1The idea in one paragraph

Two countries can produce the same output per person and feel completely different to live in, because output has to be shared out and the sharing is never even. This subtopic is about measuring how uneven it is, about the people at the bottom of the distribution who cannot afford what they need, about why the gaps open in the first place, and about the tools a government has to close them. The measuring is technical and you can get it exactly right. The policies are contested, and the marks in part (b) go to students who know which parts are settled and which are not.

2Equality, equity, and why poverty is a different question

Start with three words that are used loosely everywhere except in an exam.

Equality means sameness. An equal distribution of income is one in which everybody receives the same amount. It is a fact about a distribution, and it can be measured.

Equity means fairness. It is a judgement about whether a distribution is just, and different people, looking at the same numbers, reach different answers. Most people think it is fair that a surgeon who trained for eleven years earns more than a school leaver, so most people think some inequality is equitable. Most people also think a child should not go without food because of who their parents are, so most people think some inequality is inequitable.

The relationship between the two is the point. Equity does not require equality. A perfectly equal distribution would pay the same to someone working seventy hours and someone working none, which few would call fair. But large inequality is usually judged inequitable, because past some distance the gaps stop reflecting effort and start reflecting where people started. So equity is the reason we care about equality, and equality is one of the things we measure when we ask whether a society is equitable. Equality is positive, meaning it can be settled by data. Equity is normative, meaning it cannot.

Now the distinction the whole subtopic rests on.

Inequality is about the gaps between people. Poverty is about the floor that some people fall below. A country can move one without moving the other.

Imagine Brindal grows for a decade. Every household's income rises, so far fewer households sit below the line that buys food and shelter: poverty has fallen. But the richest tenth gained far more than the poorest tenth, so the gaps are wider than before: inequality has risen. Both sentences are true at once, and section 6 draws the picture. An answer that treats "inequality fell" and "poverty fell" as the same claim has thrown away the distinction the examiner is testing.

3What is unequally distributed: income and wealth

The guide splits economic inequality into two, and they behave differently.

Income is a flow of money to a person over a period: wages and salaries, rent from property, interest from savings, profit from a business, and transfer payments from the government. You measure it per week, per month or per year, and the period is part of the number.

Wealth is the stock of assets a person owns at a moment, minus what they owe: a house, land, shares, savings, a pension pot, a business, less mortgages and loans. You measure it on a date.

Figure 1 shows how the two feed each other.

Income you do not spend becomes saving, and saving adds to wealth. Wealth pays income back, as rent, interest and profit. Both loops run faster for whoever already owns assets, which is why wealth is more unequally distributed than income in every country that measures both. A household can have a decent income and no wealth at all; a household with wealth keeps receiving income even when it stops working.

Two consequences for your answers. Wealth inequality is the more persistent of the two, because assets pass between generations and incomes do not. And a policy that changes incomes, such as a higher minimum wage, may leave wealth inequality untouched for years.

4Measuring inequality: the Lorenz curve and the Gini coefficient

The Lorenz curve turns a whole distribution into one line. Line the population up from poorest to richest. Along the horizontal axis put the cumulative percentage of the population, starting with the poorest. Up the vertical axis put the cumulative percentage of total income they receive. Both axes are cumulative, which means each point adds everyone to its left.

If income were shared perfectly equally, the poorest 10% would receive 10% of income, the poorest half would receive half, and the curve would be the 45° straight line. That line is the line of equality, and no real country is on it. The further a country's curve sags below the line, the more unequal it is. Figure 2 reads one off.

In Figure 2 the poorest 50% of the population receive 22% of the income. Read that sentence off the curve again, because reading a single point off a Lorenz curve is a standard Paper 2 mark.

The Gini coefficient puts a number on the sag. It is the area between the line of equality and the Lorenz curve, area A, divided by the whole triangle underneath the line of equality, areas A and B together:

Gini = A ÷ (A + B)

It runs from 0 to 1. A Gini of 0 means perfect equality, where the curve sits on the line and area A disappears. A Gini of 1 means one person receives everything. Real countries run roughly from the high 0.2s to the low 0.6s. Some data sources multiply by 100 and call it the Gini index, so 0.38 and 38 are the same statistic; check which one an extract is using before you comment on it.

Figure 3 shows what a rise in inequality looks like.

The population has not changed. The distribution has. L₂ sags further from the line than L₁, so area A is bigger, so the Gini rises from 0.38 to 0.58. Whenever an extract reports a rising Gini, this is the picture behind it.

Redistribution runs the same picture backwards. Figure 4 shows a country's distribution before tax and transfers, and the same country after.

Nobody moved along the horizontal axis: the same people are there in the same order. What moved is the income they end up with, so the curve climbs towards the line of equality and the Gini falls from 0.50 to 0.31. This is the single most useful diagram in the subtopic, because almost every policy in section 11 is an argument about how far this curve can be pulled up and at what cost.

One limitation to keep for evaluation. The Gini compresses a whole distribution into one number, so two countries with the same Gini can be unequal in different places: one with a very poor bottom tenth, another with an extremely rich top 1%. And because the Gini is usually calculated on income, it understates inequality of wealth, which is larger.

6Poverty: what it means and how it is measured

Poverty is the condition of not having enough resources to meet needs. The guide splits it in two, and the difference is which needs.

Absolute poverty is having an income below the amount needed to buy the basic necessities of life: enough food, safe water, shelter, clothing, basic health care. The line is fixed in real terms, so it means the same basket this year as last. Because it is fixed, absolute poverty can in principle fall to zero, and in a growing economy it usually falls.

Relative poverty is having an income far below what is normal in the society you actually live in — commonly set at 60% of the median household income of that country. It is a statement about distance from your neighbours, not about your basket of goods. Because the line moves with the median, relative poverty can persist in a rich country and can rise even while everybody is getting better off.

Figure 6 puts both lines on one distribution of households. Note where the relative line sits: it is tied to the median, so if every income doubled overnight the absolute line would leave almost nobody below it while the relative line would simply double too and leave the same households below it.

That is the mechanism behind section 2's claim, and Figure 7 draws it properly.

On the left, growth has pushed the whole distribution of Brindali incomes to the right, so far fewer households now fall below the fixed absolute line: poverty has fallen. On the right, the same decade's Lorenz curve has sagged further from the line of equality, because the top gained proportionately more than the bottom: inequality has risen. The two panels describe the same country in the same years. Neither is wrong.

Measuring poverty: single indicators. A single indicator uses one variable, almost always income.

• International poverty lines set one income line, the same for every country, after converting currencies for what money actually buys locally rather than at the exchange rate. The line is a small sum per person per day, and it is revised as prices and data improve, so use the figure an extract gives you rather than one you memorised. Its purpose is comparison: it lets you say poverty is higher in one country than another, which a national line cannot.
• Minimum income standards ask a different question, inside one country: what does a household here need to reach a standard of living that people here agree is the minimum acceptable? A panel of ordinary members of the public agrees a basket — food, clothing, housing, heating, transport, enough to take part in ordinary social life — and the cost of that basket is the line. It is higher than an absolute line, specific to one country, and built from what people judge necessary rather than what keeps a body alive.

Measuring poverty: composite indicators. A composite indicator combines several variables into one figure, because income is not the only thing people go without. The Multidimensional Poverty Index (MPI) is the one the guide names.

The MPI works across three dimensions — health, education and standard of living — measured by ten indicators, as Figure 8 shows. A person counts as multidimensionally poor if they are deprived in at least a third of the weighted indicators. The index then combines how many people are poor with how many deprivations the average poor person suffers, so it captures both the reach of poverty and its depth. A household with a small but adequate cash income, no electricity, no sanitation and a child out of school does not appear in an income headcount and does appear here.

Difficulties in measuring poverty. Every figure in this section comes with problems, and naming them is an evaluation mark.

1. The line is a judgement. Whether the relative line is 50% or 60% of the median changes the headcount without changing anybody's life.
2. Income data are incomplete. Subsistence farming, informal work, cash in hand and self-employment are badly recorded, and in the poorest countries they are most of the economy.
3. Households are treated as one person. Income is measured per household, so unequal sharing inside a household, often along lines of gender or age, is invisible.
4. Non-money income is missing. Food you grow, a free clinic, a free school place and a subsidised bus all raise a standard of living without touching an income figure.
5. Prices differ from place to place. The same money buys much more in a village than a capital city, so one national line misdescribes both.
6. A snapshot hides movement. People fall into poverty and climb out of it, so a one-year headcount does not tell you whether the same families are poor for a decade or different ones each year.
7. The data are worst where they matter most. Surveys are expensive and infrequent in the poorest countries, so figures are often several years old and heavily estimated.

7What causes inequality and poverty

Eight causes. Learn all eight with one concrete sentence each, because a Paper 1 part (a) asks for two and expects each to be explained rather than listed.

Inequality of opportunity. Two children of equal ability do not face the same chances if one is born into a household that can pay for books, tutoring, a quiet room and university and the other is not. The cause is not income today but access to what turns ability into earnings.

Figure 9 shows why this cause is so stubborn: low income leaves less to spend on schooling and health, which lowers human capital, which leads to low-paid work, which leaves low income again. The loop runs at the speed of a generation, so the effect shows up in the data long after the cause. Nothing in the loop requires anybody to be lazy or unlucky. The starting point does the work.

Different levels of resource ownership. Land, buildings, shares and businesses pay rent, interest and profit to whoever owns them. Owners therefore receive income without working, non-owners do not, and since ownership is concentrated, this widens both income and wealth gaps.

Different levels of human capital. Human capital is the stock of education, skills, training and health that a worker carries. More of it usually means higher productivity, and higher productivity usually means higher pay. Unequal schooling and unequal health therefore turn into unequal earnings.

Discrimination. Where workers of the same productivity are paid differently, or not hired at all, because of their gender, race, ethnicity, religion, age or disability, the gap has nothing to do with what they can do. Discrimination also works earlier, by closing off the education and the contacts that build human capital in the first place.

Unequal status and power. Income depends on bargaining. A worker in a strong union, a profession that controls entry to itself, or a firm with few competitors captures more than a worker with none of those. Power also decides who writes the rules — which taxes are collected, which subsidies are paid, whose property rights are enforced.

Government tax and benefits policies. The state can narrow the distribution or widen it. A progressive income tax paying for transfers to the poorest narrows it; a system that relies on indirect taxes and spends little on transfers widens it. This cause is unusual because it is also a policy, so it appears again in sections 9 and 11.

Globalisation and technological change. Both raise total output and both change who gets it. Technology has raised the return to high skills and replaced routine tasks that used to support middle-income jobs, and trade puts workers producing tradeable goods into competition with workers abroad. Consumers gain from cheaper goods; the workers displaced lose, and their losses are concentrated in particular regions and occupations.

Market-based supply-side policies. Deregulation, privatisation, lower marginal tax rates and weaker employment protection aim to raise efficiency and output. Whatever they do to the size of the economy, they tend to widen its distribution, because they cut the taxes and the labour protections that compress it. Notice the trade-off the guide is pointing at: the same policy can raise the total and worsen the share.

8What inequality does: growth, living standards and social stability

This is an AO3 line, so the exam wants a judgement, and a judgement means saying what the effect depends on. Economists genuinely disagree here. Presenting one side as settled fact is the commonest way to lose evaluation marks.

Economic growth. There are serious arguments in both directions.

Inequality may support growth. Higher-income households save a larger share of their income, and saving funds investment. The prospect of a large reward is also an incentive to work harder, train longer, start a business and take risks, and an economy that flattens rewards flattens those incentives too.

Inequality may hold growth back. A talented child in a poor household cannot borrow against future earnings to pay for education, so ability goes unused and the economy produces less than it could. Low-income households spend a larger share of any extra income, so shifting income towards the top can weaken consumption and therefore aggregate demand. And instability, which the third impact covers, deters the long-term investment growth depends on.

The evidence does not settle it cleanly, and the most defensible position in an essay is conditional: the effect probably depends on how much inequality there is, on where in the distribution it sits, and on what causes it. Inequality that reflects differences in effort and innovation is a different thing from inequality that reflects inherited assets and closed doors, even when the Gini is the same.

Standards of living. A standard of living covers material things — income, housing, food, goods — and non-material things — health, life expectancy, education, safety, security. Inequality matters for both, and mostly through the bottom of the distribution. A country with high average income and high inequality can have a large group living far below what that average suggests, so GDP per capita flatters it. Where health care and education have to be paid for, the effect compounds: less income now means worse health and less schooling, which means less income later. Some economists argue further that inequality itself damages health and well-being, through the stress of low relative status, and others reply that what looks like an effect of inequality is really an effect of poverty. Report the disagreement; do not resolve it by assertion.

Social stability. When a large group believes the distribution is unfair and their position cannot improve, the results include falling trust, higher crime, protest and sharper political division. Instability has an economic price: firms facing the risk of disruption or of sudden changes in the rules invest less, which feeds back into growth. Two qualifications matter for a balanced answer. What seems to do the damage is perceived unfairness rather than a number, so a country whose citizens believe the gaps are earned and that mobility is real can hold together at a Gini that would strain another. And the direction of causation is hard to establish, because instability also causes inequality.

Across all three impacts, the honest summary is that the connections are real, widely studied and still argued over. Write the argument, weigh it, and say what your judgement depends on.

9Taxation

Tax does two jobs here: it raises the money that pays for the policies in section 11, and the way it is collected changes the distribution by itself.

Direct and indirect. A direct tax is paid to the government by the person or firm on whom it is charged, and it falls on income or wealth. An indirect tax is charged on spending and collected by the seller, who passes it to the government, so the person bearing it pays it as part of a price.

The guide names three direct taxes.

• Personal income tax, charged on wages, salaries and other personal income. Usually the most progressive tax a country has and the main tool for reshaping the distribution of income.
• Corporate income tax, charged on company profits. It reaches income that flows to owners of capital, so it bears on wealthier households, though firms can shift profits between countries to avoid it, which limits how much a single government can collect.
• Wealth taxes, charged on assets rather than on the income from them: taxes on property, on inheritance, on gifts, on capital gains when an asset is sold. These reach the stock rather than the flow, so they are the direct answer to the wealth inequality in section 3, and they are the hardest to collect, because assets can be moved, hidden and valued differently.

Indirect taxes include value added tax or goods and services tax charged on most purchases, and excise duties charged on particular goods such as fuel, alcohol and tobacco. They are cheap to collect and hard to evade, which is why governments rely on them.

Progressive, proportional, regressive. These three words describe a tax system, and the test is always what happens to the average rate as income rises.

• A progressive tax takes a larger percentage of income as income rises, so the average rate rises.
• A proportional tax takes the same percentage at every income, so the average rate is constant.
• A regressive tax takes a smaller percentage of income as income rises, so the average rate falls.

The test is the percentage, not the amount of money. Under a 10% proportional tax a household earning 200,000 hands over ten times as much cash as one earning 20,000, and the system is not progressive, because both pay 10%.

Average and marginal rates. This is the pair students confuse most, and one figure fixes it.

• The average tax rate (ATR) is the share of your whole income that goes in tax: ATR = (total tax paid ÷ total income) × 100.
• The marginal tax rate (MTR) is the rate charged on your next unit of income: MTR = (change in tax paid ÷ change in income) × 100.

They are different because a progressive income tax is charged in bands, and moving into a higher band raises the rate on the income inside that band only, never on the income below it. Take Brindal's bands:

A household earning $100,000 pays nothing on the first $10,000; 20% on the $30,000 between $10,000 and $40,000, which is $6,000; 40% on the $40,000 between $40,000 and $80,000, which is $16,000; and 50% on the last $20,000, which is $10,000. Total tax is $32,000. Its marginal rate is 50% but its average rate is 32,000 ÷ 100,000 × 100 = 32%.

Figure 11 draws both rates against income. The marginal rate is the step function; the average rate is the smooth line that climbs towards it but never reaches it. The gap between them is all the income taxed in the lower bands. Because the average line rises, this system is progressive, and that is the proof — not the fact that the top rate is 50%.

The practical consequence is worth stating, because students and newspapers both get it wrong: a pay rise that takes you into a higher band never leaves you worse off, since only the income above the threshold pays the higher rate.

Why indirect taxes are usually regressive. The rate is the same for everybody, so at first sight the tax looks proportional. It is not, once you compare the tax with income.

In Figure 12 both households pay a 20% value added tax at the till. The household earning $30,000 spends $20,000 on taxed goods, because a low income is nearly all spent, and pays $4,000 of tax, which is 13.3% of its income. The household earning $100,000 spends $35,000 on taxed goods, saving the rest, and pays $7,000 of tax, which is 7.0% of its income. The same rate, the smaller share of income for the richer household: that is exactly the definition of regressive.

Two refinements for a strong answer. It is the burden relative to income that is regressive, not the rate, so say it that way. And governments soften the effect by exempting necessities such as basic food, children's clothing and medicine, or by taxing luxuries at higher rates, which makes the system less regressive without making it progressive.

11The policies that reduce poverty and inequality

Six further policies, each with a mechanism and a cost. Part (b) questions ask you to weigh them, so learn the objection alongside the policy.

Policies to reduce inequality of opportunity, and investment in human capital. Free or subsidised schooling, early years places, health care, adult training, and affordable credit for people with nothing to pledge as security. These attack the loop in Figure 9 at its weakest link, and they raise long-run productive capacity as well as equity, which is why they attract the least disagreement. The objection is time: a child starting school today enters the labour market in fifteen years, so a government in trouble now gets nothing from it now.

Transfer payments. Payments from the government to households for which nothing is produced in return: unemployment benefit, state pensions, child benefit, disability payments, income support. They are the fastest way to raise incomes at the bottom, and they are what pulls the curve up in Figure 4. Two objections are standard. They cost a great deal, so they need tax revenue that must come from somewhere. And where a benefit is withdrawn sharply as earnings rise, the household can lose most of each extra pound earned, which weakens the incentive to work more — the argument for withdrawing benefits gradually rather than at a cliff edge.

Targeted spending on goods and services. Rather than cash, the government provides the thing itself: free school meals, subsidised housing, free health care at the point of use, subsidised public transport, free textbooks. Because the help arrives in kind, it reaches the intended use and cannot be spent elsewhere, and it is often cheaper for the state to buy at scale than for a household to buy alone. The objections are that it removes the household's own judgement about what it needs most, and that targeting requires means testing, which costs money to administer, misses people who do not apply, and carries stigma.

Universal basic income. A fixed payment made to every citizen, regardless of income, wealth or whether they work. The case for it: nobody falls through the gaps, nothing is withdrawn as you earn so the work incentive is not destroyed, there is no means test to administer and no stigma in claiming, and it fits an economy where work is increasingly insecure. The case against it: paying everybody is extremely expensive, so either the payment is too small to lift anybody out of poverty or the tax rise needed to fund it is very large; money goes to people who do not need it unless it is clawed back through tax; some argue labour supply would fall; and it may replace targeted support that served particular needs better. Trials have been small and short, so the evidence on a full national scheme is thin. This is a live argument, not a settled one, and an answer that treats it as either obviously right or obviously foolish is not evaluating.

Policies to reduce discrimination. Equal pay legislation, laws against discrimination in hiring and promotion, requirements to publish pay gaps, parental leave available to both parents, and quotas or targets for under-represented groups. They address a cause that markets do not correct on their own. The objections are that the law reaches formal employment much more easily than informal work, that enforcement is difficult and slow, and that quotas are themselves contested on grounds of fairness.

Minimum wages. A legal floor under the hourly wage. Figure 13 is the standard analysis.

In a competitive labour market, a floor set above the equilibrium wage W* means Qs workers want jobs at that wage and only Qd are hired, so the gap Qs − Qd is excess supply of labour, which is unemployment. Everyone still employed earns more, which is the point of the policy; those who lose their jobs, or never get hired, are worse off, and they tend to be the least experienced workers — exactly the group the policy meant to help.

That is the model, and you must be able to draw it. It is not the last word, and saying so earns evaluation marks. The prediction depends on the assumption that the labour market is competitive and that the wage already equals what the work is worth to the employer. Where a small number of large employers dominate hiring in a town or an occupation, they may be paying less than that, and a floor can raise pay without cutting employment. Studies of modest increases have found effects on employment ranging from small negative to negligible, while large increases relative to local wages look riskier. So the defensible judgement is conditional: the size of the increase relative to typical wages in that labour market matters more than the existence of a minimum wage.

Judging the set. Two trade-offs run through all six. The first is equity against efficiency: taxes and benefits that compress the distribution also change what people gain from working, saving and investing, though how much they change it is disputed. The second is targeting against simplicity: money aimed precisely at the poorest goes furthest per pound but costs more to administer and misses people, while universal payments reach everybody and spend a great deal on households that do not need it. A part (b) answer that names the trade-off it is choosing, and says what its choice depends on, is doing what evaluation means.