Policy options for student loan schemes: lessons from five Asian case studies

Transcription

1 Volume 1 Number 6 Policy Research and Dialogue Student Loans Schemes in Asia Policy options for student loan schemes: lessons from five Asian case studies Adrian Ziderman United Nations Educational, Scientific and Cultural Organization UNESCO Bangkok

2 Policy options for student loan schemes: lessons from five Asian case studies

3

4

5 Ziderman, Adrian Policy options for student loan schemes. Bangkok: UNESCO Bangkok/ IIEP, p. 1. Student loans. 2. Educational systems. 3. Higher education. 4. Educational grants. 5. Financial aid. I. Title. The views and opinions expressed in this booklet are those of the author and do not necessarily represent the views of UNESCO, UNESCO Bangkok or of the IIEP. The designations employed and the presentation of material throughout this review do not imply the expression of any opinion whatsoever on the part of UNESCO or IIEP concerning the legal status of any country, territory, city or area or its authorities, or concerning its frontiers or boundaries. Published by: United Nations Educational, Scientific and Cultural Organization UNESCO Bangkok Mom Luang Pin Malakul Centenary Building 920 Sukhumvit Road P.O. Box 967, Prakanong Post Office Bangkok 10110, Thailand UNESCO Bangkok web site: and International Institute for Educational Planning 7-9 rue Eugène Delacroix, Paris IIEP web site: Cover design: Keen Publishing Typesetting: Linéale Production ISBN: UNESCO 2004 Printed in Thailand

6 Contents List of abbreviations 7 List of tables 9 List of figures 10 Preface to the series 11 Chapter 1. Introduction Background Outline of the paper 17 Chapter 2. The five case studies Case studies: general description China Hong Kong SAR, China Republic of Korea The Philippines Thailand Dominance of mortage-type schemes 27 Chapter 3. Why student loans schemes? Alternative loans scheme objectives Country case studies: loans scheme objectives Policy alternatives to student loans schemes 38 Chapter 4. Organizational framework Unitary scheme versus multiple schemes Centralized or decentralized loans schemes? 43 Chapter 5. Assigning institutional roles: funding Provision of capital Who pays the interest subsidy, who bears the risk? 51 Chapter 6. Assigning institutional roles: borrower selection and loan distribution The loan distribution process Alternative distribution pathways: case study experience Weighing the alternatives Adequacy of loan size 60 5

7 Contents Chapter 7. Assigning institutional roles: loan repayment collection Institutional roles in collection Case study practice Minimizing default 67 Chapter 8. Financial viability of loans schemes Factors leading to low loan recovery The individual loan account Loan recovery: the overall perspective Loans scheme viability 76 Chapter 9. Equity and assistance to the poor: the role of subsidies When are loan subsidies justified? Targeting Case study experience with targeting Defining eligibility Reach of loans schemes: case study experience 87 Chapter 10. Case study loans schemes: major strengths and weaknesses China Hong Kong Republic of Korea The Philippines Thailand 90 Chapter 11. Key issues in design and reform: lessons from the case studies Applying lessons from international experience Issues emerging from the case studies Towards a typology of good practice 104 References 107 Glossary 109 Appendix 115 6

8 List of abbreviations ADB CHED COE COI GCSLS GDP GECP GFIs GSSLS Asian Development Bank Commission on Higher Education (Philippines) Centres of Excellence Cost-of-living index General Commercial Student Loans Scheme (Chinese commercial scheme) Gross domestic product Government Employees Pension Corporation (Republic of Korea) Government financial institutions (Philippines) Government Subsidized Student Loans Scheme (Chinese subsidized scheme) HECS Higher Education Contribution Scheme (Australia) IIEP International Institute for Educational Planning KLWC Korea Labour Welfare Corporation KRF Korean Research Foundation KTB Krung Thai Bank KTP Korea Teachers Pension Corporation LSFS Local Student Finance Scheme (Hong Kong SAR subsidized scheme) MOE Ministry of Education / Ministry of Education and Human Resources Development (Republic of Korea) MOL Ministry of Labour MUA Ministry of University Affairs (Thailand) NLS Non-means-tested Loan Scheme (Hong Kong SAR nonsubsidized scheme) 7

9 List of abbreviations ONC SFAA SLSC SNPL TRF UNESCO Office of the National Education Commission (Thailand) Student Financial Assistance Agency (Hong Kong SAR) Student Loans Scheme Committee (Thailand) Study Now, Pay Later scheme (Philippines) Thailand Research Fund United Nations Educational, Scientific and Cultural Organization 8

10 List of tables Table 2.1 Table 3.1 Table 3.2 Table 3.3 Table 4.1 Table 4.2 Table 5.1 Table 7.1 Table 8.1 Table 8.2 Table 9.1 Table 9.2 Table 9.3 Table 10.1 Table 11.1 Table A1.1 Table A1.2 Table A1.3 Case studies: general description of loans schemes Alternative objectives of student loans schemes Country case studies: loans scheme objectives Policy alternatives to student loans schemes Unitary or multiple schemes: case study practice Centralized or decentralized schemes: case study practice Financing loans schemes: functions and agents Measures/sanctions against repayment default Factors leading to less-than-full loan recovery Case studies: repayment and recovery ratios Justification for subsidized loans Targeting and reach: case study loans schemes Setting the income ceiling for loan eligibility Loans schemes: major strengths and weaknesses Characteristics of good practice loans schemes Treatment accorded to various loans schemes in case study reports Case studies: major features of selected loans schemes Lending conditions in case study loans schemes 9

11 List of figures Figure 5.1 Figure 6.1 Figure 6.2 Figure 7.1 Figure 9.1 Figure 9.2 Higher education budget allocations, with cost sharing The loan distribution process: simplified presentation Alternative scenarios for loan distribution Alternative scenarios for loan repayment collection The egg diagram : reaching the poor Defining eligibility 10

12 Preface to the series This synthesis study is part of a series of in-depth studies on the functioning of government-sponsored student loans schemes in Asia. The study assesses the lessons learned for national policy design and reform from a regional comparative policy review on student loans schemes undertaken in countries in Asia between 2001 and The review was carried out in the context of a project undertaken by the Bangkok-based UNESCO Asia and Pacific Regional Bureau for Education (UNESCO Bangkok) in association with the International Institute for Educational Planning (IIEP). It covered five countries and territories including P.R. China, Hong Kong S.A.R. China, the Republic of Korea, the Philippines and Thailand. Additional studies have also been commissioned by IIEP. Most countries in Asia are experiencing a dramatic increase in demand for higher levels of education at a time of both severe public budget constraints and profound overhaul of education systems aimed at significantly increasing their impact and relevance. In an attempt to ease the burden on public budgets, a number of countries have introduced student loans schemes, hoping to recover costs and increase the revenue base for the expansion of education, while at the same time providing opportunities for poorer segments of the population to access higher levels of education. In an attempt to increase the knowledge base available to governments and provide practical insights which might be useful for national education policy, the UNESCO Asia and Pacific Regional Bureau for Education initiated in 2001 a regional comparative study to examine the performance of student loans schemes in a number of countries in Asia. The study is intended to be instrumental in improving the efficacy and financial efficiency of existing schemes and in providing a comparative information base for countries intending to introduce a student loans scheme. A joint endeavor by UNESCO Bangkok and IIEP, this regional policy study was initiated and its implementation co-ordinated by the Education Policy and Reform Unit (EPR) 1 at UNESCO Bangkok. It benefited from 1. Former Planning and Sector Analysis Unit (PSA), restructured in

13 Preface to the series technical support by IIEP and was able to draw on a wide range of regional and international expertise available at research institutes, universities and ministries in the participating countries. Five research teams led by academics and senior-level practitioners were involved in producing the monographs. UNESCO Bangkok and the Korean Educational Development Institute (KEDI) hosted several research seminars bringing together research teams and practitioners from the countries participating in the project. Partner institutions involved in the study included the Korean Educational Development Institute (KEDI); Huazhong University of Science and Technology (Graduate School of Education), Wuhan; Peking University (Graduate School of Education); the Chinese University of Hong Kong (Department of Educational Policy and Administration); the Commission for Higher Education (CHED) in the Philippines (Office of Student Services, Office of Policy Planning, Research and Information) and the Asian Development Bank collaborating in the framework of an ongoing Education Sector Development Program in the Philippines. Researchers and officials from several universities, education and finance ministries and national agencies such as student loans offices in the participating countries collaborated in the preparation of the case studies. UNESCO Bangkok and IIEP would like to thank all those individuals who provided their expertise and professional experience to this research and therefore helped to assemble a considerable cross-sectoral information base required for comparative loans policy analysis. The important contributions by individual researchers and authors are acknowledged in this book. The series of studies has benefited from the technical expertise of Adrian Ziderman, Professor of Economics at Bar-Ilan University, Israel, and UNESCO lead consultant, who provided research guidance and prepared the synthesis study. The common framework and methodology for the case studies was derived from a study of the new Thailand student loans scheme, written by Adrian Ziderman in the context of a UNESCO- Asian Development Bank Technical Assistance Project in Igor Kitaev, Programme Specialist (education financing), served as resource person from the IIEP in addition to authoring one of the studies. Dominique 2. Thailand education management and finance study, UNESCO-Asian Development Bank, Technical Assistance Project (T.A THA),

14 Preface to the series Altner, Chief, Education Policy and Reform Unit (EPR), UNESCO Bangkok, with support from Toshiyuki Matsumoto, Assistant Programme Specialist, EPR, initiated the project and ensured the professional coordination throughout the study. Sheldon Shaeffer, Director UNESCO Asia and Pacific Regional Bureau for Education, Bangkok Gudmund Hernes, Director International Institute for Educational Planning, Paris 13

15

16 Chapter 1 Introduction 1.1 Background State-supported student loans are increasingly on the policy agenda in many countries. Student loans schemes are in place in over 50 countries around the world, almost all of them are concerned with tertiary education. In many other countries, proposals to introduce a student loans scheme constitute a recurring theme in political debate. Loans schemes are of particular interest to policy makers because they are able to contribute to the solution of a range of pressing policy problems that governments are facing. They are able to relieve pressures on national budgets by facilitating greater cost sharing through the raising of tuition and other university fees; liberated resources may be used in areas of greater priority for society, both outside and within the education sector and notably basic education (Psacharopoulos, Tan and Jimenez, 1987). Greater cost recovery can provide additional funds for the expansion of the university system, to accommodate ongoing increases in the social demand for tertiary education. Targeted at disadvantaged groups, subsidized loans schemes can lead to greater access of the poor to university education, thus contributing to improved social equity. And loans offered at favourable conditions for study in particular fields can lead to the loosening of skilled manpower bottlenecks that inhibit national and social development. The record of success, particularly for loans schemes in developing countries, has been mixed. While some schemes have proved to be broadly successful, the outcomes of loans schemes have frequently been disappointing, both in terms of meeting set objectives and in terms of financial sustainability. Where schemes have been less than successful, the lack of success has stemmed from weaknesses in the process (administrative deficiencies, excessive default or poor targeting) or the causes were in-built and, in particular, related to excessively generous loan conditions and high subsidies. 15

17 Policy options for student loans schemes: lessons from five Asian case studies Government-sponsored student loans schemes operate in a number of countries across Asia. Student loans schemes have been established in Australia, Mainland China, Hong Kong SAR, India, Indonesia, Japan, Kazakhstan, the Republic of Korea, 3 Malaysia, Mongolia, Pakistan, the Philippines, Sri Lanka, Singapore, Uzbekistan, Viet Nam and Thailand. As in other regions, these schemes differ markedly from country to country. Some schemes are well established and of long standing (India) while others have been set up only recently (Kazakhstan and Thailand); some reach a large percentage of enrolled students (Hong Kong SAR, China and Thailand) while others, relatively minor in scope, cover rather few (Philippines). Major objectives, too, differ from scheme to scheme, with the sharpest division drawn between those countries where loans schemes aim mainly at cost recovery (Singapore, Hong Kong SAR, China) and those with such social objectives as increasing higher education participation and access amongst less well-off groups (Philippines and Thailand). In some instances, administrative and other difficulties have led to the eventual abandonment of loans schemes (Indonesia and Sri Lanka); in other countries, planned schemes were stillborn though since revived (Mongolia). In a number of cases, governments have displayed an interest in reforming existing loans schemes that are in place or introducing new schemes. Policy reform in the field of student finance, as in other areas, should be guided by international experience of good practice (on what has worked well) and of mistakes to be avoided. But the general lack of available indepth appraisals of existing loans schemes in Asia has constituted a serious barrier to successful reform in the region. While some detailed appraisals relating to the workings of loans schemes in a number of countries have been carried out, these are not readily accessible; they are either unavailable in English or, being restricted internal documents of international organizations, are not in the public domain. The central aim of the UNESCO Bangkok/IIEP Regional Study on Student Loans in Asia has been to take some steps forward in filling this knowledge gap. The project sponsored five in-depth case studies of loans schemes in Asia, set in four countries: Mainland China and the Special Administrative Region of Hong Kong (SAR, China), the Republic of Korea, the Philippines and Thailand. The aim was to include loans schemes with 3. The adjective Korean is used in this booklet to refer to the Republic of Korea. 16

18 Introduction differing objectives, institutional structures and financial arrangements, as well as contrasting economic and political backgrounds. The case studies, focusing on the performance of these schemes, employed a broadly common framework and methodology, thus giving the opportunity for comparative analysis of the strengths and weakness of loans schemes operating within differing contexts. An earlier study of the new Thailand student loans scheme (reported in Ziderman, 1999), was used as a model for the other case studies conducted by local partners, 4 thus providing the necessary framework and methodology for the other case studies. These studies, which have already been published in the present monograph series, have been valuable in adding significantly to available case study material on student loans scheme operation and outcomes in Asia. Taken as a whole, the five case studies provide a broad canvas of differing experiences, practices and degrees of success. They constitute a rather unique database on the workings of student loans schemes in a defined region. No less important, however, is the opportunity afforded for integrated, comparative analysis the aim of the present study based on contrasts and similarities found across the case studies. The overall purpose of the comparative analysis in this monograph is not to provide an integrated summary of the five case studies but rather to draw out, from the loans scheme experience in the five cases, broader lessons for policy. 1.2 Outline of the paper The paper consists of 11 chapters. Following this introduction, a brief synopsis of the main workings of the five cases is provided in Chapter 2. Chapter 3 discusses alternative objectives of student loans schemes. Subsequent chapters deal with the organizational structure of loans schemes (Chapter 4), the funding of loans schemes (Chapter 5), the allocation of loans and repayment collection (Chapters 6 and 7, respectively), the financial viability of loans schemes (Chapter 8), equity, loans scheme subsidies and targeting (Chapter 9). In Chapter 10, an analysis of the strengths and weaknesses of case study loans schemes is provided while the final chapter, based on lessons learned from the case studies, discusses key issues in the design and reform of student loans schemes. 4. This study was carried out in the framework of the Thailand Education Management and Finance Study, UNESCO Asian Development Bank, Technical Assistance Project (T.A., 2996-THA), 1999, executed in association with the Office of the National Education Commission (ONC) and Thailand Research Fund (TRF). 17

19

20 Chapter 2 The five case studies 2.1 Case studies: general description This chapter describes in summary form the principal features of the main loans schemes in the five case studies. A brief description of the loans schemes in place in each case study is given in Table 2.1. Table 2.1 Case studies: general description of loans schemes Case study China Description Two national schemes, still being developed. One scheme is subsidized by government, while the second operates on commercial lines. Limited coverage but growing. Hong Kong SAR Republic of Korea The Philippines Thailand Veteran, centralized loans scheme (part of a comprehensive framework, providing grants and loans to poor and non-poor students). Broad regional coverage. A number of separate schemes, run independently and aimed at different target populations. Most schemes are based on existing financial funds. Very small, long-established national loans scheme, complemented by two new schemes (regional and high-level institutions). Marginal in total size and impact. Virtually no loan repayment. New, highly subsidized, national loans scheme with broad coverage; includes students enrolled at the upper secondary and tertiary levels. Low financial viability. 19

21 Policy options for student loans schemes: lessons from five Asian case studies The reader is referred to the original case studies for a fuller, indepth treatment: China (Shen and Li, 2003); Hong Kong, SAR (Chung, 2003); the Republic of Korea (Kim and Lee, 2003); Philippines (Kitaev, Nadurata, Resurrection and Bernal, 2003); and Thailand (Ziderman, 2003). 5 Apart from Thailand, which has a single loans scheme with very broad coverage, all the remaining case studies exhibited multiple loans schemes. Not all of these schemes are discussed in detail in the case study reports, since the central objective was to concentrate on lessons to be learned rather than to provide comprehensive coverage of all the schemes. A similar, eclectic approach is adopted in this report. Table A1.1 in the Appendix lists all the significant loans schemes in place, in each case study, with an indication of the level of detail provided in the case study reports for each scheme. The exclusion of the innovative and much discussed Australian scheme from the comparative study may occasion some surprise. Australia is not included for two reasons. First, the Australian case has already been extensively analyzed and documented; a recent account is provided in Chapman and Ryan (2002). Second, some of the central aspects of the scheme itself, particularly the system of income-contingent loan repayments, are not readily transferable to the rather different institutional setting in other countries in the region (Johnstone, 2001); this would be an example of inappropriate institutional transfer across countries. Indeed, apart from the Australian and New Zealand income-contingent schemes, all of the loans schemes at present in operation in the region (including all those examined in the case studies) are mortgage-type loans schemes. Thus our focus will be on mortgage type loans schemes (to be defined subsequently), which is the prevailing form of student loans scheme present in developing countries. However, some consideration of the efficacy of income-contingent loans schemes in developing countries will be provided in Chapter After the case studies were completed, institutional development has continued apace in some of the loans scheme cases, including both the introduction of new schemes and ongoing reforms of existing schemes. Thus, while they are not discussed in the case study reports, reference will be made to these developments at appropriate points in this paper. 20

22 The five case studies 2.2 China Student loans schemes were first introduced in China in The two national schemes, both formally established in 1999, are still under development; one is subsidized by government, the second operates on commercial lines. The Government Subsidized Student Loans Scheme (GSSLS) is the main loans scheme in China. It is aimed at poor students enrolled full-time in regular public universities. Loan capital is provided by four state-owned commercial banks. While educational institutions initially process loan applications, the commercial banks are responsible both for selection, lending out of loans and collection of due repayments; they also bear most of the default risk. The banks receive the commercial rate of interest on loans, half of which is paid by government. While the commercial banks put up the loan capital, the total loan volume is constrained by the system of institutional quotas, based on the total amount of interest support available from government and by the willingness of commercial banks to provide loans (if a bank does not want to lend more, it will offer fewer loans than indicated by the quota). There are no formal guarantors on loans; students own personal credit acts by way of guarantee, with no consideration of an applicant s credit history. Repayment is due four years after graduation. By the end of 2001, 272,000 students had received loans from the subsidized loans scheme over 30 per cent of applicants. But overall coverage is very low: only 3.8 per cent of students were in receipt of a loan. The relatively low earnings differential received by graduates, together with the short four-year repayment period imposes a relatively high burden on borrowers, of about 24 per cent of annual income. The scheme does not act very equitably, since the chances of receiving a loan are lower for very needy students and differ amongst students of similar economic status depending on the educational institution. Since banks shoulder most of the default risk, they tend to discriminate against the more default-prone students the poorer students and those enrolled in institutions of lower standing. In addition, since local governments provide the interest subsidy for local institutions, poorer local governments are less able to supply interest rate subsidies to local universities. 21

23 Policy options for student loans schemes: lessons from five Asian case studies Unlike the government-subsidized scheme, the General Commercial Student Loans Scheme (GCSLS) operated by commercial banks (and rural credit co-operative unions) is open to students in private as well as public universities, and regardless of socio-economic status. Interest on loans is charged at the commercial market rate, without government subsidy. Repayment periods differ, because the various participating banks have their individual loan regulations. Since loans are guaranteed through the assets of parents/guardians, thus minimizing default risk, the scheme is limited in practice to students from middle and upper class families with the required assets for collateral. 2.3 Hong Kong SAR, China The student loans scheme in Hong Kong SAR is a major component of the governmental system of financial aid to students that has evolved over the past decades. It is at once the oldest of the five cases and, in many ways, the most successful. The scheme is operated by an autonomous public loans organization, the Student Financial Assistance Agency (SFAA). At present, the scheme consists of interest-free loans under the Local Student Finance Scheme (LSFS) (which are means-tested and take into account the financial situation of the student s family) and of the interest-bearing NLS loans (for which having student status is the only requirement). The loans are meant to cover living expenses, complementing the grants given towards academic expenses. The interestfree LSFS loan is sustained by continuing government subsidy, while the interest-bearing NLS loan is expected to operate on a no gain, no loss basis. Loan coverage is relatively high: In 2000/2001, some 28,000 students took up an LSFS loan and over 10,000 took up an NLS loan, together constituting some 50 per cent of the eligible student population. The allocation of student loans in Hong Kong has been mainly based on considerations of equity, efficiency and adequacy. First, students from less well-off families receive greater financial assistance. The loan entitlement varies according to a formula, which takes each applicant s family financial situation into consideration. The goal of the system is to ensure that no qualified student is deprived of higher education because of lack of funds. Second, the maximum amount of a loan is adjusted so as to correspond to the general living needs of a student. This adjustment is achieved through regular surveys of student expenses and the compilation of a Student Price Index. 22

24 The five case studies The allocation of financial assistance has also been used to encourage development in areas of study required by society. At different stages, various grants and loans in Hong Kong have been targeted at students in teacher training, information technology, financial services and creative media. Increases in university tuition fees in the 1990s which generated a vast amount of additional student applications, were also followed by a considerable increase in the number of awarded grants and loans. Indeed, a large portion of costs recovered through the tuition hike was offset by increases in grants and loans during the same period. Thus the development of the government student loans scheme has not reduced the government s financial commitment to higher education. Some commercial student loans schemes have emerged recently in Hong Kong. However, the scope of these schemes is rather limited and these loans are usually contingent upon the acceptance of other services provided by the banks offering the loans. 2.4 Republic of Korea There are six government-supported loans schemes for higher education in the Republic of Korea, covering about 16 per cent of student enrolment. Among these schemes, the Ministry of Education and Human Resources Development (MOE) and the Government Employees Pension Corporation (GECP) loan programmes are the largest, accounting for almost 86 per cent of all student loans. The MOE scheme is the largest of the six loans schemes run with government financial support. It disburses interest-subsidized loans to the poor over the entire higher education sector; it targets poor students, giving priority to students from unemployed and low-income groups. The MOE scheme uses commercial banks for a number of the loans scheme functions: loan fund provision, loan administration, repayment collection, etc. The Ministry only decides on the size of loans and on aspects of general policy, such as interest rates, selection criteria and quota for each institution, and makes up the interest difference. The GECP loans are available to government employees and their children and do not target the poor specifically. The GEPC scheme uses its own fund for the financing of loans and transfers the loan amount to the recipient s individual bank account. 23

25 Policy options for student loans schemes: lessons from five Asian case studies Both schemes usually cover tuition fees (including registration fees) but not living costs. The responsibility for repayment of MOE loans falls on the students themselves. They are able to delay payment until they receive higher earnings. In contrast, recipients of a GECP loan are government employees, who have to pay back the loan received for financing their children s education. A notable feature of the student loan system in the Republic of Korea is that the government budget is not the source of initial loan capital: this is put up by the commercial banks in the MOE scheme or provided from existing funds in the others. All schemes except one use commercial banks to administer loan payment and repayment collection. In 2003, the Korean government introduced a new student loan and scholarship programme, specifically targeted on engineering students. The objective of the programme is manpower needs and covers around 37,000 students (about 1 per cent of student enrolment). This scheme was introduced after the Korean case study had been completed. Student loans cover tuition fees only, and not living expenses; this is a particular problem for the poor. Of the six loans schemes, only two (including the MOE scheme) directly target the poor; the other four are directed at specific population groups, such as government officials, teachers or industrial workers. The loan programme of the MOE, the only sizeable scheme directed towards the poor, is not successfully targeted on this group. In practice, it may happen that loans are allocated to students that are not necessarily in financial need. This is because some universities and colleges allocate loans on a first-come, first-served basis and also, because some institutions apply merit-based and needs-based criteria in combination in the selection process. The total amount of loans provided by the six schemes came up to 970 billion won (or US$750 million) in The number of recipients was 488,900 in 2000, which is approximately equivalent to 30 per cent of four-year university/college students in the Republic of Korea. The average loan amount per student is around 2,000,000 won for all programmes. Nominal interest rates for the six listed programmes vary from 0 per cent to 5.75 per cent; the typical repayment period is 4 years. Overall, student loans in the Republic of Korea are highly subsidized and have not been used as a tool to support cost recovery in higher education. 24

26 The five case studies 2.5 The Philippines Student loans schemes in the Philippines have a relatively long history: The Study Now, Pay Later (SNPL) scheme was set up in Restricted to poor students enrolled in public universities, the scheme has never operated on a large scale and has exerted only a marginal influence on higher education financing; its record of performance has been low. The SNPL has operated in two phases. In the first phase ( ), government financing institutions (GFIs) including the Social Security System and various government banks were required to provide the initial loan capital and were charged with the task of operating the loans scheme, including the lending and repayment mechanisms. The GFIs displayed active resistance to the scheme: over the period from 1976 to 1986, they disbursed only 40 per cent of assigned loan funds and, given government default guarantees, collected only about 40 per cent of due loan repayments. After the disengagement of the GFIs in 1989, the scheme has been administered by the Office of Student Services, within the Commission on Higher Education (CHED). However, these new bureaucratic arrangements both lacked administrative capacity and incentives to operate a viable loans scheme. Stringent budgets continue to restrict coverage to a few thousand new borrowers each year and loans are often insufficiently large to cover tuition fees. With a disinclination to collect loan repayments and negative borrower attitudes towards repayment, the repayment rate has now dropped to a mere 2 per cent. Two new government schemes have emerged recently, focused on a specific (economically deprived) region and on particular institutions the university Centres of Excellence (COE). While both schemes are government funded (via CHED), the educational institutions concerned are responsible for the direct administration of the scheme, including disbursement and, notably, loan repayment. Some institutions have declined to participate in the COE scheme. Student loans have never operated on a large scale in the Philippines; their impact on higher education finance has been minimal and their performance record poor. Current experimentation with governmentfunded but university-based loans schemes is also operated on a small 25

27 Policy options for student loans schemes: lessons from five Asian case studies scale with questionable success. No clear plans are afoot to develop any nationally based scheme of broad coverage and sizeable impact in the foreseeable future. 2.6 Thailand The Thai loans scheme, which began operating in 1996, is aimed at disadvantaged students, enrolled in upper secondary general and vocational schooling as well as tertiary education, in both the public and the private sector. The coverage of the scheme is extensive, reaching about a quarter of the students in upper secondary, a third of university students (excluding the two open universities) and up to a half of all students enrolled in Rajabhat (teacher training) institutes. The scheme is run through the national Student Loan Scheme Committee (and its loans office), which receives annual subventions from the national budget. Borrower selection is made by individual educational institutions (schools, colleges and universities), which receive loan budgets through a system of top-down budget allocation, from the central loans committee, via the Ministries of Education and University Affairs (now divisions in a combined Ministry of Education). Individual educational institutions have considerable autonomy in the lending process, including decisions on the size of individual loans (up to a designated maximum) and on the purpose of a loan (tuition fees, accommodation and other living expenses); this highly decentralized method of loan distribution leads to considerable inequities across the educational system. While the scheme is aimed at the needy student, targeting is not effective. The family income ceiling set for loan eligibility is three times the income officially designated as defining poverty. Loan budget allocation to educational institutions is only very loosely tied to the social profile of the student population at a given institution (in the case of universities, allocation criteria are not based at all on student poverty within the university or related to need). The rapid growth of the scheme, considerably in excess of plans, has led to budgetary cutbacks in the system for new borrowers. This is reflected by the falling average loan size overtime (even in nominal terms) rather than by the number of new borrowers (which has continued to rise at tertiary level). Institutions have evidently preferred to spread declining 26

28 The five case studies loan budgets over a broader student population, again blunting the effect of the scheme in assisting the most needy students. The Thai loans scheme receives a considerably higher level of government subsidy than the loans schemes in the other case study countries; with a one per cent rate of interest and repayment in nominal terms, students are in practice required to repay only a small fraction of the value of the original loan. 2.7 Dominance of mortgage-type schemes From the foregoing summaries, it is evident that there are considerable differences across the various case study schemes, in terms of such parameters as schemes organizational structure, student coverage, loans schemes objectives, funding sources, loan allocation procedures and collection methods. 6 An analysis of the salient differences between loans schemes and how these differences impact upon loans scheme performance are central concerns of the present study. However, all the schemes examined in the case studies have one common element: their underlying loan arrangements are, as in the case of virtually all student loans schemes in developing countries those of a traditional mortgagetype loan. The central features of mortgage loans are presented in this section, as a background to the comparative discussion in the chapters that follow. In the conventional mortgage-type student loan, repayment is made over a specified time period in fixed, usually monthly payments. A designated rate of interest on the loan and a maximum loan repayment period are employed to calculate the fixed periodic payments. The higher the rate of interest that is set and the shorter the repayment period, the greater the monthly sum required as repayment will be. One advantage of mortgage-type loans, from the perspective of the body charged with loan repayment collection, is the ease in ascertaining the monthly repayment obligations of the individual borrower, which are time-invariant. However a regime of fixed-level repayments may impose a heavy burden on new graduates in the initial years following graduation, given low starting salaries of new graduates and the reality of graduate 6. Additional comparative information on the case study loans schemes is summarized in Table A

29 Policy options for student loans schemes: lessons from five Asian case studies unemployment in many developing countries. A heavy repayment burden, it is often claimed, will lead to repayment default. Yet, in practice, a number of measures may be employed, within mortgage loan repayment procedures, to obviate the occurrence of heavy repayment burdens; some of these are standard practice, others could be easily adopted. These measures include, for all borrowers, the delay of repayment obligations for a period of grace following graduation and the introduction an extended repayment deadline. The introduction of a lower interest rate charge (acting as a general loan subsidy) would also lighten the burden of monthly repayments, but this would lead to a shortfall in total repayments. 7 Finally, repayment deferral may be made available to borrowers on an individual basis, in cases of financial hardship: to the unemployed and to low-income workers (whose income falls below a designated low income ceiling). 7. Financial issues are discussed in detail in Chapter 8. 28

30 Chapter 3 Why student loans schemes? 3.1 Alternative loans scheme objectives National student loans schemes may be established for diverse reasons. We may identify five different sets of objectives for student loans schemes, which in turn will influence the design and operation of the scheme as a whole, as well as its financial sustainability (this topic is discussed in greater detail in Ziderman, 2002). These are: 1) budgetary objectives (income generation), 2) facilitating the expansion of higher education, 3) social objectives (improving equity and access for the poor), 4) meeting specific manpower needs, and 5) easing student financial burdens. In practice, any given scheme may incorporate more than one objective. Budgetary objectives (income generation) Public universities throughout the world, and particularly in developing countries, are under-financed. Leaving aside situations where the university system is expanding (we deal with this subsequently), constrained government budgets may lead to a general under-funding of public universities. This situation may arise for a number of reasons; three of which are discussed here (Table 3.1). First, additional government funding may not be available to enable universities to maintain enrolment levels and quality in the face of rising unit costs. Second, across-the-board cuts in overall government expenditures, including higher education, will pressure the public university sector to seek alternative funding. Third, many countries have adopted policies that favour basic, over tertiary, education, leading to a reallocation of funding away from the universities to other sectors of the education system that display higher social rates of return. In all these cases, budgetary parsimony has resulted in public universities turning to greater cost recovery, in an effort to tap alternative sources of funding. The main thrust of these policies is to be seen in the 29

31 Policy options for student loans schemes: lessons from five Asian case studies introduction, or increase, of student payments for services received. These may take the form of higher, more realistic tuition fees or increased payments for subsidized lodgings and meals. Recourse to the banking system for a regular loan to ease this payment burden may be unavailable; banks are notoriously loath to lend for educational courses a clear case of market failure. Hence there is a role for a government-backed student loans scheme, offered at commercial rates, to fill this gap. This would mean that students are able to finance their education and living expenses through borrowing; the repayment burden is eased by the expected enhancement of earnings that the additional education makes possible. Table 3.1 Alternative objectives of student loans schemes Loans scheme objectives 1. Budgetary objectives (income generation from tuition fees) Generate income to maintain university enrolments and output or quality, in response to rising public university unit costs (additional government funding not available). Funding replacement: in response to reduction in overall government expenditures, including the education sector. Funding replacement: in response to reallocation of public educational budgets, from universities to other sub-sectors with higher societal returns. 2. University system expansion Generate additional tuition fee revenues to (partially) finance expansion of the public university sector. University expansion via growth of the private university sector (to minimize the state s role in financing expansion. 3. Social objectives (equity and access for the poor) Loans targeted towards needy students. Cross-subsidization: grants for needy students financed by income from higher fees. 4. Manpower needs Meet specific occupational or regional manpower needs. 5. Student assistance Ease student financial difficulties during study. Increase student commitment. Improve students financial independence. Loans schemes primarily concerned with cost recovery are also frequently subsidized and targeted on the poor. But these elements are 30

32 Why student loans schemes? not an integral part of a cost recovery loans scheme, which, in principle, should be offered at market interest rates and available to all, not only the poor. The availability of additional revenues from student fees, facilitated by the introduction of a loans scheme, may not lead to increases in net funding to the education institutions. Whether it does so or not will depend on whether additional revenues from fees are offset by commensurate reductions in public funding of the institutions. Facilitating higher education expansion Governments have responded to the growing social demand for higher education through policies leading to expanded student enrolments; yet, due to national budgetary constraints, the growth in student numbers is largely unmatched by commensurate additional government funding. Responding to the pressures of growing social demand for education expansion will require sizeable increases in public expenditure. These increases could be contained by offsetting additional revenues from increases in student fees, perhaps matched by reduced public institutional support. A complementary measure is to encourage the growth of private educational institutions. Students pay full costs at private universities, with a minimal burden on the public purse. However, full-cost fees are likely to be sizeable and beyond the reach of large segments of the population. A student loans scheme may have a central role to play in easing the burden of private fees, particularly if private education is to be widely available and not to remain the purview of the rich. Social objectives (increasing access for the poor) None of the above reasons for introducing a student loans scheme constitute a case for subsidized student loans. The widely held objective of increasing the educational participation of the poor does. In many countries the relatively low enrolment of poor and disadvantaged youth in non-compulsory education is a cause of social concern; increasing the access to schooling among these segments of the population has become a major element in educational and social policy. There is a broad consensus that clear financial incentives need to be offered, not only to overcome the burden of fee payments and living expenses but also to offset both 31

33 Policy options for student loans schemes: lessons from five Asian case studies parental resistance to reductions in family income and the risk that the benefits of the educational process may not be sizeable. The traditional, and most effective, method of enhancing the educational access of the poor has been through the provision of meanstested grants (scholarships) to cover tuition fees (where schooling is not free) and usually living expenses as well. This remains the dominant approach for secondary education, although there are exceptions in some countries; we have noted for example that Thailand provides student loans for upper-secondary students. However, a widespread scholarship scheme is likely to be expensive; the use of loans rather than grants offers a method that both increases access for the poor and reduces, or at least contains, public expenditure over the longer term, as loan repayments build up. To be effective in increasing the education access of the poor, loans need to be offered with attractive conditions. Hence the justification for subsidized loans, in terms of grace periods for repayment, belowmarket rates of interest and repayments not fully linked to inflation. Loans schemes meant to assist the poor should be designed so as to reach this population, otherwise the central objective of the scheme will be lost. A less direct role for student loans under this objective is the provision of grants for needy students, financed by income generated from (higher) tuition fees paid by more affluent students, for whom non-subsidized loans are available (i.e. at market or near-market interest rates). Manpower needs Loans schemes may aim specifically at providing support for students who are willing to study in fields of national manpower priority (e.g. engineering) or to work in areas of social importance (doctors or teachers servicing remote rural areas). Loans schemes may either be developed specially for these groups (such as medical students) or advantageous repayment conditions may be offered within the context of a general, non-subsidized loan programme. Easing student financial burdens The final group of objectives is more typical of loans schemes in developed countries. Even when tuition fees are minimal, students (both the more affluent and the disadvantaged) may face considerable financial 32

34 Why student loans schemes? burdens. Potential earnings are foregone while studying, and living expenses may be considerable, especially when the student does not attend a university close to home. Financial pressures, which may have negative effects on a student s academic performance (thus compromising the process of effective human capital investment), can be mitigated by the broad availability of student loans. While such burdens may fall relatively heavily on the poor, in principle loans for this purpose could be made available to all students, including more affluent as well as poorer students, as long as these loans are not subsidized. There are two additional arguments under this category of objectives, which are often voiced: increasing student commitment and offering university-level students financial independence. Students, particularly from more affluent backgrounds, who benefit from overly subsidized university education, may fail to develop a sufficiently mature approach or commitment to studies without the discipline imposed by payment of realistic university fees; loans schemes make possible the imposition, or augmentation, of tuition fees. Finally, it is argued that students who have reached adulthood, should not be forced to rely upon parental financial support, which might not be forthcoming even in the wealthier families. Multiple objectives Some schemes may be aimed at more than one objective. Often multiple objectives can be accommodated, without compromising any particular objective; for example, loans available to all students to facilitate fee increases and cost sharing, may also offer greater student independence. A loans scheme aimed at greater cost sharing between the government and students may be constrained by insufficient loan budgets and some applicants will be denied a loan. Loans might be made available on a first-come, first-served basis, but it may be preferable to ration loans objectively, using parental income as a screen for selection, in order to give priority to poor students. The importance of defining objectives A clear distinction can be drawn between objectives 1 and 2, and between the group of objectives 3, 4 and 5. The first group of objectives is concerned with generating funding for the university sector, in the case of tight government budgets and ensuing increase in the social demand 33

35 Policy options for student loans schemes: lessons from five Asian case studies for higher education; they have constituted the major rationale for the spread of student loans schemes in industrialized countries. The second group of objectives, however, is not primarily concerned with university funding as such, but is broader in scope, with a societal perspective. Most student loans schemes are highly subsidized, particularly in developing countries. However, in most cases extensive subsidies cannot be sufficiently justified: The aim should be full loan recovery. Loans schemes targeted to the poor constitute an exception, as possibly do those aimed at easing particular manpower shortages. We note that loans for objectives 1 and 2 should be restricted to students enrolled in public universities (except for those loans aimed specifically at private university students). However, in meeting the remaining three objectives, loans should be available to students in the public and private sectors on an equal basis. Implications for loans scheme evaluation The preceding discussion carries important implications for loans scheme evaluation. There can be no standard approach to evaluating the efficacy of individual loans schemes. A given student loans scheme will need to be evaluated in the context of the central objective(s) that it is designed to achieve. For objectives groups 1 and 2, the most relevant form of evaluation is a financial appraisal of the scheme (as discussed in Chapter 8) in particular, to probe the level of loan recovery achieved by the scheme. In contrast, loans schemes with equity objectives, aimed at increasing university access for the poor, should be evaluated, primarily, on the basis of their success in reaching these populations and of the extent to which the availability of loans results in raising the participation of these disadvantaged groups in higher education. Testing the efficacy of loans schemes with manpower needs objectives consists in gauging their success in encouraging students to enrol in these priority areas of study and to subsequently enter employment in these fields or designated regional locations. Loans schemes with student assistance objectives (objective group 5) will probably need to be appraised by survey methods, aimed at eliciting information on the easing of student financial burdens and enhanced commitment. 34

36 Why student loans schemes? 3.2 Country case studies: loans scheme objectives Against this background we examine the objectives, either declared or implicit, of the loans schemes comprised in our five case studies. These are summarized in Table 3.2. Perhaps surprisingly, the major objectives for all schemes listed in the table are clearly societal mainly social and also student assistance (objectives 3 and 5, respectively); they do not relate directly to the financial crisis facing universities in these countries. For the loans schemes in China (GSSLS), Republic of Korea (MOE and HRD), the Philippines and Thailand the major objective is social, aimed at offering financial help to students from poor and disadvantaged backgrounds. Among the Hong Kong schemes, the largest scheme (LSFS) is social in aim, while the complementary scheme (NLS) is available to students from all income groups, including poor students who wish to topup subsidized LSFS loans. The highly subsidized Korean Government Employees Scheme (GEPC) is available to government employees and their families, regardless of family income. Some schemes do meet university finance objectives too. The lowcoverage loans schemes in China facilitated the introduction of tuition fees after a 40-year history of free tuition. In this case, however, as in the case of other countries, financial objectives, rather than being a declared central objective of the scheme, tend to be embedded in the way the schemes in practice operate. Thus the provision by commercial banks of initial capital for student loans in both China and the Republic of Korea (MOE and HRD) means that, in practice, the major financing burden of the loans scheme is shifted from the government to the banking system. (The issue of the source of capital for student loan funding is discussed in Chapter 4.) In Thailand, the average amount of loans given to students at private education institutions exceeds by a substantial margin that of loans given to students enrolled in public institutions, because of the sizeable differences in tuition fees in the two sectors. Since all loans in Thailand are very highly subsidized (see Chapter 5), the availability of loans to students in private institutions results in the subsidization of the costs of private education, thus encouraging the growth of private institutions (particularly universities). Student loans in the Philippines are restricted to students enrolled in a range of defined priority courses, the aim of which is to produce qualified manpower in fields necessary for economic 35

37 Policy options for student loans schemes: lessons from five Asian case studies and social development. Finally, a new Korean loans scheme, aimed at increasing the supply of engineering manpower in the Republic of Korea, has dominantly manpower objectives. 8 The central objective of a loans scheme may change over time. In Hong Kong it was in the context of university system expansion and increases in university fees that the various loans schemes were either improved or formally constituted at different stages. In the early 1990s university tuition fees in Hong Kong were raised (by a factor of 2.65) to generate greater cost recovery for university expansion; about two thirds of revenue generated from by policy of cost recovery through tuition fees was offset by increases in LSFS expenditure in the mid 1990s (Chung, 2002). In Thailand, current discussions on reforming the student loans scheme envisage a shift in emphasis from social objectives (assisting the poor) to budgetary objectives (cost recovery). Today, all the major schemes examined in the case studies are predominantly social in aim, justifying government loan subsidies. This raises a number of issues, all which will be addressed in subsequent sections of this monograph. What effect does subsidization have on the financial efficacy of the various loans schemes? The higher the government subsidy is, the more effective the loan will be in assisting the poorer students; however, excessively generous levels of subsidy may compromise the financial viability (and sustainability) of a loans scheme since loan recovery through repayments may be unduly low. A financial appraisal of selected loans schemes from the case studies is provided in Chapter 7. Subsidized loans are justified where successfully targeted on the needy student. But how effective are the loans schemes in question in reaching target groups? Is there a sizeable leakage of subsidized loans to non-targeted (non-poor) students? This issue is discussed in Chapter 8. Where loans schemes are well targeted, are they effective in assisting needy students through the provision of adequate individual levels of support? Chapter 5, concerned with institutional roles assigned to the main actors in loans schemes, also takes up this issue. 8. The scheme was introduced after the Korean case study had been completed and therefore is not discussed in the Korean report. 36

38 Why student loans schemes? Table 3.2 Country case studies: loans scheme objectives Case study Major objective Secondary (implicit) (loans scheme) objective China Social: Budgetary: (GSSLS subsidized aimed at students facilitated introduction of tuition scheme) from poor families. fees; funding burden shifted from the government to the banking system China Student assistance: Budgetary: (GCSLS commercial loans are not means- facilitated introduction of tuition scheme) tested. fees; funding burden shifted from the government to the banking system Hong Kong (LSFS subsidized scheme) Social: aimed at students from poor families. Hong Kong Student assistance: (NLS non-subsidized loans are not meansscheme) tested. Republic of Korea Social: Budgetary: (MOE) aimed at students from funding burden shifted from the poor families. government to the banking system Republic of Korea Manpower: Student assistance: (MOE Engineering) loans and scholarships loans for engineering majors for engineering students. Republic of Korea Student assistance: (GECP Government loans for families of Employees Scheme) government officials. Philippines Social: Manpower needs: (all schemes) aimed at students loans restricted to students in from poor families. priority courses Thailand Social: University expansion: aimed at students subsidizes students in private from poor families. institutions 37

39 Policy options for student loans schemes: lessons from five Asian case studies 3.3 Policy alternatives to student loans schemes This chapter concludes on a cautionary note. While recognizing that student loans schemes may differ as to the objectives that they are designed meet, it should also be noted that a student loans scheme may not necessarily constitute the optimal mechanism for meeting a particular objective. Prior to introducing a loans scheme, or carrying out a major reform of an existing scheme, policy makers should weigh the advantages and disadvantages of available policy alternatives to loans schemes, which may be better able to achieve the same objectives. Such an analysis may indicate that these alternative measures and policies should be introduced in concert with a loans scheme to strengthen its effect or, if seen preferable, they may constitute an alternative approach. Careful social and economic analysis of policy alternatives is a prerequisite for sound policy reform. Table 3.3 summarizes some of the policy alternatives for achieving the desired outcomes. 38

40 Why student loans schemes? Table 3.3 Policy alternatives to student loans schemes Loans scheme objectives Policy alternatives (or complementary policies) Budgetary objectives (income generation from tuition fees) Revenue generation: rising public Provide incentives for greater university unit costs Funding replacement: reduction in As above government expenditures Funding replacement: reallocation As above of public educational budgets institutional internal efficiency; seek other sources of university income generation University system expansion Revenue generation to fund public Other sources of income generation university expansion Growth of private university sector Direct subsidies to private universities Social objectives (equity and access for the poor) Loans targeted to needy students Scholarships for needy students Cross-subsidization Alternative sources of finance for scholarships to needy students Manpower needs Meet specific manpower needs Other incentives: e.g. subsidized housing, income tax reductions Student assistance Ease student financial difficulties Opportunities for student employment; flexible study programmes Increase student commitment Work-study and national service schemes Financial independence Opportunities for student employment 39

41

42 Chapter 4 Organizational framework The most striking differences between the provision of student loans across countries, and which are apparent in the loans schemes examined in the case studies, stem from the contrasting organizational frameworks adopted. Yet the extensive literature on student loans, while dealing with the financial aspects of loans schemes in considerable detail, remains largely silent on issues of organizational structure. In this chapter, we examine two dimensions of organizational structure, both of which may impinge upon the efficacy of loans schemes in operations. 4.1 Unitary scheme versus multiple schemes Central issues An examination of the provision of student loans across countries shows considerable variation in the number loans schemes in place. A country may have set up a single, national loans scheme to cater for eligible students; in contrast, there may be a number of separate loans schemes in operation, usually each serving a different client group. What are the relative strengths and disadvantages of these two alternative approaches? Can a hybrid scheme be developed to retain the advantages of each? There are a number of advantages in setting up a single, central national loans scheme; these include: size of impact in achieving the loans scheme s central objective; economies of scale, allowing low administrative costs per loan; administrative efficiency and effectiveness; specialized staff and functions, such as the ability to undertake evaluation and follow-up studies. Indeed, where governments have set up major loans schemes to meet a central policy objective, these schemes have been of the unitary type; notable examples are the Australian HECS scheme and the new British incomecontingent scheme (both to facilitate increases in university tuition fees), as well as the Thailand scheme (designed to assist poor students). 41

43 Policy options for student loans schemes: lessons from five Asian case studies Where multiple loans schemes are in place, these have generally been the result of ad hoc or historical processes rather than of deliberate design. Yet, while they may lack potential national impact, comprehensive coverage and may be poorly co-ordinated, they can be effective where there are multiple loan objectives and can serve the particular needs of very different target groups. Contrasting case study practice The case studies display a picture of contrasts. As shown in Table 4.1, we may order the five case studies on a notional scale, ranging from a purely unitary scheme (Thailand) to a case of multiple, independent schemes (Republic of Korea), with mixed cases falling in between. However, our interest lies in the mixed cases (particularly Hong Kong) rather than in those at the extremities. The Hong Kong case is instructive in demonstrating how it is possible to build on the strengths of a robust unitary scheme, by adding the policy flexibility afforded by multiple schemes. Table 4.1 Unitary or multiple schemes: case study practice Scale Case study Description Unitary scheme Multiple schemes Thailand Hong Kong Philippines China Republic of Korea Unitary scheme run under the umbrella of a national loans agency the Student Loans Scheme Committee A national agency the Student Financial Assistance Agency runs complementary loans programmes aimed at different target groups Three schemes, all of limited scope, are loosely administered by the Office of Student Services, at the Commission on Higher Education Two independent loans schemes one government subsidized, the other run commercially Six separate loans schemes, aimed at different target groups, but with no national co-ordination 42

44 Organizational framework In Hong Kong, the Student Financial Assistance Agency (SFAA) runs a number of student finance assistance schemes. The largest, aimed at the poor, is the means-tested Local Student Finance Scheme, providing a combination of grants (to cover tuition fees) and loans (for living expenses) for needy students. This is complemented by a non-meanstested loan programme which is available to all students ( student assistance objectives) and operated on a no gain, no loss basis (i.e. not subsidized). The Hong Kong experience is instructive in exemplifying a robust, well-financed loans scheme that benefits from the advantages of both unitary and multiple loans schemes by operating complementary student assistance programmes (with differing client groups), under a common organizational umbrella. 4.2 Centralized or decentralized loans schemes? Advantages and risks: illustrations from the case studies The case study loans schemes differ markedly in the extent of functional decentralization (Table 4.2). The Hong Kong scheme is the most highly centralized, while decentralization has been taken furthest in the Thai scheme and, to a lesser extent, in the Philippines. Our discussion of the relative advantages and risks associated, respectively, with centralization and decentralization draws upon experience from these three cases. In Hong Kong, the well-funded Student Financial Assistance Agency (SFAA) is responsible for all facets of the loan process; 9 these arrangements work to good effect. One major advantage of centralization is that, through centralized procedures and the use of common criteria for loan allocation, horizontal equity in the distribution of loan budgets is attained; this is particularly important for the major LSFS scheme, which is aimed at serving the poor. The task of loan repayments collection also falls within the remit of the SFAA, which acquits itself of this task well. But this finding of successful repayment collection by the loans agency may not be applicable to other regional contexts, where much larger geographical areas and more diffuse populations are concerned. 9. This is also the case in the Korean loans scheme for government employees (GEPC). 43

45 Policy options for student loans schemes: lessons from five Asian case studies Table 4.2 Centralized or decentralized schemes: case study practice Scale Case study Advantages Risks Centralized Hong Kong Facilitates horizontal equity Less easy to identify and in distribution of loans target poor students (employs common criteria) No comparative advantage in repayment collection, but collection eased by small geographical area China, Rep. Commercial banks can assume of Korea many roles, in both distribution of loans and particularly in collection, where they have expertise Philippines Educational institutions Loan repayments collection perform a useful role in educational institutions promoting the scheme and unlikely to be effective (COI initial vetting of applicants and Region 5 schemes) Thailand Educational institutions Loan budgets of educational perform a valuable role in institutions unrelated to promoting the scheme, poverty profile of student Decentralized identifying poor students and enrolment conducting the initial vetting Selection of applicants and of applicants fixing of individual loan size by educational institutions leads to horizontal inequities across institutions (in both coverage and loans size), possible nepotism and abuse At the other extreme lies the considerably decentralized Thai scheme. Rather than employing a centralized clearing house to service loan applications from potential borrowers as in the Hong Kong case, loans are distributed to students by the individual educational institution. Education institutions receive a loan budget through a top-down budget allocation process, which is unrelated to the poverty status of students enrolled in the institutions. A precondition for a properly functioning loan distribution system at the institutional level, lacking in the Thai case, is a well designed formula for allocating the total loan budget across education institutions: such a formula would assign budgets in relation to potential student need (poverty level) in each institution. The system further lacks horizontal equity because of the widely differing loan distribution policies across educational institutions. 44

46 Organizational framework In the two new Philippine loans schemes restricted to Region 5 and to Centres of Excellence respectively responsibility for repayments collection has been placed in the hands of the educational institutions themselves. Institution-based, decentralized repayment collection was introduced in an attempt to improve on the dismal collection record and massive repayment default of the veteran Study Now, Pay Later scheme, where repayments is administered by the lending body itself (the Commission on Higher Education). However, the collection of loan repayments by educational institutions is unlikely to be effective. Some institutions have been hesitant to buy into the scheme; both incentives and administrative capacities may be absent. However, some decentralization in the distribution system may be desirable. In the Thai scheme and also in the Philippine schemes, educational institutions perform a positive role in promoting the scheme, identifying poor students and vetting application; these are useful roles, which could be incorporated, to good effect, into centralized loan distribution regimes. In the Chinese and Korean Ministry of Education schemes, commercial banks play a central role in loan capital provision, distribution and collection. A broader discussion of these issues appears in the following three chapters. Within the chosen organizational structure of a loans scheme there are, potentially, many institutional actors that may be assigned specific roles in running the scheme. In an attempt to draw out lessons for policy to identify good practice as well as weaknesses to be avoided we analyze existing practice in the five case studies, pertaining to three sets of institutional issues, which are seen as central to the operation (and success) of any loans scheme. These are: the funding for loans schemes; arrangements for borrower selection and loan budget distribution (including the decision on individual loans size); and arrangements for collection of loan repayments (including minimizing repayment default). These institutional issues are discussed, respectively, in the three chapters that follow. 45

47

48 Chapter 5 Assigning institutional roles: funding Commercial bank loans for financing education (at reasonable rates of interest) are usually not available, because most potential students (or their families) cannot provide collateral to cover the loan nor offer proof of creditworthiness. In most cases (though not all, as we shall see) the government intervenes by supplying the initial capital for the loan, subsidizing interest payments, bearing the risk of default and absorbing much of the administrative costs of operating the loans scheme. In this chapter, we discuss variations in practice, across the five case studies Provision of capital Every type of loan needs to be financed initially and subsequently to be repaid over time. This reality has led to two widely spread misconceptions regarding capital provision for student loans. The first misconception is that student loans will require the government to provide additional funding for the loans, in the absence of another funding source. The second is that this additional governmental financing burden is only temporary and will diminish over time as repayments accrue to the revolving loans fund; ultimately, loans for new student cohorts will be financed from repayments on previous loans. In the pure cost-sharing model, no additional government financing is necessary to finance a new loans scheme, as long as student enrolments are constant. Cost-sharing entails a shifting of the existing financing burden from the government onto other beneficiaries of higher education (in this case, students or their parents). The process is as follows. The government reduces its budgetary subvention to university institutions, with universities maintaining their revenues by introducing (or raising) fees for tuition and living items. As illustrated in Figure 5.1, the government sets up a student 10. A complementary discussion of the issues in this chapter are given in an excellent paper by Johnstone (no date). 47

49 Policy options for student loans schemes: lessons from five Asian case studies loans scheme, from which students obtain loans to pay the newly introduced fees. In the event that all students take out loans to pay their fees, the cost to the government of funding the loans scheme is equal to the total reduction in its direct budgetary support to the universities. 11 Indeed, to the extent that some students do not take a loan, total government expenditure on higher education may be reduced, at the outset. Where a loans scheme is introduced in the context of university system expansion (as is usually the case), then an initial financing burden does fall on government; this is also the case where loans are introduced for reasons of equity (to help the poor gain access), for manpower needs or to ease student burdens generally, as discussed above. Figure 5.1 Higher education budget allocations, with cost sharing Government budget Loans scheme Budgetary allocations Repayments Loans Students Fees Universities 11. In some countries (such as the United Kingdom), a system of student loans has replaced student grants for living expenses; in this case, too, additional government funding is not necessary. 48

50 Assigning institutional roles: funding The second misconception is that a loans fund can be regarded as a revolving fund, in the sense that annual receipts from loan repayment, building up over time, will finance new loan commitments. This view is often expressed in support of the establishment of new loans schemes. But even in the pure cost-sharing model, a student s repayments will fall short of the original sum borrowed because of loan subsidies, repayment default and government-absorbed administration costs; these issues are taken up further in Chapter 8. More usually, however, growing student enrolments will exert pressure on the government to increase student loans scheme annual budgets; in this case, even full repayment of loans (without leakage from the scheme) would require continued government funding for the loans scheme. Yet this misconception is deeply ingrained; in the national budget in Thailand, for example, the budgetary item relating to the student loan fund (the Education loan Fund) is referred to as a revolving fund. In summary, a new loans scheme is likely (though not necessarily) to require an initial infusion of capital, usually provided by the government (from the current budget or from borrowing); given less than full loan recovery, the financing burden will remain, though perhaps diminish, overtime. While most of the loans schemes presented in the case studies derive their capital funding from government, there are some alternatives (Table 5.1, column 1). Before considering these, we should note that there is considerable variance in the adequacy of government loan capital provision across the countries in which the government assumes this role. The Hong Kong system is well financed and provides adequate loan support for well over a third of student enrolment; in contrast, the no less extensive Thai scheme has been constrained by inadequate government budgeting, thus tempering the growth of the scheme and levels of support. The long established Philippine Study Now, Pay Later scheme has always operated on a very small scale, with little impact on the higher education system as a whole, due to meager government budgetary provision; this is also true of the new COE and Region 5 schemes. 49

51 Policy options for student loans schemes: lessons from five Asian case studies Table 5.1 Financing loans schemes: functions and agents Financing functions Case study (1) (2) (3) (loans scheme) Capital provision Loans Risk-bearing subsidies (interest rates etc.) China State-owned commercial Government Banks (GSSLS banks subsidized scheme) China Commercial banks None Parental assets (collateral) (GCSLS commercial scheme) Hong Kong Government budget Government Government, (LSFS parents subsidized scheme) Hong Kong Government budget None None: risk factor is (NLS incorporated in interest rate non-subsidized scheme) Republic of Korea Commercial banks Government Government, (MOE) students/parents Republic of Korea Government Government Government, (GECP Government (previously, government parents Employees Scheme) employees pension scheme Philippines Government budget Government Government (Study Now Pay Later) (previously, government financing institutions) Philippines Government budget Government Government (Region 5) Philippines Government budget Government Government, (Centres of Excellence) co-signatory Thailand Government budget Government Government, parents Turning now to non-government loan capital provision, in the Chinese scheme run by the commercial banks 12 and in the Korean loans schemes run by the Ministry of Education (including the new loan and scholarship scheme for engineering students), initial loan capital is provided by 12. The scheme is directed, overall, by the Central Lending Group for Loans. 50

52 Assigning institutional roles: funding commercial banks. This obviates the need for initial funding by the government, though not for ongoing subsidy, as will be discussed. The considerable advantage of employing the commercial banking sector to provide the loan capital is evident. The central issue that still needs to be addressed is: What are those particular institutional and other conditions present in these two schemes that make it possible for banks to assume this role, which is shunned in most other schemes, both in the region and elsewhere? Another option is to use an existing fund or quasi-government institution to provide the loan capital. These arrangements have not always worked well. The Korean loans scheme for government employees and their families was financed by borrowings from the government employees pension fund. 13 Since the scheme is highly subsidized (it provides loans free of interest) and full recovery is not possible, it would seem that the overall effect of the scheme was to weaken the financial robustness of the government employees pension scheme. After temporary suspension in 1999, loans were resumed using the financial resources of government. In the early years of the Philippine Study Now, Pay Later scheme, an obligation to fund the scheme was imposed on various government financing institutions, including the Social Security System, the Government Service Insurance System and various public banks. The reluctance of these agencies to disburse funds earmarked for student loans, given the high levels of required subsidy and the consecutive agency losses, led to the disengagement of these agencies from the scheme, while the government took up the role of capital provider and operator of the scheme. 5.2 Who pays the interest subsidy, who bears the risk? Virtually all student loans schemes are subsidized, in the sense that borrowers do not repay the full value of their loans; the causes and effects of subsidizing student loans will be discussed in subsequent chapters. The repayment shortfall stems from a number of sources, particularly belowmarket interest rates and grace periods, repayment default and administration costs not passed onto borrowers. Who bears these costs? 13. Loan capital in other Korean loans schemes are also provided by existing financial funds: the Ministry of Labour scheme for industrial employees is financed from the Employment Insurance Fund; teachers and their children from the Korean teachers pension fund; and the Korean Labour Welfare Corporation for industrial accident victims from the Industrial Accident Compensation Insurance Fund. 51

53 Policy options for student loans schemes: lessons from five Asian case studies A central element of loan subsidy is the provision of student loans at rates of interest below those that would be payable on commercial loans. Subsidized rates of interest derive from the political judgment that a nonsubsidized interest rate would impose too heavy a repayment burden on student borrowers. The government subsidizes the interest rate in almost all student loans schemes; this is true also in the loans schemes examined in our case studies (Table 5.1, column 2). A notable exception is the supplementary Hong Kong Non-means-tested Loan Scheme (NLS), which operates on a full cost recovery basis, charging the government s no gain, no loss interest rate. 14 Where student loans are financed by commercial banks (and offered at below-market interest rates) as in China (GSSLS) and in the Republic of Korea (MOE scheme), the government finances the difference. In the Chinese scheme (GSSLS), students are charged half the market interest rate and the government transfers the shortfall (equivalent to half the market rate) to the participating banks. The commercial China scheme receives no subsidy. A major constraint on the willingness of the banking system to provide commercial loans to students is the inability of most students, particularly from a poor background, to supply loan collateral. Given the risk of repayment default, the government acts as guarantor in virtually all student loans schemes; an exception, again, is the Hong Kong NLS, which incorporates a risk premium into the interest rate to cover default (Table 5.1, column 3). Where commercial banks put up the loan capital, the government acts as direct guarantor. China is a notable exception. In the Chinese commercial scheme, parents mainly the better-off are required to mortgage assets as loan collateral (this effectively restricts these loans to middle and upper-class students); however, some government subsidy is available to the commercial banks in this scheme in the form of preferential tax treatment on student lending. In the Chinese subsidized scheme, universities in some provinces are now required to deposit from 10 to 20 per cent of tuition revenues into loan risk foundations, thus passing part of default risk onto higher education institutions. 14. This rate is now set at 2 per cent below the average best lending rate of note-issuing banks, plus a risk-adjustment factor (currently 1.5 per cent) to cover the loans agency s risk in disbursing unsecured loans. 52

54 Assigning institutional roles: funding In the Korean MOE scheme, banks require those students without creditworthy guarantors to take out loan insurance, amounting to 5 per cent of the loan amount (which is deducted from the loan size provided). As the economic climate began to deteriorate in recent years, the number of student-borrowers required to take out loan insurance increased, reaching some 40 per cent of borrowers in Where the government provides the loan capital, it acts indirectly as guarantor by absorbing repayment losses due to default. While it is true that many student loans schemes require parents (or other relatives) of the borrower to co-sign as loan guarantors, this is in practice largely cosmetic as they usually do not put up loan collateral. In many loans schemes (including the Thai scheme) the authorities are loath to turn to guarantors, often from low-income backgrounds, in the event of nonrepayment. In most loans schemes loan administration costs are not passed on to students. 15 While these costs are usually covered by the government, in many schemes substantial administration costs are absorbed by other participants in the loan process. In Thailand, administration costs are paid to individual educational institutions and to the public bank paying out loans and collecting repayments. However, these fees cover only a fraction of total administrative costs; the remaining costs are absorbed by these institutions. 15. Exceptionally, the Hong Kong NLS charges students an annual administrative fee to cover the full costs of processing and administering the loans. 53

55

56 Chapter 6 Assigning institutional roles: borrower selection and loan distribution In this chapter, we address an issue central to all student loans schemes: How are student loans distributed? What are the advantages and disadvantages of alternative institutional arrangements for borrower selection and loan distribution? We draw upon differing case study experience to derive lessons for best practice. 6.1 The loan distribution process Every loans scheme must develop a mechanism whereby funding available to finance student loans is distributed to student applicants. Figure 6.1 provides a generic chart describing the essential elements of this process. Students apply for a loan (light-shaded arrow) to the institution(s) responsible for distributing a loan budget amongst applicants; loan budgets are received from a funding source (dark arrow). For example, the funding source could be a central loans agency which, in turn, allocates loan budgets to educational institutions to which students apply for loans. In this case there will be a need to define both loan eligibility criteria for students and an objective for the allocation of loan budgets to institutions. In highly centralized systems, there are no intermediary institutions and loan applications are made directly to the loans scheme operator, obviating the need for intermediary institutions to distribute loans. 55

57 Policy options for student loans schemes: lessons from five Asian case studies Figure 6.1 The loan distribution process: simplified presentation Student loan applications Applications Loans distribution institutions Loan budget allocation Funding source 6.2 Alternative distribution pathways: case study experience The case studies displayed considerable variation in the process by which borrowers are selected and loans distributed; institutional roles differ from case to case. In Figure 6.2 we trace for each scheme the process of loan distribution in two directions. As in Figure 6.1, the downwardfacing light arrows show the path taken by student loan applications, until the point when a loan application is approved. The dark, upward-facing arrows show the institutional movement of loan budget allocations, until the point at which loan budgets are distributed amongst applicants. 16 The dark boxes indicate the two sources of loan capital in the schemes under discussion the commercial banks and the government which budgets the Philippines Commission on Higher Education (CHED) and the various central loan funds. An alternative funding source (not shown in the figure) is existing financial funds (such as pension funds); these provide the loan capital for some of the loans schemes in the Republic of Korea and for the Study Now, Pay Later scheme in the Philippines in its early years. In most cases, the educational institutions play an intermediary role in the process of loan distribution. 16. We are not concerned here with mechanical banking functions, such as expediting loan payments. 56

58 Assigning institutional roles: borrower selection and loan distribution Figure 6.2 Alternative scenarios for loan distribution Student loan applications China subsidized scheme Rep. of Korea MOE Philippines SNPL Region 5, COE Thailand MUA MOE Hong Kong Rep. of Korea government employees China commercial scheme Education institutions MOE Regional offices, CHED MUA Provincial offices, MOE Commercial banks (Not funded by government) MOE Commission on higher education (Government funded) Loan scheme funds (Government funded) The Hong Kong scheme is the most straightforward. Students submit loan applications directly to a central autonomous public loans organization the Student Financial Assistance Agency (SFAA); there are no intermediaries in the process. The Korean government employees loan programme, as three other loans schemes in the Republic of Korea, was funded from an existing financial fund (government employees pension fund), until financial difficulties led to the government taking up the role of loan capital provider for the scheme. The decentralized loans schemes in Thailand and in the Philippines stand in sharp contrast to the centralized Hong Kong arrangements. The Thai scheme is the most intricate. Students apply for a loan to the educational institution they attend whether universities under the Ministry of University Affairs (MUA) or secondary and tertiary institutions under the aegis of the Ministry of Education (MOE). These educational institutions decide who should receive a loan, the purpose for which the loan is given (tuition fees, accommodation, living expenses) and the loan amount, up to a ceiling set by the central Student Loans Scheme Committee (SLSC). The SLSC allocates the total annual loan budget provided by government 57

59 Policy options for student loans schemes: lessons from five Asian case studies between the two education ministries. The MUA provides an annual loan budget directly to each university, while the MOE allocates loan budgets to the 75 provincial offices, which in turn allocate loan budgets to the schools and other institutions under its control. The Philippines loans schemes operate through the Commission on Higher Education (CHED), which is financed by government subvention. In all three schemes, loan applications are initially vetted by the educational institution. In the case of the veteran Study Now, Pay Later scheme, applications are forwarded to CHED regional offices where a decision is made on the application. In the still largely experimental Region 5 and Centres of Excellence schemes, the universities distribute loans to students on the basis of the number of loan slots allocated by the regional offices of CHED. In the case of the Chinese subsidized scheme (GSSLS) and of the Korean MOE scheme, where commercial banks provide the loan capital, applications are initially received by universities. In the China scheme (GSSLS) the final decisions on loans are made by the lending banks. Finally, the new Chinese GCSLS, which operates as a fully commercial loans scheme, applicants for student loans deal directly with the banks. 6.3 Weighing the alternatives Given this diversity of practice, what are the advantages and shortcomings of each case? There are two main advantages to schemes in which an autonomous public lending body deals directly with student applications, as it is the case in Hong Kong. One advantage is the administrative simplicity of such schemes. The second advantage is that this arrangement fosters horizontal equity; loans are distributed across-the-board on the basis of objective, transparent criteria an important consideration in schemes that target the poor. In most schemes, the educational institutions themselves play a role in the loan distribution process. But the nature of this role may vary considerably from case to case. There are three major tasks that educational institutions may perform: 58

60 Assigning institutional roles: borrower selection and loan distribution Student loans offices within institutions may act as a post-office, in distributing and receiving application forms, vetting for eligibility and forwarding applications to a central loans agency, commercial bank or education ministry. This is broadly the case in a number of the schemes represented in Figure 4.2: the Philippines Study Now, Pay Later scheme (SNPL), the Chinese subsidized scheme (GSSLS) and the Korean MOE scheme. Student loans officers may adapt a more pro-active, targeting role: in actively seeking out and identifying students in need or at risk, and encouraging them to take advantage of available loan facilities. This function may be important in the case of loans schemes where targeting the poor is important. Student loans officers or councillors in upper secondary schools in Thailand, given the relatively small institution size and their familiarity with the students, are well placed to perform this task. As in the Thailand and Republic of Korea (MOE) loans schemes, educational institutions may be assigned a more direct distributive role, in distributing an assigned institutional loan budget amongst its applicants. The benefits of the post-office and targeting roles are manifest; these are notably lacking in direct, centralized schemes where the intermediary role of educational institutions is minimal. However, the distributive function is usually more problematic, particularly where the central objectives of the loans scheme are social (equity and access for the poor). The problem often stems from the use of inappropriate criteria for the top-down loan budget allocation, through the various administrative levels to the education institution. Given that the central objective in most of the schemes examined is to assist poorer students, each educational institution should receive a budgetary quota that reflects the number of poor students (or potential students). This would assure that all eligible students have equal chances of receiving a loan, regardless of institution. However, institutional budgetary allocation pays scant attention to this issue in the schemes under scrutiny. In the Thai scheme, the formula used by the Ministry of University Affairs to allocate loan budgets amongst the universities does not take into account the social profile of the student enrolment in each university; 17 thus relatively well-off students in some 17. Admittedly, such data are not available at present but they could be generated. 59

61 Policy options for student loans schemes: lessons from five Asian case studies universities can gain a loan that is denied to poorer students in other institutions. 18 Similarly, in the Korean MOE scheme, allocation of budget quotas to universities by the Ministry of Education is based on individual university enrolment. The problem is exasperated in schemes where institutional budgets are allocated by regional or provincial offices. In the Thai scheme, the MOE allocation to provincial offices employs a formula where the weight accorded to the relative poverty level in each province is only 20 per cent (subsequent allocation of funds to individual schools by the provincial office is arbitrary). In the Philippine schemes, budgetary allocation to regional offices done by the Commission on Higher Education, as reported in the case study, does take regional poverty level into account, but the logic of the formula used is not easy to grasp. The upshot of the foregoing discussion is that the particular administrative arrangements and institutional roles assigned in borrower selection and loan distribution are essential to the efficacy of a loans scheme in meeting declared objectives. 6.4 Adequacy of loan size We have noted that a successful loans scheme must incorporate appropriate selection mechanisms. It must also provide loans that are sufficient to meet the needs of the students at whom the scheme is directed; whether or not the students can afford these loans, in terms of the repayment burden, is a different issue which will be addressed in the next chapter. Of all the schemes examined, the Hong Kong case is the most satisfactory in terms of adequate level of student assistance. Students may receive an interest-free LSFS loan for living expenses, based on an evaluation of the family financial situation; this complements means-tested outright grants for academic expenses. In addition, all students, including the poor, may avail themselves of an interest-bearing NLS loan, allowing students to top up individual loans according to their need. Loan size (depending on family income) is adjusted periodically, via surveys of student expenses and the compilation of a Student Price Index, to ensure an adequate level of support. 18. A more severe equity problem in the Thai scheme is the considerable degree of autonomy given to individual educational institutions, not only in relation to selection but also concerning the size of individual loans; we return to this issue subsequently. 60

62 Assigning institutional roles: borrower selection and loan distribution The relatively generous Chinese commercial scheme offers maximum-size loans of up to some three times the maximum loan size in the more sparing Chinese subsidized scheme (GSSLS); recipients of GSSLS loans may find them insufficient even to cover tuition fees, given the variation in fees across institutions and study disciplines. The issue is more severe in the budgetary-deficient Philippine scheme; the loan amount often falls short of the tuition fee. In Thailand, falling budget allocations for new borrowers, combined with the considerable autonomy granted to education institutions in fixing the size of individual loans, results not only in considerable inequities across the system but also in a tendency (particularly in private institutions) for the size of actual loans to fall considerably short of the maximum loan size recommended by the Student Loans Scheme Committee. Yet even these SLSC loan size ceilings, evidently, are set arbitrarily and take no account of real costs or of available student expenditure surveys; moreover, there is no link between the amount of loans and rising educational costs (as there is in Hong Kong): as a consequence the real value of loans falls over time. In the Korean case, the problem is not severe because in many schemes the source of loan capital is a financial fund. These funds have sufficient financial resources to accommodate the demand for loans from client (member) groups. In most of the schemes examined in the case studies, the loan frequently falls short of student needs for education and living expenses; given the social orientation of most of the schemes, the implication of a small loan size is that higher education remains beyond the means of the very poor, thus largely defeating the purpose of the scheme. Since initial loan capital cannot be augmented in most of the cases, the next best alternative is to target poor students more effectively and ensure that they receive adequate support, rather than spreading the loan budget more thinly amongst a larger number of borrowers, which includes those in lesser need. 61

63

64 Chapter 7 Assigning institutional roles: loan repayment collection No student loans scheme can operate successfully unless an effective loan repayment collection mechanism is in place. In this section, we discuss alternative collection methods, examine collection methods in use in the case study schemes and consider what steps can be taken to minimize repayment default. 7.1 Institutional roles in collection We may define two major types of collection mechanisms: selfcollection and agency collection. Self-collection refers to situations where the organization operating the loans scheme also takes responsibility for the mechanics of repayment collection. With agency collection, the task of collection and follow-up is outsourced to a specialist agency. In addition to banks, other possible collection agency options are available; these include national institutions concerned with collection, such as the income tax administration and social security. The possible use of the income tax authorities as a collection agent in the present context should not be confused with the central role played by the tax system in income contingent loans schemes. Under income contingent loans, which have been introduced in a small number of industrialized countries, periodic loan repayment obligations are determined as a proportion of the graduate s income in each period. The participation of the tax authorities in collection is necessary in this case because information on individual income is required. Thus while income contingent loans imply the use of the tax system, the opposite is not true. If the tax authorities proved to be an effective collection agent in the context of any given country, then they could be thus used for conventional mortgage-type loans schemes (where repayments may not be income- 63

65 Policy options for student loans schemes: lessons from five Asian case studies contingent). The tax system could be effective where PAYE systems are widespread and effective. Likewise, employers who deduct social security payments from employees wages on behalf of the social security (national insurance) administration, could be used to make wage deductions for student loan repayments. However, these mechanisms are insufficiently well developed for this purpose in most developing countries. 7.2 Case study practice Figure 7.1 illustrates five alternative scenarios for loan repayment collection, based on case study practice. Self-collection Most collection arrangements in the schemes under study are of the self-collection type. Not surprisingly, banking institutions are responsible for collection in a number of cases. For the three schemes where commercial banks provide the loan capital (the Chinese schemes and the Korean MOE scheme), these banks are also responsible for loan repayment collection. In the case of Hong Kong and the Republic of Korea (Government Employees Scheme), borrowers repay directly to the central lending agency. In the Philippines, repayments are made to the decentralized institution which makes the loan: the regional offices of the Commission on Higher Education (CHED) for the Study Now, Pay Later scheme, and the relevant universities in the case of the Region 5 and the Centres of Excellence schemes. In the case of the schemes where collection is executed by commercial banks (which also supply the loan capital) such arrangements may be expected to be effective, given the expertise of the banking system in these activities. However, it is not sufficient that banks have a comparative advantage in collection; they must also have an incentive to collect. Where governments act as guarantor against repayment default, the incentive for banks to collect may be weakened, particularly in the case of individuals where collection costs are high. A fine balance must then be struck between the dual needs of providing incentives for rigorous collection and of offering last-resort government guarantee against default. The efficacy of other forms of self-collection is less clear. The failure of self-collection in the Philippine Study Now, Pay Later scheme, where 64

66 Assigning institutional roles: loan repayment collection repayment is virtually non-existent, is particularly notable. The Office of Student Services within CHED, which is responsible for administering a wide variety of grant and loan programmes, has neither the staff nor the capacity to effect repayment collection; more importantly, there is little incentive to do so, as annual subventions for loans and grants are protected by legislation. In parallel, low repayment rates also reflect the attitude of students and their parents, who regard SNPL loans as outright grants, which indeed they are, in practice. CHED regional offices seem to do little to correct these attitudes. The two new, essentially pilot loans schemes (Region 5 and COE) may achieve better results, as they are institutionbased with responsibility for repayment collection, as well as banking functions and fund management, delegated to the participating institutions. However, their success seems unlikely. There are no indications that the institutions have the capacity or the know-how to perform these functions well. Figure 7.1 Alternative scenarios for loan repayment collection Loan Repayments Loan Student repayments Loan Repayments Rep. of Korea MOE China Subsidized scheme China Commercial scheme Universities Philippines Region 5, COE Philippines SNPL Regional offices, CHED Thailand Public bank Hong Kong Rep. of Korea government employees Commercial banks Commission on higher education (CHED) Central Loans Fund 65

67 Policy options for student loans schemes: lessons from five Asian case studies The two centralized loans schemes (Hong Kong and Republic of Korea, Government Employees Scheme), which both employ forms of self-collection, are more successful. In Hong Kong, the Student Financial Assistance Agency (SFAA) issues quarterly demand notices for repayment; payment to SFAA is effected by post draft or direct bank debit. The reported default rate is slightly over 1 per cent, a very low rate by international standards. There are probably several reasons for this success, which, however, are not related to the selfcollection repayment mechanism. The small geographical area of Hong Kong is one factor: While the number of students served is large, they can be located readily and repayment notice can be served quickly. The repayment burden, in relation to income, is not high (this general issue is further discussed in the next chapter) and the healthy economy has generated good employment conditions easing the repayment process. But most important perhaps is the ethos of Hong Kong society, including a positive attitude towards debt obligations. Agency collection The only example of agency collection in the case studies is that of Thailand. The public bank that administers loan payments is also responsible for collection and the follow-up of defaulters. Loan budgets are passed from the central Student Loan Scheme Committee, via the education ministries, to the educational institutions, which allocate individual loans; however, none of these institutions engage in repayment collection. This task, as well as regular banking functions relating to the loans scheme (including initial loan payment), is taken on by the public Krung Thai Bank (KTB). KTB responsibilities include maintaining individual borrower accounts, dispatching repayment notices when due and chasing borrowers who fail to repay. An agreed schedule of fees has been established for these services. Though Hong Kong does have a well oiled and straightforward personal income tax system, it does not provide a service for collecting student loan repayments; there are no pressures to change the existing self-collection system which appears to work well. However, current recommendations to introduce a new income-contingent loans scheme in Thailand include the assignment of repayment collection responsibilities to the Revenue Department (Ministry of Finance); repayments due would be deducted automatically by employers from the wages of employees/ borrowers. 66

68 Assigning institutional roles: loan repayment collection Conclusions The foregoing discussion may be summarized in this way. Choosing an appropriate collection institution is central to effective loan recovery. Self-collection by bodies such as autonomous public student loans institutions, government ministries or universities might work well, as in Hong Kong. However, while such institutions may possess comparative advantages in selecting students and targeting, their capacity for effective repayment collection is less evident. Banks often have the necessary infrastructure and expertise that such bodies may lack. Strong repayment collection institutions are not enough to ensure successful repayment collection. The incentive for those institutions to collect pro-actively can be undermined by public policy. A full government guarantee on loans may encourage private or public banks to collect from the government rather than the debtor. Thus a government guarantee below 100 per cent is advisable, with commercial banks taking on a small part of the risk (as in the Korean MOE programme). Similarly, overreliance by loan organizations on government budgetary support may undermine the incentive to follow up on delinquent borrowers (as in the Philippine SNPL scheme); loan bodies may tend to rely solely on public funds to supply new loans rather than stepping up efforts to secure repayment to finance loans for new borrowers. 7.3 Minimizing default Successful repayment collection depends not only on the efficiency of the collecting institution: it also is a function of borrower attitudes and behaviour. Borrowers failure to repay loans may have two main causes, each requiring a different treatment. There are borrowers who cannot repay and there are borrowers who do not want to repay (they evade repayment). 19 Lightening the repayment burden Let us first consider the case of students who encounter repayment difficulties. Properly designed repayment plans may help those students who do not earn large salaries after graduating. Obviously, the simplest 19. This section draws on a parallel discussion presented in Ziderman and Albrecht (1995). 67

69 Policy options for student loans schemes: lessons from five Asian case studies way of reducing the burden of repayment is to provide additional loan subsidies (lower interest rates and extended grace periods). This, however, would be self-defeating, as the benefits of reduced default would be offset by lower repayment obligations per student. Again, screening out higherrisk borrowers may reduce default but will tend to exclude poorer students who are the target of most of the schemes discussed. More positively, temporary repayment remission for borrowers with low income is a key measure to avoid classifying students with repayment difficulties as being in default: When a graduate s income falls below a threshold level, the borrower is exempt from repayment, while still accruing interest charges. Similarly, an increasing repayment schedule (albeit based on the mortgage loan principle) may help to lessen the extra burden on young graduates during the early repayment years (as in the Thai scheme). In some cases, the amortization period may need to be increased. In China (GSSLS), the short repayment period of only four years imposes a particularly heavy repayment burden on the new graduate, with repayments absorbing a quarter of annual graduate income (this is discussed in greater detail in the following chapter). Penalizing repayment evasion Let us now turn to evasion. Table 7.1 provides a list of measures taken in the case study schemes to lower repayment evasion. The most widespread measure, simulating commercial loans, is to require the cosignature of a guarantor (usually a family member) who undertakes to repay the loan in the event of default. Requiring a guarantor can have negative consequences that may defeat the purpose of loans schemes with social objectives: The very individuals who are most in need of support may be the least able to provide guarantors. Perhaps for this reason, little more than lip service is accorded to guarantor provisions in many schemes (Thailand and Philippines (SNPL)); in practice, the task of a guarantor may vary from little more than that of moral counsellor to that of a provider of financial collateral (China, commercial scheme). Further measures include barring access to further credit among defaulters and moral pressure, exerted through the publication of the names of inveterate defaulters. 68

70 Assigning institutional roles: loan repayment collection Table 7.1 Measures/sanctions against repayment default Case study Sanction Comments (loans scheme) China Students individual credit It is unfair to penalize new students (GSSLS subsidized Moral pressure: publication of because previous loan recipients scheme) defaulter list of students and have defaulted institutions Denial of loans to new students enrolled at universities with high default rates China Family co-signatures and pledged Effective, but restricts loans to (GCSLS assets as collateral better-off students commercial scheme) Hong Kong Loan guarantor Serious and prolonged defaulters are referred to the Department of Justice for legal recovery Republic of Korea (MOE) Republic of Korea (GECP Government Employees Scheme) Joint surety (parents) Loan insurance Bar access to further credit if in default Joint surety (parents) Moral pressure: publication of defaulter list Bar access to further credit if in default Philippines Loan guarantors Ineffective: guarantors not required (Study Now Pay Later) to pay-off bad loans Virtually no loans repayment Philippines (Region 5) None Philippines Loan guarantor (Centres of Excellence) Thailand Loan guarantor SLSC reluctant to turn to Legal action against recalcitrant guarantors defaulters Currently, SLSC is considering use of credit rating to discourage default Finally, measures to inculcate a more positive attitude towards repayment could be developed. In some cases, this may be tantamount to changing social norms and therefore not feasible in the short run. But 69

71 Policy options for student loans schemes: lessons from five Asian case studies universities could play an important role in emphasizing the necessity for repayment. Indeed, this is likely to occur increasingly in the Chinese subsidized scheme; new loans will not be available to new students enrolled in institutions where the defaulted amount exceeds 20 per cent and there are 20 or more defaulters. It might be useful for the loans organization to maintain contact with students during the borrowing period to remind them of their repayment obligations. This approach could be extended, so that students do not forget their repayment obligations during the lengthy period before repayment starts. Thus students could be required to begin repayment (albeit nominal amounts only), while studying and during grace periods, to develop early on the concept of repayment as an obligation, rather than an activity to be avoided. These early payments would be kept small. 70

72 Chapter 8 Financial viability of loans schemes 8.1 Factors leading to low loan recovery The financial viability of any loan programme depends on the extent to which loan outlays are recovered by the lending body. A number of factors hinder full recovery of loans: these may be divided into two groups (Table 8.1). First, there are factors that are built-in to the scheme, as elements of its design. Lending conditions on virtually all government-sponsored loans schemes are softer than those on regular commercial loans schemes; this difference represents a subsidy received by the student, in the sense that the borrower is not required to pay back the full value of the loan received. As shown in the table, these conditions include below-market interest rates on the loan, periods in which no interest is levied on outstanding debt (both during study and in grace periods after study completion) and repayments that are not linked to the rate of inflation. The effect of these built-in subsidies is amplified where amortization periods are long. The larger these built-in subsidies are, the less of the loan the borrower is required to repay; the difference between the original loan and the actual required repayment represents, effectively, a hidden grant to the student taking out a loan. Table 8.1 Factors leading to less-than-full loan recovery Built-in design factors Below-market interest rates Interest-free study and grace periods Repayment in nominal terms Long amortization period Administrative factors Payment in arrears Non-repayment (evasion) Forgiveness Administration costs 71

73 Policy options for student loans schemes: lessons from five Asian case studies Even if student loans were not subsidized and the individual student was required to repay the loan in full, not all of the sums loaned would be recouped by the loan authorities. The extent of such a shortfall would be dependent on the level of administrative efficiency with which the loans scheme is run. Thus, overall loan recovery depends not only on cash repayments but also on administrative costs that are not passed on to borrowers and on the level of repayment default. Repayment default is broadly defined to include payment in arrears, repayment evasion and the cancellation of individual repayment obligations ( forgiveness ) for such reasons as physical disability and the encouragement of graduates to enter shortage occupations. Against this background we discuss in the next two sections the issue of loan repayment in relation to original loan size, respectively from the contrasting perspectives of the individual borrower and of the loans scheme as a whole. 8.2 The individual loan account Repayment ratio We have noted that, in most loans schemes, the individual borrower is not required to repay the full value of the loan received, because of the advantageous loan repayment conditions which are not present in commercial schemes. These built-in factors (as listed in Table 8.1) result in subsidized loans; the subsidy represents a hidden grant to borrowers. The loan repayment ratio, defined as the ratio of required repayments to the size of the loan received, 20 measures how much of a loan a borrower is required to repay. Using a common methodology, repayment ratios were computed for a number of case study loans schemes. Results are presented in Table 8.2, column Both measured in terms of present values, i.e. discounted to a common time point. 72

74 Financial viability of loans schemes Table 8.2 Case studies: repayment and recovery ratios REPAYMENT RECOVERY Repayment burden (%) Case study (1) (2) (3) (4) (loans scheme) Repayment Related to Related to Recovery ratio, ratio (%) annual annual with default and income incremental administration income costs (%) China (GSSLS * 53 subsidized scheme) Hong Kong 100 Initially 11, Initially 47, n.a. (?100) (LSFS subsidized falling to 5 falling to 13 scheme) Rep. of Korea (MOE) n.a. 64 Rep. of Korea (GECP Government Employees Scheme) n.a. 40 Philippines (Study Now Pay Later) n.a. n.a. n.a. 0 Thailand 2 1 Males: 2.3 Males: (1st degree programs) Females: 3.5 Females: 6.1 n.a.: not available. * This new estimate corrects an erroneous figure shown in the case study report. While repayment ratios were not computed in the Hong Kong study, the repayment ratio for the non-subsidized NLS scheme ( no gain, no loss ) is taken to be 100 per cent the borrower repays the loan in full. In both China (GSSLS) and the Republic of Korea (MOE) borrowers repay about 80 per cent of the loan, a fairly high rate by international standards; the interest charged in both schemes is about half of the market rate. 21 In the Korean scheme for government employees (GECP), the effect of a zero interest charge is to lower the repayment ratio to 55 per cent. In the Thai scheme, the repayment ratio on loans for first-degree programmes is little over 20 per cent, 22 indicating a massive loan subsidy (i.e. most of the loan represents, in fact, a hidden grant to students). A number of 21. Lending conditions in the case study schemes indicating the extent of built-in subsidies are presented in the Appendix, in Table A The Thailand study reported similarly low repayment ratios for secondary schooling loans. 73

75 Policy options for student loans schemes: lessons from five Asian case studies converging factors account for the low repayment ratio in Thailand: the detrimental effect on the repayment ratio of the one per cent interest rate charged and of repayments not linked to inflation is incremented by the long, post-study, repayment horizon of 14 years. Repayment burden Soft loan repayment conditions are sometimes justified by the need to avoid imposing unduly harsh repayment burdens on students, especially the poor. But the price of minimizing the disincentive to take up loans is a heavy government subsidy. Excessive subsidies tend to compromise the very purpose of a loan and, in effect, convert loans into hidden grants. In a number of the case study reports, estimates were made of the extent to which the repayment of a loan does in fact constitute a burden. The repayment burden is measured in two ways. First, annual loan repayment is expressed as a percentage of annual income, providing an indication of the repayment burden falling on the borrower each year. The second measure calculates the ratio of annual repayment to annual incremental income (the additional income received as a result of completing the course of education) and measures how much of the annual incremental income resulting from additional education remains after annual repayment is made. It should be noted that the repayment burden is a function of both the loan size and the degree of loan subsidy. Results are given in Table 8.2, columns 2 and 3. Looking first at column 2, the repayment burden (based on annual income) at about 10 per cent or less is not excessive in Hong Kong (NLS) or the Republic of Korea (MOE). Surprisingly, the repayment burden in the interest-free Korean GEPC scheme is greater than that in the Korean MOE scheme; this is because loans in the GEPC scheme are typically bigger. The heavy repayment burden in the China GSSLS (about a quarter of annual salaries are absorbed in loan repayment) reflects both the low level of graduate salaries and the short repayment deadline of four years following graduation. This suggests that a lengthening of the repayment period would be advantageous. Repayment in the heavily subsidized Thai loans scheme is not burdensome at all; only a little over 2 per cent for men and 3.5 per cent for women, whose wages are lower. 74

76 Financial viability of loans schemes The general picture is similar for the repayment burden relatively to incremental income (Column 3). In the case of China (GSSLS), annual repayments are equal to over twice the increase in annual wages resulting from additional education. In Hong Kong (NLS) nearly half of the return on investment in university education is lost in repayment in the initial year or two following study, but this subsequently falls to a manageable 11 per cent. For Thailand, only 3.5 per cent (for males) and 6.5 per cent (for females) of incremental income is absorbed by the repayment of loans.. The loans scheme in Thailand is seen to be excessively generous, providing borrowers with unnecessarily highly subsidized loans, which may be effortlessly repaid out of higher income received on completing courses of schooling. Far stricter repayment conditions could be introduced, with a few disincentive effects on students willingness to borrow and a healthy reduction in government budgetary support over the longer term. 8.3 Loan recovery: the overall perspective The repayment ratio, related to the established repayment conditions of the loans scheme, relates to the typical borrower; it fails to show the extent of overall recovery to the loan fund, from an overall viewpoint. Loan recovery focuses more on the scheme as a whole, not on the individual borrower. It is concerned with the issue of how much of the total outlays of the loans scheme (loan disbursement plus all other costs including administration) will be recovered. It takes into account all of the items listed in Table 8.1, both the fixed, built-in design factors as well as the effects of administrative efficiency in running the scheme. Thus if a borrower defaulted, total repayment receipts would fall; general administration costs are included in total outlays of the loans scheme. The recovery ratio is measured by the ratio of total (discounted) repayments to total (discounted) outlays; results are given in Table 8.2, column 4. As expected, the recovery ratios reported are lower than repayment ratios, because the latter take no account of the probability of repayment default and do not include any apportionment of administrative costs. The recovery ratios shown for China and Korea, while not untypical of recovery ratios for other countries (Ziderman and Albrecht, 1995, Table 4.2), should signal a warning against complacency. The poor showing of Thailand must call into question the future of the scheme as now constituted; however, a reform of the scheme (both in terms of its loan conditions and its management) is now on the public agenda in Thailand. 75

77 Policy options for student loans schemes: lessons from five Asian case studies 8.4 Loans scheme viability There are a number of lessons that can be derived from the analysis provided in this chapter. First, the widely held view that loans schemes can act as revolving funds which, once capitalized, can finance themselves through repayments from earlier loans is seen to be a myth. Given less-than-full loan recovery in almost all schemes and, in many cases, very heavy losses, government subsidy will remain a continuing feature of loans schemes. Even where the higher education system is not expanding, governments will need to inject funds annually to make up for losses from nonrepayment. Second, the financial viability of loans schemes can be improved, with lowered administrative costs and, particularly, reduced loss due to default; but this will require the authorities to adopt an open-minded and eclectic stance with regard to the assignment of institutional roles in running the scheme. Third, the justification of heavy government subsidies provided through large hidden grants the major source of recovery losses needs to be scrutinized case-by-case. While some element of subsidy can be justified in most cases, the question is whether or not the amount of government subsidy provided is excessive. This is the central concern of the chapter that follows. 76

78 Chapter 9 Equity and assistance to the poor: the role of subsidies 9.1 When are loan subsidies justified? In most countries student loans schemes are highly subsidized. Yet in contrast, when we examine particular loans schemes in terms of their major objectives, in most cases little justification is found for heavy subsidies. In Table 9.1, for each of the eleven major loans scheme objectives (as outlined in Chapter 3), the question is asked: Is a loans scheme subsidy justified? In most instances, there is no compelling case for loan subsidy and, in principle, the aim should be full loan recovery. Only for loans schemes aimed at directly helping poor students is there a clear case for subsidy. For two of the objectives, the case for loan subsidy is ambiguous: student loans that aim at facilitating the growth of the private university sector and those intended to ease manpower shortages. In principle, loans for students enrolled in private universities, aimed at facilitating the growth of the private university sector (and, therefore, the expansion of the university system), should not be subsidized. But in many countries (including Thailand), high tuition fee levels at private institutions may imply an overly heavy repayment burden in the case of full-cost, unsubsidized loans (particularly for less well-off students who are unable to gain entry to the prestigious public universities). In this case, the subsidization of student loans in the private sector may remain a more cost-effective method of generating university expansion (via private university growth) than expanding public sector provision. However, this is true only if the loan subsidy is not excessive. Student loans may be used as a device to encourage enrolment in courses leading to employment in needed occupations. In cases where there is no general student loans scheme in place, the availability of non- 77

79 Policy options for student loans schemes: lessons from five Asian case studies subsidized loans (in combination with other incentives) may be sufficient to attract students to these courses of study. But where loans are available to students in other courses of study, subsidization of loans may be necessary for these specialized fields of study, either in terms of softer repayment conditions or even loan forgiveness, with the loan being converted to a grant. A similar policy may be adopted, ex post, to encourage specialized workers (teachers, doctors) to accept employment in peripheral locations. Table 9.1 Justification for subsidized loans Loans scheme objective (1) (2) Loan subsidy Effective loan justified? recovery expected? Budgetary objectives (income generation from tuition fees) Revenue generation: rising public university No Yes unit costs Funding replacement: reduction in government No Yes expenditures Funding replacement: reallocation of public No Yes educational budgets University system expansion Revenue generation to fund public university No Yes expansion Growth of private university sector No(?) Yes(?) Social objectives (equity/access for the poor) Loans targeted on needy students Yes No Cross-subsidization No Yes Manpower needs Meet specific manpower needs Yes(?) No(?) Student assistance Ease student financial difficulties No Yes Increase student commitment No Yes Financial independence No Yes Given the justification for subsidized loans, particularly in loans schemes with social objectives, two issues arise. The first point relates to the need for targeting. A subsidized loan (incorporating a hidden grant 78

80 Equity and assistance to the poor: the role of subsidies element) lies somewhere between an outright grant and a non-subsidized loan. This gives rise to the central policy issue: Given the higher administrative costs of loans compared with grants, particularly collection costs, and the possibilities of repayment default, at what level of subsidy does a grant become a more cost-effective instrument for assisting the poor than a subsidized loan? In the case of the loans scheme in Thailand, the low recovery rates might indicate the need to substitute carefully targeted grants for loans, particularly at the upper secondary education level. This relates to the second point. A system of highly subsidized loans, aimed at needy students, can be justified only if these loans do, indeed, serve the needs of the poor only, and not of the better-off students. But many such schemes, including some in our case studies, are not well targeted to reach the poorer students. Targeting thus becomes a critical issue in appraising the efficacy of loans schemes that are directed at disadvantaged students. 9.2 Targeting Targeting as a rationing device A distinction needs to be drawn between targeting as a rationing mechanism given limited loans funds, and between deliberate targeting of the poor in pursuance of the objective of reaching the needy student. In some countries, student loans (set up for other than social purposes) are available to all students, regardless of need or ability. But open access to loans can be prohibitively expensive to governments, particularly if governments provide the loan capital and/or the scheme is highly subsidized. Growing student numbers and less-than-full loan recovery will result in increasing, and unsustainable, pressures on limited loans funds. This indicates the need for some form of loans rationing. If rationing is not to be arbitrary (on the basis of who applies first) a policy of deliberate targeting should be adopted. Rationing via targeting could be based on various rationing devices; we consider three possibilities here. First, rationing could be based on targeting poor students those deemed most deserving of support. But, in this case, reaching the poor student as such is not the central objective of the scheme; it is simply a 79

81 Policy options for student loans schemes: lessons from five Asian case studies rationing device that is judged to be appropriate. But in terms of the financial efficacy of the loans scheme this may not be the preferred approach as poorer students may be regarded as higher risk borrowers, with a greater propensity of default on repayment obligations. Thus other rationing devices might be preferred. In terms of enhancing the internal efficiency of the loans scheme, it would be better to target students of greater academic ability. These students are less likely to repeat or drop out during study and, securing better jobs, are less likely to default on loans repayment. Indeed, with the aim of limiting the extent of loans repayment default, targeting might be aimed at excluding various categories of high-risk students, including the poor. A third rationing device is the restriction of loans (or giving priority) to students enrolling in course of study leading to occupations in short supply or to students willing to work in less attractive regions. Loans rationing based on such manpower considerations are likely to lead to greater external efficiencies of the scheme, as the economic payoff of the scheme is likely to be enhanced as manpower shortage bottlenecks are loosened. Reaching the poor The main consideration in loans schemes where the central objective is to provide loans to students from poor families, is whether the scheme does indeed reach this target group. Targeting the poor requires not only the setting of definitions for loan eligibility (usually in terms of family income ceiling) but also the ability to check and test the veracity of such information supplied by applicants. Where it is difficult to authenticate this information, the scheme is subject to abuse, with non-poor students gaining access to subsidized loans. This is doubly unfortunate in terms of the scheme s objectives. It is both wasteful and inequitable to provide students who are not in poverty, with these unneeded financial benefits. And without careful targeting, it may well be that it is the student most in need who will be excluded. The concept of targeting and reach may be clarified by reference to Figure 9.1: the egg diagram. The outside (largest) egg relates to the total population of university age and includes both youngsters who have left the school system and those currently attending tertiary education (the middle egg). Within the schooling population, some students are in 80

82 Equity and assistance to the poor: the role of subsidies receipt of loans (the smallest egg). On the basis of the particular poverty measure adopted (the dotted curve), only part of this total youth population is defined as poor located in the North-East area of the chart. Total loan recipients are shown in Areas A and D of the chart. Those in Area A are poor and thus entitled to a loan, while D are not poor but are, nevertheless, in receipt of loans. This occurs because loan distribution does not conform to formal eligibility criteria or the loans scheme is being abused. The targeted school population consists of all poor youngsters enrolled in tertiary education (Areas A and B); of these only A receive loans and B are not reached. The total target population consists of A and B plus a portion of C (the poorer youngsters that have left school but are qualified for university entrance and could be drawn back into the education system through the availability of loans). Monitoring, on a continuing basis, will show how successfully the targeted population is being reached. Thus in considering the success of any loans scheme aimed at assisting the poor, attention must be paid to the reach of the scheme. Does the scheme indeed reach the target group? Which deserving groups are not reached? Are some ineligible students in receipt of loans? Figure 9.1 The egg diagram : reaching the poor Income ceiling Nonpoor Poor Non-student population C B Students non-recipients A Students loan recipients D 81

83 Policy options for student loans schemes: lessons from five Asian case studies Targeting versus screening This leads to the important distinction between targeting and screening. A loans scheme that is based on screening (rather than real targeting) focuses on the vetting of applications for their eligibility for inclusion, in terms of the defined entry criteria. It is less concerned with the actual composition of those accepted for a loan (and whether they constitute the group most in need of a loan), and with the self-excluded but eligible individuals who do not apply. It is essentially passive in approach, emphasizing legal entitlement rather than attempting to satisfy aneed. In contrast, a well targeted scheme actively focuses on those most in need; it aims at reaching out to a target population defined in terms of those most deserving of help, including such less readily reached populations as student dropouts in need and the rural poor. 9.3 Case study experience with targeting In terms of the foregoing discussion, how well have the case-study loans schemes performed? Summary information is provided in Table 9.2. Leaving aside the Hong Kong (NLS) and the Republic of Korea (GECP) schemes which are not aimed at needy students (Column 1), experience in the other schemes (all aimed at the poor student) is mixed. The case study reports on the Republic of Korea (MOE) and Thailand schemes both indicate leakage of loans to non-poor students; this seems to be the case in the China scheme too; the Philippines and the Hong Kong (LSFS) schemes are more successful in directing loans solely to poor students only. Since coverage of the Philippines schemes is too marginal to have any significant impact on the poor, the Hong Kong (LSFS) scheme emerges as the most successful in meeting its objective of reaching poor students: There is little leakage of loans to non-poor students and the scheme reaches a substantial percentage of poor students. 82

84 Equity and assistance to the poor: the role of subsidies Table 9.2 Targeting and reach: case study loans schemes Case study (1) (2) (3) (loans scheme) Loans scheme Loans Reach of aimed at poor reach only loans scheme students? poor students? (% of poor students) China Yes Yes? Low (GSSLS subsidized scheme) Hong Kong Yes Yes Substantial (LSFS subsidized scheme) Hong Kong Not targeted (NLS non-subsidized scheme) Republic of Korea (MOE) Yes No Low Republic of Korea Not targeted (GECP Government Employees Scheme) Philippines Yes Yes Marginal (all schemes) Thailand Yes No Substantial 9.4 Defining eligibility A loans scheme targeted on needy students must define the eligibility criteria for receiving a loan. Most student loans schemes aimed at the poor do so in terms of a maximal family income ceiling; this is also the practice in the case study loans schemes aimed at the poorer students. The central policy question is: What level of family income should constitute the eligibility ceiling? If this ceiling is set too high, with a consequent enlargement in the number of eligible students, there is a danger that the less disadvantaged amongst eligible students, may be those who receive loans rather than those most in need. This is likely to arise in those schemes where the number of eligible, and potential, borrowers greatly exceeds the supply of available loan slots. Thus many schemes define the eligibility ceiling in terms of the official poverty line, to ensure that only those most in need are eligible to take up loans. 83

85 Policy options for student loans schemes: lessons from five Asian case studies In the case studies, the Chinese subsidized scheme and the Philippines schemes confine eligibility to students from families in poverty, as defined by the official poverty line (Table 9.3, column 2). 23 However, in both the Korean (MOE) and Thai schemes, both formally aimed at poor students, the designated eligibility ceiling is an income level above of the official poverty line; in Thailand, the 150,000 Baht income ceiling defining eligibility (formerly 300,000 Baht) is three times above the official poverty line for a family with three children. The result of this policy is that many students, not drawn from the ranks of the very poor, are eligible to receive a loan. This issue is further complicated by the difficulty of checking the veracity of the information on family income that is provided by loan applicants; this is a central problem in places where an effective system of income taxation is lacking, or where the rural and informal sectors are relatively large. As indicated in Table 9.3 (column 3), the case study schemes employ alternative methods to check the validity of declared family income. In some cases, the educational institution certifies that the declared income is correct, on the basis of institutional familiarity with the socio-economic background of the applicant; more usually, a civil servant provides certification. In Thailand the village head may provide this certification; however, if the loyalty of the village head is stronger towards the student and his family than to the loans scheme, this certification may constitute little more than a rubber stamp. More generally, the use of family income as the sole criterion for eligibility, whether or not set at the official poverty level, is itself questionable. The family income measure fails to take into account a number of factors such as the number of children or elderly in the family, the number of other students in the family, whether the head of household is female and the amount of private property owned. Broader and more sophisticated measures for defining students in need, probably based on an index related to the foregoing factors, will need to be developed in due course. 23. In the Chinese case, the poverty line is not mentioned in official government eligibility criteria; however, in practice institutions and banks define as priority eligible borrowers those students whose family income falls below the local poverty line. 84

86 Equity and assistance to the poor: the role of subsidies Table 9.3 Setting the income ceiling for loan eligibility Case study (1) (2) (3) (loans scheme) Eligibility How is the Who provides eligibility certification income of declared ceiling fixed? family income? China Poor students Local government Local authority (GSSLS subsidized poverty line scheme) Hong Kong Poor students Amount of loan Applicant provides (LSFS subsidized offered is in inverse documentation scheme) proportion to family such as salary income and assets statement, profit-and-loss account, etc. Hong Kong All students n.a. n.a. (NLS non-subsidized scheme) Republic of Korea Poor students Ministry of Educational (MOE) Education (eligibility institutions; ceiling is above banks poverty line) Republic of Korea Government n.a. n.a. (GECP Government employees and Employees Scheme) their children Philippines Poor students Official poverty line Parents employer (all schemes) Thailand Poor students Loans Scheme Government official Office (eligibility (level 4 and above); ceiling is above village head. poverty line) n.a.: not applicable. The problem of defining an appropriate eligibility criterion may again be illustrated by an egg diagram (Figure 9.2). It is intention of policy makers to provide loans to the very poor students and potential students (groups A1, B1 and C1 the target group). However, difficulties in defining this target group (the dashed curve) and the adoption, in practice, of a more generous eligibility criterion (the dotted curve) usually based on 85

87 Policy options for student loans schemes: lessons from five Asian case studies family income but set above the poverty line means that eligible, but less deserving, students are in receipt of a loan (A2). The problem is not only that non-eligible students succeed in receiving a loan (D) but, more importantly, that non-targeted but eligible students (A2) may crowd out the target groups. A1 may be relatively small in practice and B1 and C1 are excluded; consequently relatively few of the targeted population are reached. Figure 9.2 Defining eligibility Income screen Target group Noneligible Non-student population Eligible C1 B2 Students nonrecipients B1 A2 D A1 Students loan recipients 9.5 Reach of loans schemes: case study experience Let us turn to three schemes, all aimed at serving the poor, where both targeting and the reach of the scheme are below par. China (GSSLS) 24 It is estimated that 19 per cent of Chinese students may be classified as poor (and 8 per cent as very poor). The question is whether those most in need are able to receive a loan. By the end of 2002, some 526,000 students had received a loan; this is nearly 40 per cent of all loan 24. These figures, taken from Shen (2004), update the data provided in the case study report by Shen and Li (2003). 86

88 Equity and assistance to the poor: the role of subsidies applicants and 5.5 per cent of total student enrolment. However, even if only poor students were in receipt of GSSLS loans as intended, this would imply, at best, that less than 29 per cent of poor students received a loan. Republic of Korea (MOE) Since the major purpose of this loans scheme is to aid poorer students, it is intended that priority should be given to the children of the unemployed and low-income citizens. In practice, targeting is blunted by the malfunctioning of the selection process; any student can apply for a loan and the selection (the issuing of letters of recommendation to the participating commercial banks) is usually made on a first-come, firstserved basis. Thus, only some 13 per cent of students from the lowest family income group receive a loan (7 per cent of all loans) and 15 per cent from the second lowest tier. On the other hand, some 9 per cent of students from families in the highest-income group receive a loan (4.3 per cent of all loans). While 35 per cent of all students receive a scholarship (42 per cent for public institutions), the scholarship system focuses on the student s academic achievement rather than financial status. Thailand The loans scheme s ground rules require that priority should be given to poorer students. This is often not the case in practice. Since loans are distributed by educational institutions whose loan budgets are not fixed in proportion to the number of poor students enrolled, there are considerable differences in the cut-off point for family income across institutions. In the case of upper secondary students, 452,000 loans were allocated in 1999, representing less than a third of eligible students (with family annual income below 150,000 Baht). We have no data on the reach of the scheme. Were the 313,000 very poor students (from families living below the official poverty line) in receipt of loans or did most of the loans accrue to the 429,000 better-off students in the 101, ,000 Baht family income bracket? The upshot of this section is the need to design more precise yet operational definitions of poor and to develop sharper mechanisms for targeting the poor; these should be combined with steps to generate better data on the reach of loans schemes, as an indispensable tool for appraising and improving the efficacy of loans scheme targeting. 87

89

90 Chapter 10 Case study loans schemes: major strengths and weaknesses In this penultimate chapter, we compare the main findings from the five case study loans schemes; these major findings from the case study reports are summarized in Table The aim of this chapter is to identify the major strengths and weaknesses of the case study loans schemes in view of drawing conclusions for policy (to be presented in Chapter 11) China The Chinese experience with student loans is notable for the extensive use made of the banking system in providing funding for loans, thus minimizing the financial burden falling on government. However, loan coverage is still very restricted and additional bank funding for extensive expansion may be limited, thus constraining the growth of student loan coverage. The favourable estimated loan repayment level and recovery ratios for the subsidized scheme, which are high by international standards, may not be sustainable, given the high repayment burden resulting from the short four year repayment period. Indeed, evidence of repayment difficulties appeared in 2003, when the first student borrowers graduated. Shen (2004) argues that it may be necessary to extend the repayment period to nine years in order to lighten the repayment burden and reduce the risk of high repayment default. However, as more material becomes available upon which to judge the efficacy of the new scheme, it may appear that broader systematic adjustments of other parameters are necessary Hong Kong Of the five loans schemes examined, the comprehensive Hong Kong scheme emerges as the most successful overall. The main feature of the scheme is its broad scope. It is a well-funded means-tested scheme, run 89

91 Policy options for student loans schemes: lessons from five Asian case studies by a centralized loans agency and offering tuition grants for poor students, supplemented by loans for living expenses; non-subsidized loans are available for the other students. While over a third of all students are covered by the scheme, the small geographical area facilitates effective administration and repayment collection. Student expenditure surveys provide an objective reference for updating loan amounts. Less favourable economic climate and labour market conditions may lead to a higher repayment burdens and greater default Republic of Korea Unlike in the other cases, there is no centralized loans scheme in the Republic of Korea. Six independent schemes operate in parallel, which have different objectives, target population groups and sources of finance. The largest scheme, run by the Ministry of Education and covering about 6.4 per cent of student enrolment, is aimed at assisting the poor; funding is provided by commercial banks. Other major schemes are run for the benefit of particular groups, such as teachers or government employees (without being restricted to the poor); loan capital comes from existing financial funds such as pension funds. Overall student coverage is about 15 per cent The Philippines This veteran scheme run by the government on a small scale, suffers from inadequate government budgetary provision, leading to very restricted coverage. The size of individual loans could also be readjusted so as to better suit students needs. A particular effort should be made to improve collection, which would lead to a much-needed increase in repayment. Two new schemes are currently being experimented, which are directed at a particularly poor region and to university Centres of Excellence; they aim at improving performance by placing repayment collection in the hands of participating universities Thailand This ambitious scheme, aimed at needy students, has the largest coverage amongst the case studies; it covers upper secondary and vocational education, as well as all of tertiary education. Run through a centralized loans office, it is operated as a top-down scheme, with loan 90

92 Case study loans schemes: major strengths and weaknesses budgets allocated to individual schools and other educational institutions, which then distribute loans to selected student applicants. The system leads to considerable horizontal inequities and appears to target the poor well enough. Built-in subsidies are extensive, and the repayment ratio and loan recovery are very low. The scheme having over-reached original plans, budgetary constraints have led to a retrenchment, reflected in a reduced number of new borrowers and smaller-sized loans. A major restructuring of the scheme is currently being planned. Table 10.1 Loans schemes: major strengths and weaknesses Case study Major strengths Major weaknesses China Funding provided by banking Small coverage (weak national system (minimal financial impact) burden on government) Heavy borrower repayment High repayment ratios burden Promising experiment with Recent evidence of difficulties commercial loans in repayment collection Funding for loan scheme expansion may be limited Hong Kong Extensive student coverage Uncertain economic and The various loans schemes are political future well integrated by central overseeing agency Loans schemes for both poor and non-poor students Use of student expenditure surveys to determine loan size objectively Small geographical area: facilitates repayment collection Republic Funding provided by banking Numerous schemes: of Korea system or existing financial rationalizing and integration funds (minimal financial burden desirable on government) The schemes aimed at the poor Multiple schemes, with differing have low coverage, only cover objectives, able to meet tuition and are insufficiently the needs of different well-targeted sub-populations Scheme for government Satisfactory loan repayment officials is excessively and recovery ratios generous 91

93 Policy options for student loans schemes: lessons from five Asian case studies Table 10.1 (continued) Case study Major strengths Major weaknesses Philippines New schemes being designed Very limited funding and experimented in order to Small national impact: covers reach very specific population very few students (less than groups (CEO and Region 5) 1% of enrolment) Need for greater administrative skills in management of scheme Virtually no loan recovery in SNPL scheme (Region 5 and COE schemes are largely untried) Individual loan size is too small to meet student needs No loans scheme reform in the offing Thailand Extensive coverage of upper Over-extended in relation to secondary and tertiary student plans; weak central control population Inadequate funding has led to Loan scheme restructuring is retrenchment of the scheme now on the public agenda Over-generous built-in loan subsidies Minimal loan repayment and recovery Poorly targeted on needy population Massive horizontal inequities Default sanctions weak 92

94 Chapter 11 Key issues in design and reform: lessons from the case studies In this final chapter, we bring together some of the major lessons that may be distilled from the foregoing comparative analysis and from the case studies themselves. We discuss the major topics that have emerged from the discussion in preceding chapters as well as key issues which ought to be considered in the process of designing a new loans scheme or in the reform of an ongoing one Applying lessons from international experience A central idea underlying the UNESCO Bangkok/IIEP Regional Study of Student Loans in Asia, is the value of in-depth comparative analysis across countries, as a tool for policy makers in charge of identifying the strengths and weaknesses of alternative loans scheme structures and financing arrangements. But in many cases, a comparative analysis takes a very different form. Rather than adopting the eclectic approach of drawing careful lessons from a wide range of international experiences scrutinizing what has worked well and under what conditions the analyst seeks out good practice in the form of successful practice in a particular country, which then becomes a model, even a blueprint to be followed. But the efficacy of relying too freely on institutional borrowing across countries has been much questioned in recent years, both in relation to student loans schemes and other policy areas. Differences in institutions, administrative capabilities and socio-cultural norms all inevitably interfere with successful institutional transfer across countries to a greater or lesser extent. In discussing the strong influence of American practice on the design of top-up student loans in the UK in the 1980s, McFarland (1993) emphasizes the importance of undertaking a systematic examination of any foreign scheme s appropriability ; the latter he defines as the chances of success in transferring a loans scheme, effectively and faithfully, from 93

95 Policy options for student loans schemes: lessons from five Asian case studies one country to another, without serious distortions in the scheme s substance or impacts. Problems of appropriability loom in a number of wellfunctioning national loans schemes, where some central aspects of which are not readily transferable to different social and institutional settings in other countries. A case in point is the successful income-contingent Australian loans scheme (the Higher Education Contribution Scheme HECS), which has been seen as a model of good practice to be emulated in other countries. Recent student loan reform in the UK and particularly the introduction of income-contingent loan repayment has been strongly influenced by HECS; the presence in both countries of a sound public administrative framework, and of a well-developed and effective system of income tax collection with broad national coverage, is a necessary condition for success of HECS and of the new UK scheme. But can the HECS model be applied successfully in more diverse settings, particularly in developing countries where well-functioning regimes of personal taxation are generally lacking? Bruce Chapman, the principal architect of HECS and a leading advocate of income-contingent loans schemes for advanced Western countries, is less than sanguine about the prospects for the successful adoption of HECS-type loans schemes in developing countries, where well functioning regimes of personal taxation are generally lacking (Chapman, 2003). Thus it is perhaps not surprising that, to date, almost none of the loans schemes in developing countries is of the incomecontingent type; noteable exceptions are the fairly successful South African scheme and a new and unsure scheme in Namibia. 25 It is in this context that current government moves to introduce HECS-type incomecontingent loans in Thailand should be regarded as highly problematic, constituting an inappropriate form of institutional transfer across countries. 25. Indeed, Bruce Chapman, a principal architect of HECS and a leading advocate of incomecontingent loans for advanced Western countries, has expressed doubts about the prospects for the successful adoption of HECS-type loans schemes in developing countries (Chapman, 2003). Loan capital in other Korean loans schemes are also provided by existing financial funds: the Ministry of Labour scheme for industrial employees is financed from the Employment Insurance Fund; teachers and their children from the Korean teachers pension fund; and the Korean Labour Welfare Corporation for industrial accident victims from the Industrial Accident Compensation Insurance Fund. 94

96 Key issues in design and reform: lessons from the case studies 11.2 Issues emerging from the case studies Achieving national impact In some countries loans schemes remain very small in size, often constrained by government budgetary limitations; they cover only a small percentage of enrolled students. While these schemes no doubt provide a useful service to the students who receive a loan, their small size limits their efficacy in achieving the national objectives set for the loans scheme in question. A student loans scheme cannot be regarded as successful in the national context, unless it is sufficiently sizeable to achieve an impact nationally. Loans scheme objectives should be clearly defined National student loans schemes may be established for different reasons. The central objective of a loans scheme will influence the design and the operation of the scheme as a whole, as well as its subsequent financial sustainability. For this reason it is important that the central objectives for introducing a loans scheme are carefully defined. Five different sets of objectives for student loans schemes may be identified: 1) budgetary objectives (income generation); 2) facilitating the expansion of higher education; 3) social objectives (improving equity and access for the poor); 4) meeting specific manpower needs; and 5) easing student financial burdens. In practice, any given scheme may incorporate more than one objective. A clear distinction may be drawn between the first group of objectives 1 and 2, and between the second group of objectives 3, 4 and 5. The first group is concerned with generating funding for the university sector, in the light of tight government budgets and the ensuing increase in social demand for higher education; they have constituted the major rationale for the spread of student loans schemes in industrialized countries. The second group of objectives is not centrally concerned with university funding as such, but rather has a broader, societal perspective. Consider policy alternatives to loans schemes While recognizing that student loans schemes may be designed to meet different objectives, it should be also noted that a student loans scheme may not constitute the optimal mechanism for meeting any particular 95

97 Policy options for student loans schemes: lessons from five Asian case studies objective. Prior to the introduction of a loans scheme, or a major reform of an existing scheme, policy makers should weigh the advantages and disadvantages of available policy alternatives to loans schemes, which may be better able to achieve the same objectives. Such an analysis may indicate that these alternative measures or policies should be introduced in concert with a loans scheme to strengthen its effect or, if seen preferable, they may constitute an independent solution. Careful social and economic analysis of policy alternatives is a prerequisite for sound policy reform. Mortgage versus income-contingent loans Mortgage-type loans are the common standard for loans schemes in developing countries worldwide; repayment is made over a specified time period, usually in fixed monthly payments. A designated rate of interest on the loan and a maximum loan repayment period are employed to calculate the fixed periodic payments. A regime of fixed-level repayments may impose a heavy burden on new graduates in the initial years following graduation, given low starting salaries of new graduates and graduate unemployment in many developing countries. A heavy repayment burden, it is often claimed, will lead to repayment default. Under income-contingent loans, which have been introduced in a small number of industrialized countries, periodic loan repayment obligations are determined as a proportion of the graduate s income in each period; this would lower the repayment burden in the earlier repayment years. The participation of the tax authorities in collection is necessary in this case because information on individual income is required. Yet while the tax system can be successfully used where PAYE systems are widespread and effective, these mechanisms are insufficiently well developed for this purpose in most developing countries. If tax authorities proved to be an effective collection agent in the context of any given country, then they could be also used in conventional mortgage-type loans schemes (where repayments may not be income-contingent). Unitary schemes versus multiple schemes There are a number of advantages in setting up a single, central national loans scheme; these include: increased impact in achieving the loans scheme s central objective(s); economies of scale allowing low administrative costs per loan; administrative efficiency and effectiveness; 96

98 Key issues in design and reform: lessons from the case studies specialized staff and functions, such as the ability to undertake evaluation and follow-up studies. Where multiple loans schemes are in place, these have generally been the result of ad hoc or historical processes rather than of deliberate design. Yet, while they may lack potential national impact, comprehensive coverage and may be poorly co-ordinated, multiple loans schemes can be effective where there are multiple objectives to be reached; they can serve the particular needs of very different target groups; finally, they can be effective where it is possible to tap alternative sources of loan funding. Loan capital provision: ensure sufficient government-sourced funding for sustainability Commercial bank loans for financing education (at reasonable rates of interest) are usually not available, because most potential students (or their families) cannot provide collateral to cover the loan nor offer proof of creditworthiness. In most cases (though not all, as we note in ), the government intervenes by supplying the initial capital for the loan, either from the current budget or from borrowing. (It also subsidizes interest payments, bears the risk of default and absorbs much of the administrative costs of operating the loans scheme). A common misconception is that initial loans scheme capital can be regarded as a revolving fund, with the build-up of annual loan repayment receipts providing the funds for new loan commitments. But even in a pure cost-sharing model, repayments will fall short of the original sums borrowed because of loan subsidies, repayment default and governmentabsorbed administrative costs. Thus a new loans scheme is likely to require more than an initial infusion of capital by the government. Given less than full cost recovery and growing student enrolment, the commitment of the government to continue funding will be necessary to ensure sustainability in the long term. Diversification of sources of loan capital It is advantageous to use non-governmental sources for initial loan capital, where available. This obviates the need for government initial funding, though not for ongoing subsidy as has been discussed. The considerable advantage of employing the commercial banking sector to 97

99 Policy options for student loans schemes: lessons from five Asian case studies provide loan capital is evident. But some preconditions are necessary for student loan financing by commercial banks. Where this is the case (as in the Korean Ministry of Education and Chinese schemes) they are offered to students at below-market interest rates, with government funding providing the difference (though in the Chinese commercial scheme banks receive no interest rate subsidy). Again, given the inability of most students, particularly from a poor background, to supply loan collateral, the banking system is loath to provide commercial loans to students, thus requiring the government to step in as guarantor. A central issue that still needs to be addressed is: In schemes where banks assume this role, what enabling institutional and other conditions are present, that are shunned in most other schemes, both in the region and elsewhere? This issue is on the agenda for future research. Another option is the use of an existing fund (such as a public pension or insurance fund) or of a quasi-governmental financial institution to provide the loan capital. Such arrangements are present in the Republic of Korea but have not always worked well and must be employed with care. Since full loan recovery is not achieved in many of these schemes (partly because loans are offered at below market rates of interest), the overall effect may be to weaken the financial robustness of the funds in question if state subsidies are not available. Measuring financial viability of loans schemes Measurements of the loan repayment ratio and of loan recovery level provide important indications as to the financial viability of a given loans scheme; these measures constitute valuable policy tools for improving the efficacy of loans schemes. The loan repayment ratio (measuring how much of a loan an individual borrower is required to repay) is contingent on the amount of subsidy (low interest, grace periods) that is built into the scheme; a low repayment ratio may indicate an overgenerous loan subsidy. The overall financial viability of a loan programme depends on the extent to which loan outlays (loan disbursements plus all other costs including administration) are recovered by the lending body; extensive built-in subsidies, high administration costs and repayment default will result in lower recovery ratios. Very few developing countries are known to have estimated these indicators in the planning and design stages of new loans schemes; an 98

100 Key issues in design and reform: lessons from the case studies unduly low repayment ratio might indicate the need to tighten up on planned conditions of repayment (and provide less generous subsidies). And while these measures are not widely used subsequently, it is strongly recommended that they are employed as ongoing tools for financial evaluation and monitoring of loans schemes. Centralized or decentralized loans schemes? Loans schemes differ markedly as to the extent of functional decentralization. In some schemes, a central loans agency is responsible for all facets of the loan process. One major advantage of centralization is that, through centralized procedures and the use of common criteria for loan allocation, horizontal equity in the distribution of loan budgets is attained; this is particularly important for a loans scheme aimed at serving the poor. The task of loan repayments collection may also fall within the remit of the loans agency but this task is best left to institutions with comparative advantage in collection, such as banks or income tax collection institutions. At the other extreme are considerably decentralized schemes. Rather than employing a centralized clearing-house to service loan applications, loans are distributed to students by individual educational institutions. Educational institutions receive a loan budget through a top-down budget allocation process, which is unrelated to the poverty profile of students enrolled at a given institution. A precondition for a properly functioning loan distribution system at the institutional level is a well designed formula for allocating the total loan budget across educational institutions: such a formula would assign budgets in relation to potential student need (poverty level) in each institution. A decentralized system may further lack horizontal equity if loan distribution policies differ widely across educational institutions. Some decentralization in the distribution system may be desirable. Educational institutions may perform a positive role in promoting the loans scheme, identifying poor students and vetting applications; these are useful roles which could be incorporated to good effect into centralized loan distribution mechanisms. Responsibility for repayment collection may be placed in the hands of the educational institutions themselves, in an attempt to lower repayment 99

101 Policy options for student loans schemes: lessons from five Asian case studies default. However, loan repayment collection by educational institutions is unlikely to be effective, since both incentives and administrative capacities may be absent. Adequacy of loan size Individual loans should be of sufficiently size to meet the needs of the students at whom the scheme is directed. In many schemes, the loan falls short of student needs for education and living expenses. Given the social orientation of many of the schemes, the implication of small loan sizes is that higher education remains beyond the means of the very poor, thus largely defeating the purpose of the scheme. Three elements are required to ensure adequacy of loan size: availability of sufficient total funding; information on students financial needs, probably via surveys of student expenses; and equitable distribution policies. Need to justify subsidized loans In most countries, student loans schemes are highly subsidized. Whether subsidization is justified or not largely depends on the loans scheme s principal objective. The level of subsidy should be analyzed relatively to the objectives to be achieved. For most objectives, there is no compelling case for extensive subsidies; there is a clear case for strong subsidies only where loans are aimed at directly helping the poor. The government subsidizes the interest rate in almost all student loans schemes. Even those loans schemes which are not aimed specifically at the poor provide loans at rates of interest below those payable on commercial loans. Subsidized rates of interest derive from the political judgment that a non-subsidized interest rate would impose too heavy a repayment burden on student borrowers; this assumption should be tested by computing the repayment burden (annual repayments in relation to expected income) and if the burden is found to be light, interest rates can be raised. Ensure horizontal equity in loan provision A major concern should be to ensure that all potential loan recipients are treated equitably. This issue is twofold. First, all potential (eligible) individuals should have equal chances of receiving a loan. Thus, regardless 100

102 Key issues in design and reform: lessons from the case studies of area of residence or educational institution attended, all youngsters of similar socio-economic background should have equal opportunities for securing a loan. Second, all loan recipients of the same economic status and level of need should receive loans of the same size and under the same repayment conditions. These tenets of horizontal equity do not hold in practice in many countries, especially where administration is decentralized or where multiple schemes are in place. Unify grant and loan policy Student grant and loan programmes are often operated within separate policy and administrative frameworks. This does not constitute good practice. A subsidized loan (the subsidy element being equivalent to a hidden grant ) may be seen to lie somewhere on the continuum between an outright grant and a non-subsidized loan. Given the lower administrative costs of grants as compared to loans (which also incorporate collection costs), an outright grant might represent a more cost-effective instrument for assisting the poor than a highly subsidized loan. A unified programme for grants and loans would offer a menu of options: outright grants for the very poor, combined with subsidized loans for the needy (the degree of subsidy depending on socio-economic background); student loans for nonequity objectives would carry low levels of subsidy only. Develop effective eligibility and screening criteria Where the central objective of a loans scheme is to provide loans to students from poor families, procedures should be adopted to ensure that loans do indeed reach this target group. Effective screening mechanisms need to be put in place. This requires not only the setting of definitions for loan eligibility (usually in terms of family income ceiling) but also the ability to check and test the veracity of such information supplied by loan applicants. Where it is difficult to authenticate this information, the scheme is subject to abuse, with non-poor students gaining access to subsidized loans. In countries where student loans (set up for other than social purposes) are available to all students, regardless of need or ability, open access to loans can be prohibitively expensive to governments, particularly if governments provide the loan capital and/or the scheme is highly subsidized. Growing student numbers and less-than-full loan recovery will 101

103 Policy options for student loans schemes: lessons from five Asian case studies result in increasing, and unsustainable, pressures on limited loan funds. This indicates the need for some form of loan rationing. If loan distribution is not to be arbitrary (on the basis of who applies first) a policy of rationing through deliberate targeting (probably of the poor) should be adopted. Proactive targeting Ensuring that loans reach only eligible students is but one component of effective targeting. The latter is also strongly concerned with the actual composition of those accepted for a loan (and with determining whether they constitute the group most in need of a loan), as well as with reaching out to the self-excluded but eligible individuals who do not apply. Too often it is the eligible but only moderately poor students that apply for loans, rather than those most in need. Thus, a proactive targeting focuses on those most in need; it aims at reaching the most deserving sections of the target population defined in terms of those most in need of help, including such less readily reached populations as student dropouts in need and the rural poor. How should loan repayments be collected? Choosing an appropriate collection institution is central to effective loan recovery. Self-collection by bodies such as autonomous public institutions for student loans, government ministries or universities could be effective in relatively confined contexts. But such institutions are unlikely to possess comparative advantage in repayment collection and to have effective capacity to do so. Agency collection, where the task of collection and follow-up is outsourced to a specialist agency, is preferable. Commercial banks usually have the necessary infrastructure and expertise that loan institutions may lack. Strong collection agencies are not enough to ensure successful repayment collection. The incentive for collection institutions to collect pro-actively can be undermined by public policy. Full government loan guarantees may encourage private or public banks to collect from the government rather than the debtor. Thus a less-than-full government guarantee is advisable, with commercial banks taking on a small part of the risk of repayment default. In addition to banks, other possible collection agency options may be available; these include national institutions concerned with collection, 102

104 Key issues in design and reform: lessons from the case studies particularly the income tax administration. While such collection has worked well in a few industrialized counties as part of an income-contingent loans scheme, sufficiently well developed income tax mechanisms are not available in most developing countries. Reducing default: avoiding heavy repayment burdens Repayment difficulties lead to default. A number of measures may be employed, within mortgage loan repayment procedures, to obviate heavy repayment burdens; some of these are standard practice, others could be adopted with ease. These measures include, for all borrowers, the delay of repayment obligations for a period of grace following graduation and the introduction a longer repayment deadline. The introduction of a lower interest rate (acting as a general loan subsidy) would also lighten the burden of monthly repayments, but this would lead to a shortfall in total recovery. An increasing repayment schedule (albeit based on the mortgage loan principle) may help to lessen the extra burden on young graduates during the early repayment years. Finally, repayment deferral may be made available to borrowers on an individual basis, in cases of financial hardship: to the unemployed and to low-income workers, whose income falls below a designated low-income ceiling. Reducing default: penalizing repayment evasion The most widespread measure, simulating commercial loans, is to require the co-signature of a guarantor (usually a family member) who undertakes to pay the loan in the event of default. Requiring a guarantor can have negative consequences that may defeat the purpose of loans schemes with social objectives: The very individuals who are most in need of support may be the least able to provide guarantors. Perhaps for this reason, little more than lip-service is accorded to guarantor provisions in many schemes; in practice, the task of the guarantor may vary from little more than that of moral councillor to a provider of financial collateral. More effective measures include undertaking legal action against defaulting borrowers or their guarantors and barring access to further credit. Moral pressure could be exerted through the publication of the names of inveterate defaulters. Finally, measures to inculcate a more positive attitude towards repayment could be developed; in some cases, this may be tantamount to changing social norms and therefore not feasible in the short run. But universities could play an important role in emphasizing the requirement for repayment. 103

105 Policy options for student loans schemes: lessons from five Asian case studies Technical experts for evaluation; develop data bank Effective loans scheme administration is often constrained by a lack of adequate financial appraisal, forward planning, monitoring and evaluation. The need to build up the research and evaluation capacities of loans schemes organizations is evident: this will require the construction of an appropriate database and the development of technical expertise. Learning from international experience Policy reform in the field of student finance, as in other areas, should be guided by international experience of good practice (on what has worked well) and of mistakes to be avoided. However, instant institutional borrowing, based on a particular country s practice, should be avoided Towards a typology of good practice Based on the discussion from the previous section, a summary of these issues is presented, in Table 11.1, constituting a typology of good practice. It consists of a basic list of twenty characteristics, central to good practice loans schemes. Table 11.1 Twenty characteristics of good practice loans schemes 1. Loans schemes are sufficiently large in coverage to achieve national impact 2. Loans scheme objectives are clearly defined 3. Policy alternatives to loans scheme have been considered 4. Any planned departure from mortgage-type loans scheme is closely analyzed 5. Preference is given to unitary schemes 6. Loan capital provision: governmental funding sufficient for sustainability is ensured 7. Loan capital provision: wherever possible non-government funds are used 8. Financial viability of loans schemes is appraised before implementation and monitored on a continuing basis 9. Appropriate decentralization of administrative roles is implemented 10. Individual loans are sufficiently high to meet student needs 104

106 Key issues in design and reform: lessons from the case studies Table 11.1 (continued) 11. Subsidized loans are justified 12. Horizontal equity is respected in loan provision 13. Grants and loans are part of a single policy and are administered in common 14. Appropriate and effective eligibility and screening criteria have been developed 15. Pro-active targeting mechanisms have been developed 16. Loan repayment collection is outsourced to collection agency whenever possible, usually through commercial banks 17. Reducing default: heavy repayment burdens are avoided 18. Reducing default: measures are taken to penalize repayment evasion 19. Sound information base is developed and of technical expertise is employed for evaluation and appraisal 20. Lessons are learned from international experience; instant institutional borrowing is avoided 105

107

108 References Albrecht, D.; Ziderman, A Deferred cost recovery for higher education: student loans schemes in developing countries. (World Bank Discussion Paper No. 137.) Washington, DC: World Bank. Chapman, B.; Ryan, C Income-contingent financing of student charges for higher education: assessing the Australian innovation. In: The Welsh Journal of Education, 11(1), Chapman, B Income contingent loans for higher education: international reform. Draft chapter for Economics of Education Handbook, forthcoming. Chung, Y.P The student loans scheme in Hong Kong. Bangkok: IIEP; UNESCO Bangkok. Johnstone, D.B Sharing the cost of higher education: student financial assistance in the United Kingdom, the Federal Republic of Germany, France, Sweden, and the United States. New York: The College Board. Johnstone, D.B. No date. Student loans in international perspective: promises and failures, myths and partial truths. Buffalo: Center for Comparative and Global Studies in Education. (Available at / Kim, A.; Lee, Y Student loans schemes in the Republic of Korea: review and recommendations. Bangkok: IIEP; UNESCO Bangkok. Kitaev, I.; Nadurata, T.; Resurrection, V.; Bernal, F Student loans in the Philippines: lessons from the past. Bangkok: IIEP; UNESCO Bangkok. McFarland, L Top-up student loans: American models of student aid and British public policy. In: Oxford Studies in Comparative Education, 3(1). 107

109 Policy options for student loans schemes: lessons from five Asian case studies Psacharopoulos, G.; Tan, J.-P.; Jimenez, E Financing education in developing countries. Washington, DC: World Bank. Shen, H.; Li, W A review of the student loans scheme in China. Bangkok: IIEP; UNESCO Bangkok. Shen, H The most urgent problem in student loans scheme in China: repayment. In: Harvard China Review, Spring, Ziderman, A.; Albrecht, D Financing universities in developing countries. (The Stanford Series on Education and Public Policy.) London: The Falmer Press. Ziderman, A (July). The student loans scheme in Thailand: a review and recommendations for efficient and equitable functioning of the scheme. Report prepared for UNESCO Bangkok, as part of the Asian Development Bank Loan Programme, in the framework of the Education Management and Finance Study (Project TA THA). Bangkok: UNESCO. Ziderman, A Financing student loans in Thailand: revolving fund or open-ended commitment? In: Economics of Education Review, 21(4), Ziderman, A Alternative objectives of national student loans schemes: implications for design, evaluation and policy. In: The Welsh Journal of Education, 11(1), Reprinted in: Peking University Journal of Education, April 2004 (in Chinese). Ziderman, A Student loans in Thailand: are they effective, equitable, sustainable? Bangkok: IIEP; UNESCO Bangkok. 108

110 Glossary (* indicates that that a term is defined elsewhere in the Glossary.) Administration cost: The total cost of running a loans scheme; administration costs are not usually charged to the student-borrower but absorbed by the lending body (and thus subsidized by the government). Amortization period: The period over which repayment of the loan plus interest on the loan is made, until the loan is repaid in full. Case studies of loans schemes: Detailed investigation of the functioning of individual country loans schemes in particular countries. Centralized/decentralized loans scheme: In centralized loans schemes a central agency is responsible for carrying out all the major facets of the loan process, including the receipt and processing of loan applications. In decentralized loans schemes, active roles are played by other institutional players, such as banks and educational institutions, in key areas such as loan allocation. Co-signatory: An individual who adds his signature to a loan agreement, agreeing to act as guarantor*. Cost recovery: The introduction or raising of tuition fees from nominal levels, in order to generate additional revenues for educational institutions. Cost sharing: A process whereby a greater share of the costs of instruction and living expenses is borne by parents and students. Cross-subsidization: A process whereby grants for needy students are financed ( cross-subsidized ) by higher fees paid by the majority of students. 109

111 Policy options for student loans schemes: lessons from five Asian case studies Default risk: The risk (for a lending institution) that a borrower will default on repaying a student loan (i.e. the borrower does not repay the loan at times specified in the loan agreement). Defaulter: A borrower who fails to make repayments when due (i.e. the borrower does not repay the loan at the times specified in the loan agreement). Excessive default: Some repayment default is to be expected in all student loans schemes (the costs of which are absorbed by the loans agency); but a high level of default ( excessive default ) will result in a low recovery ratio* and financial difficulties for the loans scheme as a whole. Financial viability: A loans scheme is unable to survive over the longer term (it is not financially viable ) if income from repayments is insufficiently high in relation to loan disbursements, or if the shortfall is not met by budgetary allocations. Full loan recovery: Full loan recovery is achieved when the present value of the total repayment stream equals the total cost of providing the loans; this would imply no loans subsidization* (i.e. the full repayment of sums originally lent plus interest and administrative cost) and zero default. Grace period: Period of time, immediately following the completion of studies, during which the borrower is not required to make loan repayments; it is assumed that graduates need time to find suitable employment. Grants versus loans: Student financial aid may take the form of outright grants to students to cover tuition fees and/or living expenses (also called bursaries or scholarships) or the form of loans; they may also be used in combination. Grants are cheaper to administer (there is no repayment collection) and provide stronger incentives to poor students to remain in study; but loans are recoupable and therefore impose a lighter burden on educational budgets over the longer term. Guarantor: An individual, usually a parent or other close relative, who undertakes (in writing) to be legally responsible for repaying a loan in the event of default on loan repayment. 110

112 Glossary Horizontal equity: Horizontal equity in loan distribution requires that loan applicants, of same economic status and level of need, are given equal chances of receiving a loan. Inappropriate institutional transfer: This applies when institutional arrangements, that work well in other countries, are transferred to a country with substantially different institutions, administrative capacities and socio-economic norms; they are thus unlikely to work well in this new setting. Income-contingent loans scheme: Under income-contingent loans schemes, the periodic loan repayment obligation is fixed as a percentage of the borrower s earnings each year; this may be constant for all income levels or progressively increasing for higher levels of income. Either repayments continue until the loan has been repaid or a repayment period* is stipulated. Loan eligibility: Most loans schemes are not available to all students but to specific categories of students who are eligible to apply for a loan: eligibility is usually defined in terms of student s family income. Loan rationing: A method for allocating student loans when the total value of eligible loan applications exceeds the supply of funding available. Rationing may be based on student/family income level (and employing a means-test*), on academic merit or on a firstcome, first served basis. Loan repayment ratio: The loan repayment ratio measures how much of the full value of a loan a borrower is required to repay. Loan repayment ratio = required repayments/loan received. Required repayments fall short of the value of the loan received because of such built-in loan subsidies as below market interest rates and repayment in nominal terms. Loan repayment: The process whereby a borrower makes regular, periodic payments to the lending body until the original loan is repaid. Loan size ceiling: The maximum amount that can be lent to an individual borrower. Means test: An enquiry into the means (income or economic resources) of a person applying for a student loan. 111

113 Policy options for student loans schemes: lessons from five Asian case studies Means-tested loans scheme: A scheme in which borrowers are selected on the basis of the applicant s means (income or economic resources) or those of his family. Mortgage type loans scheme: In mortgage type loans schemes, repayment is made over a specified repayment period*, usually with equal (nominal) monthly payments; less often, repayments are graduated, so that initial repayments do not impose an excessive repayment burden* on borrowers. The vast majority of loans schemes are of this type. Poverty line: A designated level of family income; families with income below this level are defined as being in poverty. Reach of loans scheme: Ascertaining which student groups are in receipt of a loan, in order to make sure that the beneficiaries of a loan programme are those designated as most deserving of support. Recovery ratio: The recovery ratio measures the extent to which a program of student loans is repaid. Recovery ratio = total repayments/total outlays. Loan recovery depends on three factors: the amount of interest subsidies on the loans, repayment losses due to default and non-recovered administration costs*. Repayment burden: The repayment burden defines the amount that an individual borrower is required to repay as a percentage of his current annual income; too heavy a burden (approximately in excess of ten per cent) is likely to result in repayment default* while a light burden may indicate that the level of subsidy is excessive. Repayment collection: The process of collecting repayments due on outstanding student loans, performed by the student loans body or by a specialized collection agency*. Repayment deferral: Decision to allow a borrower to postpone required repayments to a later period (usually because the borrower experiences temporary low income). Repayment period: The period of time set out in the loan agreement by the end of which the borrower has to repay a loan. 112

114 Glossary Revolving fund: A fund set up for a specific purpose (such as student loans) in which current receipts into the fund (i.e. repayments, in the case of a loans scheme) are able to cover current outgoings, so that the size of the fund is maintained over time. Screening: The process of verifying which loan applicants are eligible for a loan. Self-collection/agency collection: Self-collection relates to arrangements where the central institution operating the loans scheme also assumes responsibility for carrying out the collection of loan repayments. With agency collection, the task of repayment collection and follow-up is outsourced to a specialist collection agency such as a bank or a national institution concerned with collection (e.g. the income tax administration). Social rate of return: The social rate of return on a given educational expenditure measures the relationship between all the social costs of education that must be borne by society as a whole, and the benefits that are expected to accrue to society. It is a useful tool in identifying educational investment priorities. Student expenses survey: A survey carried out amongst a representative sample of students, to measure the level of educational and living expenses incurred by students: this provides a guide in determining the typical loan size and a loan size ceiling*. Student loans scheme: A framework, usually set up and subsidized by the government for providing students with loans, so as to enable students to meet educational and/or living expenses in the absence of commercial loans from the banking system. Student price index: A price index measures the average increase in prices over time of a defined collection of goods and services; a student price index relates to goods and services purchased typically by students. Subsidized/non-subsidized loans scheme: A non-subsidized loans scheme is one which operates according to market principles, with the borrower meeting all of the costs associated with the loan. 113

115 Policy options for student loans schemes: lessons from five Asian case studies Targeting: The process of directing limited public resources to narrowly defined deserving categories of people; in the case of student loans, targeting is usually directed towards the poor. To default: To fail to make repayments at due date (i.e. at as specified in the loan agreement). Unitary/multiple schemes: In a unitary loans scheme, a single national loans scheme is established to cater for eligible students; in contrast, in the case of multiple schemes, there are a number of separate loans schemes, run by different bodies, each usually serving a particular client group. 114

116 Appendix Table A1.1 Treatment accorded to various loans schemes in case study reports Case study Loans scheme Loans scheme Treatment in intended for: case study report China Government-Subsidized Student Poor students Detailed Loan Scheme (GSSLS) subsidized scheme General Commercial Student Loan All students Partial Scheme (GCSLS) commercial scheme Hong Kong Local Student Finance Scheme Poor students Detailed SAR (LSFS) subsidized scheme Non-means-tested Loan Scheme All students Partial (NLS) non-subsidized scheme Republic Ministry of Education and Human Poor students Detailed of Korea Resources Development loan scheme (MOE) Ministry of Education and Human Engineering students Not discussed Resources Development loans (new scheme) for engineering students Ministry of Labour loans scheme Industrial employees Partial (MOL) Korea Research Foundation (KRF) Poor and meritorious Partial students Korea Teachers Pension Teachers and their Partial Corporation loans scheme (KTP) children Government Employees Pension Government employees Detailed Corporation loans scheme (GEPC) and their children Korea Labour Welfare Corporation Victims of industrial Partial (KLWC) accidents Philippines Study Now, Pay Later scheme Veteran scheme aimed Detailed (SNPL) at poor students Region 5 loans scheme Poor students in Region V Partial Student Loan Programme for Poor students at Centres Partial Centres of Excellence (COE) of Excellence Thailand Student Loans Scheme (SLS) Poor students Detailed in secondary and tertiary institutions 115

117 Policy options for student loans schemes: lessons from five Asian case studies Table A1.2 Case studies: major features of selected loans schemes Case Study (1) (2) (3) (4) (5) (loans scheme) Year Scope of Covers public Purpos Number of scheme scheme or private of loan* borrowersas established university % of student enrollment? enrollment China 1999 National Public Tuition Limited (GSSLS + living (3.8%, subsidized end of 2001) scheme) Hong Kong 1969 Regional Public Living Extensive (LSFS (tuition (> 35 %) subsidized scheme) fees are covered by a grant) Hong Kong 1998 Regional Public Tuition Limited (NLS non- +living (about 14 %) subsidized scheme) Rep. of Korea 1975 National Both sectors Tuition Limited (MOE) (6.4%) Rep. of Korea 1976 National Both sectors Tuition Limited (GECP (6%) Government Employees Scheme) Rep. of Korea 2003 National Both sectors Tuition Limited (MOE (1%) Engineering) Philippines 1976 National Both sectors Tuition Marginal (Study Now (< 1%) Pay Later - SNPL) Philippines 1999 Limited Both sectors Tuition Marginal (Region 5) to Region 5 (< 1%) Philippines 2001 Limited to Both sectors Tuition Marginal (Centres of Centres of (< 1%) Excellence) Excellence Thailand 1996 National Both sectors Tuition Extensive + living (> 35%) * Tuition fees may include other fees such as registration. 116

118 Appendix Table A1.3 Lending conditions in case study loans schemes Case study (loans scheme) Interest rate (%) Lending conditions Loans Market Grace period Amortization scheme rate rate (years) (years) China Half of (GSSLS) market rate China Market rate Varies with (GCSLS) (no subsidy) 6.21 participating banks Hong Kong (LSFS) Hong Kong Market rate (NLS) (no subsidy) Rep. of Korea Long term: 3 Long term: 7 (MOE&HRD) Short term: 0 Short term: 2 Rep. of Korea year college: 4 (GEPC) 2-year college: 3 Philippines (SNPL) Philippines (Region 5) Philippines (COE) Thailand

119

120 Copies of this publication may be obtained on request from: UNESCO Bangkok, Education Policy and Reform Unit (EPR) and

121 ISBN The book This book is part of a series of in-depth studies on student loan schemes. It makes a comparative review of five case studies undertaken in Asia, including PR China, Hong Kong SAR China, the Republic of Korea, the Philippines and Thailand. At a time when most Asian countries are experiencing both a dramatic increase in demand for higher education and severe public budgetary restraints, loan schemes are increasingly on the policy agenda. This synthesis study provides a broad canvas of practices and experiences, looking at aspects such as organization, funding, financial viability and targeting. It also provides an analysis of the strengths and weaknesses of each scheme studied, drawing valuable lessons for policy makers. The author Adrian Ziderman is Professor of Economics at Bar-Ilan University, Israel, where he holds the Sir Isaac Wolfson Chair in Economics and Business Administration, and Editor of the International Journal of Manpower. His main fields of specialization are labour markets, the economics of vocational training and university funding, including student loans and cost-sharing. He has served as senior economist for human resources at the World Bank and consultant for several international organizations, including UNESCO, the World Bank, the British Council, DANIDA and the Asian Development Bank. He has also frequently acted as advisor to government ministries, notably in Asia and Africa. UNESCO Bangkok Asia and Pacific Regional Bureau for Education 920 Sukhumvit Road, Prakanong, Bangkok, Thailand [email protected] Website: International Institute for Educational Planning (IIEP) 7-9 rue Eugene Delacroix Paris, France