The research project “Ways out of Poverty, Vulnerability and Food Insecurity” is funded by the German Federal Ministry for Economic Cooperation and Development (BMZ) and carried out by the Institute for Development and Peace (INEF) at the University of Duisburg-Essen. Within this project, examples of successful outreach to poor people and their sustainable exit from poverty with the support of development cooperation (DC) are to be presented in the form of Good Practices. This also includes agricultural financing. In many countries agricultural financing is only offered to a limited extent, in contrast to the availability of credit offers for industry, trade and commerce.

The situation is different in Cambodia, where there has been a steady, recently very large increase in credit offers and loans since the early 2000s, in addition to a rapid increase in the amount of individual loans. The offer also and especially exists in rural areas and includes agriculture to a considerable extent. The credit sector, with 81 registered microfinance institutions (MFIs), 47 commercial banks, twelve specialist banks and 246 Rural Credit Institutions in the formal sector alone, as well as thousands of informal, i.e. non-registered private money lenders, has grown rapidly even in recent years and has led to a steadily increasing proportion of massively overburdened debtors. Hence microfinance (MF) as a whole has come under strong criticism from academics and, above all, from non-governmental organizations (NGOs). A special role is played here by the fact that in Cambodia, even for relatively small loan amounts, the loans are usually secured with land ownership or land use titles, which are put at risk in the event of repayment problems.

Cambodian NGOs in particular even accuse the MFI sector of violating human rights, among other things, because of land title losses among debtors. Their fierce criticism of the conditions in Cambodia’s MF sector has also led to international protests, including German NGOs, and to two minor inquiries in the German Federal Parliament (Deutscher Bundestag) in the meantime. German state development cooperation, which until a few years ago was heavily involved in Cambodia’s MF sector and still indirectly supports the sector today through investment funds, is thus also at least indirectly under criticism.

The NGOs’ criticism is based primarily on four findings: Firstly, over-indebtedness leads not only to massive losses of land, but particularly to the loss of arable land, which is particularly important for many households (hh), and thus to the loss of livelihoods. Secondly, the compulsion to (punctually) repay the loans and their high interest rates is said to lead to child labour, and thirdly, to debt-related more or less involuntary labour migration, e.g. to neighbouring countries. Fourthly, a further consequence is said to be food insecurity (which did not exist before in this form), as money that had previously been invested in feeding the families had to be used for repayment.

Initially, according to further NGO findings, supported by numerous documents, MF in Cambodia started as a poverty reduction project with the aim of helping people who previously had no access to (bank) loans. Today, however, most of the large MFIs are closely linked to or even owned by foreign banks, investment firms and Western development agencies, which make considerable profit from them. In 2017 alone, profits amounted to US$130 million. It should be added that even in 2020, the year of the outbreak of the COVID-19 pandemic, this profit was as high as about US$453 million, according to the National Bank of Cambodia, calculated on the basis of 81 MFIs, including banks also active in the MF sector.

Extensive figures from the documentation of financial service providers (FSPs) as well as our research results indeed prove that at least some of the MFIs and FSPs that converted from MFIs to banks have long since ceased to pursue poverty reduction through MF as their primary objective but have tried to establish themselves in the broad area of small and medium-sized enterprise (SME) promotion and in even higher market segments. The goal of poverty reduction continues to be carried in the “visions” of MFIs and even banks like the important ACLEDA. In reality, however, the loan volumes are far higher than what was originally understood as “micro” financing, even by donors.

The National Financial Inclusion Strategy 2019-2025 (NFIS) also continues to cite access to financial services as an important contribution to poverty reduction in the country (cf. KoC 2019). However, the strategy only emphasises the benefits of financial inclusion, but does not explain how poor hh in particular can benefit from it and what exactly should be done by the state and the FSPs to achieve this.

Both the federal government and other domestic and foreign actors in the MF sector have taken up the NGO criticism and see a real need for reform in a number of points, especially in the regulatory area. Some of the NGOs’ criticism, on the other hand, is rejected by the Cambodian side as well as by donor organizations, combined with the reference to the merely qualitative studies of the Cambodian NGO LICADHO in particular, which bases its serious accusations primarily on compiled individual examples. Statistical studies commissioned by several donors and MF funds operating in Cambodia on the practice of lending and the consequences of over-indebtedness (2017 and most recently 2021) are said to show quite a different picture.

However, since both of these “donor” studies have also been criticised, and the extent of over-indebtedness, but above all the repayment problems and the associated negative effects on the debtors, is relatively unknown so far, the INEF was asked by the BMZ to include Cambodia as part of the studies on agricultural financing and to conduct an empirical study on the overall picture of the debt problem. This investigation, which was postponed several times because of COVID-19, was then carried out between January and April 2022 in Germany and for several weeks on site in six Cambodian provinces as well as in the capital Phnom Penh.

The core piece of the study is a hh survey of 1,388 randomly selected hh. This was supplemented by a total of around 100 interviews on debt and debt consequences. Among other people, these were held with the village chiefs and representatives of rural communities, who are always involved in loan applications that claim land titles as collateral, as well as with the management staff of important MFIs and banks, representatives of the National Bank, the Association of MFIs, with Cambodian NGOs and in focus group discussions with 23 groups of debtors.

In contrast to existing studies, the survey was not to be limited to the circle of MFI borrowers and was also to be conducted absolutely anonymously. The number of hh surveyed was therefore 1,388 – a significantly larger sample than would have been necessary to investigate hh known to have current loans. The interviews revealed that of the total number of hh, 770 or 55.5% of the sample had current loans, of which 672 had only one loan, 78 hh (11.3%) had two, and another 20 hh (2.5%) had three and more loans. Especially on the basis of these 770 hh, details were asked about the loans, their purpose, the positive and negative effects of taking out the loan, the problems of repayment and the solutions found or not found in the process. 648 out of 1,388 hh surveyed were also able to provide information on loans taken out in the last five years. From the results of the hh surveys, the additional interviews as well as the focus group discussions, a relatively clear picture can be drawn of the connection between loans, over-indebtedness and its consequences including the loss of land among borrowers.

Four observations are particularly important in this context:

• (i) The MF market in Cambodia is largely saturated due to the very numerous providers of financial services, apart from a few regional exceptions and one sectoral exception (= smaller agricultural loans for pre-financing cultivation). Without persuading clients to take out new, larger loans or poaching borrowers from other MFIs, it is difficult to significantly expand business in the microcredit sector (even with a definition that still calls volumes between US$2,500 and over US$4,000 “micro”).
• (ii) Debt predominantly, but by no means exclusively, serves investment purposes and the majority achieves its purpose, be it the expansion of income-generating measures, investment in housing or the acquisition of higher-value consumer goods. However, a considerable part of the loans also serves to finance (longer-term) uncovered living costs, which should not be the case with responsible financing and inevitably leads to over-indebtedness.
• (iii) In the context of an oversupply of financial services, the possibility of being able to secure loans with land titles tempts individual MFIs or banks to lend irresponsibly. This happens even in cases where merely examining the cash flow in the context of the loan application assessment would show that repayment from the income secured alone would hardly be possible. In other words, over-indebtedness, with the consequences listed, is foreseeable here.
• (iv) For the reasons mentioned above, in an unacceptable number of cases the debtors have to sell land in order to be able to repay their debts according to the contract, to avoid being categorized as defaulting payers and thereby being excluded from further borrowing. Or they have to sell other goods or are forced to take other measures such as reducing the quality of their food or, in very rare cases, child labour or forced labour migration.

What could not be confirmed is an interest of the FSPs in the land of the debtors and their engagement in land grabbing via purposefully driving borrowers into over-indebtedness. On the contrary, the study shows that MFIs and banks try by all means to prevent expropriation of land titles by the courts, also in order to avoid public criticism regarding the loss of land by defaulting debtors. The fact that instead there is a certain pressure on the debtors to sell land in advance, however, is accepted and approved by the more irresponsible members of the MFIs or the loan officers in the institutions. This occurs even in those cases where the ability to repay the loan was already questionable at the time it was granted.

Whether land sales and other problematic solutions always violate the human rights of those affected must remain a matter of debate. On the one hand, a number of the ultimately problematic loans are by no means the result of persuasion by MFIs or banks alone, but are the result of bad investments, unfortunate coincidences or even risky speculation. In some reported cases, borrowers deliberately concealed their inadequate repayment capacity when applying for a loan or even took out several loans at the same time. This is something which even the recently introduced, relatively strict monitoring of the Credit Bureau of Cambodia (CBC) was unable to prevent. However, regardless of the question of guilt, it should also be noted that the consequence of over-indebtedness must never be food insecurity for a family, child labour or forced labour migration.

A key recommendation of this study is to set the floor for real estate collateral on loans reviewed by the CBC at an amount that could range from US$2000 to US$3000, depending on the purpose of the loan. In any case, this limit should apply to land titles as collateral, below which loans would not be allowed to be secured by land titles.

An immediate measure needing to be initiated through the MF funds supported by German DC is the urgent demand to completely separate the credit assessment by MFIs and banks from the question of whether the loans can be secured with land titles. If cash flow calculations show that it is highly unlikely that a loan can be serviced, i.e. that there is a risk of over-indebtedness, the possibility of securing it through land titles must not lead to the awarding of a loan. This principle should be explicitly included in future contracts between investors and FSPs, even if they are only topping up existing credit lines.

The door-to-door canvassing by representatives of MFIs and banks, which is currently very aggressive, should also be stopped quickly. This step would be particularly welcomed by the majority of the village chiefs and representatives of rural communities interviewed. Another immediate measure would be to position links to compliance mechanisms more prominently on the home pages of MFIs and banks. In this way, even inexperienced internet users would quickly find a way to contact the responsible FSP staff in case of problems.

Since a considerable number of currently over-indebted hh, including those classified as poor (ID Poor), were granted loans not on the basis of cash flow analyses, but because of the presence of land titles as collateral, a serious restructuring or (partial) reversal of loans is recommended. Loans that were clearly granted through gross negligence should at least have their interest cancelled. Loans where the repayment ability of the borrowers was even deliberately ignored (i.e. where clear data from the CBC were completely ignored) should be written off. In both cases, the responsible FSP would have to pay for the costs.

In order to be able to check the loans of over-indebted hh in this respect, a neutral monitoring agency could be set up in Cambodia relatively quickly under the supervision of the National Bank, which on the one hand would check the contracts of over-indebted hh considering ID Poor status or the data of the CBD, and on the other hand could take on the role of a consumer protection agency for the financial sector in the future. In view of the large number of existing MFIs and banks and their financial resources, further involvement of German governmental DC seems unnecessary, at least in the area of general and particularly higher-end MF, especially since the institutions that have so far been supported directly or, more recently, only indirectly through funds, are pursuing the lower segment, i.e. classic MF, with less and less interest.

There is still a need for microloans (in the range of less than US$1,000 or US$2,500), especially for the pre-financing of agricultural production, and these are offered by the MFI sector rather subordinately and at less favourable conditions. New partners should therefore be sought for cooperation in the MF sector, such as agricultural cooperatives, the umbrella organization of agricultural cooperatives, and / or the cooperative development fund under the cooperatives act.