- Types of unemployment
- Causes of inflation
- Sustainability — growth vs environment · coming soon
Richer, and harder to breathe
Over the last forty years, the number of people in China living in extreme poverty fell by close to 800 million. That is the World Bank's estimate, using its international poverty line of $1.90 a day. It is almost three quarters of the fall in extreme poverty in the whole world over that time. Most of it came from growth. Since 1978, China's real GDP has grown by more than 9% a year on average.
Now look at the air in its capital. In 2013, the average level of fine-particle pollution in Beijing, called PM2.5, was about 89 micrograms per cubic metre (µg/m³). The World Health Organization's guideline for a safe yearly average is 5. The same factories, power stations and cars that produced the growth also produced the smoke.
So was China better off or worse off? The honest answer is: both, in different ways. The last lecture ended with exactly this point. Growth brings large benefits. It also brings costs, and who gains and who loses matters as much as how fast the total rises. This lecture weighs the two sides. By the end, you will have three questions that let you judge any country's growth for yourself.
What growth actually gives a country
Economic growth is a rise in real GDP: the economy produces more goods and services. Each benefit below comes from that one fact. Read each one as a mechanism, not a list item.
1. Higher material living standards. If real GDP per head rises, the average person can have more food, housing, healthcare and schooling. This matters most where people are poor. An extra $1000 a year changes very little for a family earning $50 000. For a family earning $2000, it can mean a child stays in school instead of going to work. So the same percentage growth does far more good in a poor country than in a rich one.
2. More tax revenue, at the same tax rates. Governments collect taxes as a share of incomes and spending. When incomes and spending grow, the tax collected grows too, even if no rate changes. That money can pay for hospitals, schools, roads and pensions. Without growth, spending more means higher tax rates or more borrowing.
3. Fewer people out of work. When demand grows, firms produce more, and to produce more they hire more workers. This is the short-run growth of the last lecture, a move towards the production possibility frontier. Unemployment caused by weak demand falls. The next topic studies unemployment in detail.
4. Investment becomes affordable. A richer economy can save more and pay for roads, power grids and research. The last lecture showed that investment is spending today and capacity tomorrow. So growth pays for investment, and investment produces more growth.
5. New technology. Rich economies spend more on research, and some of it spreads far. The Global Positioning System (GPS) began in 1973 as a project of the United States Department of Defense, which still runs it. Today it guides farmers, ambulances and drivers in every country, free of charge. This spreading of benefits to people who did not pay for them is a positive externality.
Why the total can hide the people
Now the costs. The first is the easiest to miss, because it hides inside the numbers themselves.
GDP per head is a mean: total output divided by the number of people. A mean tells you nothing about how the total is shared. To see what happens to a typical person, economists look at the median income, the income of the person exactly in the middle when everyone is lined up from poorest to richest.
A tiny invented economy has five households. Their incomes, in thousands of dollars a year, are 8, 12, 20, 30 and 80.
The mean is . The median is the middle value, 20.
Next year, the richest household's income rises to 110. Nobody else's income changes. The new total is 180, so the new mean is . That is a rise of 20%. The median is still 20.
What a news report would say: "Income per head grew by 20% this year."
What four of the five households would say: "Nothing changed for us."
Check: the total rose by 30, all of it to one household. , and . ✓
Real economies are less extreme, but in many countries over the last forty years, incomes at the top grew faster than incomes in the middle. So growth in GDP per head can overstate what most people feel. The claim that gains at the top always "trickle down" to everyone else has been tested many times, and the evidence for it is weak. Growth usually does help reduce poverty, but how much depends on who gets the new income.
The costs that GDP does not count
Environmental damage. This is the largest long-run cost. Growth as it has usually happened has meant burning more coal, oil and gas, cutting more forests and producing more waste. In the language of an earlier lecture, pollution is a negative externality. The factory sells its output, and GDP counts it. The smoke harms people who are not part of the sale, and GDP does not subtract it. So a country can grow while the air, rivers and climate that people depend on get worse. Wider measures, such as the Genuine Progress Indicator, try to subtract these costs.
Lost jobs in old industries. Growth is not only more of the same output. New industries appear and old ones shrink. The economist Joseph Schumpeter called this creative destruction. When a town's main mine or mill closes, the workers' skills may not fit the new jobs, which may be in another city. This is structural unemployment. Across the whole economy, the change raises output. In one town, it can mean years of hardship.
Inflation, if demand runs ahead of capacity. The last lecture showed this on the AD-AS diagram. If spending grows faster than the economy's capacity, firms cannot produce enough, and prices rise.
In October 2022, consumer price inflation in the United Kingdom reached 11.1%, its highest in 41 years. Is this a clear example of growth "pushing output past capacity"?
Show the answer
Only partly. Demand did recover strongly after the pandemic. But the Office for National Statistics named rising energy and food prices as the main drivers. Gas prices jumped after Russia's invasion of Ukraine in February 2022. That is a rise in costs, not mainly a rise in demand. So most of this inflation was not a cost of growth at all. The inflation lectures separate these two causes, demand-pull and cost-push.
Can it last? The deepest question is about sustainability. If growth needs ever more coal, metal and land, then on a finite planet it must stop one day. Some economists answer that modern growth is increasingly in services and digital goods, which need less material for each dollar. Others reply that every service still runs on energy and machines. The rest of this lecture gives you the tools to follow this debate.
Does pollution always rise with output?
If growth and pollution moved together exactly, every extra dollar of GDP would bring a fixed amount of extra smoke. Test your own belief on the Beijing numbers.
Between 2013 and 2023, China's real GDP grew by about 80%. What do you think happened to the yearly average of PM2.5 pollution in Beijing over the same ten years? Choose one.
China's real GDP and Beijing's yearly average PM2.5 in 2013 and 2023, each shown as an index with 2013 = 100.
Output and this kind of pollution moved in opposite directions. So pollution is not tied to output by a fixed law. It depends on how the output is produced.
Real case: Beijing's air (2013 to 2023)
The policy. In 2013, China's national government launched an action plan against air pollution. Around Beijing, it cut the use of coal: many coal boilers and home stoves were switched to gas or electricity, and some heavy industry was closed or moved. It also tightened standards for vehicles and fuel.
The prediction. There were two predictions. A simple one: cleaner air would cost growth, because cheap coal powered the factories. And a more hopeful one, from an idea called the environmental Kuznets curve (below): as people become richer, they demand cleaner air and can afford to pay for it.
What happened. Beijing's yearly average PM2.5 fell by 64.2% between 2013 and 2023, to 32 µg/m³ (Beijing Municipal Government, January 2024). Over the same years, China's real GDP grew by about 80% (International Monetary Fund data, through FRED). But two facts limit the success. First, 32 µg/m³ is still more than six times the WHO guideline of 5. Second, carbon dioxide behaved differently. China's CO₂ emissions kept rising, and in 2023 China produced about 35% of the world's CO₂ (International Energy Agency, 2024).
The verdict. The case shows that one cost of growth, local air pollution, can be cut sharply while output still grows. But it did not happen by itself. It took deliberate policy, paid for partly by a richer economy. It also shows the limit. The harm from smoke falls on the people who breathe it, so they press for action. The harm from CO₂ is spread across the whole world and the future. That makes it much harder for any one country to act.
The Kuznets curves: does growth fix its own costs?
In 1955, the economist Simon Kuznets suggested a pattern for inequality. In early industrialisation, workers move from farms into factories, and most of the gains go to the owners of the new factories. So inequality rises. Later, workers become scarcer, wages rise, and governments spend more on schools and welfare. So inequality falls. Plotted against GDP per head, inequality would trace an upside-down U. This is the Kuznets curve.
Some countries fit this story for a time. But in many rich countries, inequality has risen again since about 1980. So the curve is not automatic. Countries that invest in education and protect workers' bargaining power tend to share growth more widely.
In the 1990s, economists applied the same shape to pollution. This is the environmental Kuznets curve.
The evidence fits some local pollutants well. In the United States, total emissions of six common air pollutants fell by 78% between 1970 and 2023 while the economy kept growing (US Environmental Protection Agency). Beijing's PM2.5 is another example. For carbon dioxide, the evidence is much weaker. Notice what moved the curve down in every good example: laws, standards and new technology. The curve describes what richer societies often choose to do. It is not a force that cleans the air on its own.
Decoupling: growing without growing the damage
The key word in today's debate is decoupling: breaking the link between GDP and environmental damage. There are two kinds, and confusing them is the commonest mistake in this topic.
- Relative decoupling: emissions still rise, but more slowly than GDP. Each dollar of output is cleaner, but the total damage still grows.
- Absolute decoupling: emissions actually fall while GDP rises.
Whether a country gets one or the other depends on a race between two rates. GDP growth pushes emissions up. A fall in the emissions from each dollar of output, called the emissions intensity, pushes them down. Growth factors multiply, as in the last lecture. So each year, emissions are multiplied by (1 + growth rate) × (1 − rate of fall in intensity). Emissions fall only if that product is less than 1. Run the race yourself below.
Set how fast GDP grows and how fast each dollar of output becomes cleaner. The graph runs the economy for 30 years, starting with both GDP and emissions at 100. Find the line between relative and absolute decoupling.
Presets:
GDP and emissions after 30 years in an invented economy, both starting at 100, for four pairs of rates.
| GDP growth | Cleaner per $ | GDP, year 30 | Emissions, year 30 | Result |
|---|---|---|---|---|
| 6% | 2% | ≈ 574 | ≈ 313 | Relative |
| 5% | 3% | ≈ 432 | ≈ 173 | Relative |
| 3% | 3% | ≈ 243 | ≈ 97 | Absolute (just) |
| 2% | 4% | ≈ 181 | ≈ 53 | Absolute |
Emissions fall only when (1 + growth) × (1 − cleaning) is less than 1. Roughly, each dollar must become cleaner a little faster than the economy grows.
What should you have seen? When GDP grows at 5% and each dollar becomes 3% cleaner, emissions still rise by about 73% in 30 years. That is relative decoupling. For emissions to fall, the cleaning must slightly outpace the growth. At 3% and 3%, emissions fall only a little, because 1.03 × 0.97 is just under 1. Notice also what the model leaves out. It assumes steady rates for 30 years. It ignores emissions that move abroad when a country imports goods instead of making them. And for the climate, what matters is the total CO₂ added to the air over time, not the level in one year.
Some countries have achieved absolute decoupling. The UK government reports that the country's greenhouse gas emissions in 2023 were 53% below their 1990 level, while its economy grew by 80% over the same period. These figures count emissions produced inside the UK. Part of the fall happened because the UK now imports many goods that it once made, so their emissions happen in other countries. Measures based on what the UK consumes show a smaller fall.
A politician says: "The UK cut emissions by half while its economy grew by 80%. This proves that the whole world can keep growing with no harm to the climate." What is right, and what goes too far?
Show the answer
What is right: absolute decoupling inside one country has happened, so growth and falling emissions are not impossible together. What goes too far: first, part of the UK's fall moved emissions abroad through imports, and the world as a whole cannot import from another planet. Second, the UK did this mainly by replacing coal in electricity, which is the easiest step; later steps, such as heating, steel and flying, are harder. Third, a falling rate is not enough for the climate if the total keeps adding up. So the evidence shows that decoupling is possible, not that it is happening fast enough everywhere.
Three traps
GDP per head is a mean. If most of the gain goes to a few people, the median income can stay flat, as in Worked Example 1. Ask who received the growth before you say who benefited.
Neither curve says that growth fixes its own costs. Where pollution or inequality fell, laws, standards, technology and public spending made it happen. And for carbon dioxide, the evidence for a downturn is weak.
"Each dollar of output is 3% cleaner every year" sounds like progress. But if GDP grows faster than 3%, total emissions still rise. Always ask whether the total is falling, not only the amount per dollar.
Work each one on paper first. Open a hint only when you are stuck.
-
An invented country's real GDP is $500 billion, and its government collects 20% of GDP in tax. Real GDP grows by 4% a year and tax rates do not change. How much extra tax is collected after one year? Roughly how much is collected each year after ten years?
Hint 1
Tax collected is 20% of GDP. Find GDP after one year first.
Hint 2
After ten years, GDP has been multiplied by 1.04 ten times, and so has the tax collected.
Full answer
Now: 20% of $500 billion = $100 billion. After one year GDP is $520 billion, so tax is $104 billion: an extra $4 billion, with no change in tax rates. After ten years: 1.04¹⁰ ≈ 1.48, so tax is about $148 billion a year. Check with the rule of 70: at 4%, revenue doubles in about 17.5 years, so after 10 years it should be well under double. ✓ This is why governments care so much about growth: it pays for public services without raising rates.
-
Five households in an invented economy have incomes (in thousands of dollars) of 5, 10, 15, 25 and 45. Next year the economy gains $9 thousand of income in total. In plan A, it all goes to the richest household. In plan B, it is shared equally among the three poorest. Find the mean and the median after each plan.
Hint 1
The total rises by 9 in both plans, so the mean changes by the same amount in both.
Hint 2
In plan B the three poorest each gain 3. Line the new incomes up in order before you find the middle one.
Full answer
Before: total 100, mean 20, median 15. Plan A: incomes 5, 10, 15, 25, 54. Total 109, mean 21.8, median 15. Plan B: incomes 8, 13, 18, 25, 45. Total 109, mean 21.8, median 18. Both plans show the same growth in income per head (9%). But the median, the typical household, gains nothing in plan A and 20% in plan B. Check: 18 ÷ 15 = 1.2. ✓ The same growth figure can describe two very different countries.
-
An invented country's GDP grows by 5% a year, and its emissions per dollar of output fall by 3% a year. By what percentage do total emissions change each year? How fast would emissions per dollar have to fall for total emissions to stay level?
Hint 1
Growth factors multiply: emissions are multiplied by 1.05 × 0.97 each year.
Hint 2
For emissions to stay level, 1.05 × (1 − r) = 1. Solve for r.
Full answer
1.05 × 0.97 = 1.0185, so emissions rise by about 1.85% a year. That is relative decoupling. Over 30 years, 1.0185³⁰ ≈ 1.73, a rise of about 73%, which matches the model. For level emissions: 1 − r = 1 ÷ 1.05 ≈ 0.952, so r ≈ 4.8% a year. Notice that the simple subtraction 5% − 3% = 2% gives almost the right answer here, but the exact method multiplies.
-
An invented country plans a new coal mine and coal power station in a region where 12% of workers are unemployed. Officials expect national GDP to be 1.5% higher as a result. Use three questions from this lecture to evaluate the plan: who gains, what is lost that GDP does not count, and can it last?
Hint 1
Take each question in turn and give at least one point for and one point against.
Hint 2
Think about local jobs, local air, global CO₂, and what happens to the region when the coal runs out or the world moves away from coal.
Full answer
Who gains? Local workers gain jobs in a region with high unemployment, and the government gains tax revenue. But if most profits go to the owners, the median household may gain much less than the 1.5% suggests. What is lost? Local air pollution harms the health of people nearby, an external cost GDP does not subtract. The CO₂ adds to climate damage that falls on the whole world. Can it last? Coal is finite, and many countries are moving away from it. The region could face creative destruction later, with structural unemployment when the mine closes. A balanced verdict: the plan gives real short-run benefits, but its value depends on pollution controls, on how the gains are shared, and on a plan for the region's future after coal. A cleaner project with similar jobs might pass all three questions better.
Growth is a means, not an end. GDP measures how much an economy produces. It does not measure who receives it, what was used up to make it, or whether it can continue. So every claim about growth can be tested with three questions. Who gains? What is lost that GDP does not count? Can it last?
These questions do not make growth bad. For a family in extreme poverty, growth is the difference between hunger and enough food, and nearly 800 million people in China alone left extreme poverty this way. The questions make growth something to manage, not just maximise. The next topic turns to one of the costs that people feel most directly: unemployment.
- Growth raises living standards, tax revenue, employment and investment, and the benefit is greatest where people are poorest.
- GDP per head is a mean, so it can rise while the median income, the typical household's, stays flat.
- Pollution is a negative externality that GDP counts as output but never subtracts as damage.
- Creative destruction raises output overall but can cause structural unemployment in declining industries and towns.
- The Kuznets curves describe a rise and then a fall in inequality or pollution, but the fall depends on policy and is weak for carbon dioxide.
- Relative decoupling means emissions grow more slowly than GDP, while absolute decoupling means emissions fall as GDP grows.