Growth against development, GDP against GNI, purchasing power parity, how the Human Development Index is built from health, education and income, what Qatar and Costa Rica show about income and development, and the limits of the HDI. Through the 2025 Human Development Report.
The UN’s Human Development Report 2025, published on 6 May 2025, ranks 193 countries on more than income. Iceland tops the table with an HDI of 0.972; the United Kingdom is 13th at 0.946; South Sudan is last at 0.388, with life expectancy of 57.6 years and an income per head of $688.
The interesting cases are in the middle of the money table. Qatar, with income per head of $105,000 at purchasing power parity, almost double the UK’s, ranks 43rd. Costa Rica, with less than half the UK’s income, has a life expectancy of 80.8 years, close to the UK’s 81.3, and ranks 62nd, above several far richer states. India, 130th, has raised life expectancy from 58.6 years in 1990 to 72 and expects a child starting school now to get 13 years of education; its adults have had 6.9.
If income does not tell you how developed a country is, what does, and how do you put a number on it?
Source: UNDP, Human Development Report 2025, 6 May 2025, table 1 (data for 2023); GNI per capita in 2021 PPP dollars.
Economic growth is a rise in real output. Economic development is a rise in living standards and welfare more broadly: health, education, freedom, security and the distribution of income, as well as its level.
GDP measures output produced within a country. GNI measures income received by its residents: GDP plus income from abroad minus income paid abroad. Per capita figures divide by population. Purchasing power parity (PPP) converts currencies at rates that equalise the cost of a basket of goods, so that $1 buys the same in India as in Iceland; without it, poor countries with cheap haircuts and rents look poorer than they are.
The Human Development Index combines three dimensions into one number between 0 and 1: a long and healthy life (life expectancy at birth), knowledge (expected years of schooling for a child and mean years of schooling for adults) and a decent standard of living (GNI per capita, PPP, on a log scale). Each is turned into an index from 0 to 1 and the HDI is the geometric mean of the three, so a weakness in one dimension cannot be fully offset by strength in another.
Bands: very high 0.800 and above, high 0.700 to 0.799, medium 0.550 to 0.699, low below 0.550.
Qatar’s income per head is almost double the UK’s: $105,000 against $54,000 at purchasing power parity.
On the UN’s Human Development Index the UK ranks 13th in the world. Where does Qatar rank?
Qatar ranks 43rd with an HDI of 0.886. Adults there have had 10.8 years of schooling on average against 13.5 in the UK, and a child can expect 13.1 years against 17.8. Development is more than income, and the HDI is built to say so.
Start with income. The chart puts GNI per capita across the bottom, in thousands of PPP dollars, and the HDI up the side. South Sudan: $688 a year. India: $9,046. The UK: $54,372. The income gap between the UK and South Sudan is eighty to one. Whether the gap in living standards is that large is the question the rest of the chart answers.
Check: A country’s GDP is far larger than its GNI. The most likely reason is:
GNI subtracts income paid to foreign residents. Ireland is the standard example: multinationals book output there and repatriate the profit.
All eight countries. Incomes are at purchasing power parity, so a dollar buys the same basket everywhere; at market exchange rates India’s figure would be about a third as large and the gap would look wider than it is. Now the vertical axis. Iceland tops the HDI at 0.972, Norway 0.970, the UK 0.946. Qatar, far to the right at $105,000, sits at 0.886. Costa Rica, at $23,000, at 0.833.
Check: India’s GNI per capita is about $2,500 at market exchange rates but $9,046 at purchasing power parity. The PPP figure is higher because:
PPP rates equalise what money buys. Non-traded goods such as rent, food and services are far cheaper in India, so the same rupee income represents more real consumption.
The dotted curve shows roughly what income alone would predict. It rises steeply at low incomes and flattens: diminishing returns to income, which the HDI builds in by using the log of GNI. Going from $700 to $9,000 moves a country from South Sudan to India; going from $54,000 to $105,000 adds almost nothing. The bands mark the UN’s categories.
Check: The HDI is a geometric mean of three dimension indices rather than a simple average. The effect is that:
The geometric mean (the cube root of the product) penalises imbalance: 0.9 × 0.9 × 0.5 gives 0.74, below the simple average of 0.77.
Now the two outliers. Qatar is far below what its income predicts: a child there can expect 13.1 years of schooling against 17.8 in the UK, and adults have had 10.8 years against 13.5. Costa Rica is well above: it abolished its army in 1948 and spent on health and schools, and its people live to 80.8 on an income less than half the UK’s. Income buys development only if it is spent on it.
Teacher tip: This pair is the best evaluative example in the topic. Learn the four numbers: Qatar $105,000 and 0.886; Costa Rica $23,000 and 0.833. One sentence with those in it earns an application mark in any development question.
Check: Qatar lies below the curve and Costa Rica above it. What explains the gap between them in the HDI despite Qatar’s income being four times higher?
Costa Rica: life expectancy 80.8, expected schooling 16.3 years. Qatar: 82.4 and 13.1. With mean schooling of 8.8 against 10.8, Costa Rica’s education index is close to Qatar’s while its income index is far lower, and the HDI gap is only 0.05.
What the HDI leaves out. It is an average, so it hides inequality: the inequality-adjusted HDI cuts India’s score from 0.685 to 0.475. It says nothing about freedom, security or the environment: Qatar’s workforce is mostly migrants on temporary contracts; Norway’s score rests on oil. Expected schooling counts years in a classroom, not learning. And a national figure hides regions: London alone would score 0.982, above Iceland.
Check: India’s HDI is 0.685 but its inequality-adjusted HDI is 0.475, a loss of 30.7%. What does this show?
The IHDI discounts each dimension by how unequally it is distributed. A 30.7% loss means that if achievements were shared equally the HDI would be as recorded; because they are not, the typical person’s development is lower.
So which measure? GNI per capita exists for every country every year and drives much of the rest. The HDI adds the two outcomes income is meant to buy, and its rank minus the income rank shows who converts money into lives well lived: Costa Rica plus, Qatar minus. Neither is development itself; both are indicators of it.
Check: Which statement is the strongest case for using the HDI rather than GNI per capita alone to compare countries?
Income is instrumental. The HDI measures two of the things it is supposed to buy, and the gap between income rank and HDI rank is itself informative.
The line is straight: it says each extra dollar adds the same to development. The points show diminishing returns: the HDI rises steeply at low incomes and flattens, which is why the index uses the log of income. A straight line would put India far below where it is and Qatar above.
Explain why GNI per capita is an incomplete measure of a country’s development, and assess the usefulness of the Human Development Index as an alternative.
Economic development is a sustained improvement in living standards and wellbeing: health, education, security and the distribution of income, not only its level. GNI per capita, income received by residents divided by population and converted at purchasing power parity, measures only the level of income.
That is incomplete for several reasons. Income is a means, not an end: it says nothing about whether it is spent on hospitals and schools or on palaces, and nothing about how it is shared. It omits health, education, freedom, security and the environment. It misses unpaid and informal work, which is a large share of output in poor countries. Qatar shows the problem: income per head of $105,000 at PPP, almost double the UK’s, yet children can expect 13.1 years of schooling against 17.8 in the UK, and much of the workforce is migrant labour on temporary contracts whose living standards a national average does not describe.
The Human Development Index, published by the UN since 1990, combines life expectancy at birth, expected and mean years of schooling, and the log of GNI per capita into a geometric mean between 0 and 1. It measures two of the outcomes income is supposed to buy, and because it uses the log of income it gives diminishing weight to money. The gap between a country’s income rank and its HDI rank is itself informative: Costa Rica, with less than half the UK’s income but a life expectancy of 80.8 years, ranks 62nd with an HDI of 0.833, while Qatar ranks 43rd at 0.886 despite four times the income.
**However,** the HDI is an average, so it hides inequality: adjusting for it cuts India’s score from 0.685 to 0.475. Its equal weights are a value judgement, and because income enters as a logarithm it barely distinguishes rich countries from one another: the top twenty lie within 0.04 of each other. It omits freedom, security and the environment, and years of schooling say nothing about what was learned. The UN’s companion indices, the inequality-adjusted HDI, the Multidimensional Poverty Index and the planetary-pressures-adjusted HDI, each repair one gap. The HDI is a better single measure of development than income alone, and most useful for comparing poor and middle-income countries; for a full picture it should be read alongside income, inequality and environmental measures rather than instead of them.
All figures: United Nations Development Programme, Human Development Report 2025 (published 6 May 2025), statistical annex table 1, data for 2023. GNI per capita in 2021 PPP dollars. The dotted curve on the chart is an illustration of the typical relationship, not a UNDP series. Diagrams drawn with Rough.js and perfect-freehand, both MIT licensed.