Additional Financing to the Third Primary Education Development Project (RRP BAN 42122) ECONOMIC AND FINANCIAL ANALYSIS 1. This document provides an analysis of the economic rationale for additional financing of the ongoing government Third Primary Education Development Program (PEDP3), which is supported by a consortium of development partners and the Third Primary Education Development Project of the Asian Development Bank (ADB). 1 It analyzes the financial and budgetary affordability of the total financing, including additional financing under PEDP3; poverty impacts; and the sustainability of the primary education system beyond the program period. PEDP3 was designed to improve the quality and efficiency of and access to a full 5-year primary education. However, because of slower revenue collection arising from the deteriorated political situation and the government s decision to nationalize a large number of schools, an additional year will be needed to complete the program, which will also incur additional costs. An updated economic analysis has been conducted to reassess the economic justifications of investment in additional financing and quality interventions needed to complete PEDP3 successfully. The economic analysis is an extension of the one completed during original PEDP3 project processing; it updates actual spending in the first 3 years of PEDP3 (FY2012 FY2014) and planned spending during FY2015 FY2017, and assesses the benefits and costs associated with the entire program. A. Background 2. Improving students learning, reducing disparities, and enhancing the capacity of education service delivery were the basic goals of PEDP3. Since it launched in 2011, reform activities including the Each Child Learns initiative, pilot testing of the Diploma in Primary Education, improvement of the Grade 5 examination, pre-primary education, school level improvement plan grants, and nationalization of registered nongovernment primary schools (RNGPSs) have been started. Despite these, disbursement of the development budget part of PEDP3 has been slow, particularly for key quality and disparity related reforms. Disbursement of the non-development budget, which includes recurrent expenditures such as teacher and staff salaries, has been on track disbursing about 94% of what was planned. However, only about 35% of development budget funds planned to be disbursed by year 3 (FY2012 FY2014) has been spent. Extra time will be needed for implementing activities under PEDP3 as a result of delays in disbursement in the early stage. Taking these into consideration, the medium-term budget framework (MTBF) assumes an extension of the development budget implementation period by 1 year until FY2017. B. Economic Analysis 3. Rationale for additional financing. Although the MTBF expects continuously stable growth of public expenditure on education, a funding gap may emerge during the remaining PEDP3 period primarily because of the (i) increased recurrent cost caused by a rise in teacher salary levels, and (ii) increased unit cost of infrastructure. Under the ongoing PEDP3, potential funding gaps arise because (i) the deteriorated political situation in the middle of FY2014 slowed economic activities, resulting in a reduction in revenue collection and lowering the budget allocation for PEDP3; (ii) the government decided to nationalize around 35,000 RNGPSs, placing fiscal pressure by funding additional 115,000 teacher positions associated with these schools; and (iii) the unit cost of construction rose as a result of the change in standard school 1 ADB. 2011. Report and Recommendation of the President to the Board of Directors: Proposed Loan to the People s Republic of Bangladesh for the Third Primary Education Development Project. Manila (Loan 2761-BAN).
2 design and the impact of inflation on building materials, which increased the cost of nondevelopment and development expenditures under PEDP3. 4. One of the most important policy changes implemented since the inception of PEDP3 is the nationalization of RNGPSs (now called newly nationalized government primary schools [NNGPSs]), starting with around 23,000 schools in 2012. While the government had provided salary subsidies to RNGPSs, teachers salaries increased from an average Tk6,000 per month per teacher to Tk10,500 for around 115,000 NNGPS teachers who were officially on the civil service payroll from FY2014. In addition, the overall national pay scale revision increases the salary level of teachers by 40% 60% from FY2015. As a result of these policy changes related to teacher salaries, the total salary costs for all government primary schools (GPSs and NNGPSs) is estimated to increase by 65% within two years from FY2015. Thus, nationalization of RNGPS teachers and the national pay scale revision have made today s primary education system require more operating budget than before. In addition to the expected rise in recurrent costs, the need for an increased capital budget is due to higher unit costs of infrastructure. Two factors contributed to the increase in unit costs of infrastructure: (i) change in the standard design for schools, and (ii) inflation in the cost of building materials. What was originally planned for activities including reconstruction of around 2,700 schools and construction of around 32,000 additional classrooms would not be possible to complete as a result of significant cost increase. The average unit cost per classroom has also already increased by 22%. 5. Rationale for intervention. The anticipated resource gap demands additional funds for implementing the ongoing quality and disparity focused reforms under PEDP3. Without additional financing, funds could be shifted from the quality and disparity components, which were underspent in the first 3 years of the program. However, if the shift takes places to cover costs of the infrastructure component, PEDP3 would be re-characterized as an infrastructure project. The increased recurrent and capital expenditures must be accommodated while maintaining the focus on quality and disparity so that PEDP3 meets its results objectives.. Timely implementation of quality improvement activities is crucial for generating higher economic benefits from PEDP3. Delays in the quality improvement reforms could hamper the future of millions of students in Bangladesh. Therefore, the additional financing is necessary for maintaining the momentum of improving the quality of education. 6. The economic viability of PEDP3 is assessed based on the following internal rate of return (IRR) analysis and cost benefit analysis. C. Internal Rate of Return Analysis 7. The potential benefits of PEDP3 can be quantified using the latest Bangladesh Household Income and Expenditure Survey (HIES) 2010. 2 Table 1 shows the distribution of median monthly earnings by level of education. Population with no education have median monthly earnings of about Tk5,921 per month. As the education level rises, people earn higher incomes the median earnings are more than doubled with tertiary education. Therefore, the returns on primary education are expected to be high, and even higher if the pupil proceeds to tertiary level. 2 Bangladesh Bureau of Statistics. 2010. Household Income and Expenditure Survey. Dhaka.
3 Table 1: Median Monthly Earnings Education Level Amount (Tk) No education 5,921 Incomplete primary 4,670 Complete primary 6,334 Incomplete secondary 6,565 Complete lower secondary 7,932 Higher secondary 10,365 Tertiary 12,333 Source: Bangladesh Bureau of Statistics. 2010. Household Income and Expenditure Survey. Dhaka. 8. The distribution of average monthly per capita income by educational attainment of head of household and location shows that heads of households with higher education earn more, on average, than heads of households with a lower level of education (Table 2). The average monthly per capita income of heads of households is about Tk1,966 for no class passed compared with Tk6,246 for postgraduates. Table 2 also shows that heads of households in urban areas earn more on average than heads of households in rural areas, reflecting more educated households in urban areas as well as lower wages in rural areas. Table 2: Average Monthly per Capita Income by Educational Attainment of Head of Household (Tk) Education Level National Urban Rural No class passed 1,966 2,461 1,858 Primary school 2,505 3,294 2,227 Secondary school 2,892 3,358 2,684 Higher secondary school 4,510 5,579 3,646 Graduate 5,642 6,534 4,080 Postgraduate 6,246 7,101 4,140 Source: Bangladesh Bureau of Statistics. 2010. Household Income and Expenditure Survey. Dhaka. 9. The IRR analysis based on the 2005 HIES data using a Heckman selection indicated that the average social rate of return to an additional year of schooling for Bangladesh was 14.3% for age 15 45, which is higher than the 10.4% average private rate of return (Table 3). 3 The average private returns to schooling for females are higher than those of males in both age groups whereas the private returns to another year of education in rural areas remain as low as 7.8% 8.2% (Table 3). Table 3: Private and Social Rates of Return by Gender and Location All Ages Age 15 45 Population Private Social Private Social Male 0.098 0.131 0.095 0.132 Female 0.128 0.158 0.131 0.172 Rural 0.078 0.108 0.082 0.121 Urban 0.118 0.155 0.115 0.154 All 0.106 0.139 0.104 0.143 Source: Bangladesh Bureau of Statistics. 2005. Bangladesh Household Income and Expenditure Survey (from World Bank, Secondary Education Quality Enhancement Project, Bangladesh). 10. Based on the results of estimating a Mincerian wage regression using the 2005 HIES data, the earnings of a Grade 5 9 completer for all ages was estimated to be 13.6% higher than 3 Economic and Financial Analysis. ADB. 2011. Report and Recommendation of the President to the Board of Directors: Proposed Loan to the People s Republic of Bangladesh for the Third Primary Education Development Project. Manila (Loan 2761-BAN).
4 the estimate for a worker with no formal education. 4 Because of the poor quality of schooling in Bangladesh and the cost of attending school, many children especially the poor and those living in rural areas are constrained to withdraw or drop out of school. Hence, by improving the overall quality of primary education, the additional financing under PEDP3 also intends to improve the labor market outcomes for poor boys and girls in rural areas, raising the returns on schooling in those areas. D. Cost Benefit Analysis 11. The cost benefit analysis (CBA) has been conducted on the whole program ($9,808 million) original program ($7,476 million), including additional financing ($2,332 million). It uses the methodology and techniques in ADB guidelines, including the Guidelines for the Economic Analysis of Projects and Framework and Criteria for the Appraisal and Socioeconomic Justification of Education Projects. 5 For ease of computation, the CBA calculates the net present value (NPV) and economic internal rate of return (EIRR) for primary school students enrolled in Grade 5 in 2012. 12. General assumptions. The following approach and assumptions have been used in the analysis: (i) The official average exchange rate of Tk77.7 = $1.00 (FY2015) from Bangladesh Bank is used for converting foreign exchange costs to local currency equivalent. (ii) All prices are valued in a domestic price numeraire in FY2015 constant prices. (iii) Economic prices of investment costs and recurrent costs are estimated by converting the financial prices with a shadow exchange rate factor of 1.05 for traded goods (net of taxes and duties), a factor of 1.0 for non-traded goods, a shadow wage rate factor of 1.0 for skilled labor, 6 and a shadow wage rate factor of 0.75 for unskilled labor. 7 (iv) Price contingencies and interests are excluded in the calculation of the EIRR, while physical contingencies are included because they represent the monetary value of additional real resources that may be required beyond the base cost to complete the program. (v) Taxes and duties are excluded because they represent transfer payments. (vi) A 12% economic discount rate has been used. (vii) The economic life of the project is assumed at 30 years. 13. Additional assumptions on cost estimation. Costs include total program costs incurred by the government and development partner sources. The estimated direct costs of the program comprise the project investment, and the public and private costs of each additional pupil-year, comprising both recurrent and capital costs. The annual recurrent and capital costs 4 World Bank. 2008. Project Appraisal Document: Proposed Credit to the People s Republic of Bangladesh for A Secondary Education Quality & Access Enhancement Project. Washington, DC. 5 ADB. 1997. Guidelines for the Economic Analysis of Projects. Manila; ADB. 1994. Framework and Criteria for the Appraisal and Socioeconomic Justification of Education Projects. Manila. 6 It is assumed that there are no significant distortions in the wage rates of skilled labor. 7 The shadow exchange rate factor is the ratio of the economic price of foreign currency to its market price, while the shadow wage rate factor is the ratio of the economic to financial cost of labor. Conversion factors for the shadow exchange rate factor and shadow wage rate factor are based on recently approved ADB-financed projects in Bangladesh: (i) ADB. 2014. Report and Recommendation of the President to the Board of Directors: Proposed Loan and Administration of Loan to Bangladesh for the Third Urban Governance and Infrastructure Improvement (Sector) Project. Manila; and (ii) ADB. 2014. Report and Recommendation of the President to the Board of Directors: Proposed Multitranche Financing Facility to Bangladesh for the Skills for Employment Investment Program. Manila.
5 are estimated by applying a proportion (Grade 5 students enrolled comprise 14.7% of total enrollments in 2012) 8 to the total cost of primary education allocated over the 6-year program. 14. Assumptions on economic benefit. As in the original PEDP3 CBA analysis, the benefits are changes in the quantity and quality of education, and a reduction in the internal inefficiency of education produced over the period as a result of PEDP3. The benefits are broadly assumed from two possible sources: (i) an increased number of primary school completers who earn higher wages (relative to non-completers); and (ii) an increase in the quality of education, resulting in a higher wage premium for all primary school completers. 15. Quantification of benefits. For the benefit estimations, the base case assumptions are as follows: 9 (i) The CBA is specific to the primary school students enrolled in Grade 5 in 2012 (a total of 2,793,922), graduating in 2013. Benefits were assumed to be realized progressively from the first year of implementation, as Grade 5 completers will be able to be employed right after graduation. (ii) The project capital and recurrent costs are spread over the 6-year implementation period (with an additional year), while the project benefits from lifetime earnings for the Grade 5 completers (in the form of higher employability and a higher wage premium) are assumed to continue for 30 years. (iii) Of the Grade 5 students enrolled, 85% will complete primary school. 10 (iv) Of the completers, 65% will be employed for an average of 25 days per month, 11 months per year. 11 This number of days and months considers employment in either formal and/or informal sectors. It is also assumed that progressive uptake of employment by primary education graduates will take place as a result of the improved quality of education. This accelerates gradually from year 2 of project implementation: year 2, 20%; year 3, 30%; year 4, 40%; year 5, 50%; year 6, 60%; year 7, 70%; year 8, 80%; year 9, 90%, and year 10, 100%. (v) The CBA assumes the future additional daily wage for primary education completers compared with non-completers. The incremental daily wage resulting from completion of Grade 5 schooling is conservatively estimated at Tk50.0 (or $0.6) per day for the base case assumption. 12 8 Ministry of Education, Bangladesh Bureau of Educational Information and Statistics. 2013. Bangladesh Education Statistics 2012. Dhaka. 9 The assumptions for estimating benefits for the current CBA differ from the assumptions due to the following reasons: (i) the program cost, both in terms of local currency and dollar terms changed considerably with added financing and exchange rate fluctuations; (ii) data on new cohort of primary graduate is available; (iii) both primary school completion rates and wage rates have changed over the period with updated data. 10 Directorate of Primary Education, Monitoring and Evaluation Division. 2014. Bangladesh Primary Education Annual Sector Performance Report 2014. Dhaka. The actual primary school completion rate was 80.0% in 2013. However, as the dropout rate declines with grade progression, a higher completion rate of 85% is assumed for Grade 5 enrollments. 11 Ministry of Education, Bangladesh Bureau of Educational Information and Statistics. 2013. Bangladesh Education Statistics 2012. Dhaka. The actual secondary school gross enrollment rate was 54.0% in 2012. However, a dropout rate of 15.0% from Grade 6 to Grade 7 is also considered while choosing a rough estimate for the employment rate of 55.0% (65% of 85% completers) for Grade 5 completers. 12 Estimated based on the income data from Bangladesh Bureau of Statistics, Statistics Division. 2011. Report to the Household Income & Expenditure Survey 2010. Ministry of Planning, Bangladesh, or HIES 2010. As shown in Table 1, those with complete primary education earn Tk66 per day higher than those with incomplete primary education. It is also evident from HIES 2010, while the average income gap of households between no education and any Grades 1-5 completers is Tk20 per day, the highest level of income groups between these two (as this analysis only considers the wage rate of Grade 5 completers, which is supposed to be higher than lower grade completers), the gap rises to Tk106 per day. In addition, with economic expansion since 2010, a higher wage gap is expected during this period.
6 16. Net present value, economic internal rate of return, and sensitivity analysis. Based on a discount rate of 12% for the benefit and cost streams described above, economic rate of return analysis shows that the PEDP3 interventions, including additional financing for FY2015 FY2017, are economically sensible. The EIRR of the entire program, which includes investing an additional $2,332 million over the next 3 years of PEDP3, is 18.8% indicating an economically sound investment (Table 4). The results, including the sensitivity analysis with low case scenarios for wage rate, and employment condition, with an EIRR of 12.2% 15.7% (Table 5), show that PEDP3 (original financing with additional financing and extension of one additional year) is expected to be a very sound investment. Table 4: Economic Internal Rate of Return based on Base Case Scenario (Tk 000) Year Aggregate Costs Capital Recurrent Total Benefits Additional Wage of Trainees Net Benefit NPV at 12% 1 58,773 11,189,904 11,248,677 0 (11,248,677) (10,043,462) 2 1,690,477 11,596,507 13,286,984 4,245,015 (9,041,969) (7,208,202) 3 2,598,183 13,101,721 15,699,904 6,367,523 (9,332,382) (6,642,605) 4 2,878,233 17,089,420 19,967,653 8,490,030 (11,477,623) (7,294,237) 5 3,090,434 19,825,744 22,916,178 10,612,538 (12,303,640) (6,981,416) 6 3,726,291 19,664,919 23,391,209 12,735,046 (10,656,164) (5,398,744) 7 0 0 0 14,857,553 14,857,553 6,720,803 8 0 0 0 16,980,061 16,980,061 6,857,962 9 0 0 0 19,102,569 19,102,569 6,888,578 10 0 0 0 21,225,076 21,225,076 6,833,906 11 0 0 0 21,225,076 21,225,076 6,101,702 12 0 0 0 21,225,076 21,225,076 5,447,948 13 0 0 0 21,225,076 21,225,076 4,864,240 14 0 0 0 21,225,076 21,225,076 4,343,071 15 0 0 0 21,225,076 21,225,076 3,877,742 16 0 0 0 21,225,076 21,225,076 3,462,270 17 0 0 0 21,225,076 21,225,076 3,091,312 18 0 0 0 21,225,076 21,225,076 2,760,100 19 0 0 0 21,225,076 21,225,076 2,464,375 20 0 0 0 21,225,076 21,225,076 2,200,335 21 0 0 0 21,225,076 21,225,076 1,964,585 22 0 0 0 21,225,076 21,225,076 1,754,094 23 0 0 0 21,225,076 21,225,076 1,566,155 24 0 0 0 21,225,076 21,225,076 1,398,353 25 0 0 0 21,225,076 21,225,076 1,248,529 26 0 0 0 21,225,076 21,225,076 1,114,758 27 0 0 0 21,225,076 21,225,076 995,320 28 0 0 0 21,225,076 21,225,076 888,678 29 0 0 0 21,225,076 21,225,076 793,463 30 0 0 0 21,225,076 21,225,076 708,449 ( ) = negative, EIRR = economic internal rate of return, NPV = net present value. Source: Asian Development Bank estimates based on project data and projections. NPV = 34,778,063 EIRR = 18.8%
7 Table 5: Net Present Value, Economic Internal Rate of Return, and Sensitivity Analyses Scenario NPV (Tk 000) EIRR (%) Low incremental wage scenario Tk13,951,566 14.8 (Tk40 per day) Low employment scenario (55.0%) Tk18,757,681 15.7 Switching scenario: low incremental wage scenario (Tk40 per day) and low employment scenario (55.0%) Tk1,135,261 12.2 EIRR = economic internal rate of return, NPV = net present value. Source: Asian Development Bank estimates based on project data and projections. 17. The EIRR based on the CBA is reasonable, reflecting positive externalities and longerterm intergenerational social benefits that come with increased levels of education. The benefits are estimated for completion of the primary cycle only. No allowance is made for the fact that the improvements in primary education will enable more children to attend secondary school and thereby have access to higher earning power. It is also assumed that children currently enrolled in primary school, especially in senior grades, will not benefit because they will complete their primary education (or drop out) before the full impact of the project is felt. E. Financial Analysis 18. Budget envelope scenarios. The estimates of resource envelopes likely from the government are based on (i) the likely share of the education sector in the total budget estimates, (ii) likely shares of the primary and mass education subsector, and (iii) the likely contribution to PEDP3. 19. The fast and stable economic growth of Bangladesh has generated a positive influence on public expenditure on education. Gross domestic product (GDP) growth of 6.1% in FY2014 (nominal growth rate of 13.5%) was similar to the assumption at the original project appraisal; and in FY2015, the education sector continues to receive one of the largest shares of the national budget. The combined allocation ($3,746 million) to the Ministry of Education and the Ministry of Primary and Mass Education (MOPME) is 7.6% of total public expenditure in FY2015. Of the total education sector budget, MOPME has received a constant share of about 45.0%, on average, since 2005. In FY2015, the allocated budget for MOPME is $1,534 million. 20. Third Primary Education Development Program financing. It is estimated that development partners would commit $1,055 million (total [$1,722 million] less realized expenditure [$554 million] during FY2012 FY2014) during the next three fiscal years (FY2015 FY2017) with additional financing of $667 million (Table 6). It is estimated that the government will finance an additional $1,665 million (71.4% of the total PEDP3 additional financing). Development partners contribution for additional financing would be 29.0% of the total PEDP3 additional financing. Table 6: Third Primary Education Development Program Financing Plan ($ million) Description FY2012 FY2014 FY2015 FY2016 FY2017 Total Original Budget Additional Financing Total (amount) 3,693 1,840 2,154 2,121 9,808 7,476 2,332 Government share (%) 85 75 86 81 82 86 71 Government share (amount) 3,139 1,387 1,851 1,708 8,185 6,420 1,665 Development partner share (%) 15 25 14 19 18 14 29 Development partner amount 554 453 303 412 1,722 1,055 667 Source: Asian Development Bank estimates based on project data and projections.
8 21. Fiscal viability. The long-run fiscal viability of PEDP3 lies within the original framework of PEDP3, which depends on adequate budget to the MOPME to maintain the investment plan. The estimated resource envelope and the current gross domestic product growth trend indicate that it will not be difficult for the Ministry of Finance to allocate the required amount for PEDP3 (FY2015 FY2017). The allocation of a higher share of budget to primary education by the government is ensured since (i) the government s national education policy proposes a significant increase in budget allocation to the education sector; and (ii) development partner support is linked to both results and assurances that sector budget allocations are in line with the medium-term budgetary framework. 22. Fiscal sustainability. Sustainability considerations are built directly into the PEDP3 design through improvements in budgetary processes to refocus program implementation under a results-based management model and strengthening institutions to continue the flow of improvements in quality and efficiency of the sector. Efficiency gains are expected to benefit Bangladesh s primary education sector as a result of the reforms being supported by PEDP3, including improved financial management, electronic procurement, and strengthened monitoring and evaluation, which also keep costs down while improving overall education outcomes. These will continue reinforcing the long-term fiscal sustainability of Bangladesh s primary education sector. Moreover, reliance on external financing will continue to diminish if systemic reforms and efficiency gains are realized as envisaged and the agreed fiscal framework is implemented.