Low Cost Private Schools (LCPS)

GUIDANCE NOTE Engaging the Low Cost Private Schools in Basic Education: Issues, Challenges and Opportunities Throughout the world, non-state schools provided the first formal educational opportunities for children - whether established by individuals, the private sector or religious organizations. Unsurprisingly these were often tuition-based private schools that were only accessible to the wealthy. More recently though, Low Cost Private Schools (LCPS) have become a distinct reality in nearly all developing countries and there has been an increasing awareness that these LCPS can play a significant role in education service delivery through either public private partnerships (PPPs) or purely private sector led development. This Guidance Note provides information on how LCPS can become involved in supporting access to a quality basic education for all. It identifies the contribution that LCPS can make to increasing both the quality and quantity of schooling, complementing as well as challenging state provision either through a PPP contracting arrangement or through private sector led development. A variety of possible interventions is described, as well as the outcomes of current or recent reform initiatives and research in developing countries. These outcomes and the resulting recommendations will certainly not all be relevant for all countries all of the time. Decisions on what path to follow must take into account the stage of development of a country’s economy and current governance and operational contexts. This Guidance Note is particularly focused on the engagement of LCPS in the development and delivery of the primary and secondary education cycles. For the purposes of this Guidance Note, LCPS include 'for profit' and 'not for profit' providers and a range of provision – from conventional schools to contracting LCPS to provide services under public funding arrangements. The Guidance Note is in six sections: Section 1: Executive Summary. Section 2: Policy and Practice Background. This section provides a brief background on the policy reasons why the LCPS market segment has become significant and an exploration of the practical value of investing in partnerships with LCPS, including some of the research supporting and questioning such partnerships. Section 3: Definitional and Conceptual Issues. This section considers some of the semantic and conceptual difficulties involved in analysing LCPS and a look at some of the main similarities and differences between 'for profit' and 'not for profit' LCPS. Section 4: Private Sector Involvement and Partnerships. This section provides an assessment of recent major shifts in approach to LCPS, particularly regarding contracting of education services in low income countries and the relevance of these to future initiatives. 2 Section 5: Potential Initiatives. This section outlines some key lessons learnt and possible initiatives going forward across three focus areas: the partnership framework, the operating environment (with a specific focus on regulation), and finance. Section 6: References and Further Reading. SECTION ONE: SUMMARY 1. Overview Drawing on a number of source documents and in particular ongoing research being conducted under DFID's Operational Research and Impact Evaluation Capability in South Asia programme, 1 this Guidance Note draws the following preliminary conclusions: Strategic  There are key roles for LCPS to play in the delivery of a quality basic education, but their involvement is not a simple undertaking that can quickly reform a failing state system;  Involvement of LCPS in PPPs can be effective, but these arrangements have been difficult to implement because the government has historically decided which partnership model would be the most appropriate;  For sustainable results, there needs to be a comprehensive system for dialogue in which both the state and the private sector participate as partners;  Public consensus is also critical because, in practice, opposition can arise due to ideological concerns, and concerns about the equity and fairness of privatisation; and  The respective roles of the state and the private sector must be clearly delineated according to the country context, including governance, financial management and administrative capacity of public and private providers, the size and nature of the LCPS sector and the government’s fiscal situation. Regulatory  The partnership decision depends on the government’s capacity to regulate and effectively control the quality of education delivery; and  Different contracting models, with several common characteristics including the Government as funder/regulator but not necessarily the provider, provision of perstudent funding to schools, autonomous management, accountability and a strong focus on outcomes, are currently being used. Financial  Disadvantaged groups – such as girls/women in some contexts, or the very poor – still require specific, targeted demand-side support as they cannot be reached by LCPS that charge the relatively high fees needed to operate their school;  While 'for profit' and 'not for profit' LCPS share characteristics in their educational approaches, there are differences in their commercial approaches which have implications for state funding (reference the Bangladesh LCPS experience);  It is especially difficult to find private investors willing to fund LCPS, particularly if those LCPS operate as 'not for profit' entities (reference the Gyan Shala experience); and  LCPS are delivering education at a unit cost that is comparable to or less than the unit cost of similar public education providers (reference the Foundation Assisted Schools [FAS] experience); 1 Funded by DFID Bangladesh, DFID India, DFID Nepal and the DFID South Asia Research Hub. 3 Research  There is still a lack of data and comparative analysis on education outcomes to be able to assess value for money from the government perspective and for prospective students and their families to distinguish between high-quality and low-quality public and private providers;  Different approaches to partnerships with LCPS have been and are being tried in countries with contrasting development and governance characteristics. The results must be interpreted taking those different contexts into account; and  The lack of comprehensive and comparable data on both 'for profit' and 'not for profit' LCPS continues to be problematic for educators, researchers, aid organizations and policymakers alike. SECTION 2: POLICY AND PRACTICE BACKGROUND 2. Why is there growing partnership with the Low Cost Private Schools? The international community has recognized the centrality of basic education to development. The second Millennium Development Goal (MDG) seeks to achieve universal primary education and 164 governments committed to the Education for All (EFA) global initiative under the Dakar Framework for Action in 2000. Despite these major commitments and targets, access to basic education, its quality, and literacy levels remain major challenges in developing contexts throughout the world - as evidenced in Figure 1 with the number of projected children that are estimated will be out of school in 2015. Figure 1: Projected Numbers of Out-of-School Children to 2015 Source: UIS database (data for 1999-2007) According to the UNESCO Institute of Statistics (UIS), nearly 68 million children are not in school worldwide, more than half of whom are girls and, of those, 25 percent reside in South Asia trapped in the vicious cycle of poverty and extreme disadvantage. Globally, 795 million adults, 64 percent of whom are women, still lack basic literacy skills (UNESCO, GMR 2011).The capacity of education systems to provide quality education is hampered by a number of variables, such as a lack of resources, governance gaps, uneven capacities and unpredictable emergencies (e.g. natural disasters, conflicts and post-conflicts). Although primary education is often seen as a state responsibility, non-state providers can play an important role in expanding access to education where public provision is not meeting local needs (Rose, 2006). In South and West Asia, and particularly in Pakistan, alternative education providers, such as LCPS, are 4 increasingly being called upon to fill this vital gap in educational services in order to help improve access to education in the region (Andrabi et al., 2002). Figure 2 illustrates the considerable growth in FAS in the Punjab Province of Pakistan. Figure 2: Growth in FAS in the Punjab, Pakistan - 2005-2013 Source: Punjab Education Foundation data (2013) This growth of LCPS in the developing world raises some significant questions including: What role can and should LCPS play in supporting the global drive for basic primary education? How sustainable and replicable are these alternative forms of educational provision? And how do LCPS contribute to an equitable access for all to a quality basic education? “Low-cost private education is an important, complementary element of education in developing countries and should be seen as an active partner with governments looking to ensure all children have access to a high quality education. We are convinced that affordable schools, operated on a for-profit basis, can make a big difference. We have examined carefully the challenges and opportunities of this sector, and we’re committed to sharing our knowledge with governments, donors and aid agencies and to working with them on this urgent, global mission. 2 There is now a growing appreciation that non-state providers of education can play a significant role in the delivery of education services. Alongside the problems of public delivery, there are also often problems in the quality of the education and parents have sometimes turned to private education as an alternative. Amongst those turning to private schools are families from poor backgrounds who are served by the emergence and proliferation of these LCPS (Harma, 2011). Recent research in South Asian countries such as India, Pakistan and Bangladesh and African countries such as Nigeria and Kenya indicate that private schools have emerged which cater for the poor.3 These schools extend across a continuum of charity or non-government organization (NGO) funded schools to 'for profit' schools. Despite this increasing awareness and popularity, private schools in developing and fragile states are still viewed with some suspicion. Private funding and delivery of education services are often perceived as a threat to state authority, rather than complementary or agents of government programmes. In the case of 'for -profit' institutions, the profit motive is often viewed as contrasting with the perception of education as a social rather than commercial good. This has resulted in reluctance amongst some governments and NGOs to support private schooling 2 Sir Michael Barber, Pearson’s Chief Education Advisor and Chairman of the new Pearson Affordable Learning Fund sourced from www.affordable-learning.com 3 http://www.dfid.gov.uk/Site-search/?q=low+fee+private+schools 54 204 1,111 1,085 1,406 1,725 1,749 2,153 2,153 0 500 1,000 1,500 2,000 2,500 2005 2006 2007 2008 2009 2010 2011 2012 2013 Number of Schools Year 5 and has meant that some governments have banned the existence of private schools or have limited the number of schools that can be established. Despite these challenges, the number of LCPS is increasing and parents are choosing to educate their children there - often stretching their meagre household incomes to do so - and increasingly organizations like UNICEF and its development partners are interested in advancing the partnership agenda to help ensure the realization of the rights of all children, especially the poor, envisioning the state as an enabler as well as one of the providers of education (UNICEF, 2011). DFID is also increasingly committed to flexible and responsive approaches to education challenges, including partnerships with non-state providers. SECTION 3: DEFINITIONAL AND CONCEPTUAL ISSUES 3. Definitions Several key terms and concepts must be defined in order to ensure the Guidance Note's consistency. 1. Low-cost: There is no universally accepted definition of low-cost or 'affordable' private schools in the literature. This lack of agreed terminology clearly has an impact on the generalizability of studies on the sector. LCPS can be viewed either in the context of the cost incurred by the user relative to their household income or the cost of operating the school. USAID (2011) defined LCPS as those with tuition rates less than 50 percent of the minimum wage while Barakat (2012) considers that schooling costs should not exceed 4 percent of the household budget to be considered low-cost. 2. Private school: UNESCO regards as “private” any educational institute that is controlled and managed by an NGO (e.g. religious group, association, enterprise) or if its governing body consists mainly of members not selected by a public agency (UNESCO, 2005). There are many types of private school - some are solely dependent on the financial support of a development partner or charitable organization; others are run by for-profit companies while some private schools are associated with religious organizations and offer a religious-based curriculum, while others are secular. Figure 3 illustrates the three main categories (public, LCPS 'for profit' and LCPS 'not-for-profit') according to the following three criteria - ownership, management or provision/funding. 6 Figure 3: Allocation of School Categories by Ownership, Management and Provision/Funding 3. Public Private Partnerships: There is no single, internationally accepted definition of PPP. This Guidance Note takes a broad view, with PPPs defined as 'A long-term contract between a private party and a government agency, for providing a public asset or service, in which the private party bears significant risk and management responsibility' (World Bank, 2012). Figure 4 illustrates the differences between the 'not-for-profit' and 'for profit' in an alternative financing and provision arrangement. This Figure illustrates how the LCPS purely private segment (Category C) falls into the top left quadrant while the partnership models combining public finance and private provision fall into the bottom left quadrant. Figure 4: Finance and Provision by School Type Provision Financing Private Public Private Private schools Home schooling Tutoring User fees Public Vouchers Contract schools Charter schools Contracting out Public schools Source: Patrinos et al (2010) This Guidance Note refers to LCPS that are not solely dependent on government financial assistance in the long-term, or, if they are currently dependent on such assistance, have a clearly defined plan to become self-sustaining within a specified amount of time. In addition, the Guidance Note is limited to LCPS whose curriculum is not solely religious. Figure 5 illustrates the complexity of classifying the different types of private schools in India (Bangay & Latham, 2012). 7 Figure 5: Classification of Private Schools in India Private schools in India can be categorised into: Government schools: wholly state financed often including the provision of free uniforms, school meals and textbooks. Government schools follow the state curriculum (aligned to the national curriculum) to a set timetable and school hours. Teachers are hired and assigned by state department of education; Government aided schools: these are privately managed and follow government regulations on curriculum, timetable, school hours, textbooks and eligibility criteria for teacher recruitment. Up to 95% of funding can come through state ‘grants-in-aid’ including: teacher salaries and recurrent non teacher spending. Private un-aided (recognised) schools are self-financing; however, they have registered with government having fulfilled a minimum set of standards. This category of schools includes a diverse range from India’s elite schools to English medium schools catering for the emerging middle class. The last category is the Private un-aided (unrecognised) school. These are self-financing schools not registered with any government agency and they include the burgeoning LFPS sector. 4. Research Background Despite the proliferation of LCPS in developing countries, there is a paucity of high quality published comparative research in the area covering either alternative modes of public financing and LCPS provision or LCPS private sector segment alone. Improving efficiency, access, and quality, along with attracting private sector funds, are the prevailing motivations for pursuing PPPs with LCPS as an instrument in project implementation, but assessments of how PPPs work out and deliver on their promise of efficiency gains and increased access and service levels are rare, and if available are only partial. Similarly, delivering quality education at an affordable price point in a manner that provides a profit or a small development surplus for the operator are the 'modus operandi' of the LCPS private sector segment. Drivers of Growth While the relative size and impact of LCPS vary by country, there are two seemingly ubiquitous reasons for the rise of the sector in developing countries. The demand for LCPS has risen during the past decade and continues to rise because of inadequacies in the public sector (i.e., insufficient supply, low-quality public provision, and/or differentiated demand). These participation and quality inadequacies range from access issues such as the lack of supply of public spaces, other problems such as overcrowding, violence, insufficient attention paid to slow learners or students from other religions to perceived quality issues such as teacher absenteeism, teacher engagement and absence of English language or IT instruction. Figure 6 provides a summary of the main drivers and barriers for the three types of schools that are enrolling children from the lower quintiles. 8 Figure 6: Summary of Growth Drivers and Barriers Type of school Drivers Barriers Government school  Free education (although many charge fees via diverse means)  Lower income families can potentially benefit from various government schemes  All schools are recognised hence students can sit the board examinations  Poor quality of education provided due to (i) insufficient infrastructure and manpower; (ii) attitude and lack of accountability of teachers; and (iii) teacher absenteeism  Accessibility as schools are located at central locations and not close enough to young users Low Cost Private School (For Profit)  Perceived to impart higher education (ie. use of English and higher accountability of teachers)  Easy accessibility and choice  School fees represent high proportion of household income Low cost Private School (Not for Profit)  Lower costs than LCPS for profits and some public schools  Beholden to donors in that are unable to access fee income and only partially subsidised by government Evidence of this strong growth in India is highlighted in this extract from the Gray Matters Capital (GMC) Report Affordable Private Schools (APS) Sector Analysis (2012): The success of the APS sector and increasing opportunities in it has resulted in rising competition in the APS sector. There are about two to five new schools starting every year in growing communities. Larger chain schools (called corporate schools) that were primarily focused on the lower middle-income market (fees >INR1500 [USD 30]) are now attracting relatively affluent APS parents (eg. Gowtham Model Schools, Sri Chaitanya Schools and Narayana Concept Schools etc. in Hyderabad). Another interesting trend is the growth of LCPS chains. Currently, nearly one-third of the 200 APS that GMC assessed had at least 2 schools. This provides clear evidence of the success of APS but also the huge demand for such schools in low income communities and proof of the risk taking abilities of these APS leaders who have ventured out to meet this rising market demand. An explanation for the growth of these chains becomes clear when one considers the revenue earnings per child enrolled. Figure 7 shows financial analysis from GMC's Report (2012) that was based on a sample of these 200 LCPS operating out of Hyderabad, Delhi and Bangalore that followed the State Board of their respective states. It shows that more than 80 percent of the schools are making an operational surplus. The operational margin is also impressive at an average of 36 percent (median 32 percent). A typical APS earns revenue of INR 5,300 (USD 106) per student per annum while it incurs expenses of INR 3,600 (USD 72) per student per annum. This allows for a margin of INR1,700 (USD 34) per student per annum – a figure that increases with scale. Figure 7: Financial Ratios by APS Class Segments Ratios Pre-primary Primary Secondary Revenue per Student INR 5,200 (USD104) INR 6,000 (USD120) INR 7,700 (USD156) Expense per Student INR 2,800 (USD 56) INR 4,100 (USD 82) INR 6,400 (USD 128) Surplus Margin 35% 33% 24% Source: GMC (2012) 9 The analysis shows that for the majority of schools, pre-primary classes are the most profitable, followed by primary classes. Secondary classes are the least profitable. This is predictable as the pre-primary classes: (i) do not invest a lot on learning aids, (ii) hire teachers who are below graduate level, and (iii) have the highest student teacher ratio, allowing for less expense per child. LCPS and the Quality of Provision In some quarters, the perceived quality of LCPS is high: ‘‘Low fee private schools’ superiority in terms of ensuring some learning is occurring, along with the drive for English instruction, results in a great boost in demand for LFP schooling and has a direct bearing on parental choice. A majority of the 95 percent of parents stated that their preferred school type was LFP, while only 42% of children were actually accessing them.’’ (Harma, 2009, p. 157). Yet there are mixed empirical results with regard to the quality of LCPS outputs and inputs. No countries have shown consistent findings of private school quality, while the majority has shown that empirical approaches and sampling play a large role in the ultimate outcomes of studies on school quality. Without high-quality studies that correct for selection bias, it is difficult to make any significant claims about the sector in the aggregate, except to say that the current results are mixed. Some studies have shown that LCPS produce low-quality outputs (Bray, 1997 and Rose, 2005), and others such as Gauri & Vawda (2004) have found that a quasi-voucher system does not improve completion and promotion rates. Conversely, some studies have shown that the private system provides more efficient student outcomes after controlling for background characteristics (Lassibille & Tan, 2003). In India ‘‘Extant Indian studies are consistent in suggesting that private schools in India are, on average, more internally efficient than government schools. They are more cost efficient on average costing only about half as much per student as public schools. Private schools are also more technically efficient, producing higher achievement levels (after controlling for student intake) and making for more efficient use of inputs for example having more students per class and lower teacher absenteeism’’ (French and Kingdon, 2010, p. 6). From the perspective of partnerships and the use of public finance and private provision, two of the largest and most thoroughly studied programs in the region (the Chilean and Colombian voucher programs) have both found mixed results regarding the quality of outputs. In Chile, for example, while some researchers have found that the voucher system had positive impacts on test scores and pre-college examinations (Gallegos, 2004; Sapelli, 2003), others found that there was no impact on test scores, repetition rates, or secondary school enrollment rates (Hseih and Urquiola, 2006). As for Colombia, while Angrist et al (2002) found standardized test score gains for voucher recipients and Angrist et al (2004) found that voucher recipients were more likely to take the college entrance exam, Angrist et al (2006) also found no significant difference between college entrance exam results of those receiving vouchers and those who did not receive them. LCPS and Equity of Provision In all six countries reviewed in the USAID (2011) study, low-income families were found trying to place their children in public schools.4 Furthermore, the difference in cost between a free public school and a low-tuition private school was found to be modest in some of the countries while the difference in outcomes was often substantial. 4 The countries were Ghana, Kenya and Tanzania in Africa, Jamaica in the Caribbean, and Indonesia and Pakistan in Asia. 10 As reported above in Ghana, the cost of sending a child to an unregistered private school is 12 percent of the minimum wage. The cost at a registered private school is 20 percent of the minimum wage. The cost of sending a child to a free public school (where the private cost of uniforms, etc is still significant) is 16 percent of the minimum wage. In this instance, it may be less expensive for a low-income family to send its child to an unregistered private school, and it may be more cost-effective to send its child to a registered private school. (USAID. 2011, page 56) This cost differential between a public and LCPS is shown starkly in the case of the FAS Programme in the Punjab Province of Pakistan where a place in a private school cost the Punjab Education Foundation around PKR 6,000 (US $60) a year while a place in a government school cost around PKR 15,000 (US $150) [Barber, 2013]. Yet evidence does suggest that there are serious equity and choice barriers associated with the growth of LCPS. Firstly, there is the 'location' barrier and the fact that the predominant growth of LCPS has been in urban contexts. Secondly, there is poverty. Woodhead et al.’s work (2013) on seven years of data from Andhra Pradesh also points to the critical importance of disaggregating the poor by wealth, finding: ‘‘While attendance at private school increased for all wealth tertiles, the poorest group only achieved 14 percent and the gap in attendance rates between this group and the least poor rural households actually widened from 25 percent percentage points to 42 percentage points during the seven year period.’’ (Woodhead et al., 2013). This inability of the LCPS 'for profit' or 'Affordable' schools to provide a quality basic education to the bottom quintile is evidenced in this extract from the GMC 2012 Report: It appears that below a certain threshold of fees, APS might find it difficult to perform well. From the sample of 200 APS, it is observed that none of the schools charging below INR400 per month (USD 8) are able to perform well. They either lack basic facilities and resources required to create conducive learning environment or are financially unsustainable. Furthermore, Harma finds that poorer households have larger families, thus increasing the education cost burden (Harma, 2011; CREATE, 2011). A third barrier to participation is gender, as reported in work by Kingdon (2007); Rangaraju et al. (2012) and Woodhead et al. (2013). Finally, there is access for the very vulnerable such as those without families at all, street children and AIDS orphans. Figure 8 highlights how spending on LCPS involves many aspects, not least transportation and the provision of midday meals, which a recent DFID SARH study conducted in Bangladesh found were comparable to the tuition fee. 11 Figure 8: Annualized Educational Expenditure in LCPS in Dhaka 2013 Source: DFID SARH (2013b) Summary of LCPS Research Findings Figure 9 provides a brief summary of some of the studies that show differing findings - for and against - LCPS provision in the areas of access, quality and equity. The fact that there is contradictory evidence surrounding the case for a role for LCPS in supporting equitable participation in a quality basic education highlights once again the need for more robust comprehensive and comparable data on both 'for profit' and 'not-for-profit' LCPS. Figure 9: Summary of Research For and Against Role of LCPS Research supporting LCPS Author(s) and Research LCPS can leverage public funds in order to provide access to schooling at rates faster than are possible with public funds alone Kim, et al., (1999) LCPS provide the potential to give access to education for poorer communities Alderman, et al., (2001) LCPS can contribute to addressing gender imbalances in access Andrabi, et al., (2004) and Kingdon (1996) LCPS can provide better mechanisms for accountability to service users Tooley, (2007) Have greater flexibility than their public counterparts, Alderman, et al., (2003). Research querying LCPS Author(s) and Research The quality of provision in LCPS is not proven to be superior than that provided by the public schools Alderman, et al., (2001) and James, (1991) LCPS provision raises concerns about equity of access primarily in terms of income disparities and access of the poor Andrabi, et al., (2002) Gauri & Vawda (2004) LCPS exacerbate gender disparities in access, as well as access inequalities between urban and rural communities Andrabi, et al., (2006) and Harma (2011) Considerable number of LCPS are operating outside of the recognised legal and regulatory frameworks Muralidharan & Kremer (2007) Admission 8% Tuition Examinations 16% 3% Materials 12% Transport 38% Tiffin 17% Uniform 5% Others 1% 12 5. Comparing Low Cost Private Schools Some Examples of the Two Types of LCPS 'For Profit' LCPS: Some Examples5  Omega schools is a social enterprise as well as a 'for-profit' business, creating private schools for the benefit of low-income families. The Omega Schools chain has grown to 20 schools and 11,000 students over the past two years. The separate Omega Schools Foundation was set up in 2009 to undertake research and development in innovative learning methods and the affordable learning sector.  Bridge Academies is a chain of LCPS operating in Kenya, has plans to reach over a million primary students from very poor communities for $4 per child per month. Bridge International has developed a “school-in-a-box” model. This model includes all of the tools, curriculum, materials, systems, processes, training programs, technology, research and monitoring needed to open and run an affordable but high-quality school. Efficient and scalable, the model ensures a high-quality, transparent, supportable, and auditable system. 'Not for Profit' LCPS: Some Examples  BRAC is one of the largest NGOs in the world, supports affordable education in Bangladesh by running nearly 22,618 primary schools and accommodating 750,000 students of which 63% are female, 20,550 are children with special needs, 77,533 are from urban poverty and 46,770 from ethnic communities. In addition BRAC supports progression to secondary schools for 4.38 million students out of the total 4.66 million who have finished primary school.  Gyan Shala, which opened its first school in 2001, offers low-cost basic education to children from very poor backgrounds in urban slums in the states of Bihar, Gujarat, and West Bengal. Gyan Shala’s one-room schools have provided education to children in the urban slums of Ahmedabad city (Gujarat) since 2002, Patna city (Bihar) since 2008, and Kolkata (West Bengal) since 2011. LCPS Similarities and Differences The education offerings among these four LCPS exhibit common features including a distinct school design developed in response to users' needs and circumstances, a specialised curriculum and a scheme for employing a reductionist or prescriptive lesson planning, inclusive provision of a large number of workbooks and learning materials, regular use of assessment and data collection, and an ongoing/'on the job' teacher and manager training and mentoring for a workforce that is often lacking teaching qualifications. Figure 10 shows the levels of qualifications among 142 teachers in 10 public, 10 LCPS and 10 Gyan Shalas in similar urban settings in Ahmedabad (A) and Patna (P). It highlights the small difference between Gyan Shala (a 'not-for-profit' LCPS) and the other 'for profit' LFPS in the same urban area regarding qualifications, but both the private operators have a much lower qualified cadre of teachers in the classroom compared to the public schools where only 16 staff lack a BA or above. 5 http://www.affordable-learning.com/ http://www.brac.net/content/bangladesh-education-primary-schools. 13 Figure 10: Levels of Qualifications of the Different Types of Provider by Percentage Qualifications Gyan Shala LFPS Public Schools A P A P A P Post Graduate - - 5 5 29 29 B.Ed - 5 - - 62 13 BA 45 36 36 37 - 50 Class 12 55 59 42 42 8 8 Class 10 - - 16 16 - - Source : DFID SARH (2013a) Many of the problems that arise in assisting LCPS are shared across the 'for profit' and 'not-forprofit' models. They include a lack of management experience (many directors are new to the profession and have little business experience); inadequate financing (tuition revenues are usually the sole source of finance and there are few opportunities to borrow or to invest in new capital infrastructure); multiple purposes of assistance (religious, street children or orphans, etc.); reliance on a single, charismatic leader (often the founder is uniquely important to a school’s success) and inter-school segmentation – what makes for success in one instance is not easily generalised. These problems are then further compounded by the lack of organisation among LCPS as a group. Although there are commonalities across the education offering, there are perhaps three key differences between the types - compared to their 'for profit' LCPS counterparts, the 'not for profit' LCPS pay their staff lower salaries, charge lower fees and are more likely to be operating outside of the main urban areas (as highlighted in Figure 11). Figure 11: Comparison of Features of the LCPS Education Models Name of business or programme School Fees (monthly) Typical Teacher Qualifications Initial Teacher Training Teacher Salary Geographical Context Omega Initial fee of USD 1.40 + daily USD 0.70 Unknown Unknown USD 44 Urban Bridge USD 3.50 Secondary school 7 weeks USD 59 Urban BRAC Free Secondary school 12 days USD 18-24 Rural and semi rural Gyan Shala USD 0.90 Secondary school 12 days USD 27 Urban and rural Source: http://www.bridgeinternationalacademies.com/Bridge_International_Academies/Our_Model.html and http://www.ratio-magazine.com/201005042964/Kenya/Kenya-A-Commercial-Approach-to-SlumEducation.html; UNESCO Institute for Lifelong Learning (2009), “BRAC Education Programme; DFID (2011) ‘Preliminary study into low fee private schools and education’. LCPS provide different resourcing models in order to meet demand. Figure 12 shows the three types of provider - public and the two types of private provider - across the four different parameters of source of finance, ownership, salaries and main drivers: 14 Figure 12: Comparing School Types by Finance, Ownership, Salaries and Growth Drivers Type Examples Main Source of Finance Ownership Teacher salaries Main driver Government NA Government Government Higher than ‘low cost private’ Education for All Low Cost Private - ‘for profit’ Bridge International Academies – Kenya Omega Schools - Ghana School Fees Private business Lower than government schools Profit Low Cost Private - 'not for profit' BRAC primary schools - Bangladesh Gyan Shala schools - India Grants from development organisations and charities NGOs Lower than government schools Education for disadvantaged communities Sourcing Funding LCPS financial sustainability is a key issue. LCPS that accommodate low-income families are inevitably at risk of financial failure. 'For profits' depend on a range of different sources of income - primarily tuition or tuition in kind - while the 'not-for-profit' LCPS are often overly dependent on a single donor. Because they are fiscally a major risk, banks and other lending institutions are reluctant to engage them in long-term planning or investment. In addition, they sometimes face a tax and regulatory structure that is administratively counter-productive to fiscal stability. The DFID SARH (2013) Study into the potential long-term viability and expansion of Gyan Shala through private sector investment gathered investors' responses regarding investing in the LCPS sector and these investors - ranging across the continuum from private capital to social investors - categorised barriers to investment into the following three areas: finance, legal and regulatory, and operational. Figure 13 provides a summary of their responses: 15 Figure 13: Barriers to Investment in Low Cost Private Education Financial  Minimum expectations of an Internal Rate of Return of 15%6  High capital expenditure requirements in K to 12 education investments - essentially a real estate play  Few models that are investable and scalable  Most models would be better off receiving grant or debt funding  Low Fee Private Education is an urban phenomenon that has yet to spread into the rural localities Legal and Regulatory  Discomfort with the education management company phenomenon as it could come under public scrutiny as a means of evading the 'no profits' stipulation for education  The Right to Education Act Clause 12 has made it compulsory for every fee-paying school in India to admit at least 25 percent of its pupils from poor and low-income families, with the state government reimbursing schools for the fee. This Clause has placed low fee private education operators with a few unpleasant options - raise fees and lose students, lose recognition or, perhaps most critically, lose poor students who can now gain access to higher cost private providers Operational  Little innovation in education delivery models  Very few promoters who are able to manage an operation professionally Source: DFID SARH (2013a) SECTION 4: PRIVATE SECTOR INVOLVEMENT AND PARTNERSHIPS 6 PPPs and the LCPS There are a number of alternative ways in which the government can partner with the LCPS in order to deliver equitable access to a quality basic education. Each way entails the use of public funds and provision of education services by the private operators. These include private management of public schools, government contracting with private schools for the delivery of teaching services, voucher or subsidy programmes, contracting with the private sector for the delivery of educational support services and contracting with the private sector for the provision of school infrastructure and ancillary services. Country Examples of Partnership Example #1: Contracting out of Delivery The FAS Programme in the Punjab Province of Pakistan. Under the FAS Programme, the Punjab Government contracts with private schools to teach students at public expense. The FAS is managed by the Punjab Education Foundation (PEF) that was established by the PEF Act of 1991. The PEF is an autonomous statutory body with the mandate to encourage and promote education in the private sector operating on non-commercial/not-for-profit basis. It is governed by board of directors comprising a private sector chair and public and private membership. Key features include: 6 The internal rate of return (IRR) or economic rate of return (ERR) is a rate of return used in capital budgeting to measure and compare the profitability of investments. 16  Schools paid up to $3.50/month/primary student and $4.00 per month/secondary student they enroll;  Students do not pay tuition or other fees;  Operates at all levels of education – primary to higher secondary;  Schools must have at least 100 students and not more than 500 students (unless approval is sought);  Financial assistance to recipient schools is tied to satisfactory performance on the Quality Assurance Test - administered every 6 months;  Schools that do not meet quality standards can no longer participate in the programme; and  Schools are regularly monitored by PEF staff to ensure basic amenities and teaching aids Figure 14 illustrates the tremendous growth in enrolment in the these FAS and the fact that this growth has impacted equally on girls and boys - in contrast to the findings in India described in Section 2.4: Figure 14: FAS Enrolment (Boys and Girls) – 2005-2013 Source: FAS (2013) Example #2: Payment of Government Subsidy to Private Providers The Education Service Contracting Scheme (ESC) in the Philippines. The Fund for Assistance to Private Education (FAPE) receives an administrative service fee - currently $2.50/grantee for management of ESC and other private education programs. FAPE’s role in respect of ESC is certifying participating private schools, determining the allocation of ESC places to schools based on quotas, collating, reviewing, and forwarding billing statements from participating private schools to the Department of Education, providing in-service training for teachers and principals and monitoring program implementation. In 2012, the ESC budget was USD109 million, there were 722,000 public students receiving grants to study in 2,812 participating private schools - which comprised 40% of all the private secondary schools in the country. Figure 15 shows the considerable increase in enrollments in private schools under the ESC Scheme from 1996/97 to 2012/13: 0 100,000 200,000 300,000 400,000 500,000 600,000 2005 2006 2007 2008 2009 2010 2011 2012 2013 Enrolments Year Male Female 17 Figure 15: Number of ESC Recipients, 1996/97-2012/13 Source: FAPE data (2013) There are many potential models for school contracting. Countries have adapted the general PPP model to a variety of country and developmental contexts. The country context though is critical and it includes the country's governance structures, the financial management and administrative capacity of both the public and private education sectors, and the size and nature of the LCPS. While there are many different contracting models, they tend to exhibit the following main characteristics: the Government has the role of funder and regulator but not provider; funding is provided on a per-student basis; LCPS have greater autonomy in their management; there is strong accountability and the contract is outcomes focused. Figure 16 provides a summary of the perceived benefits and risks to these forms of contracting: Figure 16: Benefits and Risks of Education Contracting Benefits Risks  Efficiency  Quality of service delivery  Specialized skills  Overcome public service operating restrictions rules  Innovative service delivery  Government focus on areas of comparative advantage  Increase access, especially for poorly served groups  Transparency  Complex arrangements, requiring considerable design, implementation and monitoring capacity for public and private sector  Inadequate regulatory environments  Lack of competition in some markets  Quality of service delivery  Difficulty of finding private investors willing to invest • Difficulty in that it is the government that decides which partnership model would be the most appropriate SECTION 5: POTENTIAL INITIATIVES Governments are increasingly turning to PPPs in hopes of improving the efficiency and quality of public basic education delivery. The traditional model of publicly financed and publicly provided basic education service delivery is unsatisfactory. PPPs with the LCPS have gained greater acceptability for a number of reasons: (i) governments find themselves falling short of 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 Number of Grantees Year 18 the large and growing demand for public education investments particularly with the increased pressures on the post primary education sub sectors; (ii) the LCPS sector has deepened its presence and has demonstrated its ability to deliver educational services for the lower economic quintiles particularly in urban slum locations; and (iii) governments are increasingly criticised for poor service delivery and have been compelled to consider other models of basic education service delivery. Partnerships with the LCPS are obviously not going to be the panacea for all the ills that prevent provision of quality educational access for the underserved. In addition, structuring of the partnership needs to address trade-offs between efficiency and equity and strike a fine balance through effective contractual agreements to ensure optimal levels of efficiencies without comprising equity. This Section offers some key lessons learnt and possible initiatives going forward across the focus areas of the partnership framework, the operating environment (with a specific focus on regulation, accountability and risks), and finance. 7. The Partnership Framework Current scenario We know from the weaknesses outlined previously in the Guidance Note that many developing countries struggle to deliver equitable access to a quality basic education despite a large and growing school-age population and that a LCPS sector has increasingly intervened at this basic education. Further, we know that the LCPS market segment is also responding to drivers and barriers in its delivery of a quality basic education for all quintiles. This Guidance Note posits that it requires support and partnership by all types - public and LCPS 'for profit' and 'not-forprofit' alike - in order to attain and maintain access to a quality basic education particularly for the lower economic quintiles and the 'hard to reach'. There are a number of interrelated conditions already influencing the strength of the relationship between governments and LCPS. There is the extent to which they are dependent on each other - in the case of the PEF/FAS for example LCPS are receiving full subsidies for taking in poor students. There are the characteristics of the state which clearly play a role in determining the extent to which LCPS are supported or discouraged. There are also the characteristics of civil society which will influence the extent to which LCPS can develop to deliver services to underserved groups. In addition, relationships will be influenced by the extent to which government provision is falling short of achieving international and national goals - whether this is widespread, limited to access for specific groups, or related to filling gaps in the quality of public provision. There are a number of features that distinguish a partnership with LCPS in the basic education sector that have important implications for any strategic framework. Figure 17 highlights six of the most significant features: Figure 17: Summary of Partnership Features Feature Details Segmentation Education service delivery is vulnerable to segmentation between a public sector providing subsidised services to the poor while the private sector provides paid services to the richer segments of society who can 'afford' private services Incentives Given the limited potential of the LCPS to earn third party revenues, the government may need but lack the necessary budgetary allocation to promote sustainable and bankable basic education partnerships 19 Complexity Any partnership with LCPS will have to involve different tiers of government and communities Sensitivity Partnerships can be rolled back due to political reasons and anticipating political sensitivities and ramifications is difficult and critical Assessment As if often also the case in the public system, service delivery structures are difficult to monitor and there is often a lack of data upon which to monitor compliance with performance parameters and assess payment mechanisms Policies and ideologies In education the size of the teaching workforce, often coupled with the existence of unions gives political clout to key stakeholders to resist change. This situation is exacerbated by the fact that greater partnership can be misconstrued as the government abrogating its responsibilities to provide universal basic education. Any proposed partnership with the LCPS thus needs to be built on solid data and communicated to all stakeholders yet measuring the 'value added' in education is difficult and it takes time to produce results. Going Forward The LCPS partnership framework presented here has three distinct elements: the service delivery value chain; LCPS sector constraints, risks and opportunities; and structuring of the partnership. The service delivery value chain. Various inputs through a value adding process lead to outputs and in turn into outcomes/impact. The key inputs in the education sector are physical, human resources and financial and they are results of a complex value chain process - for example teachers are key inputs and effective teachers are produced through a value chain process of pre and in-service teacher training processes. A paradigm shift is needed to unlock the hidden values in this value chain and, while there are a number of ways to unlock the value of the real estate and/or human resources, there may be only a few opportunities that are politically and socially acceptable with regard to LCPS. This Guidance Note has provided some examples where opportunities for unlocking value of different assets of the LCPS have been explored. Figure 18 provides other types of arrangements on the risk sharing spectrum along the fully public to fully private continuum that could be explored, particularly the management and services contracts on the left of the continuum. 20 Figure 18: The Spectrum of PPP Arrangements Source: World Bank (2012) LCPS Constraints, Risks and Opportunities. A number of factors aggravate the constraints and risks in the delivery of education services. Four of these risks and constraints are highlighted: (i) public provision of basic education and transferring of public funds as user fees to a LCPS - or even more so the transfer of management of public education institutions to the private sector - is highly sensitive; (ii) the poor quality of teaching and the lack of accountability and performance systems make it high risk for LCPS to engage with public sector schools and their staff; (iii) while some countries have developed a PPP policy or framework, many have not and unfamiliarity with education contracting arrangements needs specialised capacity building; and (iv) the ability of LCPS to meet the mandated school registration and operational requirements. Structuring for Value for Money. The 'Value for Money' (VfM) proposition for structuring partnership with the LCPS in the provision of more effective, efficient and inclusive basic education needs to be examined through a number of different dimensions such as increasing access and equity, improving quality and relevance, enhancing cost effectiveness and efficiency and supporting financial and fiscal soundness. Building up the strength of this partnership structure starts with a number of critical and desirable features within the policy environment, including:  Clear, focused, leadership of the reform process from the national government;  Government recognition of the role of LCPS in attainment of quality basic Education for All;  Willingness by government to commit adequate human resources to the partnership process;  Support to the existence of LCPS representative bodies; and  A facility for generating market information on LCPS. In a systemic basic education reform in which partnership with LCPS is supported, then the role of the private education sector in turn needs to include: contributing towards policy development; co-operating with the gathering and analysis of school enrollment and academic and non-academic performance data; articulating future education market trends indicated by 21 demand and new technology; mobilising support at a local level; and disseminating information on the LCPS costs and performance. 8. The Operating Environment In the development of policy, special attention needs to be paid to ensuring that regulatory and legislative obstacles do not limit access to those who could benefit from participation and that there will be no discrimination against the mode of skills acquisition, be it full-time, part-time, distance-learning, evening classes, on-the-job or a blend of any of these. The important thing is that learners can demonstrate that they have the skills, regardless of where or how they were acquired. Some principles for effective regulations include:  Legal recognition of private education;  Effective and efficient school registration process (Realistic and objective registration criteria; Output/outcome focused; No limits on private school numbers; Provisional/graduated registration; Independent authorizing agency; Streamlined and time limited registration process; and Online criteria, forms and guidance materials);  Provision of information on school performance (School ratings, test scores, parent/community reporting; and diversity in supply – allow for-profit and not-for profit);  Provision of financial support to private schools (Contracting, private management of public schools, vouchers/subsidies, etc);  Funding on per-student basis; and  Provision of tax holidays, customs duty rebates, free or subsidized land, grants – Ensuring flexibility in the teacher labour market (Hiring/firing and flexibility in pay scales) – Ensure government regulatory capacity (Dedicated unit, training/mentoring, information/ databases, fraud detection, on-time payment systems). 7 Figure 19 provides a summary of important actions required to support LCPS operate either as purely private sector providers or in a partnership arrangement with the government. Figure 19: Important Actions for the LCPS Operating Environment Statistics  LCPS should be registered without charge and included in the national education database.  Ministries of Education should assign specialized staff to work with the LCPS helping to collect and incorporate statistics on non-government schools  Each country should have available the number and categories of LCPS, their enrollments, rates of completion, examination passes, and tuition levels.  Reports on LCPS should be published annually including trends regarding their quality and costs Regulations Governing LCPS  Regulations are clearly justified on curriculum objectives, completion, or admissions tests for further education, and health or safety  Other regulations should be reduced or eliminated, hence enabling maximum flexibility. This should include regulations that control tuition, regulate teacher certification, and pedagogy. Innovation depends on freedom to experiment in these areas. These extracts from the Terms and Conditions of Partnership under the Punjab Education Foundation’s FAS Programme illustrate some of these actions: 7 See LaRocque (2008) for further explanation of the LCPS operating environment. 22  The school must register with the District Registration Authority within a period of one year after entering into a partnership with the Punjab Education Foundation. In case of failure, PEF may discontinue financial assistance.  The financial assistance per child will be Rs. 450/ per month to cover tuition fees and allied charges. This value is subject to change, to account for Inflation and/or any other considerations as decided by the Board of Directors of PEF.  After entering into this partnership, the recipient educational institution shall not charge any fee in any form from any student whatsoever  In case of outstanding performance by the private educational institution in terms of enrolment and/or quality education in the shape of learning outcomes assessed on the basis of examination and assessment, through Quality Assurance Tests (QAT), administered on six monthly basis under the auspices of PEF-ADU, the Board of Directors, PEF may give awards to the institution and/or its staff. The rules and criteria for the award of such incentives will be determined by the Board of Directors of the PEF.  In case of increase in enrollment, the Foundation will pay the fee for the enhanced student body of the recipient educational institution.  Physical infrastructure of schools in terms of building, classrooms, toilets, library and laboratories (in case of elementary and secondary schools) must be hygienic, congenial and conducive. Conversely, the Right to Education Action (2010) in India provides a good example of how stipulations in the regulatory requirements for private schools to meet specific physical and human resource conditions have considerable implications for LCPS, most particularly the fact that they drive up the minimum costs of operation. These conditions include:  Land (biggest item of expenditure)  Playground  All weather building that includes: Kitchen for midday meals (wherever midday meal is cooked in the school), Drinking water, Separate toilets for girls and boys, At least one classroom for each teacher, An office-cum-store-cum-head teacher’s room, Boundary walls  Teacher salaries as per 6th Pay Commission  Pupil teacher ratios (not exceeding 30:1 up to 120 students; 40:1 thereafter)  At upper primary stage, at least one teacher per class, such that there is at least one teacher each for Science & Maths, Social Studies and Languages respectively  Full-time head teacher for school with over 100 students  At least part-time instructors for Art, Health/Physical and Work education  Minimum working days/instructions hours: 200 days or 800 hours for primary; 220 days or 1000 hours for upper primary; 45 hour week for teachers  Minimum qualification of a Bachelor of Education degree Accountability Ensuring ‘accountability’ of an education system - be it public or private - is fraught on a number of accounts. For example, there is the fact that there are diverse approaches to monitoring and evaluation that draw on other disciplines beyond education. Three types of monitoring might include: (i) Compliance monitoring, derived from the literature on bureaucratic control, where the objective is to establish and maintain state standards; (ii) Diagnostic monitoring, derived from behavioural psychology, where the objective is to diagnose and remediate deficiencies in student learning; and (iii) Performance monitoring, derived from an attempt to apply the principles of micro economic theory to public sector organisations – in this case schools – by using competition to focus on academic achievement and on school outputs in the belief that this will drive improvement. 23 There is also the fact that schools are complex organisations; not all their members are on the same wavelength and the conclusions from evaluation may be reinterpreted to maintain the status quo. Moreover, while external evaluators undergo quite extensive training, schoolfocused training in the use and follow-up to evaluation tends to be less systematic. Schools need guidance on improvement in addition to information on their performance. While there are quite comprehensive evaluation frameworks (including criteria and scales) and even monitoring regimes, systematic approaches to the implementation of change and improvement strategies are less evident. Moreover, devolution of responsibility to schools requires that heads are able to take on leadership of evaluation and innovation. Four measures are outlined for improving accountability and increasing the focus for LCPS on supporting equitable access to a quality basic education. These measures include: (i) improving the way the LCPS sector is regulated; (ii) improving information as a means of supporting accountability in the LCPE sector; (iii) strengthening capacity and capability in the LCPS sector; and (iv) enhancing dialogue between the public and the LCPS sector. (i) Improving the way the LCPE sector is regulated. One way of promoting growth in the LCPE sector is to improve the way it is regulated. The current approach to LCPE regulation can often be overly input focused, prescriptive, cumbersome and heavy handed. A related concern is that school regulations may be applied in an 'ad hoc' and uneven manner (both over time and across different regions). This, combined with over-zealous and arbitrary application of the rules, can create an environment that is not conducive to effective school establishment or operation. It also creates uncertainty for potential investors and can limit private investment in the LCPS sector. A common approach to regulation that is currently employed can be characterised as a LooseTight-Loose approach where there is little specification of performance (i.e. it is loose), strict control over the day-to-day operation of schools (tight) and there is weak accountability for performance (loose). The new approach would instead involve a Tight-Loose-Tight approach to regulation whereby the government provides clear specification for performance (tight), grants schools considerable autonomy in their day-to-day operation (loose) but holds them to account for performance (tight). Under the new approach, the decisions of investors, school operators, and informed consumers would replace bureaucratic decree as the key driver of the private education sector. A key plank would be the introduction of a new system for licensing private schools. Under this system, all schools would be given more freedom to manage their institutions, while at the same time being subject to different levels of reporting and accountability depending on their circumstances. One option might be to even differentiate schools on the basis of whether they were accredited by an outside agency. Under such a system, schools that were accredited could be excluded from much of the regulation that currently applies to private schools. Those schools that were not accredited would have greater accountability and reporting requirements. Figure 20 provides an example of a possible LCPS licensing system. 24 Figure 20: Example of a Possible Private School Licensing System LEVEL 1: Independent Schools LEVEL 2: Subsidized Schools LEVEL 3: Contract Schools  No government funding  Subject to generic regulation only  Provide information to government  Must register  Management and operational flexibility  Receives government funding  Funding determined by funding criteria  Accountability requirements and authorisation process  Management and operational flexibility  Government funding specified by contract  Outcomes specified by contract  Accountability requirements specified by contract  Management and operational flexibility The school licensing process could be further streamlined by allowing private sector organisations to play a greater role in school licensing. Under such a system, these organisations could be given accreditation to license private schools, if these organisations met certain criteria. An equally important issue is to ensure that private school operators have greater certainty about the regulatory environment in which they operate. Too often, regulations can be applied differently depending on the locality and the supervisor. This is especially problematic where regulations are unclear and where decision-making is decentralized. In order to provide greater certainty, a list of activities that require government approval could be prepared. Figure 21 provides details of the school rating tool established by Gray Matters Capital (GMC), the philanthropic foundation of Gray Ghost Ventures. Figure 21: Example of an Independent School Review Agency GMC has established a school rating initiative in an effort to understand the barriers to basic education and gaps in student learning outcomes. The aim of this rating is twofold: to allow the establishment of key “Levers of Change” that improve the student learning outcome in the LCPS sector and make it sustainable; and to empower the parents and the community at large by giving them access to information so that market demand can drive quality improvement in this sector. Their School Ratings Program is a comprehensive 360 degree assessment of a school that covers student learning outcome, parent satisfaction, teacher competency, school management, school infrastructure & financial performance. These reforms to the LCPE regulatory environment should not be seen as a relaxation of regulations but rather as a better way of regulating the sector and as a means for providing greater incentives for delivering quality education in LCPS. The proposed approach alters the government's role from regulator to facilitator and shifts the focus to providing parents and students with the information they need to make informed decisions; providing private school owners and managers with greater freedom to establish and manage private schools; and promoting competition among public and private schools. (ii) Improving information. Well-informed consumers are an important building block in an effective regulatory framework. Access to timely, high quality information upon which to base schooling, policy and regulatory decisions can assist market processes such as parental choice to be more effective regulatory tools. For LCPS, there is limited information available on variables such as fees and programmes offered by schools. In addition, there are few 25 independent measures of school quality – whether exam scores, results of school reviews and assessments or other indicators – upon which parents and students can make decisions about which school to attend. While centrally collected quality indicators are not the only basis upon which parents and students make schooling decisions (affordability, location and reputation might be just as important), they do provide an independent and common benchmark to assess the quality of a school. At the same time, there are many private means of providing information to consumers, such as accreditation, reputation and membership in a network. A number of measures could be introduced to improve the information available to consumers and indeed to regulators on the LCPE sector, including: requiring that private schools disclose information to regulators and the general public (e.g. fees, performance indicators, exam results), as a condition of licensing; maintaining a directory of private schools. These could be posted on the internet and publishing information on public and private school performance – exam results, school reviews, etc. (iii) Strengthening LCPE Capacity and Capability. Moving toward a more light-handed regulatory approach requires an accompanying change in the regulatory “culture” within the regulatory bodies. To assist in changing the present regulatory culture, a number of initiatives could be undertaken, including: developing a resource and training programme outlining good practice in regulation, with a focus on approaches to “light-handed” regulation; and providing training and mentoring in good regulatory practice for government officials, as well as for private education sector representatives. A second key capacity concern is the variable quality of governance and management at LCPS. This doubtless underlies much of the desire to regulate in a heavy-handed way. One initiative that should be undertaken to overcome this is to provide training aimed at improving the management and governance of the LCPS. A necessary ingredient to good oversight of the LCPE sector is that regulators have access to timely, high quality information upon which to base schooling policy and regulatory decisions. Regulatory authorities collect a considerable amount of information on the private education sector. However, two issues are of concern: (i) information from different sources (e.g. on student numbers at different education levels) is often inconsistent and is available only with a time lag; and (ii) information is scattered across a range of departments and agencies, reflecting the distribution of regulatory responsibilities. Information on the private sector could be collated by one agency and released in a timely, organized manner – perhaps via an annual private education annual report. (iv) Enhancing dialogue. Building greater linkages and enhanced dialogue between the public and private sectors needs attention if the LCPE sector is to be able to fulfil its potential as a full partner in helping the state meet its education and wider policy objectives. There is often a lack of trust between the public and private sectors, as well as an absence of mechanisms upon which to build such trust. Dialogue is required to develop and attain this framework. Such a dialogue (whether in the form of a national forum or consultative mechanism) is necessary to provide the required focus on broad policy themes, a suitable climate for public-private relations as well as access to the required information. The potential outcomes from this dialogue include, inter alia, improved information for decision-making, a broader policy consensus and greater accountability of the public and private decision-makers. There are clearly risks and certain key questions that need to be addressed when designing and implementing the public-private consultative dialogue, including: whether the dialogue should be focused on particular policy issues over a defined lifetime or should a permanent consultative mechanism; what internal processes and procedures can be used to alleviate concerns of 26 participants for whom some of the reform initiatives will have a critical impact; and whether it is it possible to establish and fund a secretariat to facilitate the dialogue. Private sector associations should be encouraged to move beyond simply a representational role. They could, for example, play a greater role in regulating the sector and lifting the quality of education provided in the private sector. Further advantages accruing from such an association are that it would provide a vehicle for: linking funding from a possible education finance facility to accreditation standards that have been set by the Association and verified by the MoE; sharing best practice among private sector operators, advisors and teachers; and developing and coordinating school improvement initiatives, including staff and management training. Figure 22 highlights the formation of the National Independent Schools Alliance in India. Figure 22: The National Independent Schools Alliance NISA aims to create an enabling ecosystem for budget private schools that are working to improve access to affordable and quality education for the economically weaker. NISA believes in concerted efforts for advocacy with the government, amplifying the voice of budget private schools through access to the media and building evidence for our claims about relatively better quality of budget private schools as compared to many government schools and the threat posed to them by regulations through systematic linkages with the academia. Source: http://nisaindia.org/sls2012.php Sustainability Assessing the financial sustainability of a quality LCPS can be undertaken by reviewing success across key three parameters: (i) Efficacy with a particular focus on the evidence of impact on learning outcomes and an assessment as to the critical factors that are behind this performance; (ii) Equity and a review of the school catchment; and (iii) Scalability or replication and the ability of the LCPS to grow 'horizontally' (across geographic locations) and/or 'vertically' (across education sub sectors). Efficacy. Learning outcomes can be measured through different assessments of learning outcomes from which evidence of impact can be drawn. Four main areas of focus for this assessment include whether the school is child-centric, there is continuous teacher training and support, there is benchmarking of attainment and professional teaching and learning processes are in place. Eight common quality indicators often used to assess these areas include: How good are the students’ attainment and progress? How good is the students’ personal and social development? How good are the teaching and learning? How well does the curriculum meet the educational needs of all students? How good are the staffing, facilities and resources for learning? How good is the partnership between the parents and the school? How good are the leadership and management of the school? And how well does the school perform overall? While much has been debated and written regarding the profile of LCPS including their quality of provision, their impact on equity of access and how they perform in comparison to public schools, this Guidance Note recognizes that there are both good as well as poor performing schools in both the LCPS and the public sectors. Both types of school exhibit some examples of the delivery of acceptable quality education to low income communities yet there are other examples which perform below expectation. The crucial need is to grow the capability for the assessment ecosystem to differentiate between the good and the bad and to provide support accordingly. LCPS that are not performing well need more targeted support and handholding while progressive LCPS needs solutions, products and services which can help them improve further. 27 Equity. Even if a LCPS is unable to provide specific details regarding the socio-economic status of its students, it is possible to obtain a preliminary assessment as whether the students are from the lowest quintile based on the following aspects - the timetable, the location, and the price point. Timetabled to support beneficiaries and clients  Is the class schedule designed to minimise idle time and maximise the time-on-learning task for each child?  Is the school schedule set to minimise the interference by the local social life cycle in the schools’ functioning? Location  Are classes held close to the children's home so that young children can come to the school unescorted? Price point  Are the classrooms rented or owned and are there playgrounds or other amenities?  What rate are the teachers' wages in comparison to the salaries of public teachers in the formal sector or teachers in the for profit or not for profit LCPE sector?  What proportion of the household income is allocated to education - for the direct and indirect expenses? Scalability. The critical issues of the scalability and sustainability for the LCPS can be viewed across the following four parameters, namely organizational culture, structural design for accountability, management structure and ability to grow. Figure 23 provides an assessment of the Gyan Shala programme across these parameters. Figure 23: Scalability Assessment of the Gyan Shala LCPE Programme Organizational culture  The teachers, who come from low-income backgrounds, are provided annual and monthly training in a good quality training-centre  The design team is required to be in direct touch with the teachers and classrooms so there remains the least possible gap between the design parameters and actual classroom practice. Structural design for accountability  A multi-tier supervisory chain oversees the performance.  The availability of financial resources to the Gyan Shala team is linked to the measurable performance of children in an independently held examination at the end of the 3-year module.  A mechanism to replace non-performing staff without disrupting routine performance is built-in. Management structure and processes  The organisational structure is designed to integrate the management of the programme with the development and supply of learning material and teaching guides, the annual and monthly teacher training and weekly supervisory support to the teachers.  This is done in a decentralised mode so as to fit the learning needs of a chosen group of less than 15000 children with similar socio-economic profiles.  The teachers are supported/supervised by a team of senior teachers.  A core team of subject specialists is responsible for the design and development of learning materials and teacher training, all of which is linked to the feedback from the classes.  The design establishes a chain of supervision/mentoring for quality assurance. Ability to grow  Use only such level of talent and staff that are available in large numbers 28 at the given salary level.  The core competence of the organisation is to induct and train new staff of modest formal education to deliver teaching and learning to an acceptable level of quality.  A decentralised self-contained education design and delivery unit whose effectiveness/success can be measured un-ambiguously.  Use of private sector contracts for employment with competitive salaries and weekly supervision. Source: http://r4d.dfid.gov.uk/pdf/outputs/mis_spc/60912-GyanShalaFinalReport.pdf LCPS serve the low-income marketed and to do so successfully they need to operate ‘No Frills’ models that enable economizing at every stage in the provision of the education offering: (i) In setup and service, non-core capital and expenses provide a basic service while the education quality is kept sufficiently high to ensure the education service is comparable to or superior to other options; (ii) there is a high utilization of the asset base with facilities being used on double shifts and the same teacher often working on two shifts; and (iii) there is a focus on service specialization with the delivery of all procedures with regard to personnel and training being highly standardized, documented, routinized, and easy to deliver for lower-skilled staff. Risks and Mitigating Strategies There are a number of significant internal and external risks that LCPS must overcome when delivering education to the lowest socio economic quintile in different locations across a large area. Some of the main risks are summarised below with brief outlines of how such risks might be mitigated. The following three main ‘internal’ risks have been identified, they include:  A possible fiduciary risk which can be addressed by strengthening financial management systems (e.g., procurement, audit, output-based budgeting);  A risk of weak implementation capacity resulting from poor management systems and inadequate number of staff with management capabilities – a risk that can be contained through the delivery of a capacity building programme that supports the creation of a cadre of specialist managers as well as encouraging greater participation from the community in order to strengthen school management through PTAs or SMCs;  Access to teachers and their retention for the primary programme is a substantial risk and access and retention of qualified teachers require a recruitment drive and TA to possibly support an ‘on the job’ qualification programme. Then there a number of significant ‘external’ risks that include:  A lack of political will to support LCPS;  Opposite to a lack of the political will is the risk of undue political interference - interference over such matters as teacher recruitment and transfers, curriculum reform and school construction;  A serious national economic downturn might affect the ability to fund the programme either through government, corporate social responsibility or private funding streams;  Lack of private sector interest to support the initiative. There are few incentives for the private sector foundations to offer support while the beneficiaries do not have the willingness to meet the proposed fee charges due to the lack of a clear policy and collection system. 29 9. Finance Differentiation of LCPS LCPS - be they 'for profit' or 'not for profit' models - exhibit some common features. Commercially their offerings are similar in that their capacity for growth and expansion is constrained by various factors, including lack of access to capital, property rights, and availability and cost of labour. Both LCPS models are also dependent on the low cost of labour. Whereas common educational offerings include the fact that the LCPS are increasingly offering a package of interventions that are subject to robust quality control processes, regular 'on the job' training for teachers that have low levels of qualification, and considerable provision of supporting teaching and learning materials. However, in order for there to be a partnership approach that leverages the strengths of the government and the private sector, there needs to be a disaggregation between the three different types of providers that cater to these parents. This disaggregation using these categories is necessary as the 'for profit' and 'not-for-profit' LCPS will require different responses across three key areas of possible PPP interventions namely: Enhancing the Operating Environment; Promoting any Supply side Education Market; and Supporting any Demand side Incentive Programme. Figure 24 summarises these three interventions and some of the main aspects. Figure 24: PPP Interventions and the Main Features Interventions Features of the Interventions Enhancing the operating environment  improving the regulatory environment  establishing a robust assessment and quality education assurance system  providing capacity building to strengthen the ability within and across the public and private sectors to monitor and support the transition Promoting the supply side of the education market  setting up customised low cost private lending facilities  supporting tax exemptions and utility fees  supporting non state operators to access a portfolio of school improvement services Supporting a demand side incentive programme  establishing an education service contracting subsidy model  establishing a universal or targeted voucher model In each of these three interventions the two LCPS types need to be treated differently. For example, 'not-for-profit' providers often fail to comply with the regulatory requirements and from a supply side, these same providers have greater difficulty than their 'for profit' competitors in accessing funding due to the fact that they lack fee revenue. Whereas from the demand side intervention - because they are 'not-for-profit' providers - they are possibly the more suitable 30 recipient of any public financing because they charge less and thus have greater reach to the bottom quintile. The importance of making this differentiation can be illustrated by the situation in Bangladesh where there are thirteen types of providers allocated across these three categories (see Figure 25). 31 Figure 25: Types of Education Provider by Category A) Government B) Not for Profit Private (NFP) C) For Profit Private (LFP) i. Government Primary School ii. Registered Nongovernment Primary School iii. Experimental School i. NGO School ii. Non-Registered Non-Government School iii. Ibtedayee Madrasha iv. High Madrasha Attach Ibtedayee v. Community School vi. Building Resources Across Communities Center (BRAC) vii. Reaching Out of School Children Project viii. Shishu Kollan i. Kindergarten School ii. High School Attached Primary Source: DFID SARH Study on Bangladesh LCPS (2013b) Figure 26 shows the percentage share of schools, teachers and students in primary schools by category over the period 2007 to 2011. This Graph illustrates three significant points: (i) how the public sector share has remained static across this time with few new schools being opened over the past three years; (ii) how the 'for profit' LCPS share met the obvious supply constraints more quickly than their 'not for profit' LCPS competitors; and (iii) how the 'for profit' LCPS fee market share grows at the expense of the 'not for profit' LCPS since the latter is unable to access funding and thus respond as swiftly. Essentially, with minimal increase in the number of public schools and the number of teachers, the student enrolment has increased 14 percent over the five years and in response to this growing 'supply' driver the 'not for profit' LCPS sub sector has decreased by almost half a million students whereas the 'for profit' LCPS sub sector has seen an increase in enrolment of over a million students. Figure 26: Percentage Share of Schools, Teachers & Students in Primary Schools by Category Source: Compilation from DPE and BANBEIS data (2007 - 2011) - DFID SARH (2013b) 0 10 20 30 40 50 60 70 80 Schools Teachers Students Schools Teachers Students Schools Teachers Students Category A Category B Category C 2011 2010 2009 2008 2007 32 Example of Supply side Financial Support Access to capital is a major constraint for LCPS that are privately owned and operated on a 'for profit' basis and for LCPS that are 'not for profits' engaged in the delivery of basic education services for public benefit. The International Finance Corporation (“IFC”), the private sector investment arm of the World Bank Group, has developed an innovative programme to support the private schools sectors in different regions. In Africa, this programme was launched first in Ghana in 2005, in Kenya in 2007 and then in Rwanda in 2008. Similar programmes are now planned for additional countries in sub-Saharan Africa (Liberia, Sierra Leone, Senegal, South Africa, Uganda and Zambia) as well as for rollout in the South Asia and MENA Regions. The IFC has designed the Private Sector Support Program to assist private schools overcome their lack of access to appropriate finance and services to help them better manage their operations. The programme has two parallel primary activities: A. Financing Facility: Through a partial risk sharing guarantee, signed with a local Bank, IFC assists the selected Bank to expand its lending from short term financing to a few schools, to medium term loan financing for a bigger portfolio of private schools in the education sector. The loans are to be applied to classroom and other relevant schools infrastructure expansion, purchase of educational materials, and other capital expenditures. B. Technical Assistance: 1. Delivery of direct and wholesale technical assistance to schools as follows: i) Direct Business Development Services (BDS) via education business experts to conduct a school business diagnostic and determine in a prioritised order the key issues necessary for growth in financial performance and school capacity and quality outputs; ii) Delivery of School Development Services (SDS) to groups of schools through workshop training on relevant topics; the development of a series of school operating manuals and the development of a sustained local capacity (a local service provider) to deliver these services after the end of IFC’s program. 2. Delivery of training to the local Bank in Education SME Project Finance and due diligence of private schools. 8. Concluding Remarks In the West, those interested in aid fall into two camps: those who want less of it because they think it is ineffective and creates dependency; and those who want more because they believe it is fundamental to global justice. The visionary leaders among recipient countries don’t hold either view. They want their province or country to succeed and ultimately to do so without any dependence on aid. In the meantime they want aid of the right kind but they don’t want the entire (occasionally self-serving) aid industry that sometimes comes with it......It would be wonderful if when the end of aid comes to pass we could look back and say, “That worked.” Rather than “That’s over.” Or, worse still, “That was a waste of time.” Sir Michael Barber, 2013 page 73 The key message from this LCPS Guidance Note is that the distinction between public and private is less important than the perceived public good of each set of institutions and the 'rules of the game' to which the critical actors of the system respond. Further, with the right policy framework, there is no contradiction between high quality public education and encouragement 33 of the expansion of low cost private education. A significant obstacle against low cost private solutions which serve the public good is the incapacity of the state to design and implement an environment and appropriate incentives where the private gain more when they behave in ways that promote the public good. To improve, the public and the low cost private education sectors require clear and coherent standards, the means to achieve those standards and feedback on how well they are doing. There is no single answer to the question of how to create self-sustainable LCPS for the poor. The best thing to do is to be pragmatic and heed the following five main messages: 1. It is vital that countries rich and poor alike, deliver education through a wide variety of mechanisms, public and private 'for profit' and 'not for profit'. 2. Donors and governments need to diversify their approach in which the government funds, provides, owns and regulates the basic education sector to an approach that includes private schools managed by non-governmental organizations or for-profit edupreneurs 3. Where programmes already exist of demand side assistance to families or children of the poor (be it through cash transfers, subsidies or vouchers), these programmes should be extended to the children of the poor who attend registered LCPS. 4. Publicly financed programmes in LCPS should be expanded on specific grounds and justified on the basis that the public system cannot currently accommodate these specific needs. Such grounds may include a lack of public supply of secondary education (eg., the ESC in the Philippines); children with specific learning needs; orphans, street children (e.g., the Gyan Shala programme in India) and children with considerable ability from low-income homes who deserve a better opportunity (e.g., the FAS programme in Pakistan). 5. Alternative models and financing schemes need to be examined in depth. This is especially important in countries with public educational opportunities in short supply. Whether it is self-association schemes (such as the National Independent School Alliance which is supporting LCPS to form a national association) or government-assisted programming (such as the Sindh Education Foundation which provides support directly to parents through vouchers that they can spend in low-cost private schools that have been accredited to meet minimum quality standards) alternative models and partnerships need to be tested. At stake is the long-term financial sustainability of a sub-sector that is responding to a demand that cannot be met by the public sector alone. At the outset this Guidance Note stated that there is a lack of data collection and comparative analysis to be able to assess value for money from the government perspective and for prospective students and their families to distinguish between high-quality and low-quality public and private providers; and that this lack of comprehensive and comparable data on both 'for profit' and 'not for profit' LCPS continues to be problematic for educators, researchers, aid organizations and policy makers alike. The Guidance Note concludes by posing some further questions worthy of further research in the area of LCPS:  Without a subsidy or a possible PPP arrangement, is it possible for a private education provider to deliver quality education based only on fee collection from the lowest socioeconomic quintile?  To what extent are very poor parents prepared to choose low-cost schooling over free schooling, even when they have very little disposable income? At what ‘price’ in quality will families pull their children out of public schools and seek a LCPS alternative?  Why do parents choose LCPS? 34  Without reengineering the mode of delivery, is it possible for the private or public provider to deliver quality education that meets the particular physical and social needs of these communities?  To what extent does the regulatory environment impact upon both this significant consumer choice for the poor and the operation of these LCPS?  Why do countries with similar basic education statistics, government expenditures on education and policies toward LCPS exhibit differences in attendance rates or rates of expansion?  On what grounds should state support be provided to LCPS? Is it equity and ensuring the access of disadvantaged groups to quality training? Is it efficiency and driving improvement through greater competition? Or is it participation and enabling access through complementing public provision?  What mechanisms should be put in place to deliver such support ranging from direct subsidies, vouchers, scholarships and fiscal incentives? 35 SECTION SIX: REFERENCES AND FURTHER READING Alderman, H., Orazem P., and Paterno, E. 2001. School Quality, School Cost, and the Public/Private School Choices of Low -Income Households in Pakistan, The Journal of Human Resources, Vol. 36, No. 2, pp. 304-326. Alderman, H., Jooseop, K. and Orazem, P. 2003. Design, Evaluation, and Sustainability of Private Schools for the Poor: the Pakistan urban and rural fellowship school experiments’, Economics of Education Review, Vol. 22, pp. 265-274. Andrabi, T., Das, J. and Ijaz Khwaja, A. 2002. The Rise of Private Schooling in Pakistan: Catering to the Urban Elite or Educating the Rural Poor? Working Paper, Harvard University, March 2002. Andrabi, T., Das, J., Khwaja, A. I., Vishwanath, T., and Zajonc, T. 2007. Learning and Educational Achievements in Punjab Schools (LEAPS): Insights to inform the education policy debate. World Bank, Washington, DC. Andrabi, T., Das, J., and Khwaja, A. I. 2008. A dime a day: The possibilities and limits of private schooling in Pakistan. Comparative Education Review, 52(3):329–355. Angrist, J., Bettinger, E., Bloom, E., King, E., and Kremer, M. 2002. Vouchers for private schooling in Colombia: Evidence from a randomized natural experiment. The American Economic Review, 92(5):1535–1558. Angrist, J., Bettinger, E., and Kremer, M. 2004. Long-term educational consequences of secondary school vouchers: evidence from administrative records in Colombia. Working Paper 10713. Washington, DC: National Bureau of Economic Research. Asadullah, M. N. 2005. The effectiveness of Private and Public schools in South Asia . Mimeo Barakat, S. 2012. Low-Cost Private Schools in Afghanistan and Pakistan: What evidence to support sustainable scale-up? Protocol written by Sultan Barakat et al. Bangay, C. and Latham, M. 2012. Are we asking the right questions? Moving beyond the state vs non-state providers debate: ‘Reflections and a Case Study from India’. International Journal for Education Development. Barber, M. 2013. The good news from Pakistan - How a revolutionary new approach to education reform in Punjab shows the way forward for Pakistan and development aid everywhere. http://www.reform.co.uk/resources/0000/0688/The_good_news_from_Pakistan_final.pdf Bray, M. 1997. Community financing of education: Rationales, mechanisms, and policy implications in less developed countries. In: Colclough, C. (ed.) Marketizing education and health in developing countries: Miracle or mirage? Oxford University Press, USA. CfBT, 2011. Preliminary Study into Low Fee Private Schools and Education. Prepared on behalf of the DFID South Asia Research Hub. DFID SARH, 2013a. Study into the potential long-term viability and expansion of Gyan Shala through private sector investment. Yet to be published DFIS SARH, 2013b. Study into the role of the private sector in primary education for the urban poor in Bangladesh. Yet to be published Fennell, S. 2007. Tilting at windmills: Public-Private partnerships in Indian Education Today , RECOUP Working Paper 5. Fielden, J., and LaRocque, N. 2008. The Evolving Regulatory Context for Private Education in Emerging Economies." The World Bank Group International Colleqium on Private Education, Washington, DC. French, R., and Kingdon, G., 2010. The Relative Effectiveness of Private and Government Schools in Rural India: Evidence from ASER Data. Institute of Education, London. Gallegos, F. 2004. School Choice, Incentives, and Academic Outcomes: Evidence from Chile. Boston: Massachusetts Institute of Technology. Garg, N .2011. Low Cost Private Education in India: Challenges and Way Forwards 36 http://www.bridgeinternationalacademies.com/Bridge_International_Academies/Our_Model.html http://www.dfid.gov.uk/r4d/Output/190230/Default.aspx http://www.omega-schools.com/ http://www.periglobal.org/subject/low-fee-schooling http://www.ratio-magazine.com/201005042964/Kenya/Kenya-A-Commercial-Approach-to-SlumEducation.html Gauri, V. and Vawda, A. 2004. Vouchers for basic education in developing economies. World Bank Research Observer, 19(2):259–280. Gray Matters Capital Report. 2012. Affordable Private Schools (APS) Sector Analysis Report Harma, J., 2009. Can choice promote Education for All? Evidence from growth in private primary schooling in India. Compare: A Journal of Comparative and International Education 39 (2), 151–165. Harma. J. 2011. Low cost private schooling in India: Is it pro poor and equitable? International Journal of Education and Development doi 10.1016/j.ijedudev2011.01.003 Hsieh, C. and Urquiola, M. 2006. The effects of generalized school choice on achievement and stratification: Evidence from Chile’s voucher program. Journal of Public Economics, 90(8- 9):1477–1503. James, E. 1991. Public Policies Toward Private Education: An International Comparison, International Journal of Educational Research , Vol. 15, pp 359–376 Jimenez, E. and Peng Tan, J. 1987. Decentralised and Private Education: The Case of Pakistan, Comparative Education, Vol. 23, No. 2, pp. 173 190 Kim, J., Alderman, H. and Orazem, P. F. 1999. Can Private School Subsidies Increase Enrollment for the Poor? The Quetta Urban Fellowship Program. The World Bank Economic Review, Vol. 13, No. 3, pp. 443-465 Kingdon, G., 2007. The progress of school education in India. Global Poverty Research Group. Working Paper 71. LaRocque. N. 2008. "Public-Private Partnerships in Basic Education: An International Review", CfBT Education Trust, Berkshire. Lassibille, G. and Tan, J. P. 2001. Are private schools more efficient than public schools? Evidence from Tanzania. Education Economics, 9(2):145–172. Lassibille, G. and Tan, J. P. 2003. Student learning in public and private Schools in Madagascar, Economic Development and Cultural Change, 51:699-717. Lewin, K. (2007). “The Limits to Growth of Non-government Private Schooling in sub-Saharan Africa,” pp. 41 – 67. In: Srivastava, P. and Walford, G. (Eds.) Private Schooling in Less Economically Developed Countries: Asian and African Perspectives. Oxford: Symposium Books. Lewin, K. and Caillods, F. 2001. Financing secondary education in developing countries: Strategies for sustainable growth. UNESCO. Lewin, K. M. and Sayed, Y. 2005. Non-Government secondary schooling in Sub-Saharan Africa. Exploring the evidence in South Africa and Malawi. Researching the Issues: DFID Educational Research Series No, 59. Muralidharan, K. and Kremer, M. 2007. Public and Private Schools in Rural India. http://www.economics.harvard.edu/faculty/kremer/files/Public%20and%20private%20schools%2 0in%20rural%20india%20(Final%20Pre-Publication).pdf accessed 20.06.2011 Niaz Asadullah, M. 2009. ‘Returns to private and public education in Bangladesh and Pakistan: A comparative analysis’, Journal of Asian Economics, Vol. 20, pp 77–86 Rangaraju, B., Tooley, J., Dixon, P., 2012. The Private School Revolution in Bihar: Findings from a Survey in Patna Urban. India Institute, New Delhi. Rasool Memon, G. 2007. Education in Pakistan: The Key Issues, Problems and The New Challenges. Journal of Management and Social Sciences, Vol. 3, No. 1, pp 47-55 37 Rose, P. 2005. Privatisation and decentralisation of schooling in Malawi: Default or design? Compare, 35(2):153–165. Rose, P. 2006. Collaborating in Education for All? Experiences of Government Support for non - state provision of basic education in South Asia and Sub Saharan Africa. Public Administration and Development , Vol. 26, pp 219–229 Srivastava, P. 2008. The shadow institutional framework: towards a new institutional understanding of an emerging private school sector in India. Research Papers in Education Vol. 23, No. 4, December 2008, 451–475 Srivastava, P. and Walford, G. (Eds)., 2007. Private Schooling in Less Economically Developed Countries: Asian and African perspectives. Symposium Books. Sigsgaard, M. 2009. Education and Fragility in Afghanistan: A Situational Analysis. INEE, UNESCO and IIEP. Tooley, J. 2007. Could for -profit private education benefit the poor? Some a priori considerations arising from case study research in India’, Journal of Education Policy, Vol. 22, No. 3, pp 321–342 Turrent, V. 2009. Expanding Support for Education in Fragile States: What Role for the Education for All –Fast Track Initiative? CREATE Monograph No. 30 UNICEF. 2011. Non State Providers and Public Private Partnerships in Education for the Poor. UNICEF East Asia and Pacific Regional Office, Bangkok. USAID. 2011. The Search for Effective EFA Policies: The Role of Private Schools for Lowincome Children. Washington DC. Woodhead, M., Frost, M., James, Z., 2013. Does growth in private schooling contribute to Education for All? Evidence from a longitudinal, two cohort study in Andhra Pradesh, India. International Journal of Educational Development 33 (1), 65–73. World Bank, 2009. The Role and Impact of Public-Private Partnerships in Education. Washington DC. World Bank 2010. Education Service Contracting. Washington DC. World Bank, 2012. Public-Private Partnerships Reference Guide. Washington DC.