Chapter 1. Colombia and its education system1

Increased access and participation

Recent education policies have led to a significant expansion of education participation in Colombia, in particular at those levels where enrolment had been lagging. Between 2000 and 2012/13, the share of children enrolled in pre-primary education (i.e. programmes for children below the age of entry into primary education, but with an educational component) increased from 36% to 45%, which remains below the OECD average (84%) but higher than Turkey (28%) (UNESCO-UIS, 2015). A series of measures, from the provision of free public schooling to major investments in infrastructure and education resources, have enabled a dramatic increase in secondary education coverage. Between 2002 and 2012/13, net enrolment rates rose from 59% to 70% in lower secondary and from 30% to 41% in upper secondary (UNESCO-UIS, 2015). Enrolment in tertiary education has also surged, with the gross enrolment rate doubling from 24% to 48% between 2000 and 2013. The OECD average increased from 52% to 72% during the same period (UNESCO-UIS, 2015; see Annex 1.A1).

Colombia has increased school life expectancy by 2 years in the past decade (from 11.4 years in 2001 to 13.5 years in 2010), whereas OECD countries and Latin American countries extended school life expectancy by 1 year on average, to 16.5 years and 13.4 years, respectively (Figure 1.3). Despite this increase, only children aged 7-13 years are close to enjoying universal access to education (defined as enrolment rates above 90%), whereas the majority of OECD countries ensure universal or high enrolment for all 5-14 year-olds (OECD, 2014a).

Figure 1.3. School life expectancy across all educational levels (2000-11)

Note: Missing links suggest data for that particular year were not available.

Source: UNESCO-UIS (2015), “Browse by theme: Education”, Data Centre, UNESCO Institute for Statistics, www.uis.unesco.org/DataCentre/Pages/BrowseEducation.aspx.

Several factors contribute to low attendance rates in Colombia, including poor transitions between levels, high dropout rates and a significant number of children never entering the education system at all. Although the dropout rate from pre-school to secondary education has more than halved between 2002 and 2013, the share of out-of-school primary-aged children has actually increased, from 4% in 2000 to 9% in 2010. About one in five students in Colombia do not continue past primary school (OECD/CAF/ECLAC, 2014). The share of lower secondary-aged youth out of school has declined, from 15% in 2005 to 9% in 2013, and is now below that in Mexico (13%) (UNESCO-UIS, 2015). However, lower secondary education remains the weakest link in Colombia’s education system, with an annual dropout rate of 4.5%, higher than in primary (3.2%) or upper secondary education (3.1%) (MEN, 2015a; UNICEF, 2012). Only 30% of youth make the transition from school to tertiary education and of these many will drop out before completion; annual dropout rates for tertiary programme range from 10.4% at the university level to 22.2% in technical and technological institutions (DNP, 2015; MEN, 2015a). Persistent dropout rates and weak transitions mean Colombia has one of the highest shares of 15-19 year-olds who are not in education – some 36% in total, which is more than double the OECD average (13%) and similar to the situation in Mexico (35%) (OECD, 2015b). Moreover, 19% of 15-19 year-olds are not in education, employment or training (NEET) compared with the OECD average of 7% (OECD, 2015b). Many obstacles limit young people’s access to education, including a lack of educational opportunities, poverty, pressure to work, conflict and violence. Supporting students at all levels, in particular during transition from primary through to the end of lower secondary education, will be fundamental to ensuring students acquire the minimum level of knowledge and skills required for social and economic participation (OECD, 2011).

Colombian students’ access to and participation in education are also hampered by late entry to school and high levels of grade repetition, which can affect academic achievement and increase the risk of dropping out (OECD, 2013b; Garcia Villegas et al., 2013; Barrera, 2014). Many Colombian students are not enrolled in an age-appropriate grade. While the gross enrolment rate5 among 5-year-olds in Colombia is 89%, only 57% are in the appropriate grade (the transition year), which is almost 20 percentage points below the OECD average. In primary education, 84% of students are enrolled in the age-appropriate grade, against a gross enrolment rate of 107%, which is the largest gap among OECD countries. The grade repetition rate in Colombia is also one of the highest among countries participating in the OECD Programme for International Student Assessment (PISA): in 2012, 41% of 15-year-olds had repeated at least one grade, compared with the OECD average of 12% (OECD, 2013b). The high rate of grade repetition is a major factor behind the large share of over-age students. Grade repetition has steep costs, is ineffective at addressing low performance and can increase the risk of dropping out, with older students facing higher opportunity costs of remaining in education (OECD, 2012a).

Evidence suggests that inadequate educational infrastructure and materials hinder access and achievement. This is a problem at every level of the system. While there does not appear to be an overall shortage of teachers in basic education, there are gaps in the teacher workforce at the pre-school and upper secondary levels, and across the system the level of teachers’ qualifications is a concern. Improving both school infrastructure and teaching quality are important government priorities.

Low learning outcomes

Quality is improving, but the majority of students leave school with weak basic skills. PISA assesses what students aged 15 around the world know and are able to do with what they know. Colombia has participated in PISA since 2006. Results suggest that student reading skills have improved across assessment cycles, but remain low compared with the OECD average and other Latin American countries (Figure 1.4). Performance in mathematics and science has remained unchanged. In mathematics, 15-year-old Colombian students are, on average, more than 3 years (118 score points) behind their peers in OECD countries (OECD, 2014b). The Third Regional Comparative and Explanatory Study (Tercer Estudio Regional Comparativo y Explicativo, TERCE), which assesses the mathematics, science and writing skills of Latin American students in Grades 3 and 6,6 shows how Colombian students start to fall behind neighbouring countries such as Chile, Costa Rica and Mexico in the earlier years of education (UNESCO Office of Santiago, 2015). National SABER assessments in Grade 9 (theoretically at age 14) and Grade 11 (at age 16) also indicate low overall standards, with negative trends in reading and mathematics scores in some grades and years.

Figure 1.4. PISA performance and annualised change across all subjects (2006, 2009 and 2012)

Note: Non-OECD countries are shown in blue.

Source: OECD (2014b), PISA 2012 Results: What Students Know and Can Do (Volume I, Revised edition, February 2014): Student Performance in Mathematics, Reading and Science, OECD Publishing, Paris, https://doi.org/10.1787/9789264208780-en.

Colombia has made progress in raising the achievement of students at the bottom of the performance scale although its share of low performers remains one of the highest among OECD countries. Students with the lowest reading performance (bottom 10th percentile) improved their PISA scores by more than 50 points to 295 points between 2006 and 2012, which amounts to slightly more than a school year, but the share of low performers did not significantly decrease across PISA cycles. In PISA 2006, 56% of 15-year-old Colombian students performed below proficiency level 2, compared with an OECD average of 21%. In PISA 2012, the share remained statistically unchanged at 51.4%, far above the OECD average of 18% (OECD, 2014b). National 2014 SABER literacy assessments found 49% of students in Grade 3, 67% in Grade 5 and 73% in Grade 9 did not meet minimum standards. A significant proportion of Colombian students are not well prepared for entry to tertiary education or the labour market; 27% of individuals who took the SABER 11 national assessment in 2013 performed at the low or inferior level, with important implications for tertiary retention and success (DNP, 2015).

Attainment rates and labour market outcomes indicate whether an education system is able to develop individuals who can support their country’s social and economic development. Colombia has greatly improved upper secondary attainment rates across generations; 40% of 25-34 year-olds have obtained an upper secondary education or post-secondary non-tertiary education, more than double the rate among 55-64 year-olds (16%) (OECD, 2015b). Household survey data suggest that students who progress beyond lower secondary school have better wages and a greater chance of employment in the formal economy than those who do not attain an upper secondary education (CEDLAS/World Bank, 2014). The labour-market returns on a tertiary education are significant, and have remained constant despite the rising number of graduates (see Chapter 5). Labour market rewards for education are limited, however, by the size of the informal economy and the weak skills of many graduates. The difference in employment rates among 25-34 year-olds in Colombia with a below upper secondary education (73%) and those with an upper secondary or post-secondary non-tertiary education (76%) is small; the OECD average difference is 18 percentage points (OECD, 2015b). Breaking this cycle of poor education quality and limited labour market returns will require co-ordinated efforts to stimulate the supply, demand and use of skills in the workforce.

Inequities in access and learning

Colombia has implemented a range of initiatives to tackle disparities in education access and learning. ECEC policies have aimed to increase school participation and performance among poor and vulnerable students. From 2012, education has been free from the transition grade to the end of upper secondary education, though indirect costs such as transport and learning materials remain. Early student selection into different pathways, such as academic or pre-vocational, can hamper equity efforts, but Colombia provides comprehensive education until the end of lower secondary education at age 15 and upper secondary certification, whether academic or technical, enables entry to tertiary education. Funding policies, such as conditional cash transfers and tertiary loans, have also improved equity. As well as targeting children from poor socio-economic backgrounds, the government has introduced tailored measures to meet the needs of other vulnerable groups, such as ethnic communities, children with special educational needs, individuals affected by violence and rural populations. Flexible school models, ethnic education, school meals and school transport have all helped to reach disadvantaged groups. However, socio-economic background, geographic location, ethnicity and gender still heavily determine the educational opportunities of Colombian children.

Social inequality has contributed to deep disparities in access to education and achievement in Colombia. According to Colombia’s Demographic and Health Household Survey 2009-10, a student from the poorest socio-economic level (strata level 1)7 has a school life expectancy of 6 years, which is half that for an individual from the wealthiest socio-economic level (strata level 6), and is much more likely to be out of school (Garcia Villegas et al., 2013; UNESCO-UIS, 2015). National and international assessments indicate that socio-economic background is a strong determinant of student performance in Colombia (Barrera, 2014; Garcia Villegas et al., 2013). PISA 2012 results show that Colombia has one of the lowest percentage of resilient students compared to OECD and Latin American countries, meaning that few disadvantaged children overcome their background to achieve high results (OECD, 2013c and Figure 1.5). Just 9% of disadvantaged 17-21 year-olds are enrolled in tertiary education compared with 62% of advantaged ones (ECV, 2012 in DNP, 2015). An assessment of SABER results indicates that socio-economic background and parents’ educational level has a determinant effect on student achievement in Colombia (Barrera, 2014; Garcia Villegas et al., 2013; ICFES, 2012).

Figure 1.5. Share of resilient students across OECD and Latin American countries (2012)

Notes: A student is classified as resilient if he or she is in the bottom 25% of the PISA index of economic, social and cultural status (ESCS) in the country of assessment and performs in the top 25% of students among all countries, after accounting for socio-economic status. Non-OECD countries are shown in blue.

Source: OECD (2013c), PISA 2012 Results: Excellence through Equity (Volume II): Giving Every Student the Chance to Succeed, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264201132-en.

Socio-economic inequality in Colombia is also evident in the differences between education institutions. Evidence from Bernal (2014) indicates that 61% of children from advantaged families (in the top quintile) attend private early childhood education (preescolar jardin) compared with 2% from disadvantaged families. Families from higher socio-economic levels tend to have higher participation rates in early childhood education for three to five year-olds (Bernal, 2014). Disadvantaged students also tend to attend disadvantaged schools, which often lack human and financial resources (OECD, 2011). Evidence from an analysis of SABER assessments indicates that the socio-economic level of schools explains a significant portion of the difference in achievement, suggesting a high rate of segregation between schools based on the socio-economic level of students (Duarte et al., 2012; Barrera, 2014). PISA 2012 results also indicate that, after taking the socio-economic context of the schools into account, there was no significant performance difference between students attending public or private schools, suggesting the socio-economic context in which a school is located can negatively impact student performance (OECD, 2013b).

Individuals in rural areas tend to have less access to educational opportunities and lower academic achievement than their peers in urban areas (Figure 1.6). Colombians living in rural areas suffer from multiple disadvantages, including higher rates of poverty (43% compared with 27% in urban zones), malnutrition, teenage pregnancy and violence, as well as insufficient infrastructure (CIPI, 2013). A majority of indigenous people and Afro-Colombians live in rural areas, and face additional barriers to education (Barrera, 2014). Participation in education is on average lower in rural departments and municipalities than more urbanised areas, in particular at the ECEC and upper secondary levels (MEN, 2015a; DNP, 2015; Garcia Villegas et al., 2013; Bernal, 2014; OAS, 2010). In 2013, the average number of years of schooling among rural populations aged 15 and older was just 5.47 years compared with 9.36 years in urban areas (MEN, 2015a). At the tertiary level, several rural departments (Amazonas, Arauca, Guainía, Guaviare, Huila, Putumayo, San Andrea, Vichada and Vaupés) had gross enrolment rates below 20%, while the more urbanised departments of Quindío and Santander had rates close to 60% (MEN, 2015a).

Figure 1.6. Regional disparities across Colombia as indicated, by SABER 11 assessment (2013)

Source: DNP (2015), Bases Plan Nacional de Desarrollo 2015-2018, Departamento Nacional de Planeación (National Planning Department), Bogotá, https://www.dnp.gov.co/Plan-Nacional-de-Desarrollo/Paginas/Que-es-el-Plan-Nacional-de-Desarrollo.aspx, Graph IV-2, page 62.

In Colombia, students with special educational needs have lower participation rates, though quality data are lacking (MEN, 2012). In 2014, 119 060 students diagnosed with special educational needs were enrolled in basic education according to MEN data. Yet data from the 2005 census indicated that there were 426 425 children (0-17 year-olds) with some form of disability in Colombia, which suggests that many are not participating in education and that data gathering should be improved to ensure these students are identified and supported (MEN, 2012). Guidelines have been developed to better support children and young people with special needs and the MEN allocates additional funding to educational institutions to co-finance the cost of providing education services to disabled or exceptionally gifted students (MEN, 2015a). Improved data would help make these policies more effective.

Males tend to outperform females in Colombia on a number of indicators, and these disparities are deeper than in most OECD countries. Boys significantly outperformed girls in PISA 2012 mathematics and science assessments by one of the largest differences among OECD countries: 25 points for mathematics and 18 points for science compared with the OECD averages of 11 points and 1 point, respectively. The large difference may in part be explained by the fact that by the age of 15, poorly performing boys are more likely to have dropped out of school than girls. On the reading assessment, girls outperformed boys by 19 points, whereas in OECD countries the average gap was twice as much, at 38 points (OECD, 2014b). Tertiary attainment rates among 25-34 year-old women were 29%, outperforming men at 23%, but both were below the OECD average of 46% and 35%, respectively. However, 25-34 year-old Colombian women face far higher levels of unemployment (13%) than the OECD average (8%) and Colombian men in the same age group (8%, OECD average of 7%) (OECD, 2015b). Young women (15-29 year-olds) are also more likely to be excluded from education, employment and training than their counterparts in other OECD countries (31% compared with the OECD average of 18%) (OECD, 2015b). Improving gender equality in education and the labour market need to go hand-in-hand if Colombia is to widen its talent pool and achieve more inclusive growth.

Ensuring institutional coherence and clear leadership

The MEN is the main body responsible for education in Colombia and its mandate covers the whole sector, from pre-school to tertiary education. According to Decree 5012 of 2009, the MEN sets national policy and criteria, defines standards, guarantees and ensures the right to access education, formulates directives, supports the territories, and inspects and evaluates the quality of the education system. The MEN has two vice ministries: one responsible for pre-primary/ECEC, basic (primary and lower secondary education) and upper secondary education; and the other responsible for tertiary education. Within each vice ministry, departments responsible for each educational level direct and plan education policy.

Many other bodies also help shape education policy in Colombia, including the presidency’s office, other ministries, various multi-sectorial commissions, and multiple agencies, institutes and councils (Annex 1.A2). While such institutional complexity is a common feature of education systems in all countries (OECD, 2015c; Cerna, 2014a), what stands out in Colombia is the extent to which these other bodies play a decisive and largely autonomous role at some education levels (Figure 1.7). Although the MEN defines national policy on ECEC, the main implementing agency is the Colombian Institute for Family Welfare (Instituto Colombiano de Bienestar Familiar, ICBF), the national body responsible for children under five. The ICBF is largely responsible for regulating its own institutions and co-ordination with the ministry and local secretaries of education is limited. At the tertiary level, the MEN is responsible for setting national policies, and for the promotion, supervision and inspection of public and private tertiary education institutions (TEIs). However, the 1991 Constitution grants TEIs far-reaching autonomy and the MEN has few mechanisms to ensure its policies are implemented. Moreover, SENA, which operates under the Ministry of Labour, is responsible for offering and implementing VET, as well as non-formal education and training (known in Colombia as education for work and human development). SENA provides 70% of technical and technological education at the tertiary level and is playing a growing role in delivering education in upper secondary education as well. SENA manages its own funding and decides its own policy and regulation (Saavedra and Medina, 2012).

Figure 1.7. Colombia’s governance structure across all educational levels

Note: See Annex 1.A2 for further details on governance actors.

Source: Updated from Piñeros, L., B. Caro Acero and L. Villa Arcila (2013), “Diagnostico de los sistemas de información de la educación inicial, básica, media y superior”, ESOCEC, July 2013.

Such a breadth of engagement and level of autonomy can be a strength. Most countries have a range of institutions involved in ECEC and in VET, and there is little doubt that in Colombia the ICBF and SENA have expanded resources for education, enhanced delivery capacity and enabled services to better meet the distinct needs of different populations. There are also historical and political reasons for the independence of tertiary education institutions in Colombia, as in many other Latin American countries, which are important to understand. These strengths, and also the potential challenges arising from them, will be discussed in more depth in subsequent chapters. What is important to note from a system-wide perspective is the extent to which leadership across the education sector is fragmented. In Colombia, the MEN has limited means of influencing reform outside of basic education. This makes it difficult to effect systemic improvement and create consistent pathways for learning from early childhood to the tertiary level. It creates gaps in accountability and in data collection and monitoring. The existence of different institutions with separate funds and budgets also creates inefficiencies and makes it more difficult to pool capacity and resources and avoid duplication. Complexities at the central government level can be magnified at the local government and school level, where the capacity to co-ordinate multiple actors is often very weak (OECD, 2013a).

The educational reforms to which Colombia has committed itself will require clear leadership and strong co-ordination across the sector. The current lack of clarity in the roles and responsibilities of important institutions involved in education delivery will make this difficult.

Ensuring effective decentralisation

There is a growing trend towards education decentralisation in OECD countries. International experience suggests that the process of decentralisation can improve local responsiveness, encourage creativity in the use of resources, promote innovation and create incentives for quality improvement (Waslander, Pater and van der Weide, 2010 in OECD, 2012a). However, decentralised governance is only effective when adequate capacities, co-ordination measures and accountability mechanisms are in place (OECD, forthcoming). Without such conditions, decentralisation can deepen inequity and make quality improvement more difficult.

Colombia was one of the first countries in Latin American to begin decentralising its education system and is the most decentralised unitary country in the region. The process was rapid and far-reaching but remains incomplete. The 1991 Constitution and the 1994 General Education Act paved the way for the devolution of responsibilities to departments and municipalities, which include guaranteeing education services from pre-primary to secondary education, managing education personnel and educational funding, and supervising private institutions (Cortes, 2010). The General Education Act also gave schools considerable autonomy over curricula and assessment. The Law 715 of 2001 further clarified the responsibilities of each level of government and reformed the system of allocation for education resources to better align funding and the needs of the regional authorities.

Decentralisation has been managed by a process of certification, whereby all departments irrespective of their capacity, and any districts or municipalities that meet specific criteria, are granted the status of Certified Territorial Entities (Entidades Territoriales Certificadas, ETCs). Certification is determined by several factors including adequate physical and financial resources, a basic information system, and capacity to manage the teaching profession. As of 2015, there are 95 ETCs, each led by a Secretary of Education and guided by an education development plan (Plan de Desarrollo para la Prestación del Servicio Educativo, PED). ETCs are responsible for defining and implementing education policy, monitoring quality and the educational provision of all public and private schools, and supporting non-certified municipalities.

This system has allowed ETCs with the capacity and funding to innovate and improve performance (García Jaramillo et al., 2014). Decentralisation has been a positive factor in increasing enrolment rates, in particular in regions where local demand for education reform is strong (Faguet and Sánchez, 2008). Decentralisation has also enabled municipalities to use subsidised private education to expand the provision of education (Cortes, 2010). Several departments and territories, such as Bogotá, Medellín and Manizales, have benefited from increased autonomy to both expand access and raise standards (Box 1.2). However, the majority of ETCs, particularly those in rural areas, lack the human and financial capacity to effectively manage their school system and drive improvements.

The capacity of local certified governments in Colombia to carry out their functions varies greatly (World Bank, 2014; Garcia Jaramillo et al., 2014). The MEN has identified 48 ETCs as lacking capacity according to a range of indicators covering enrolment, quality and management (MEN, 2015c). Many municipalities lack the human, financial or technical resources to ensure the quality of schools under their responsibility. In extreme cases, such as Chocó, the MEN has taken over ETCs that have been unable to govern effectively (Cortes, 2010; MEN, 2015c). Capacity is usually weakest in those regions where education needs are greatest, such as remote and isolated areas and those affected by the conflict, with the result that decentralisation has contributed to deepening inequalities. This was evident in the review visit where the team observed that larger urban municipalities had relatively sophisticated and promising strategies and partnerships, which were less evident in poorer areas.

The MEN has made efforts to strengthen the capacity of the ETCs and their secretaries of education through a variety of support mechanisms. The Modernisation of the Secretaries of Education Project (Proyecto de Modernización) was introduced in 2004 to provide differentiated support to meet the specific management and contextual needs of the ETCs (MEN, 2015c). Under the project, technical support plans were developed to assist the weakest territories, and a team of advisors set up, including former secretaries of education, to provide guidance on implementation (MEN, 2015c). However, the project was discontinued in 2014 and a recent evaluation of flexible education models indicated only half of ETCs collaborated with the MEN on policy implementation, suggesting that relations between central and local government remain insufficient (Econometría/SEI, 2014).

Tackling disparities in institutional capacity between regional and local governments is a central priority of the PND 2014-2018 (DNP, 2015). The National Planning Department (Departamento Nacional de Planeación, DNP) is designing a differentiated approach, built on a careful assessment of institutional competencies across municipalities and focused on supporting those where capacity is weakest (OECD, forthcoming). New education programmes, such as full-day schooling, also include explicit provision for capacity development and technical assistance (DNP, 2015). In addition, the MEN offers training to help weaker ETCs apply for national grants and take part in the new competitive financing schemes that aim to incentivise quality improvement, though more support is likely to be needed to ensure results-based funding mechanisms do not leave less developed departments even further behind (see Chapter 3).

It will be important for Colombia to continue these efforts if it is to raise performance standards across the country and ensure greater equity in learning opportunities. Strong co-ordination mechanisms are important in decentralised governance structures, both to ensure coherence and alignment among actors and to support capacity development (Charbit, 2011; OECD, forthcoming). Peer learning can be a powerful catalyst for improvement, and there is scope for the MEN to play a stronger role in facilitating knowledge exchange across ETCs. The MEN regularly convenes secretaries of education to discuss policy implementation; these meetings could be structured to enable more systematic sharing of good policies and practice. Several OECD countries with decentralised education systems, such as Canada, Germany and Switzerland, have used national councils of education secretaries to forge a collective vision for reform and encourage knowledge exchange (OECD, 2015c). They have also created collaborative platforms to strengthen capacity at the local level. In Germany, for example, the Learning Locally programme (Lernen vor Ort) has supported the development of networks to strengthen local government capacity in areas such as education monitoring, management and consulting. Decentralisation will only be an effective force for improvement if the MEN has the right information to identify and monitor the ETCs and the ETCs and local governments have the capacity to drive change and effectively support schools and school leaders.

Forging a shared, long-term vision

Effective decentralisation requires a shared vision across national actors and governance levels. This is particularly important when managing improvement in a complex system, where it is necessary to bring together a range of institutions and actors at different levels behind shared goals for reform (Hargreaves and Shirley, 2011; Fullan, 2011). Creating a shared vision requires clear targets that are commonly agreed and effectively communicated. Platforms for dialogue, where different stakeholders can come together to discuss policy objectives, are also important to ensuring broad ownership of reform and coherent implementation.

The elevation of education as a presidential priority, and the overarching ambition to make Colombia the most educated country in Latin America by 2025, provides a strong basis for building a shared vision for reform in the country. The PND 2014-2018 defines a set of strategic priorities with measurable medium-term targets as well as longer-term objectives for change across all education levels, which are now being developed further in a four-year Sectorial Plan for Education (see Box 1.2). Long-term planning is important in the education sector, where change requires time to take effect, and major reforms, such as the introduction of a new curriculum, need careful planning and consultation. Such a longer-term planning horizon is supported in Colombia by a ten-year education plan (Plan Decenal de Educación), a new version of which is due to be developed for 2016-2026.

Colombia has invested efforts in strengthening national dialogue on reform. The priorities presented in the PND have been developed through consultation with regional authorities and a wide range of stakeholders, including the Federal Teachers’ Union (Federación Colombiana de Educadores, FECODE) and the private sector. The National Council of Social and Economic Policy (Consejo Nacional de Política Económica y Social, CONPES), led by the Department for Social Planning, provides an additional platform for multi-sectorial consultation on key issues in education and other policy areas. At the tertiary level, the National Council of Higher Education, an advisory body of the MEN, has facilitated broad-based debate on the future of higher education, resulting in a shared long-term vision for the sector, detailed in the Agreement on Higher Education 2034 (CESU, 2015) (see Chapter 5). A similarly wide consultation process informed the development of the From Zero to Forever strategy for early childhood education. This inclusive approach is important to forging a common understanding of strategic priorities. The challenge is to maintain this collaborative momentum in the critical stage of implementing reforms.

Engaging education secretaries to get behind a common agenda for change will be essential to successful reform in Colombia. Within multi-level governance structures like Colombia’s, close co-ordination between national and local authorities is essential to effective implementation (Charbit, 2011). The main instrument for policy alignment in Colombia are the ETCs’ development plans for education services (Plan de Desarrollo para la Prestación del Servicio Educativo, PEDs), which all education secretaries are required to develop detailing how they intend to improve their education system in line with national objectives. For priority programmes, such as full-day schooling, the MEN is establishing separate agreements with education secretaries to ensure a shared understanding of targets and accountability for results. The MEN has also introduced financial incentives to encourage local governments to give higher priority to education. Such proactive engagement is important, as experience with other national initiatives, like the From Zero to Forever strategy, suggests that scepticism and resistance at the local levels have presented obstacles to reform (Bernal, 2014; CIPI, 2013), (see Chapter 2).

Regular dialogue between the MEN and secretaries of education is vital to building the relationships and trust that provide the foundations for successful reform. Secretaries of education also need to take a proactive approach to build political support for education among municipalities under their responsibility, especially in large rural territories, where school systems are widely dispersed. Antioquia, which is the second largest department in Colombia, has made local mayoral involvement a central pillar of its campaign to make the department Colombia’s “most educated”. Such local leadership is essential to carry the strategic vision of the government into Colombia’s classrooms.

International evidence shows that reform is most successful when every level of society is engaged (OECD, 2015c). Engaging families and communities, teachers unions, non-governmental organisations (NGOs), and the business community is crucial to achieving sustained improvement. Those regions that have advanced education reforms, such as Manizales and Bogotá, have invested in building partnerships and creating a social movement for change. The MEN has sought to generate such engagement at the national level though the creation of National Excellence Day; every year, on 25 March, all Colombians will be encouraged to reflect on the quality of education and its importance for individual and national development. The introduction of a performance index for every school is also intended to mobilise public demand for improvement. However, as subsequent chapters of this report will suggest, further efforts are needed in Colombia to engage civil society, in particular parents and communities, in the process of education reform. OECD countries have developed various vehicles to encourage this kind of engagement. The example of the Student Success/Learning to 18 Strategy in Ontario, Canada, is instructive of how one government has mobilised a wide range of stakeholders behind common goals and seen positive results in terms of student performance (Cerna, 2014b).

Creating a shared education vision can help secure reform over the long term, and avoid unnecessary changes of direction when a new government takes office. Colombia has embarked on an ambitious path of improvement that extends until 2025. It is important that it stays the course. The revision of the 2006-2016 decennial plan for education will be an opportunity to embed the objectives laid out in the PND in a longer-term strategy and ensure sustained political and financial support for the far-reaching reforms that are required.

Mobilising adequate funding for education

Despite increased investment in recent years, public funding for education in Colombia is below that of other major emerging economies in Latin America and per-capita spending is well below the OECD average. Continuing to expand the public resource base for education, while harnessing additional funds and improving the effectiveness of resource use, will be important if Colombia is to reach its ambitious education goals.

Colombia devotes a considerable and rising share of national resources to education. In 2011, total spending (public and private) across all education levels amounted to 6.7% of GDP, which is slightly above the OECD average of 6.1% (OECD, 2014a). Between 2000 and 2013 government spending on education rose from 3.5% of GDP to 4.9% (UNESCO-UIS, 2015). This puts Colombia in the middle of the 4-6% international benchmark for public expenditure on education in developing and emerging economies, above Chile (4.5%), but below Costa Rica (6.8%), Brazil (6.3%), and Mexico (5.1%) (UNESCO-UIS, 2015). In OECD countries, average public expenditure on education was 5% in 2011 (OECD, 2014a).

Surging private spending suggests that public funding has not kept pace with rising demand for education in Colombia. Between 2005 and 2011, private expenditure increased by 31% to account for 35% of overall education expenditure, more than double the OECD average (16%) and close to Chile’s high level of private spending (40%) (OECD, 2014a). As in Chile, the majority of private spending in Colombia comes from households and is particularly high at the pre-primary and tertiary levels, where it accounts for 56% and 46% of all resources (compared with the OECD average of 31% and 19%). As subsequent chapters show, this high degree of private funding contributes to significant inequities in access to early childhood and tertiary education in Colombia.

Public spending per student in Colombia is far below the OECD average and that of other major emerging economies in Latin America. This is true at all levels, but in particular for upper secondary and tertiary education (Figure 1.8). Almost all public funding for primary and secondary education is absorbed by teachers’ salaries (83%, compared to the OECD average of 63%). This leaves limited funds for investment in resources that could improve quality, such as teachers’ professional development and text books. There are also scarce resources for the renovation and expansion of education infrastructure: 1.7% of spending is on capital expenditure compared with 7% in the OECD (OECD, 2014a),8 and for spending on food, transportation and accommodation, which are important for reaching Colombia’s disadvantaged students.

Figure 1.8. Annual public expenditure on public institutions, per student in Colombia and selected countries (2011)

Note: Non-OECD countries are shown in blue.

Source: OECD (2014a), Education at a Glance 2014: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2014-en, Table B3.3.

OECD analysis shows that excellence in education requires much more than financial resources (OECD, 2013b). Countries such as Turkey and Uruguay have a similar total expenditure per student to Colombia but have much higher learning outcomes, showing that how funds are allocated is just as important as the level of resources available. However, the data do suggest that there is a threshold of per-student expenditure below which achievement is likely to suffer (Figure 1.9). For countries such as Colombia, with cumulative expenditure below USD 50 000 per student, higher expenditure on education is correlated with higher student scores (OECD, 2013b). The experience of many developing and emerging countries suggest that in contexts with high levels of poverty, additional spending, when used effectively, can improve learning outcomes for the poorest students (UNESCO, 2015).

Figure 1.9. Spending per student age 6 to 15 and PISA mathematics performance (2012)

Note: Non-OECD countries are shown in blue.

Source: OECD (2013b), PISA 2012 Results: What Makes Schools Successful? (Volume IV): Resources, Policies and Practice, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264201156-en, Figure IV. 1.8, 2013a.

Colombia has committed to sustain funding for education despite recent fiscal pressures caused by the drop in global commodity prices. Budget cuts have significantly affected many areas of government but education priorities set in the PND 2014-2018 will continue to be supported at the levels planned. The national education budget increased from COP 27.3 trillion (Colombian pesos, about USD 9.8 billion)9 in 2014 to an estimated COP 28.9 trillion (approximately USD 10.4 billion) in 2015, which is higher than expenditure on defence (COP 28.2 trillion or USD 10.1 billion).10 The ring-fencing of education in the current budget is a welcome move. The government is also implementing a range of measures to improve the efficiency of education spending, such as reorganising school networks (see Chapter 3). However, this is unlikely to be sufficient to achieve the educational improvements Colombia needs and is committed to in the PND. The construction of new classrooms alone under the full-day schooling initiative is estimated to require COP 4.5 billion (about USD 1.5 million) by 2018.

Additional measures will be needed to expand the funding base for education. Increasing national tax revenues represents a major structural challenge to improving public service delivery in Colombia, and is examined in more detail in the OECD Economic Survey of Colombia (OECD, 2015a). Other policies will also be important to widen the public resource base for education. Local resources are the most significant untapped source of funding for education in Colombia, and encouraging greater local investment in the sector will be essential to consolidating the decentralisation process. Stronger public-private partnerships and greater use of innovative financing models can also help to increase the range and level of funds available.

Stronger local leadership in education financing is essential if Colombia is to improve education access, equity and quality. The current financing system remains heavily centralised, with many local governments lacking the commitment, capacity and incentives to fund education improvements. The majority of public resources for education in Colombia are transferred by the Ministry of Education to departments, municipalities and schools through the System of General Participations (Sistema General de Participaciones, SGP), which is the main channel for redistributing funds from national taxation to support public services across the country (Figure 1.10). The share of public funding allocated through the SGP has decreased in recent years, falling from 80% in 2002 to 60% in 2013, while the share of public funding mobilised by ETCs has increased (from 8% to 13%), as has the share of funds distributed by the MEN for specific education policies, such as free education (Álvarez et al., forthcoming; MEN, 2015a). However, local spending on education remains extremely low compared with OECD countries. In 2011, central government accounted for 84% of all education public resources in Colombia, compared to the OECD average of 52% (OECD, 2014a).

Figure 1.10. Distribution of funding, by source and by departments and selected ETCs

Source: Ministry data.

There is a clear need, and also an opportunity, to mobilise greater local resources for education in Colombia. Departments and municipalities with strong local economies and political commitment to education, such as Antioquia, Barranquilla, Bogotá, Cali, Manizales and Medellín, have succeeded in allocating considerable local funds for education. However, evidence suggests that the majority of ETCs lack both the fiscal capacity and political will to invest in education (Melo, 2005; Sanchez Torres and Pachon, 2013; Galvis, 2014). Stimulating greater local investment will require the right mixture of incentives and capacity development. The MEN is already doing much more to use national funds to encourage increased local spending, such as through match funding and competitive calls for grant proposals, and is providing technical assistance to help weaker ETCs benefit equally from such opportunities. However, inadequate local investment in education remains a major barrier to improvement (Álvarez et al., forthcoming; see Chapter 3).

Stronger public-private partnerships offer another channel for increased education investment, though one that needs to be navigated with care. Several ETCs have outsourced education service delivery to private providers as a means to create additional capacity to expand access to schooling. Publicly funded private organisations play an important role in providing ethnic education for specific populations and communities and delivering Colombia’s many flexible school modalities. The majority of ECEC services provided by the ICBF are also delivered by private providers. The experience of Medellín, where around half of students attend private institutions supported by government vouchers, is instructive of the mechanisms needed to avoid some of the risks of public-private partnerships, such as segregation across schools and uneven quality (Patrinos, Barrera-Osorio and Guaqueta, 2009). For example, Medellín actively fosters networking between public and private schools to ensure weaker institutions can learn from the best.

Colombia has also been active in developing innovative funding models for education. Countries are increasingly looking for new ways of financing education to expand the resource base and use funds more effectively to spur improvements (IFF, 2013). In the SENA, Colombia has a strong and long-standing system for capturing corporate resources to support vocational education and training. Funded through the Income Tax for Equality (Impuesto sobre la Renta para la Equidad, CREE), it is an example of how Colombia is seeking to improve mechanisms to redistribute returns from economic growth to support greater equity in education, and close the financing gaps in sectors that are under-resourced, such as ECEC. More recently, in 2012, the government allocated 10% of oil and mining royalties to fund science, technology and innovation (see Chapter 5). This is good example of how natural resource windfalls can be invested in long-term human capital gains. In order to help finance the full school day, the MEN intends to create a National Education Infrastructure Fund (Fondo de Financiamiento de la Infraestructura Educativa) to pool public and private resources, and is looking to incentivise further private investment in school infrastructure development. At the local level, some governments have introduced special taxes to capture revenue for education. In Antioquia, for example, 2% of all alcohol sales are reserved for education and the department has been proactive in mobilising local corporate spending on tertiary education and training. This is an example that more local governments could follow.

Providing more effective financing to support equity and quality

Increased funding for education will only deliver improvements if resources are allocated effectively. Evidence shows that how resources are distributed and used has a major impact on education outcomes (OECD, 2013b). Funding strategies, if designed well, can help steer improvements in both equity and quality. The major strides Colombia has made in expanding access have been driven by deliberate policies to remove financial barriers to education. Measures such as the abolition of school fees, conditional cash transfers, and school nutrition and transport programmes have brought increasing numbers of young Colombians into basic education, while tertiary loans have enabled many disadvantaged students to be the first in their family to progress beyond school to higher levels of learning. As well as improving the targeting of resources to redress inequities, the government has sought new ways to use funds to incentivise quality reforms and encourage achievement. Despite these efforts, socio-economic background still determines learning outcomes in Colombia far more than it does in the majority of OECD countries. The stubbornly wide gap in performance between the poorest and richest regions and students suggests that funding policies could be improved further to ensure all children in Colombia have equal opportunities to learn.

Recent years have seen important improvements in the SGP, the financing mechanism that determines how resources are distributed across territories and accounts for the majority of education spending, especially in the poorest municipalities. The funding formula has been revised several times to address the additional costs of educating low-income students better, with the introduction of weightings to compensate for rurality, poverty, minority status and other markers of disadvantage (Barrera-Osorio, Maldonado and Rodríguez, 2012). Additional efforts have been made recently to simplify the methodology used to allocate resources. The fact that SGP funds now follow the student has removed the distortions that occurred in the past, when the hiring of teachers was disconnected from student numbers and staff costs escalated. In an effort to drive up standards, the MEN has also sought to develop a more results-based approach, with a proportion of the SGP tied to performance indicators such as dropout rate, grade repetition and achievement on SABER assessments (see Chapter 3). However, only 8% of the SGP is available for such incentives. The fact that 83% of overall resources in primary and secondary education are still absorbed by teacher salaries severely limits the scope to use national transfers to improve equity and quality (Álvarez et al., forthcoming). Moreover, stakeholders interviewed during the review visit suggested that the increasing sophistication of the funding formula created undue complexity and unpredictability for local governments, and created challenges for transparency and accountability. This risk is underlined by international research, which points to the need for such funding schemes to be matched with complementary policies, particularly to strengthen school capacity, if they are to have the desired impact (Fazekas, 2012).

Figure 1.11 relates the achievement of Grade 5 students in SABER to expenditure per student. It points to the need for a multidimensional approach to address the overlapping factors affecting education performance in Colombia. Both spending and results vary hugely across territories. It is evident that the majority of departments and territories spend very little per student, and their results are equally low. The level of student spending also seems to be poorly related to the level of student needs. Many departments with high levels of unsatisfied basic needs are allocating very little to support disadvantaged students (Santa Maria, et al., 2009). For example, rural Chocó, with a poverty rate of over 60%, spends the same per student as Bogotá, one of the richest municipalities in the country with one of the lowest poverty rates (DNP, 2015). However, those departments and municipalities allocating more funds to support students, like Amazonas, Guaviare and Vichada, are not seeing commensurate improvements in quality. Poor management, corruption and a lack of complementary policies to support local capacity to improve schooling, impede the effective use of additional funds to enhance performance (OECD, 2012a, 2012b).

Figure 1.11. Relationship between SABER performance and per-student spending, by departments and selected ETCs (2014)

Source: Data provided by the MEN (Ministry of National Education); DANE (2015), “Educación formal”, Departamento Administrativo Nacional de Estadistica (National Department of Statistics) website, www.dane.gov.co/index.php/esp/cultura-estadistica/89-sociales/educacion/3901-educacion-formal.

More effective funding in Colombia will require a range of measures across all levels of government. The capacity of the MEN to use the SGP to steer change will remain limited unless national and local authorities have greater flexibility over the allocation of teachers and the means to channel more and better qualified teachers to the poorest regions and schools. Given the limited discretionary funds available under the SGP, resources channelled through national programmes, like full-day schooling (jornada única) and “Let’s All Learn” (Todos a Apprender), will be critical to improving equity and quality in the country. It will be important for these programmes to reinforce each other, and be effectively integrated with other equity strategies, such as the More Families in Action (Más Familias en Acción) programme, to achieve better results (Box 1.3). Increased resources for poorer regions and schools will also need to be matched with better auditing and controls, and adequate investment in management capacity, to ensure funds are used to support their intended objectives.

Effective funding requires strong accountability mechanisms. Colombia has an established system of audits and yearly progress reports which ETCs and educational institutions are required to comply with. Recent efforts have been made to tighten such controls, notably in the auditing of tertiary institutions. Stronger information systems have also helped to improve accountability but more needs to be done (Barrera, 2014; see below). There is a particular need to improve information, accountability and management capacity at the school level. Unless school leaders and teachers are empowered to use additional resources to support learning in the classroom, then funding policies will have limited impact (see Chapter 3).

One final consideration is how public resources are allocated across education levels. As is natural for an emerging economy with a young population, Colombia invests more in primary and lower secondary education (3.6% of GDP) than the OECD average (2.5%). However, public investment in pre-primary education is low, in particular when compared with spending on tertiary education, which is above the OECD average (2% in Colombia compared with the OECD average of 1.6%). With 10% of Colombia’s population below the age of 4 and around 60% of young children living in poverty, this allocation seems regressive, especially considering that tertiary education remains out of reach for the majority of students from low-income backgrounds. Both international evidence and national research from Colombia indicate that investment in quality early years learning yields high rates of return in terms of student performance in school and across a range of outcomes in adulthood, and that these rates of return are highest for those from the most marginalised groups (OECD, 2012a; Bernal, 2014). Prioritising spending on the early years is one of the most effective funding decisions Colombia could make to drive improvements across the education system (see Chapter 2).

Creating a more integrated information system

Colombia’s education information system is shaped by its governance structure. It comprises a large number of databases that are often poorly co-ordinated, creating both duplication and gaps in information and making it difficult to create a clear picture of education performance (Figure 1.12). The MEN is legally responsible for developing and implementing the national education information system but many other national actors also play a role. The three prominent information databases for early childhood education and care are managed by different actors: the MEN, the ICBF and the Inter-sectoral Commission for the Comprehensive Care of Early Childhood (Comisión Intersectorial de Primera Infancia, CIPI). The MEN manages multiple databases in pre-primary, primary and secondary education, and each certified territory is also required to have its own information system. While the MEN provides guidelines for the latter, secretaries of education can also design their own systems. Tertiary education institutions manage their own information; they are legally required to share information with the MEN, although until recently the ministry had limited means to ensure they did. SENA manages its own information system, applying its own criteria and standards to the monitoring of technical and technological institutions (OECD/IBRD/The World Bank, 2013; MEN, 2014). An overview of the main data systems in Colombia is provided in Annex 1.A3. A recent analysis of this complex matrix has shown that the majority of databases operate independently of each other, with their own structures, indicators and objectives (Piñeros et al., 2013). Access and use are often difficult and there are few tools to enable users to link data systems.

Figure 1.12. MEN’s information system including databases and sources of education information

This lack of integration creates serious challenges for education stakeholders seeking to use information for accountability and improvement. Without common indicators and standards it is difficult to ensure consistency in the data. This affects the ability to track trends across education levels, assess performance within and between the ETCs, and exploit information for policy-making purposes (Piñeros et al., 2013; OECD/IBRD/The World Bank, 2013). The MEN has begun efforts to address these disconnects. For example, the CIPI has recently introduced an information system which aims to connect data from the various early childhood services, and the Early Childhood Information System (Sistema de Información para la Primera Infancia, SIPI) and the Enrolment Information System (Sistema Integrado de Matrícula, SIMAT) can now be linked to provide information on 5-year-olds entering the transition year. Further reforms are in the pipeline. The government plans to create a national registry of education (Registro Nacional de Educación, RENE) that will be available to the public and provide information on education from early childhood to the tertiary level (MEN, 2015d). The creation of the index of school quality (Índice Sintético de Calidad Educativa, ISCE), which integrates SABER and MEN data, is also intended to provide a more holistic picture of school performance (MEN, 2014). Another step that Colombia might consider is establishing a shared framework of standards, indicators and methods of reporting that apply to all institutions (OECD, 2013d; Piñeros et al., 2013). This would not only help to facilitate system integration and the assessment of performance across institutions and regions, but also support equal standards in information quality.

Improving data quality

Good quality data are the backbone of a strong information system. In order to be an effective support for policy making, data need to be consistent, complete, reliable, timely and precise. Countries are constantly striving to improve the quality of their education data through better indicators, more systematic collection and stronger quality assurance (OECD, 2013d). In Colombia, the information system faces a number of data quality issues, the most important of which are weak capacity for data collection and use, in particular at the local level, and the absence of a strong, independent national body responsible for quality assurance.

Variations in statistical capacity across regions result in important differences in the quality and timeliness of data collected by ETCs and reported to the MEN. In 2012, the MEN carried out an audit of education institutions to verify school enrolment rates; over 300 000 students who were said to be enrolled turned out to be “ghost” students. Anecdotal evidence suggests that many secretaries of education have difficulty in reaching rural areas to audit the reliability of data and often lack the technical capacity to do so. Information on infrastructure, in particular in rural areas, is limited, although the MEN is taking steps to address this through a comprehensive inventory between 2015 and 2018. Although private schools are legally required to provide data to ETCs regarding tuition fees, enrolment and other school data is frequently lacking. A recent review of ECEC information systems found that many were incomplete and unreliable (Bernal, 2014). At the tertiary level, institutional capacity to provide information varies considerably across the country (Piñeros et al., 2013). Across all levels, disaggregated information is often missing, even though this is essential for designing effective policy responses in a country with a high degree of inequality and local diversity. For example, data on schools do not include equity variables which would enable policy makers to identify and address the factors influencing the performance of disadvantaged institutions or regions (Piñeros et al., 2013). In addition to weak capacity to collect data, there also appears to be a lack of understanding of the value of data for policy making and how information can be used to steer improvement. Local stakeholders interviewed during the review visit referred to national reporting and monitoring mechanisms as checklists, and other studies have highlighted gaps in understanding at the local government and individual institutional level about how such exercises might support them in their daily work. This points to the importance of ensuring that efforts to build statistical capacity are combined with measures to explain the purpose of monitoring and reporting processes and help key stakeholders, in particular education secretaries and school principals, understand the results.

To improve the quality of data, many countries have introduced an independent quality assurance body or mix of agencies responsible for the information system and other areas of evaluation and assessment. Central quality assurance bodies can co-ordinate complex information systems, help set standards, ensure data quality and hold various stakeholders accountable (OECD, 2013d). An independent central agency can also be impartial and help inform the national debate without being subjected to political changes. Among OECD countries, for example, Mexico granted the National Institute for Educational Assessment and Evaluation autonomy to develop a strategic and coherent vision of assessment and evaluation in collaboration with the Secretariat of Public Education and decentralised bodies (OECD, 2015c). Colombia might consider strengthening and expanding the role of ICFES to oversee monitoring and evaluation at all education levels, including across institutions managed by SENA and ICBF, as a means to both enhance integration and improve data quality (Garcia Jaramillo et al., 2014; OECD/IBRD/World Bank, 2013). ICFES could have the responsibility for establishing data and reporting protocols and providing capacity development to different data providers.