Ethics as an important aspect of microfinance
Not so long ago, the provision of small and very small loans and other financial services to relatively poor people in developing countries and the former socialist countries of Eastern and Central Europe, known as microfinance, was hailed as a fascinating and absolutely positive idea. It was supported by almost all policy-makers and development experts.
The hype about microfinance reached its peak in late 2006, when the Nobel Peace Prize was awarded to Professor Muhammad Yunus and the Grameen Bank, the microfinance institution (henceforth abbreviated as MFI) in Bangladesh that he had founded in the 1970s. The Noble Peace Price had the effect that microfinance became suddenly widely known to the general public and regarded by many as the most humane part of the international financial system, perhaps even the only humane part. And indeed, the prize was awarded for good reason: anyone who strives for a fairer distribution of the opportunities for personal and economic development by providing loans as widely and effectively as Yunus and his bank have done for years is indeed promoting world peace, since lack of access to financial services is one of the main reasons why poverty is perpetuated, making massive and long-term poverty one of the greatest threats to peace.2 Thus, there is obviously a manifest connection between microfinance and ethics.
But as I write in the summer of 2011, the situation has changed in a fundamental way, and this is not only because the financial crisis has affected some MFIs as seriously as many other banks. Much more important is a plainly moral issue. In India, a series of suicides among borrowers who had obtained loans from MFIs occurred in 2010. These tragic events attracted great public attention and tarnished the formerly unambiguously positive image of microfinance. Suddenly general scepticism concerning microfinance has spread. Does microfinance work at all? Can it have any positive effects on poverty and development? Is the business model that many MFIs had adopted in recent years really appropriate? And is it justified to use public funds to support microfinance, as had been done on a considerable scale in past years? All of these concerns also touch on ethical issues.
Thus the relationship between microfinance and ethics is certainly more complex than the Nobel Peace Prize for Yunus and his bank as a reward for discovering microfinance as a means of combating poverty suggests. The complexity goes beyond the doubts that had already been expressed by competent observers for quite some time as to whether microfinance is really a suitable instrument for combating poverty; even if this claim were inappropriate it would not imply that microfinance lacked any developmental or ethical value.
My contribution to this volume discusses several aspects of the relationship between microfinance and ethics. The questionable role of microfinance for poverty alleviation, and the exaggerated claims made in this respect by Yunus and many of his followers are only a minor aspect of this debate. Of greater importance are the tensions that seem to exist between moral standards and economic imperatives in setting up and operating MFIs. More specifically, it is the role and the merits of what has become known as the commercial approach to microfinance.
Lack of access to financial services is one of the main reasons why poverty is perpetuated, making massive and long-term poverty one of the greatest threats to peace.
The term commercial approach to microfinance means a strategy of setting up and running MFIs in such a way that they can cover their full costs, including the cost of equity, after a short start-up period, and be permanently independent of subsidies from development aid institutions. Interestingly, Grameen has not adopted this approach, and Yunus has for many years been one of its most outspoken critics. However, since the turn of the millennium, the vast majority of microfinance experts have been convinced that there is no alternative to the commercial approach. For a few well-known institutions like Grameen Bank with an extremely eloquent and highly respected spokesperson like Muhammad Yunus, it may be a viable business model to simply do good and socially highly-relevant work while relying on outside support for covering the expected deficits of these activities. But for the hundreds and even thousands of MFIs that have come into existence during the past 30 years, this is not the case. Therefore, any serious MFI must to some extent be commercially oriented. But the crucial question is this: how strong should, or even must, the commercial orientation of an MFI be?
That too little commercial orientation can be a problem is self-evident, because a lack of financial viability would threaten the very existence of an MFI. The more interesting cases are MFIs whose activities reflect what I call excessive commercialization. Such cases exist, and they have recently attracted great attention. This is because those events that caused microfinance to lose most of its reputation and its moral appeal have precisely occurred at MFIs which I regard as excessively commercialized.
This paper is built around three propositions: the first is that microfinance continues to be a socially, economically and morally valuable undertaking, provided that it is done properly and is based on a healthy dose of a commercial orientation. The second is that recent developments have caused a moral crisis of microfinance resulting from excessive commercialization. The final proposition is that this excessive commercialization should and can be avoided.
In order to develop and support these three propositions, section two describes how microfinance has developed and become a success story. Section three explains the substance of the 1990s controversy about the commercial approach. Then, in section four, I address recent developments which have led to the current moral crisis of microfinance. I conclude with what may appear to be an inappropriately optimistic and seemingly paradoxical outlook for microfinance for the next five or even ten years.
The development of microfinance
Promoting financial systems in poor countries has long been an important part of Western development-aid policy. It was based on the plausible assumptions that poor countries lack capital and that having a better financial system would spur economic growth. However, views as to what constitutes a good financial system and how the financial system of developing countries can be improved have changed over time. One of these changes led to the emergence of microfinance.
The old policy of development finance that had been pursued since the 1950s consisted of channeling large amounts of capital from Western countries to government-related development banks in the recipient countries, which would then on-lend these funds to state-related projects or agencies and large firms. The optimistic belief underlying this policy was that the promotion of so-called growth poles would ultimately trigger a process of general, self-sustaining growth and that this would also benefit the poor and small businesses because a rising tide would lift all boats.
However, in most recipient countries the hoped-for trickle-down effect failed to take place. The ethical lesson from this failure is clear and simple: it is not enough to rely on illusory and poorly substantiated theories as to what the measures in question will achieve and to hope that the initiatives undertaken will somehow simply lead to positive results.4 Moreover, in most countries the traditional form of development finance had the unintended consequence that income, wealth and economic opportunity became more unequally distributed than before. In the period after 1968 and the end of the Vietnam War this effect was not politically acceptable.
In 1973, Robert McNamara, at that time the president of the World Bank, renounced development finance as it had been practiced so far. His famous speech delivered in Nairobi, Kenya, marked the starting point of the second phase of development finance and the beginning of microfinance, as we call it now. The new policy ushered in by his speech still consisted of channeling funds to developing countries. But now capital was no longer only provided in large amounts to large banks and central agencies but also directly, parceled into small sums, to the newly identified target groups of the poor and of small enterprises.
The idea of bringing small sums of capital directly to those who were supposed to need it most required finding new delivery channels to ensure that these target groups were reached. Given the general interest-rate restrictions that were still in place in almost every country at that time, banks were rightly considered to be unable and unwilling to perform this function: it would never pay for them to serve poor clients. Thus other institutions that were really interested in reaching and supporting small enterprises and poor people were needed; and indeed many such novel institutions were created quite rapidly with foreign support. Most of them were not-for-profit institutions that had high ethical standards and great ambitions. But because they also had very high costs and lacked a well-designed business model, they became the proverbial bottomless pits and almost none of them survived their first years.5 Moreover, a considerable fraction of these new institutions were managed by politically inspired social scientists from Western countries who were not competent to run financial organizations and had little understanding of the financial problems of the owners of small businesses in developing countries. Thus also the second approach to development finance failed, leaving behind another ethical lesson: it is simply not enough to have good intentions.
The advent of modern microfinance
Another group of new development-finance projects had better prospects for success. They consisted of externally financed credit programs of NGOs which were not shaped and dominated by sociologists and their like from the industrialized countries, but rather by politically liberal men and women, often themselves entrepreneurs, from the respective host country. The initiators of these programs were aware of the needs of micro-entrepreneurs, the main target group of the new policy, and they could be assumed to understand their economic and financial problems.
Towards 1990, the conviction emerged among development-finance experts that small-scale and very small-scale entrepreneurs in the developing countries were the ideal target group, and NGOs set up and operated by local business people were the ideal organizational form for a sponsoring organization, thereby creating the perfect match on the development-policy scene. Since the NGOs were business-oriented and also socially committed and had as little to do with the state as did their customers, this constellation fitted almost perfectly into the political landscape of the time. A number of international development agencies took up this idea and sponsored several of these new credit-granting NGOs.
It is important to distinguish between two types of MFIs.6 One type consists of those which were concerned about limiting costs and increasing revenue in order to eventually reach a level of efficiency that would make them financially self-supporting. In principle, this commercial orientation would imply the requirement to contain costs by focusing on credit as the sole service provided and increase revenue by charging higher, and possibly even cost-covering, interest rates for loans.
The policy of MFIs of the second type consists of providing many kinds of services that poor people might need, charging lower interest rates on microloans and being less stringent with respect to repayment. For outside observers, this softer approach had some appeal because it appeared to be more client-friendly and more in line with the ability of poor clients to bear the full cost of borrowing.
In 1992, the Inter-American Development Bank commissioned a study of the efficiency of credit-granting NGOs covering MFIs of these two types. The findings of this study, which were published in 1996, were quite sobering.7 The yearly deficits of those MFIs that followed the soft approach were horrendous. But even those MFIs that had some aspirations to break even on their credit operations were far from reaching this objective. The conclusion was straight-forward: with the high levels of costs reported in the study, no MFIs would be able to survive and would be permanently dependent on subsidies. However, shifting their high costs, which were in the range of almost 100%, to customers would have been equally impossible for economic, political and ethical reasons.
Some international donor agencies required a much higher level of efficiency from the MFIs they proposed to support. Given this pressure, some credit-granting MFIs started to undertake serious efforts to reduce costs and become efficient.
If microfinance had to be as expensive as it seemed at the time, it would not be a suitable instrument either for alleviating poverty or for creating jobs and stimulating growth. Over the long term, the organizations in question could not have created even an illusion of greater social equality.
In the form in which it was carried out at the time, microfinance could only be political window-dressing, and advocating it as a means of development policy – and spending donor money on it which could have been allocated to other, more effective development-aid measures – appeared ethically questionable.
These findings inspired some international donor agencies to require a much higher level of efficiency from the MFIs they proposed to support. Given this pressure, some credit-granting MFIs started to undertake serious efforts to reduce costs and become efficient. When he learned about the results of this study, J. D. Von Pischke, one of the leading experts on development finance, suggested a goal which at the time seemed utopian: the sum of administrative and risk costs of a good MFI should not amount to more than 20% of its loan portfolio. Costs of this magnitude would be low enough to be passed on to their customers, thus allowing MFIs to cover their costs and expand the scope of their activity in accordance with their dual goals of sustainability and increasing outreach. Those who accepted the challenge, certainly a minority among the MFIs in existence in the early 1990s, adopted the new label of being commercial MFIs.
The old debate about the ethics of commercial microfinance
Commercial microfinance and its critics
The second half of the 1990s saw dramatic developments in microfinance. This was thanks to a series of innovations in credit technology8 and organizational structure of MFIs and of the types of development projects that served to create new microfinance institutions and programs.9 The best MFIs were able to reduce their costs to a level of about 20% of their loan portfolios. A few even managed to cover their full costs entirely through current revenues.
Naturally, there were serious debates regarding the best way of achieving this ambitious aim. Two main approaches, as they were referred to in the technical jargon, crystallized out of these debates. One is the institution- building approach, which considers creating viable institutions that can achieve sound performance as the key issue and a prerequisite for financial and developmental success. The main protagonist of this approach was the German consulting firm Internationale Projekt Consult (IPC). The other approach was called the commercial approach. Among experts, it was mainly associated with ACCION, a United States-based microfinance support organization. But the labels do not mean much since there is hardly any contradiction between the two approaches: financially viable MFIs must also be commercially oriented, and if they are to achieve the commercial success necessary to make a lasting impact, they must have a suitable institutional form. I therefore continue to use the more widespread term commercial approach for both of them.
Towards the end of the last century, the commercial approach had clearly won out over the more traditional well-intentioned but inefficient softer ways of doing microfinance. However, it would be a mistake to assume that the views shared by IPC and ACCION and most other leading practitioners and academics in microfinance were universally accepted. The key figures of many MFIs, including Yunus, as well as many experts working in development-aid institutions or academia, had strong reservations vis-í -vis the new approach, or were even openly hostile to it.10 In their view, trying to achieve profits through development projects was morally reprehensible a priori. They even did not consider covering costs as an appropriate goal, because in their eyes the work of MFIs was socially and politically so important that the industrialized countries were morally obliged to finance their activities and to cover the deficits that developmentally respectable MFIs would necessarily incur. This criticism was ethically motivated, but as is shown now, it is not justified on the basis of facts and figures.
Using simple numbers to illustrate the conflict
The conflict of the 1990s regarding the ethical value of the commercial approach can be illustrated by comparing two hypothetical MFIs called MFI-1 and MFI-2, modeled after what may have been good institutions of their respective types for that era.
MFI-1 is commercially oriented, while MFI-2 is not. Both were founded four years earlier, using the same amount of development-aid funds, and both issue loans of an average size of US$1,000. In its fourth year of operations, MFI-1 has administrative and risk costs (per year) which together amount to 15% of its average loan portfolio. Including funding costs of about 10%, it has total costs of about 25%. Owners of small and very small businesses are offered loans at terms and conditions which correspond to an effective interest rate of 30% if one includes customers own transaction costs. MFI-1 is thus covering its costs, and as it has already been doing so for some time it also has the equity that is required for an expansion of its operations. It can also easily obtain additional loans from development-aid organizations which it may need for expansion. Thanks to its better access to funding, MFI-1 has 30,000 borrowers. However, its strict orientation aimed at covering costs implies that its management cannot always be nice to borrowers in arrears and employees who do not cooperate.
MFI-2 rejects the commercial approach on ethical grounds and operates accordingly. This is why it has higher total costs of about 40%. Its managers do not find it acceptable to pass these costs on to its customers. The effective interest costs for the customers are only 20%. But this is not enough to cover MFI-2’s costs. Therefore, it is necessary to find some development-aid organization willing to offset the deficit. This can be managed with a certain amount of effort and skill, but since the organizations which come into question to cover the deficit do not approve of the fact that MFI-2 is still failing to cover its full costs, they cannot be persuaded to provide the additional funds that would make it possible to expand the level or the scope of activity. Therefore MFI-2 can only serve 10,000 customers.
I also make the following assumption that will make the evaluation easier: the clients of both institutions take out loans not out of high spirits but because they really need them for their businesses. Their only alternative is to borrow on the informal credit market where the effective interest rate on small loans is at least 100% per year, a realistic figure in the poorer parts of the world.
The ethical problem is now easily demonstrated: MFI-1 can at times be harsh, and it demands higher interest rates from its poor clients, but it can serve three times as many customers as MFI-2 by offering them loans at 30%, or in other words enables 20,000 more individuals to avoid taking out expensive informal loans that would cost 100%. MFI-2 is not as rigorous in dealing with its customers (and possibly its employees), and is thus a friendlier institution, but it serves far fewer customers.
The monetary value of the costs which the customers of a microfinance institution save by not taking out a loan on the informal market can be used as a very simple quantitative measure of the impact of its operations. If one multiplies the average loan size by the number of loans outstanding, the interest savings (100% minus interest paid, that is 30% or 20%) per dollar borrowed, yields a benefit for the clients of US$21 million in one single year. The corresponding calculation for MFI-2 yields a benefit of only US$8 million – and thus US$13 million less.
Being a strategic shareholder – or simply an owner – of an MFI requires more than merely putting up equity.
Other benefits offered by the two MFIs could also be calculated, and it would also make sense to add the net benefits over the course of several years. But for what I want to show, these refinements are not necessary. A decision-maker acting in the interest of the officially declared target group of the poor people and of small businesses would hardly find it difficult to decide which one of the two MFIs should be supported: MFI-1 is clearly preferable on both economic and ethical grounds. This simple comparison of the expected benefit to the borrowers is a clear ethical endorsement for the commercial approach: it is ethically superior simply because it generates a much stronger development impact.
New ways of creating microfinance institutions
By the middle of the 1990s, the economic and regulatory conditions for financing small and very small enterprises had changed. Most importantly, the upper limits for interest rates that banks could charge their clients had been removed in most countries. In view of the serious efficiency problems that most microfinance NGOs had encountered, microfinance experts started to think that it would be more effective and more efficient to have specialized microfinance banks instead of NGOs for providing microfinance services. The most important advantage of banks would be that they are permitted to take deposits. This would allow them to provide an additional service that their clients could be expected to value highly, and also to use client deposits as a source of funds for making small loans. Another advantage is that formal, licensed banks as specialized providers of microfinance services would be regulated as supervised institutions, and for this reason alone could be expected to be more efficient. But where would these banks come from? How could they be identified and possibly created?
One way of creating the desired kind of banks is a new type of development-finance project called upgrading. An upgrading project consists of three steps. The first step is identifying a credit-granting NGO whose founders or current leaders would be willing to make their NGO undergo a profound institutional transformation. As the second step, foreign aid is used to strengthen the NGO and turn it into a good credit- granting institution able to cover its full costs or at least well on its way towards full cost-coverage. The final and decisive step consists of changing the legal status of the NGO into that of a corporation, obtaining a banking license and then starting banking operations within the new legal structure. A consideration that had made some development finance experts expect a great deal from upgrading projects was the presumed importance of locally available knowledge and the active involvement of the key players in the former NGO.
Some upgrading projects were quite successful. But the number of success cases was smaller than had been expected,12 and in many cases it turned out to be very difficult to integrate the former NGO leaders into the new organizations since they were very attached to the old NGO model. Thus most of the presumed benefits of transforming existing NGOs into microfinance banks were unattainable.
Because upgrading was less successful than expected, a new type of development-finance project was developed. Instead of transforming an existing NGO into a bank for local small-business clients, aid agencies and their advisors started to build target-group oriented MFIs from scratch – that is, without a precursor institution. This approach was called the greenfield approach. By now, this approach has become the preferred way of creating new efficiency-oriented and also target-group oriented MFIs.
The crucial question of ownership
The creation of a microfinance bank in the legal form of a corporation, be it through the upgrading of a former NGO or through the founding of such a bank from scratch, became more or less a routine operation after the first cases had been successfully implemented. However, in all of these cases, there is one very important problem, which evidently entails an ethical aspect: who should be the owner or owners or shareholders of the new institution? One possible candidate in the case of an upgrading project could be the former NGO. But this would rarely be enough; additional shareholders would be needed. Experience also showed that the former NGO leaders would hardly be able to manage a bank, which is after all a much more complex enterprise than an NGO. The international donors who in most cases have supported the conversion or the creation of the new institution financially might be other candidates. But as it turned out in practice, they were neither able nor willing to take over the full responsibility for making the new bank become successful in both financial and developmental terms. Purely commercial investors were also not deemed suitable as owners of MFIs simply because it was feared that they would only focus on profitability and neglect the social and developmental role and mission of microfinance banks.
Being a strategic shareholder – or simply an owner – of an MFI requires more than merely putting up equity. A qualified owner must feel responsible for the institution and its success in two dimensions – both a financial and a developmental dimension, and act accordingly. The new MFI should of course be commercially successful because otherwise it cannot become a stable institution. However, its task is ultimately to contribute to the country’s economic development and to serve the less prosperous members of its society. Therefore, asking who is best suited to exercising ownership and control amounts to asking who would use his ownership rights to take decisions that fairly balance short- and medium- term financial goals and long-term development aims.
At the time when creating new MFIs had become the rule of the game, there was no good answer to the question of who could be competent, patient and socially responsible owners of commercially oriented MFIs. Interestingly, the two organizations mentioned above, IPC and ACCION, that had been the most fervent advocates of the commercial-cum- institution building approach for years and that had also been the most successful players in creating MFIs that are both financially viable and developmentally oriented, also were first aware of the problem that it is not enough to build up such institutions. For their continued stability and growth, they needed dedicated and competent strategic owners. And they finally assumed this role themselves. ACCION created a number of institutional and equity-supported links to the large number of MFIs in whose creation it had played a role, and IPC founded an equity participation company called ProCredit Holding-AG. The latter by now holds the major part of the equity of 22 small business banks in 22 countries and acts as the central coordinator of this group of banks.
Recent developments and excesses of commercial orientation
The apparent victory of the commercial approach as the starting point
The time between the late 1990s and the middle of the last decade were the golden years of microfinance. What had started as a domain of well- intentioned but hardly competent amateurs became more and more a domain of professionals. In parallel, microfinance also became more and more effective in financial and developmental terms. Many new MFIs were set up and managed in the spirit of the commercial approach, and a considerable number of them became quite successful in purely financial terms and at the same time started to have a sizable impact. The institutions and the networks of MFIs set up by ACCION and IPC/ProCredit served as models, and their best practice was copied by several imitators. When Yunus and his Grameen Bank received the Noble Peace Price in 2006, the new world of microfinance – to quote from the title of an influential book edited by ACCION14 – seemed to show that the commercial approach had scored an all-out victory in the battle of finding the best solution for an important problem in social and economic development.
It was part of the apparent success of commercial microfinance that commercially oriented investors and their advisors also started to be interested in microfinance as a new and interesting asset class.15 This interest was welcome because the loan portfolios of the new MFIs were growing fast at that time. Since most new MFIs are regulated banks, a growing portfolio requires a growing equity base. Lack of equity had become a constraint on growth and potential new investors might have offered a solution to this problem – as long as they only provided equity and did not try to upset the delicate balance between profit and development-orientation that the best MFIs had established for themselves.
Apparent victory can lay the seeds of defeat. Not only those who had for many years worked in microfinance for the purpose of mitigating development problems noticed that commercial microfinance can be profitable. New players started to invade the field, and they seemed to have widely different intentions from those of the traditional players. Who were the new players, and what did they do?
The dual face of commercial microfinance
The term commercialization has more than one meaning. Until now, I have only used it in the sense of doing microfinance with a keen interest in covering the full costs of running an MFI, and even making a moderate profit in order to attract new funds. These could constitute equity or loans, and thus eliminate the former dependency on subsidies and development-aid funds.
A second meaning of the term refers to the (possible) fact that purely commercial agents such as private commercial banks or other purely profit-oriented investors engage in business operations with poor people and small and very small enterprises, and that they do it in a way in line with their one-sided objectives. Since the middle of the last decade, commercialization in both senses of the term is taking place. The indisputable success of commercial microfinance in the first sense of the term commercial laid the foundation for commercialization in the second sense of the term, because it demonstrated to purely commercial players that the microfinance market can be a profitable market.
There had previously been some episodes of commercial actors invading a country’s microfinance market and operating there in ways that not only did not solve poor people’s problem of access to finance on reasonable terms, but instead created problems of over-indebtedness. But these were rare events and their negative effects were soon overcome. However, since the middle of the last decade, this new kind of commercialization has become a widespread phenomenon. So-called microfinance in Mexico is a well-known example. As the American business magazine Business Week reported several times in recent years (one of which under the title the ugly side of micro-lending),16 some large international banks have recently engaged in microfinance in Mexico. Unfortunately, it is necessary to add, elsewhere as well. Although they call their lending operations microfinance, this business has little to do with financing small and very small enterprises. Instead, it is simple consumer lending. In general, some scepticism is justified as to whether it is ethically defensible to finance consumption on the part of poor people, all the more so when such financing is provided under a misleading label. When the supply of consumer loans is targeted at poor people it constitutes irresponsible lending if, as reported by Business Week, it occurs with usurious interest rates. The gradual transformation of microfinance from microenterprise financing to consumer lending has started to tarnish the reputation of microfinance, and it is questionable from an ethical standpoint.
The Compartamos IPO of 2007
Unfortunately, the invasion by some commercial banks into the domain of microfinance is not the only recent development that casts a shadow over microfinance and raises ethical concerns. There is also the growing tendency of some genuine MFIs to adopt what I call excessive commercialization.
By issuing shares to the public and having their shares listed on a stock exchange to allow trading, an MFI tries to attract private, profit-oriented investors. A successful initial public offering (IPO) – to use financial terminology – of an MFI indicates that microfinance has reached a high degree of maturity since it requires that the MFI and the investment bankers supporting it should be able to convince new investors that their investment will pay off. As investors should not be misled, an IPO also implies a commitment by the MFI that it will ensure that these shareholders will receive the returns they were led to expect.
Over the past few years, five MFIs have undertaken IPOs – three of them located in Asia, one in Africa and one in Mexico.17 I shall first discuss the IPO of the Mexican MFI Compartamos, a case that serves to highlight the ethical dimension of excessive commercialization.
In early 2007, 30% of the existing ordinary shares of Compartamos, a former NGO which had been converted into a joint stock corporation in 2000, were sold to American and Mexican private and institutional investors and the shares were listed on the Mexico City Stock Exchange. Since no new shares were created in the course of this IPO, the MFI itself did not receive any new funds. Only those who had invested earlier were given the opportunity to sell out at a profit.
By issuing shares to the public and having their shares listed on a stock exchange to allow trading, an MFI tries to attract private, profit-oriented investors
The IPO was extremely successful in financial terms. The issue price of Compartamos shares was 13 times higher than their book value,18 which corresponds to an extremely high price-earnings ratio.19 Assessed at the offering price, Compartamos was worth about US$1.5 billion, although at that time it was still a rather small institution.20 Despite the high issue price, the issue was oversubscribed by a factor of 13. After the issue, the price rose once again, by about 50%.
Among those who sold shares in the course of the IPO were ACCION and the development finance institution IFC, a subsidiary of the World Bank, as well as several private shareholders drawn from among the founders and managers of Compartamos. They had all bought their shares at their nominal value in 2000, when Compartamos was converted from an NGO into a corporation, and their investment turned out to be very lucrative. Over a period of seven years, between 2000 to, and including, 2006, the value of their investment had doubled each year.
I think that the financial aspects of this IPO should be a cause of concern. However, this should not be based on the mere fact that shares in an MFI were sold to a broad range of investors or that the IPO was extremely profitable for the investors. Both features of the IPO have the positive effect of increasing the reputation of microfinance as an investment opportunity, making it easier for other MFIs to access the capital market and use it as a source of much-needed equity capital.
There is cause for concern, however, over the reason why Compartamos shares became so valuable. In this case, it was the extremely high profits earned over the time span since the conversion from an NGO into a corporation. The high issue price can only be explained by the expectation of the purchasers of the shares that this stunning level of profitability would be maintained over the coming years and the implicit promise of the initial owners and managers of Compartamos to maintain this exorbitant level of profits would be kept.
The past profits of Compartamos were the result neither of low-operating nor of low-funding costs. Rather, it was its pricing policy which consisted of charging an average interest rate of close to 100% on its loans. There had been an outburst of inflation in Mexico in the late 1990s. When inflation shot up to about 100% Compartamos did what every good MFI is supposed to do. It raised the interest rates on its loans accordingly so that the inflation-adjusted interest rates remained positive. However, very soon the inflation rate in Mexico fell back to its normal level. Nevertheless, Compartamos maintained its high interest rates, which suggests that its interest-rate policy may have been chosen with an eye to the planned IPO.
Real interest rates of almost 100% would constitute a political and ethical problem at any financial institution. They are all the more problematic at an institution which presents itself as a development-oriented MFI, as Compartamos has done until now. In my view, as well as that of many other observers, the excessive profits are the result of plain and simple exploitation of customers.21 This leads to the conclusion that at least until 2007 the commercial orientation of Compartamos was simply excessive and incompatible with the ethical mandate of microfinance – constituting a major ethical problem.
Was there reason to expect that the exploitative pricing policy of Compartamos would change after the IPO and become more consistent with what constitutes best practice among commercially oriented but also ethically motivated microfinance institutions? This was unlikely from the start because roughly half the shares were sold to hedge funds in the course of the IPO.22 Hedge funds are not the kind of shareholders likely to show a strong commitment to social and development-related aims. Transferring power to them by issuing ordinary shares with full voting rights amounted to definitively losing power and renouncing the original developmental aspirations.
The Compartamos IPO led to a wave of very critical comments from within the microfinance community. One of the harshest comments came from Muhammad Yunus. As he argued, microfinance had been started to combat loan sharks and not to create new ones. One might be inclined to reject Yunus’s criticism as inappropriate since he had never accepted the wholly commercial approach to microfinance. But he was not alone. Others like Rosenberg (2007), who had always been an advocate of the commercial approach, were equally appalled by the Compartamos IPO and the example it seemed to set for commercial microfinance, as well as its damage to the image of microfinance in general.
The IPO of SKS in 2010
Interestingly, though, the general public did not really become aware of the Compartamos IPO and its questionable aspects. It may simply not have had enough appeal for the media. However, this attitude did not prevail after the second major IPO by an MFI in India in 2010.
India was a latecomer to microfinance, despite its proximity to Bangladesh, the home of Grameen Bank, ASA and BRAC, three of the world’s largest and most well-known MFIs. One reason for this was that government-owned banks and government-related programs had dominated the microfinance scene in India for many years. Therefore private MFIs, and especially those with a commercial orientation, had a very slow start. But once it began, microcredit almost exploded in India. Between 2006 and 2010, the number of borrower-clients of commercially oriented MFIs grew from 8 million to 28 million, and the volume of outstanding loans grew by a factor of 8. A year previously, India, and especially the State of Andhra Pradesh in South India, had become the home to some of the world’s fastest-growing and thus ultimately also the largest MFIs. Among them was SKS Microfinance.
Two events took place in this part of India in 2010 that led to a situation termed by observers a major crisis (Wallstreet Journal) and even the death of a microfinance (Global Post, India). One of these was the IPO by SKS, and the other was a series of about 80 suicides that were linked to repayment problems of microfinance borrowers and the pressure that seems to have been applied by MFIs debt-collection agents.
SKS is a large strictly commercially oriented MFI headquartered in Hyderabad in the State of Andhra Pradesh. It operates in that state and several neighboring states of South India. It was founded as an NGO in 1997 by Vikram Akula, an entrepreneur who had worked as a consultant in the United States. In 2003, SKS was transformed into a Private Limited Company – that is, a for-profit institution, and converted into a Public Limited Company in 2009. Over the years, the ownership of SKS has changed dramatically. It was mainly owned by its clients until 2003. Then before and after the IPO most of the shares were held by American private-equity companies. Measured by the size of its loan portfolio at the time of the IPO, SKS was about twice as large as Compartamos. Shortly before the IPO, Akula and other top managers sold their shares, cashing in substantial profits.
On 28 August 2010, 23% of the SKS shares were issued to the general public, including more institutional investors. In financial terms, the IPO was at least as successful as that of Compartamos. On the basis of the issue price, SKS had a total market value of around US$1.5 billion, almost exactly the same as Compartamos at the time of its IPO, and the issue was also 13 times oversubscribed.
Like Compartamos, SKS was, and still is, an institution that only grants loans and does not take deposits. Equity financing comes mainly from institutional investors of the kind mentioned above, and the bulk of SKS’s funding is in the form of debt from Indian banks. There is a legal requirement in India that banks must provide a certain fraction of their loans to low-income borrowers. However, they can meet this requirement if they lend money to microfinance institutions; and that is what most Indian banks do. Therefore, for a long time funding was not a problem for the major MFIs in India despite their enormous growth rates. In the case of SKS, the average annual growth rate of the portfolio was more than 160% during the five years preceding the IPO.
The high issue price of the SKS shares and the high price-earnings ratio of over 40 are hard to explain. Because of heavy competition from other MFIs, and possibly also because SKS management wanted to avoid incurring the criticism that Compartamos had attracted three years earlier, the interest rates charged by SKS were lower and within the normal range obtaining in good MFIs. Thus the claim that it overcharged clients, the main thrust of the criticism against Compartamos, does not apply to SKS. Its return on equity and return on assets were also not particularly high. Given its moderate profits, it seems that those who bought the shares at a very high price must have been convinced that SKS would be able to maintain the enormous growth rates of client numbers, loan volumes and staff it had achieved in the five years before the IPO. Another possibility is that they simply overvalued the new shares offered to them, as Chen et al. (2010) also believe.
The Indian microfinance crisis
As I have said, recent growth rates of microcredit in South India were enormous, and a number of fast-growing MFIs competed fiercely for market share. Largely similar to the subprime crisis in the United States, this situation incited negligence in the credit assessments undertaken by loan officers and a policy of unrestricted credit expansion on the part of MFI management. As a consequence, poor borrowers felt encouraged to take out loans from several MFIs. Multiple borrowing was widespread, leading to clients inability to repay their loans. With surging repayment problems, debt-collection practices became increasingly ruthless.
As has been extensively reported in the press worldwide, a growing number of borrowers were unable to stand the pressure exerted on them by debt collectors working for SKS and other MFIs, and 80 people committed suicide, allegedly on account of this pressure. Of course, it is never possible to know the true reason for a suicide in any given case, and even more difficult to understand what it means in a culture like that of South India. But it seems plausible to assume that these sad cases were related to over-lending and over-borrowing. For the affected families it was a tragedy, and for microfinance, it was certainly also an ethical disaster.
It was also a disaster from a political and commercial perspective for Indian MFIs. Politicians in Andhra Pradesh and some other states reacted fast by recommending that MFI clients stop repaying their loans. They also introduced legal means to stop the questionable operations of debt collectors and those responsible for granting new MFI loans. Repayment rates fell drastically, as did the flow of new loans. In simple words, the microloan market came to a stop. Also the stock price of SKS dropped to about one third of the issue price only one year earlier.
Looking at events in the heartland of Indian microfinance, one might be inclined to think that the crisis in Andhra Pradesh was also an all-India microfinance crisis and might even signal the beginning of a similar crisis in other countries. Certainly, practitioners and observers were now alerted to the danger of over-lending and multiple lending by MFIs, as well as of MFI loans that were merely consumer loans as opposed to micro-business loans that would help people generate more income. However, the Indian crisis has so far not developed into a general microfinance crisis. It is restricted to a few national microfinance markets or regions, and even there it mainly affects the operations of only a few MFIs.
However, even though it is premature to diagnose a general microfinance crisis, as some commentators do, the damage is done. Apart from the fact that it means great hardship for the people directly affected, the general effect of the events in South India is that microfinance has lost almost all its former ethical and political appeal. As Chuck Waterfield, an expert in microfinance and an advocate of the commercial approach puts it in a blog24 maintained by the well-known microfinance support organization CGAP, we run the risk of the world seeing no difference between microfinance and the moneylenders we set out to displace.
There can hardly be any question that many of the unethical practices that have so severely tarnished the former reputation of microfinance can be traced back to excessive commercialization. It is not surprising that the two MFIs that have undertaken IPOs which were spectacularly successful in purely financial terms and meant enormous gains for those who had set them up are at the center of the crisis. Milford Bateman, a British expert, paraphrases Shakespeare by writing something is rotten in the state of microfinance and blames the international-development community’s preferred model of microfinance, the commercial model for the current situation.25 As he sees it, hidden behind the focus upon extending outreach, was the awkward fact that a MFI’s senior managers were quietly turning the institutional gains into private gains taken out in the form of spectacular salaries, bonuses, dividends and, eventually, windfall profits arising from an IPO.
There can hardly be any question that many of the unethical practices that have so severely tarnished the former reputation of microfinance can be traced back to excessive commercialization.
Bateman’s remark was precisely aimed at those two MFIs at whose IPOs I have taken a closer look. Their common features are the heavy involvement of private equity companies and hedge funds as investors and the huge gains of those inside owners who cashed in during the course of the IPO. This raises the question of whether the involvement of this type of investor and the enrichment of the founders and top managers are related in some way. This seems to be the case. Or more generally, is an IPO necessarily the end of microfinance as we had come to know it, and of all of its moral aspirations, or is it only a problem of how an IPO is implemented? And can traditional commercial microfinance avoid running into the problems of multiple lending, over-borrowing and a loss of focus on development? I shall address these questions in the concluding section.
Conclusion and outlook
Are the negative effects of the capital market inevitable?
It is my impression that in both cases of MFI IPOs, people who may once have created MFIs with the noble intention of making a contribution to development, put the MFIs they had created into the hands of the capital market. Apparently they thereby lost control over the MFIs and became unable or unwilling to maintain the developmental orientation that they may once have had. More precisely, they put their MFIs into the hands of hedge funds and private equity companies – that is, of investors who presumably are only interested in profit and are indifferent to developmental and social effects and aspirations. Is what happened to these MFIs an unavoidable consequence of turning to the capital market? And is undertaking an IPO necessarily the end of ethically motivated microfinance for the MFIs in question, thus precluding the option of using the stock market to attract more equity and thereby to increase the possible positive effects that microfinance can have?
Using the example of the IPO of an MFI that had been contemplated but because of the general financial crisis has so far not been executed, it is possible, I hope, at least to sketch an answer. The case I have in mind is that of ProCredit Holding-AG, or for short PCH. PCH is the German holding company of a group of 22 small business banks in 22 countries already mentioned above. I am familiar with this case because until not long ago I was a member of PCH’s supervisory board.
The holding company and the MFIs under its umbrella have in the past consistently followed the commercial approach, and as a group they have been moderately profitable for a long time. They are also keenly aware of their development policy mandate and try to adhere to it as much as possible. PCH can therefore be regarded as successful in both regards. Since the middle of the past decade, PCH had considered undertaking an IPO in order to raise new equity, which it urgently needed to support the lending business of its fast-growing affiliated small business banks. When the financial crisis hit in 2007, preparations were quite advanced. However, the crisis has completely stopped all IPO activity in Western Europe and America. Therefore the plans had to be put on hold, at least for the time being.
For PCH, the financial success of the Compartamos IPO had ambiguous implications. On the one hand it was considered encouraging because it showed that the capital market is receptive to MFI shares. On the other hand, the managers and owners of PCH feared that after the controversy surrounding the Compartamos IPO, socially and developmentally oriented investors, with whom PCH had planned to place a part of its shares, would turn away from microfinance entirely as they might no longer trust commercially oriented MFIs to maintain their commitment to development aims once they had submitted themselves to the dictates of the capital markets. Moreover, the managers and current owners may have also feared that their organization and possibly they themselves as people could change in the way that appeared to have happened at Compartamos – and also at SKS – if they took PCH public, or even if they only seriously considered doing so.
Could PCH do something to prevent this from happening? There is a positive answer to this question. However, this answer requires a thorough understanding of the problems that have to be solved. What has happened at Compartamos – and possibly also at SKS – could have been expected in principle, because at each point in time the exact thing happened that could have been expected, given the incentives for the participants involved and the options available to them at all points in time.
But that this mechanism is effective is also predictable in principle and as a principle/: the incentives and opportunities that will be encountered at a later point in time can be anticipated. But they can also be structured in advance. One can try to create binding commitments in advance that will take effect later if one knows in advance that one does not wish to act in accordance with the incentives and opportunities that will exist later. However, this is not an easy matter, and it has economic costs.
Thomas Schelling analyzed the principle of self-commitment years ago in a masterly way.27 It can be applied directly to the case of the IPO of a commercially oriented MFI like PCH that also has a social and developmental objective and wants to retain it after an IPO. Before the IPO, the existing owners can create a binding commitment regarding the orientation that their MFI will follow after the IPO by signing a shareholder agreement as a contract between each other. For example, they can agree that the level of profit earned by their MFI may not exceed a specific limit. Having such an agreement in place would limit the incentive to later shift to providing consumer loans or similar products of limited social value in the face of the pressure of the capital market. Of course, it is necessary to ensure that this commitment cannot easily be revised. Therefore, the current owners of the MFI must find a way of giving their agreement a binding or constitutional status.
But wouldn’t the new distribution of voting rights among the owners automatically change once the IPO had taken place so that the contract would become ineffective? The answer depends on whether new shareholders would receive voting rights or not. Non-voting shares exist in almost every legal system. In order to make their commitment irreversible, the existing owners of PCH, who wanted to preserve the orientation of their MFI, planned to agree among themselves to issue only non-voting shares in the envisioned IPO, thus leaving the existing distribution of decision rights intact and ensuring that the existing development orientation of PCH would remain in place. This naturally comes at a price, and one which may even be very high, because non-voting shares cannot be issued at the same price as voting shares. It might even be feared that non-voting shares could not be placed at all. The PCH owners were willing to take this risk. However, the good lesson of the Compartamos IPO which is relevant here is that the capital market seems to have a considerable appetite for MFI shares. It could thus be expected that the market would take up even non-voting shares in an MFI with a long-term self-imposed commitment to development goals. This is in essence what the present owners and managers of PCH had planned to do before the crisis derailed their plans to issue shares to the general public. With such a concept, PCH can still today try, and indeed hope, to maintain its identity and its development orientation which is important to the owners, while at the same time taking advantage of the capital market.
In concluding, I venture to put forth the optimistic proposition that as long as an MFI does not approach the dangers of the capital market in a blind and naí¯ve manner, the commercial orientation need not lead to a situation in which it will be induced to maximize profit at the cost of all ethical and development aims.
The future role of microfinance
In mid-2010, a number of new research studies were published which claimed that the effectiveness of microfinance as a means of overcoming poverty had been greatly overestimated in the past. This assessment is most probably correct. However, it is by no means new. Among many others, Jonathan Morduch had made this point before and argued that microfinance – at least that of the commercially oriented kind, is a means of generating employment and stabilizing the lower-middle-income group of a developing country rather than an instrument to combat poverty (Morduch 2000). After all, the typical MFI client is the local baker in a small town and not his poor neighbor who might work in the bakery from time to time. The local baker is a member of the local lower middle class. He is poor by Western standards but not by local standards. Really poor people do not need loans, which inevitably put a debt burden on their shoulders, but rather other services such as access to clean water and health care for which they would have to pay little or even nothing.
But why should it be important to improve the baker’s access to credit through better financial services? There are three reasons. One is that the baker as a borrower offers certain prospects of creating income and employment for himself and his family and also for others. Thus at least an element of the old trickle-down concept needs to be resuscitated in thinking about microfinance or – as it should rather be called in order to avoid misleading expectations – small-business finance. There is sufficient evidence that small-business finance can have positive effects on income and employment.
Second, the recent events in India and some other places have taught the bitter lesson that really poor MFI clients can hardly be prevented from also taking out loans from other MFIs and from private moneylenders. Therefore, it is almost impossible to avoid multiple borrowing and over- indebtedness if a development-finance institution focuses on granting credit to almost any poor person without regard for how the borrowed funds will be employed. In the case of lending to the baker, there is at least a certain possibility of monitoring the use of funds and avoiding over-lending and multiple borrowing.
The third argument for lending to clients who belong to the lower local middle class rather than to really poor people is that small business owners might be more interested in demanding and also supporting democracy, a well-functioning legal system and an efficient bureaucracy and that they might be more inclined and better able to voice their interests. Thus successful lending to small business owners can have the positive effect of helping to establish and stabilize well-organized societies and thereby indirectly also benefitting other groups of the population.
What does all of this imply for the future of microfinance? Many or even most of the existing MFIs will have to turn into small business banks. Of course, these banks must be stable and profitable institutions, and this requires that they continue operating in the spirit of the commercial approach. The crucial ethical challenge for them consists of finding ways of committing themselves to supporting those parts of a local society and economy that so far have insufficient access to finance. Finance for the really poor may not have much of a future, at least as far as the existing larger and commercially oriented MFIs are concerned. The idea of having MFIs that mainly cater to the really poor people may in the end prove to be an illusion. And for the really poor, new solutions need to be devised by the next generation of ethically motivated and innovative experts and social entrepreneurs.
ACCION. 2007. The Banco Compartamos Initial Public Offering. ACCION InSight 23.
Bateman, M. 2011. “Over-Indebtedness and Market Forces.” CGAP Microfinance Blog, February 11.
Chen, G., S. Rasmussen, X. Reille and D. Rozas. 2010. “Indian Microfinance Goes Public: The SKS Initial Public Offering.” CGAP Focus Note 65.
DB-Research. 2007. Microfinance: An Emerging Investment Opportunity. Frankfurt: Deutsche Bank.
Easterly, W. 2001. The Elusive Quest for Growth: Economists’ Adventures and Misadventures in the Tropics. Cambridge, MA: MIT Press.
Hulme, D., and P. Mosley. 1996. Finance against Poverty. London, New York: Routledge.
Lieberman, I., A. Anderson, Z. Grafe, B. Campbell and D. Kopf. 2007. Microfinance and Capital Markets: The Initial Listings/Private Offering of Four Leading Institutions. Calmeadow and the Microfinance Equity Fund.
Morduch, J. 1999.“The Microfinance Promise.” Journal of Economic Perspectives 37: 1569–1614.
Morduch, J. 2000. “The Microfinance Schism.” World Development 28: 617–629.
Nair, A., and J. D. Von Pischke. 2007. “Commercial Banks and Financial Access.” In Building Inclusive Financial Systems: A Framework for Financial Access, edited by M. S. Barr et al., 89–116. Washington, DC: Brookings.
Otero, M., and E. Rhyne (eds.). 1994. The New World of Microfinance. Bloomfield, CT: Kumarian Press.
Rosenberg, R. 2007. “CGAP Reflections on the Compartamos Initial Public Offering: A Case Study on Microfinance Interest Rates and Profits.” CGAP Focus Note 42.
Schelling, T. C. 1960. The Strategy of Conflict. Cambridge, MA: Harvard University Press.
Schelling, T. C. 1984. Choice and Consequences. Cambridge, MA: Harvard University Press.
Schmidt, R. H. 2010. “Microfinance, Commercialization and Ethics.” Poverty and Public Policy 2: 99–137.
Schmidt, R. H., and C.-P. Zeitinger. 1996. “The Efficiency of Credit-Granting NGOs.” Savings and Development 20: 353–385.
Waterfield, C. 2010. “Has the Microfinance Market Come of Age?” CGAP Microfinance Blog, October 18.
Woller, G., C. Dunford and W. Woodworth. 1999. “Where to Microfinance?” International Journal of Economic Development 1 (1).
Yunus, M. 2006. “Poverty is a Threat to Peace, Nobel Peace Prize acceptance speech by M. Yunus, given on 10.12.2006,” published as an Epilogue in: M. Yunus. 2007. Creating a World Without Poverty. New York: Public Affairs.
- The author holds the chair of international banking and finance at Goethe University in Frankfurt, Germany. His email address is firstname.lastname@example.org. This paper is an updated and greatly shortened version of an article published in the e-journal Poverty and Public Policy 2, 2010.
- As Yunus rightly pointed out in his acceptance speech for the Nobel Prize; see Yunus (2006, 285)
- Scepticism concerning the role of microfinance in poverty alleviation is expressed, among others, by Morduch (1999).
- See also Easterly (2001) for a cogent critique of this lack of a theoretical basis of this policy.
- One of the few exceptions that managed to survive and even to grow considerably due to Yunus’s charismatic leadership and fund-raising skills was the now famous Grameen Bank.
- See also Morduch (2000), who vividly describes what he calls the “microfinance schism.”
- See Schmidt and Zeitinger (1996).
- The debate at that time was about the merits of three technologies. These were granting loans to small groups, called group lending, to large groups, called village banking, and lending to individual small businesses owners and other poor people.
- The focus of this debate was about the relative merits of three types of development projects in microfinance called downscaling, upgrading and greenfield projects; for details see section three below.
- The critical positions of Hulme and Mosley (1996) and Woller et al. (1999) are particularly outspoken and also noteworthy since both teams of authors were the editors of the two leading journals in this field, the Journal of International Development and the Journal of Microfinance, in the 1990s. Woller and his co-authors explicitly attack the commercial approach. They expressed serious “concerns about the direction in which [the proponents of the commercial approach] are attempting to push the industry,” and mention the Grameen Bank as the “most prominent example” of the only type of MFI which they consider appropriate from a developmental and ethical standpoint.
- For a more explicit calculation, see Schmidt (2010).
- In 2007, Nair and Von Pischke made the same point in looking back at the upgrading strategy of the past ten years.
- Information about ACCION, IPC and ProCredit Holding and the networks of microfinance banks affiliated with ACCION and ProCredit Holding can be found on their respective homepages. ProCredit now no longer calls the banks in its network microfinance banks but rather small-business banks.
- See Otero and Rhyne (1994)
- See DB-Research (2007).
- See Business Week (2007). The article, which is available on the Internet, appeared on pp. 38–46 of the Dec. 24, 2007 U.S. edition of Business Week.
- Four of these cases are compared and analysed in Lieberman et al. (2007); the Indian case occurred later.
- All the figures given here are taken from the careful assessment of the Compartamos IPO by Richard Rosenberg, a CGAP staff member; cf. Rosenberg (2007). The equally extensive documentation of the IPO published by ACCION (2007) arrives at very different conclusions from those of Rosenberg, but provides no reason to doubt Rosenberg’s figures.
- The price-earnings ratio of the Compartamos shares at the time of the IPO was close to 25. Generally, price-to-book ratios of listed banks are in the range of 1 to 4, and their price- earnings ratios are between 10 and 20.
- At the end of the year that preceded the IPO, the credit portfolio of Compartamos was only slightly more than half as large as that of the Grameen Bank and only one-eighth the size of that of the ProCredit group. However, the Compartamos’ loan portfolio grew faster than those of the other two institutions.
- A simple calculation presented in Schmidt (2010) shows that the high value of Compartamos at the time of its IPO was essentially a reflection of the wealth transfer achieved by keeping interest rates high after the inflation spike.
- This information is provided in ACCION (2007, 10) and confirmed in Lieberman et al. (2007, 32).
- For details and especially the figures used in this section, see Chen et al., 2010.
- This blog can be found at http://microfinance.cgap.org/tag/chuck-waterfield/. The entry dates from Oct. 18, 2010
- This statement, dated from 11 Feb. 2011, is also found at http://microfinance.cgap.org/author/Milford-Bateman/.
- The information concerning the plans of PCH had, for understandable reasons, been kept largely secret for a long time. However, attentive readers can now find it in remarks on the homepage of PCH.
- For the basic argument, see Schelling (1960). Schelling received the Nobel Prize in Economics in 2005 primarily for this work. In a later book (Schelling 1984, 57), he explained his principle of pre-commitment using the example of Odysseus who wanted to hear the enchanting song of the Sirens, yet knowing that, once he had heard it, i.e. later, he would have only one wish—namely to go straight to them—and would die on the cliffs of their island.