OECD monthly unemployment rate fell slightly to 6.0 % in August 2021

The OECD area monthly unemployment rate fell for the fourth consecutive month, albeit slightly, in August 2021, to 6.0 % (from 6.1 % in July). While the rate remains 0.7 percentage point above the pre-pandemic rate as at February 2020[1], this latest update represents a continuation of the downward trend since the peak of 8.8 % in April 2020, with the exception of April 2021. However, some care is needed in interpreting the fall in the OECD area unemployment rate when compared with the April 2020 peak, as it largely reflects the return of temporary laid-off workers in the United States and Canada, where they are recorded as unemployed.[2] The number of unemployed workers across the OECD area continued to fall (by 1.0 million) in August 2021, reaching 39.7 million, still 4.3 million above the level in February 2020.

In the Euro Area, the unemployment rate reached 7.5 % in August 2021, down slightly from 7.6 % in July, getting closer to pre-pandemic rates. There were falls of 0.3 percentage point or more in Greece (to 13.2 %, from 14.2 % in July), Finland (to 7.2 %, from 7.8 %), Spain (to 14.0 %, from 14.5 %), Latvia (to 7.1 %, from 7.5 %) and Lithuania (to 7.2 %, from 7.5 %).

Outside the euro area, the largest decreases in unemployment rates in August (from July) were observed in Colombia (to 12.7 %, from 13.6 %), Korea (to 2.8 %, from 3.3 %), Canada (to 7.1 %, from 7.5 %) and the United States (to 5.2 %, from 5.4 %), while slight changes or no changes were observed in most other countries, including Australia (at 4.5 %, from 4.6 %), Japan (2.8 %), Mexico (at 4.1 %, from 4.2 %), Poland (at 3.4 %) and Turkey (at 12.1 %). More recent data show that in September 2021 the unemployment rate declined further in the United States (by 0.4 percentage point, to 4.8 %) and Canada (by 0.2 percentage point, to 6.9 %).

In August 2021, while the OECD area unemployment rate declined among people aged 25 and over (to 5.1 %, from 5.3 % in July), it increased slightly (to 12.5 %, from 12.4 %) among youth (people aged 15 to 24), after 3 months of decline. There remain large differences across OECD countries in youth unemployment rates. For example, among the European economies, Spain (33.0 %) and Greece (30.8 %) show very high youth unemployment rates, compared with the Netherlands (7.4 %) and Germany (7.5 %).

OECD area unemployment rates for both women and men declined slightly in August to 6.2 % (from 6.3 % in July) and 5.8 % (from 5.9 %), respectively. However, declines of more than 1.0 percentage point in women unemployment rates were observed in Colombia (to 16.6 %), Finland (to 6.5 %) and Greece (to 15.1 %).

It should be noted that unemployment statistics do not account for the full amount of labour market slack due to the pandemic, as some non-employed people may be classified as “out of the labour force”, because they are either not able to actively look for a job or are not available to work.

Note: Employment and unemployment statistics during the COVID-19 crisis

The broad comparability of unemployment data across OECD countries is achieved through the adherence of national statistics to International Guidelines from the International Conference of Labour Statisticians (ICLS) – the so-called ILO guidelines.

Departures from these guidelines may however exist across countries depending on national circumstances (e.g. statistical environment, national regulations and practices). Typically, these departures have only a limited impact on broad comparability of employment and unemployment statistics. However, the unprecedented impact of Covid-19[1] is amplifying divergences and affects the cross-country comparability of unemployment statistics in this news release.

This concerns in particular the treatment of persons on temporary layoff or employees furloughed by their employers. These are persons not at work during the survey reference week due to economic reasons and business conditions (i.e.lack of work, shortage of demand for goods and services, business closures or business moves).

According to ILO guidelines, ‘employed’ persons include those who, in their present job, were ‘not at work’ for a short duration but maintained a job attachment during their absence (ILO, 2013 and 2020). Job attachment is determined on the basis of the continued receipt of remuneration, and/or the total duration of the absence. In practice, formal or continued job attachment is established when:

  • the expected total duration of the absence is up to three months (which can be more than three months, if the return to employment in the same economic unit is guaranteed and, in the case of the pandemic, once the restrictions in place where applicable – are lifted)


  • workers continue to receive remuneration from their employer, including partial pay, even if they also receive support from other sources, including government schemes.

In turn persons are classified as ‘not employed’ if: 

  • The expected total duration of absence is greater than three months or there is no or unknown expected return to the same economic unit


  • People in this condition do not receive any part of their remuneration from their employer.

Not-employed persons are classified as ‘unemployed’ if they fulfil the criteria of active “job search”[2] and “availability” specified for the measurement of unemployment.

However, departures from these guidelines in national practices do exist. In North America persons on temporary layoff are considered to be “only weakly or not at all attached to their job and are to be counted as unemployed” (Sorrentino, 2000). In the United States, people on temporary layoff are classified as ‘unemployed’ if they expect to be recalled to their job within six months. If they have not been given a date to return to work by their employer and if they have no expectation to return to work within six months, they need to fulfil the “job search” criteria to be classified as ‘unemployed’. For the latest US figures “people who were effectively laid off due to pandemic-related closures were counted among the unemployed on temporary layoff” without further testing for their return to their previous job (BLS, 2020). In Canada, persons in temporary layoff are also classified as ‘unemployed’ if they have a date of return or an indication that they will be recalled by their employers.

Conversely, persons on temporary layoff are classified as employed (not at work) in Europe, as recommended by the ILO Guidelines (Eurostat, 2016). In practice, formal job attachment is tested on the basis of (i) an assurance of return to work within a period of three months or (ii) the receipt of half or more of their wage or salary from their employer. Somewhat stricter than ILO guidance, absences during COVID-19 crisis whose duration is unknown are treated as absences longer than three months. Those failing to satisfy these two criteria are classified as unemployed if they are “available to start work” (over the next two weeks) and have actively searched for a job in the last four weeks. All other persons on layoff are classified as inactive.

Working with over 100 countries, the OECD is a global policy forum that promotes policies to preserve individual liberty and improve the economic and social well-being of people around the world.


[1] Broad comparability is ensured during normal business conditions, while divergences are potentially exacerbated during economic and financial crises, such as the Great Recession or the Covid-19 crisis.

[2] Some not-employed persons may be classified as “inactive/out of the labour force” because, due to the pandemic, they are either not able to actively look for a job even if they are available to work or are not available to work because of family responsibilities as schools and care services are closed.