Assessing Microfinance: The Bosnia and Herzegovina Case
Anne Welle-Strand Kristian Kjollesdal Nick Sitter
Microfinance is often hailed both as a tool for fighting poverty and as a tool for post-conflict reconciliation. This paper explores the use of mi-crofinance in post-civil war Bosnia and Herzegovina, assessing its results in terms of both goals. As it combined high unemployment with a highly educated population in an institutionally open context, Bosnia and Herzegovina provides a crucial test of the effect of microfinance. If unambiguous signs of success cannot be found in a case with such favorable conditions, this would raise serious questions about the potential benefits of microfinance. The paper draws together evidence from a series of independent reviews of microfinance in Bosnia and Herzegovina, to assess its impact in terms of economic performance, the economic system, social welfare and post-conflict integration. Based on this case study, microfinance appears a better tool for dealing with poverty than with social integration or institution building.
Key Words: micro finance, post-conflict, poverty alleviation, economic development, Bosnia and Herzegovina jEL Classification: g21, oi
Introduction
Microfinance is often hailed both as a tool for fighting poverty and as a tool for post-conflict reconciliation. Since the 2000s microfinance has increasingly been employed as a means for poverty-reduction. Its international profile as a tool for poverty alleviation was secured in 2006, when Muhammad Yunus and Grameen Bank were awarded the Nobel Peace Prize. Yet, since the 1970s, microfinance has also been seen as part of a post-conflict reconstruction strategy. It is in recent years that the strategy has been more frequently used (Nagarajan and McNulty 2004). In
Dr Anne Welle-Strand is a Professor at the Norwegian School
of Management bi, Norway.
Kristian Kjollesdal is a reseacher at the Comte Analysebyrä, Norway.
Dr Nick Sitter is a Professor at the Norwegian School of Management
b i, Norway, and Central European University, Hungary
Managing Global Transitions 8 (2): 145-166
Bosnia and Herzegovina (hereinafter bh), both approaches of microfi-nance were used after the 1992-1995 war left the country in ruin.
The bh microfinance sector developed rapidly, transforming microfi-nance institutions (mfis) from donor-funded institutions into financially-sustainable microcredit organizations (mcos). The sector gradually became institutionalized. Many commentators have therefore proclaimed a case of the bh microfinance sector as successful. However, there are also voices raised against such evaluation, due to the country's stagnation in developing the small and medium enterprise (sme) sector (Bateman various years).
Microfinance is extensively used as a development tool, and is largely supported by existing financial, technical and political resources. However, less is known about microfinance in post-conflict situations. For instance, Woodworth (2006) argues that more research is needed on the complexities of managing mfis in times of conflict and post-conflict, while Nagarajan and McNulty (2004) find that implementing effective microfinance in specific post-conflict contexts is not yet well understood.
These and similar studies suggest that while post-conflict microfi-nance strategies may indeed resemble normal strategies, there are also important differences - for example the trust-building aspect of micro-finance emphasized by Doyle (1998). However, we can go even further in exploring the relationship between microfinance and conflict. For example, we can explore the relationship between the social environment and microfinance in order to discover what impact microfinance has on the post-conflict environments. Such an inquiry would be in line with the general approach applied in the microfinance research field, which is mainly preoccupied with scrutinizing the social and financial impact of microfinance. In the present study, evaluation of the financial impact is carried out in terms of: (i) the immediate effect on profit, business start-ups and economic performance, as well as (ii) the effect on the development of the broader economic system including institutions and the rule of law. In terms of its social impact, the focus is on (iii) the welfare and social equality of the users of microfinance, as well as (iv) on post-conflict reconciliation and integration.
bh represents a key test case for this type of inquiry because it mirrors a broader debate about the impact of microfinance - a divided opinion among scholars and practitioners about the impact and contribution of microfinance to the country's overall development. Therefore, this report has three main objectives: first, it will provide a brief overview of
microfinance as a tool for poverty reduction and as used in post-conflict situations, as well as a critique of microfinance. Second, it will outline a case of the bh microfinance sector - its establishment, development, current trends and impact. Third, it will identify what we can learn from this case as well as offer avenues of further research, focusing on issues of evaluation.
Microfinance, Poverty and Post-Conflict Reconciliation
Microfinance (mf) is a development tool designed to address issues of poverty, under-development and marginalization. It is based on a simple idea: to provide poor individuals with access to microloans. These microloans allow a client to start a micro-business. In the best-case scenario, such micro-businesses develop into a small-medium enterprise. Such mfis are often supported by government or international donor funding. Over time, these institutions aim at commercializing their operations and achieving their own financial self-sustainability. As a development tool, microfinance has been used both to reduce the poverty and to support countries recovering from a conflict or a major disaster.
mf is seen as an important factor in reaching the Millennium Development Goals (Littlefield, Murduch and Hashemi 2003). Donor funding provided to mfis usually includes poverty reduction in their mission. Based on a review of impact studies that took place in the period 1994-2002, Littlefield et al. (2003) find that microfinance goes beyond just business loans - it affects investments in health and education, management of household requirements, and other cash needs.
For example, since 1989 'Freedom from Hunger' has worked with local partners to develop and distribute a cost-effective strategy to improve the nutritional status and food security of poor households in rural areas of Africa, Latin America and Asia. MkNelly and Dunford conducted an impact evaluation study of these programs in Ghana in 1998 and Bolivia (CRECER - Credit with Education Program) in 1999. In Ghana, clients' economic, social and health status were shown to improve due to micro-finance (MkNelly and Dunford 1998). In Bolivia, MkNelly and Dunford (1999) documented that microfinance led to improved nutritional and health status of the clients' families, as well as higher involvement in local government.
These studies point to what Boudreaux and Cowen (2008) named 'the micromagic of microcredit.' Boudreaux and Cowen (2008, 31) explain: 'With microcredit, life becomes more bearable and easier to manage. The
improvements may not show up as an explicit return on investment, but the benefits are very real.' According to Boudreaux and Cowen (2008), microcredit is an alternative to money lending, which is a traditional way of borrowing and lending money to the poor part of the population. As such, microcredit is a more humane way of providing access to credit for the poor. They compare the microcredit and moneylender approaches and suggest some advantages for the microcredit approach, such as easing debt burdens as well as formalizing and institutionalizing the business relationships. Compared to the traditional banks, on the other hand, the microcredit institutions lend to people who work also in the informal sector (Boudreaux and Cowen 2008). In short, microfi-nance initiatives reduce poverty, promote education of children, improve health and empower women.
However, because poverty and war are sometimes linked, microfi-nance is often used in post-conflict contexts. Conflicts cause degradation of both quality of life and the economic situation. They lead to dysfunctional states; in some cases, states practically even cease to exist. Many post-conflict environments suffer from a lack of financial and social capital, infrastructure and functioning relationships (Nagarajan 1999). Business activities are adversely affected by such instable macro-economic frameworks. Humanitarian aid projects are suitable for the immediate post-conflict period, helping the population overcome starvation and diseases. Microfinance can be a means for managing the transition from humanitarian relief to economic reconstruction and sustainable development (Seibel 2006; Hudon and Seibel 2007). It can be an integral part in 'jump-starting' a crippled economy, aiming to rebuild and recover local economies (Nagarajan 1999). It can also foster legitimate business activities by facilitating linkages between refugee and local markets, as well as undermining informal lending.
However, microfinance may also have a broader impact in terms of peace and reconciliation. For example Nagarajan (1999) emphasizes the possibility of MFis playing a role in the restoration of social capital when they provide long-term viable services. Doyle (1998) shows how microfinance initiatives in Rwanda helped local Hutu and Tutsi populations to overcome their differences following the civil war in the 1990s, and to find common ground in business development through microfi-nance. Generalizing about such effects may still be premature, but there are signs that microfinance might help restore and build social capital through creating a meeting point and facilitating business relations.
The main criticism of microfinance is that it may have an adverse effect on broader development. This applies particularly to post-communist states, because the legacy of the communist (the 'second world') may be very different from conditions in developing countries (the 'third world'). A 2007 issue of the Development and Transition^ addressed the current state of the private sector in post-communist Eastern Europe and asked how development agencies can strengthen their work in five areas, one of which was microfinance. A central argument focused on the inadequate attention paid to the local and global conditions and post-socialist legacies when designing 'blanket' reforms such as generic sme support, commercializing the microfinance sector and liberalizing the business environments (Hughes and Slay 2007).
A stronger critique asks whether microfinance can actually undermine medium-term economic development because it supports inefficient activities. This relates specifically to commercial microfinance. Bateman (2007a) criticized 'new wave commercial microfinance' institutions that avoid international and government financial support and supervision, while extending microcredit to many poor individuals and communities. These commercial mfis tend to support microenterprises that operate well below the minimum efficient scale and have little realistic chance of long-run survival. This, Bateman argues, has two negative consequences: first, a high rate of microenterprise exit caused by the saturation phenomenon within the informal sector; and second, high opportunity costs for the countries, since a standard commercial microfinance business model does not make it possible for microenterprise to deploy advanced technologies, skills and product and process innovations. Bate-man found that there is little solid evidence to confirm that commercial microfinance facilitates sustainable economic and social development. Reviewing the data from Central and Eastern Europe, Galusek (2007) found that mfis in the region had served over 4 million clients (predominantly low-income families and microenterprises) by the end of 2005, but that although there was rapid growth in microfinance in the region, there was no evidence that mfis are successful in reaching the low-income groups on a large scale (Galusek 2007). He emphasized that a commercialized approach of mfis is a consequence of the donor community that encouraged such a model, and called for microfinance strategies and products to respond to the needs of the region.
In a recent paper Roodman and Morduch (2009) argue that many of the oft-cited positive impacts of microfinance originate in two studies
from Bangladesh, Pitt and Khandker (1998) and Khandker (2005), and are in fact based on faulty statistical methods. After replicating the analyses of the same datasets, with slightly different statistical techniques, Roodman and Morduch found that 'reverse or omitted-variable causation is driving the results, and that the endogenous credit-consumption relationship varies substantially by subsample, as well as borrower sex, which can explain the seeming gender differential in impact' (Roodman and Morduch 2009, 3). Their paper argues that, with few exceptions, the idea that microfinance effectively reduces poverty, that poverty-reduction is especially effective when the borrowers are female, and that the poorer the recipient the greater the benefit, are results of incorrectly applied statistical analyses rather than evidence of the true effect of mi-crofinance.
A broader debate concerns the issue of whether microfinance fits a country's strategy for economic growth. Calling for 'jobs, not microcredit,' Karnani (2007a) reviewed macroeconomic data and found that although microcredit yields some non-economic benefits, it does not significantly eradicate poverty. Even though microfinance can make life better at the 'bottom of the pyramid,' creating jobs and increasing worker productivity is a better way to get rid of poverty. Unless microfinance directly affects the jobless, it is merely a way of transforming employees into micro-entrepreneurs - simply by replacing old businesses with microcredit-funded micro-businesses. Such crowding-out results neither in net job nor in income gains (Storey 1994). Therefore, Karnani (2007b) suggested that romanticizing the poor as 'resilient and creative entrepreneurs' harms the same poor individuals in two ways: it under-emphasizes modern enterprises as well as legal, regulatory and social mechanisms to protect the poor; and it overemphasizes the impact of microcredit. Therefore, Karnani (2007a) suggested that governments, businesses and civil society should work together to reallocate their resources away from microfinance and instead support larger enterprises in laborintensive industries. This formula, he claims, worked well in China, Korea, and Taiwan.
Even though the microfinance concept has been present for more than thirty years, the evidence on its effects is at best mixed. It is still difficult to identify any sustainable development trajectories arising from this experience. Countries such as India, China, Thailand, Vietnam, Malaysia, Taiwan, Brazil and South Korea, have been successfully dealing with underdevelopment by employing a variety of state and non-state interventions as well as institutional innovations not related to microfinance. China,
Vietnam and South Korea have significantly reduced poverty in recent years with little microfinance activity. Bangladesh, Bolivia and Indonesia have not been as successful at reducing poverty despite the influx of microcredit (Karnani 2007a). Somewhere in between these extremes lie the countries of Eastern Europe, which started channeling of donor funds into microfinance in the 1990s as a response to a number of processes, ranging from the conflict and post-conflict era to deindustria-lization.
Microfinance in Bosnia and Herzegovina
Using Bosnia and Herzegovina as a case study allows for testing microfinance both as a tool for post conflict mitigation and as a means for economic development and poverty reduction in a post-communist economy. bh represents what can be labeled a best-case scenario for testing the impact of microfinance initiatives.
First, bh still had a highly educated work force when the conflict ended - what Demirguc-Kunt et al. (2007) called an 'entrepreneurial middle class.' However, war activities resulted in a per capita gdp drop of 75 percent (from 2,400 usd to 600 usd) registered over a five-year period, from 1990 to 1995 (Demirguc-Kunt et al. 2007). Consequently, three out of four could not afford basic family necessities.
Second, most commonly identified favorable conditions were present. The state-owned enterprises (soes) had been broken up during the war - leaving highly educated people unemployed and displaced. There was also a lack of capital for business start-ups (sdn 2001). Consequently there was an opportunity for microfinance to provide sought-after capital to suitable borrowers. There may be parallels here to the importance of trust (social capital) and affordable credit (loans to small enterprise) in post-conflict reconstruction in other European states, such as Italy after the Second World War (Weiss 1988; Putnam 1993).
Third, weak political and economic institutions meant that mfis could be set up with relative ease. In the formative years of the microfinance institutions the regulatory framework was very limited, which allowed the mfis significant leeway to establish their lending policies.
These factors meant that the 'new poor' in post-conflict bh represented a different clientele than the poor in Africa and Asia, since they were educated and usually possessed some physical assets. Hartarska and Nadolnyak (2007, 4) offer their description of the new poor:
The potential microentrepreneurs were people who before the war might have had sophisticated private businesses but were displaced
or, alternatively, people who, before the war, were factory workers but became unemployed after the industry collapsed (post-war unemployment reached as much as 85%).
Painting a less optimistic picture, Ohanyan (2002,398) has underlined the complexity of bh's post-socialist, post-conflict context in the following way:
The post-Communist conflict cases are characterized by the problem of multiple transitions, which refer to the variety of policy goals that international organizations working in the country bring to the table. Specifically, transitions (1) from war to peacetime economy, (2) from command to market economy, and (3) from humanitarian assistance to long-term development assistance are the major policy goals driving the international involvement in bh.
Although it warrants some focus on possible favorable conditions, it is argued that microfinance in a post-conflict environment is not dramatically different from other, 'normal' environments (Larson 2001). The core idea still is to provide small amounts of credit to those deemed not creditworthy in a traditional, commercial banking setting. A core condition is that there must be a low intensity of conflict. This is amongst other things necessary in order to ensure the safety and working conditions of the mfi staff, their offices as well as the clients. This is the first of three essential conditions set out by Doyle (1998). Doyle's second condition is that markets be reopened. In order for an mf initiative to be successful there must be a certain resumption of economic activity, so that the local population can believe in the changes at hand. The third condition concerns long-term displacement, which is needed in order to ensure some reiteration of relations. Doyle finds that displacement of at least 18 months is reasonable for creating a possibility for the economic investments to yield returns. This is especially important if the main objective of MF is to initiate economic growth, not to serve humanitarian goals in the wake of disaster. Adopting a long-term view also makes the MF seem more stable, rather than a quick, one-off fix (Woodworth 2006). All three conditions were present in bh.
Doyle also lists a range of other helpful conditions, including the existence of a functioning commercial banking system; an absence of hyperinflation, which secures a predictable value of money over time; and a certain density of population that increases the possibility for economic linkages. Enabling legislation for mfis is also preferred in order to make
the future more predictable for the microfinance lenders. It is also preferable for the providers of mf to have access to locally skilled and educated labor to staff the organization, while more social capital is preferable to less, due to its effect on demand, scale, training needs and operational efficiency. Finally, trust in the local currency and financial institutions is preferred. mfis should also be cautious about their client target when offering credit in a post-conflict setting (Woodworth 2006). They should not offer credit to everyone as a peace-building measure, but keep in mind the probability for repayment and economic development.
bh is sometimes hailed as an excellent example of success of a microcredit enterprise in post-conflict situations, considering its financially sustainable microfinance institutions (Woodworth 2006). The mfis initiated their activities shortly after the 1995 Dayton Peace Agreement ended the war, and over the ten following years 50 mfis were established (Ribic 2005). The increase in the number of mfis in bh followed a general trend of increased funding for microfinance activities. This happened in an environment in which financial sectors officials accepted NGos giving loans and government officials made the registration process for ngos quick and simple (Woodworth 2006). Today, the bh mfis are some of the largest in Eastern Europe, financially self-sustainable and operating in a competitive environment.
Both short- and medium-term microfinance strategies have been applied in bh (Ohanyan 2002). While the first approach sees microfinance as an immediate peace-building instrument targeting segments of the population and offering financial services to these specific clients, the second approach is more commercially focused on the long-term sus-tainability of the microfinance institutions. The first approach tends to be favored by humanitarian organizations - such as the unhcr - in order to reintegrate minorities, offer a social safety net and minimize the social cost of transition (Ohanyan 2002; Hudon and Seibel 2007). The second approach was favored by some of the international donors funding the mfis, such as the World Bank. Overall there seems to be a consensus that the environment for growth in microfinance in bh was good. Financial officials accepted that ngos were providing the credit, and government officials facilitated the registration process.
The brief history of microfinance in bh can be divided into two periods, before and after 2000. The initial period, from 1996 to 2000, covered the early years of microcredit during which legislation was nearly nonexistent. This coincided with the first of two World Bank-financed mi-
crofinance institutions in bh: Local Initiatives Project i and ii, covering the time periods 1996-2000 and 2002-2005, respectively.
The World Bank was joined by the United Nations High Commissioner for Refugees (unhcr), the Netherlands, Italy, Japan, Switzerland and Austria in financing the lip i project, which had the following three objectives (World Bank 2001):
•	lip i should respond to urgent needs by targeting demobilized soldiers, displaced persons, returning refugees and widows.
•	It should commence a process of establishing financially sustainable mfis in 5-10 years time.
•	Further it should improve the business and regulatory environments for self-employment, micro- and small-enterprises, as well as the regulatory environment for non-financial mfis.
In the World Bank report (2004), lip i was qualified as a successful project and had produced results beyond original expectations: Through micro-credit organizations (mco) established under the project, some 20,000 micro-enterprises with up to five employees had received 50,261 loans with maturity ranging between 6 and 18 months. The loaned sums were small, averaging 1,600 usd. Repayment records were very good, likely helped by the incentive of receiving larger loans if repaid on time. 50 percent of recipients were females (war widows), 21 percent were displaced persons, while 5 percent were returning refugees. 17 ngos were funded in order to initiate microcredit activities in 1996, and by the end of 2000, eight of nine mfis funded under lip i became self-sustainable. On the basis of these results the second phase of the project was launched in 2001.
The second period is the post-2000 era, when a more advanced legal framework was elaborated. The first two years, 2000-2001, saw the adoption of the microcredit organization (mco) laws that were put in place to cope with the institutions that had developed in the immediate postwar period. It was not an easy political environment in which to operate, and the adopted legislative agenda was limited. Cicic and Sunje (2002) analyzed the proposed regulation, and found that it proved difficult to have mfis assume the form of a finance company - a privately owned lending institution not necessarily limited to microenterprise lending. Neither was it possible to regulate them as savings and credit associations that would allow them to mobilize capital from their members in the form of savings, nor as microfinance institutions that would be authorized to ac-
cept deposits from the general public (Cicic and Sunje 2002). Rather, the legislation allowed the mfis to be regulated as microcredit organizations that were non-profit, credit only-institutions. The mco-law was passed in bh parliament in 2000 and rs parliament in 2001, allowing mcos to operate in both the Federation of Bosnia and Herzegovina, and the Republika Srpska (Lyman 2005). The World Bank was an important driver in furthering this legislative agenda. In 2002-2003 the broader legislative developments concerning the financial system took place. The Central Bank is the main monetary authority of bh, while the Banking Agencies conduct supervision and enforcement of banking regulations (de Montoya et al. 2006). Since 2004 the microcredit sector has been maturing, and the Central Bank has emerged as the likely dominant force in the future development of the bh financial system (Lyman 2005).
During this period, the World Bank's second Local Initiatives Project (lip ii) focused on facilitating the transformations of mcos from ngos to commercial-legal organizations. This furthered financial market legislation and regulation, as well as mco market consolidation (Lyman 2005). These developments gave the mcos the opportunity to operate as mfis - offering both credit and deposit, as well as making it possible for them to transform into for-profit institutions. The lip ii came into effect in the spring of 2002, with a budget of some 24 million usd of which the World Bank/International Development Agency financed 20 million usd and the two Entity governments the remainder (Dunn 2005). The project's aim was to raise incomes, develop businesses and increase employment through the use of microfinance institutions. This was to be done in two specific ways (World Bank 2005):
•	lip ii should finance the growth and institutional development of high-performing microfinance institutions with the potential to provide sustainable financial services to the 'unbankable.'
•	It should also support the transition of the microfinance sector towards sustainable sources of financing.
The World Bank found the lip ii to produce satisfactory outcomes, as the project had significantly influenced the entrepreneurial poor, strengthened the mcos' capacity for providing high-quality credit services to their clients, and had helped 'create and/or sustain more than 200,000 jobs in nearly 100,000 micro-businesses' (World Bank 2005, 4). Consequently, lip i and ii have had major influences on the development of a microfinance sector in bh (Dunn 2005). The lips were seen
as necessary for providing micro entrepreneurs with access to loans.
With the transition from the first to second phase, microfinance institutions in bh gradually acquired more autonomy and financial independence. Initially, the Ministry of Finance wanted direct control of the microcredit project, while the World Bank preferred the mfis to be autonomous entities. The two compromised by using the existing Employment and Training Foundations to create the Local Initiative Departments (lids) in both the Federation and Republika Srpska. After this proved successful they agreed to channel all the funds through the lids (sdn 2001). Most MFis became financially self-sufficient shortly after their establishment. Under the lip i Project, eight mfis became self-sustainable and five were fully financially sustainable - perhaps most notably among them the Pro-Credit Bank (Cicic and Sunje 2002). Five MFis thus had a positive fully adjusted return on assets, and hence sufficient income to cover all costs including inflation, market costs of funds and adjustments of subsidies. mfis tripled their total assets and gross loan portfolio in the 1999-2003 period (Berryman and Pytkowska 2003). Long-term sustainability can be challenged when the mfis operate in limited markets, have similar groups of beneficiaries, offer similar product portfolios, and share similar missions. In bh, however competition both inside the mfi-sector and from the commercially oriented banking sector prevented this outcome, and prompted major consolidation of microfinance institutions (Berryman and Pytkowska 2003). Therefore, the lip ii could reduce the number of organizations that received funds to 8, a significant downscaling from the 17 receiving funds through lip i (Dunn 2005). Since the introduction of new Mco legislation in 2006, the organizations can be transformed into non-profit microcredit foundations (mcf) or for-profit microcredit companies (mccs) (Mehmedovic and Sapundzhieva 2009). This legislation opens the capital structure of the MFis/Mccs to investors - securing further access to finance.
The Impact of Microfinance in Bosnia
Several independent studies have assessed the impact of microfinance activities in bh. The results vary from positive, through neutral, to negative. It is likely that the varying assessments have to do with issues of methodology. For example, as found by Matul and Tsilikounas (2004), the impact on entrepreneurs' finances has been underemphasized in bh, as the social impact was the main interest in the post-war environment. The following section provides an overview and assessment of this range
of studies and reports. These findings are organized under four labels: (i) the immediate effect on profit, business start ups and economic performance, as well as (ii) the effect on the development of the broader economic system including institutions and the rule of law. In terms of its social impact, the focus is on (iii) the welfare and social equality of the users of microfinance, as well as (iv) on post-conflict reconciliation and integration.
First, assessments of the success of mfis and their effect on growth have generally been positive. Although it is inadequate to assess the success of the microfinance initiative by assessing the sustainability of the mfis themselves, because profitability is not necessarily correlated with the sustainability of the businesses to which they lend, this is a first step. Because the mfis in bh competed for clients, they quickly learned the use of methods like focus groups, exit interviews and market research in order to better learn clients' needs and how to meet them with improved financial services (Hartarska and Nadolnyak 2007). This market-solution, in turn, led to a growth in products offered, and as a result, the range of mfis was better able to meet the needs of a diverse group of entrepreneurs (Goranja 1999). Thus, mfi profitability can be seen as a measure of selling successful products to the client population.
The World Bank's own evaluation (2004) of lip i found that the project had made an important contribution to the resumption of economic activity in the post-war bh, through supporting start-ups and the expansion of small businesses (World Bank 2004). Specific data on business start-ups were not collected in the baseline survey, however. In the evaluation of lip ii, the World Bank (2005) found that microcredit services had helped create and/or sustain more than 200,000 jobs and served 98,852 active clients. However, the lip ii, and consequently the self-evaluation, focused mainly on the development of the sector rather than its broader impact on the economy and society. The sustainabil-ity of these micro-businesses is also drawn into doubt as World Bank researchers have found that just under half of new microenterprises established in bh between the 2002 and 2003 failed within just one year (Demirguc-Kunt et al. 2007). It is also noteworthy that the project's accomplishments on documenting its impact on a client-level were rated not higher than satisfactory.
There have also been independent assessments of the lips, as well as of the bh microfinance initiative as a whole. In a study of 1,437 microcredit clients and rejected clients,^ Cicic and Sunje (2002) found that microcre-
dit is positively correlated with new business-generation, the addition of new lines to existing businesses, a growth in production, and, less regularly, employment. By using data collected from the World Bank's Living Standards Measurement Survey, Hartarska and Nadolnyak (2007) evaluated the impact of the bh microfinance industry as a whole. Their findings showed that mfis improved access to credit in municipalities where two or more mfis offered competing financial products.
The main critique has centered on the broader impact of microfi-nance on the bh economy. Bateman (2007b) argues that the commercial microfinance approach has led to deindustrialization and infantiliza-tion of the bh economy. It has atomized the local enterprise sector, programmed to reach only the minimum efficient scale of operations, while simultaneously facilitating trade deficits and destruction of local social capital. He argues that there was never an attempt to establish a local industrial policy for bh, although 'the comparatively high level of industrial development, skills and technology in 1995 represented a once only opportunity to establish a core of small-scale, innovative, relatively technology-intensive, industry-related ventures' (2007b, 214). These high opportunity costs are due to the commercial microfinance model, which represented 'the only major local financial support structure in bh' and was used for funding of 'largely unsustainable trade-and household-based economic activities.' Bateman has identified an adverse selection of the microfinance clients, and an anti-industrial bias characterized by '[f]iltering out those potential entrepreneurs wishing to work in the industrial sector but who cannot hope to service the onerous terms and conditions offered by the commercial microfinance institutions, and filtering in those ventures incorporating only the very simplest of non-industrial business ideas that just about can' (Bateman 2007b, 214). Rather than providing credit to microenterprises, it has been argued that credit should be offered on affordable terms and maturities to small businesses, drawing on the experiences from the military-industrial complex in Northern Italy, amongst other places, that was central in rebuilding, restructuring and developing the post-Second World War economy (Weiss 1988).
Bateman (2007a) argues that microfinance has been mostly directed towards the informal sector. That way, he argues, it does not create businesses but rather subsidizes microenterprises and that it has a redistributive effect. Microfinance has 'undermined most local economic and social development triggers, such as cumulative and coordinated
investments, capturing economics of scale and scope, technological innovation, inculcating social capital, or incorporating technical skills and knowledge' (Bateman 2007a, 4). Bateman (2007a) has tried to nuance the impression of microfinance in bh by pointing out that 30 per cent of borrowers from mfis during the lip i-period that were surveyed in 2002 had failed within two years.
Second, most studies point to a positive effect of microfinance on bh's economic and legal system, although there are some broader criticisms of its small-scale and liberal focus. Again, the World Bank's own assessment is the most positive. The transitions for mcos to increasingly rely on the market as a source of finance, as well as the mcos' project management, which were both rated highly satisfactory. Its impacts in furthering legal and regulatory reform, providing institutional support for mcos, and documenting microcredit impact were rated satisfactorily (World Bank 2005). This indicates a highly professional microfinance sector that has been granted due attention by the World Bank. In a study of the mf experience in bh, Dunn (2005) analyzes data on the effect of mf for clients and non-clients alike. In her impact assessment of the lip ii, Dunn found that microcredit indeed has positive impacts on household welfare, business development and business start-ups, employment, workers' wages, as well as livestock development, amongst other things. However, isolating the impact of mf is difficult. mf lending is not the only source of finance for the micro-businesses. Dunn found that 29 percent of entrepreneurs had loaned finances from alternative sources, in addition to the mfis with which they were associated at the time. For example, foreign-owned commercial banks, incentivized by the profitable microcredit sector, did increase their supply of microloans to bh. Chen and Chivakul (2008) attribute much of the real growth in credit to bh households between 2001 and 2006 to the emergence of these banks on the bh scene, standing at nearly 50 percent compared to the real growth rate for credit to enterprises of 13.5 percent.
Critics of microfinance in bh point to the close association with neo-liberalism, trade and privatization, and its adverse effects on investment and the overall economic system. Bateman and Chang (2009) criticize the microfinance model on several accounts. For example, the focus on mf ignores economies of scale by stimulating individual start-ups. This leads to economic activity being organized in the least efficient way, they argue. Proponents of mf lending have also drawn attention away from other perverse effects. For example, Bosnian mfis provided financing for
its clients to purchase cows, resulting in over-supply of milk and consequently decreasing prices (Bateman and Chang 2009). The indirect effect of MF lending thus adversely affected the one-cow farmers and incumbent milk producers alike. mf favors easy business ventures that are able to meet the short maturity of the loans, while capital-intensive businesses are under-represented in terms of mf (Bateman and Chang 2009). The lack of capital-intensive investments further leads to a trade deficit since such goods must be imported.
Third, in terms of the overall effect of micro-finance on welfare and social equality, the verdict is even more mixed. But measurement is also more problematic. It is not unusual for mfis to be designed for reaching certain especially sought-after groups. However, as Meissner (2005) points out, reaching only these groups may have the perverse effect of creating jealousy among other groups - thus undermining the very restoration of social capital. Rather good programs should be available for everyone - including the conflict-affected groups. In designing products and strategies, the institutions offering mf should take care not to discriminate between potential clients, as mf can best help refugees integrate into society by offering them access to credit along the same lines as non-refugees. Furthermore, mere mf outreach does not constitute MF impact; while reaching the sought-after groups is a necessary condition for impact, it is the actions undertaken with the borrowed funds that result in real impact. Measuring and determining any causal effects is difficult - perhaps especially so with regard to the social goals of MF in post-conflict settings. Determining whether the rebuilding of social capital can be credited as mf, or whether it is due to resumption of 'normal' social activity, is challenging to measure. As noted by Meissner (2005), it might take years before the effects of social capital construction initiatives emerge. Matul and Tsilikounas (2004) also point towards methodological issues arising from an evident selection bias. They found that, in most cases, the non-clients either did not know of the possibility for receiving microcredit, or else were concerned that their 'lack of entrepreneurial spirit, skills, or ability to plan' would negatively affect their ability to repay the loans (Matul and Tsilikounas 2004, 8).
The World Bank's self-evaluations of the lip projects concluded that implementations were more or less successful. The World Bank's (2004) self-evaluation of lip i found that the project had successfully assisted both the economically disadvantaged and other underserved groups in resuming economic activities, as well as supporting the establishment of
an institutional framework for micro-credit. The evaluation also found bank performance and borrower performance to be highly satisfactory. Dunn (2005) found that over the period 2002-2004, mfi clients' annual per capita income increased, while the change in non-clients' income was insignificant. Over the same period, the figures showed that only the clients that had a relationship with the mfis for over more than two years experienced a lower share of households below the national poverty level - set at km ii00 per person. This effect was not identified with the clients that had a relationship extending for less than two years. Dunn's data indicate that 'income trends were generally positive' (2005, 23). But an alternative reading is also possible, inasmuch as the positive income effect from mf does not extend beyond the client base. Furthermore, the poverty-alleviating effect is not well documented. The data are somewhat unclear on this point, and may indicate that time is of the essence and that such effects may take a few years to materialize. Critics such as Bateman (2007b, 220) argue that:
Very little evidence has emerged in bh to suggest that the commercial microfinance model actually possesses the required 'transformative capacity' to secure genuinely sustainable poverty reduction, through genuinely sustainable local economic and social development. On the contrary, the commercial microfinance model is quite centrally implicated in the evolution of the disturbingly weak, unsophisticated, anti-social, disconnected and unfair economic and social structures we see in bh today.
Fourth, and finally, the evidence that microfinance has contributed to post-conflict reconstruction is limited, but positive. Conflicts and acts of war traumatize the entire civilian population. After the conflict has ended, the demobilized soldiers may experience issues related to participating in conflicts, and in reconciling with former enemies. Generally speaking, one can say that post-conflict societies tend to be characterized also by low social capital. As trust is normally expected, at least partially, to be built on iteration of relations (Hardin 2003), microfinance may have a positive impact as it facilitates network development by providing an arena for clients to meet. Goronja (1999) finds evidence that mf in bh did indeed produce such reconciliatory effects. Dunn's 2005 study found that microfinance did indeed reach the displaced persons - 36 percent of beneficiaries and 34 percent of new beneficiaries were dislocated - the demobilized, disabled and widowed.
Conclusion
The overall picture is mixed. bh may be a good critical test case, but the results are not unambiguous. It is clear that mf has had some positive contributions in this post-conflict situation, and that mfi sustainability in bh has been remarkable. However critics suggest that mf may actually have impeded sme development in the country, and that it remains a challenge to make mf have an impact beyond the client base. The mfi-model applied in bh has produced self-sustainable mf institutions with, amongst other improvements, acceptable repayment rates. But equating the success of mf institutions and their clients with the success of mf is not enough. mf also affects the broader legal and economic system, and seems to have had a positive impact in the bh case. Even so, mf does not necessarily translate into success in effect among the population. This point might be especially true in post-conflict situations where some of the goals are of a social nature. There is therefore some concern that a focus on mfi commercialization can be seen as a mission drift. For a poverty alleviation and post-conflict strategy to be truly successful, it cannot be limited to improving the standard of living for the mf clients alone. Such a strategy would be significantly impacted by the issue of adverse selection of mf clients. It can hardly be based on mf institutions' assessment of the potentially most profitable clients. mf may indeed have a positive effect on both the economic and social arena, and working to maintain and enhance these beyond the immediate post-conflict situation should still be a priority for mfis and governments alike. However, the evidence reviewed in this article should indicate that mf is no silver bullet, neither for securing economic growth nor for post-conflict reconciliation. It should therefore be carefully evaluated in each setting how mf is to be designed, and how it can best be complementary to other programs for economic and social enhancement. Yet, as Christen and Drake (2002, 16) point out, the 'ultimate irony of microfinance is that broad outreach is possible only if mfis are commercialized.'
The principal lessons of the short history of microfinance in Bosnia-Herzegovina seem to be that microfinance is more impressive as a tool to improve the economic performance of the mf institutions and their clients than to reach broader social goals. It appears to be a better tool for aiding its participants' quest for wealth and escape from poverty on an individual basis, than for addressing wider problems of social integration or institution building. The bh case provides an almost ideal test
case, since it featured well-educated citizens with incentives to re-build wealth after the destruction of war, and the evidence suggests that this did indeed take place. Moreover, in a context where political and economic institutions had been weakened by war, mf and mfo legislation provided a building block toward rebuilding a small- and medium-scale financial system. Broader social goals, from integration and reconciliation to inclusion, seem mainly to have been achieved (if at all) through mf's effect on its individual participants. Perhaps the clearest conclusion from the mf experience in bh is that even microfinance can only be used to pursue one goal directly; the broader goals may be achieved indirectly, if at all.
Acknowledgments
The authors would like to thank Dr. Dijana Tiplic and Professor Arild
Tjeldvoll for their valuable comments and contributions.
Notes
1	This is an online journal published by the United Nations Development
Program (undp) and the London School of Economics and Political Science.
2	1,032 were current microcredit clients, and 405 were people who applied
for loans but were not approved.
References
Bateman, M. 2007a. Undermining sustainable development with commercial microfinance in Southeast Europe. Development and Transition, no. 7:2-4.
-. 2007b. De-industrialization and social disintegration in Bosnia.
In What's wrong with microfinance? Ed. T. Dichter and M. Harper, 20723. Warwickshire: Intermediate Technology Publications.
Bateman, M., and H. J. Chang. 2009. The microfinance illusion. Mimeo, University of Juraj Dobrila and University of Cambridge.
Berryman, M., and J. Pytkowska. 2003. A Review of the Bosnian micro-finance sector: Move to financial self-sufficiency. Microfinance information exchange. Http://www.mfc.org.pl/doc/Publication/Review.pdf.
Boudreaux, K., and T. Cowen. 2008. The micromagic of microcredit. Wilson Quarterly 32 (1): 27-31.
Chen Chen, K., and M. Chivakul. 2008. What drives household borrowing and credit constraints? Evidence from Bosnia and Herzegovina. imf Working Paper 08/202.
Christen, R. P., and D. Drake. 2002. Commercialization: The new reality of microfinance? In Commercialization of micro finance: Balancingbusi-ness and development, ed. D. Drake and E. Rhyne, 2-21. Bloomfield, ct: Kumarian Press.
Cicic, M., and A. Sunje. 2002. Micro-credit in transition economies. In Small enterprise development in South-East Europe: Policies for sustainable growth, ed. W. Bartlett, M. Bateman, and M. Vehovec, 145-70. Boston, ma: Kluwer.
Demirguc-Kunt, A., L. F. Klapper, and G. A. Panos. 2007. The origins of self-employment. Washington, dc: World Bank.
De Montoya, M. L., R. Deshpande, and J. Glisovic-Mezieres. 2006. Bosnia-Herzegovina: Country level savings assessment. Washington, dc: Consultative Group to Assist the Poor.
Doyle, K. 1998. Microfinance in the wake of conflict: Challenges and opportunities. Bethesda, md: Microenterprise Best Practices.
Dunn, E. 2005. Impact of microcredit on clients in Bosnia-Herzegovina: Final report. Washington, dc: World Bank.
Galusek, G. 2007. Commercial Microfinance: Too much, or not enough? Development and Transition, no. 7:4-7.
Goronja, N. 1999. The evolution of microfinance in a successful post-conflict transition: The case study for Bosnia-Herzegovina. Paper presented at the Joint ilo/unhcr Workshop: Microfinance in Post Conflict Countries, Geneva.
Hardin, R. 2003. Gaming trust. In Trust and reciprocity: Interdisciplinary lessons for experimental research, ed. E. Ostrom and J. Walker, 80-102. New York: Russell Sage Foundation.
Hartarska, V., and D. Nadolnyak. 2007. An impact analysis of microfinance in Bosnia-Herzegovina. World Development 36 (12): 2605-19.
Hudon, M., and M. D. Seibel. 2007. Microfinance in post-distaster and post-conflict situations: Turning victims into shareholders. Savings and Development 31 (1): 5-22.
Hughes, J., and B. Slay. 2007. The private sector and poverty reduction. Development and Transition, no. 7:9-12.
Karnani, A. 2007a. Microfinance misses its mark. Stanford Social Innovation Review 5: (3): 34-40.
-. 2007b. Romanticizing the poor harms the poor. Ross School of
Business Working Paper Series 1096.
Khandker, S. R. 2005. Microfinance and poverty: Evidence using panel data from Bangladesh. World Bank Economic Review 19 (2): 263-86.
Larson, D. 2001. Searching for differences: Microfinance following conflict vs. other environments; mbp Microfinance following conflict. Bethesda, md: Microenterprise Best Practices.
Littlefield, E., J. Murduch, and S. Hashemi. 2003. Is microfinance an effective strategy to reach the millennium development goals? Focus Note, cgap.
Lyman, T. R. 2005. Legal and regulatory environment for microfinance in Bosnia-Herzegovina: A decade of evolution and prognosis for the future. Essays on Regulation and Supervision 2. Washington, dc: cgap.
Matul, M., and C. Tsilikounas. 2004. Role of microfinance in the household reconstruction process in Bosnia and Herzegovina. Journal of International Development i6 (3): 429-66.
Mehmedovic, D., and R. Sapundzhieva. 2009. Bosnia and Herzegovina microfinance analysis and benchmarking report 2008. Association of Microfinance Institutions in Bosnia and Herzegovina / Microfinance Information Exchange.
Meissner, L. K. 2005. Microfinance and social impact in post-conflict environments. Washington, dc: School of International Service of American University.
MkNelly, B., and C. Dunford. 1998. Impact of credit with education on mothers and their young children's nutrition: Lower rural bank credit with education in Ghana. Freedom for Hunger Research Paper 4.
-. 1999. Impact of credit with education on mothers and their young
children's nutrition: crecer credit with education program in Bolivia. Freedom for Hunger Research Paper 5.
Nagarajan, G. 1999. Microfinance in post-conflict situations: Towards guiding principles for action. Paper presented at the Joint ilo/unhcr Workshop on Microfinance in Post-Conflict Countries, Geneva.
Nagarajan, G., and M. McNulty. 2004. Microfinance amid conflict: Taking stock of available literature; Accelerated microenterprise advancement project. Washington, dc: usaid.
Ohanyan, A. 2002. Post-conflict global governance: The case of microfinance enterprise networks in Bosnia-Herzegovina. International Studies Perspectives 3 (4): 396-416.
Pitt, M. M., and S. R. Khandker 1998. The impact of group-based credit on poor households in Bangladesh: Does the gender of participants matter? Journal of Political Economy 106 (5): 958-96.
Putnam, R. 1993. Making democracy work: Civic traditions in modern Italy. Princeton: Princeton University Press.
Ribic, M. 2005. Mikrofinansiranje u BiH od 1996 do 2005 godine. Mikrofi-nansije 5:1-4.
Roodman, D., and J. Morduch 2009. The impact of microcredit on the poor in Bangladesh: Revisiting the evidence. Working Paper 174, Center for Global Development.
SDN. 2001. Innovative approaches to microfinance in post-conflict situ-
ations: Bosnia local initiatives project. Social Development Notes 50, Social Development Family in the Environmentally and Socially Sustainable Development Network of the World Bank. Seibel, H. D. 2006. Financial linkages in Mali: Self-reliance and liquidity balancing versus liquidity supply and donor dependence. Small Enterprise Development 17 (1): 40-9. Storey, D. J. 1994. Understanding the small business sector. London: Rout-ledge.
Weiss, L. 1988. Creating capitalism: The state and small business since 1945.
Oxford: Blackwell. Woodworth, W. 2006. Microcredit in post-conflict, conflict, natural disaster,
and other difficult settings. Provo, ut: Brigham Young University. World Bank. 2001. Project evaluation for credit of 20 million usd to Bosnia-Herzegovina for project local (microfinance) initiatives ii. World Bank Report. -. 2004. Project performance assessment report Bosnia and Herzegovina. World Bank Report. -. 2005. Implementation completion report local initiatives (micro-
finance) project ii. World Bank Report.