1 Costing Basic Education in Country X Various methods differ in sophistication when it comes to estimating the cost implications of attaining basic education. The following simulation model offers a simplified and useful method to estimate annual costs of gradually attaining the long-term goal of basic education for all. For a more sophisticated approach, please see the Millennium Project: Preparing National Strategies to Achieve the Millennium Development Goals: A Handbook (www.undp.org). Goals and Targets Before we go into the details of the simulation model attached as an Excel file in Annex A1, we will establish some basics about goals and targets, the entry points for all cost estimation. The first step and entry point in this model is that stakeholders agree on a long-term goal, its definition and milestones. This goal might be the MDG to achieve universal education by 2015. Goals and targets should be SMART, that is, they must be Specific, Measurable, Achievable, Realistic, and Time-bound (see Section X.X). Specific means that the goal is well defined and says exactly what stakeholders mean. In this case, the MDG goal of education for all was deemed to be pretty straightforward by Country X stakeholders, who agreed to adopt this as a long-term goal. Measurable means that stakeholders must be able to measure and verify if the goal is being attained. Although some stakeholders in Country X argued that the MDG goal should be measured as 100% completion rate, due to limitations in data they agreed to measure the target in gross enrolment rate (GER) data that has been collected regularly by the Ministry of Education in Country X during a number of years. Stakeholders also agreed that about 108% GER by 2015 would approximate a 100% completion rate. Achievable. Stakeholders agreed that the goal will be possible to achieve in the time specified (2015) and that it will be possible to break down the goal into annual targets/milestones. Stakeholders in Country X disagreed about the Realism of the goal, given that the current GER and capacity to expand basic education are extremely limited in Country X. In order to increase the likelihood of achieving the goal, stakeholders discussed and agreed to integrate and make sure that teacher-training and other activities would take place simultaneously. But even if such measures will generate positive results in terms of increased enrolment, the entry point in Country X is only 30% GER in 2006, and in combination with the currently extremely low capacity, it did not seem realistic that the GER could rise to 108% in 2015 (MDG goal for education). Stakeholders therefore agreed to aim for 90% GER in 2015. They also agreed on the importance of regularly following-up the realism of the goal in relation to outcome indicators. Since the goal will be monitored each year, it can be modified at a later stage if capacity and enrolment rates would improve more rapidly than currently estimated.
2 In order to be able to monitor the long-term goal, stakeholders agreed on annual Timebounded targets/milestones. Stakeholders agreed that teacher training and curricula development would be necessary in order to increase quality of the education, which was seen as important for motivating parents to send their children to school. In addition, a school-feeding program would promote increased enrolment. Country X imposes neither school fees nor compulsory payments for school uniforms. If parents would have these private expenses, it would have been important for Country X to strive for a policy of abolishing school fees as financial barriers are often a binding constraint that bar parents from sending their children to school. As the results from these initiatives to increase quality in education likely would take some time, stakeholders agreed that the GER would probably not increase that much during the first two years, but thereafter it would take off. Annual GER targets were therefore decided hand-in-hand with annual targets for teacher-student ratios, teachertraining activities, and number of children to be included each year in the school-feeding program, amongst others. Summary Table: Gross Enrolment Rate Goals and Targets Goal Baseline Targets for Projection Years 2015 2006 2007 2008 2009 2010 2011 2012 2013 2014 90% 30% 32% 35% 40% 45% 55% 65% 75% 85% Simulation Model Now when the main goal and annual targets have been established, we are turning to the simulation model in Appendix A1. This is a simple Excel sheet with long-term goals in the second column (2015) and base years 2005 and 2006 in the next two columns. Thereafter, the model consists of projection years up to 2012. The last column considers comments such as major assumptions, targets, unit costs and/or data sources. All numbers are in real prices and in USD. Should the user wish to estimate costs in local currency and at current prices, he/she can easily modify the model. The user should observe that he/she would need to fill in fields marked in yellow (goals and targets) and light green (baseline statistics or unit costs). Red text marks monetary terms, and purple background marks estimated financial gaps. The base years in this model are 2005 and 2006, and they have blue background. PRSP costing often covers three years, why 2007-2009 is marked in red and framed in bold. But PRSP costing is a long-term and iterative process that must be connected with long-term development goals. Therefore, the simulation model covers a longer time horizon that one should work backwards from. Anticipated Domestic Public Revenue the First Block of the Model The first block of the model is to find out domestic public resources available for education in Country X. The country s total production (GDP) was 2,04 billion USD in 2004. In 2006, the real growth rate of the economy was 5%, hence GDP increased to
3 2,142 billion USD in 2006. Current projections are that the economy will grow 5% annually. The latest census showed that Country X s population was 6,8 million in 2005 and that population growth has been around 2,8% during the past few years. If the annual population growth continues to be 2,8%, Country X will have about 8,3 million inhabitants by 2012. Based on the figures for GDP and population, the model calculates GDP per capita, which was 300 USD in 2005. Country X has been in war for many years, which has affected its domestic revenue mobilization very negatively. In 2005, public revenue was estimated at about 5% of GDP, and there were no indications of increased revenue rate in 2006. For 2015, however, Country X aims for a revenue ratio of 20% of GDP, which is the regional average. Due to war, Country X has also been unable to provide sufficient education for its citizens, which shows up in low GER and a low percentage of education as a percentage of public revenue. Moreover, education spending has up to now not focused on basic education only 30% of resources for education has targeted basic schooling. Country X has instead focused its education spending on higher education. To change priorities to reflect the importance of basic education, stakeholders agreed on a long-term goal of doubling the proportion (60% by 2015). Summary Table: Macroeconomic Goals, Baselines and Targets Variable Goal Baseline Targets for Projection Years 2015 2006 2007 2008 2009 2010 2011 2012 2013 2014 GDP growth 5% 5% 5% 5% 5% 5% 5% 5% 5% 5% Population 2,8% 2,8% 2,8% 2,8% 2,8% 2,8% 2,8% 2,8% 2,8% 2,8% growth Domestic 20% 5% 8% 10% 12% 14% 16% 18% 20% 20% resource mobilization (% of GDP) Education 25% 17% 19% 21% 22% 23% 24% 25% 25% 25% budget (% of revenue) Basic education (% of education budget) 60% 30% 35% 38% 40% 42% 45% 48% 52% 56% Recurrent Expenditure the Second Block of the Model This block starts with baseline statistics about the current number of school-aged children 6-13 years old. The user is expected to fill in the latest statistics and an estimate on the growth of the age group. Country X had statistics about the number of school-aged children (around 2,2 million), but lacked data on the cohort s growth rate. For that reason, an assumption was made that this age group has the same growth rate as the population in general. The model uses this information to calculate the actual number of children in school by multiplying the latest GER and targets for GER to the projected number of children (ages 6-13) for every year between 2006 and 2012.
4 Schoolchildren in Country X are in general enrolled in large classes, where each teacher has 60 students on average. In order to raise quality, stakeholders agreed that classes should be smaller and adopted a goal of 40 students per teacher by 2015. Stakeholders worked backwards from the goal to agree on annual PTR targets that are realistic. As it would likely take time before planned teacher training activities would result in a fast increase of trained teachers, the PTR was projected to increase slowly during the first few years, and thereafter more rapidly. Besides the school-feeding program, teachers salaries are the single most important variable for this model. In Country X, teachers had about 100 USD per month, equaling 4 times GDP per capita. The benchmark used for high EFA achievers 1 is slightly lower 3,6 times GDP per capita. If data on teacher s salary is available, the user is advised to use it and calculate the relation between it and per capita GDP. Goals and targets for teacher cost in percent of GDP per capita should then relate to the entry point and the realism of raising or lowering salaries. The total cost for teachers salaries was about 13,5 million USD in 2005 according to the model s calculation. Stakeholders checked this amount against the actual total wage bill of Country X budget follow-ups and found it realistic since teachers salaries normally take up about 50% of the total wage bill. Schoolchildren in Country X normally share a set of textbooks with fellow students. In order to increase quality, stakeholders in Country X have agreed that each student should have his/her own set of books. This is captured in the model, by filling in 100% as the proportion of students that will have textbooks by 2015. This goal relates to the current state in Country X and is broken down in annual targets. Stakeholders were careful to relate to actual data on the proportion of schoolchildren that get their own books, but also, stakeholders carefully assesses bottlenecks in the delivery of schoolbooks. The unit cost for books, which includes also the Government s costs for logistics and delivery, is estimated at 10 USD per set. Costs for schoolbooks therefore constitute an essential part of total expenses for schooling. Next field in the simulation model regards the proportion of non-salary recurrent costs to salary costs. This includes administration costs, small repairs, material, etc. Please observe that the model estimates separately costs for curricula development and teacher training, which are also recurrent non-salary expenses but not included in the field on other recurrent. The model assumes a goal of 10% other recurrent costs, which together with the separately estimated non-salary recurrent costs would equal about 33% of teachers salaries. This proportion is used as a benchmark in Gurria and Gershberg 2, and necessary for assuring quality in education. It is a higher proportion than in industrialized countries, mirroring the large quality problems with education in Country X. 1 See Costing the Education MDGs: A Review of Leading Methodologies, by Martin Gurria and Alec Ian Gershberg 2 See Costing the Education MDGs: A Review of Leading Methodologies, by Martin Gurria and Alec Ian Gershberg
5 Summary Table: Interventions, Goals and Baselines Intervention Goal Baseline Number of Teachers 40 Students per Teacher by 2015 60 Students per Teacher Teacher Salary 4 times per capita income 4 times per capita income Textbooks 100% of Students have 1 set of 30% of Students have textbooks textbooks by 2015 Curriculum Development 20% of teachers trained annually in new Curriculum needs to be updated curriculum Teacher Training 30% of teachers trained annually 5% of teachers trained annually Teacher Training Centers 10 new centers constructed and 5 centers rehabilitated by 2015 Dire need for teacher training centers to be constructed and rehabilitated School Feeding Program All students receive one meal per day in No school feeding school Classroom rehabilitation 10,000 classrooms rehabilitated by 2015 75 classrooms rehabilitated in 2006 Classroom construction 20,000 classrooms constructed by 2015 75 classrooms constructed in 2006 Curriculum Development the Third Block of the Model The curriculum had not been modernized for many years and needed urgently to be updated in Country X. The model assumes that the unit price, 500 USD, covers training and upgrading of curricula. 20% of all teachers will get trained in the new techniques and learning process every year. Teacher Training the Fourth Block of the Model Costs for teachers training include mainly the training modules and investments in teacher training institutes. In Country X, teacher-training unit cost is about 500 USD per teacher, rehabilitation and equipment of a teacher-training center is estimated at 133,000 USD, whereas it costs about 445,000 USD to build and equip a new center. Stakeholders assessed the needs to rehabilitate 5 centers and build 10 new centers up to 2015. School-feeding Program the Fifth Block of the Model Country X is extremely poor after the long war. Stakeholders believe that a schoolfeeding program would increase parents incentives for sending their children to school. It will also improve health and nutrition, which is important for the learning capacity of children. The long-term goal is that all schoolchildren will receive meals at school during the 200 days that a normal school year in Country X covers. Each meal costs about 50 US cents (all inclusive). Several donors have expressed their interest in contributing to this program, which otherwise Country X could not afford. Capital Expenditure the Sixth Block of the Model The war in Country X has resulted in an extremely low net enrolment rate and children have been taught in multiple shifts in few available classrooms. Moreover, there is a big gap between boys and girls currently enrolled, suggesting that additional construction such as separate lavatories will be required to reduce restrictions on girls enrolment. There is thus an urgent need to rehabilitate and build new classrooms, including sanitary
6 and water facilities. During 2006, only 75 classrooms were rehabilitated and another 75 classrooms constructed. Stakeholders agreed that as a proxy for the need of classrooms, a ratio of 1,3 teachers in each classroom should be used. With this as a base, they agreed on the need to rehabilitate at least 10,000 and build 20,000 classrooms by 2015. Although this is an ambitious goal, it will not be sufficient for all schoolchildren and some will still receive their education in multiple shifts and/or outdoors. The next step for stakeholders is to estimate how much it will cost to rehabilitate and construct classrooms and facilities. Relevant and reliable unit costs were not available in Country X. Moreover, unit costs were believed to differ from region to region depending on transportation and security. Stakeholders therefore contacted NGOs and public authorities that had recent experience of rehabilitating and constructing classrooms. They found that unit costs also varied a lot depending on the constructor. In those cases where local communities had taken responsibilities for construction (under sound technical supervision), unit costs had been brought down considerable. A reasonable average unit cost was agreed to be 5,000 USD for rehabilitation and 10,000 USD for construction. Financial Gaps the Seventh Block of the Model This block in the Model adds and consolidates information. As expected, there is an increasing trend in recurrent as well as capital costs, reflecting the increased enrolment rate and quality of education. Unit costs (excluding the school-feeding program) will increase from current levels of about 25 USD per student to about 78 USD per student in 2011. This implies that it will triple as a percentage of per capita income from currently 8% of per capita income to 23% in 2012. Although public domestic resources are projected to steadily increase for basic education, Country X will need to seek additional finance to cover annual gaps of between 12% to 15% of total domestic public revenue every year. In the three years that are covered by the PRSP, the financial gap amounts to 21,9 million in 2007; 23,4 million in 2008; and 28,8 million in 2009. Adding the schoolfeeding program would increase financial gaps considerable. As a proportion of GDP, however, Country X expected spending on basic education is slowly increasing from very levels (about 1,3% of GDP excluding school-feeding program) in 2007 to current Sub-Saharan average levels of about 4,8% of GDP in 2012. Technical and Vocational Education Add-On to the Model The model also consists of two additional blocks on vocational education and alphabetization. Stakeholders in Country X felt strongly that they would need both. The number of students in vocational training was estimated by multiplying a target percentage (1%) to the actual number of persons aged 15-19 years. Stakeholders agreed on aiming for 20 students per teacher in the long run. This goal was based on the current average of students per teacher (25), gradually declining. Vocational teachers salaries were estimated to be 25% higher than those for teachers in basic education. The model also includes teacher training and curriculum development for vocational education.
7 Finally, it also includes goal and unit costs for rehabilitation and construction of vocational training centers. Stakeholders realized that unit costs for vocational training per student were very high, especially when compared to those for basic education. In Country X, vocational training was estimated at about 160 USD per student, that is about 3 times the unit cost for basic education. Adult Literacy and Non-Formal Education Programs Add-On to the Model The adult literacy rate is low in Country X (30% in 2006), despite the start-up of smallscale initiatives for adult literacy in 2000. The goal is to increase the literacy rate to 50% by 2015, starting from 30% in 2006. The target age group is those above 15 years of age. The plan is to continue using community teachers, but to add on some incentives as well as training and updated teaching material for them. In addition, stakeholders agreed on the need to rehabilitate and construct training centers. After stakeholders agreed on goals, targets and unit costs, the model estimated the total annual costs for adult literacy programs to 650 900 thousand USD. Due to large classes and the use of community teachers, the unit cost per student is low. In 2007, when the program is planned to take off, the unit cost per student is estimated at about 11 USD. As the program proceeds with economy of scale effects, the unit cost per student declines to about 5 USD by 2012.