Every growth model is an answer to a question about the numbers: why do incomes per head differ so much, and why do they change at such different speeds? This lesson sets out the numbers. It covers what is measured, why growth compounds, how real countries have grown since 1990, and the regularities Nicholas Kaldor asked every theory to explain.
Two numbers describe an economy’s output over time, and they are easy to confuse. The level is how much is produced: India’s real GDP, or income per head. The growth rate is how fast the level is changing, as a percentage of where it started. A country can have a very low level and a very high growth rate, which is what “catching up” looks like, or the reverse.
The growth rate between two years is
Total output can rise simply because there are more people. What a person can buy depends on output per person. Per-head growth is output growth adjusted for population, and for small rates it is simply the difference:
Growth theory usually works with output per worker, , because that is what the models determine. Output per head and output per worker move together unless the share of the population that works changes. The lessons that follow say which one they mean each time.
Growth of 5% this year is 5% of this year’s output; growth of 5% next year is 5% of a bigger number. So growth does not add, it multiplies. After t years at a steady rate g, a level x0 has become
| Growth a year | Exact doubling time | Rule of 70 |
|---|---|---|
| 1% | 69.66 years | 70.0 years |
| 2% | 35.00 years | 35.0 years |
| 3% | 23.45 years | 23.3 years |
| 4% | 17.67 years | 17.5 years |
| 5% | 14.21 years | 14.0 years |
| 6% | 11.90 years | 11.7 years |
| 7% | 10.24 years | 10.0 years |
| 8% | 9.01 years | 8.8 years |
| 10% | 7.27 years | 7.0 years |
The shortcut is within about a year at any rate you will meet in the exam. Now set two economies growing at different rates and watch the gap open. The preset One point of growth compares 3% and 4%.
A starts at 100 and B at 25. After 30 years A is at 181 and B at 144; after 60 years, 328 and 825. B overtakes A after about 36 years.
Two things to take away. First, switch to the log axis: each curve turns into a straight line, and a steeper line means a faster rate. That is why growth is plotted on log scales. Second, try Fast, but too far behind. Growing twice as fast as the leader is not enough if you start at a tenth of its income, because the leader is also compounding. The catch-up time is
Here is real GDP per head for twelve countries from the World Bank, in dollars of constant purchasing power (2021 international dollars, which adjusts for price differences between countries). Switch on and off the countries you want, and between levels and an index that starts every country at 100 in 1990.
| Country | 1990 | 2025 | Multiple | Growth a year | Doubles every |
|---|---|---|---|---|---|
| India | 2,140 | 10,039 | 4.7× | 4.5% | 16 years |
| China | 1,667 | 25,067 | 15× | 8.1% | 9 years |
| United States | 44,384 | 76,931 | 1.7× | 1.6% | 44 years |
| Korea, Rep. | 14,378 | 55,697 | 3.9× | 3.9% | 18 years |
Source: World Bank, World Development Indicators, series NY.GDP.PCAP.PP.KD (GDP per capita, PPP (constant 2021 international $)), read from the World Bank API on 2026-10-06; the series was last updated 2026-07-13. The latest years are the World Bank’s own estimates and get revised. “Growth a year” is the compound annual rate between the first and last year shown.
Read the table under the chart, not just the lines. Between 1990 and 2025, income per head multiplied by 4.7 in India (about 4.5% a year) and 15.0 in China (8.1% a year), against 1.7 in the United States (1.6%). South Korea, already in the middle of its rise in 1990, went from $14,378 to $55,697, a multiple of 3.9. Nigeria grew slowly (1.2% a year), and Ethiopia (3.7%) grew faster, but from a very low base.
Three facts stand out, and each is a question for a theory:
In 1961 the Cambridge economist Nicholas Kaldor argued that growth theorists should not start with elegant assumptions. They should start from the broad, stable patterns in the data, which he called stylised facts, and ask whether a model reproduces them. His list, as it is usually summarised:
| # | The fact (for the long run, in industrial economies) | What a model has to deliver |
|---|---|---|
| 1 | Output per worker grows at a steady rate | Sustained growth of y = Y/L, not a one-off rise |
| 2 | Capital per worker grows at a steady rate | k = K/L rising in step with output per worker |
| 3 | The rate of return on capital is roughly constant | No fall in the return, so no running out of investment |
| 4 | The capital–output ratio is roughly constant | K/Y steady while both K and Y grow |
| 5 | Labour and capital keep roughly constant shares of income | Shares that do not drift, as in Cobb–Douglas |
| 6 | Growth rates of output per worker differ widely between countries | A reason why, rather than one rate for all |
Facts 1 to 5 together describe balanced growth: output, capital and wages grow at the same rate, so ratios such as K/Y stay put. The Solow model in Lesson 5 is built to produce exactly this in the long run, and the Harrod–Domar model in Lesson 4 cannot. Fact 6 is what the endogenous growth models of Lessons 10 and 11 were built to explain.
Treat the list with care. It is a stylisation, not a law. Jones and Romer (2010) offer an updated set of facts that emphasise population, ideas and institutions, and Karabarbounis and Neiman (2014) document a fall in the share of income going to labour in many countries since the 1980s, which strains fact 5.
Growth is a rise in output per head. Development is wider: whether people live longer, are educated, are free of poverty, and can choose how to live. Growth is usually a means to development, since a richer country can afford more schools and clinics, but the link is loose. Two countries with the same GDP per head can differ greatly in life expectancy or in how widely income is shared. That is why the Human Development Index combines income with health and education, and why exam papers ask students to separate the two ideas. The development module that follows this one takes up the measures.
If the question is about how much output, or how fast it is rising, it is growth. If it is about whether lives are going well, it is development.
A neighbour of this question appears often. “A lily pad doubles every day and covers the lake on day 48; when did it cover half?” (JNU 2012) is not 24 but day 47, because doubling means the day before was exactly half. And “a person’s wealth doubles every two years and is ₹10 lakh fifty years after the start; when was it ₹5 lakh?” (JNU 2017) is two years earlier, at year 48. Neither 24 nor 25 is an option, so both are “none of the above”.
Questions from CUET, JNU and UGC NET on compounding, doubling time and the history of growth theory.
Definitions, one-step arithmetic and which-model questions. Most CUET and Hyderabad questions sit here.
Growth accounting, steady-state algebra, statement questions and matching, up to JNU, IIT JAM and UGC NET level.
These are questions from the TestStop bank. A free account answers 10 a day, each with its worked solution; All-Access opens every set with no daily limit.
Economic growth is a sustained increase in the quantity of goods and services an economy produces. It is measured as the percentage change in real GDP, the value of output at constant prices, from one period to the next. What matters most for living standards is the growth of real GDP per head, which is output growth minus population growth, approximately.
Growth rate = (real GDP this year − real GDP last year) ÷ real GDP last year, multiplied by 100 to give a percentage. Use real GDP, which strips out price changes, not nominal GDP. For the average growth rate over several years, use the compound annual growth rate: (end value ÷ start value)^(1 ÷ years) − 1.
The rule of 70 says that something growing at g per cent a year doubles in about 70 ÷ g years. At 2% a year income doubles in about 35 years, at 7% in about 10. It comes from the exact doubling time, ln 2 ÷ ln(1 + g), because ln 2 is about 0.693. The rule is accurate for growth rates up to roughly 10%.
GDP growth is the growth of total output. GDP per capita growth is the growth of output per person, which is approximately GDP growth minus population growth (exactly, (1 + g_Y) ÷ (1 + g_N) − 1). A country can have fast GDP growth and little improvement in income per head if its population grows almost as fast.
In 1961 Nicholas Kaldor summarised the long-run regularities that a growth theory should explain: output per worker grows at a steady rate; capital per worker grows at a steady rate; the rate of return on capital is stable; the capital–output ratio is stable; labour and capital keep roughly constant shares of national income; and growth rates of output per worker differ substantially between countries. Later work, such as Jones and Romer (2010), updates the list, and the stability of factor shares has been questioned.
Because growth is a percentage change. On a log scale equal distances mean equal percentage changes, so a straight line is steady growth and a steeper line is faster growth, whatever the starting level. On an ordinary scale a rich country's steady 2% looks steeper than a poor country's steady 6% simply because 2% of a large number is a large number.
No. Growth is a rise in output, usually measured by GDP per head. Development is wider: health, education, freedom to choose, poverty and inequality. Growth is usually necessary for development and does not guarantee it. Measures such as the Human Development Index combine income with life expectancy and schooling to capture the difference.