How is Unemployment Measured? Approaches and Challenges in Developing Countries

Unemployment numbers shape policy, budgets, and the lives of millions, yet measuring joblessness is surprisingly tricky. Who counts as unemployed? Someone who hasn’t worked for a year, or just for a week? What about a farmer who works only during the harvest season? The answers depend on which framework you use, and in a country like India, where most work happens outside formal contracts, different methods can give wildly different results.

Why measuring unemployment is harder than it looks 🔗

At first glance, counting the unemployed seems simple: just add up everyone without a job. But this overlooks a critical detail. A student, a retired person, or a homemaker may not have a paid job, yet none of them are “unemployed” in the economic sense. Unemployment statistics only count people who are part of the labour force, meaning those who are either working or actively looking for work.

The unemployment rate itself is calculated using a simple formula: the number of unemployed persons divided by the total labour force, multiplied by 100. The challenge lies in deciding who qualifies as “unemployed” and over what time period. Different definitions can dramatically change the final number, which is why standardised methods exist.

The ILO’s global approach to unemployment 🔗

The International Labour Organization (ILO) sets the global benchmark for measuring unemployment. Its definition is used by most countries to ensure data is comparable across borders. According to the ILO’s operational definition, a person must satisfy three conditions simultaneously to be classified as unemployed.

The three criteria of unemployment 🔗

An individual is counted as unemployed only if they meet all of the following conditions during a specific reference period:

Without work: The person did not engage in any paid employment or self-employment during the reference week, not even for an hour.

Currently available for work: The person is ready to take up a job or start a business in the reference week or within a short subsequent period, typically up to two weeks.

Actively seeking work: The person has taken specific steps in a defined recent period (usually the last four weeks) to find paid employment or self-employment. A general statement of wanting to work is not enough; concrete job-search activity is required.

As the ILO guidance on analysing unemployment makes clear, all three criteria must be met at the same time. If even one is missing, the person is classified differently, either as employed or as outside the labour force.

Age range and the working-age population 🔗

The ILO recommends measuring unemployment only among the working-age population. Most countries set this range between 15 and 64 years, though the upper limit may vary depending on retirement norms. In India, the Periodic Labour Force Survey (PLFS) typically reports indicators for persons aged 15 years and above.

How India measures unemployment: the NSSO framework 🔗

India has developed its own approaches that build on the ILO framework but address the country’s specific labour market realities. The National Sample Survey Office (NSSO), now operating under the National Statistical Office, uses three distinct approaches recommended by the expert committee headed by Professor M.L. Dantwala. Each method uses a different reference period and captures a different aspect of joblessness.

Usual status approach 🔗

The Usual Status approach uses the longest reference period of 365 days preceding the date of survey. It captures long-term or chronic unemployment. A person is classified as unemployed only if they had no gainful work for the major part of the year and were seeking or available for work.

This approach has two components. The Usual Principal Status (UPS) looks at the activity in which the person spent the longest part of the year. The Subsidiary Status (SS) picks up any economic activity the person pursued for 30 days or more during the same year, in addition to their principal activity. Together they form the Usual Status (ps+ss), which gives the broadest picture of who was employed at some point during the year.

Because the reference period is so long, this approach tends to produce the lowest unemployment rate. Anyone who managed to find some kind of work for a substantial part of the year is counted as employed, even if they were jobless for several months.

Current weekly status approach 🔗

The Current Weekly Status (CWS) approach narrows the reference period to the seven days preceding the date of survey. A person is considered employed if they worked for at least one hour on at least one day during this week. If no economic activity was pursued but the person was seeking or available for work, they are classified as unemployed.

If a person engaged in multiple activities during the week, the activity status is decided using a priority-cum-major time criterion, where being employed takes priority over being unemployed, which in turn takes priority over being outside the labour force. This approach captures seasonal unemployment better than the usual status approach.

Current daily status approach 🔗

The Current Daily Status (CDS) approach is considered an Indian innovation. Instead of assigning a single status for the whole week, it records the activity status for each day of the reference week, treating each day as either two “half days” or one “full day”.

The rules are precise. A person working four hours or more on a day is recorded as employed for the full day. Someone working between one and four hours is counted as employed for half a day. A person with no gainful work even for an hour is recorded as unemployed for the full day.

This method captures even short spells of unemployment within a working week and is especially useful in sectors like agriculture, where work fluctuates daily. As a result, CDS usually shows the highest unemployment rate among the three approaches, because it counts intermittent and seasonal joblessness that the other methods miss.

Why the three numbers differ so much 🔗

Because each approach uses a different reference period, the same population can show very different unemployment rates depending on which method is applied. According to data discussed by the Centre for Economic Data and Analysis at Ashoka University, in 2022-23 the unemployment rate in India ranged from about 2.1 percent under the usual status to roughly 5.5 percent under the daily status. The labour force participation rate also varied from 61.6 percent (usual status) to 54.1 percent (daily status).

This is why analysts examine all three figures together rather than relying on just one. Each tells a different story: usual status captures chronic joblessness, weekly status captures shorter spells, and daily status reveals the seasonality and irregularity of work.

The Periodic Labour Force Survey: a modern upgrade 🔗

Until 2017, the NSSO measured employment and unemployment through quinquennial (every five years) household surveys. To provide more frequent data, the Periodic Labour Force Survey (PLFS) was launched in April 2017. The PLFS generates quarterly estimates for urban areas using the Current Weekly Status, and annual estimates for both rural and urban areas using both Usual Status (ps+ss) and CWS.

This shift was significant. With quarterly data, policymakers can now track changes in the labour market in near real time rather than waiting half a decade. The PLFS is also India’s first computer-based survey on this scale, improving both accuracy and speed of reporting.

Why measuring unemployment is so difficult in developing economies 🔗

Even with these refined methods, accurate measurement remains a serious challenge in countries like India. The standard ILO framework was designed with the labour markets of industrialised economies in mind, where most people hold formal, full-time jobs. Conditions on the ground in developing nations are very different.

The dominance of the informal sector 🔗

The single biggest complication is the size of the informal sector. According to Britannica, more than 90 percent of India’s workforce is employed in the informal sector, which operates outside formal regulation and largely escapes the tax base. Informal workers rarely have written contracts, fixed working hours, or social security coverage.

This creates measurement headaches. A street vendor who sells vegetables for two hours a day, a daily-wage labourer who finds work three days a week, or a domestic worker employed in multiple households all qualify as employed by ILO standards, even if they earn very little and want more work. Their situation is closer to underemployment than full employment, but the unemployment rate does not capture this distinction.

Disguised unemployment in agriculture 🔗

Indian agriculture has long suffered from disguised unemployment, a situation where more people work on a piece of land than are actually needed. The output would not fall if some of these workers left, yet they are all formally counted as employed. Standard ILO and NSSO definitions cannot easily separate productive employment from this kind of hidden joblessness, which is why additional indicators on underemployment are also reported.

Seasonal work and fluid employment status 🔗

In agriculture, construction, and many service jobs, work is highly seasonal. A farm worker may be busy during sowing and harvesting but idle for months in between. The Usual Status approach would call this person employed for the whole year, while the Current Daily Status would record many days of unemployment. Without using multiple measures, the true picture of seasonal joblessness remains hidden.

Women, social norms, and the discouraged worker problem 🔗

Another challenge is the treatment of people who want work but have stopped looking. The ILO definition requires active job search, so someone who has given up after months of failure is classified as outside the labour force, not as unemployed. In India, this affects women disproportionately. Many women who would take a job if one were offered nearby do not actively search because of social restrictions, lack of safe transport, or family responsibilities. They are statistically invisible, even though they represent a significant pool of potential workers.

The mismatch between data and reality 🔗

Put together, these factors mean that headline unemployment rates can look surprisingly low even when distress is widespread. As research published in Heliyon highlights, in developing economies the informal sector functions as the primary job provider, often absorbing workers who would otherwise be unemployed, but doing so at very low productivity and income levels. This is one reason why analysts now increasingly look at additional indicators such as the labour force participation rate, the worker population ratio, the quality of employment, and the share of informal versus formal jobs alongside the unemployment rate itself.

Reading unemployment data with a critical eye 🔗

Numbers about jobs are never just numbers. They depend on definitions, reference periods, and the realities of the economy being measured. A 3 percent unemployment rate in a country where 90 percent of work is informal means something very different from a 3 percent rate in an economy of full-time, formal jobs. Understanding what is being counted, and what is being missed, is essential to interpreting any labour market statistic.

For students and researchers in population studies, this is where the topic becomes most interesting. Unemployment is not just an economic indicator; it is a window into how a society organises work, who gets recognised as a worker, and whose contributions remain invisible in official statistics.

What do you think? If you had to design a single unemployment measure for India, which features of the labour market would you prioritise capturing, and which trade-offs would you accept? Should “want of work” alone be enough to count someone as unemployed, even without an active job search?

References

1. https://www.ilo.org/media/445251/download
2. https://www.ilo.org/media/351701/download
3. https://prepp.in/news/e-492-measurement-of-unemployment-indian-economy-notes
4. https://www.pib.gov.in/PressReleasePage.aspx?PRID=1966154
5. https://ceda.ashoka.edu.in/daily-status-weekly-status-usual-status-why-labour-force-participation-rates-vary-by-reference-period/
6. https://www.britannica.com/topic/socioeconomic-challenges-in-India
7. https://www.cell.com/heliyon/fulltext/S2405-8440(24)09409-X