Policy Research Working Paper 10763 Ensuring an Equal Start for All Pakistani Children What Will It Cost? Abdullah Alam Amer Hasan Elizabeth Hentschel Education Global Practice A verified reproducibility package for this paper is May 2024 available at http://reproducibility.worldbank.org, click here for direct access. Policy Research Working Paper 10763 Abstract Quality early childhood education improves childhood delivery mechanisms, such as community construction development outcomes and has long-term implications for and vouchers. To ensure 100 percent net enrollment in school readiness, workforce participation, and economic early childhood education by 2035, Pakistan must increase growth. Despite this, in Pakistan, the net enrollment rate of the amount of the education budget spent on early child- children ages 3 to 5 in early childhood education was only hood education from the existing allocation of 5.3 percent 31 percent in 2022. This paper estimates the cost of expand- to 10.4 percent by 2035. This means increasing the early ing access to early childhood education using an adapted childhood education budget from PKR 71 billion (US$0.3 version of the early childhood education Accelerator Cost- billion) in 2022 to PKR 418 billion (US$1.85 billion) in ing and Simulation model. Using available administrative 2035, suggesting an average annual increase of 14 percent. data, the paper presents cost estimates for three packages: Using alternate delivery mechanisms, such as community (i) a business-as-usual package, (ii) a core service delivery construction and vouchers, the required budget can be package, and (iii) an augmented service delivery package. reduced to PKR 311 billion (US$1.37 billion) in 2035. It considers how these costs might vary using alternate This paper is a product of the Education Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at ahasan1@ worldbank.org. A verified reproducibility package for this paper is available at http://reproducibility.worldbank.org, click here for direct access. RESEA CY LI R CH PO TRANSPARENT ANALYSIS S W R R E O KI P NG PA The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Ensuring an Equal Start for All Pakistani Children: What Will It Cost? Abdullah Alam ♦ Amer Hasan ♣ Elizabeth Hentschel• Keywords: early childhood education, Pakistan, cost of universal access JEL Classifications: H52, I10 ♦ Aflatoun International and Education Global Practice, South Asia Region, World Bank. ♣ Education Global Practice, South Asia Region, World Bank. • Harvard T.H. Chan School of Public Health and Education Global Practice, South Asia Region, World Bank. Introduction Approximately 20 million school-aged children in Pakistan did not attend school in 2021 (Pak Alliance for Maths and Science, 2021). Learning poverty in Pakistan is also extremely high, estimated to be 78 percent for 10-year-old children (World Bank, processed). Low participation in pre-primary education has major implications for human capital development, productivity, and skill acquisition (Holla et al., 2021). Research suggests that the availability of quality education in the early years could be one pathway for countries to achieve SDG-4 targets and tackle challenges such as poor access to education, retention in school, and marginalization (Rad et al., 2022). However, the net enrollment rate of children aged 3-5 in early childhood education (ECE) was only 31 percent in 2022 in Pakistan (Tomlinson et al., 2023) – far short of the SDG 4.2 target of universal access to quality pre-primary education. If Pakistan fails to close the universal ECE coverage gap, each consecutive cohort remaining at the present enrollment rate will cost Pakistan 3.4 percent of its GDP (Tomlinson et al., 2023). Data limitations do not allow realistic planning and decision-making for the ECE sub-sector in Pakistan’s education system. There are currently no mechanisms in place to record the budgetary allocations and spending on ECE in federal or provincial budgets. Pre-primary and primary education budgets are combined into one budget and cannot be disaggregated. Cognizant of the cost of inaction, Pakistan is committed to expanding ECE services for all children. To do so, it is imperative to have a robust strategic plan in place that relies on a realistic understanding of the existing investments in pre-primary education and scenarios for additional investments in the upcoming years. Despite the gaps in existing data, there are mechanisms through which this information gap can be filled in the short to medium term until such systems are developed and implemented. This paper is one attempt to fill this information gap. It presents a simulation exercise using an adapted version of the ECE Accelerator Costing and Simulation Model 1 to outline the ECE costs, financing gap, and human resources and infrastructure needed to achieve government goals and SDG 4.2 targets. The aim is to facilitate evidence-based planning and decision-making for the ECE subsector in the upcoming years. The model also considers alternative scenarios for meeting targets by 2035. The discussion section considers system capacities and additional support mechanisms required for these endeavors. This paper draws on administrative data sources, the population census, and publicly available data on government spending to consider what a core package of ECE services might cost in Pakistan. It considers goals such as those articulated in SDG 4.2 as well as prospective goals being considered by the government. 2 This paper presents cost estimates for three packages: (i) a business-as-usual package, (ii) a core service delivery package, and (iii) an augmented service delivery package. Various approaches to service delivery are costed including community construction and vouchers. 1 The development of this tool was supported by UNICEF in collaboration with others. The original tool is available at https://www.ece-accelerator.org/resources/early-childhood-accelerator-simulation-model. 2 The analyses presented in this paper were motivated by a government request to understand the costs of expanding ECE provision across Pakistan. 2 Cost of Inaction Improving human capital development in Pakistan by improving ECE attendance and the quality of ECE provided has immediate costs and long-term benefits. In 2022, Pakistan’s Human Capital Review calculated the cost of inaction of failing to invest in ECE (Tomlinson et al., 2023). The cost society pays for failing to implement such an intervention was estimated for three scenarios in Pakistan. 3 Low coverage was estimated by equalizing enrollment rates between households in the highest income quintile and those in the other quintiles to achieve 27.2 percent enrollment across income levels (based on the Pakistan Social and Living Standards Measurement Survey 2019-20 enrollment rates for children in the highest income quintile). Medium coverage was estimated using the enrollment rate of a regional peer, Nepal, at 61.9 percent, and full coverage was estimated as achieving one year of universal ECE enrollment, meeting Pakistan’s commitment to SDG 4.2 and 2009 National Education Policy objectives. Continuing coverage at present levels and not raising coverage levels would be expensive. The cost of inaction would be 0.34 percent of GDP for the low-coverage scenario (failing to boost enrollment to 27.2 percent of all 3- to 5-year-olds in preprimary school), 1.79 percent of GDP for the medium-coverage scenario (failing to enroll 61.9 percent of 3- to 5-year-olds), and 3.39 percent of GDP for the full-coverage scenario. The cost of inaction on medium coverage is conservatively estimated at US$ 4.7 billion for each cohort that fails to receive medium coverage. The cost of failing to invest in ECE is high. As such, the next step is understanding the upfront costs of investment in ECE. Evidence from Other Countries Costing and simulation models are fairly common in education sector planning. However, because of data availability and consolidation challenges, ECE is often not included as its own education level. In most cases, ECE is grouped together with primary education. However, there are examples of countries that have been able to achieve significant milestones through better planning and budgeting. For example, Uzbekistan developed dedicated strategies and costing simulations for preschool education as part of its Education Sector Plan 2019-2023 (Government of Uzbekistan, 2019). The preschool enrollment rates increased from less than 30 percent in 2017 to 69 percent in 2022 (Ministry of Preschool Education, Uzbekistan & UNICEF, 2022). In addition, more than 12,000 non-state family preschools were established between 2018 and 2021 using public-private partnership models. Given the need to estimate ECE costs, a variety of different ECE costing and simulation modeling tools have been developed and utilized in various periods and contexts. Some well-known tools 3 The approach used a benefit–cost ratio from Engle et al. (2011), data on the population aged 3–5-years from the 2020 Pakistan Census, government expenditure per primary student from World Bank (2015), and GDP from World Bank (2020). The benefit-cost ratio was estimated at a conservative 6.4:1 and a discount rate of 6 percent was applied based on a study modeling the impact of increased ECE attendance on school attainment and subsequent wage changes in 73 low- and middle-income countries (LMICs) (Engle et al., 2011). 3 include the UNICEF WCARO tool, 4 CARICOM costing model, 5 Estimator Cost Analysis Tool,6 Brookings Childhood Cost Calculator, 7 ILO Care Policy Investment Simulator, 8 and UNESCO’s SDG 4.2 costing model. 9 For the case of Pakistan, UNICEF’s ECE Accelerator Costing and Simulation Model was preferred over other available models because it is open source, easy to adapt, and allows the planners and decision-makers to use the model iteratively. The ECE accelerator simulation model was developed in 2021. Governments have been using the tool for ECE sub-sector planning. For example, Lesotho used the results of the simulation exercise to undertake focused advocacy with the government and public representatives to increase funding for ECE. The resulting analysis suggested that universal expansion would not be feasible given available resources. The government instead opted to support the expansion of one year of ECE. The costing exercise also resulted in leveraging additional funding from development partners for ECE. Data and Methods The approach This paper analyzes data from the Pakistan Bureau of Statistics (2017 Census) 10 and the provincial Education Management Information Systems 11 on the population size, enrollment rates, and achievement rates (dropout and retention rates) in ECE. Data on existing infrastructure, human resources, ECE and education financing, and unit rates for key inputs such as teachers and basic ECE supplies were also collected from each provincial government. These inputs were categorized under 3 heads: salary, non-salary (such as payment of electricity bills), and development (such as construction of classrooms). Policy options and targets for ECE were then defined based on best practices and in line with national and provincial policies and implied service standards. Enrollment estimates from the PSLM were updated. Using data on population growth and enrollment in public and private ECE we use an estimated enrollment rate of 31% for all scenarios in this paper. Enrollment rates were then projected for 2023-2035 using the population projections contained in the most recent published census files. 12 Infrastructure and human resource 4 https://www.ece-accelerator.org/resources/unicef-west-and-central-africas-regional-prototype-supporting-cost- effective-policy-development-early-childhood-development 5 https://www.humanitarianlibrary.org/sites/default/files/2014/02/ECD-CoNo30-rev.pdf 6 https://www.costtoolkit.org/ 7 https://www.brookings.edu/articles/the-childhood-cost-calculator-c3/ 8 https://www.ilo.org/globalcare/?language=en#simulator 9 https://bangkok.unesco.org/sites/default/files/assets/article/Early%20Childhood%20Care%20and%20Education/file s/Costing%20and%20Financing%20SDG%204.2%20Webinar/2.1_Kristy%20Bang.pdf 10 https://www.pbs.gov.pk/content/final-results-census-2017 11 http://www.emis.gob.pk/; https://semis.rsu-sindh.gov.pk/; https://open.punjab.gov.pk/schools/; https://ese.kp.gov.pk/page/education_management_information_system_emis; http://hrmis.fde.gov.pk/; 12 The rates used for each province are available online at https://www.pbs.gov.pk/content/final-results-census-2017- 0, Table 1. These are 2.4 for Pakistan overall, 2.13 percent for Punjab, 2.89 percent for KP, 2.41 for Sindh, and 3.37 for Balochistan. While newer population growth rates are available from the 2023 census, age-wise population estimates are not yet published. Hence these newer growth rates were not used. 4 requirements were based on prevailing policies and service standards in each province. 13 An example of the calculations undertaken for enrollment is presented below. 1 2024 = 1 2023 ∗ (1 + %) = ℎ 1 2025 = 1 2024 ∗ (1 − %) = 1 2 The dropout rate is the difference between the enrollment in pre-primary year 1 (2022) and year 2 (2023) taken as a proportion of pre-primary year 1 (2022) enrollment. Using unit rates, the costs for ECE provision were calculated for the 2023-2035 period, incorporating inflation – which was assumed to be 5 percent per year over this period (except for salaries for which a 10 percent inflation was assumed). The costs were disaggregated by salary, non-salary, and development heads. The required financing for ECE and additional investments needed were then calculated as follows: = � + � − + � An illustrative calculation The model presents various policy scenarios from a planning, decision-making, and investment perspective. An illustrative calculation using this approach is presented below. • Balochistan has a baseline (2022) population of 1.4 million boys and 1.3 million girls in the 0-5 age group. The enrollment at the pre-primary level includes 156,011 boys and 104,809 girls (including the public and private sectors), with a net enrollment of 19 percent at the pre-primary level. The government allocated PKR 4.5 billion (US$ 0.02 billion) for ECE, which is 4 percent of the education budget allocation for 2022. 14 • Policy decisions in Balochistan suggest that schools must have a Student-Classroom Ratio of 40:1 and a Student-Teacher Ratio of 30:1. Given existing enrollments, classrooms and teachers – there is a shortfall of both rooms and teachers in Balochistan. • The model estimates the cost of addressing the current gap of teachers and classrooms in the next twelve years as well as of achieving 62 percent enrollment by 2035. This requires a total enrollment of 871,253. To deliver quality ECE in public schools and to comply with existing service standards, 16,248 additional teachers will need to be hired in the next 12 years. The model assumes these salaries grow with a 10 percent inflation rate. The current average annual salary of teachers in the system is estimated to be PKR 384,000 (US$ 1,696). For instance, a student to classroom ratio of 40:1 was used. 13 14 For the purposes of estimates reported in this paper, the ready exchange rate of the State Bank of Pakistan as of December 30, 2022 is used. This was PKR 226.4 to USD 1. 5 • Likewise, to ensure these students are not enrolled in overcrowded rooms – an additional 22,358 rooms will be needed over the next 12 years. • To provide an augmented service delivery package (enhanced infrastructure, improving quality of service delivery, providing school meals, etc.), the current budget for ECE in Balochistan must be increased from PKR 4.5 billion (US$ 0.02 billion) to PKR 28.5 billion (US$ 0.13 billion) by 2035. This corresponds to an annual increase of 14 percent on average. 15 o If alternate delivery mechanisms (community construction and vouchers) are used, the budget requirement by 2035 will be substantially reduced to PKR 19.4 billion (US$ 0.09 billion), with an annual increase of 11 percent on average. The simulation modeling relied heavily on data availability, which varied from province to province. Due to data limitations, several assumptions are made throughout data consolidation and analysis. Population, public enrollment, and unit rates are largely available for the exercise. However, data on human resources and infrastructure was unavailable at the pre-primary level. The figures were, however, available at the primary level. We assume that if a primary school has six classrooms and six teachers, one of them should be for the pre-primary level. So, the total number of teachers at primary level was divided by six to calculate the pre-primary level figures. Similar assumptions were used for baseline education financing data. The baseline enrollment in private schools was estimated based on private to public enrollment at the pre-primary level ratio using the latest available Pakistan Education Statistics Report 2020-21 (Pakistan Institute of Education, 2022). The data on teacher training was also not available. It was assumed that public school ECE teachers are currently untrained in ECE teaching and must undergo extensive training. Table 1 provides an overview of the baseline data used for simulation modeling. 15 These calculations assume that the baseline year is 2022. Estimated figures are projected up to 2035 – resulting in 14 periods. The compounding formula used in this case is ((28.5/4.5)^(1/14))-1. 6 Table 1: Baseline data used for simulation modeling, 2022 New Classroom WASH Teacher Population Public Private Construction Facility ECE ECE Salary Province (3-5y) Enrollment Enrollment Cost Provision Classrooms Teachers (annual) millions (3-5y) (3-5y) Millions, Millions, Millions, PKR PKR PKR Balochistan 1.36 185,814 66,766 1,430 2,025 3.1 0.3 0.7 Islamabad Capital 0.15 11,241 79,615 174 298 3 0.42 0.8 Territory Khyber Pakhtunkhwa 3.54 693,349 453,363 2,216 2,216 3 0.46 0.21 Punjab 9.1 796,756 1,776,028 19,895 18,560 3 0.54 0.6 Sindh 4.57 822,675 822,675 6,264 16,784 2.5 0.42 1.6 Notes: Public enrollment statistics are from the Annual School Census of each province. Private enrollment statistics are computed based on the share of the population that reports being enrolled in a private school. The number of ECE teachers and classrooms is estimated based on the figures reported for the primary level in each province's Annual School Census. The cost of classroom construction, teacher salary and the cost of providing a WASH facility are provided by provincial education departments. A WASH facility is comprised of one enclosed flush toilet and one sink connected to piped water. 7 Costing Scenarios To build and sustain a well-functioning ECE system in Pakistan, financing support across actors will be needed. It is, therefore, essential to consider costs from multiple angles, including donor support, public financing, the role of civil society organizations, leveraging private sector contributions, and exploration of innovative multisectoral approaches that will maximize investments in the ECE sub-sector. A preliminary examination of the costs of continuing business- as-usual and undertaking a realistically ambitious financing approach for ECE in Pakistan follows. The paper presents costs for three packages of service delivery: a business-as-usual package, a core package, and an augmented package. It also presents costs for alternate delivery mechanisms that do not solely rely on the public sector. The business-as-usual approach means that provincial governments follow existing budgetary trends for ECE and allocate ECE budgets in the future based on them. This is without routing any significant additional resources for the sub-sector. It does, however, rely on construction of additional classrooms using existing approaches. The core package focuses on using existing ECE facilities to the maximum – thereby minimizing additional construction. 16 This package assumes no additional investments into infrastructure. The augmented service delivery package includes services that are not currently offered to ensure quality ECE delivery in the provinces. For example, capacity building of ECE human resources and providing school meals to a subset of children (on the basis of need) have been included in the simulation models. These are in addition to constructing additional classrooms and providing additional human resources for the ECE sub-sector. The alternate delivery approach introduces cost-effective service delivery mechanisms that can also deliver quality ECE. For this paper, we consider providing education vouchers to parents for enrolling their children in private schools and community construction as alternate modalities. 17 Results The simulation exercise was undertaken for the four provinces of Pakistan and the Islamabad Capital Territory (ICT). The results are presented below at the provincial and ICT levels, and then aggregated at the country level. The costs are calculated using two criteria: a) provinces reaching 62 percent net enrollment in ECE (medium coverage) and b) 100 percent net enrollment in ECE (full coverage). In 2022, Pakistan earmarked a budget of PKR 70.64 billion (US$ 0.31 billion) for ECE out of the total education budget of PKR 1,345 billion (US$ 5.94 billion). The allocation for ECE equals 5.3 percent of the total education budget for the year. While there are efforts underway to ensure that governments allocate at least 10 percent of the education budget to ECE (UNICEF, Education Commission & LEGO Foundation, 2022), in reality, few countries achieve this threshold. The allocations for ECE in Pakistan are well below this global benchmark (Figure 1). 16 For example, no new classrooms are constructed during the simulation period and existing classrooms are used to accommodate new enrollment. This could be through double shift schools as one example. 17 There are various other alternate modalities possible. 8 Figure 1: ECE budget as a proportion of the education budget, 2022 10% 9% 8% 6.9% 7% 6% 5.6% 5.3% 4.9% 5% 4.3% 4% 3% 2% 1% 0.5% 0% ICT Balochistan Sindh Khyber Punjab Pakistan Pakhtunkhwa A provincial analysis highlights that none of the provinces currently achieve this benchmark target. As per the figure above, in 2022 Punjab spent the highest portion of its education budget on ECE (6.9 percent), followed by Khyber Pakhtunkhwa (5.6 percent), Sindh (4.9 percent), Balochistan (4.3 percent), and ICT (0.5 percent). Business as Usual The business-as-usual approach means that the provinces allocated ECE budgets per the ongoing trends in previous years. This is without routing any significant additional resources for the sub- sector. This approach also sets a baseline for our analyses to see what the additional investments will look like if the government opts to deliver only a core package, an augmented package, or if it chooses to deploy alternate service delivery mechanisms. Table 2 outlines the budgetary projections for ECE at the provincial levels for a business-as-usual approach which would result in a net enrollment rate of 43 percent in ECE by 2035; an increase of 12 percent points compared to the current ECE net enrollment rate of 31 percent. As an illustration, based on ECE budget trends in the last five years, Khyber Pakhtunkhwa is expected to demonstrate the highest average annual percentage increase (13 percent) in the ECE budget. In 2035, it will also be the province with the highest allocation for ECE in absolute terms. The remaining three provinces and ICT are projected to increase their ECE budgets by 5-7 percent annually. Overall – across Pakistan, using a business-as-usual approach, the ECE budget will rise from PKR 71 billion (US$ 0.3 billion) in 2022 to PKR 217 billion (US$ 0.96 billion), representing an increase of 8 percent annually. 9 Table 2: ECE budget projections for a business-as-usual approach, billions of PKR Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 4.7 5.0 5.3 5.6 6.0 6.3 6.7 7.1 7.5 8.0 8.4 8.9 9.5 6% Islamabad Capital 0.6 0.7 0.7 0.8 0.9 0.9 1.0 1.1 1.1 1.2 1.3 1.4 1.6 1.7 7% Territory Khyber 15.5 17.7 20.2 23.2 26.5 30.3 34.7 39.8 45.6 52.3 60.0 68.8 79.0 90.7 13% Pakhtunkhwa Punjab 33.8 36.2 38.8 41.5 44.4 47.5 50.9 54.5 58.3 62.4 66.8 71.5 76.6 82.0 7% Sindh 16.3 17.2 18.1 19.1 20.2 21.3 22.5 23.7 25.0 26.4 27.9 29.4 31.1 32.8 5% Pakistan 70.6 76.5 82.8 89.8 97.5 106.0 115.4 125.7 137.1 149.9 164.0 179.6 197.1 216.6 8% Notes: A business-as-usual package assumes that the provinces allocate ECE budgets per the ongoing trends in previous years without significant additional investments for the sub-sector. 10 Core Package Considering resource limitations, it is also essential to consider the scenario where the existing facilities are utilized to the maximum. This means no additional investments in infrastructure. Table 3 shows the budget projections for achieving 62 percent net enrollment in ECE. To achieve the medium coverage of 62 percent net enrollment in ECE through a core package of services, Pakistan must increase its ECE budget by 9 percent on average annually. In aggregate terms, if this percentage increase is maintained, the budget for ECE will increase from PKR 71 billion (US$ 0.3 billion) in 2022 to PKR 237 billion (US$ 1.05 billion) in 2035. Khyber Pakhtunkhwa needs the highest annual growth of 14 percent to achieve the target of 62 percent net enrollment. Sindh, Balochistan, and Punjab follow Khyber Pakhtunkhwa with an 7 percent yearly increase. ICT must enhance its ECE budget by 7 percent annually. To achieve universal enrollment in ECE (full coverage) by 2035 and provide a core package of ECE services means an annual increase of 10 percent in the ECE budget. Using this annual increase, the budget earmarked for ECE will grow from PKR 71 billion (US$ 0.3 billion) to PKR 261 billion (US$ 1.15 billion) in 2035. Khyber Pakhtunkhwa will need the highest increase of 14 percent, while ICT, Sindh, and Punjab need a yearly increase of 8 percent on average. Balochistan has to enhance its ECE budget by 9 percent annually to achieve universal enrollment at the pre- primary level. Table 4 shows the budgets needed to achieve a universal enrollment scenario by providing core ECE services. 11 Table 3: ECE budget projections for medium coverage (62 percent enrollment) using a core package, billions of PKR Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 5.5 5.9 6.3 6.7 7.2 7.7 8.3 8.9 9.6 10.3 10.2 10.9 11.8 7% Islamabad Capital 0.6 0.7 0.7 0.8 0.9 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 8% Territory Khyber 15.5 19.5 22.2 25.4 29.0 33.1 37.9 43.3 49.5 56.7 64.9 71.5 81.8 93.9 14% Pakhtunkhwa Punjab 33.8 39.5 42.5 45.7 49.1 52.8 56.9 61.3 66.0 71.2 76.8 78.2 84.0 90.6 7% Sindh 16.3 18.5 19.2 20.3 21.5 23.0 24.9 26.4 28.5 30.4 32.6 33.9 36.6 39.3 7% Pakistan 70.6 83.8 90.5 98.4 107.2 117.1 128.3 140.4 154.1 169.1 186.0 195.3 214.9 237.3 9% Notes: A core package assumes using the existing ECE facilities to the maximum such that there are no additional investments into infrastructure and curtailing supporting interventions like school meals, which are added as part of the other packages considered in this paper. 12 Table 4: ECE budget projections for full coverage (100 percent enrollment) using a core package, billions of PKR Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 5.6 6.0 6.4 6.9 7.5 8.1 8.8 9.6 10.5 11.5 11.7 13.0 14.5 9% Islamabad Capital 0.6 0.7 0.8 0.8 0.9 1.0 1.1 1.1 1.2 1.3 1.5 1.6 1.7 1.9 8% Territory Khyber 15.5 20.0 22.9 26.2 30.0 34.3 39.4 45.1 51.8 59.5 68.4 75.9 87.2 100.5 14% Pakhtunkhwa Punjab 33.8 40.0 43.1 46.5 50.2 54.2 58.7 63.5 68.8 74.7 81.2 83.7 90.8 99.1 8% Sindh 16.3 19.9 20.8 22.2 23.8 25.5 27.3 29.3 31.6 34.1 36.8 38.3 41.4 45.0 8% Pakistan 70.6 86.3 93.6 102.2 111.8 122.5 134.5 147.9 163.0 180.1 199.3 211.1 234.0 260.9 10% Notes: A core package assumes using the existing ECE facilities to the maximum such that there are no additional investments into infrastructure and curtailing supporting interventions like school meals, which are added as part of the other packages considered in this paper. 13 Augmented Package The simulation models for all the provinces include additional strategies to ensure quality ECE delivery in the provinces. These include capacity building of teachers, caregivers, and parents, for example, and providing school meals to a select number of children. Including these strategies in the ECE planning and implementation processes in the coming years is represented here as the augmented package. This also means the construction of additional classrooms and providing additional human resources for the ECE sub-sector. Table 5 presents the budget projections to achieve medium coverage (achieving 62 percent net enrollment in ECE) through an augmented package. To achieve the target of 62 percent net enrollment in ECE, Pakistan must increase the ECE allocations by 11 percent annually on average. For ICT, an increase of 18 percent will be needed annually. At the provincial level, Balochistan needs to increase its ECE allocations in the upcoming years to the tune of 14 percent annually on average. For Khyber Pakhtunkhwa, an annual increase of 15 percent will be needed. Punjab and Sindh will need to increase their ECE budgets by 9 percent annually on average. For the universal access scenario with an augmented package of services, a significant increase is warranted from all the provinces (Table 6). All four provinces must increase their ECE budgets by more than 10 percent per annum on average to achieve the 100 percent net enrollment target. The ask is most challenging for Balochistan and ICT, which will need to boost the per-year increase from 6 percent and 7 percent (business-as-usual approach) to 18 percent and 19 percent, respectively, to achieve universal access to pre-primary education by 2035. An annual increase of 17 percent by Khyber Pakhtunkhwa, while Punjab and Sindh will need to increase their annual allocations for ECE by 11 percent each. In aggregate terms, Pakistan will need to increase the ECE allocations from PKR 71 billion (US$ 0.3 billion) in 2022 to PKR 418 billion (US$ 1.85 billion) in 2035, highlighting an average annual increase of 14 percent. 14 Table 5: ECE budget projections for medium coverage using an augmented package Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 10.5 11.3 12.5 13.8 15.3 16.9 18.7 20.7 22.9 25.1 23.3 25.7 28.5 14% Islamabad Capital 0.6 1.4 1.7 2.1 2.5 3.0 3.5 4.0 4.6 5.2 5.5 5.8 6.2 6.6 18% Territory Khyber 15.5 27.6 30.7 34.8 39.4 44.5 50.4 57.1 64.7 73.3 82.8 84.9 96.3 109.7 15% Pakhtunkhwa Punjab 33.8 51.6 55.2 59.8 64.8 70.2 76.1 82.5 89.5 97.1 105.1 100.8 109.0 118.4 9% Sindh 16.3 24.9 25.9 27.5 29.8 32.5 36.5 39.1 43.4 46.6 50.5 47.5 52.1 56.3 9% Pakistan 70.6 116.0 124.8 136.7 150.3 165.4 183.4 201.4 222.9 245.2 269.0 262.3 289.3 319.5 11% Notes: An augmented package assumes an augmented service delivery package and includes additional strategies to ensure quality ECE delivery, like capacity building of ECE human resources and providing school meals to children. 15 Table 6: ECE budget projections for full coverage using an augmented package Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 11.4 12.5 14.1 15.9 17.9 20.3 23.1 26.4 30.2 34.2 34.9 40.4 47.0 18% Islamabad Capital 0.6 1.5 1.9 2.2 2.7 3.1 3.6 4.2 4.8 5.4 5.8 6.1 6.5 6.9 19% Territory Khyber 15.5 31.4 35.2 40.1 45.8 52.3 59.8 68.3 78.1 89.3 101.8 107.6 123.4 141.8 17% Pakhtunkhwa Punjab 33.8 54.7 58.9 64.4 70.4 77.0 84.4 92.6 101.8 112.0 123.1 122.5 135.1 149.9 11% Sindh 16.3 30.9 32.5 35.2 38.2 41.5 45.1 49.1 53.5 58.4 63.5 60.4 66.1 72.5 11% Pakistan 70.6 130.0 140.9 156.0 172.9 191.9 213.2 237.3 264.5 295.3 328.4 331.5 371.4 418.1 14% Notes: An augmented package assumes an augmented service delivery package and includes additional strategies to ensure quality ECE delivery, like capacity building of ECE human resources and providing school meals to children. 16 Alternate Modes of Service Delivery The needs of the education system, including for the early years, are substantial. It is, therefore, essential to identify means of making education financing effective and efficient. One such approach is to introduce alternate service delivery mechanisms that are cost-effective but can also deliver quality ECE. For this paper, we have considered two mechanisms: a) providing education vouchers to parents for enrolling their children in private schools and b) community construction, which is more cost-effective and relies on localized services. Table 7 presents the budget projections for medium coverage (achieving 62 percent net enrollment in ECE) through an alternate mode of service delivery. For Pakistan to achieve the target of 62 percent net enrollment in ECE through a mix of traditional and alternate delivery mechanisms, the ECE allocation would need to increase by 10 percent annually on average. At the provincial level, ICT would need to increase its ECE allocations by 18 percent annually on average in the upcoming years. The average increase required for Khyber Pakhtunkhwa and Balochistan will be 14 percent and 11 percent, respectively. For Punjab and Sindh, an additional allocation of 8 percent per year will be needed, respectively. Table 8 describes the costs of achieving universal enrollment in ECE by 2035. Using a mix of traditional and alternate models of ECE service delivery will result in an annual increase of 11 percent in Pakistan’s ECE budget. Using this annual increase, the budget earmarked for ECE will grow from PKR 71 billion (US$ 0.13 billion) to PKR 311 billion (US$ 1.4 billion) in 2035. ICT will need the highest increase of 18 percent, while Khyber Pakhtunkhwa and Balochistan will have to increase the ECE budget by 15 percent and 13 percent annually, respectively. Punjab and Sindh need a yearly increase of 9 percent on average. The table below looks at the budgets required to achieve a universal enrollment scenario by providing ECE services through a mix of traditional and alternate mechanisms. 17 Table 7: ECE budget projections for medium coverage using an alternate delivery mechanism Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 7.3 8.0 8.8 9.7 10.7 11.8 12.9 14.2 15.7 16.9 16.6 17.9 19.4 11% Islamabad Capital 0.6 1.4 1.7 2.1 2.5 2.9 3.4 3.9 4.5 5.1 5.4 5.7 6.1 6.4 18% Territory Khyber 15.5 21.7 24.7 28.2 32.3 36.9 42.2 48.1 55.0 62.8 71.3 77.4 88.1 100.6 14% Pakhtunkhwa Punjab 33.8 42.6 45.9 49.5 53.5 57.8 62.5 67.6 73.1 79.0 85.1 85.9 92.3 99.6 8% Sindh 16.3 20.7 21.7 23.2 24.9 26.9 29.6 31.7 34.5 37.0 39.7 40.2 43.3 46.4 8% Pakistan 70.6 93.8 102.0 111.8 123.0 135.3 149.4 164.3 181.3 199.7 218.5 225.9 247.8 272.4 10% Notes: An alternate delivery package introduces cost-effective service delivery mechanisms that can also deliver quality ECE. For example, providing education vouchers to parents for enrolling their children in private schools and community construction as the alternate modalities. 18 Table 8: ECE budget projections for full coverage using an alternate delivery mechanism Average Province 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Annual Inc % Balochistan 4.5 7.6 8.4 9.3 10.4 11.6 12.9 14.5 16.2 18.1 20.0 20.6 23.0 25.9 13% Islamabad Capital 0.6 1.5 1.8 2.1 2.6 3.0 3.5 4.0 4.6 5.2 5.5 5.9 6.2 6.6 18% Territory Khyber 15.5 23.1 26.3 30.2 34.7 39.8 45.6 52.3 60.0 68.8 78.5 86.0 98.4 112.9 15% Pakhtunkhwa Punjab 33.8 43.8 47.3 51.3 55.7 60.5 65.8 71.5 77.9 84.9 92.3 94.7 102.9 112.4 9% Sindh 16.3 23.2 24.5 26.4 28.5 30.7 33.2 35.9 38.9 42.2 45.4 45.9 49.5 53.6 9% Pakistan 70.6 99.2 108.2 119.3 131.8 145.6 161.0 178.2 197.5 219.2 241.7 253.1 280.0 311.4 11% Notes: An alternate delivery package introduces cost-effective service delivery mechanisms that can also deliver quality ECE. For example, providing education vouchers to parents for enrolling their children in private schools and community construction as the alternate modalities. 19 Discussion While resources are scarce, it is worth noting that multiple traditional and innovative funding mechanisms are available to fund ECE. In order to tap into these innovative resources, it is important to know the quantity of the challenge and funding gaps. This calls for developing a strategy to agree on the vision for ECE, outline the targets, and maintain a balance between ambitiousness and fiscal and operational realities. Pakistan spends 5.3 percent of its education budget on ECE. To ensure 100 percent net enrollment in ECE by 2035, Pakistan needs to boost its ECE spending up to 10.4 percent of its education budget by 2035. This would mean increasing the ECE budget from PKR 71 billion (US$ 0.13 billion) in 2022 to PKR 418 billion (US$ 1.85 billion) in 2035, allowing an average annual increase of 14 percent. This increase will allow for human resources and infrastructure improvements, implementing strategies such as the capacity building of teachers, caregivers, and parents, for example, and providing school meals to a select number of children. ICT would need to increase its ECE budget by 19% percent on average for this to happen. Balochistan and Khyber Pakhtunkhwa would need to increase their annual ECE spending by at least 18 percent and 17 percent, respectively. On the other hand, to reach 62 percent net enrollment (enrollment rate in peer country, Nepal), the resource requirement by 2035 is estimated to be PKR 319 billion (US$ 1.4 billion). The largest increase needed would be from ICT (18 percent), followed by Khyber Pakhtunkhwa (15 percent), Balochistan (14 percent), Punjab (9 percent), and Sindh (9 percent). Given the resource limitations and the steep financial requirement, the federal and provincial governments could also consider a core package of services. The core services will exclude additional strategies to improve the quality of services and service delivery. In this package, the governments would curtail extra infrastructural investments and undertake commitments such as school meals. To do this, Pakistan would need to increase its current ECE spending from PKR 71 billion (US$ 0.13 billion) in 2022 to PKR 261 billion (US$ 1.15 billion) in 2035. For the medium coverage (62 percent), the resource requirement would be PKR 237 billion (US$ 1.05 billion) in 2035. To do this, Khyber Pakhtunkhwa must increase its annual ECE spending by at least 14 percent. Balochistan, Punjab and Sindh would need to increase the ECE budget by 7 percent, while ICT would need to increase its ECE allocation by 8 percent annually. Over the years, Pakistan has experimented with numerous alternate delivery mechanisms for ECE and education service delivery in general. Given the financial requirements for achieving the committed enrollment targets as part of national and international commitments, these alternate modes are worth considering. Through a combination of education vouchers and community construction, the ECE budget requirement could be reduced from PKR 418 billion (US$ 1.85 billion) in 2035 for the augmented package to PKR 311 billion (US$ 1.37 billion) for achieving 100 percent net enrollment at the pre-primary level. The annual increase would decrease to 11 percent from the earlier stated 14 percent. To achieve the 62 percent net enrollment target employing alternate delivery mechanisms, the national ECE budget will need to be increased from PKR 71 billion (US$ 0.13 billion) in 2022 to PKR 272 billion (US$ 1.2 billion) in 2035. 20 This paper provides targets and estimates of the resource requirements for ECE in the next 12 years (See Table 9). Notwithstanding the budgetary implications, good planning, political prioritization, and consistency can facilitate reaching such targets. A robust ECE system will improve school readiness, student retention in primary and secondary education, transition to higher educational levels, and education outcomes. Given the benefits of ECE, considering the current state of ECE financing in Pakistan, and because of the resource requirements outlined above, this paper provides decisionmakers with a framework to invest additional resources in ECE and ensure they are used effectively and efficiently. Different policy options should be considered to develop a long-term ECE service delivery and financing strategy. The example scenarios described in this paper all call for a substantial increase in ECE financing over the next decade. However, the analysis considers modalities that could help lessen this burden and facilitate a sequenced approach. There is also an opportunity for federal and provincial governments to invest in tracking expenditures on pre-primary education at aggregate, functional, and operational levels. Doing so would allow Pakistan to report ECE spending to international databases, such as the UIS – something it does not currently do. Such tracking is crucial for understanding the state of ECE financing, gaps, and compliance status with international and national benchmarks. The government, civil society organizations, and development partners might also consider focusing on implementing community-based approaches and public-private partnership modalities for ECE. This could lower the burden on the public education system and reduce the funding requirements for achieving SDG 4.2 targets. Policy makers can use simulation exercises such as those used in this paper to better plan service delivery for children in the early years. Limitations This exercise is not without constraints and limitations. First, pre-primary budgets are grouped with primary education in Pakistan; therefore, the ECE budgetary allocations and expenditures are unknown. To address this, we used a part of the primary education budget proportionate to the number of grades at the primary level as the pre-primary education budget. However, this is not a robust way of estimating budgets. Second, the human resource and infrastructure figures are not disaggregated for ECE in the provincial EMIS and other data systems. For provinces where this data was unavailable, we made proportionate calculations like the ones mentioned above. Third, the enrollment data for private schools is not readily and accurately available in the EMIS/NEMIS or other data systems. Nor is data on children with special needs (population, enrollment, etc.) accurately available. Fourth, multigrading at the ECE level is not identifiable through EMIS systems. Lastly, there are inconsistent ECE policies within provinces: e.g., caregivers are a mandatory part of the public ECE system in Punjab but not in other provinces. Future Directions This paper presents a preliminary status of ECE financing in Pakistan, outlining estimates not available before at the national or provincial level. However, the data systems in the country must be tailored to capture the planning and costing needs of the ECE sub-sector. Updating these estimates regularly and with more precise data would allow an opportunity to track progress and adjust as needed. 21 Table 9A: Summary of scenarios in billions of PKR Modality 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Business as usual (without additional 76.5 82.8 89.8 97.5 106.0 115.4 125.7 137.1 149.9 164.0 179.6 197.1 216.6 interventions) Augmented package (without additional 130.0 140.9 156 172.9 191.9 213.2 237.3 264.5 295.3 328.4 331.5 371.4 418.1 support from the private sector) Alternative delivery modalities (combine public provision, targeted vouchers to 99.2 108.2 119.3 131.8 145.6 161.0 178.2 197.5 219.2 241.7 253.1 280.0 311.4 leverage private sector and community- based models) “Core” package (without new major 86.3 93.6 102.2 111.8 122.5 134.5 147.9 163.0 180.1 199.3 211.1 234.0 260.9 capital expenses) Table 9B: Summary of scenarios in percent of GDP 18 Modality 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 Avg Business as usual (without additional interventions) 0.09 0.10 0.10 0.11 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.13 Augmented package (without additional support from the private sector) 0.15 0.16 0.18 0.19 0.21 0.22 0.24 0.26 0.29 0.31 0.31 0.34 0.37 0.25 Alternative delivery modalities (combine public provision, targeted vouchers to leverage private sector and community-based models) 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.20 0.21 0.23 0.24 0.25 0.28 0.19 “Core” package (without new major capital expenses) 0.10 0.11 0.12 0.12 0.13 0.14 0.15 0.16 0.18 0.19 0.20 0.21 0.23 0.16 18 Percent of GDP computed using GDP at current market prices of 84,068.76 in 2023 from the SBP and an average annual growth rate of 2.5% as per SBP. 22 References Denboba, A.; Mclin, Monica Ellen; Neuman, Michelle J.; Sayre, Rebecca Kraft; Ying, Wang Y.; Wodon, Quentin T., (2014). Investing in young children: an eLearning course (English). Education notes; No. 4 Washington, D.C.: World Bank Group. Engle, P. L., Fernald, L. C., Alderman, H., Behrman, J., O’Gara, C., Yousafzai, A., De Mello, M. C., Hidrobo, M., Ulkuer, N., Ertem, I., and Iltus, S. (2011). Strategies for Reducing Inequalities and Improving Developmental Outcomes for Young Children in Low-income and Middle-income Countries. The Lancet 378 (9799): 1339–1353. Government of Uzbekistan (2019). Education Sector Plan of Uzbekistan 2019-2023. Holla, A., Bendini, M., Dinarte, L., & Trako, I. (2021). Is Investment in Preprimary Education Too Low? Lessons from (Quasi) Experimental Evidence across Countries. Policy Research Working Paper 9723. Ministry of Preschool Education, Uzbekistan & UNICEF (2022). Towards Achieving Universal Early Childhood Education in the Republic of Uzbekistan. Pak Alliance for Maths and Science (2021). The missing third: An out of school study of Pakistani 5-16 year olds. Islamabad: Pak Alliance for Maths and Science. Pakistan Institute of Education (2022). Pakistan Education Statistics 2020-21. Islamabad: Pakistan Institute of Education. Pakistan Bureau of Statistics. (2021). Pakistan Social and Living Standard Measurement Survey (PSLM), 2019–2020. Government of Pakistan. Rad, D., Redes¸ A., Roman, A., Ignat, S., Lile, R., Demeter, E., Egerau, A., Dughi, T., Balas, E., Maier, R., Kiss, C., Torkos, H., & Rad, G. (2022). Pathways to inclusive and equitable quality early childhood education for achieving SDG4 goal—a scoping review. Frontiers in Psychology, 13. Tomlinson, H., Hentschel, E., Chowdry, M., Ansari, A., Zamand, M., Yousafzai, A., & Hasan, A. (2023). Pakistan Human Capital Review. Fostering early childhood development, 65–91. UNICEF, Education Commission & LEGO Foundation (2022). Add today multiply tomorrow: Building an investment case for Early Childhood Education. World Bank (processed). Pakistan Learning Poverty Brief, Oct 2023. 23