A quarterly magazine of the IMF

June 2006, Volume 43, Number 2

Adding Depth
Anne-Marie Gulde and Catherine Pattillo

Africa can achieve even higher growth through financial sector reform

Economic growth in sub-Saharan Africa (SSA) is forecast to be above 5 percent in 2006 for the third consecutive year. But a key obstacle to further growth is the region's weak financial sectors—among the least developed in the world. Sustained improvement in growth is contingent on the establishment of a supportive financial infrastructure. Sound, deep, and efficient financial sectors are critical for improving the business climate and creating the conditions for the private sector to be the engine of growth.

Financial sectors in low-income SSA countries are generally not performing adequately. The range of institutions is narrow, and most of the countries' assets are smaller than those held by a single medium-sized bank in an industrial country. Most people do not have access to even basic payment services or savings accounts, and the largest part of the productive sector cannot obtain credit. Limited finance lowers welfare and hinders poverty alleviation, and a lack of credit to the economy impedes growth. In addition, implementing monetary policy in the context of shallow markets is costly and inefficient. To date, the region's efforts to address these challenges have borne little fruit.

Through financial sector reform, SSA could build on its efforts to increase economic growth: effective financial sectors mobilize and pool savings; produce information on possible investments so that resources can be channeled to their most productive use; monitor the use of funds; facilitate the trading, diversification, and management of risk; and ease the exchange of goods and services. Because SSA is a high-risk environment exposed to terms of trade shocks and a volatile climate, it would benefit from greater risk sharing through portfolio diversification, consumption smoothing, and insurance, which are facilitated by financial development.

Progress since the 1990s

Banking is the most developed part of the financial sector in low-income SSA countries and accounts for more than 80 percent of the assets. Insurance, the few stock markets, nonbank financial intermediaries, and microfinance are all small sectors, although the latter two are growing rapidly. By contrast, SSA's few middle-income countries, whose performance is considerably better, have larger and more diversified financial sectors (see box).

SSA's banking systems have come a long way since the 1990s, when the region experienced a number of crises. At that time, politically motivated lending to public enterprises and political insiders, as well as a blind eye from supervisors, created the problem of rampant nonperforming loans, insolvent banking systems, and financial crises. Now, most of the region's banking systems are, on average, adequately capitalized and highly liquid. However, some countries' financial systems and many individual banks are still weak, with many banks failing to meet tests of basic capital adequacy, a sign of persistent problems in banking supervision. Average nonperforming loans in 2004 stood at 15 percent (see Chart 1).

Less diversified countries are more vulnerable to credit risk; a big shock to the major export sector could turn most of their banking loans bad. To reflect this risk, SSA countries need higher capital-adequacy ratios, although few set their minimum ratios above the standard 8 percent. Some frequently violate other prudential ratios, such as those limiting banks' exposures to a single client or sector, because of the shortage of bankable customers. Having rules on the books that everyone knows cannot be met, given the economies' structure, can unintentionally create a culture of noncompliance.

SSA banking systems are concentrated but are showing signs of becoming more competitive. Some concentration is inevitable because institutions must achieve economies of scale and scope. Many SSA economies can support only a small number of banks. Although the number of competitors in a banking market rises with the population and the economy's size, a number of SSA countries are outliers. Ethiopia—a populous country with one dominant commercial bank—is the most striking case. In contrast, Cameroon, Ghana, and Senegal have a large number of banks in relation to their population and income. SSA concentration ratios are generally moving in the right direction: they have declined in the wake of recent bank restructurings and privatizations. The growing presence of foreign banks—which represent a larger share of banking systems than in other low-income countries—also indicates some market contestability.

But the banking sectors of low-income SSA countries are not becoming more efficient. For all SSA banks, including foreign-owned institutions, low efficiency is reflected in high overhead costs and higher net interest margins than those prevailing in other low-income countries (see table). Net interest margins are used as an efficiency indicator, albeit an imperfect one, since banks with high operating costs need a high spread—the gap between what banks can charge borrowers and what they pay savers—to cover those costs. Cross-country research has found that high inflation, corruption, and concentration reduce bank efficiency (Detragiache, Gupta, and Tressel, 2005; Gulde and others, 2006).

Despite high overhead costs, SSA banks are profitable. Their main source of income is interest-related items. Given noncompetitive market structures, banks can charge high interest margins and remain profitable even in a difficult operating environment.

Economies are still cash based

Consumers in many parts of the world are increasingly using debit and credit cards and electronic banking, but cash transactions are still dominant in SSA, and Africa has made less progress than other regions in moving to noncash forms of money.

Banks' core business of financial intermediation—mobilizing deposits and on-lending them to borrowers—is less pronounced in SSA than in other low-income countries. Bank deposits in low-income SSA in 2004 were only 19 percent of GDP, compared with 38 percent on average in other regions (see Chart 2), and private sector loans were only 13 percent of GDP. The reasons for the sluggish growth in banking activities are manifold. In addition to legal and institutional weaknesses, the aftereffects of banking crises—which often caused banking activity to grind to a halt—also appear to have played a part. For example, private deposits and credit in the CFA franc countries have been stagnant since the sharp contraction that followed the banking crises of the late 1980s.

In more advanced countries, most financial sector activities take place between financial institutions, promoting market efficiency and the trading of risk; deepening interbank markets are signs of maturing financial systems. While interbank markets exist in most SSA countries, banks use them infrequently and trade in small amounts. One reason is that banks in many SSA countries have excess liquidity—cash holdings that exceed the prudential reserves required by the central bank, often because they do not find enough attractive projects to finance—and thus do not need to borrow short-term funds from other banks. Other reasons are the lack of instruments that can be used as collateral and the lack of mutual trust among banks, especially since it is hard to get information on financially weak ones.

Many countries are trying to improve their financial market infrastructure—the "plumbing" of the financial system—but have not yet achieved greater financial market activity. For example, the West African Economic and Monetary Union (WAEMU) and the Central African Economic and Monetary Community (CEMAC) have either recently put in place or are implementing state-of-the-art real-time gross settlement and retail payment systems to speed up check clearances. Most SSA countries that introduced treasury bills are moving from paper-based securities to more modern, book-entry systems. Once the remaining technical problems are resolved, the availability of these systems should contribute to financial deepening.

Difficult operating environment

Several factors impede progress in developing stronger financial sectors in the region.

Legal environment is weak. Legal and institutional frameworks are generally poor in SSA, and improvements have been slow. Legal systems are underfunded, and the public often has little confidence that legal proceedings are objective and will take a reasonable amount of time. A credit information index, measuring the ability of financial institutions to obtain information on client creditworthiness, and a legal framework index are lower for SSA than for other low-income countries. Studies have shown that improvements in these indices are strongly associated with higher shares of private loans to GDP (Gulde and others, 2006; Detragiache, Gupta, and Tressel, 2005).

Weak property rights and poor contract enforceability constrain financial market activity. Financial institutions are reluctant to lend because of the difficulty of securing collateral and seizing assets if loans default. Borrowers often have difficulty presenting collateral because of unclear land titles as a result of inadequate documentation and overlapping systems of rights and ownership. Registering titles of movable property (such as cars and other durable goods) is also problematic. For example, when borrowers in Rwanda and Senegal use movable property as collateral, they must often physically surrender it for the duration of the loan.

SSA businesses are ranked the lowest in the world on indicators necessary for efficient financial system operation: registering property, getting credit, protecting investors, and enforcing contracts. On average, to enforce a commercial contract through the courts, creditors must go through 35 steps, wait 15 months, and pay 43 percent of the country's per capita income before receiving payment. Banks' concerns about credit guarantees seem natural in such an environment.

Supervision is constrained. In a big stride forward, SSA countries have brought many regulatory and supervisory requirements into line with international norms. Actual supervision, however, is often constrained. Supervisors tend to be subject to political pressure and thus have little power to demand "prompt corrective actions." They also tend to exhibit substantial discretion for tolerating violations of prudential rules. This forbearance often reflects underlying pressures, such as banks' inability to meet prudential requirements (for example, on loan concentration) given structural features of the economy (limited lending opportunities) that are slow to change. Politically influenced supervisors also sometimes worry about the possible fiscal costs of bank restructuring. Resource constraints in supervisory agencies and the generally weak accounting and auditing systems in place also hamper supervision.

Rules-based monetary policy instruments impose costs on banks. Reserve ratios in SSA are high and have been climbing since the mid-1990s. In 2004, the average reserve ratio was 11.3 percent and ranged from 0 percent in the Central African Republic to about 50 percent in Zimbabwe. Increases in required reserves reflect the region's heightened focus on stabilizing inflation and maintaining financial system stability. With many SSA countries only partially remunerating required reserves, if at all, this instrument amounts to a heavy tax on banks. Because of the lack of remuneration, banks do not have the incentive to seek deposits or to develop products against which they must hold reserves. And empirical studies find that high reserve requirements contribute to high interest rate spreads.

Despite increases in required reserves over the past years, most SSA banking systems hold significant unremunerated excess liquidity amounting to more than 13 percent of total deposits, on average (see Chart 3). Why do banks hold these funds if they are not required to and receive no return on them? In some countries, persistent excess liquidity is linked to growing oil and aid inflows. Capital controls, structural problems in financial systems (such as interest rate restrictions, perceived limited and risky lending opportunities, and asymmetric information), and underdeveloped government securities and interbank markets also play a role. Excess liquidity is both higher and more volatile in oil-producing countries and is higher than the SSA average in the CEMAC and the WAEMU zones.

What can be done?

How can SSA countries make more rapid progress in their financial sectors so that the improved economic performance can be sustained? While many countries are taking measures to address some of the challenges—through promotion of microfinance and efforts to enhance entreprise financing and improve the operating environment—these efforts have yielded only partial results.

A number of priority actions can be identified. A harmonized approach to regulation in the context of low restrictions to market entry would allow financial firms to benefit from economies of scale and scope in larger markets. An elimination of distortions, such as the forbearance practice of bank supervisors, could improve banking soundness and facilitate greater interbank activities. Similarly, reducing the excessive use of costly monetary instruments (high reserve requirements) could spur development of the banking sector.

SSA countries could benefit from alternative instruments (such as leasing) or alternatives to collateralization (for example, group guarantees, reversible equity stakes) to overcome bottlenecks created by weak property rights. While governments' interest in promoting access to finance is well founded, new specialized state institutions should generally be avoided. Finally, evenhanded application of the legal framework—important for improving lending to the private sector—could benefit from development of commercial courts and, in some cases, specialized judges.