1The idea in one paragraph

A government wants four things at once: output growing, few people out of work, prices rising slowly and steadily, and a sustainable level of public debt. Each of those has a way of being measured, a set of causes, and a set of costs when it goes wrong, and that is most of what this subtopic contains. The last part is the part that carries the evaluation marks: the four objectives pull against each other. Spending to put people back to work pushes prices up. Growing fast burns fuel and widens the gap between rich and poor. There is no setting of the controls that gives you all four, so a government chooses, and your job in an exam is to say what it gave up.

2Economic growth: actual and potential

Economic growth is an increase in real output over time. Two very different things are called growth, and telling them apart is the first thing examiners look for.

Actual growth is a rise in output that the economy was already capable of producing. In the production possibilities curve model it is a movement from a point inside the curve towards it, as in the left panel of Figure 1. Nothing has been added to the economy. Idle machines have been switched on and unemployed workers have been hired, so the resources that were sitting there are now being used.

Growth in production possibilities, also called potential growth, is a rise in what the economy is capable of producing at all. On the PPC it is a shift of the whole curve outwards, as in the right panel. That happens when the quantity or the quality of the factors of production rises: more workers, better-trained workers, more capital, better technology, newly discovered resources.

The same two ideas appear in the AD/AS model, and the guide asks for both diagrams.

The role of AD. When aggregate demand rises and the economy has spare capacity, firms meet the extra spending by producing more, so real output rises. Figure 2 shows AD₁ shifting to AD₂ along an upward-sloping SRAS, with real output rising from Y₁ to Y₂. Notice two things. Output rises, which is the growth. And the price level rises too, from PL₁ to PL₂, which is the cost. That second point comes back in section 14.

The role of LRAS. Long-run growth is an outward shift of LRAS, drawn in Figure 3. Full-employment output rises from Yf₁ to Yf₂, so the economy can now produce more without running into shortages. With AD unchanged, real output rises and the price level falls. Growth from the supply side is the only growth that does not push prices up, which is why supply-side policy is the answer to almost every "how can the government have both?" question.

Measuring growth. The rate of economic growth is the percentage change in real GDP from one year to the next.

growth rate = ((real GDP this year − real GDP last year) ÷ real GDP last year) × 100

Take Tavene. Real GDP, measured at constant prices, was 480 billion in year 1 and 494.4 billion in year 2.

growth rate = ((494.4 − 480) ÷ 480) × 100 = (14.4 ÷ 480) × 100 = 3.0%

That number flatters Tavene, because its population also grew, from 20.00 million to 20.24 million. Real GDP per capita is real GDP divided by population: 24,000 in year 1 and 24,427 in year 2, a rise of 1.8%. Output grew by 3.0%, but output per person grew by 1.8%, and it is output per person that tells you whether the average Tavenen is better off. If population had grown faster than 3.0%, output per person would have fallen while the headline growth rate was still positive.

3What growth does: living standards, the environment, income distribution

The guide asks you to evaluate three consequences, which means saying "it depends" and then saying what it depends on.

Living standards. More output means more goods and services per person, more tax revenue for schools and hospitals, and more jobs. That is the strong case, and it is usually right. But real GDP per capita is an average. It says nothing about how long people worked for it, whether the extra output was weapons or vaccines, or how much of it went to a small number of households. A country whose output per person doubles while its working hours rise by half and its air becomes unbreathable has grown without its people living better.

The environment. Most growth so far has been fuelled by burning things and by extracting resources faster than they replace themselves. That raises output today and lowers what the economy can produce later, which is the definition of unsustainable. The counter-argument matters and earns marks: rich countries can afford cleaner technology and stricter regulation, so growth is what pays for the clean-up. Whether growth harms the environment depends on the kind of growth, not on growth itself.

Income distribution. Growth does not arrive evenly. It usually rewards the owners of capital and the workers with skills that the new output needs, and it can leave behind workers whose industry is shrinking. Growth can therefore widen the gap between rich and poor even while it lifts everyone's absolute income. Whether it does depends on the tax and benefit system: the same growth in two countries with different governments produces different distributions.

4Unemployment: measuring it, and why the number lies

A person is unemployed if they are of working age, without a job, available to start work, and actively looking for work. All four conditions must hold. Someone who is not looking is not unemployed, however much you might think they should be.

The labour force is everyone employed plus everyone unemployed. It is not the whole population and it is not the working-age population; a retired person or a full-time student is in neither part of it. Figure 4 sorts everyone into the right box.

unemployment rate = (number unemployed ÷ labour force) × 100

In Kessia, a separate and larger economy, the working-age population is 40 million. Of those, 28.5 million are employed and 1.5 million are unemployed, so the labour force is 30 million.

unemployment rate = (1.5 ÷ 30.0) × 100 = 5.0%

Difficulties of measuring unemployment. The 5.0% is an underestimate, an overestimate and an average all at once.

• Hidden unemployment. Kessia also has 0.4 million discouraged workers: people who want a job but have given up looking, so they are outside the labour force and outside the figures. Count them and the rate becomes (1.9 ÷ 30.4) × 100 = 6.25%. A recovery that brings discouraged workers back to searching can make the unemployment rate rise.
• Underemployment. A part-time worker who wants full-time hours, and a trained engineer driving a taxi, both count as fully employed. The waste of their skills does not show up anywhere.
• The informal economy. Some people counted as unemployed are working and being paid in cash, which pushes the figure the other way.
• It is a national average. One national rate of 5% can hide a 3% rate in the capital and a 20% rate in a region whose factory has closed, or a 4% rate for workers over thirty and a 15% rate for school leavers. Every policy conclusion drawn from the average is wrong for somebody.
• Definitions differ. Countries count differently, so international comparisons are shakier than the tidy table in the data booklet suggests.

5The four causes of unemployment, and what cures each

There are four causes on the syllabus, and each one needs a different response. Figure 5 is the map; the diagrams that follow are the three the guide asks you to draw.

Cyclical unemployment, also called demand-deficient unemployment, is caused by a fall in aggregate demand. When spending falls, firms across the whole economy sell less, produce less and need fewer workers. Figure 6 shows it as a deflationary gap: equilibrium output Yₑ sits below full-employment output Yf, and the horizontal distance between them is output the economy could have produced but did not. The workers who would have produced it are the cyclically unemployed. This is the only type that vanishes when AD recovers, and the only type that demand-side policy can treat.

Structural unemployment is caused by a permanent change in the pattern of demand or in technology, so that the jobs that exist no longer match the workers who are available. It is specific to an industry or a place, not general. Figure 7 draws it in one labour market: demand for labour falls from DL₁ to DL₂. If the wage were free to fall, the market would clear at a lower wage. Wages are sticky, so at the old wage W₁ firms want only Qd workers while Q₁ are still there, and the difference is out of work. Retraining and help with relocation are the cures, and both are slow and expensive. A structurally unemployed worker whose skill has no buyer is not helped by a tax cut.

Wages can also be held above the market level by an institution rather than by habit. Figure 8 shows a minimum wage set above the equilibrium wage Wₑ. At Wmin, firms wish to hire Qd workers and Qs workers offer their labour, so the horizontal gap Qd to Qs is unemployment created by the wage floor. Two honest sentences of evaluation belong beside this diagram: the workers who keep their jobs are better paid, and how many jobs are actually lost depends on how responsive labour demand is to wages, which is an empirical question rather than a diagram.

Frictional unemployment is the unemployment of people between jobs: a graduate looking for a first post, a worker who has quit one job and not yet started another. It is short term and it is the sign of a working labour market, not a broken one. Better job information shortens it; nothing abolishes it.

Seasonal unemployment is caused by demand for particular work rising and falling with the time of year: ski instructors in July, fruit pickers in February. It is predictable and it is mostly untreatable, other than by helping a local economy depend on more than one season.

Match the cure to the cause. Only cyclical unemployment answers to demand-side policy; everything else needs supply-side work, and supply-side work is slow.

6The natural rate, and what unemployment costs

The natural rate of unemployment is the unemployment that remains when the economy is at full employment, so it is the sum of the structural, seasonal and frictional unemployment. It is not zero and it is not a target of zero. An economy sitting exactly at its full-employment level of output still has people retraining, people between jobs and people waiting for the season to start. Cyclical unemployment is the part that sits on top of the natural rate, and full employment means cyclical unemployment of zero, not unemployment of zero.

The costs divide three ways, and a question that asks for "the costs of unemployment" wants all three.

Personal costs fall on the unemployed person: lost income, lost savings, skills that decay while they are not used, and the well-documented damage to physical and mental health. Long spells make the next job harder to get, so the cost compounds.

Social costs fall on everyone: higher rates of crime and family breakdown in areas of long-term unemployment, and the loss of social cohesion in a town whose main employer has gone.

Economic costs fall on the economy as a whole: output that is never produced, so the economy operates inside its PPC; lost tax revenue at the same time as higher spending on unemployment benefits, which worsens the government's budget; and a poorer distribution of income, because unemployment is concentrated among people who already had least.

7Inflation, and measuring it with the consumer price index

Inflation is a sustained increase in the general price level. Two words carry the weight. Sustained, so a one-off jump is not inflation. General, so a rise in the price of one good is not inflation; that is a relative price change and it belongs in Unit 2.

The consumer price index (CPI) measures the price of a basket of goods and services bought by a typical household, weighted by how much of each the household buys. The weights are what make it an index of the cost of living rather than an average of unrelated prices: a 10% rise in the price of housing matters far more to a household than a 10% rise in the price of shoelaces, and the weights say so.

Here is Tavene's basket, with base-year quantities used as the weights.

The base year is given an index of 100. Every other year's index is that year's basket cost as a percentage of the base year's.

CPI = (cost of the basket this year ÷ cost of the basket in the base year) × 100

CPI in year 2 = (555 ÷ 500) × 100 = 111.0, and since the base year index was 100, inflation in year 2 was 11.0%.

Once you have a series of index numbers, the inflation rate between any two years is the percentage change in the index:

inflation rate = ((CPI this year − CPI last year) ÷ CPI last year) × 100

Limitations of the CPI. It is the best measure available and it is still wrong in five known ways.

• The basket goes out of date. It is fixed for years at a time, so it keeps measuring what households used to buy while their spending moves on.
• It ignores changes in quality. A phone that costs the same as last year's but does twice as much has become cheaper in every sense except the one the CPI records.
• It is an average household's basket. A pensioner who heats a house and buys no petrol faces a different inflation rate from a commuter, and neither is the published figure.
• It excludes some prices that matter. House purchase prices and interest payments are treated differently in different countries, and the choice moves the number.
• International comparison is unsafe, because baskets, weights and methods differ from country to country.

9Demand-pull and cost-push inflation

Prices can be pushed up from either side of the market, and the guide wants a separate diagram for each.

Demand-pull inflation is caused by an increase in aggregate demand. Households, firms, the government or foreign buyers want to spend more than the economy can comfortably produce, so prices are bid up. In Figure 9, AD₁ shifts to AD₂ along an upward-sloping SRAS: the price level rises from PL₁ to PL₂ and output rises from Y₁ to Y₂. The closer output is to full employment, the steeper SRAS becomes, so the same increase in AD delivers more inflation and less extra output. Causes to name: a rise in consumer confidence, lower interest rates, tax cuts, higher government spending, a rise in exports, a depreciating currency.

Cost-push inflation is caused by an increase in the costs of production, which makes firms willing to supply less at every price level. In Figure 10, SRAS₁ shifts left to SRAS₂: the price level rises from PL₁ to PL₂ and real output falls from Y₁ to Y₂. Rising prices together with falling output is called stagflation, and it is what makes cost-push the harder problem: raising AD to rescue output makes the inflation worse, and cutting AD to stop the inflation makes the recession worse. Causes to name: higher oil or raw material prices, wage rises not matched by productivity, a depreciating currency raising import costs, higher indirect taxes.

Reading the diagram backwards is a reliable exam skill. If prices and output moved the same way, it was AD. If they moved in opposite directions, it was SRAS.

10Disinflation and deflation

Three words, and students swap them constantly.

• Inflation is the price level rising.
• Disinflation is the price level still rising, but more slowly than before. The inflation rate falls; prices do not.
• Deflation is the price level actually falling. The inflation rate is negative.

Figure 11 puts all three on one line. Tavene's CPI runs 100.0, 111.0, 114.8, 116.0, 114.8 over five years, giving inflation of 11.0%, then 3.4%, then 1.0%, then −1.0%. Years 3 and 4 are disinflation: the line is still climbing, just less steeply, and prices in those years were the highest Tavene had ever seen. Only in year 5, when the line turns down, is there deflation.

The causes of deflation are the two mirror images of the causes of inflation, and the guide wants both, because they are not equally bad.

Figure 12 draws them side by side. On the left, aggregate demand falls: the price level falls and real output falls with it. That is deflation caused by a collapse in spending, and it arrives with a deflationary gap and cyclical unemployment. On the right, short-run aggregate supply rises, from better technology or cheaper raw materials: the price level falls and real output rises. That deflation is a symptom of an economy getting better at producing things.

An answer that says "deflation is bad" without asking which panel it is in has thrown away the evaluation marks.

11What inflation costs, what deflation costs, and which is worse

Costs of a high rate of inflation. The guide names six.

• Uncertainty. Firms cannot forecast costs and revenues, so they postpone investment, and less investment today means slower growth in LRAS tomorrow.
• Redistributive effects. Inflation takes from lenders and gives to borrowers, because a loan is repaid in money worth less than the money that was lent. It takes from anyone on a fixed income, such as a pension that is not index-linked, and from workers with weak bargaining power, while those who can negotiate keep up.
• Effects on saving. If the interest rate on savings is below the inflation rate, the real return is negative and saving falls. Less saving means less funding for investment.
• Damage to export competitiveness. If Tavene's prices rise faster than its trading partners' prices, its exports look dearer abroad and imports look cheaper at home, so net exports fall.
• Impact on economic growth. The three effects above run together: less investment, less saving, weaker exports, and therefore slower growth.
• Inefficient resource allocation. Prices are meant to be signals. When every price is rising, firms cannot tell a genuine rise in demand for their product from general inflation, so resources are allocated to the wrong places.

Costs of deflation. The guide names seven, and this is the longer list because deflation is harder to escape.

• Uncertainty, in the same way, and worse, because falling prices are rarer and less understood.
• Redistributive effects, running the other way: deflation takes from borrowers and gives to lenders.
• Deferred consumption. If a washing machine will be cheaper next month, households wait. Waiting cuts AD, which pushes prices down further, which is a reason to wait again. This is the deflationary spiral.
• High cyclical unemployment and bankruptcies. Demand-side deflation arrives with a deflationary gap, so firms fail and workers lose jobs.
• An increase in the real value of debt. A debt is fixed in money terms. If prices and wages fall, the debt does not, so the burden of every existing loan rises. Households and firms cut spending to service it, which deepens the fall in AD.
• Inefficient resource allocation, for the same reason as under inflation: the signal is drowned out.
• Policy ineffectiveness. Interest rates cannot fall far below zero, so a central bank facing deflation runs out of its main instrument at the moment it needs it most. There is no equivalent ceiling on the way up.

Relative costs of unemployment versus inflation. A judgement is asked for, so make one and defend it. The case for unemployment being worse: its costs fall heavily on a small group of identifiable people who lose income, health and skills, while moderate inflation is spread thinly and is partly compensated by index-linked wages and benefits. The case for inflation being worse: it damages everybody at once, it destroys the value of savings, and once expectations of inflation set in it is very expensive to remove, usually by deliberately creating the unemployment you were trying to avoid. Most economists rank high unemployment as the more urgent harm and high inflation as the more corrosive one, and the honest answer depends on the size of each: 3% inflation with 12% unemployment is not the same problem as 25% inflation with 4% unemployment.

13When the objectives conflict

This is where part (b) marks live. Each conflict below is a real one, and each has a way out that is worth naming.

Low unemployment and low inflation. Cutting unemployment usually means raising aggregate demand. Figure 2 already showed what that does: output rises and so does the price level. The nearer the economy is to full employment, the steeper SRAS is, so the more of the extra demand comes out as inflation rather than output. The way out is supply-side: policies that raise LRAS, such as training and investment in infrastructure, cut unemployment without raising the price level, as Figure 3 showed. They are slow, and they cost money now for output later.

High economic growth and low inflation. Demand-driven growth carries inflation with it for exactly the same reason. Supply-driven growth does not — Figure 3 shows the price level falling as output rises. The evaluation point is that governments reach for demand-side policy because it works within an electoral cycle, and supply-side policy does not.

High economic growth and environmental sustainability. Growth that comes from burning more fuel and cutting more forest raises output now and lowers the economy's future productive capacity, so the PPC that shifts out today shifts back in later. The counter-argument: growth funds the research and the regulation that clean production up, and poor countries cannot afford to care about emissions until they are richer. Both halves are needed for a balanced answer.

High economic growth and equity in income distribution. Growth rewards those with capital and with the skills the new industries want, so it can widen the income gap while raising everyone's income. Redistribution through progressive taxation and benefits narrows the gap, but high marginal tax rates may weaken the incentive to work and invest, which slows the growth being redistributed. Where the balance lies is a value judgement about equity, and saying so is worth a mark.