A macro-framework for successful development banks

Abstract

The large-scale failures of development banks in the 1970s and 1980s meant that they all but disappeared from the development agenda. However, there are still a large number of development banks worldwide that operate with various degrees of success. Some governments are also looking to re-establish such banks to address the shortage of finance for higher-risk market segments. To avoid a repeat of the earlier failures, government policy needs to be informed by an objective framework for the success of these banks. This article, based on economic theory and informed by case studies, outlines such a framework. It addresses the following six dimensions of these banks: enabling environment, mandate, regulation and supervision, governance and management, financial sustainability and performance assessment. Development banking remains a risky initiative but, managed appropriately, and using this framework, it can help achieve development objectives.

Keywords:

• Development finance
• development bank
• governance

1. INTRODUCTION

After the much-publicised failure of many national development banks in the 1970s and 1980s, the future of this type of development finance was in question (World Bank, 1989). Poorly controlled spending by development banks and other state-owned banks delivered little in terms of development but contributed to fiscal crises in several developing countries (Nellis, 1986). As the policy trend shifted towards financial liberalisation, many of these countries privatised, restructured or closed their state-owned banks (Fitzgerald, 2006). The then prevailing view of development banks is summed up particularly well by Krahnen and Schmidt (1994:6): ‘Development banks still have to find their proper role in a world where financial repression is on the retreat … They might not find such a role and gradually die.’

However, there are at least two reasons why the role of national development banks is being reassessed. First, many governments persisted with these banks, with mixed results. Although there are still failures, a few development banks have successfully adhered to their mandates in a financially sustainable way. Second, where the banks have been closed or privatised, the market failures that they had been meant to correct still hamper economic growth and development. Despite government efforts to address these failures (for instance, through supporting financial development), many developing countries still experience a shortage of long-term finance and finance for high-risk sectors such as new or small and medium enterprises. This has rekindled interest in national development banks, which raises the possibility that, if there is no clear understanding of the role of these banks, the problems that beset the early ones may well recur (Yirga-Hall, 1998; Yaron, 2004; Bruck, 2005).

Despite the obvious need for a better understanding of their role, however, development banks have not featured strongly in the recent literature on development finance. Much of the relevant literature is either dated, deals with national development banks only in passing, or is restricted to particular aspects affecting their performance (see, for instance, World Bank, 1976; Diamond & Raghavan, 1982; Yaron, 2005; Scott, 2007). On the whole, the literature offers very little structured guidance on how best to use, position and govern these institutions.

There is a clear need for an objective framework for development banking that can inform government policy and help prevent a repeat of the earlier problems. There is no universal model for development banking as it is influenced by a variety of factors, such as a country's level of development and the sophistication of its financial system. Still, a general framework may be derived for the role of such banking. The application of this framework then depends on a country's particular developmental and financial needs.

This article outlines such a framework for what Nembelessini-Silue (2006) calls ‘classic’ national development banks – state-owned and wholesale (which means that they fund institutions or projects rather than individuals). The framework does not address private development banks, deposit-taking institutions, multilateral or regional banks or microfinance institutions. The framework draws on economic theory and case studies of development banks in Africa, Asia and Latin America (Thorne, 2008). It sets out principles for six dimensions of development banking: enabling environment, mandate, regulation and supervision, governance and management, financial sustainability and performance assessment.

The article is structured as follows. Section 2 defines development banks and outlines their role in the financial system, Section 3 reviews the evolution of development banking, Section 4 assesses some principles of the six dimensions, Section 5 draws these together into a macro-framework, and Section 6 concludes by summarising the issues and highlighting areas for further research.

2. ROLE OF DEVELOPMENT BANKS

Development banks are a form of government intervention in the financial system that aims to address market failures in the provision of finance or, more generally, to help achieve socio-economic objectives such as equity or poverty reduction. Development finance can broadly be defined as the provision of finance to those market segments (such as projects, sectors or sections of the population) that are not well served by the financial system. These segments include projects whose social benefits exceed their commercial ones; long-term projects or projects with a long lead time; new or risky ventures, such as new technologies; and small and new borrowers who lack collateral (Diamond, 1996; De Aghion, 1999; Marston & Narain, 2004; Micco & Panizza, 2004; United Nations, 2005).

As Figure 1 shows, development finance complements government resources and market funding by filling the niche between fiscal funding for projects from which no cost recovery is possible and private funding for projects with profit opportunities. Development finance aims to lower the risks of investment in certain sectors or areas and hence crowd in private sector investment (Marston & Narain, 2004; Musasike et al., 2004; United Nations, 2006a). It is channelled through a variety of institutions. Most countries have more than one development finance institution; many have a complex system comprising a range of private and public institutions. Worldwide, about 520 of these institutions are development banks, and they tend to focus primarily on providing long-term finance to projects with large social benefits. They also provide developmental services such as research, advocacy and technical assistance (Bruck, 2005).

Figure 1: Role of development finance institutions

3. EVOLUTION OF DEVELOPMENT BANKING

Development banks played an important role in facilitating the industrialisation, and later the post-war reconstruction, of Europe by providing long-term finance. Among the most important early development banks was the privately owned Crédit Mobilier, established in 1848, which played a large role in the European economy and is credited with increasing per capita income, disseminating skills in long-term finance and fostering competition. It was the model for the Industrial Bank of Japan in 1900, which in turn was the model for later development banks in poorer countries. However, despite its success, the Crédit Mobilier faced a significant problem: that of conflict between its evelopmental objectives and the need for short-term profit (Cameron, 1953; De Aghion, 1999; United Nations, 2006a). Even now, 150 years later, the issue of conflicting objectives remains at the heart of the problems facing development banks.

Two events created a demand for development finance in the early twentieth century: World War I, which led to the need for reconstruction; and the Great Depression of 1929, which led to a shortage of long-term funds in the US and Europe. Several development finance companies were set up after the war, with government support. After World War II, the renewed need for reconstruction led to the establishment of a second group of development finance institutions. These included the generally successful Kreditanstalt für Wiederaufbau and the Japan Development Bank, the forerunner of the Development Bank of Japan (De Aghion, 1999).

The success of the early development banks can be attributed to factors such as private (co)ownership, co-financing of projects with the private sector, commitment to skills dissemination, autonomy, hard budget constraints, and highly professional staff and managers. They also benefited from the post-war economic stability in developed countries (Diamond, 1996; Siraj, 2004). This success encouraged poorer countries to establish their own development banks. But many of these banks failed, leading to large fiscal losses and poor development outcomes (Micco & Panizza, 2005). In the words of Nellis (1986:ix), they ‘present a depressing picture of inefficiency, losses, budgetary burdens, poor products and services, and minimal accomplishment of the non-commercial objectives so frequently used to excuse their poor economic performance’. A range of financial, political and management problems contributed to their failure (World Bank, 1989; Yirga-Hall, 1998; De Aghion, 1999; Siraj, 2004):

• The institutional environment in developing countries was weak and critical skills in management, finance and operations were limited.

• Governments or corrupt officials often interfered in their activities.

• They were poorly managed and regulated, and did not operate on commercial principles.

• Their mandates were rigid and often inappropriate, and they were stand-alone banks instead of being integrated into the financial system.

• Even well-managed banks struggled to reconcile their conflicting objectives of maintaining financial sustainability while pursuing socially desirable outcomes. The World Bank (1989:106) sums it up well: they ‘found it difficult to finance projects with high economic but low financial rates of return and remain financially viable at the same time’.

The 1980s and 1990s saw the widespread restructuring, closure or privatisation of development banks. Despite subsequent efforts to strengthen capital markets, many countries still face a lack of long-term finance and developmental services such as technical assistance and research (Yirga-Hall, 1998; Siraj, 2004; International Monetary Fund [IMF], 2006; World Bank, 2006). At the same time, there have been visible successes: some development banks have maintained financial sustainability while adhering to their mandates. They have provided counter-cyclical funding, facilitated access to credit, created employment, strengthened the capital market, built capacity in project appraisal and evaluation, and influenced government policies (Hulme & Mosley, 1996; United Nations Economic Commission for Latin America and the Caribbean [UN-ECLAC], 2002; Micco & Panizza, 2005; United Nations, 2005). This combination of successful development funding and insufficient private funding has led to calls for a re-examination of the potential role of development finance institutions (see, for instance, UN-ECLAC, 2002; Siraj, 2004; Yaron, 2004; United Nations, 2005, 2006a; Fitzgerald, 2006; IMF, 2006; Nembelessini-Silue, 2006).

This history shows that development banks have an unenviable role – they operate as financial institutions within the constraints imposed by the implementation of (largely unprofitable) government policy. Likewise, the governments that own these banks are also in an unenviable position – the banks may incur considerable fiscal liabilities while delivering little by way of development (Caprio et al., 2004). This raises two questions: on the one hand, how can a government create an environment that ensures the bank will make the best use of the state's scarce resources for development? On the other, how can the development bank best manage its conflicting objectives and avoid undue government interference? The next section proposes some principles for addressing these problems.

4. SUCCESSFUL DEVELOPMENT BANKS: DIMENSIONS AND PRINCIPLES

The success of particularly the early development banks and the widespread failure of many of the later ones are a rich source of lessons for development banking. On the basis of these lessons, and case studies of development banks, this section outlines six interdependent dimensions of the success of a development bank, and identifies and discusses the principles involved in each of these dimensions.

4.1 Enabling environment

A development bank requires an enabling environment in which to operate. Its role is determined primarily by the country's socio-economic environment and its particular priorities. But this environment also influences the bank's ability to carry out its functions. In the words of Diamond (1996:12), ‘no factor is more important in influencing a development bank's “success” than the situation of the economy in which it operates’.

Put simply, while the mandate of the development bank may require it to address problems in the economy, it cannot operate in a largely dysfunctional environment. This is one of the paradoxes of development banking – these banks are needed most in poor countries, but the often weak economic and political systems in these countries make it more difficult for the banks to succeed. For example, macro-economic instability contributed to the failure of 16 development banks in Francophone Africa during the 1980s (Yirga-Hall, 1998). More recently, Malawi has been described as a ‘fundamentally flawed contextual basis’ for development banking owing to its poor economic prospects, high levels of corruption and limited political will to foster good governance (Benefit Advisory Research [BAR], 2006:62).

Thus arguably the primary requirement for a development bank is macro-economic stability, without which financial development is virtually impossible as it affects the risks associated with all types of finance. Development banks have proved unable to perform without a reasonably functional micro-economic environment with proper regulation, acceptable infrastructure, sufficient skills and adequate competition, for example. They also require political stability, leadership without undue intervention, and adequate capacity in the organs of state. In addition, well-functioning legal and regulatory institutions are as much a prerequisite for development banks as for the rest of the private sector (Shirley, 1989; De la Torre, 2002; IMF, 2004; Bruck, 2005; United Nations, 2005; BAR, 2006).

4.2 Mandate

A development bank needs an appropriate mandate to ensure it is correctly positioned within the environment. The lessons drawn from the experience of development banking have highlighted the disastrous effects of inappropriate mandates, but countries such as Malaysia, Brazil and Rwanda show that banks with appropriate and flexible mandates can contribute significantly to development (BAR, 2006).

The first principle is mandate clarity: the mandate of the development bank must be clearly articulated, as a vaguely defined mandate creates uncertainty for the bank, its stakeholders and the private sector. It allows the bank to pursue activities not intended by the government (‘mission drift’), gives the bank more scope to avoid difficult or costly activities (‘mission shrink’), reduces accountability, and increases the opportunities for political interference (Diamond & Raghavan, 1982; Shirley, 1989; Caprio et al., 2004; BAR, 2006).

The next two mandate principles relate to tailoring the bank to the specific needs of the country: local relevance and institutional fit. Since the bank aims to fill the gap between the public and the private provision of finance, its role is influenced by the reach of both local and foreign finance, and by the government's fiscal policies and its view on intervening in the financial sector (Diamond, 1957; Kritzinger-Van Niekerk, 1995; Musasike et al., 2004). The bank must also fit into the local economic, political and institutional environment, and complement other financial institutions. Diamond (1957:18) puts it well: ‘A development bank is one instrument among many, all of which need to be used consistently and in conjunction.’

This leads to the fourth mandate principle: complementarity of funding. As an integral part of the financial system, the development bank should restrict itself to funding only those activities in which it has a comparative advantage (and that are not funded by commercial banks) in order to avoid crowding out the private sector. Typically, the development bank will have a better understanding of high-risk markets and an in-depth knowledge of the clients in these markets. By restricting itself to funding according to such comparative advantages, a development bank is less likely to compete with commercial banks.

There are two other aspects to complementarity: first, the bank should aim to mobilise private co-funding of its projects, whether by demonstration or more concretely through risk mitigation measures. Petersen and Crihfield (2000:71) say that development finance institutions ‘should always be looking for an exit strategy and a shifting of obligations to the commercial credit markets’. Second, the bank should assist borrowers only until they are financially strong enough to obtain private funding. In the words of the UN-ECLAC (2002:160): ‘[It] should be run in a way designed to avoid building up a permanent, stable customer base’. This will help to ensure that the bank's scarce resources are not captured by stronger borrowers at the expense of weaker ones.

Development banks have struggled with this principle owing to the moral hazard and adverse selection effects of concessionary finance (Diamond, 1957; Hulme & Mosley, 1996; Yaron, 2004; United Nations, 2006a). Examples of these effects include the following:

• Because they often provide technical assistance to clients, banks build up strong relationships with their clients over time, which both may be keen to preserve.

• The client may have a perverse incentive to understate its financial strength to continue to obtain concessionary finance.

• The bank may be measured in terms of the volume of its funding. Hulme and Mosley (1996:183) say that ‘lenders who are under strong pressure to meet lending targets have no incentive to be rigorous in refusing a promising borrower’.

• Since existing borrowers are cheaper to finance, as information costs are lower, a bank may avoid spending time and money on developing new clients.

• The requirement for development banks to be financially self-sustainable creates a powerful incentive for them to compete with the private sector for profitable projects that can cross-subsidise losses on their more developmental projects.

One option for encouraging complementarity may be to create a formal mechanism where the private sector can table complaints about uncompetitive behaviour by development banks. Another would be for co-financing with the private sector to be made compulsory, although this could affect the bank's financial sustainability. The growth in public−private partnerships, especially in the provision of infrastructure, has created good opportunities for co-funding with the private sector. For example, the Brazilian development bank, BNDES, is supporting significant public−private partnerships in the transport sector with the aim of furthering regional integration (Micco & Panizza, 2004; Jackson, 2006; United Nations, 2006a; Scott, 2007).

The complementarity principle is linked to the fifth mandate principle: flexibility. The mandate should be reviewed regularly to take account of changing circumstances. Such changes could stem from a general deepening of the financial system, exogenous influences such as new policy directions, or the success of the bank's efforts to strengthen the private financial sector. The bank's role in changing its own environment is captured in the lifecycle theory: once the market failure for which it was designed has been addressed, whether by changes in the environment, in policy or in private sector capacity, this role should be reduced and eventually eliminated. This does not necessarily mean closing the bank – given the difficulties of building strong institutions, the best use of a successful bank may be to privatise it or to refocus it on another under-served sector (Shirley, 1989; Stanton, 1999; Musasike et al., 2004). There is an increasing awareness of the need to adjust mandates on a regular basis. Malaysia recently reviewed the mandates of its development finance institutions to ensure that they focus on ‘niche’ sectors, while South Africa reviewed the mandates of its development finance institutions in 2007 (BAR, 2006; Thorne, 2008).

The sixth mandate principle relates to the scope of the development bank. There are no easy answers as to whether a bank should be narrowly focused (and therefore small) or multi-sectoral (and large). Although most ‘successful’ development banks are multi-sectoral (Diamond, 1996), such as BNDES in Brazil, each form has both advantages and disadvantages.

Multi-sectoral banks run the danger of being ineffective and unfocused, and are more prone to mission shrink or drift. They may present more problems of corporate governance, be less transparent and be more susceptible to political interference. Also, the failure of a large development bank in a weak financial system could have drastic systemic consequences (Diamond, 1957; Murinde & Kariisa-Kasa, 1997; Graham et al., 1999; Scott, 2007).

Specialised development banking also has a number of drawbacks, the primary one being that of covariant risk. Mistry (1999:7) puts it bluntly: ‘[T]he worst thing you could do to any development bank was to … give it a mandate that automatically led to concentrated covariant risk in terms of portfolio concentration’. Development banks focusing on agricultural lending are particularly prone to this problem, since bad weather conditions would affect the whole portfolio. Such covariant risk contributed to the collapse of the National Development Bank of Botswana in the early 1990s (BAR, 2006).

Given the range of development finance needs, most countries would have to set up more than one specialised development bank. This could overstretch a poor country's limited fiscal, human and managerial resources. Individual banks would have a narrow financial base, coordinating their activities would be challenging, and they could have less of an impact on government policy or in interacting with donors than a large multi-sectoral bank might have had (Diamond, 1957; Kritzinger-Van Niekerk, 1995; Graham et al., 1999; Yaron, 2004; Thomas, 2006).

In short, the appropriate scope of a development bank depends on a range of factors, such as a country's macro-economic conditions and capacity, the strength of the financial system and of the supporting institutions, the size of the potential market for the bank (and hence its options for diversifying its portfolio), and the ability of the government to regulate, coordinate and monitor the development finance system. A government needs to weigh up the relative advantages and disadvantages of a narrow or a broad mandate in the light of these factors.

4.3 Regulation and supervision

Poor regulation and supervision by a government have contributed to the downfall of many development banks, including the Development Bank of Zambia in the 1990s (BAR, 2006). Even now, members of the Association of African Development Finance Institutions (AADFI, 2006:3) still regard the policies and practices of their owners (i.e. the government) as their biggest single problem. A primary concern here is that the ownership role of the state creates a potential conflict of interest in the regulation and supervision of development banks. Caprio et al. (2004:8) warn that this ‘inherent conflict of interest in both owning and supervising banks is difficult to resolve’. The discussion below examines ways of addressing this issue.

4.3.1 State as owner

The corporate governance guidelines for state-owned enterprises of the Organisation for Economic Co-operation and Development (OECD) focus specifically on this potential conflict of interest. Calling on the government to act as an ‘informed, accountable and active owner’ (OECD, 2004:6), the guidelines suggest that a government should:

• ensure that its ownership role does not distort its policy decisions;

• create a clear and simple set of legal rules governing state-owned enterprises;

• make the developmental roles of these institutions and any funding for such roles clear and transparent; and

• ensure that state-owned institutions do not enjoy special privileges.

In addition to such legal rules, best practice is moving towards a formal, published ownership policy that defines the objectives of the state as owner, the legal forms of the enterprises under its control, its role in governance, and how it will implement its ownership role. Several European countries, including Finland, France, Poland and Sweden, have adopted formal ownership policies (Scott, 2007).

The legal rules or ownership policy must establish checks and balances in the way the government exercises its ownership role. This is generally done by sharing the responsibility between different departments. In this process, care should be taken to avoid a regulatory ‘overburden’ (Reddy, 2006:10): if state-owned institutions are overseen by a range of entities with different requirements, they face a heavy reporting burden or, worse, conflicting instructions.

One option for avoiding conflicting instructions is to create a single ownership entity, which would also improve coordination between state-owned institutions. An ownership entity should be independent of government while still being accountable, for example, to a legislative assembly. It should have only a limited participation in the boards of the institutions under its control, allowing the management full operational autonomy. To keep the government's ownership and regulatory and supervisory roles separate, the ownership entity should not be involved in any regulatory or treasury functions (Shirley, 1989; Pannier, 1996; Reddy, 2001; OECD, 2004; Scott, 2007). Norway and Finland have established single ownership entities, while Sweden, Singapore, Poland and Chile have assigned the function to one minister who acts as shareholder representative. Many countries provide dedicated professional support to such representatives (Scott, 2007).

4.3.2 State as supervisor

The counterpart of the ownership entity is the supervisory entity, which should be separate and independent of the ownership function. Fletcher and Kupiec (2004) apply the Basel core principles for banking supervision (Basel Committee on Banking Supervision, 2006) to state-owned financial institutions. They suggest the establishment of an independent supervisory capacity to protect the state against both credit and reputational risk, while also protecting the private sector from unfair competition from state-owned banks.

It is a perennial question whether development finance institutions should be subject to the same rules as the private sector (UN-ECLAC, 2002; United Nations, 2005). On the one hand, these institutions pose more regulatory challenges than do private firms: for the state, the conflict of interest noted above; and for the institution, the problems of political interference and the state's poor regulatory capacity (IMF, 2004; Scott, 2007). On the other hand, given that they operate in under-served markets and under difficult conditions, over-regulation may be counterproductive as it could inhibit innovation and risk-taking. Thus the IMF and World Bank (2003:4) called for a form of regulation of specialised financial institutions that has ‘a sufficiently light touch so as not to crush them’.

Best practice seems to imply that, once the state has clearly defined and separated its ownership and regulatory roles, development banks should be regulated and supervised along the same lines as the private sector, possibly with a caveat regarding capital adequacy. The main regulatory requirement is arguably that the playing field the bank shares with the private sector should be level. The remaining requirements are similar to those of the private sector: an independent board, high levels of disclosure, effective monitoring and evaluation, and sound risk management (De la Torre, 2002; Fletcher & Kupiec, 2004; Marston & Narain, 2004; OECD, 2004; Basel Committee on Banking Supervision, 2006).

Returning to the issue of capital adequacy, there is considerable debate about the effect on development banks of adhering to the Basel Capital Accord (Basel II), which aligns capital requirements with risk. Basel II has a pronounced pro-cyclical effect: during economic downturns, the overall risk of default is higher. In Basel II, this triggers higher capital adequacy requirements and reduces the supply of credit (Griffith-Jones & Persaud, 2005). Reisen (2002:3) says that ‘linking bank lending to bank equity acts as an automatic amplifier for macroeconomic fluctuations’. This would significantly undermine the ability of development banks to provide counter-cyclical funding. Fletcher and Kupiec (2004) suggest that only the commercial activities of development banks should be subject to the Basel requirements, while others propose that they should be exempted altogether (Gottschalk & Sen, 2006).

4.3.3 Market supervision of development banks

Finally, government regulation and supervision could usefully be supplemented by market-based measures such as credit ratings. Although credit ratings are not a formal element of external governance, they help both the government and the bank to gauge the quality of the bank's financial management. There is growing consensus that development banks should submit themselves to the discipline of credit ratings, while also encouraging their clients (especially subnational governments) to obtain such ratings to help them access private capital markets. Few development banks would initially qualify for a commercial rating, but they can take measures to improve their financial standing. In this regard, both the World Federation of Development Financing Institutions and the Association of Development Financing Institutions in Asia and the Pacific are developing rating systems for development banks. The rating questions of the AADFI play a similar role (UN-ECLAC, 2002; Bruck, 2005; AADFI, 2006; Thomas, 2006; United Nations, 2006a).

In summary, the primary principles for the regulation and supervision dimension are a combination of market oversight and a clear separation of the ownership and supervisory roles of the government.

4.4 Governance and management

The quality of governance and management has often meant the difference between the success and failure of development banks functioning in the same environment. For example, while the Brazilian development bank, BNDES, is seen to be successful owing to its strong management, the perception of the management of the Caixa Econômica Federal is far more critical (BAR, 2006; United Nations, 2006a). The analysis below does not deal with the general principles of governance and management in any detail, but focuses on aspects that are specific to development banks.

4.4.1 Role of the board

A properly functioning board is a critical success factor for a development bank as it prevents undue political interference in the bank's day-to-day management. Early on, Diamond (1957:71) called it ‘a very useful screen and protection for management’. The board annually contracts with the government on the objectives of the institution and has a fiduciary duty to oversee performance against those objectives. It also oversees the management of the institution and is accountable to both the government and the stakeholders for ensuring high standards of corporate governance. It must ensure that the bank has a clear performance contract, a strategic plan for achieving the objectives of this contract, proper financial controls and auditing, and a high level of disclosure. It must also ensure the ethical functioning of the organisation through a written code of ethics and adequate measures to prevent corruption (Scott, 2007).

To fulfil this critical role, board members must be objective and independent, act in the best interests of the bank and its shareholders, and have the highest levels of integrity and competence. A board needs an enabling environment, such as a well-defined mandate, independence from government, an appropriate balance of skills and experience, a clear legal exposition of its functions and fiduciary duties, written job descriptions, proper procedures, and a code of ethics for board members. These aspects can be combined into a board charter. The board may need training, and its performance should be evaluated annually through self-evaluation or by the shareholder(s). Boards may be supported by specialised committees, the most common ones being audit, risk management and remuneration committees. External members with specialised skills can be co-opted onto these committees (Pannier, 1996; Micco & Panizza, 2004; OECD, 2004; United Nations, 2006a; Scott, 2007).

4.4.2 Internal management

The management of the development bank, overseen by the board, must set up appropriate internal governance systems to ensure that the institution achieves its financial and developmental objectives while meeting regulatory requirements. The principle is simple: adhere to the best practice requirements for private sector banks. From a corporate governance perspective, this implies professionalism in all aspects of operations, as well as fairness, transparency, accountability and responsibility towards staff, government and stakeholders. From a financial management perspective, development banks should adhere to the general principles of sound financial management. These include the requirements of their charter or founding legislation, relevant central bank requirements and, as far as possible, international norms. In this regard, the Basel core principles state that ‘all banks should be subject to the same operational and supervisory standards regardless of their ownership; however, the unique nature of government-owned commercial banks should be recognized’ (UN-ECLAC, 2002; OECD, 2004; Marston & Narain, 2004:66; see also AADFI, 2006).

4.4.3 Performance management

One of the most intractable problems of supervising a development bank is maintaining the balance between accountability and autonomy. Failure to achieve such a balance could lead to political interference and/or poor funding decisions. This makes performance management a critical part of the governance process.

It is generally agreed that the government and the development bank should conclude a performance contract that sets out clear objectives. The bank is then given the operational autonomy to pursue these objectives. Performance contracts are intuitively attractive as they incentivise additional effort, clarify the requirements of the stakeholders, facilitate monitoring and evaluation, and increase transparency (Nellis, 1989; Pannier, 1996; Shirley & Xu, 1998). But these contracts could have unintended consequences, since there is an agency relationship and there are large information asymmetries between the government and the bank. For example, a financial sustainability requirement could lead the bank to reduce its costly developmental services and compete with the private sector for lucrative contracts. There may also be elements of a power game between the bank and the government, especially when the latter has limited capacity. This problem may be addressed in part by publishing the contract to enhance oversight by the public and the media (Shirley, 1998; Shirley & Xu, 1998). This should be complemented by transparency and adequate disclosure of the performance of the bank in terms of this contract, including, for example, policies, portfolio structure and the projects that are being financed.

In conclusion, the main requirement for the governance and management dimension is adherence to principles similar to those of the private sector.

4.5 Financial sustainability

Diamond (1996:12) defines financial sustainability as ‘the capacity [of development finance institutions] to attract, on the basis of their own performance, the capital they required to pay their creditors, sustain their shareholders’ interest, and support their own growth'.

A financial sustainability requirement protects the government against the kind of large-scale losses experienced during the collapse of development banks in the 1970s and 1980s. In addition, the requirement encourages efficiency, reduces the cost of funding, safeguards the independence of the bank and, through demonstrating profitability, attracts private investment to development projects. The principle of financial sustainability, at least for the commercial operations of a development bank, is supported by a variety of observers (see, for example, Diamond, 1957; Bruck, 2005; BAR, 2006; United Nations, 2006a). However, as noted, a financial sustainability requirement may incentivise inappropriate behaviour, such as a lower risk appetite, competition with the private sector and a reduction in developmental services.

There is a fine line between financial sustainability and profitability, which may imply that developmental objectives are not being met. One way of treading this line is for banks to agree with the shareholders on a specific capital adequacy ratio (Buyskes, 2005). An alternative may be for a bank to earmark returns from highly profitable investments for social investment purposes, such as grants. Neither of these options addresses the question of project choice: the bank may still take on projects that could have been funded by the private sector. One option for avoiding this problem is fiscal support such as subsidies and tax or dividend exemptions. Such support can be earmarked for developmental activities with positive externalities, provided that the social benefits of these activities exceed the costs and that the fiscal support is equitable and transparent.

This is not to argue for blanket subsidies. Even development banks reject subsidies as ‘hazardous’: dependency on the fiscus exposes them to political interference and uncertainty about future income streams. It also undermines their independence, the credibility of their advocacy role, and their ability to influence government policy. Instead, the call is for well-targeted and transparent fiscal support for specific activities that can be ring-fenced and separated from the commercial activities of the organisation (Development Finance Forum, 2004; United Nations, 2006b:14).

4.6 Performance assessment

Proper performance assessment contributes to clear policy decisions: when the public and the government are well informed about the costs and benefits of the activities of development banks, these institutions have a powerful incentive to become more efficient. Over the past decade or so, the performance of development finance institutions has increasingly been assessed on the basis of the twin measures of outreach and sustainability (Yaron, 1992, 2004, 2005; Schreiner, 2002).

Schreiner's (2002) six ‘aspects of outreach’ are a proxy measure for the social benefits of development finance. The ‘aspects of outreach’ are the worth of the service to the clients, the cost to the client, the depth and breadth of the service (the poverty status of clients and the number of people reached), the length or sustainability of the service and the scope of the service. Woller (2006) combines these aspects of outreach with a social audit (a review of processes such as leadership, staff development, performance management and strategic planning) to design a ‘social performance measurement tool’. Yaron (2004) uses an ‘outreach index’ customised to individual institutions to reflect the relative priority of the different aspects of outreach. The index is based on quantifiable indicators, such as the size and growth of the portfolio or average loan sizes, and can be adjusted in line with new policy requirements.

On the sustainability side, development banks are assessed in terms of standard financial ratios of efficiency and on the basis of their subsidies. Yaron's (1992, 2004, 2005) Subsidy Dependence Index takes account of all sources of funding. The Subsidy Dependence Index is calculated by dividing the annual net subsidies received by the average annual yield on the loan portfolio. A negative Subsidy Dependence Index shows that the institution generates sufficient profit to cover all subsidies at market interest rates, a zero value shows it is financially self-sustainable, and a positive value shows that it needs subsidies to survive. In view of the social objectives of development banks, a value of around zero is advocated.

5. MACRO-FRAMEWORK FOR DEVELOPMENT BANKS

The preceding sections have shown that the success of a development bank depends on a variety of dimensions, and that failure in any one of these can significantly undermine the bank. Drawing on the principles for these dimensions, a macro-level framework can be derived for the success of development banks (for a detailed discussion of the framework, see Thorne, 2008). The framework is illustrated in Figure 2 and can be summarised as follows:

• Their environment: Development banks need a climate of macro-economic stability without too many micro-economic distortions; they require political stability and a variety of complementary institutions. Although by definition their role is to address some of the weaknesses in the environment, they cannot succeed in a largely dysfunctional climate.

• Their role in it: They must be integrated into the financial system and operate along commercial lines, with a flexible mandate. They must not compete with the private sector, but rather aim to develop it. Once the private sector has the capacity to fund sectors previously funded by a development bank, the latter should be refocused on other areas of operation.

• How they are controlled: The ownership role of the state needs to be carried out circumspectly, allowing the bank to have operational autonomy while ensuring that it adheres to its mandate. The combined ownership and oversight role of government creates a potential conflict of interest that requires careful management. In general, the regulation and supervision of development banks should be along private sector lines.

• How they are run: Sound governance and management may be the single factor most likely to determine the success of a development bank. This involves issues such as the role and independence of the board, the accountability and capacity of management, the availability and retention of skilled staff, and sound operational, risk and financial management.

• How they are funded: The government needs to capitalise new (or restructured) development banks adequately, and then limit additional fiscal support to ring-fenced non-commercial activities undertaken on behalf of the state. Development banks should be encouraged to approach donors and obtain a credit rating to enable them to raise funds on the capital markets.

• Do they make a difference? Development banks should be assessed on a regular basis against an agreed set of objectives, both financial and social or developmental. Government must also be convinced that it could not have achieved these socially desirable outcomes in another (less expensive) way.

Although adhering to the principles underlying this framework could contribute significantly to the success of a development bank, some caution is required. In the words of Diamond (1996:11):

There is no such thing as permanent ‘success’. Perhaps success may be applied to a particular period of time but too much depends on the economic environment and on government policy, over which the institution has no control, and on its own (changing) management to apply the characterisation of ‘success’ over a very long period or with assurance that it will long continue to apply.

There is clearly a need for continued vigilance and flexibility in the governance and operations of development banks.

Figure 2: Macro-framework for development banks

6. CONCLUSION

The failure of many national development banks in the 1970s and 1980s led to them all but disappearing from the development agenda. However, many governments persisted with these banks, with mixed results. The success of some banks and the continued need for the services they provide have rekindled interest in national development banks. This raises the possibility that, without a clear understanding of the role of these banks, more failures could occur.

Despite this obvious need for a better understanding, however, the literature does not offer an updated and comprehensive assessment of all the factors that affect the performance of development banks. This article offers a starting point for such an assessment by providing a macro-framework for the successful functioning of development banks, derived from economic theory and informed by case studies. The analysis focused specifically on wholesale, state-owned, national development banks.

The framework sets out principles for six dimensions of development banking: enabling environment, mandate, regulation and supervision, governance and management, financial sustainability, and performance assessment. Since development banks operate under different conditions and in different markets, the framework can be adjusted to suit the development priorities of individual countries. For example, it has been applied to the South African development finance environment to derive policy recommendations. This case study will be the subject of a subsequent article.

Although it serves as a useful starting point, the framework would be enriched by additional detailed research on a range of concerns, including the following:

• A detailed study of each dimension and of particular principles, such as the appropriate conditions for a narrow or broad mandate.

• The interaction between development banks and regional or multilateral institutions such as the World Bank.

• The interaction between the various development banks in a country, and the shaping and coordination of their activities.

• Benchmarking the efficiency of development finance institutions.

• Sectoral analysis, for example, of agriculture or infrastructure development banks.

It is unlikely that development banks will gradually die off, as was once predicted, but their continued existence is no cause for complacency. The dangers inherent in their conflicting objectives mean that development banking remains a risky initiative for both the government and the bank. However, under the right circumstances, with appropriate supervision and governance, development banks can be a useful instrument for achieving the development objectives of a government and society. The macro-framework outlined above could help create just such an appropriate set of circumstances for these banks.

Notes