Indicator D7. What proportion of teachers leave the teaching profession?

This indicator analyses teacher attrition; that is to say, the proportion of teachers (between pre-primary and upper secondary levels of education) leaving the teaching profession during their career. Several methods can be used to measure attrition, and this indicator focuses on an indirect measure of attrition, computing attrition based on the number of teachers in two successive reference years and the number of teachers who entered the teaching profession between these two reference years. This follows the method proposed by the Sustainable Development Goal (for comparability issues related to this method, see Box D7.1; for more information on other methods to estimate attrition, see Box D7.2).

Across the OECD and partner countries and economies with comparable attrition rates estimated with the proposed method, attrition rates of all teachers from pre-primary to upper secondary public institutions range from 3.3% in Israel to 11.7% in Norway. In a half of these countries and economies, attrition rates exceed 8%: Brazil, Chile, England (United Kingdom), Finland, the Netherlands, Norway and Sweden (Figure D7.1).

Attrition rates of all teachers can also be expanded to analyse attrition by type of institution, gender, age and/or level of education. Except in the analysis by level of education, attrition rates in this analysis cover pre-primary to upper secondary levels of education combined.

Expanding the scope to all types of institutions combined (both public and private institutions combined) shows a similar range of attrition rates in public institutions for pre-primary to upper secondary education combined: from 3.8% in Israel to 11.7% in Norway. In eight out of the ten countries and economies with comparable data, differences in attrition rates between teachers in public institutions and teachers in all types of institutions combined do not exceed 1 percentage point (Figure D7.1). This may result from the fact that public institutions enrol most students, and then most teachers teach in public institutions, resulting in a similar level of attrition rates in public institutions as the level of attrition for all types of institutions combined (Table D7.6, available on line).

However, attrition may vary between types of institutions. In Sweden, teachers in private institutions show lower attrition than teachers in public institutions (because the attrition rate of teachers in all types of institutions combined is lower than that in public institutions) (Figure D7.1). Chile is an example of the opposite case. In Chile, where teachers in private institutions make up more than a half of all teachers, the attrition rate for teachers in private institutions is higher than that in public institutions (because the attrition rate of teachers in all types of institutions combined is higher than that in public institutions only) (Figure D7.1 and Table D7.6, available on line). Higher frequency of changes in the teaching population (due to recruitment and dismissal) in government-dependent private schools than in public (municipal) schools may explain this phenomenon (Ávalos and Valenzuela, 2016[3]).

In a majority of countries with available data, male teachers show higher attrition rates than their female colleagues, in both public institutions and in all types of institutions combined (Figure D7.2). However, this statement should not be generalised across all OECD countries considering the small number of countries with available data (for more information on the relationship between gender and the intention to leave teaching within five years among lower secondary teachers, see Box D7.3).

Among the 14 countries and economies with available data, the extent of the gender difference in teacher attrition rates in public institutions varies across countries. Gender differences in attrition rates exceed 5 percentage points in Estonia, but are less than 3 percentage points in other countries and economies. In Brazil, Colombia and Ireland, the attrition rate of female teachers in public institutions is larger than that of their male colleagues, though the extent of the difference is less than 1 percentage point. In 11 countries and economies with data on all types of institutions combined, the pattern of gender difference is more consistent than that of public institutions: attrition rates of male teachers are equal to or higher than than those of female colleagues (Table D7.1).

Higher attrition rates among male teachers compared to female teachers may not lead to high attrition rates for all teachers combined, due to the smaller proportion of male teachers in the teaching profession. For example, in Estonia, while attrition rates of male teachers in public institutions are 9.3%, the attrition rate of all teachers is 4.8%, as only 17% of teachers are male at primary and secondary levels combined (Table D7.1 and Education at a Glance Database (OECD, 2021[8])).

The higher attrition rate of male teachers may increase the gender imbalance in the teaching profession that is already observed in many OECD countries. In fact, between 2005 and 2019, the share of female teachers in the teaching profession in primary and secondary levels combined increased from 69% to 72% on average for all OECD countries with data (see Indicator D5). A gender imbalance in itself may not be a critical problem, as its impact on students’ performance is unclear (Cho, 2012[9]). Nevertheless, the higher attrition rate of male teachers may signal other reasons, which implicitly differ between gender (e.g. flexibility in working arrangements, social perceptions of teaching jobs, differences in relative salaries to similarly educated workers in the labour market), that better explain why certain groups of teachers leave the teaching profession (see Box D3.3 for teachers’ actual salaries relative to earnings of tertiary-educated workers and Box D5.1 for potential sources and implications of gender imbalances in the teaching profession).

Regardless of the type of institution, teachers of different age groups show varying levels of attrition, which may reflect differences in expectations of staff from their jobs and also the situation of the job market. Among countries with available data, attrition rates are high for younger age groups (aged 24 or below and 25-34 year-olds), reach the lowest levels during the mid-career years (35-44 year-olds and 45-54 year-olds), then increase again when teachers approach legal retirement age (aged 55 or over). The general trend is similar across all countries with available data, though there are some differences due to differences in the education systems (Figure D7.4).

Among the countries and economies with available data, attrition rates of teachers in the youngest age group (aged 24 or below) are the highest among all the age groups in 5 out of the 12 countries and economies with data (Austria, Brazil, Finland, Norway and Sweden) (Figure D7.4). In these countries, attrition rates for the youngest age group are 20-40 percentage points higher than those of teachers aged 35-44 (Table D7.1). The prevalence of short-term contracts and lack of qualification as fully qualified teachers for teachers of younger ages may explain the high attrition rates of the youngest age group in some of these countries. For example, in Austria, teachers usually begin their teaching career under fixed-term contracts up to five years. In Sweden, a large proportion of young teachers have short-term contracts and/or are not yet qualified for the national teacher certification, which is required for a permanent contract. Other plausible reasons for high attrition among younger teachers could be that young teachers realise that the teaching profession does not meet their expectations or the fact that female teachers may leave the teaching profession to care for their young children.

However, the level of attrition among the youngest age group needs to be interpreted with caution in some countries, considering the small absolute number of teachers included in this group. For example, many teachers in Finland begin their teaching career beyond age 25, because the minimum qualification required for primary and secondary teachers is a master’s qualification (ISCED 7) for most teachers. Consequently, teachers aged 24 or below make up only about 1% of all teachers in pre-primary to upper secondary levels combined and high levels of attrition will not translate into a large number of teachers leaving the profession.

In contrast, the attrition rate of teachers in the oldest age group (aged 55 or older) is the highest among all age groups in Chile, the Flemish and French Communities of Belgium, England (United Kingdom) and Ireland, regardless of the type of institution (Figure D7.4). In most countries, legal retirement ages are 60 or over, though there can be some differences between teachers (e.g. depending on gender or year of birth) and prerequisite conditions (see Annex 3 for more information). The younger legal retirement age (55 in Ireland) and existence of pre-retirement systems allowing teachers to prepare for retirement before they reach the legal retirement age (the Flemish and French Communities of Belgium and England (United Kingdom)) could contribute to explain high attrition rates in the oldest age group.

The variation of attrition rates by level of education differs across the countries and economies with available data. For example, for teachers in public institutions, differences in attrition rates between primary and secondary education is less than 1 percentage point in Brazil, whereas the difference exceeds 4 percentage points in Lithuania (Figure D7.5).

Among the 11 countries and economies with available data, no specific level of education shows a particularly high level of attrition compared to the other levels of education. However, regardless of the type of institution, attrition rates at pre-primary level are either similar (e.g. Colombia) or higher than those at the primary level (e.g. Finland, Lithuania) in all countries and economies with available data (Figure D7.5).

In seven out of the ten countries and economies with available data, attrition rates for teachers in lower and upper secondary education combined is higher than that of teachers in primary education (Figure D7.5). In these countries, statutory work requirements (e.g. number of annual statutory working hours, task requirements) do not vary greatly between primary and secondary levels (OECD, 2016[10]). However, differences in teaching load or difficulty, actual working conditions, and/or relative salary level between the two levels may help to explain why attrition rates are higher at the secondary level than the primary level.

Attrition rate is the percentage of teachers at a given level of education permanently leaving the teaching profession during the reference period.

Reference period in this indicator refers to a period between the consecutive time points when the number of teachers is counted for the school year.

Temporary leave in this indicator refers to any form of leave during which teachers are identified “temporarily not at work” as defined in the UNESCO-UIS/OECD/Eurostat data collection on education statistics when counting the number of teaching personnel.

The indicator covers the most aggregated level of education from pre-primary to upper secondary education in headcount numbers. Most of the analysis is based on the method proposed in the Sustainable Development Goal project, unless stated otherwise.

Attrition rate(t-1,t)=Number of teacherst-1-Number of teacherst+Entrants(t-1,t)Number of teacherst-1

Please see Annex 3 for more information on other estimation methods used in this indicator and for country-specific notes.

Sixteen OECD and partner countries and economies contributed to this indicator: Austria, the Flemish and French Communities of Belgium, Brazil, Chile, Colombia, England (United Kingdom), Estonia, Finland, Ireland, Israel, Lithuania, the Netherlands, Norway, Sweden, and the United States.

Data are from the 2020 OECD-INES NESLI survey on teacher attrition, which refer to the school year 2015/16 (or 2016 for southern hemisphere countries).


[3] Ávalos, B. and J. Valenzuela (2016), “Education for all and attrition/retention of new teachers: A trajectory study in Chile”, International Journal of Educational Development, Vol. 49, pp. 279-290,

[9] Cho, I. (2012), “The effect of teacher-student gender matching: Evidence from OECD countries”, Economics of Education Review, Vol. 31/3,

[7] European Commission/EACEA/Eurydice (2021), Teachers in Europe: Careers, Development and Well-being, Publications Office of the European Union, Luxembourg.

[8] OECD (2021), “Educational personnel by institution”, Education at a Glance Database, (accessed on 12 May 2021).

[11] OECD (2020), Education at a Glance 2020: OECD Indicators, OECD Publishing, Paris,

[5] OECD (2020), TALIS 2018 Results (Volume II): Teachers and School Leaders as Valued Professionals, TALIS, OECD Publishing, Paris,

[10] OECD (2016), Education at a Glance 2016: OECD Indicators, OECD Publishing, Paris,

[6] Sims, S. and J. Jerrim (2020), “TALIS 2018: Teacher working conditions, turnover and attrition”, statistical working paper, UK Department for Education, (accessed on 19 March 2021).

[2] UNESCO (2018), Global Education Monitoring Report 2019: Migration, Displacement and Education - Building Bridges, Not Walls, UNESCO, Paris, (accessed on 19 March 2021).

[1] UNESCO-UIS (2019), SDG 4 Data Digest: How to Produce and Use the Global and Thematic Education Indicators, UNESCO Institute for Statistics, Montreal, (accessed on 19 March 2021).

[4] Weiss, E. (1999), “Perceived workplace conditions and first-year teachers’ morale, career choice commitment, and planned retention: A secondary analysis”, Teaching and Teacher Education, Vol. 15/8, pp. 861-879,

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2021

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at