rose 51.5%, the result of the base effect you learned in Lesson 8 combined with government relief payments. The more extreme the number, the more important it is to check first what it is being compared against.
2022: the first divergence between sentiment and spending. In June 2022, the consumer sentiment index fell to 50.0. Yet retail sales that same month rose 9.2%. The secret is inflation. US CPI rose 9.1% that month (Lesson 8), so the quantity of goods actually bought barely grew. Inflation was also exactly why people were gloomy.
2026: the second divergence. In May 2026, the sentiment index was 44.8, the lowest since records began, yet retail sales rose 6.7%. CPI inflation at the time was 4.2% (Lesson 13), so even after removing inflation, that is a real increase in the mid-2% range. This time inflation alone does not explain it. As seen in Lesson 19, while employment still held up and wages kept rising, households kept spending even as they voiced complaints. The answer to today's question is "mood and wallet can move differently".
The consumption engine is ultimately driven by employment and income
No matter how bad the sentiment index gets, consumption doesn't fall sharply as long as jobs and wages hold up. Conversely, if employment collapses quickly (path B in Lesson 21), sentiment and spending fall together. So when looking at consumption, it's a good habit to check for divergence in this order: 'sentiment → retail sales → employment and wages'.
For stocks, it goes like this
| Signal | Tends to react first | Relatively less shaken |
|---|---|---|
| Real consumption slows · durable goods orders fall | Consumer discretionary (autos, appliances, retail, travel, leisure) | Consumer staples (food and beverage, household goods), utilities |
| Real consumption recovers · sentiment rebounds | Consumer discretionary, consumption-related transportation | Consumer staples tend to rise relatively less |
| Core capital goods orders rise | Industrials and equipment, semiconductor equipment, power equipment | Domestic-demand defensive sectors |
| Announcements of capex cuts keep coming | Earnings forecasts cut for investment-related sectors such as equipment and materials | Defensive sectors with plenty of cash |
For Korean investors, US consumption and capital expenditure connect through the path of 'export orders'. If US households buy fewer cars and appliances, the related Korean exports wobble, and if US companies increase data center and factory investment, orders can rise in Korean sectors such as semiconductors and power equipment. Check the sector links together on the impact map on the Macro Map and the Theme Map. How important exports are to Korea's economy is covered in earnest in Part 5.
Sector reactions are general tendencies. Even within consumer discretionary, some companies, like discount stores, gain customers when the economy is bad, and even consumer staples can see earnings fall if raw material prices spike. As in Lesson 1, please check the 'earnings dial' and the 'multiple dial' together.
Quick checkNews came out that the consumer sentiment index fell sharply. But in that same week's jobs report, jobs and wages rose as expected. How should you view the chance that consumption turns down right away?
✓What we learned today
- Consumption and investment are the two engines of the economy. In 2025, consumption made up 67.6% of US GDP and 48.1% of Korea's.
- Consumption is divided into durable goods, nondurable goods and services. When the economy worsens, spending that can be put off, like durable goods, and discretionary spending such as travel and dining out are cut first, so consumer discretionary swings more than consumer staples.
- Retail sales are a nominal amount, so you need to strip out inflation, and consumer sentiment is a survey asking about mood, so it can diverge from actual spending.
- Capital expenditure is small but swings widely. The standard deviation of US real capital expenditure growth in 2008–2025 was 5.5 percentage points, more than double that of real consumption (2.2 percentage points). This is because of the accelerator principle, in which small changes in demand shake investment greatly.
- Consumption signals tend to show up first in consumer discretionary sectors such as retail, autos and travel, and investment signals in equipment, industrials and semiconductor equipment.
Having read the engines of the US economy in Part 4, now we move to the key to the Korean market. In Korea, consumption has a smaller share than in the US, and exports matter much more. In the next lesson, we start with the 'exchange rate' that sets the price of those exports.
Quiz5-question quiz
Pick an answer and it's graded right away with an explanation.
Q1. Which is closest to the share of consumption in US GDP in 2025?
In the US, consumption is 67.6% of GDP. In Korea it is 48.1%, so consumption has a smaller share and investment and exports a larger one.
Q2. When the economy worsens, which type of consumption is likely to see revenue fall first and most?
Durable goods and discretionary spending can be put off, so households cut them first. That's why consumer discretionary sectors swing more with the economy than consumer staples.
Q3. In a month when prices rose 9%, retail sales rose 9%. What is the most appropriate interpretation?
Retail sales are a nominal amount. The amount rose only as much as prices did, and after removing inflation, real consumption barely grew. June 2022 was like this.
Q4. According to the accelerator principle, what generally happens to business capital expenditure when demand rises 10%?
Companies expand their equipment in line with 'changes' in demand. When expansion machines are added on top of replacement machines, orders can double. That's why investment swings more than consumption.
Q5. In May 2026, the US consumer sentiment index was at a record low, yet retail sales rose 6.7%. What is the most appropriate explanation?
Inflation at the time was 4.2%, so consumption grew even after removing inflation. Sentiment is only a signal; what actually drives consumption is jobs and income.
NextTerms · Next lesson · Related tools
Money households spend to buy goods and services. From a company's point of view, it becomes revenue.
Capital expenditureCapital expenditureMoney a company spends to buy production assets such as factories, machinery and equipment. It becomes revenue for equipment and construction companies and increases future production capacity.
Retail salesRetail salesAn indicator that adds up the amounts sold at stores, online, gas stations, restaurants and so on. In the US it is released in the middle of each month, and in Korea in the Industrial Activity Trends report. It is mostly a nominal amount, so it should be read together with the inflation rate.
Consumer sentiment indexConsumer sentimentA survey index asking how households feel about their current and future economic situation. In the US it is published by the University of Michigan and the Conference Board, and in Korea by the Bank of Korea (CCSI, where 100 is the long-term average). It can diverge from actual spending.
Durable goodsDurable goodsGoods used for three years or more, such as cars, appliances and furniture. Purchases can be postponed and are often financed in installments, making this the consumption category most sensitive to the economy and interest rates.
Consumer discretionaryConsumer discretionarySectors tied to consumption that can be cut back or put off when times get tough, such as autos, appliances, apparel, travel, leisure and retail. Earnings swing widely with the economy.
Consumer staplesConsumer staplesSectors tied to goods bought steadily regardless of the economy, such as food and beverages, household goods and tobacco. A classic defensive sector.
Accelerator principleAccelerator principleThe principle that business capital expenditure reacts to changes in demand, not the level of demand, so even a small move in demand can make investment rise or fall sharply. It is why investment swings more than consumption.
Core capital goods ordersCore capital goods ordersOrders for nondefense capital goods within US durable goods orders, excluding aircraft. Used as a leading signal for business capital expenditure.
Private consumptionPrivate consumptionMoney spent by households (and nonprofit institutions serving households) on goods and services. It makes up about 48% of Korea's GDP and about 68% of US GDP in 2025.
Cyclical stockCyclical stockAn industry or stock whose earnings swing widely with the business cycle. Semiconductors, chemicals, steel, autos and shipbuilding are typical examples.
Defensive stockDefensive stockAn industry or stock whose earnings stay relatively steady even when the economy worsens. Food and beverages, telecommunications, electricity and gas, and pharmaceuticals are typical examples.
This is educational material and not investment solicitation. There is no guarantee that past relationships will repeat in the future.