Journal Issue

Bridging the “Great Divide”

International Monetary Fund. External Relations Dept.
Published Date:
December 2003
  • ShareShare
Show Summary Details

Poorly developed financial systems in the CIS-7 countries may jeopardize their sustained growth

“A growing and deepening divide has opened up between transition economies where economic development has taken off and those caught in a vicious cycle of institutional backwardness and macroeconomic instability. This ‘Great Divide’ is visible in almost every measure of economic performance.”

Erik Berglöf and Patrick Bolton, Journal of Economic Perspectives

Berglöf and Bolton’s stark statement was prompted by the experiences in the 1990s of several transition economies, including the seven poorest countries of the Commonwealth of Independent States, known as the CIS-7. In the past three years, these transition countries—Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan—have achieved good progress in attaining macroeconomic stability. Yet any measure of financial development would show that the CIS-7 are still on the wrong side of the divide.

During 1995–2002, the CIS-7 countries grew faster, on average, than other transition economies although they started with lower per capita GDPs in 1995. Their faster growth thus represents a partial catching up with other transition countries. But the divide in financial development has not shrunk (see Table 1). In 2002, the CIS-7 countries recorded the lowest monetary depth, as measured by the average ratio of M2 to GDP, and the lowest depth of bank intermediation activity, as measured by the ratio of total banking sector assets, deposits, and credit to the private sector, to GDP. In fact, the gap between the CIS-7 and the best-performing transition economies—Central and Eastern European countries and the Baltic States (CEE+B)—has widened since 1995.

Table 1Measures of the divide

CIS-7 countries grew quickly during 1995–2002, but financial development did not advance.

Average real

GDP growth
M2/GDPBank assets/GDPDeposits/GOPCredit/GDP
Other CIS23.515.322.
Sources: IMF, International Financial Statistics (various years), and staff estimates.

Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan.

Belarus, Kazakhstan, Russia, Turkmenistan, and Ukraine.

South Eastern Europe: Albania, Bosnia and Herzegovina, Bulgaria, Federal Republic of Yugoslavia, former Yugoslav Republic of Macedonia, and Romania.

Central and Eastern Europe and the Baltic States: Croatia, Czech Republic, Hungary, Poland, Slovak Republic, and Slovenia, as well as Estonia, Latvia, and Lithuania.

Sources: IMF, International Financial Statistics (various years), and staff estimates.

Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan.

Belarus, Kazakhstan, Russia, Turkmenistan, and Ukraine.

South Eastern Europe: Albania, Bosnia and Herzegovina, Bulgaria, Federal Republic of Yugoslavia, former Yugoslav Republic of Macedonia, and Romania.

Central and Eastern Europe and the Baltic States: Croatia, Czech Republic, Hungary, Poland, Slovak Republic, and Slovenia, as well as Estonia, Latvia, and Lithuania.

This matters greatly because a sustainable growth path for the CIS-7 may well hinge, in large part, on jump-starting financial development. Why are these countries lagging other transition economies? The root of the problem lies in weaknesses in their institutional infrastructure, which hamper their ability to intermediate savings between lenders and borrowers. Although the CIS-7 have taken some steps to correct their institutional weaknesses, they must do much more if they are to catch up with their more successful peers. Policymakers in these countries must turn their attention to strengthening the legal and judicial framework, implementing accounting and auditing standards, improving governance in the banking sector, and privatizing state-owned banks.

Roots of the divide

To help explain the origins and costs of this great divide, we examine differences in lending-deposit spreads between the CIS-7 and other transition countries. In 2002, average spreads for the CIS-7, although narrower than those in the other CIS countries, were about 200 basis points wider than those in the South Eastern European (SEE) countries and 730 basis points wider those in the CEE+B countries (see Table 2). Levels and cross-country dispersion of spreads have been declining in all transition countries, reflecting the significant reduction in inflation and macroeconomic volatility.

Table 2The gap has narrowed only slightly

Lending-deposit spreads in the CIS-7 have declined but still exceed those in the better-performing transition economies.

Other CIS63.338.120.823.229.225.315.713.5
Standard deviation26.312.
Sources: IMF, International Financial Statistics (various years), and staff estimates.Note: For countries in each group, see Table 1.
Sources: IMF, International Financial Statistics (various years), and staff estimates.Note: For countries in each group, see Table 1.

For a given inflation rate, interest rate, and phase of the business cycle, the level of spreads is determined by three factors: funding, operating, and regulatory costs; rents accruing from banks’ market power on both the lending and the deposit sides; and credit risk. High spreads may raise the cost of credit, which discourages investment, or lower the rate of return on deposits, which discourages intermediation of savings, or both. What accounts for the hefty 730 basis point differential in spreads between the CIS-7 countries and the most advanced transition economies, the CEE+B group? It is useful to examine differentials in deposit and lending rates separately.

When countries’ deposit rates are compared, differences in inflation rates and short-term interest rates need to be taken into account. In 2002, average real deposit rates in the CIS-7 countries were 480 basis points higher than in the CEE+B countries. The difference is partly due to the higher short-term interest rates in the CIS-7 countries. The remaining difference, however, cannot be attributed to lower competition in CIS-7 deposit markets because banks with market power would be able to offer low returns and obtain deposits at low cost. Nor can it be attributed to the higher costs CIS-7 banks incur in providing financial services because banks passing costs on to depositors would pay them lower rates. The likely explanation is that CIS-7 banks must pay a “confidence premium” to compensate depositors for the risk that they may be unable to redeem their deposits at par in some circumstances. Higher bank funding costs are then passed on to borrowers in the form of higher lending rates.

With a difference of 480 basis points between the average real deposit rates of CIS-7 and CEE+B countries, the corresponding differential in lending rates is a huge 1,210 basis points, about 40 percent of which is attributable to the CIS-7 banks’ higher funding costs. We examine the factors affecting lending rates—namely, operating and regulatory costs, market power on the lending side, and credit risk—to determine which ones account for the remaining 60 percent.

Banks in the CIS-7 might incur higher operating costs because they are not large enough to exploit economies of scale. As a result, borrowers would pay higher lending rates. However, the ratio of operating costs to income in the CIS-7 was about 59 percent during 1999–2001, as opposed to about 67 percent in the CEE+B countries, making it unlikely that operating costs explain the difference in lending rates. Although it has been argued that banks in transition countries are likely to exploit scale economies through mergers, recent econometric evidence for a large number of banks in the second half of the 1990s does not support this argument: banks with larger market shares were found to incur higher costs. In general, the relationship between banking system structure, bank efficiency, and soundness is more complex than policy discussions often suggest (see box, page 44).

Banks in the CIS-7 countries might also incur higher regulatory costs (such as high reserve and capital requirements) that they pass on to borrowers. As of the end of 2002, the ratio of capital to assets in the CIS-7 was about 20 percent, compared with about 10 percent in the CEE+B countries. While higher ratios sometimes result from policies designed to strengthen banks’ financial positions, there is a potential trade-off between prudential policies and the cost of credit. These higher ratios may account for some of the differential in interest rates.

Other things being equal, the stronger the banks’ market power, the wider their lending-deposit spread, because they can extract rents from borrowers, depositors, or both. Although market power is difficult to measure, simple bank concentration measures shed some light on it. When a small number of banks dominate the market, those banks may extract rents from clients because of the size of their exposure to (and connections with) large borrowers and the ease with which they might collude in setting prices and avoid competing for each other’s market share. High bank concentration can be the result of restrictions on bank entry or substantial state ownership of banks. But average bank concentration is not greater in the CIS-7 countries than in the CEE+B countries, while state ownership of banks is predominant in only two of the CIS-7 countries (Azerbaijan and Uzbekistan). Thus, differences in market power rents and state ownership are also unlikely to explain the differential in lending rates.

Is there an optimal banking system structure?

Several transition economies are considered “overbanked”—that is, they have too many small banks. It is believed that when an economy is overbanked, consolidation should be strongly encouraged or even forced.

It is important to distinguish bank consolidation through the closure of nonviable or failed banks from consolidation through mergers. Efficiency and soundness arguments strongly support the exit of failed banks. This type of consolidation has occurred in some CIS-7 countries and other transition economies, often as a result of crises that have exposed lax entry policies and weak regulatory standards.

As for bank mergers, an argument in their favor that is based on efficiency and soundness hinges on the identification of an optimal banking system structure. Yet such identification, in terms of number of banks or bank size, is difficult for at least four reasons.

• A banking system would not necessarily be more efficient or stable if economies of scale could be exploited at the firm level; there is a lack of robust evidence that larger banks in both developed and developing economies enjoy higher profitability, lower costs, and more prudent risk management.

• The minimum efficient bank size crucially depends on the ability of the bank’s management team to control the scope and composition of its activities. In a country with a weak credit culture, underdeveloped risk-management capacity, and accounting opacity, managerial diseconomies of scale might set in, even in relatively small banks.

• If economies of scale exist at a system level, it is unlikely that any subset of banks in the system is going to exploit them. For example, an adequate payment system infrastructure is a public good in that it benefits all banks by providing enhanced services at a lower cost. Setting up such an infrastructure typically requires public investment, because no subset of banks is likely to have the incentives, skills, and funds to invest in an activity that does not yield exclusive private benefits.

• Increased concentration in banking may have undesirable implications for competition and soundness, because higher lending rates arising from monopoly rents in the loan market may induce borrowers to take on excessive risk, which, in turn, may worsen the quality of bank loan portfolios.

Finally, we consider how credit risk influences spreads. Credit risk can be high because of a recession or macroeconomic uncertainty—reflected in high inflation and interest rate volatility—and because of structural factors. These factors include the lack of a credit culture, which inhibits the adoption of best practices in lending and of risk-management technologies, as well as weaknesses in the institutional infrastructure. Such weaknesses prevent banks from efficiently carrying out their monitoring functions, such as assessing borrowers’ creditworthiness (because reliable accounting standards are lacking), and from repossessing collateral rapidly if a borrower defaults (because creditors have weak legal protection). Although we believe credit risk largely explains the differential in lending rates between CIS-7 and CEE+B economies, credit risk is unlikely to arise from macroeconomic conditions. CEE+B countries have a better track record of macroeconomic stability, but both groups of countries are in an expansionary phase at present; their inflation is, on average, low, and macroeconomic volatility is subdued almost everywhere. Rather, credit risk tends to arise from a poor institutional infrastructure or one in which intermediation is costly. (See Table 3 for a comparison of institutional quality in groups of transition economies.)

Table 3How good are CIS-7 institutions?

Poorer institutional quality in the CIS-7 leads to higher credit risk, weak supervision, and lags in bank restructuring.

(CEE+B = 100)

Regulatory qualityRule of lawControl of corruptionVoice and accountabilityPolitical stabilityGovernment effectiveness
Other CIS38.548.745.249.049.639.2
Source: Kaufmann, Kraay, and Zoido-Lobaton, 2002.Notes: For countries in each group, see Table 1. Figures are for 2000/2001.
Source: Kaufmann, Kraay, and Zoido-Lobaton, 2002.Notes: For countries in each group, see Table 1. Figures are for 2000/2001.

In sum, little bank intermediation takes place because potential depositors lack confidence in the banking system, and lenders face nontransparent borrowers, a weak credit culture, and poor or costly enforcement of financial contracts. The public prefers to keep savings either in cash (usually in foreign currency) or, if feasible, in overseas banks. These problems make it difficult to monitor credit risk, and they raise the cost of credit, possibly generating credit rationing.

Policy priorities

Policymakers could learn a great deal from this cross-country examination of lending and deposit spreads. Those transition economies that were the quickest to remove impediments to banking development have achieved the most economic growth (Wachtel, 2003). These impediments were a weak legal and judicial framework (in particular, weak protection of creditor rights), poor financial sector supervision or enforcement of prudential standards, lags in the resolution of unviable or failed banks, weak accounting and auditing standards, inadequate governance in the banking sector, and state ownership of banks.

Although some CIS-7 countries have strengthened their legal and judicial frameworks, their work is not finished. Particularly pressing are the problems several CIS-7 central banks face in enforcing rules and regulations—for example, closing failed banks. Legislation incorporating best practice bankruptcy and creditor resolution procedures is either not yet in place or is unevenly or poorly enforced, partly because the courts and judges have little technical expertise in financial matters. While some CIS-7 countries are committed to introducing financial regulations and supervision practices consistent with international best practices, their weak legal and judicial systems may jeopardize implementation.

Responding to the lack of confidence in their banking systems, some CIS-7 countries have sought to introduce deposit insurance schemes. But they have not yet met three key preconditions for effective deposit insurance: a sound banking system, effective bank supervision, and legal certainty. Accounting practices make it difficult to judge the true financial strength of institutions, and some banks are nonviable. Until the countries address these weaknesses, it is premature and highly risky to introduce deposit insurance schemes.

Some CIS-7 countries have moved aggressively to strengthen accounting and auditing standards and plan to adopt the International Accounting Standards for both financial and nonfinancial firms. But for implementation to be effective, these countries need accountants and auditors with technical expertise who can understand and use the standards.

Governance in the banking sectors of most CIS-7 countries is still weak. Nontransparent ownership structures favor connected lending, which may heighten risks or crowd out funds that can be lent to finance productive projects. In addition, opaque bank ownership structures, resulting, in part, from the lack of enforcement and monitoring of “fit and proper” criteria, prevent the emergence of a market for corporate control. The lack of such a market is likely to be one key reason for the paucity of market-driven bank reorganizations in the CIS-7 countries, including entry of foreign banks, joint ventures, and mergers of existing viable banks.

Cross-country evidence indicates that state ownership of banks is associated with slow financial development, slow growth of per capita income, and slower growth of productivity in the nonfinancial sector (La Porta, Shleifer, and Lopez-de-Silanes, 2000). Intensifying efforts to privatize state-owned banks while ensuring that the financial services supplied by those banks continue to be available should be high on the financial sector reform agendas of CIS-7 countries. Transition economies that have privatized and liberalized entry for legitimate financial firms have witnessed substantial entry of foreign banks in their markets. Foreign institutions have speeded up rationalization in the provision of financial services and increased overall transparency and competition.

The banking sector is the most important financial sector in all transition economies. In most of them, several elements of financial development are missing—namely, nonbank intermediaries and capital markets. In theory, balanced growth in financial development, where bank and nonbank intermediaries develop alongside capital markets, is the most desirable scenario. In practice, the development of nonbank intermediaries and capital markets requires a sophisticated legal and regulatory infrastructure, a widespread credit and saving culture, and the institutional capacity for the efficient enforcement of financial contracts. As a result, all transition economies face important trade-offs in institutional capacity building. For the CIS-7 countries, cost-benefit analyses of development in nonbank intermediaries and capital markets aimed at detecting components of financial sector reforms with the highest value added appear necessary in identifying policies likely to be successful in bridging the great divide.

Gianni De Nicoló is a Senior Economist, Sami Geadah is an Advisor, and Dmitriy Rozhkov is an Economist in the IMF’s Monetary and Financial Systems Department.

This article draws on IMF Working Paper 03/205, “Financial Development in the CIS-7: Bridging the Great Divide,” by Gianni De Nicoló, Sami Geadah, and Dmitriy Rozhkov.


    BoninJohn and PaulWachtel2003“Financial Sector Development in Transition Economies: Lessons from the First Decade,”Financial Markets Institutions and Instruments Vol. 12 (February) pp. 166.

    • Crossref
    • Search Google Scholar
    • Export Citation

    BerglöfErik and PatrickBolton2002“The Great Divide and Beyond: Financial Architecture in Transition,”Journal of Economic Perspectives Vol. 16 (Winter) pp. 77100.

    • Crossref
    • Search Google Scholar
    • Export Citation

    De NicolóGianniPhilipBartholomewJahanaraZaman and MaryZephirin2003“Bank Consolidation, Internationalization and Conglomeration: Trends and Implications for Financial Risk,”IMF Working Paper 03/158 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation

    FriesStevenDamienNeven and PaulSeabright2002“Bank Performance in Transition Economies,”EBRD Working Paper No. 76 (London: European Bank for Reconstruction and Development).

    • Search Google Scholar
    • Export Citation

    KaufmannDanielAartKraay and PabloZoido-Lobaton2002“Governance Matters II: Updated Indicators for 2000/01,”Working Paper No. 2772 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    La PortaRafaelAndreiShleifer and FlorencioLopez-de-Silanes2000“Government Ownership of Banks”NBER Working Paper No. 7620 (Cambridge, Massachusetts: National Bureau for Economic Research).

    • Search Google Scholar
    • Export Citation

    WachtelPaul2003“How Much Do We Really Know about Growth and Finance?”Federal Reserve Bank of Atlanta Economic ReviewFirst Quarterpp. 3347.

    • Search Google Scholar
    • Export Citation

Gianni De Nicolo, Sami Geadah, and Dmitriy Rozhkov

Other Resources Citing This Publication