Macroeconomics › Section S · Macroeconomic policy around the world › Topic S.3

Causes of unemployment around the world

An unemployment rate of 5% can mean a mild recession in one country and a labour market working normally in another, while a rate of 3% can hide millions of people scraping by. This topic sorts out the causes of joblessness in rich, middle-income and poor economies, and the policies that fit each.

One word, different problems

The International Labour Organization (ILO) projected in January 2026 that world unemployment would stay at about 4.9% in 2026, about 186 million people, and estimated youth unemployment at 12.4% in 2025. Those totals hide very different situations. In high-income economies, unemployment usually means a worker with a job history searching for another. In low-income economies, few people can afford to search at all, so the real problem is often the quality of the work they do. Policy has to start from the right diagnosis, and the diagnosis differs from country to country.

Rich economies: cyclical or natural

In high-income economies, unemployment has two broad sources, covered in the topics "Short-run causes: cyclical unemployment" and "Long-run causes: the natural rate". Cyclical unemployment comes from recessions, when aggregate demand falls short of potential output. The usual remedies are an expansionary monetary policy, which cuts interest rates, and letting automatic stabilisers, such as falling tax receipts and rising benefit payments, widen the budget deficit. Extra discretionary tax cuts or spending make most sense in a deep slump, because they take time to agree and put in place. Even after growth returns, firms often wait months before hiring again.

Why natural rates differ

The natural rate depends on labour-market rules and institutions, so it differs across rich countries. ILO modelled estimates, published by the World Bank, put 2025 unemployment at about 4.2% in the United States, 3.7% in Germany, 2.5% in Japan, 7.5% in France and 10.4% in Spain, where youth unemployment was about 25%. Economists link higher natural rates to strict and costly rules on dismissal, which make firms cautious about hiring; generous benefits with weak job-search support; high taxes on labour; and wage bargaining that keeps pay above market levels. Each rule may have a sound purpose, such as security for workers, but together they can deter hiring.

Worked example: a 50-worker rule

In a hypothetical country, firms with 50 or more workers must meet extra rules costing $400,000 a year. Each worker adds $70,000 a year to output and is paid $60,000. 1. A firm with 49 workers gains \(70{,}000 - 60{,}000 = 10{,}000\) dollars a year from each extra worker, ignoring the rule. 2. Hiring the 50th worker triggers the rule: \(10{,}000 - 400{,}000 = -390{,}000\), a loss. 3. Growing past the line breaks even only when the extra workers' gains cover $400,000, which takes \(400{,}000 \div 10{,}000 = 40\) more workers, a jump from 49 to 89. 4. Interpret: many firms stop at 49. Research by Luis Garicano, Claire Lelarge and John Van Reenen, published in 2016, found this kind of bunching just below 50 workers in France.

Low unemployment, poor jobs

In poor countries measured unemployment is often low. ILO modelled estimates put it at about 4.9% across low-income economies in 2025 and about 3.1% in Nigeria. That is not a sign of healthy labour markets. Where there are no unemployment benefits, most people cannot afford to stay without work, so they farm, fish, trade on the street or take a day's casual work, sometimes paid in food or shelter. Many are underemployed, working fewer hours or at lower productivity than they would like. The ILO estimates that nearly 300 million workers worldwide live in extreme poverty, on less than $3 a day.

The informal economy

Informal work is work outside the reach of labour law and social protection: no written contract, no sick pay, no pension and no unemployment insurance. The ILO expects about 2.1 billion people to work informally in 2026. Its estimates for 2024 put informal employment at about 58% of all workers worldwide, about 88% in low-income economies, about 82% in lower-middle-income economies and about 13% in high-income economies. Because informal workers rarely specialise or use much capital, their productivity is low. Development economists find that moving into a regular wage-paying job is one of the surest routes out of extreme poverty.

The Lewis dual-sector model

In 1954 the economist W. Arthur Lewis, later a Nobel laureate, described a developing economy with two sectors. In traditional agriculture, so many people work a fixed amount of land that an extra worker adds almost nothing to output: there is surplus labour. In the modern sector of factories and services, each new worker adds much more. Moving workers from farm to factory therefore raises total output, even if factory wages stay low for a long time. When the surplus is used up, at what is called the Lewis turning point, wages in both sectors start to rise. Economists dispute whether China has passed that point.

Worked example: moving surplus workers

In a hypothetical economy, an extra farm worker adds $300 a year to output, while a factory worker adds $3,000. Ten million workers move from farms to factories. 1. Output lost on the farms: \(10{,}000{,}000 \times 300 = 3\) billion dollars. 2. Output gained in factories: \(10{,}000{,}000 \times 3{,}000 = 30\) billion dollars. 3. Net gain: \(30 - 3 = 27\) billion dollars a year. 4. Check the assumption: the gain is this large only while farm workers' extra output stays near zero; past the Lewis turning point, each move costs the farms more. 5. Note the costs: rural areas can lose their young people, and cities must find them housing.

South Africa: a structural problem

South Africa shows that high unemployment is not only a rich-country or recession problem. ILO modelled estimates put its unemployment rate at about 32% in 2025 and youth unemployment at about 60%. Economists usually describe this as structural, not cyclical. Explanations they offer include gaps in skills and work experience, long distances between townships and jobs that date from apartheid-era planning, and a small informal sector compared with other middle-income countries. Policies discussed include apprenticeships, wage subsidies for young workers, better public transport and lower barriers to starting small businesses.

Common mistakes to avoid

First, reading a low unemployment rate in a poor country as proof of a strong labour market; informality and underemployment can be widespread. Second, treating all unemployment as cyclical: stimulus cannot cure a high natural rate caused by labour-market rules or skills gaps. Third, assuming job-protection rules always protect jobs; costly dismissal can mean fewer hires. Fourth, forgetting that in the Lewis model the gain comes from the difference in productivity between sectors, not from higher wages. Fifth, comparing countries' rates without checking that they use the same definition.

Key takeaways

The ILO projected world unemployment at about 4.9% in 2026, about 186 million people. In rich economies, cyclical unemployment calls for monetary policy and automatic stabilisers, while a high natural rate reflects labour-market rules such as costly dismissal. In poor economies measured unemployment is often low, but informal work, about 88% of employment in low-income economies, and underemployment are the real problems. The Lewis model shows how moving surplus farm labour into modern jobs raises output. South Africa's high unemployment is largely structural.

Flashcards

Flashcard

World unemployment, 2026

About 4.9%, about 186 million people

ILO projection, January 2026.

Test your understanding

Question 1 of 5

A low-income country reports an unemployment rate of 3%. What is the best interpretation?

Question 2 of 5

Which policy is most likely to lower a high natural rate of unemployment caused by labour-market rules?

Question 3 of 5

A rule costs firms $200,000 a year once they reach 50 workers. Each worker adds $8,000 a year more than their wage. What is the smallest size above 49 at which crossing the line at least breaks even?

Question 4 of 5

In the Lewis dual-sector model, why does moving workers from farms to factories raise total output?

Question 5 of 5

Five million workers move from farms, where each adds $200 a year, to factories, where each adds $2,200. What is the net gain in output?

← Previous topicNext topic →