copy the linklink copied!Chapter 4. Student truancy and lateness

This chapter examines differences between countries and economies in student truancy and lateness, and how they vary by student and school characteristics. It also looks at the relationship between truancy and lateness, especially when they occur repeatedly, and reading performance. The chapter identifies some of the predictors of student truancy.

    

Every school day, many students miss learning opportunities because they skip school or arrive late for school. Doing so repeatedly may have adverse effects on the individual student and on the learning environment in school. Chronic truancy and, to a lesser extent, lateness have such adverse effects on learning that school systems around the globe are constantly devising strategies to tackle them. The European Commission, for instance, includes efforts to combat truancy as a key policy lever to reduce the share of early school leavers across Europe (European Commission, 2013[1]).

Students play truant for many reasons. Many students skip school or arrive late for school because they are academically disengaged, do not feel they belong at school, failed to wake up or are simply needed at home (Appleton et al., 2008[2]; Gottfried, 2017[3]; Lehr, Sinclair and Christenson, 2009[4]). Some victims of bullying avoid school because they are too afraid or embarrassed (Hutzell and Payne, 2012[5]; Townsend et al., 2008[6]). Good academic performance and positive relationships with peers and teachers seem critical for developing students’ attachment to school and feeding a desire to attend school every day (Gehlbach, Brinkworth and Harris, 2012[7]; Juvonen, Espinoza and Knifsend, 2012[8]; Reid, 2005[9]).

copy the linklink copied!
What the data tell us
  • On average across OECD countries, 21 % of students had skipped a day of school and 48 % of students had arrived late for school in the two weeks prior to the PISA test. In Georgia, Montenegro, Saudi Arabia and Turkey, at least one in five students had skipped school at least three times during that period.

  • The countries and economies where fewer students had skipped a whole day of school were also the countries/economies with higher average reading performance, such as Beijing, Shanghai, Jiangsu and Zhejiang (China), Estonia, Finland, Hong Kong (China), Japan, Korea, Macao (China), Singapore, Sweden and Chinese Taipei.

  • Frequently bullied students were more likely to have skipped school, whereas students who valued school, enjoyed better disciplinary climate, scored higher in the reading assessment, and received greater emotional support from parents were less likely to have done so.

Not all students are equally likely to skip school or lack punctuality. In many countries, especially middle-income countries, boys skip school and arrive late for school more frequently than girls do (OECD, 2015[10]). According to school principals, truancy problems are more likely to hinder learning in urban schools than in rural schools, on average across OECD countries (Echazarra and Radinger, 2019[11]), and urban students tend to arrive late more frequently than rural students do (OECD, 2016[12]). Previous studies have also documented the problems socio-economically disadvantaged students (OECD, 2016[12]; Ready, 2010[13]; Ready, 2010[13]) and those with disabilities (Gottfried et al., 2017[14]; Lane et al., 2006[15]) face in attending school.

Regular truancy can have adverse consequences for students. Truants are more likely to fall behind in class, drop out of school, wind up in poorly paid jobs, have unwanted pregnancies, and even abuse drugs and alcohol (Aucejo and Romano, 2016[16]; Hallfors et al., 2002[17]; Henry and Huizinga, 2007[18]; Smerillo et al., 2018[19]). Some of these unwanted outcomes are more commonly seen amongst chronic absentees from disadvantaged backgrounds (Gershenson, Jacknowitz and Brannegan, 2017[20]; Ready, 2010[13]).

If pervasive, student truancy can also hurt other students in the class. If students who arrive late for school or skip classes fall far behind in their classwork and require extra assistance, the flow of instruction is disrupted, and classmates, particularly those who might be working closely with truants, may suffer (Wilson et al., 2008[21]). Truants may also generate resentment amongst students who attend class regularly – and sympathy amongst others who may realise that they too can skip classes (Wilson et al., 2008[21]).

This chapter examines student truancy and lateness. PISA asked students to report the number of times ( “never”, “one or two times”, “three or four times”, “five or more times”) they had skipped a whole day of school, the number of times they had skipped some classes and the number of times they had arrived late for school during the two full weeks of school prior to the assessment. Skipping school/classes, student truancy and unexcused absenteeism are used interchangeably in this report since “skipping” implies dishonesty. As with other self-reported measures, students may over- or under-report the extent to which they play truant and arrive late for school, and this bias may operate differently across countries and groups of students. In addition, even when the question refers to the last two “full” weeks of school, this period may have been exceptional in some countries and economies, which could potentially affect students’ answers.

copy the linklink copied!
Figure III.4.1. Student truancy, by school characteristics
Percentage of students who reported that they had skipped a whole day of school at least once in the two weeks prior to the PISA test
Figure III.4.1. Student truancy, by school characteristics

Countries and economies are ranked in descending order of the percentage of students who had skipped a whole day of school.

Source: OECD, PISA 2018 Database, Tables III.B1.4.1 and III.B1.4.7.

 StatLink https://doi.org/10.1787/888934029470

copy the linklink copied!How student truancy and lateness vary across countries, schools and students

On average across OECD countries, 21 % of students reported that they had skipped a whole day of school at least once, and 27 % reported that they had skipped some classes at least once in the two weeks prior to the PISA test (Figure III.4.1 and Table III.B1.4.1). However, in some education systems a considerably larger share of students had skipped school. For instance, in Brazil, the Dominican Republic, Georgia, Italy, Jordan, Kazakhstan, Malta, Montenegro, Panama, Romania, Saudi Arabia, Turkey and Uruguay, more than half of students had skipped a day of school in the two weeks prior to the assessment. More worryingly, in Georgia, Montenegro, Saudi Arabia and Turkey, at least one in five students had skipped school at least three times during the same period. Large proportions of students in these countries regularly miss learning opportunities, with likely adverse consequences for both these students and their classmates. In the Dominican Republic, Panama and Peru, for instance, more than 5 % of students were enrolled in schools where at least 90 % of their schoolmates had skipped school in the two weeks prior to the PISA test (Table III.B1.4.3). By contrast, in Beijing, Shanghai, Jiangsu and Zhejiang (China) (hereafter “B-S-J-Z [China]”), Belgium, Hong Kong (China), Iceland, Japan, Korea, Macao (China), the Netherlands, Sweden, Chinese Taipei and Viet Nam, more than 90 % of students had not skipped school in the two weeks prior to the assessment.

Arriving late for school may have less serious consequences for students, but it is generally a more common occurrence (Table III.B1.4.1). On average across OECD countries, almost half of students had arrived late for school at least once in the two weeks prior to the PISA test. In 22 countries and economies, especially in Argentina, Chile, France, Georgia, Greece, Israel, Poland, the Russian Federation, Serbia and Uruguay, more than 10 % of students had arrived late for school at least five times during the same period. In Chile, the Dominican Republic, Panama, Peru and Uruguay, the problem is so widespread that, in 2018, more than 5 % of students attended a school where at least 90 % of students had arrived late for school in the two weeks prior to the assessment (Table III.B1.4.4). However, in Hong Kong (China), Japan, Korea and Macao (China), less than 30 % of students had arrived late during the same period.

Student truancy and lateness varied widely across schools (Figures III.4.1 and III.4.2). Differences related to the socio-economic profile of the schools were generally the largest. In a clear majority of countries and economies, students in socio-economically disadvantaged schools were more likely than students in advantaged schools to have skipped a whole day of school in the two weeks prior to the PISA test. The only education systems where skipping school was more common amongst advantaged students were Macao (China), Saudi Arabia, Turkey and the United Arab Emirates. Moreover, on average across OECD countries, students in rural schools were more likely to have skipped school, but less likely to have arrived late for school, than were students in city schools. On average across OECD countries, both skipping school and arriving late for school were more common in public than in private schools, and in schools with higher concentrations of immigrant students (Tables III.B1.4.7 and III.B1.4.8).

In a majority of countries and economies, boys were more likely than girls to have skipped a whole day of school, particularly in Albania, Greece, Indonesia, the Republic of Moldova (hereafter “Moldova”), the Philippines and Slovenia (Table III.B1.4.5). The only countries where more girls than boys played truant were Argentina, Costa Rica, Ireland, Jordan, Qatar, Saudi Arabia and the United States. In about three out of four school systems, skipping school was more common amongst disadvantaged students than amongst advantaged students, particularly in Denmark, the Dominican Republic, France, Italy, Lithuania, Moldova, the Philippines, Qatar and Ukraine. On average across OECD countries, 25 % of students with an immigrant background had skipped school in the two weeks prior to the PISA test, compared to 21 % of students without an immigrant background – a significant difference.

The differences between groups of students followed similar patterns in the case of lateness, though they were usually larger (Table III.B1.4.6). For example, 51 % of boys reported that they had arrived late for school, compared to 44 % of girls; 50 % of disadvantaged students reported that they had arrived late, compared to 45 % of advantaged students; and 58 % of immigrant students reported that they had arrived late for school in the two weeks prior to the PISA test, compared to 47 % of students without an immigrant background.

copy the linklink copied!Trends in student truancy and lateness

On average across OECD countries, the percentage of students who had skipped a whole day of school at least once in the two weeks prior to the PISA test increased by only one percentage point between 2015 and 2018 (Table III.B1.4.2). The share of students who had skipped some classes remained stable, and the percentage of students who had arrived late for school increased by three percentage points over that period. The largest increases in the share of students who had skipped school were observed in Austria, Greece, Peru, Portugal, Qatar, Thailand and the United Arab Emirates, while the only improvements (i.e. a reduction in the share of students who had skipped school) were observed in Estonia, Finland, Montenegro, the Slovak Republic, the United Kingdom and the United States. The share of students who had arrived late for school increased sharply in Singapore and the Slovak Republic; it shrank only in Costa Rica.

copy the linklink copied!
Figure III.4.2. Student lateness, by school characteristics
Percentage of students who reported that they had arrived late for school at least once in the two weeks prior to the PISA test
Figure III.4.2. Student lateness, by school characteristics

Countries and economies are ranked in descending order of the percentage of students who had arrived late for school.

Source: OECD, PISA 2018 Database, Tables III.B1.4.1 and III.B1.4.8.

 StatLink https://doi.org/10.1787/888934029489

copy the linklink copied!How student truancy and lateness are related to reading performance

In virtually all education systems, in 2018, skipping school or classes and arriving late for school were negatively associated with reading performance, after accounting for the socio-economic profile of students and schools (measured by the PISA index of economic, social and cultural status) (Figure III.4.3 and Table III.B1.4.9). For instance, on average across OECD countries, students who had skipped a whole day of school in the two weeks prior to the PISA test scored 40 points lower than students who had not skipped school during the same period. This association was strongest in the countries and economies where the share of students who had skipped school was the smallest, such as B-S-J-Z (China), Hong Kong (China), Iceland, Japan, Korea, Macao (China), Norway, Sweden and Chinese Taipei (Table III.B1.4.1). On average across OECD countries, skipping some classes was associated with a decline of 37 score points in reading performance; arriving late for school was associated with a drop of 26 score points, after accounting for the socio-economic profile of students and schools. Relationships between truancy and underperformance could take a variety of forms. For instance, students who skip school miss learning opportunities, and students who struggle academically may be less willing to attend school regularly.

copy the linklink copied!
Figure III.4.3. Student truancy and reading performance
Figure III.4.3. Student truancy and reading performance

1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).

Note: Statistically significant values are shown in darker tones (see Annex A3).

Countries and economies are ranked in descending order of the score-point difference associated with having skipped a whole day of school, after accounting for students’ and schools’ socio-economic profile.

Source: OECD, PISA 2018 Database, Table III.B1.4.9.

 StatLink https://doi.org/10.1787/888934029508

Students may also see their performance deteriorate when their schoolmates skip school or arrive late for school, and not only when they themselves play truant. On average across OECD countries, students scored 8 points lower in reading for every 10 percentage-point increase in the number of schoolmates who had skipped school, and 5 points lower for every similar increase in the number of schoolmates who had arrived late for school in the two weeks prior to the PISA test, after accounting for students’ and schools’ socio-economic profile and students’ own truancy or lateness (Tables III.B1.4.10 and III.B1.4.11).

Looking at it another way, students enrolled in schools with the lowest incidence of student truancy (i.e. schools in the bottom quarter of the distribution) scored 514 points in reading, whereas students attending schools with the highest incidence (i.e. schools in the top quarter of the distribution) scored 453 points – a significant difference of 62 score points. In Belgium, Croatia, France, Hungary, Luxembourg, Qatar, Romania, Serbia and Slovenia, the difference in reading performance between students in schools with the lowest incidence of student truancy and those in schools with the highest incidence was at least 100 score points.

At the system level too, student truancy and reading performance were significantly associated (Figure III.4.4). Average reading performance was lower in countries and economies with larger shares of students who had skipped a whole day of school in the two weeks prior to the PISA test than in countries/economies with smaller shares of these students.

copy the linklink copied!
Figure III.4.4. Percentage of students who had not skipped school and reading performance
Figure III.4.4. Percentage of students who had not skipped school and reading performance

Source: OECD, PISA 2018 Database, Tables III.B1.4.1 and I.B1.4.

 StatLink https://doi.org/10.1787/888934029527

However, there were a few interesting exceptions. For instance, in Brunei Darussalam and Malaysia, average reading performance was clearly below the OECD average, but student attendance in these countries was also clearly above average. In these two countries, and also in other countries like Indonesia and the Philippines, comparatively high student attendance, relative to what would be expected given these countries’ average reading performance, provides many opportunities to improve performance. By contrast, Italy and Turkey had a comparatively high rate of student truancy relative to what would be expected given these countries’ average performance in reading.

copy the linklink copied!How repeated truancy and lateness are related to reading performance

PISA has consistently found that students who skip school and arrive late for school, even if sporadically, score lower in the assessment compared to students who never skip school and always arrive on time (OECD, 2016[12]; OECD, 2013[22]). PISA 2018 results confirm this finding (Table III.B1.4.9). However, there is much less evidence on the association between repeatedly skipping school and arriving late for school, on the one hand, and PISA scores, on the other, because the questions are usually reclassified into “never” and “at least once”. To examine how chronic truancy and lateness are related to reading performance, one regression analysis was carried out for each education system where the three items ( “skipped school”, “skipped some classes”, “arrived late for school”), with all the categories ( “never”, “one or two times”, “three or four times”, “five or more times”) were analysed jointly, after accounting for the socio-economic profile of students and schools. In Japan and Korea, so few students had skipped school and classes that only the question about lateness could be examined (Table III.B1.4.1).

The findings in Figure III.4.5 clearly show that, on average across OECD countries, the number of times that students had skipped school or had arrived late for school in the two weeks prior to the PISA assessment was associated with reading performance. However, whereas the negative association with reading performance grew almost linearly with the number of times that students had skipped a whole day of school, it flattened after three or four times in the case of skipping classes and arriving late for school. The results also show that, of the three measures of truancy and lateness, skipping school was the most strongly associated with reading performance. For instance, skipping a whole day of school five or more times during the period was associated with a drop of 40 score points in reading (compared to a student who had not skipped a school day), while skipping some classes or arriving late for school the same number of times were associated with a decline of 23 score points.

copy the linklink copied!
Figure III.4.5. Repeated truancy and lateness, and reading performance
OECD average
Figure III.4.5. Repeated truancy and lateness, and reading performance

Notes: All values are statistically significant (see Annex A3).

Results based on linear regression analysis, after accounting for students’ and schools’ socio-economic profile. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).

All predictors have been included in the same linear regression model.

Source: OECD, PISA 2018 Database, Table III.B1.4.12.

 StatLink https://doi.org/10.1787/888934029546

In Croatia, Malaysia, Mexico, Norway, Sweden, Chinese Taipei and the United Kingdom, skipping school more than four times during the two weeks prior to the PISA test was associated with a decline of more than 60 score points in the reading assessment (Table III.B1.4.12). In Brunei Darussalam, Germany, New Zealand and Singapore, skipping some classes at least five times was associated with a drop in reading scores of more than 50 points. In Australia, Finland, Hong Kong (China), Japan, Macao (China), Malta, Qatar, Singapore and the United Arab Emirates, arriving late for school at least five times in the two weeks prior to the assessment was associated with a drop of more than 40 score points in the reading assessment.

copy the linklink copied!Predicting student truancy

There are multiple reasons why students skip school, including fear of being bullied, weak school attachment, lack of friends, bad relationships with teachers and academic disengagement (Appleton et al., 2008[2]; Gehlbach, Brinkworth and Harris, 2012[7]; Gottfried, 2017[3]; Hutzell and Payne, 2012[5]). This section tries to identify relevant predictors of student truancy, focusing on five areas:

  • Feeling safe at school: index of exposure to bullying

  • Being academically engaged: index of value of school; and reading performance

  • Exposure to peers’ disruptive behaviour: index of disciplinary climate

  • Receiving emotional support: index of parents’ emotional support

  • Feeling socially connected at school: index of sense of belonging at school

Other chapters in this volume describe the indices of exposure to bullying (Chapter 2), disciplinary climate (Chapter 3) and sense of belonging at school (Chapter 9). As for the other indices, PISA asked students whether they agree ( “strongly disagree”, “disagree”, “agree”, “strongly agree”) with the following statements: “My parents support my educational efforts and achievements”; “My parents support me when I am facing difficulties at school”; and “My parents encourage me to be confident”. These statements were combined to create the index of parents’ emotional support whose average is 0 and standard deviation is 1 across OECD countries. Students who sat the PISA test were also asked the extent to which they agree with the following statements: “Trying hard at school will help me get a good job”; “Trying hard at school will help me get into a good college”; and “Trying hard at school is important”. These statements were combined to create the index of value of school whose average is 0 and standard deviation is 1 across OECD countries.

The analysis in Figure III.4.6, which includes all predictors together in the same logit model and accounts for the socio-economic profile of students and schools, reveals that, on average across OECD countries, students were more likely to have skipped a whole day of school at least once in the two weeks prior to the PISA test the more frequently they had been bullied. By contrast, they were less likely to have skipped school the higher their reading performance, the stronger their belief that trying hard at school is important (index of value of school), the better the disciplinary climate in their language-of-instruction lessons, and the stronger the emotional support they received from their parents. Moreover, feeling socially connected at school was only modestly (and negatively) associated with student truancy. Interestingly, the socio-economic status of students (a control variable in the analysis), is only modestly (and negatively) associated with skipping school, even though socio-economically disadvantaged are considerably more likely to skip school than advantaged students (Table III.B1.4.5).

copy the linklink copied!
Figure III.4.6. Predictors of student truancy
OECD average
Figure III.4.6. Predictors of student truancy

1. The socio-economic status is measured by the PISA index of economic, social and cultural status (ESCS).

Notes: All values are statistically significant (see Annex A3).

All predictors have been included in the same logit regression model.

Source: OECD, PISA 2018 Database, Table III.B1.4.13.

 StatLink https://doi.org/10.1787/888934029565

Similarly, the country results in Table III.B1.4.13 show that in almost every school system, students were more likely to skip school when they were bullied more frequently, and less likely to skip school when they scored higher in reading. Moreover, in a majority of countries and economies students were less likely to skip school the higher the value attached to schooling, the better the disciplinary climate and the stronger their parents’ emotional support. However, in only 11 school systems were students less likely to skip school when they reported a stronger sense of belonging at school or when they had a higher socio-economic status.

References

[2] Appleton, J. et al. (2008), “Student engagement with school: Critical conceptual and methodological issues of the construct”, Psychology in the Schools, Vol. 45/5, pp. 369 - 386, https://doi.org/10.1002/pits.20303.

[16] Aucejo, E. and T. Romano (2016), “Assessing the effect of school days and absences on test score performance”, Economics of Education Review, Vol. 55, pp. 70-87, https://doi.org/10.1016/J.ECONEDUREV.2016.08.007.

[11] Echazarra, A. and T. Radinger (2019), “Learning in rural schools: Insights from PISA, TALIS and the literature”, OECD Education Working Papers Series, No. 196, OECD Publishing, Paris, https://doi.org/10.1787/8b1a5cb9-en.

[1] European Commission (2013), Reducing Early School leaving: Key messages and Policy support, https://ec.europa.eu/education/content/reducing-early-school-leaving-key-messages-and-policy-support_en (accessed on 14 January 2019).

[7] Gehlbach, H., M. Brinkworth and A. Harris (2012), “Changes in teacher-student relationships”, British Journal of Educational Psychology, Vol. 82/4, pp. 690-704, https://doi.org/10.1111/j.2044-8279.2011.02058.x.

[20] Gershenson, S., A. Jacknowitz and A. Brannegan (2017), “Are student absences worth the worry in U.S. primary schools?”, Education Finance and Policy, Vol. 12/2, pp. 137-165, https://doi.org/10.1162/EDFP_a_00207.

[3] Gottfried, M. (2017), “Does truancy beget truancy? Evidence from elementary school”, The Elementary School Journal, Vol. 118/1, pp. 128-148, https://doi.org/10.1086/692938.

[14] Gottfried, M. et al. (2017), “Showing up: Disparities in chronic absenteeism between students with and without disabilities”, IESP Working Paper, No. 03-17, Institution for Education and Social Policy, New York, NY, http://steinhardt.nyu.edu/scmsAdmin/media/users/lwb232/IESP_WP03-17_Chronic_1.pdf (accessed on 15 January 2019).

[17] Hallfors, D. et al. (2002), “Truancy, grade point average, and sexual activity: A meta-analysis of risk indicators for youth substance use”, Journal of School Health, Vol. 72/5, pp. 205-211, https://doi.org/10.1111/j.1746-1561.2002.tb06548.x.

[18] Henry, K. and D. Huizinga (2007), “Truancy’s effect on the onset of drug use among urban adolescents placed at risk”, Journal of Adolescent Health, Vol. 40/4, pp. 358.e9-17, https://doi.org/10.1016/J.JADOHEALTH.2006.11.138.

[5] Hutzell, K. and A. Payne (2012), “The impact of bullying victimization on school avoidance”, Youth Violence and Juvenile Justice, Vol. 10/4, pp. 370-385, https://doi.org/10.1177/1541204012438926.

[8] Juvonen, J., G. Espinoza and C. Knifsend (2012), “The role of peer relationships in student academic and extracurricular engagement”, in Christenson, S., A. Reschly and C. Wylie (eds.), Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_18.

[15] Lane, K. et al. (2006), “Academic, social, and behavioral characteristics of high school students with emotional disturbances or learning disabilities”, Journal of Emotional and Behavioral Disorders, Vol. 14/2, pp. 108-117, https://doi.org/10.1177/10634266060140020101.

[4] Lehr, C., M. Sinclair and S. Christenson (2009), “Addressing student engagement and truancy prevention during the elementary school years: A replication study of the Check & Connect Model”, Journal of Education for Students Placed At Risk, Vol. 9/3, pp. 279-301, https://doi.org/10.1207/s15327671espr0903_4.

[12] OECD (2016), PISA 2015 Results (Volume II): Policies and Practices for Successful Schools, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264267510-en.

[10] OECD (2015), The ABC of Gender Equality in Education: Aptitude, Behaviour, Confidence, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264229945-en.

[22] OECD (2013), PISA 2012 Results: What Makes Schools Successful (Volume IV): Resources, Policies and Practices, PISA, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264201156-en.

[13] Ready, D. (2010), “Socioeconomic disadvantage, school attendance, and early cognitive development”, Sociology of Education, Vol. 83/4, pp. 271-286, https://doi.org/10.1177/0038040710383520.

[9] Reid, K. (2005), “The causes, views and traits of school absenteeism and truancy”, Research in Education, Vol. 74/1, pp. 59-82, https://doi.org/10.7227/RIE.74.6.

[19] Smerillo, N. et al. (2018), “Chronic absence, eighth-grade achievement, and high school attainment in the Chicago Longitudinal Study”, Journal of School Psychology, Vol. 67, pp. 163-178, https://doi.org/10.1016/J.JSP.2017.11.001.

[6] Townsend, L. et al. (2008), “The relationship between bullying behaviours and high school dropout in Cape Town, South Africa”, South African Journal of Psychology, Vol. 38/1, pp. 21-32, https://doi.org/10.1177/008124630803800102.

[21] Wilson, V. et al. (2008), “‘Bunking off’: The impact of truancy on pupils and teachers”, British Educational Research Journal, Vol. 34/1, pp. 1-17, https://doi.org/10.1080/01411920701492191.

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.

https://doi.org/10.1787/acd78851-en

© OECD 2019

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.