Serbia – Financial Sector Notesiteresources.worldbank.org/INTSERBIA/Resources/SerbiaFSN.pdf · The team is grateful for the kind assistance of the Serbian authorities in the preparation

Serbia

Financial Sector Note

September 10, 2004

The World Bank

Finance and Private Sector Development Unit Europe and Central Asia Region

1

Table of Contents

Chapter 1: Introduction and Executive Summary .................................................................5

1. Motivation and Links to Other Bank Work.................................................................... 5 2. Outline of the Report ...................................................................................................... 6 3. Key Policy Recommendations ....................................................................................... 9

Chapter 2: Legacies of the 1990s and Actions to Date ......................................................11 1. The Disrupted 1990s..................................................................................................... 11 2. Early Policy Responses ................................................................................................ 13 3. Present Structure of the Banking System ..................................................................... 17 4. Benchmarking Serbian Financial Sector ...................................................................... 19

Chapter 3: Assessing Bank Performance...........................................................................23 1. Description of Recent Trends ....................................................................................... 23 2. Comparing bank performance ...................................................................................... 26 3. Competition and Sustainability of Growth................................................................... 30 4. Conclusions on key issues ............................................................................................ 35

Chapter 4: Implications for Policy on Banking Sector Reform............................................36 1. Policy Agenda .............................................................................................................. 36 2. Approaches to resolution of state-owned banks ........................................................... 37 3. Re-thinking the State-Bank Problem............................................................................ 39 4. Prioritizing the Supervisory Agenda ............................................................................ 42

Chapter 5: Non-Bank Financial Institutions ........................................................................44 1. Overview of Serbia’s NBFI sector development.......................................................... 44 2. Insurance Sector ........................................................................................................... 46 3. Leasing ......................................................................................................................... 50 4. Capital Markets ............................................................................................................ 51 5. Proposed reform agenda for NBFIs.............................................................................. 53

Chapter 6: Finance and the Real Economy........................................................................57 1. Patterns of Enterprise Financing in Serbia ................................................................... 58 2. Structuring the Problem and Proposed Solutions ......................................................... 61 3. Issues related to Service Innovation ............................................................................. 63 4. Role of Government-Sponsored Programs................................................................... 67

2

List of Acronyms

BEEPS - Business Environment and IOSCO - International Organization of Enterprise Performance Securities Commissions Survey IPO - Initial Public Offering BRA - Bank Rehabilitation Agency MOF - Ministry of Finance BSE - Belgrade Stock Exchange MS - Market Share CAR - Capital Adequacy Ratio NBFI - Non-Bank Financial CEE - Central and Eastern Europe Institutions DF - Development Fund NBS - National Bank of Serbia EAR - European Agency for NBY - National Bank of Yugoslavia Reconstruction OECD - Organization for Economic EBRD - European Bank for Cooperation and Development Reconstruction and OTC - Over-the-Counter Development PA - Privatization Agency ECA - Europe and Central Asia PFSAC II - Second Private and Financial EIB - European Investment Bank Sector Adjustment Credit EU - European Union PICS - Productivity and Investment FRY - Federal Republic of Climate Survey Yugoslavia PLC - Paris and London Club FSU - Former Soviet Union PSN - Private Sector Note GDP - Gross Domestic Product RS - Republic of Serbia GOS - Government of Serbia SDP - Supervisory Development IAIS - International Association of Plan Insurance Supervisors SFRY - Socialist Federal Republic of IAS - International Accounting Yugoslavia Standards SME - Small and Medium ICA - Investment Climate Enterprises Assessment SOE - Socially-owned Enterprises IFI - International Financial SSC - Serbian Securities Institutions Commission IFRS - International Financial UNICITRAL - UN Commission on Reporting Standards International Trade Law IMF - International Monetary Fund UNIDROIT - International Institute for the Unification of Private Law VAT - Value-added Tax

3

Preface

The purpose of this Financial Sector Note (FSN) is to review recent developments in Serbia’s financial system and to provide policy recommendations for its future development. The FSN builds upon, complements, and seeks to expand, the broad financial sector reform agenda pursued under the ongoing Bank program, which is anchored on two adjustment operations (PFSAC I and II) and parallel technical assistance efforts. This report was prepared by a team comprising Alexander Pankov (Task Team Leader, ECSPF), Angela Prigozhina (Financial Sector Specialist, Ukraine Country Office), Branko Radulovic (Consultant, Serbia Country Office), Anna Sukiasyan (Consultant, ECSPF), Alan Roe (Consultant, Oxford Policy Management), and Stephen Peachey (Consultant, Oxford Policy Management). The work was carried out under the general supervision of Gerardo Corrochano (Sector Manager, ECSPF). The team received valuable comments from Lajos Bokros (ECA VP), Balazs Horvath (IMF), Michael Edwards (ECSPF), and Andrew Lovegrove (UK DFID).

The team is grateful for the kind assistance of the Serbian authorities in the preparation of the report. We especially wish to thank Governor Radovan Jelasic and former Vice-Governor Dusan Stojanovic at the National Bank of Serbia for their cooperation and continued interest in this work. Much of this report’s analysis could not have been completed without the extensive data on bank performance indicators that were kindly provided by Messrs. Djordje Jeftic and Mirko Lovric, and their staff at the NBS Supervision Department. The team also benefited from helpful insights on the state of Serbian financial sector readily shared by representatives of local commercial banks and other financial institutions, multilateral and bilateral donors, professional associations, and private sector firms.

4

Chapter 1: Introduction and Executive Summary

The main objectives of this Financial Sector Note (FSN) are:

• To provide an up-to-date snapshot of the Serbian financial sector (banks as well as non-bank financial institutions (NBFIs)), including an in-depth assessment of the performance of the banking sector by analyzing both the efficiency and stability characteristics of individual Serbian banks.

• To identify existing constraints to financial intermediation, as well as potential systemic risks.

• To reinforce previous messages to the Serbian authorities and the Bank’s internal audience about the urgent necessity of implementing the next stages in financial sector reform.

1. Motivation and Links to Other Bank Work

Beginning in early 2001, the Serbian authorities, with support from the World Bank (the Bank), IMF, and other donors, embarked on an ambitious reform program that has brought a number of radical (albeit incomplete) changes in the areas of fiscal management, banking reform and enterprise privatization. These early reforms have laid a strong foundation for economic revival and growth. However, the Bank and the authorities recognize the limitations of an agenda that has necessarily had to emphasize stabilization and urgent restructuring. The preparation of the first full-fledged Country Assistance Strategy (CAS) for Serbia (FY05-07) requires a better understanding of the key determinants of, and constraints to, growth and employment. The crucial problems to be addressed include:

the failure of growth since 2000 to generate vigorous employment response and

significant improvement in living standards; the threat of higher unemployment in a setting where the further large-scale

reallocation of labor to more efficient uses is still needed as one driver of economic growth; and,

the possibly unsustainable nature of recent growth in the face of tighter macroeconomic constraints.

The Bank’s recent Country Economic Memorandum for Serbia (CEM) undertakes an in-depth assessment of reform priorities to reflect this new emphasis, including a better strategic view of the role of the financial sector. As witnessed by transition experiences elsewhere, a deeper, more stable and more efficient financial sector can play a critical role in facilitating the recovery of the real economy. The potential for a significant contribution of the financial sector to the broad objectives of the next CAS provides much of the motivation for this present assessment. This potential is based on two main propositions, namely:

i. the hyperinflation and other major disruptions of the 1990s massively reduced the level of monetization and financial sector depth in the Serbian economy. However, as experience in neighboring transition countries has shown, these losses can be

5

rapidly recovered given appropriate policies (or perpetuated if policies are incorrect). The possibility of achieving a rapid re-deepening of the financial sector is one that could contribute significantly to the broader objectives of growth and employment creation, primarily through the improved allocation of capital to productive use in the economy and through larger volumes of real credit growth for the emerging private sector.

ii. in 2001 and early 2002, the Serbian authorities have demonstrated their ability to carry out bold and rapid adjustments to their anarchic and distorted banking sector structure. The subsequent prolonged period of political turmoil has resulted in significant delays in implementation of the next round of reforms. Yet the current government’s stated commitment to accelerating the reform of financial sector gives hope that Serbia could see rapid financial re-deepening rather than an extended period of financial sector stagnation that has haunted some economies in the region.

This report should thus be viewed as part of the broader ongoing assessment of how to deepen the reform process in order to achieve faster growth and a greater impact on living standards. The findings of the FSN will serve as inputs into the design of the financial sector sections of the CEM and CAS for Serbia. Very importantly, the report endeavors to provide specific recommendations for the Bank’s future support for financial sector development in Serbia, with the focus on identifying potential policy components in planned adjustment operations. Finally, the FSN is expected to lay the analytical and quantitative groundwork for the relevant sections of the FSAP report planned for 2005.

2. Outline of the Report

Because the banking sector is presently such a dominant part of the overall financial system of Serbia (see Table 1.1), the core of the FSN will relate to an analysis of the performance of the banking sector and how it can better service the needs of the real economy.

Table 1.1 – Overview of Serbia’s financial sector

2001 2002 20031 Banking Number of banks 86 50 47 Total assets (millions of Dinars) 891,947 330,494 377,264 Total assets (millions of US$) 13,181 5,603 6,905 Assets as a share of GDP (%) 113.63 35.69 35.81

Non Bank Financial Institutions Number of insurance organizations 33 36 39 Total premiums (millions of Dinars) 17,350 21,545 23,336 Total premiums (millions of US$) 256 365 427.1 Premiums as a share of GDP (%) 2.9 3.07 2.22

Number of leasing companies2 0 1 6 Total assets (millions of Dinars) 0 … 2,070 Total assets (millions of US$) 0 … 36 Assets as a share of GDP (%) 0 … 0.28

Number of non-bank credit institutions … … 14 Total assets (millions of Dinars) … … 1,340 Total assets (millions of US$) … … 24 Assets as a share of GDP (%) … … 0.19

Number of non-state pension schemes3 … … 6 Serbian Development Fund –total assets (Dinar mln) 87,056 107,590

Source: NBS, Ministry of Finance, Department of Insurance Supervision, Serbian Development Fund 1 Banking sector, insurance and leasing indicators are as of year-end 2003, remaining NBFI data are as of June 2003

62 Experts estimate 3 Data are not available/non-existent

Chapter 2 starts by noting the uniquely unfavorable circumstances preceding Serbia’s transition. A decade of sanctions and mismanagement resulted in huge macroeconomic imbalances, large-scale demonetization of the economy, and a discredited and deeply distressed banking sector. Recognizing the importance of a functioning financial sector to reviving the economy, the new Serbian authorities, with active support from the Bank and other donors, conducted the first stages of a banking system clean-up, culminating in the closure of the four largest troubled banks in early 2002. In parallel, initial steps were taken towards the modernization of the regulatory and institutional framework for financial sector operations.

The volume of Serbia’s monetary aggregates and levels of financial depth have both shown some early improvement over the past three years in response to the reform-minded actions of the authorities and the macroeconomic stabilization that has been achieved. In particular, a number of banks have led a spectacular growth in foreign currency deposits, and credit to non-government sector has also started to increase. However, as shown by the chapter’s benchmarking of Serbia against comparator economies in the region of Central Europe and the FSU, the country’s financial sector remains severely underdeveloped and the existing level of financial intermediation is still unable to contribute fully to the economic growth agenda. The banking system is overpopulated and segmented, dominated by local banks with their historic problems, and characterized by an unsustainably high level of state ownership.

The bank-by-bank analysis of performance presented in Chapter 3 tracks specific reasons why Serbian banks in general, and certain categories of banks in particular, are so far failing to deliver adequate volumes of intermediation to the economy. Looking at the variety of bank performance indicators for the period from the beginning of 2002, this analysis demonstrates that the cost, profit and other efficiency parameters of Serbian banks, especially of those in the state-owned sector, have more in common with the weaker countries of the FSU than with the neighboring Central European economies. Perhaps more worryingly, the report finds no apparent correlation between the level of operational effectiveness and gains in market share for those (mostly domestic private) banks which have seen the fastest balance-sheet growth in recent years. On the contrary, there is some evidence of a positive correlation (not the expected negative correlation) between the rate of market share gain achieved by some private banks and their combined spread and fee yield. This could possibly indicate a supply-constrained environment where the banks capturing most of the growth in the market are able to charge a premium for the availability of much sought after services (particularly credit). It suggests that the Serbian financial system, with too many banks yet half of its assets in state-owned limbo, needs more genuine competition in order to keep pricing under control and facilitate increased efficiency of bank operations. Chapter 4 examines the main policy levers that the authorities have available to re-direct banking development in order to increase competition and close the above-mentioned efficiency gaps. This requires the creation of the conditions under which the weaker banks in the system, as defined by efficiency levels and including, but not limited to state-owned banks, lose ground rapidly to the better banks. That shift in the structure of the sector will require either the liquidation of more of the weaker banks, or their rapid re-invention, through new strategic investment, as stronger banks. Based on the analysis of bank performance indicators, the report attempts to categorize the existing state-owned banks 7

and then suggests possible resolution scenarios for each group. Specifically, it is noted that a handful of larger majority state-owned banks have relatively good effectiveness ratings, despite their problems with historic loans. Their revival through speedy privatization to strategic investors would help to introduce more viable competition at the top end of a market that today appears to be supply constrained. At the same time, the exit of the state from the banking sector activities would eliminate the well-known risks associated with state banks that tend to worsen the prospects for competitive market development and often incur significant macroeconomic and fiscal costs. The spotlight given by the authorities and the donor community to the resolution of majority state-owned banks should in no way distract attention from the potential problems posed by expansion of domestic private banks. None of the local banks driving the recent growth score unambiguously well on a mix of effectiveness, regulatory, or loan-quality indicators. This gives rise to the major concern that some of the larger local private banks are storing up future problems by lending on unsustainable terms that are likely to result in future deterioration in asset quality. In particular, the expected convergence of market, financial and regulatory pressures, such as the increased minimum capital requirement, removal of central government deposits from commercial banks, and acceleration of bank privatization, all call for stronger supervisory discipline across the whole of the banking sector. Given that Serbia had 47 licensed banks at the end of 2003, an orderly consolidation of the banking system should be seen as the end objective of a strengthened regulatory and supervisory regime. Chapter 5 provides an overview of the NBFIs sector, followed by a more in-depth look at insurance, leasing, and capital markets, which represent the most active and rapidly growing segments of the sector. The report demonstrates that Serbia’s NBFI sector is still underdeveloped and faces a number of legal, regulatory and institutional constraints. Meanwhile, the experience of other countries demonstrates that NBFIs can offer a broader array of financial services and instruments that are often unavailable from banks. Some types of non-bank credit organizations are particularly well-suited for targeting the low-income and socially vulnerable groups of the population in urban and especially rural areas, thus positioning themselves in market niches usually unattractive for mainstream banks. At the same time, the experiences of other transition countries show that poorly regulated NBFIs can easily become breeding grounds for serious financial malpractice and even systemic crises. Hence, although Serbian NBFIs are currently small relative to the banking sector, they could give rise to important risks. The chapter lays out a broad medium-term policy agenda aimed at increasing the sector’s role, efficiency, and stability. Finally, Chapter 6 considers the challenge of making the financial sector (both banks and NBFIs) more responsive to the needs of the real economy. The chapter first provides a summary of the patterns of financing used by various parts of the enterprise sector. Today’s status quo is one in which many banks, primarily those in state ownership, are seriously fettered by their backlogs of non-performing loans to large socially-owned enterprises. At the same time, most private and foreign banks find the alternative of servicing the emerging private SME sector unattractive in the present imperfect legal and institutional environment.

8

The chapter then looks selectively at the main dimensions of the task of enhancing the delivery (volumes and terms) of financial services to the growing private sector. It is argued that, in parallel to the debt workout needed to clean up the balance sheets of large banks

and SOEs, the authorities need to promote the process of service innovation, aimed at creating a more customer friendly and market-oriented culture in all surviving banks so as to extend the breadth and depth of coverage of banking services to newer private forms of productive activity. This task involves both: (i) the changes in the internal operational efficiencies of the banks themselves and in the way in which they regard private sector lending; and (ii) the provision of new institutional arrangements (public goods) that can facilitate lower cost banking and make the banks more willing to offer improved services to private sector clients. Separate sections are provided on the possible catalytic role of donor-supported credit schemes, and on benefits and risks of state-sponsored institutions such as the Development Fund.

3. Key Policy Recommendations

Overall the Serbian financial sector has made impressive progress over the past three years. The government’s reform agenda to date has been concentrated on stabilization and urgent first stage restructuring in the banking sector and parallel but also incomplete reforms in the regulatory and institutional framework for financial sector operations. However, significant further policy decisions are needed – many urgently – in order to create the dynamic for the sustainable channeling of financial resources to serve the needs of a growing economy on an efficient and affordable basis. The team recommends that in the next six to eighteen months the authorities, supported by the Bank and other donors, expand and deepen the FSD reform agenda by focusing on the following key issues, listed for each area in the approximate order of priority:1

• Completing resolution of state banks through: (i) conversion into state equity of the remaining London Club obligations (both principal and interest) ahead of the banks’ resolution; (ii) adoption of a divestiture strategy that would identify clear, time-bound resolution scenarios (privatization, merger, or liquidation) for all state-owned banks (including minority holdings), based on the careful consideration of costs and benefits in each particular case; (iii) transparent privatization of viable large state-owned banks to strategic investors through open tender process; (iv) liquidation or merger of all other state banks that cannot be sold and/or are unlikely to compete effectively even if their balance sheets could be cleaned up; and (v) sale of minority state stakes and divestiture of assets of banks in liquidation.

• Strengthening regulatory and supervisory regime across the banking system

through: (i) enhancement of NBS supervisory capacity by timely implementation of the Supervisory Development Plan; (ii) in the period leading to resolution, proper fiduciary oversight of state-owned banks through coordinated efforts of BRA, NBS and MOF; (iii) on-site examinations at the fastest growing private banks, with appropriate remedial action as may be necessary; and (iv) improvement of legal and institutional framework for bank exits.

• Laying the foundations for development of a viable NBFI sector through: (i)

urgent establishment of insurance sector supervisory capacity and examination of

9

1 These policy recommendations are reviewed and elaborated in greater depth in the concluding sections of Chapter 4 (banking sector), Chapter 5 (NBFIs) and Chapter 6 (access to credit) of this report.

insurers with the view to facilitating market consolidation; (ii) restructuring and subsequent privatization of the two largest state-owned insurance companies; (iii) adoption of medium-term development strategy for NBFI sector, highlighting the objectives and expected results of the reforms for various types of NBFIs (capital markets, leasing, mortgage providers, pension funds); and (iv) based on the above strategy, address the remaining regulatory obstacles and gaps.

• Increasing the volume of credit to real sector through: (i) improved legal and

institutional environment for lending, focusing on the establishment of a collateral registry, reform of the real estate cadastre, and strengthening of the judiciary’s enforcement capacity; (ii) encouraging changes in the banks’ internal culture and procedures to make the funds more readily available and affordable to enterprises; (iii) minimizing the crowding out effect by eliminating the directed lending to SOEs through state-sponsored institutions; and (iv) promoting the external credit lines through Serbian commercial banks.

10

Chapter 2: Legacies of the 1990s and Actions to Date

1. The Disrupted 1990s

Serbia’s financial sector was hugely impacted by the political and economic instability of the period from the hyper-inflation at the end of the 1980s to the end of the millennium. The politically-distorted and war-influenced management of the economy during those years created large imbalances in both the external and fiscal accounts. These imbalances in turn contributed to the accumulation of large domestic and external debt obligations that negatively impacted the major Serbian banks through which many of those debts were channeled. These were also major factors causing the second huge inflation after 1992. At the same time, political sanctions from mid-1992 through the change of government in October 2000 eliminated external financial support from the International Financial Institutions (IFIs). This required Serbia’s finances to rely on a range of severely distorting interventions, such as the freezing of some assets and the tolerance of payment arrears.

Monetary Aggregates and Financial Depth

During this unstable period of the late 1980s and 1990s, the growth rate of the money supply was both extremely rapid and also highly volatile. It soared in 1991 and 1992 and was still growing at around 700 percent per annum between 1995 and 2000. Not surprisingly, annual inflation for the period 1995-2000 was also high, averaging at around 50 percent even after the hyper-inflation. The monetary aggregates expressed as a percentage of GDP declined sharply as Serbs responded by increasingly substituting out of the domestic money: M2 was as low as 11 percent of GDP by end-1995, for example.2 On average, the level of financial depth remained at about half the levels seen in the 1980s. The only significant increase took place in the form of foreign exchange deposits, which are accounted for in M3, while Dinar deposits plummeted.

The high inflation rate that wiped out the real value of Dinar deposits together with the freezing of foreign currency deposits, provided a powerful cocktail to discourage the use of domestic money and domestic banks and to encourage high levels of currency substitution. Domestic transactions became largely based on hard currency even in everyday life.

The Burden on Commercial Banks

Most problems seen in Serbia’s banking sector today stem substantially from the instability and the methods of fiscal financing used during the unstable period through the late 1990s. Surprisingly, the explicit funding of the government fiscal deficit contributed little to recorded monetary expansion. But much of the necessary domestic credit expansion was associated with the quasi-fiscal deficits of state and socially owned enterprises and government agencies, together with the creation of inter-enterprise debts and guarantees.

2 The interpretation of the monetary aggregates is complicated by the revaluation effects on the foreign currency deposits. For example, the narrow money supply (M1) at end 1995 represented about 7 percent of GDP, but broad money (M3), including frozen foreign currency deposits amounted to 34 percent. By end 2000 the M3 figure valued at the then current exchange rate, had risen to almost 70 percent of GDP because of the revaluation of the foreign currency deposits.

11

The involvement of the banks in the country’s external debt management was merely one symptom of that situation (see Box 2.1).

Box 2.1 – Frozen Foreign Currency Deposits and External Debt

A. Frozen Foreign Currency Deposits Through most of the 1990s and until the Law on the Settlement of Public Debt of mid-2002, the frozen foreign currency savings deposits of households and others constituted a major liability of the Serbian banks. In 1991 most of these deposit accounts had been taken over by the government. The law of that time stipulated that all foreign exchange savings held at commercial banks had to be re-deposited into the NBY. Subsequently they represented assets on the commercial bank’s balance sheets in the form of claims on FRY. The corresponding commercial banks liabilities to the depositors remained in frozen foreign exchange accounts balances until the actions taken in mid-2002. The amounts as at end-2000 are shown in the Table 2.1 below. The 2002 Law moved the frozen savings deposits into an off-balance sheet status and so cleaned bank balance-sheets of these items. The deposits are now being re-paid on a gradual basis from 2000 on a semi-annual basis (in form of cash or negotiable bonds after 2002). The cost to the budget is estimated at 1.5 percent of GDP through 2004 but somewhat higher after that. B. External Debt The government’s long-term obligations to the Paris and London Club creditors had also been reflected in the balance-sheets of the domestic banks as a claim on government. As discussed in more detail elsewhere, these claims were swapped into shares in 2002, thereby, resulting in the government becoming a major stakeholder in eighteen banks. These together with the frozen foreign currency deposits, meant that more than 50 percent of total bank resources had previously been unavailable to be converted into income earning assets.

Table 2.1 – Monetary Aggregates (as of December 2000) Dinar billion % of GDP

Broad Money (M3) 249.6 69.1

Money Supply (M1) 26.6 7.4

Currency outside banks 10.9 3.0

Demand deposits 15.7 4.4

Time and savings deposits 5.4 1.5

Foreign currency deposits 217.5 60.3

of which: Frozen foreign currency deposits 185.2 51.3

Nominal GDP 188.0 100

Source: Data on Serbia only from the NBS

Because the banking sector (including the National Bank) was used to extend directed credits to state and socially-owned companies, today’s extensive financial distress across key parts of the banking sector is directly associated with the fiscal aspects (broadly defined) of the earlier macroeconomic instability.3 During the 1990s, large socially-owned enterprises faced soft budget constraints and financed their activities from banks, and then increasingly from arrears to the state, suppliers, and workers. The accumulated operating losses and arrears of the large enterprises (some with cross-ownership in state and socially-owned banks) were sizeable. That distress in turn is well documented in the in-depth diagnostic studies of eight major state banks conducted in the spring of 2003, and in the accounts of the four large liquidated banks. 3 The widespread partial indexation of pensions and wages has led automatically to claims for additional public expenditures, many of which were also satisfied from the banking system

12

As regards the balance-sheets of the banks, it is putting it mildly to say that the two hyper-inflations in five years and repeated currency devaluations effectively hollowed out bank balance–sheets. The main manifestations were: (i) the high rate of non-performing foreign currency loans; (ii) the loss of real value in the banks’ Dinar-based assets; (iii) the liquidation of their Dinar liabilities; and (iv) the freezing of a large block of foreign currency resources/ deposits.4

2. Early Policy Responses

Macroeconomic Stabilization

Stabilization efforts to fix these major macroeconomic problems began soon after a new government came to power after October 2000. Support came from the first IMF operation as approved in principle in December of that year. The Stand-by of June 2001 was followed by a three year Extended Agreement approved in May 2002. These agreements in turn opened the door for negotiations on FRY’s external indebtedness with a first agreement with the Paris Club being achieved in November 2001. The significant restructuring of external debts that followed had major implications for the balance-sheets of domestic banks from mid-2002 (see Box 2.1 above).

Although these programs and the parallel actions on structural reforms supported by the IFIs have helped to achieve the recovery of GDP – growth was recorded from 1999 to 2003 in the range of 4-6 percent per annum – the major macroeconomic imbalances of the economy have so far been moderated rather than eliminated. External debt is still extremely high at around 75 percent of GDP and hence Serbia’s international creditworthiness remains in question. At the same time, the country still requires capital inflows at the level of US$7.9 billion per annum to meet a combination of: (i) large current account deficits (circa US$4.7 billion per annum); (ii) the need to build reserves (circa US$1.1 billion); and (iii) the need to fulfill external debt service obligations (circa US$ 2.1 billion).5

Fiscal expenditures have been brought under much tighter discipline – and are now on a more normal basis of peace-time conditions - but fiscal deficits still run in the range of 4-5 percent of GDP6. Much of the financing for the budget is now coming from foreign sources including the anticipated revenues from enterprise and bank privatization. Inflation has declined to less than 8 percent by 2003 and this in turn has made significant reductions in

4As one WB expert has put it, “the state owned banks became empty shells serving only the war machine of the state, offering accounting to large enterprises and public utilities and partially occupied by criminal elements. Moreover, this lasted for more than a decade, which makes a huge difference, because trust in banks and the culture of prudence was completely destroyed.” Dr. Miroljub Labus has put some numbers to this. In 1996, recorded bank assets were the equivalent of 150 percent of Yugoslavia’s Total Material Product (or 200 percent if off-balance-sheet items are included). But half of all these assets were represented by estimated bad loans – i.e. 100 percent of Total Material Product had been lost. Frozen bank liabilities were the equivalent of 80 percent of the Total Material Product and 40% of this (no less than US$4 billion) were the frozen deposit claims of households. 5 GOS Letter of Sectoral Development Policy for PFSAC II, April 2003. Note that Serbia cannot access international capital markets due to the unresolved status of London Club debts. 6 New post-2000 pressures on the fiscal deficit come from among other things, the resumption of external debt servicing, the scheduled repayment of households frozen foreign currency deposits, the direct costs of resolving problems in troubled banks and enterprises either through restructuring or liquidation.

13

nominal rates of interest possible. However, the quasi-fiscal debt and deficit associated with arrears, contested claims and the obligations of social enterprises place a large cloud over the public finances.

Early progress with financial sector reforms (2001-2003)

After October 2000, the new Government of Serbia (GOS), supported by the Bank (through PFSAC program) and other donors, was quick to recognize and to try to deal with the financial distress in both the enterprise and banking sectors. The assessment of the banking system that was completed by May 2001 made it crystal clear that the rehabilitation of the large state banks – as it had been conducted in Slovenia, Croatia and Macedonia – was not going to be fiscally viable in Serbia. More importantly, the largest banks simply had no further economic purpose, little human capital, outdated technology, and no public trust.

So in parallel to decisions about the enterprise sector, the priority was to transfer the assets of the banks to new owners who could use them productively. Given the inability of the budget to re-capitalize the more seriously distressed banks, and the likely negative franchise value of most such banks, some 25 insolvent banks representing nearly two-thirds of the assets of the banking system have been closed over the past three years. This process culminated in early 2002, when the authorities made a bold decision to apply the same definitive solution to the country’s four largest banks, which at the time represented 57 percent of total recorded banking assets. The Bank Rehabilitation Agency (BRA) was given extended authority to administer banks in bankruptcy. The data in Table 2.2 summarizes the situation as it had emerged after the first round of clean-up by early 2002.7

The result of these actions was that in less than 3 years the total number of banks in Serbia decreased from 83 in early 2001 to 47 in end-2003. In the same period, the former Post-Office Savings Bank (Poštanska Stedionica) has been restructured as a universal bank, and the National Savings Bank had been founded with minority state participation. That period of rapid structural change has also seen eight new banking licenses issued to foreign banks, reflecting growing foreign investor awareness of the potential for business opportunities within the Serbian banking system.

7 It is worth noting that the timely resolution of problem banks meant that associated costs to the budget (including payouts to depositors and severance pay of bank employees ) have so far been much lower in Serbia (some US$83 million for the four largest banks, or around 0.7% of GDP) than in the neighboring economies where the authorities attempted costly rehabilitation programs with mixed results (20.6% of GDP in the Czech Republic, 26.5% in Bulgaria, 12.9% in Hungary). Source: Banking crises in transition economies: fiscal costs and related issues, by Helena Tang, et al. (World Bank, 2000).

14

Table 2.2 – The Initial Restructuring of the Banking System

(US$ million) All banks

Sept. 2001 Banks

in liquidation

Remaining Banks

Jan. 2002 Assets Liquid Assets 1,342 531 39% 811 Loans 5,354 3,935 74% 1,418 Frozen Foreign Exchange Bonds 2,190 1,149 52% 1,041 Investments 356 200 56% 156 Fixed & Other Assets 794 316 40% 478 Total Assets 10,036 6,132 61% 3,904

Off Balance Sheet Assets 6,297 5,147 82% 1,150

Liabilities & Capital Deposits 1,658 692 42% 966 Borrowings 4,889 4,203 86% 686 Borrowings from NBY 23 13 56% 10 Frozen Foreign Exchange Deposits 2,297 1,195 53% 1,102 Provisions 3,957 3,741 95% 217 Other Liabilities 366 153 42% 213 Total Liabilities 13,191 9,997 76% 3,194 Capital -3,155 3,864 710 Total Liabilities & Equity 10,036 6,132 61% 3,904

Off Balance Sheet Liabilities 6,297 5,147 82% 1,150

Source: World Bank, A Strategy for Restructuring the Enterprise and Financial sectors in the Republic of Serbia, June 2002 In July 2002, the Government initiated a debt-for-equity swap to cover the resolution of debt owed by Serbian banks to the Paris and London Club governments and banks. This conversion process remains incomplete, however, due to delays in reaching final settlements with the London Club. Nonetheless, one result of this operation was a de facto nationalization of a large part of the banking system, with the state currently holding controlling equity stakes in nine banks (including, at that time, the largest bank in the system, Vojvodjanska Banka) and significant minority stakes in seven others (including Komercijalna Banka, at that time the largest “private” bank in the country).

Although the stated objective of this move was to speed up reform of the sector through offering the state stakes for rapid sale, the privatization process has suffered significant delays during the recent period of political uncertainty. The BRA, supported by related technical assistance provided by the World Bank and other donors, is responsible for running the sale process, as well as ensuring proper controls and governance in nationalized banks to preserve their value prior to resolution. As of September 2004, the authorities have offered three medium-sized banks for sale through open tender, and preparations are underway for the sale of three more banks in early 2005. The timing and method of divestiture of state holdings in other banks are yet to be determined.

In parallel to the bank resolution process, authorities took initial steps towards the modernization of the regulatory and institutional framework for financial sector operations. With the help of the Bank and other donors, the NBS has prepared a Supervisory Development Plan (SDP) for 2003-2005, which includes among other things the need for enhancing the quality of banks’ reporting and compiling timely analytical information on

15

bank liquidity and solvency. This will allow NBS to launch proper off-site monitoring on the basis of banks’ monthly electronic reporting. Bank supervision regulations were substantially revised in 2002 and the first half of 2003 to make them more compliant with Basel core principles, first of all in such areas as capital calculation and the capital adequacy ratio (CAR), loan classification and provisioning, although the full effect of these changes will not be felt until the implementation of SDP is well underway. Starting with January 2004, the NBS increased the minimum capital requirement to Euro 10 million. This is expected to further reduce the number of banks as some small banks merge or close in response to the increase. In early 2003, the payment system was transferred from a centralized operator to commercial banks without major disruption. Finally, the new accounting law has been enacted, requiring all banks to adopt International Accounting Standards (IAS) from 2004.

Figure 2.1 – Monetary Aggregates 1998-2003

- 30,000 60,000 90,000 120,000

1998

1999

2000

2001

2002

2003

Dinar millions

Foreign Currency DepositsTime Deposits (dinar)Sight DepositsCurrency

Source: NBS data

The volume of monetary aggregates and levels of financial depth have shown some early positive reaction to the reform minded actions of the authorities and macroeconomic stabilization that has been achieved. Figure 2.1 indicates a slow restoration of confidence in the banking system albeit with a strong foreign currency component confirming the ongoing hedging of risk. Most remarkably, public holdings of foreign currency deposits rose from very low levels in 2000, to Euro 802 million by end-2001, to Euro 1.3 billion at end-2002, and to nearly Euro 1.8 billion at end-December 2003. As reserve requirements were progressively reduced and the economy started to revive, banks resumed lending to the real sector. Total credit to the non-government sector expanded from Euro 1.6 billion at end-2001, to Euro 2.4 billion at end-2002, and to almost Euro 3 billion at end-December 2003. In part, this growth can be attributed to the rapid growth in credit to households, whose share in the total credit to non-government sector has grown from 5 percent to 14

16

percent over this period, although this figure may include some credit to individual entrepreneurs and micro-enterprises.8

3. Present Structure of the Banking System

The structure of banking that has emerged from these initial actions by the authorities comprises three distinct segments. Specifically, the structure and relative importance of Serbia’s remaining 46 banks9 (as at end-March 2004) can be summarized as in Table 2.3 below.

Table 2.3 – Structure of Serbian Banking System

State-owned Banks

Foreign-owned Banks

Local Purely Private

TOTAL

Number of banks, 03/04 17 9 20 46 (of which small banks) (7) (5) (14) (26)

Total Assets, 06/03 (Dinar bln) 152.2 51.4 94.2 297.8 Share in Total 51% 17% 32%

Total Assets 12/03 (Dinar bln) 164.7 71.8 118.5 355.0 Share in Total 46% 20% 34%

Total Assets, 03/04 (Dinar bln) 162.6 85.5 124.6 372.6 Share in Total 44% 23% 33%

Total Regulatory Capital, 12/03 (Dinar bln) 39.2 10.4 34.3 84.1 Share in Total 47% 13% 40%

Total Funds-mobilized, 12/03 (deposits and borrowing - Dinar bln)

119.6 59.5 78.5 257.6

Share in Total 46% 23% 31%

Source: NBS data as indicated except Societe Generale and Raj Bank for which only end-June 2003 figures are available for capital and funds mobilized. 1 Defined as having less than Dinar 3 billion in funds mobilized Table 2.4 below helps to put Serbia’s banking sector in the regional context. It is clear that in spite of rapid expansion of a number of foreign banks in the deposit market , the Serbian banking system is still predominantly domestically-owned and controlled. At the end of 2003, the domestic banks make up almost 80 percent of total assets and 87 percent of the reported balance-sheet capital.10 By regional standards, Serbia is also characterized by a relatively high level of assets held by majority state-owned banks (33.6 percent in end-September 2003). Given the size of the economy, the banking sector is overpopulated and fragmented, and there is a room for further bank consolidation. Some of banks are very

8 Unfortunately, the monetary data published by the NBS do not provide the breakdown of bank credit between state and socially-owned enterprises, and private companies. 10 The data used here are from the NBS reports. For the period up to June 2003 as used here, the accounts comply with the standards laid down by the National Bank of Yugoslavia (NBY) and later the National Bank of Serbia (NBS). This means that they have not been adjusted to be compatible with some of the main requirements of International Accounting Standards (IAS) and especially the requirements relating to IAS 16, 19 and 36 relating respectively to Fixed assets, reporting for Hyperinflation and the Impairment of Assets.

17

small with only regional coverage or they act as “pocket banks” mainly operating as treasury divisions of large enterprises.

Table 2.4. Banking Sector Concentration Indicators and Share of Foreign and State Banksa11

Assets Share

Number of Banks (Foreign) Five Largest Foreign State1

Croatia 46 (23) 63.4% 90.2% 4.0%

Hungary 38 (27) 84.9%2 10.8% Romania 31 (24) 67.9% 52.9% 43.6% Bulgaria 34 (26) 55.5% 75.2% 14.1% FYROM 20 (7) 73.6% 44.0`% 2.0%

Slovenia 22 (6) 68.9% 18.4%2 48.6%

Serbia – 2001 86 (5) 48.7% 56.2% Serbia – 2002 50 (9) 46.6% 14.7% 38.4% Serbia – September 2003 47 (10) 49.1% 17.4% 33.6%

Source: NBS, EBRD, IMF and official estimates, as reflected in publications from the national authorities; end-2002 data unless specified otherwise 1. Figures for Serbia include all banks with majority state holding 2. Data as of end-2001

State-owned banks

Despite the dramatic closure of the four largest banks in early 2002, the Serbian banking system remains dominated by the state with 44 percent of total assets (end-March 2004 data) and 47 percent of total capital (end-2003 data) being in banks with some state ownership stake. Of the seventeen state banks shown in Table 2.3, nine have a majority state-ownership, seven have a minority state stake, and one bank is in rehabilitation.

As will be shown in Chapter 3, the market share of majority state-owned banks has been shrinking over the past two years due to a variety of internal and external factors, although the same cannot be said with regards to certain minority state-owned banks, such as Komercijalna Banka. Two of the largest four banks by total assets are owned by the state, namely Vojvodjanska Banka (majority-owned) and Komercijalna Banka (minority-owned). In mid-2003, these two banks accounted for Dinar 85 bln , i.e., nearly 28 percent of system’s total assets, although their share has decreased since then (19.5 percent in June 2004) mainly due to the stagnation of Vojvodjanska. Two other majority state-owned banks, namely Jubanka and Novosadska Banka also figured amongst the country’s seven largest banks by assets and accounted for a further 8 percent of total assets. For the purposes of further analysis in Chapter 3, Komercijalna Banka, in which GOS only has a significantly diluted minority stake and, until recently, no effective control, is treated as a local private bank.

This structure involving high state-bank dependence is recognized to be wholly unsuitable for Serbia’s long-term financial development. The financial weaknesses in several of the banks, as identified in the mid-2003 diagnostic reports commissioned by the BRA, need to be addressed with urgency. But this reality confronts the limited fiscal capacity to inject significant state funds. This means that financial resolution must take either the form of privatization of relatively viable banks to strategic investors or, for weaker banks,

18

liquidation (see Chapter 4). At the same time, the state banks continue to operate with some of the same operational and governance weaknesses that were established during the 1990s, and with insufficient capital. The large state banks, for example, are used to building their expansion on the back of state-sourced deposits rather than retail deposits.

Foreign Banks

Foreign bank penetration of the sector has been rapid since the first licenses were issued in 2001-2002 and their asset share of 23 percent at the end-March 2004 is impressive given the very recent start-up and the absence of large entry points/capital injections through privatization of local banks. Raiffeisen Bank has total assets which have grown fast and now stand at Dinar 44.7 billion (June 2004). Three other foreign banks are of medium size, namely Societe Generale of France (total assets of Dinar 13.4 billion); Hypo Alpe-Adria Bank (total assets Dinar 14.9 billion), HVB Bank of Germany (total assets Dinar 11.4 billion); and Pro-Credit Bank – a micro-finance bank with German, Dutch and EBRD capital (total assets of 9.7 billion).

Because these banks have cleaner balance-sheets than many of their domestic bank competitors; none of their ambiguities about ownership; and access to new capital from their home bases, they seem likely to be a source of considerable dynamic in the medium term. However, their early successes and the possible delays in privatization raise other key questions and dilemmas for government that are also assessed in Chapter 4.

Local Private Banks

These banks, in common with the foreign banks, have grown relatively fast in the past three years. They too have been able to take some advantage of the demoralized situation of the state-owned banks. Although it is convenient to assume that the “private” status of these banks exempts them from the stresses and distortions suffered by the state banks, the reality is unlikely to be quite so straightforward. As many private banks have been closely affiliated with specific interest groups, they lack the advantages of clean balance-sheets and the more prudential culture brought by the foreign banks. The stresses of relatively fast growth are already evident in some – e.g., in the form of strains on their capital adequacy – and their rapid growth of new lending will likely show up in problems in the years to come. It is also no secret that the uncompetitive placement of state and NBS deposits for use in directed lending schemes have partially contributed to a rapid expansion of a number of large private banks. So the situation of these banks also needs careful scrutiny during the current difficult transition period for the sector. Their performance is critically assessed in Chapter 3.

4. Benchmarking Serbian Financial Sector

Despite the early progress discussed in the sections above, Serbia’s financial sector remains severely underdeveloped and the existing level of financial intermediation is unable to fully contribute to the economic growth agenda. With banks accounting for more than 90 percent of whole financial system, their total assets amounted to only $6.9 bln at the end of 2003, or 35.8 percent of GDP. Not only this is a very small figure for an entire banking system, being smaller than the total assets of many commercial banks in Western Europe,

19

but it also represents a small system when measured on a per capita basis. At end-2003, per capita banking system assets in Serbia were only $684, compared to $1,218 in Bosnia and Herzegovina, $1,286 in Bulgaria, $7,188 in Croatia, and $5,984 in Hungary. On another key indicator, bank credit to the enterprise sector as a percentage of GDP, Serbia also lagged behind with 18.7 percent at the end of 2003, compared to 42.3 percent in Bosnia and Herzegovina, 26.6 percent in Bulgaria, 57.3 percent in Croatia, and 42.7 percent in Hungary.

Figure 2.2 – Comparing Financial Depth & Reliance on Cash

500

1500

3000

Eurozone

Croatia

B&H

SERBIA

Romania

Bulg.Est.

LatviaLithuania

Russia

Ukraine

Armenia

Slovak/CzechRepublics

Slovenia

Kazakhstan

Georgia

Poland

Hungary

Azerbaijan

Belarus

-20%

0%

20%

40%

60%

80%

100%

120%

140%

160%

-10% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90%Total NBNGS deposits as a % GDP (mid-2002)

Cas

h re

lativ

e to

mon

etar

y de

posi

ts

Bubble size denotes purchasing power parity adjusted average total deposits per head in US

$

Kyrghiz

The relative weakness of the Serbian financial system is further illustrated by the explicit benchmarking of its banking sector against comparator countries in the region of Central Europe and the FSU that have faced similar problems of adjustment in the past few years. It can be seen from Figure 2.2 that Serbia’s banking sector is comparable in size to many countries of the FSU including Russia, but still has far to go to match the size of deposits relative to GDP now found in most of the accession countries of Central Europe. The cash to deposits ratio is also higher in Serbia than in many Central European comparator countries. That fact clearly signifies some ongoing lack of faith in the domestic banks. It is true that Serbian banks have shown the ability in the past two years to mobilize significant volumes of foreign currency deposits. This suggests that ongoing anxieties about macroeconomic/monetary stability are still a major factor in the public’s perception about the appropriate forms in which to hold savings and store wealth. Banks seem to be re-established as accepted institutions for payments and transactions purposes but still need the foreign exchange guarantee to achieve significant access to domestic savings. The detailed analysis of bank-by-bank data, described in more detail in Chapter 3 and captured in abbreviated form in Figure 2.3 below, demonstrates that the cost and other efficiency parameters of Serbian banks still have more in common with the non-Baltic countries of the FSU than with the accession countries of Central Europe. Figure 2.3 shows eleven countries ranked by the ratio of deposits to GDP in 2002 (the bottom segment of

20

each chart. For each of these countries two key sets of operating efficiency indicators are shown: (i) the ratio of operating costs to productive, income-earning assets; and (ii) the ratio of operating income to the same base. The former is a measure of cost efficiency, the latter - a measure of the pricing that banks pass on to their customers. Both measures are shown separately for the most efficient quartile of banks in the system (the black bubbles) and the rest of the banking system (the gray bubbles). The size of the individual bubbles indicates the market share of the two groups of banks in each country; and the horizontal bars at the bottom refer to each country’s level of monetization. As can be seen from the figure, the better Serbian banks, while clearly outperforming the rest of the banking system, account for a minority of total banking system assets. More significantly, the efficiency indicators of even these better Serbian banks are far worse than those of leading EU banks (indicated by the dotted lines). For example, the top quartile of Serbian banks have a cost/assets ratio of around 7 percent as against an average for leading EU banks of less than 2 percent. The weaker Serbian banks (the other three quartiles) have an average cost ratio of around 13 percent - no less than 6 times the EU bank standard.

Figure 2.3 – Comparing Financial Depth & Banking Sector Effectiveness

5%

0%

OPERATING COSTS AS % EARNING ASSETS

10%

15%

20%

25%

30%

0%

5%

10%

15%

20%

25%

30%

7%14%

18%20%20%25%

32%36%37%

41%

67%

13%

Cro

atia

Hun

gary

Pola

nd

Esto

nia

Bul

garia

Latv

ia

Lith

uani

a

Rus

sia

Serb

ia

Ukr

aine

Kaz

akhs

tan

Arm

enia

OPERATING INCOME AS % EARNING ASSETS

0%

5%

10%

15%

20%

25%

30%

0%

5%

10%

15%

20%

25%

30%

13%

67%

41%37% 36%

32%25%

20% 20% 18%14%

7%

Cro

atia

Hun

gary

Pola

nd

Esto

nia

Bul

garia

Latv

ia

Lith

uani

a

Rus

sia

Serb

ia

Ukr

aine

Kaz

akhs

tan

Arm

enia

Chart Key: Circles indicate aggregate performance of different group banks in each country:• center of circle shows

level of performance (read against side axes)

• size of circle shows market share of group of banks

• black circles show performance of best quarter of banks in each country

• grey circles show performance of rest of banking system

• dotted lines indicate average performance of lending EU banks

Source: NBS and BankScope Serbia data as of mid 2003; other countries as of end-2002

The closing of those efficiency gaps (i.e., between (a) the better and the weaker Serbian banks; and (b) between the better Serbian banks and the leading EU banks) will be the key to moving Serbia’s position further to the left on Figure 2.3 and to the right on the earlier Figure 2.2. This in turn will require that the weaker banks in the system lose ground rapidly to the better banks that currently include, but are not confined to the foreign banks. That shift in the structure of the sector will need either the liquidation of more of the weaker

21

banks, or their rapid re-invention (possibly through new strategic investment) as stronger banks. The alternative of an extended period with today’s status quo cannot possibly create the dynamic to establish a much large banking system capable of providing adequate intermediation to real sector on a sustainable and affordable basis.

22

Chapter 3: Assessing Bank Performance

The previous chapter has described the current structure of Serbia’s banking system. It has also begun to identify some of the critical questions that face the authorities in developing their longer-term strategy for the sector. This has been done in relation to three main segments of the banking sector: state-owned banks, foreign banks and local private banks. In the present chapter those suggestions are developed further by invoking a detailed numerical analysis of bank performance that differentiates the situations of different groups of banks. That analysis in turn is used to provide answers to some important questions about future strategy.

The analysis uses data for individual banks kindly provided by the NBS Supervision Department.11 They make it possible to build a set of key indicators of operational effectiveness on a bank-by-bank basis. The data identify the segment to which each banks belongs but with the confidentiality of individual bank data protected. The data used relate mainly to the first six months of 2003, but with changes in market share tracked over eighteen months (i.e. from the beginning of 2002, immediately after the closure the four largest failed banks).

It is recognized that the existing NBS data take only partial account of the requirements of IAS accounting. This will mean that balance sheet size may be overstated for some banks and particularly those banks controlled by the state where balance-sheets are most impaired by the traditional close links with social-enterprise lending. For this reasons, the analysis proceeds in two stages. First it excludes these state banks from the setting of the benchmarks for bank-by-bank comparisons. Next, having defined benchmarks of “good performance” it overlays measures of the performance of the state-banks on top of the spectrum of behavior for foreign and local private banks.

The transition to IAS is currently underway but will not be fully completed until accounting years subsequent to 2004. This means that any numerical analysis conducted today will be compromised by possible weaknesses of the accounts. What is important (in this FSN and in any future analysis to be used by the NBS, such as FSAP stress-testing exercise) is to establish a sound framework for assessing individual banks performance. That framework can be gradually improved by the NBS as the data available improve. The NBS Supervision Department has been fully involved in the preparation of the data presented in this chapter in order to ensure that this continuity of analysis can become a reality.

1. Description of Recent Trends

As already noted in Chapter 2, the Serbian banking system has been expanding rapidly since the closure of the four largest failed banks at the end of 2001. Examination of three major balance sheet totals – Assets net of provisions, Funds Mobilized and Capital – shows that the expansion over the eighteen months between then and mid-2003 has been of the order of 50%.

23

11 In addition to the Reclassified Balance Sheet and Income Statement data, which banks must make publicly available, extracts were provided from the Form-KA Asset Classification data covering asset quality and from the Form-BON Bank Operating Indicators data covering capital adequacy, large loan exposures and loan rollover, liquidity, foreign exchange risk and risk weighted assets.

Obviously this overall growth rate disguises marked differences between different banks. Significantly, a number of state banks have either shrunk or barely maintained their deposit base while some local private and foreign banks have grown very fast. The distribution by bank of changes in funds mobilized over those eighteen months is illustrated in Figure 3.1 below. This figures is sub-divided to show: (i) the size of the decline in seven surviving banks that have shrunk in the period (five are state-owned and two are foreign) together with the decline in the banks that have been closed; and (ii) the size of the increase achieved in all other surviving banks. Figure 3.2 compares the distribution by banks of the resulting total stocks of outstanding bank funding, assets and capital as at mid-June 2003. The labels “S” (state – in black color), “F”(foreign – in grey) and “L”(large private – in white) on these two figure help to identify the type of bank. For this analysis Komercijalna Banka is treated as a large private bank since this seems to better reflect its de facto status in the system.

Figure 3.1 – Banks losing or gaining funding between January 2002 and June 2003

INCREASES IN FUNDS MOBILISED - GROWING BANKS

[L] Komercijalna banka

[L] Delta banka

[F] Raiffeisen bank Beograd[L] Poštanska Štedionica

[F] HVB banka

[L] Exim banka

[S] Continental banka

[L] Zepter banka

[F] Societe Generale

[S] Nacionalna Štedionica

REDUCTIONS IN FUNDS MOBILISED - SHRINKING BANKS

[F] EFG Eurobank [F] Volksbank Beograd

[S] Vojvođanska banka

[S] Panonska banka

[S] Jubanka

[S] Borska banka

Closed banks

Attention is drawn in particular to the huge gains in funds raised by the largest minority state-owned bank (Komercijalna Banka), by the largest foreign bank (Raiffeisen) and by the largest local and truly private bank (Delta Banka).

24

Figure 3.2 – Distribution by Bank of Key Balance Sheet Totals

FUNDS MOBILISED (30.06.03)



[L] Delta banka

[F] Raiffeisen bank Beograd

[F] Societe Generale Yugoslav Bank

[S] Novosadska banka[L] Poštanska Štedionica

[S] Jubanka[S] Srpska banka

[S] Agro banka

[L] Exim banka

REGULATORY CAPITAL (30.06.03)

[L] AIK banka Niš


[S] Jubanka


[L] Kulska banka

[L] Delta banka

[F] Raiffeisen bank Beograd[S] Novosadska banka[S] Srpska banka

[S] Agro banka

ASSETS NET OF PROVISIONS (30.06.03)


[L] Komercijaln

[L] Delta bank

[F] Raiffeisen bank Beog

[S] Jubanka[S] Novosadska banka[L] Poštanska Štedionica

[L] AIK banka Niš[S] Srpska banka

[F] Societe Generale Yugoslav Bank

What immediately becomes clear is that state banks are losing share fast and that growth is highly concentrated, with three banks (two local and one foreign) accounting for half of all new funds mobilized and ten banks (about a fifth of the total) accounting for over three quarters of new funds mobilized.

Market share in terms of funds mobilized can normally be expected to translate into market share in terms of total assets (measured in this analysis to include off-balance items but net of provisions and frozen foreign currency savings). This relationship is indeed seen in the Serbian data and in the relevant pie charts as shown above. The mapping of funds mobilized to capital is far less easy to assess in general terms. Banks might often expand their funding but fail to match this with appropriate additional capital. The pie charts suggest that this too appears to have been the case in Serbia since 2002. Note for example, the differences between the shares of assets of the fastest growing banks – Komercijalna, Delta and Raiffeisen and their corresponding shares of Regulatory Capital.

Figure 3.3 takes the data on funds mobilized and presents this as changes in market share, with the state banks separated out. The reason for concentrating on funds mobilized – i.e.

25

deposits plus borrowed funds – is that, of all the key balance sheet totals, it is the one least prone to distortions coming from attempts to restructure state bank balance sheets (which will shrink as necessary “catch-up” provisioning takes place and also as the Paris & London Club swaps are completed).

Figure 3.3 – Changes in market share (January 2003 – June 2003)

-8%

-6%

-4%

-2%

0%

2%

4%

6%

-8%

-6%

-4%

-2%

0%

2%

4%

6%

State-controlled banksForeign-controlled banksLocal-private banks

Source: NBS data

Figure 3.3 confirms that significant swings in market share have taken place over a very short period of time. These swings have also occurred in a period when the design and implementation of sound regulation and supervision has remained seriously incomplete. Hence questions about the quality of management and financial performance, especially in the rapidly expanding banks, should be high on the agenda for policy-makers. It is vital for the sustainable deepening of the sector that the Serbian banks gaining significant market share should be relatively efficient at turning most of the extra funds they raise into good quality income-earning assets. It is vital that: (i) they do not overstretch their capital base, and (ii) the asset growth that has accompanied increased funding should be of sound quality. The authorities can easily assure neither of these conditions while the regulatory system remains incompletely enforced. Our own initial assessment is presented in the next section.

2. Comparing bank performance

Eighteen separate indicators, all calculated from the NBS data and grouped into four main categories have been calculated for all Serbian banks. These are:

A. A set of five indicators of operational effectiveness that have been shown in other research to have significant links to the process of financial deepening. This is the process whereby formally repressed financial systems grow to establish a more appropriate relationship with the economy they support. The individual indicators calculated are:

26

• efficiency of Asset Deployment as measured by the ratio of income-earning assets to interest-bearing liabilities;

• Weight of Operating Overheads, a measure of cost efficiency captured by the ratio of operating costs (before bad debts) to income-earning assets;

• Overall Spread and Fee Yields, a measure of income efficiency captured by the ratio of operating income to income-earning assets;

• Operating Profitability as measured by the ratio of gross profit before bad-debt charges and taxes to income-earning assets; and

• Loss Wedge, which captures the degree to which gross operating profitability is dissipated in provisioning for bad loans.

B. A further set of five classical regulatory indicators provided by National Bank of

Serbia from its quarterly BON returns, namely:

• Capital Adequacy as measured by total regulatory capital divided by total risk-weighted assets;

• Continuous Placements and Large-loan Exposures, both measures of risk being taken with lending, irrespective of how much actually has gone bad; and

• Liquidity Ratios and Forex Risk, both measures of balance sheet management risk.

C. A set of five classical ratings indicators that can be derived from any published

accounts and therefore translate most easily to international experience, namely:

• Return on Assets and Return on Equity as measured by net profit relative to total assets (net of provisions) and shareholders funds;

• Two measures of the cost/income ratio, one calculated before bad debts and one after; and,

• Provisioning Rate as measured by the ratio of bad debt charges to total assets.

D. Finally, a set of three loan-quality indicators to try and capture the relative burden

of historic bad debts on state-owned compared to other banks, namely:

• Weight of Non-performing Assets as measured by the ratio of non-performing on-/off-balance assets to potentially income-earning assets;

• Provisioning Cover as measured by the ratio of long-term provisions made to the sum of performing and non-performing income-earning assets; and,

• Capital at Risk as measured by the ratio of un-provided non-performing assets to total regulatory capital.

27

Once the individual indicators are calculated, a composite scoring of each bank could be derived for each of the four sets of indicators. Banks in the table are ordered in the same way as in Figure 3.3, with state-owned banks separated out from foreign and local-private banks and banks then sorted in terms of loss or gain of share in the market for funds mobilized. This allows the reader to make a quick visual judgment on whether growth in the market is distributed in an efficient and sustainable fashion.

To arrive at the composite scores, banks were ranked under each indicator within each set (best first to worst last) separately.12 These were then used to compute an average ranking calculated for each set of indicators. Banks were then graded as to whether their composite ranking for each set of indicators was in the “Best” quartile, next best (denoted “Average +”), third quartile (denoted “Average –”) or the “Worst” quartile for that set of indicators. Possible distortions to these thresholds (i.e., what constitutes Best, Average +/– or Worst), associated with the greater data weaknesses specific to state-owned banks (see Box 3.1 for more detail), have been avoided by calculating the threshold values only with regard to foreign and local-private banks. The separate calculations for the state-owned banks therefore, in effect, show how their performance under each indicator would place them on the spectrum of private bank performance. This allows the reader to factor in particular judgments about the quality of the state bank accounting data.

It should be noted that it is perfectly possible for a bank compromised by historically bad lending to score well on operational effectiveness but poorly on regulatory and loan quality indicators. All this is saying is that the bank now lives relatively efficiently within its reduced performing balance sheet (i.e., that its real cost base has been reduced in line with its reduced real income-earning assets, what it extracts from those income-earning assets is still reasonable by market standards and that provisioning on the remaining income-earning assets is manageable), but that its capital adequacy, liquidity and large loan exposures and loan rollovers are all compromised by still carrying non-performing historic lending on its books. It is also possible for a bank to score poorly as regards operational effectiveness but well as regards traditional rating indicators. This is because the former penalize banks that make a large profit margin from charging very high prices on top of a high cost base, whereas the latter is indifferent as regards the source of profit.

Four main conclusions emerge:

• Somewhat unexpectedly, some of the biggest state-owned banks score highly on the “A indicators” - operational effectiveness even with the problems already embedded within their balance sheets. The significance of this is discussed in more detail below. It is not the same as saying that these are healthy banks. It does, however, indicate that, cleaned up and with the right owner, some of these banks have a potentially important role to play in improving competition in the growing segments of the market.

• All state-owned banks are at the bottom of the spectrum of loan quality (indicators –the “D” indicators - and for this reason most state-owned banks also rank poorly on their regulatory indicators – the “B” indicators. This is not surprising, given what is known about directed lending within state-bank portfolios. It is, moreover, virtually impossible to maintain good regulatory indicators when running a compromised loan portfolio. The implications of this for restructuring the state-owned banks are discussed in more detail below.

• Almost half of the state-owned banks score unambiguously badly on all four sets of indicators. The policy implications of this are blunt and obvious – the banks in question almost certainly cannot be saved. Even if their solvency and liquidity could be restored, they are very unlikely to be able to operate at cost and price

28

12 The ranking of individual banks is omitted from this version of the report based on the confidentiality agreement with the NBS.

ratios that would allow them to compete effectively in a consolidating market. There must therefore be a risk that if they were to be restructured (either by the state or, less likely, by new strategic investors), they would only recreate the problems that brought them under state control in the first place.

• The performance indicators of the banks gaining market share do not inspire confidence. This is particularly true of two of the three local-private banks that have gained most share. The implications of this for the quality of competition and sustainability of expansion are discussed in more detail below.

Box 3.1 – Two caveats regarding the data A critical issue in calculating the performance ratings relates to the underlying quality of disclosure, particularly as regards loan classification and liquidity at the state-owned banks and some local private banks. Two big distortions are known to be a problem in Serbia – complacency as regards the underlying quality of non-performing loans to enterprise sector (where even long overdue loans are still classified as A-grade Standard) and liquidity (where public sector deposits are directed towards banks (including, in recent years, majority privately owned banks) that have provided loans to non-performing debtors in the real sector). It is hard to ascertain the scope of the latter problem due to the very sensitive nature of the data involved. In any case such covert liquidity support can do no more than leave the banks involved with apparently adequate liquidity. It cannot boost their liquidity to the high levels typical of foreign banks in Serbia and therefore does not alter the relative rankings from which the composite scoring described below is created. The issue of misstated loan quality is more tractable, as detailed diagnostic reports were completed on eight state-linked banks in early 2003 and covering the start of the data period used for this analysis. Unfortunately, the audit companies involved in these diagnostics did not follow common rules for loan reclassification although they did all downgrade much apparently A and B graded lending to state-supported entities. If these adjustments are applied to the NBS Form-KA data used to calculate income-earning assets in the set of operational effectiveness indicators above (Set-A) and to calculate non-performing assets in the set of loan quality indicators (Set-D), two changes appear to the results:

• Two of the four state-owned banks that score relatively well on the operational effectiveness indicators slip a little in the relative rankings but not so much as to change the conclusion that they have a potentially important role to play in bring more competition to the market after their balance sheets are cleaned up. The other two banks slip markedly and all results for them must be treated with caution.

• The state banks that already score relatively badly on all indicators only get worse under this treatment. Because no changes to most of the conclusions emerging from this analysis would result from factoring in the results of the diagnostics, this has not been done in the present report. This has the benefit of making the analysis a more transparent transformation of existing regulatory indicators, thus aiding its handover to the NBS Supervision Department. It also avoids the potential distortions that come from factoring in a partial analysis that was not in any case done to common standards.

It should be stressed at this point that this analysis is no substitute for a full rigorous supervisory regime of the sort being developed in Serbia. Rather, it is a quick spectrum analysis to answer fundamental questions for any study of the economic health of a country’s banking system. Two of the most important of such questions are:

• Is growth in the market sustainable from an institutional perspective – i.e. are the growing banks turning funds mobilized into good quality income-earning assets at affordable prices and at on-lending costs bearable by sound borrowers? Are they doing this in a way that creates sufficient profit to provide the expanding capital needed to support continued growth?

• What role, if any, do the banks currently under state control have to play in the ongoing development of the banking system – i.e. are they needed to maintain effective competition and how must the banks that are needed for the future be restructured so they can take an effective place within a competitive market?

29

The next section addresses the first of these questions. The second question is picked up in Chapter 4 as one of the key policy issues facing the authorities.

3. Competition and Sustainability of Growth

The effectiveness indicators described in Section 2 have been chosen because they appear to have conditioned the re-deepening process for banking systems that have already undergone a significant transition typical of Central and Eastern Europe. In other words, they help to identify – via the best performance measures – those banks with the greatest capacity to create the profit needed to support rapid and ongoing expansion without compromising asset quality and capital adequacy. This can be illustrated graphically in two ways – first by looking at whether the bigger banks in the system (those that presently enjoy the largest market shares) perform better than smaller banks (Figure 3.4a) and secondly at whether the banks gaining market share are better than those losing share (Figure 3.4b). Both of these two sets of figures use four of the principle indicators of operational efficiency – the “A” indicators”. As before they also show separately the performance of the state banks.

Figure 3.4a – Bank-by-bank performance – banks ranked by market share as of mid-2003

30

FFICIENCY OF ASSET DEPLOYMENT

-50%

0%

50%

100%

150%

200%

IEA/IBL ratio (line)

e-earning assets as a % of interest-bearing liabilitiesRETURNS ON ASSETS

0%

5%

10%

15%

20%

25%

Mar

ket s

hare

(col

umns

)

-50.0%

-40.0%

-30.0%

-20.0%

-10.0%

0.0%

10.0%Rates of return (lines)

ROA before (dashed line) and after (solid line) bad debts and

COMBINED SPREAD & FEE YIELD

-20%

-10%

0%

10%

20%

30%

Spread & Fee yield (line)

perating income as a % of income-earning assets WEIGHT OF OPERATING OVERHEAD

0%

5%

10%

15%

20%

25%

Mar

ket s

hare

(col

umns

)

-20%

-10%

0%

10%

20%

30%

Cost ratio (line)

operating costs as a % of income-earning assets

E

0%

5%

10%

15%

20%

25%M

arke

t sha

re (c

olum

ns)

incom

0%

5%

10%

15%

20%

25%

Mar

ket s

hare

(col

umns

)

o

State Private State Private


31

Figure 3.4b – Bank-by-bank performance – ranked by change in share since start-02

EFFICIENCY OF ASSET DEPLOYMENT

-5.0%

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Mar

ket s

hare

(col

umns

)

-300%

-200%

-100%

0%

100%

200%

300%

IEA:IBL ratio (line)

income-earning assets as a % of interest-bearing liabilitiesOVERALL RETURNS ON ASSETS

-5.0%

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Mar

ket s

hare

(col

umns

)

-50.0%

-40.0%

-30.0%

-20.0%

-10.0%

0.0%

10.0%

Rates of return (lines)ROA before (dashed line) and after (solid line) bad debts

COMBINED SPREAD & FEE YIELD

-5.0%

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Mar

ket s

hare

(col

umns

)

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

Spread & fee yield (line)

operating income as a % of income-earning assets WEIGHT OF OPERATING OVERHEAD

-5.0%

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Mar

ket s

hare

(col

umns

)

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

Cost ratio (line)

operating costs as a % of income-earning assets



The first striking feature of the charts is the volatility of performance13. Nevertheless, even with the data currently available, some important conclusions can be drawn:

• The distribution of gains and losses of market share between state-owned banks appear incompatible with effective competition. In Figure 3.4b – left hand segment, those state-owned banks gaining market share when compared to those losing market share: (i) are no better at turning funds mobilized into income-earning assets; (ii) are no more profitable at a gross operating level and less profitable once bad-debts are taken into account; (iii) extract more by way of spread and fee income from their customers; and (iv) dissipate more of this on high operating costs. The absolute levels of the operating inefficiencies are also of concern. Some of the state banks gaining market share since 2002 have cost/income earning asset ratios in the range of 20-30 percent (good international practice would produce costs of around 2-3 percent) and spreads also some 10 times what one would expect in a well functioning Western bank. There is very little sign of effective competition between the state banks.

• There are no apparent economies of scale in the foreign and local private elements of the sector. This is shown in Figure 3.4a in relation, in particular to the high variability of the cost indicator and also the spread indicator across foreign and local

32

13 It is hoped that the later analysis based on a full year’s data will be less volatile, in that timing differences as to when banks accrue income/expense as well as make provisions should be eliminated.

private banks of different sizes. In other words merely achieving larger size (market shares) in these two banking segments is not of itself guaranteed to make the sector more efficient. It is also worth noting that the corresponding patterns in more strongly competitive transition banking systems (such as Poland and the Baltic states) show much more convergence of costs and spreads across different banks and at much lower levels (of costs and spreads) than is currently seen in Serbia. Notwithstanding the increase of competition that may have followed the entry of new foreign banks, the system currently exhibits few signs of true or deep competition.

• Interestingly there is much more of the expected relationship between size and efficiency measures found among the state banks – left segment of Figure 3.4a. There is some tendency for the larger state banks to be relatively more efficient than most of the smaller ones. However, the point made earlier about the poor absolute levels of the efficiency indicators of state banks remains.

• There is some indication (albeit with one clear exception) that the foreign and local-private banks most rapidly gaining market share do so at slightly lower overall spread and fee yields and with a slightly lower weight of operating overheads than those just holding market share or losing it. See Figure 3.4b. The difference, however, is not marked and nor is their performance much superior to state-owned banks conceding share to them.

• However, within the sub-group of private banks gaining market share, there is some limited evidence of a positive correlation (not the expected negative correlation) between rate of market share gain and combined spread and fee yield. This could possibly indicate a supply-constrained environment where the banks capturing most of the growth in the market are able to charge a premium for availability of much sought after services (particularly credit).

The last conclusion is potentially very important – and so needs to be confirmed. It suggests that the banks gaining share fastest need more competition. Increased competition is needed to keep their pricing under control and to force them to make more of their growing profit from spreading their costs more efficiently rather than just charging customers as much as they can bear. The probable segmentation of markets (e.g., rather different customer bases for the foreign versus the state and some of the local private banks) at the present time and the associated limitations on competition seem to be serving the needs of more efficient banking rather poorly.

The charts in Figure 3.5 illustrate the need that Serbian banks currently face to extract high margins. The two indicators shown in Figure 3.5 cover Capital Adequacy and Self-Capitalization. The former is a well-known indicator of stability. The degree to which the rapid expansion of any bank is putting capital adequacy under pressure must call in to question the sustainability of current growth. The latter measure is a relatively less well-known one. It seeks to identify the degree to which a bank is creating out of retained earnings, the extra capital it needs to support its ongoing expansion. If the calculated self capitalization ratio is positive, it shows how much the profitability of a bank’s growing business is helping to maintain capital adequacy above minimum required levels14. If the

33

14 The exact calculation is the ratio of: i) net profit minus ii) the change in required capital to maintain the crude capital adequacy ratio of 10 percent, all divided by iii) total assets at the end of the period concerned. This gives an indication of how rapidly a bank’s growth strategy would create or consume percentage points of capital adequacy. It is preferable to looking at the change in actual capital adequacy, in that it abstracts

ratio is negative it shows how much the growth of the bank’s business is being achieved by “eating” into existing capital adequacy or forcing the bank to rely on secondary forms of capital such as revaluation surpluses and subordinated loans.

Figure 3.5 – Bank-by-bank capital adequacy and sustainability indicators

CAPITAL ADEQUACY

0%

10%

20%

30%

40%

50%

Mar

ket s

hare

(col

umns

)

-20%

0%

20%

40%

60%

80%

Capital ratio (line)

regulatory capital as a % risk -weighted assetsCAPITAL ADEQUACY

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Chan

ge in

sha

re (c

olum

ns)

-20%

0%

20%

40%

60%

80%

Capital ratio (line)

regulatory capital as a % risk -weighted assets

SELF-CAPITALIZATION CAPACITY

0%

10%

20%

30%

40%

50%

Mar

ket s

hare

(col

umns

)

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

Self-capitalization (line)above broken line - capital creating / below broken line - consuming

SELF-CAPITALIZATION CAPACITY

-2.5%

0.0%

2.5%

5.0%

7.5%

10.0%

Chan

ge in

sha

re (c

olum

ns)

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

Self-capitalization (line)

above broken line - capital creating / below broken line - consuming



Again volatility is a problem in the data currently available. Nevertheless, important conclusions can be drawn. These include:

• The foreign and local-private banks that have the largest market shares (left hand upper plot in Figure 3.5) or have grown fastest since 2002 (right hand upper plot of Figure 3.5) include banks with the lowest capital adequacy ratios of that segment of the market. Nor is the capital adequacy of this segment dramatically superior to that of the state-owned banks ceding share to them – the left-hand segment of the same plots.

• Nevertheless, the banks gaining share at the absolutely fastest rates are broadly self-capitalizing probably as a result of them being able to charge a premium for availability of services. See the right hand lower plot in Figure 3.5 which shows a near zero measure for the self-capitalization ratios of most of the fastest growing private and foreign banks

34

from dividend policy, recapitalization, and below-the-line one-off bank reserves adjustments (and as such is a closer proxy to free equity cash flow used in shareholder value calculations).

This suggests that the situation is by no means critical, as the banks gaining most market share still have adequate capital. While they can continue to charge a premium for their credit and other services, the adequacy of their capital will only slowly descend to minimum required levels.

The sustainability of the present “competitive” situation is put into greater question by another factor. Although the more successful banks seem to be able to extract a premium for the availability of their services in the early stages of rapid market expansion, there is no guarantee that customers can afford to sustain such a premium as the market expands further. Adverse risk selection and moral hazard problems may be the eventual consequence. The major risk is also presented by the foreign exchange clause on practically all dinar-denominated loans, which could lead to a rapid simultaneous deterioration in banks’ asset quality in case of the marked exchange rate depreciation. The fastest growing local private banks in particular may indeed be storing up asset-quality problems in their scramble to maintain a lead over aggressive foreign competition. This is a point that needs to be kept under close supervisory review.

4. Conclusions on key issues

The analysis of this chapter has revealed a variety of stresses in: (i) the prevailing structure of Serbian banking; and (ii) the manner in which that structure is being recast by the very rapid growth of some private and foreign banks. The stresses arise from the poor operational efficiency of some of the established large banks but also of some of the banks that are most rapidly gaining market share. They also incorporate familiar concerns about the inadequate compliance of many banks - mostly in the state-owned sector - with capital adequacy and other regulatory norms.

It was concluded that the growth that has been seen since early 2002 does not create an immediate problem – most of the growing banks perform adequately in relation to capital adequacy. However, this situation relies uncomfortably on their ability, for the time being, to charge excessively for their credit and other services and to expand quickly in spite of generally low levels of operational efficiency across the system. Not only is this situation undesirable in itself. But it is also likely to be transferring problems on to borrowing enterprises that could easily show up in asset quality problems in the near future. It is also not a secure platform on which to base the ongoing growth of the sector. The next chapter identifies a number of key policy reforms that could put Serbia’s banking sector on a more sustainable growth path.

35

Chapter 4: Implications for Policy on Banking Sector Reform

The analysis of the previous chapter has revealed a number of critical inefficiencies in the way the Serbian banks operates. It is clear from this analysis that the spectacular expansion over the past two years has not been matched by the parallel improvement in banking sector performance indicators, which casts serious doubts on the sustainability of the current expansion. The next round of banking sector reforms should therefore aim at increased competition that is needed to establish the dynamic for sustainable channeling of financial resources to serve the needs of a growing economy on an efficient and affordable basis.

In this chapter we examine the main policy levers that the authorities have available to re-direct banking development in order to eliminate some of the main weaknesses that we have observed. Section 1 briefly reviews the main policy issues for various segments of the banking system. The next two sections focus on the key problem of resolution of state-owned banks. Finally, section 4 sketches the greater prioritization of banking supervision.

1. Policy Agenda

Similar to the experiences from other transition economies15, the current domination of Serbia’s financial system by state-owned banks presents significant risks, while also limiting the potential for developing viable financial markets. Given the historic role of banks in the Serbian economy, state banks are inherently prone to be misused as vehicles for patronage that worsen the prospects for competitive market development and incur significant macroeconomic and fiscal costs. This risk is particularly relevant given that the current scope of shareholder oversight and controls employed by the government at the sixteen banks wherein state has an equity stake, is neither well coordinated or resourced. At the same time, as shown by the analysis above, a number of state banks are gradually turning into ineffective shells that fail to perform a useful intermediation role and contribute little to competitive dynamics. In view of these risks, the authorities should in the nearest future adopt a strategy that would identify the clear, time-bound resolution scenario for all state stakes (including minority holdings), based on the careful consideration of costs and benefits in each particular case.

The local private banks raise concerns of their own. The stresses of relatively fast growth are already evident in some – e.g., in the form of strains on their capital adequacy – and their rapid growth of new lending will likely show up in problems in the years to come. More specifically, none of the local banks driving the recent growth score unambiguously well on a mix of effectiveness, regulatory, rating or loan-quality indicators. This gives rise to the major concern that some of the larger local-private banks are storing up future problems by lending on unsustainable terms that are likely to result in future deterioration in asset quality. This is an area of priority for greater supervisory oversight designed to anticipate and head off the possible problems that the analysis of this chapter has identified.

15 State-owned banks in the transition: origins, evolution, and policy responses, by Khaled Sherif et al. (World Bank, 2003).

36

The foreign banks have established a new and welcome competitive force in the sector. However, the evidence above indicates that this is not yet raising operational performance standards significantly. Rather the foreign banks may be adapting to the high-cost norms of the rest of the sector. Hence, their early successes and the possible delays in privatization referred to above raise another key question for government. Is it possible that the early and successful start for this first group of foreign banks will establish a competitive position for them that later arrivals may find hard to contest? This possibility seems more likely given that the next wave of foreign bank investment will be associated with the privatization of the state banks and so with the inheritance of the various management and financial problems to be found in these banks.

Table 4.1 – State Ownership in Banking Sector

Group Banks % state share

Market share (of unadjusted total assets)

Total Assets1 (US$ ‘000)

Vojvodjanska Banka a.d. Novi Sad (VB) 98.65 10.63% 733,743 1. Large Komercijalna Banka a.d. Beograd (KB) 29.08 11.78% 813,601

Jubanka a.d. Beograd (JB) 76.49 4.11% 283,542 Continental Banka a.d. Novi Sad (CB) 94.64 1.79% 123,629

2. Medium

Novosadska Banka a.d. Novi Sad (NB) 67.27 2.60% 179,736

Panonska Banka a.d. Novi Sad 82.91 2.13% 147,109 Privredna Banka a.d. Pancevo 92.42 1.00% 68,952 Credy Banka a.d. Kragujevac 60.58 0.46% 31,490

3. Small

Srpska Regionalna Banka 46.10 0.41% 28,238

Nacionalna Stedionica-Banka a.d. Beograd 10.00 1.47% 101,482 Srpska Banka a.d. Beograd 96.48 2.37% 163,865

4. Banks in which RS acquired shares on another basis JUBMES Banka a.d. Beograd 31.00 0.65% 44,888

5. Rehabilitation Niska Banka a.d. Nis 88.56 0.72% 49,963

Cacanska Banka a.d. Cacak 33.06 0.57% 39,297 Privredna Banka Beograd a.d. Beograd 16.36 0.91% 63,173

6. Banks with minority ROS shares

PB Agrobanka a.d. Beograd 14.90 2.22% 153,532

1 Total asset figures as of December 31, 2003 Source: BRA data

2. Approaches to resolution of state-owned banks

The responsibility for implementing the resolution of state-owned banks lies mainly with BRA.16 For operational purposes, the BRA has divided the banks with state ownership into the six categories shown in Table 4.1.17 These categories are largely self-explanatory. However, the state stake in many of these banks (notably in Komercijalna and Nacionalna Stedionica) is still difficult to pin down due to recent issue of shares, some of which are now being disputed by the authorities. The matter is further complicated because the London Club debts (principal and equity). when resolved, will add to the size of the state’s

16 This introduces certain limitations given that BRA’s present statutory authority (pending the approval of a new BRA Law) comes mainly from the 2002 Law on Regulation of Public Debt that guided the Paris-London Club debt swaps. So its authority over the group 4 banks as shown in Table 4.1 is limited. 17 The list of sixteen banks shown here differs from the eighteen in Table 2.3 of Chapter 2 in that it omits Borska and Pirotska banks. The first bank has already been placed into liquidation, and Pirotska is in the process of merger with Niska.

37

shares. In a few case this additional amount could be substantial and could shift a few banks from the category of minority-owned to the category of majority-owned.

The authorities have several key levers at their disposal to influence the performance of the state-owned banks. First, they can effect ownership changes via privatization. Based on the regional experience, this would involve the recruitment of major strategic investors for larger banks. Second, they could attempt the direct restructuring of some state banks – an approach that has not worked too well in the region and elsewhere.18 Third, they can liquidate particular banks whose financial weaknesses seem unlikely to be recoverable at reasonable cost.

The major point to note is that the 2002 debt-to-equity conversions that brought many banks under state control also moved their portfolio problems into the domain of the budget. To the extent that the authorities feel responsible for protecting depositors in these banks, many of the loan quality problems may have to be covered by the Serbian state whether or not the banks in question are restructured, privatized or liquidated. The question of what to do with state-owned banks must be primarily driven by an objective cost-benefit analysis of these various choices.

The costs in this analysis will come through: (i) fully recognizing unrecoverable non-performing assets and replacing them with good quality claims; and (ii) providing sufficient liquidity to allow restructured banks to free up the liability side of their balance sheets even though many of their assets will be locked into illiquid restructuring bonds and restructured claims on viable state-and socially-owned enterprises awaiting privatization. The magnitude of the first of these costs is probably little different between the two options of liquidation and restructuring. The privatization choice is only different to the extent that the new investors can be persuaded to pay a price above the true worth of the bank. (i.e., full state-sponsored restructuring before privatization would be expected to raise the price relative to the alternative of no restructuring). The second of the costs arises mainly in relation to the option of restructuring. But the magnitude of the cost would be affected by the type of restructuring bonds issued and, of course, by the amount.

The benefits side of this problem is often thought of in terms of the profit that government might make on privatizing a bank after incurring the costs of its successful restructuring. This is almost certainly illusory. The restructuring process in Serbia has stagnated for too long for there to be a significant premium over net assets in the price potential strategic investors are prepared to pay for any of the banks to be privatized. Instead, the benefits are more likely to accrue economically, in terms of introducing viable competition at the top end of a market that today appears to be supply constrained. To the extent that there might be any intrinsic value embedded in the brands of the more effective state-owned banks, the magnitude of any realizable gains from this will in any case depend on the long-term economic value of re-launching the banks as effective competitors in a supply constrained market. Regional and global best practice indicates that this impetus to banking sector development will come from the attraction of new strategic investors, including foreign banks, who will bring to the country appropriate credibility, as well as the new capital and managerial and technological know-how.

18 Sherif et al., Ibid.

38

The assessment of bank operational effectiveness indicators along the lines of Chapter 3 of the present report can be the starting point for the cost-benefit analysis by the authorities to decide which banks are worth restructuring and privatization, and which banks should be candidates for prompt liquidation. The regulatory and loan quality indicators can help to determine the costs associated with restructuring/privatization option as opposed to liquidation. Given the data imperfections acknowledged in the analysis above, the application of this methodology as described in Section 3 below should be seen as merely indicative of the types of results that seem likely rather than definitive.

3. Re-thinking the State-Bank Problem

In this section we use the analysis of Chapter 3 and the logic of the previous paragraphs to categorize the state-owned banks and then suggest possible ways forward. It is stressed that these suggestions are additional to the privatization actions that are already well advanced. More up-to-date information and better quality materials especially about asset quality will be needed to confirm the suggested dividing lines between banks that are shown (indicatively) in this present section.

Table 4.2 below compares 15 state-owned banks with the best and worst performing banks of the rest of the banking system.19 Using the banks’ ratings on four sets of indicators as explained in Chapter 3, the state banks can be divided into three distinct groups, each with its own policy implications:

Table 4.2 – Comparison of State Banks with Foreign/ Local-Private Bank Performance Grouped averages for banks under

state control

Average of best quartile foreign/

local-private banks

Average of worst quartile foreign/

local-private banks

Group 1 (some

potential)

Group 2 (inter

mediate)

Group 3 (very weak)

Number of banks in each group 8 8 3 5 7 Share of total funds mobilized 6% 8% 17% 9% 9%

Set A IEA:IBL Ratio 164% 73% 96% 102% 59% Weight of Operating Overhead 6% 24% 8% 15% 18% Combined Spread/Fee Yield 15% 28% 11% 18% 17% Gross Operating Return on IEA 9% 4% 4% 3% -1% Loss Wedge (Gross - Net ORIEA) 10% 3% 1% 2% 5%

Set B Capital Adequacy (min 8%) 100% 30% 23% 59% 4% Continuous Placements (max 60%) 10% 45% 50% 47% 349% Large Loans (max 400%) 11% 203% 278% 82% 430% Liquidity (min 1.00) 5.54 2.18 2.42 2.14 2.03 Forex Risk (max 30%) 24% 26% 14% 25% 99%

Set C Return on Assets 3% -3% 2% 1% -3% Cost income ratio (before bad debt charge) 76% 59% 68% 85% 97% Cost income ratio (incl. bad debt charge) 77% 123% 74% 93% 114% Return on Equity 12% -6% 14% 2% 1% Ratio of bad debt charge to total assets 0% 18% 0% 0% 1%

19 Two state-owned banks are excluded because the analysis in Chapter 3 showed their ratings to be “indeterminate”.

39

Set D Non-perf. share of potentially perf. assets 2% 35% 49% 64% 73% Provisioning as % of non-performing assets 189% 13% 33% 8% 13% Unprovided non-perf. assets as % of capital 0% 187% 255% 1141% 735%

Source: NBS data

Banks with apparent potential. Three state-owned banks (collectively Group 1 in table 4.2 above) have good effectiveness scores (rating “best”). Certainly these banks have problems which relate mainly to their performance on the indicators of loan quality where all three are rated among the “worse” banks. The poor quality of their loan portfolio also largely explains why these three banks score badly in terms of the regulatory indicators (ratings of “worse” or “average minus”). As noted in Chapter 3, however, to the extent that the 2003 diagnostic materials for could be incorporated into the analysis, these banks’ positioning in this group is unaffected. Given their relative effectiveness, these banks potentially have a useful role to play in filling the serious gaps in competition that the analysis above has identified. At the moment, however, they are severely constrained by their capital adequacy and ownership structures. If implemented properly, their restructuring immediately followed by privatization could be justified and could prove beneficial. This judgment does depend on full recognition of non-performance, fair valuation of collateral and long-term stability of funding. Weaker Banks. As shown by the analysis in Chapter 3, a number of the state-owned banks (seven in all) score unambiguously badly on all four sets of indicators, with ratings of “worse” or “average minus”. This group of banks is shown as Group 3 in Table 4.2. All of them are small/medium-sized banks and include two that are already in the merger/liquidation pipeline. If the data and results adduced in this report are confirmed, the initial presumption should be that these banks are candidates for liquidation unless restructuring/privatization of their large debtors in enterprise sector is going to make such a difference to their asset quality within a reasonable timeframe that their restructuring and privatization is a realistic prospect.20 As none of the banks in this group has significantly large market shares, early decisions to move to liquidation should involve few insurmountable problems. In-Between Cases. The remaining five state banks (Group 2 in Table 4.2) should be put through a four-stage evaluation as soon as possible to: (i) identify how much needs to be injected for depositors not to lose any money; (ii) add this to calculated income-earning assets and recalculate effectiveness indicators; (iii) add the same amount to capital and recalculate regulatory norms; and (iv) identify the impact on ranking relative to the spectrum of foreign/local-private bank performance. If banks could, with the minimal restructuring, become effective competitors they could then be privatized but only to investors with a sufficient depth of capital resources to cope with further possible deterioration in asset quality. 20 The case in point is Niska Banka, which has emerged as good candidate for privatization after Philip Morris had become the majority owner of Tobacco Industry Nis, and since paid Niska Banka over Euro 10.1m in debts for foreign loans guaranteed by the bank. This happened in the second half of 2003, i.e., beyond the time period covered by this report’s data analysis.

40

As noted above, all of these judgments rest crucially on the validity of the data that underpins the effectiveness and regulatory scoring. Two particular risks present themselves, both of which are known to be serious issues in Serbia:

• Directed loans to state and socially-owned enterprises may not be declared fully as non-performing even though many of these are overdue and are being rolled-over. At the same time collateral might be overvalued. The problem will not resolve itself until the privatization of state- and socially-owned enterprises allows an objective assessment of which of their obligations can be met. Both sides of this problem are part of government’s unaccounted “messy” debts. To the extent that this problem is eventually resolved, the cost of hidden non-performing assets is going to have to be met whatever choice is made on restructuring or liquidation. The paradox of this situation is that, in meeting the cost of hidden non-performing assets, through the issue of restructuring bonds, measured effectiveness and regulatory scores will be restored in a way that should leave the decision about whether to restructure and privatize or liquidate unaffected. In short, the judgments about which course of action to take can be well-informed by the analysis above even though there may be hidden non-performing loans in some of the banks.

• A second systemic problem is the non-commercial depositing of state and NBS

monies to fund the directed lending of some banks. This is thought to be preventing catastrophic liquidity failures in a number of majority and minority state-owned banks. The current authorities are currently implementing the plan to drastically reduce state deposits in commercial banks by the end of 2004. This matter is too confidential to be addressed quantitatively in this note but NBS must have the data to identify which banks are most vulnerable. Again, however, this issue should not determine the decision about whether or not to restructure or liquidate. What it will determine is what sort of investor should be selected to take over a restructured bank – where liquidity is being supported by state deposits only a major foreign bank will be able to provide the necessary back-up liquidity to allow the freeing up of these deposits.

The critical factor would seem to be the length of the transition period needed to achieve the exit of state from the banking sector. The three medium-sized banks, whose privatization is currently underway after substantial delays, account for less than 10 percent of total bank assets. In the meanwhile, the authorities have not yet decided on the timing and modality of privatization of state stakes in two of the system’s largest banks, which together account for nearly 20 percent of total assets. As neither the BRA nor MOF have sufficient capacity to effectively monitor the banks’ performance or provide timely guidance to the boards, further procrastination may lead to deterioration of banks’ condition and, ultimately, results in greater fiscal costs required for their resolution.

In this regard, the authorities need to expedite the adoption of a comprehensive divestiture strategy for banking sector. The strategy will build upon and expand the ongoing efforts under PFSAC II. Based on the cost-benefit analysis along the lines above, the strategy should include the clear, time bound resolution scenarios for all banks under state ownership. In addition to the policy actions proposed in the above discussion of three

41

groups of state-owned banks, the following broad recommendations emerge from the analysis so far undertaken:

• As several banks in Groups 1 and 2 that might have any potential to compete already have relatively good operational effectiveness scores, further operational restructuring of them is unlikely to add much value. Moreover, as noted above, there is a strong case for the early privatization of the larger of these banks to achieve a quick potential gain in competition at the top end of the market.

• Probably the economic value at risk from further delays in privatization is greater than any value that could be added by existing management or government trying to restructure the banks in lieu of strategic investor. In particular, the international best practice suggests that any recapitalization should occur only at the moment of bank’s sale, insofar as this is needed to facilitate a successful transaction.

• Because of the distortions known to exist within state bank recognition of asset quality problems and liquidity ratios, local placement of shares is unlikely to provide the financial depth or managerial controls to prevent problems reoccurring.

• A number of state-owned banks are unlikely ever to be able to compete effectively even if their balance sheets could be cleaned up (Group 3). For these liquidation will be the least painful and least costly option and should be decided upon promptly.

4. Prioritizing the Supervisory Agenda

Aside from the bank specific interventions just discussed, the authorities have an urgent task to establish and enforce stronger regulatory and supervisory discipline across the whole of the banking sector. This will need first an early re-clarification of which banks fall under which jurisdiction – the ambiguities between the BRA, MOF and NBS should be resolved with the burden of the work unambiguously assigned to the Supervision Department at the NBS. In parallel, the enforcement capacity of the NBS Supervision Department should be improved through the implementation of Supervisory Development Plan over the next two years.

As a matter of priority, BRA, MOF, and NBS should act promptly to strengthen their fiduciary oversight of state-owned banks, closing the opportunities for regulatory arbitrage by bank management which could negatively impact these banks’ sale or resolution. As a minimum, each bank should be subject to robust and regular monitoring by the BRA as an agent of the government, as well as close scrutiny by the NBS supervisory team. The shareholder’s direct representation and participation in the board of directors is a pre-requisite to protecting its interest and ability to hold management accountable for the bank’s financial and operational performance, including in cases where the state holds a minority stake.

As was made clear in Chapter 3, the problems in the banking sector are not confined to the state-owned banks. The spotlight given by the authorities and the donor community to resolution of state-owned banks should in no way distract attention from the potential problems posed by the rapid expansion of domestic private banks. None of the local banks driving the recent growth score unambiguously well on a mix of effectiveness, regulatory, or loan-quality indicators. This gives rise to the major concern that some of the larger local private banks are storing up future problems by lending on unsustainable terms that are likely

42

to result in future deterioration in asset quality. The rapidly growing large private banks should be the priority candidates for early on-site examinations. These examinations would be used, inter alia, to check whether their high growth rates are compromising asset quality and longer-term capital adequacy. Work needs to be established with a specific mandate to probe loan classifications with worse case scenarios being frankly discussed with bank managements (i.e., the worse case involving maximum loan classification and minimum assessed collateral valuations) and remedial action taken. The NBS “fit and proper” criteria for ownership should also be vigorously applied, to decrease the amount of “pocket banks” used as treasury divisions of certain industrial groups.

One of the key lessons of financial sector reform in the past decade is that the tendency of authorities to delay action on weak banks creates the potential for financial crises with an eventual need to transfer the losses to society at large.21 In this regard, the Serbian authorities should be prepared to deal promptly with potential bank failures that could result in the near future from the convergence of market, financial and regulatory pressures, such as the increased minimum capital requirement, removal of public deposits from commercial banks, possible currency depreciation, and acceleration of bank privatization. It is quite probable that in the next year or two, some banks offered up for sale may not attract a buyer, while others may be found capital deficient or may experience liquidity constraints. The NBS and government are encouraged to organize a high-level working group to study existing vulnerabilities and put in place a proper early warning system. Amendments to the bank exit/resolution mechanism may need to be developed, to minimize the fiscal cost, while protecting depositors' insured accounts and maintaining asset values and avoiding possible systemic risks. Given that Serbia had 47 licensed banks at the end of 2003, the orderly consolidation of the banking system should be seen as the end objective of changes in the regulatory and supervisory regime.22

21 What Have We Learned in the 1990s? (World Bank, 2004 (draft)). 22 Bank Consolidation in ECA Region, by Alan Roe, et al. (World Bank, 2003 (draft)).

43

Chapter 5: Non-Bank Financial Institutions

The currently limited access to finance and high cost of banking services, highlighted in Chapter 6 of this report, present a major challenge as well as an opportunity for non-bank financial intermediaries in expanding their clientele base and serving the needs of the economy. The experience of other countries demonstrates that NBFIs can contribute to a broader array of financial services and instruments, with products often unavailable from banks. Moreover, some types of non-bank credit organizations are particularly well-suited for targeting the low-income and socially vulnerable groups of population in urban and especially rural areas, thus positioning themselves in market niches usually unattractive for banking services. At the same time, the experiences of other transition countries shows that poorly regulated NBFIs can easily become the breeding grounds for serious financial malpractice and even the source of systemic crises. Hence, although the Serbian NBFIs are currently small relative to the banking sector, a sound framework for their expansion needs to be put in place as soon as possible to ensure their sound development in the future.

This chapter will provide an overview of the NBFIs sector, followed by a more in-depth look at the insurance, leasing, and capital markets, which represent the most active and rapidly growing segments of the sector. It will close with discussing existing challenges and potential policy interventions aimed at the increasing role and efficiency of Serbia’s NBFI sector.

1. Overview of Serbia’s NBFI sector development

Despite the broad consensus existing in Serbia about the potentially important role of non-bank financial services for the country’s economic development, sustainable growth and enhanced social protection, this sector is still seriously underdeveloped and faces a number of legal, regulatory and institutional constraints. As in many other transition economies of CEE, the NBFI sector is lagging behind the developments in the banking sector both in terms of its size and market share of all financial services as well as in terms of quality of services and its public recognition. As can be seen from the Table 1.1 in chapter 1, total NBFI assets represent less than 4 percent of GDP, far below the level seen in the neighboring transition economies such as the Czech Republic, Hungary and Slovenia where NBFI assets present 20 percent, 16 percent and 10 percent of GDP, respectively. Table 5.1 below compares the size of various types of Serbian NBFIs to their peers in other developed and transition economies.

There are many factors that inhibit the successful development of NBFIs in Serbia. On the supply side, these include delays with pension reform, non-conducive and EU non-compliant insurance legislation, incomplete capital markets legal and institutional framework, weak and fragmented supervisory capacity of the financial regulators, often non-existing information disclosure requirements, unreliable reporting, and low transparency of the market. Moreover, the poor quality of services, bad marketing and deep distrust of the public associated with numerous failures of credit institutions and insurance companies in 1994-1995, coupled with the poor conditions for consumer and investor protection, discourage the demand for NBFI services.

44

Table 5.1 – Assets of NBFIs in 2002 (percent of GDP) Country Insurance

Premiums Private

Pension Assets Investment/ Mutual Funds

Market Capitalization

USA 9.58 64 58.7 105.4 United Kingdom 15.1 87 18.6 119 EU 8.7 29 49.6 68.97 Poland 3.21 4.39 2.91 15.32 Hungary 2.9 2.79 4.73 19.91 Croatia 3.2 28 Slovenia 5.03 0.19 1.20 21.82 Czech Republic 4 3.12 4.98 22.84 Serbia 2.9 N/a 0 7.2 Ukraine 1.9 0.24 0.14 11.6

Source: World Insurance in 2002, Sigma, Swiss Re Economic Research & Consulting World Bank Working paper # 28, Development of Non-bank financial institutions and capital markets in European union accession countries, by Marie Renee Bakker, Alexandra Gross (Washington D.C., 2004)

Because the sector is very small, it does not yet present any systemic threat to the stability of the financial sector as a whole and the economy at large, with the possible exception of insurance arms of large banking groups. At the same time, the deepening regulatory arbitrage between still weak, but improving banking supervision on the one hand, and inadequate supervision of NBFIs on the other hand, may negatively impact market discipline, undermine the effectiveness of the government pursued development agenda and discourage the flow of local and international capital into the banking and NBFI sectors. This regulatory arbitrage arises from inadequate quality of regulation and supervision of the NBFIs compared with banks. The quality and comprehensiveness of the legislation and regulation in this area is far from the best international and EU practices. In addition, the institutional capacity of financial regulators is weak, supervisory awareness of the existing problems (based on off-site reporting and on-site exams) is at best superficial and at worse non-existent, and enforcement capacity is severely constrained.

Lessons from other countries indicate that intricate cross-ownership can cause unforeseen and rapid contagion and manipulative accounting of financial groups and its members easily flourishes in markets characterized by significant regulatory arbitrage. This in turn can result in the artificial inflation of the capital of financial intermediaries, window dressing, tax evasion schemes and other sorts of illegitimate or highly risky activities of financial intermediaries which can undermine the stability of the whole financial sector. Many regulators in the world have responded to these developments by introducing consolidated supervision of banks and their affiliated parties.

As part of the broad reform agenda, the GOS has recognized some of the existing weaknesses and made some efforts towards revising the outdated legal framework. The Law on Leasing and the Law on Registered Charges on Movable Assets, and the Law On Insurance and amendments to Law on Securities Market were passed in May/June 2003 and May 2004, respectively. Early efforts have been made towards drafting of a much-needed law on non-state pension funds. Appropriate legal and regulatory reform is a necessary but not sufficient condition for the creation of a sustainable and dynamic industry of new financial services in Serbia. A comprehensive strategy for the development of the financial services industry, with the focus on fostering the development of the sound

45

private, competitive, modern non-bank financial intermediaries also needs to be considered by the new government.

2. Insurance Sector

At the end of 2003, 36 insurance and 3 reinsurance companies were registered in Serbia. Three more companies had also received insurance license; however they were not yet operational.

A number of insurance companies were established by the Serbian banks. One of the unusual features of the Serbian financial sector is that several banks were also established by the insurance companies. As will be discussed below, these intricate arrangements call for even greater diligence in the supervision of banks and their insurance company subsidiaries/owners than might normally be necessary.

The Serbian insurance industry can be divided into four groups of companies:

• Group one consists of two large state-owned companies, Dunav and DDOR Novy Sad. These companies report a significant amount of accounted but unearned premiums from its clients – mainly socially owned companies – and are generally viewed as inefficient institutions still enjoying implicit government guarantees.

• Group two includes many companies dating back to the SFRY period. These are characterized by glaring inefficiencies and inability to adjust to the new market requirements and competition. It is expected that many such companies will exit the market with introduction of new capital requirements in the new Law on insurance.

• Group three is “pseudo insurers” which include captive insurers and companies created for tax evasion purposes. They exist largely due to the tolerance of the regulator (i.e., absence of enforcement capacity) and gaps in the legislation, especially in the area of consumer protection.

• Group four includes a few “genuine” private insurance companies which provide a variety of products more compatible with the modern insurance business. While this group is insignificant, it may soon grow into more sizeable market by encouraging entry of foreign reputable insurance companies and with development of life insurance business in Serbia.

The market is highly concentrated with the 2 largest state-owned insurers (Dunav and DDOR Novy Sad), collecting more than 70 percent of total premiums and 34 percent of life insurance premiums. The top ten companies account for 90 percent, leaving the remaining 10 percent of premiums to 26 smaller companies. While the role and the presence of foreign insurers in the Serbian market is increasing (there are now 6 majority foreign owned insurance companies as compared to 4 in 2001 and 2 in 2000), they so far account for less than 3 percent of total premiums as evidenced from Table 5.2 below.

The insurance industry is growing relatively fast, recording 24 percent increase in premiums in 2002, and 118 percent growth in 2001. The growth mostly comes from motor, property and casualty insurance, and third party liability insurance. About 26 companies operate mainly with mandatory insurance (such as motor insurance). Despite the earlier history of a large life insurance industry in the SFRY, the long years of hyper inflation,

46

market instability, closure of numerous insurance companies, excessively burdensome tax legislation and poor financial regulation jeopardized the development of this market and undermined public trust. Lengthy claims settlement processes which can take up to 5-10 years have made life and property insurance unreliable for both social protection as well as longer term investment. As a result, life insurance accounted for only 1.26 percent of total collected premiums in 2002, or just US$4 million. This is remarkably low, when compared to life insurance premiums collected in Hungary (US$779 million), Slovenia (US$252 million), Croatia (US$146 million) and Bulgaria (US$78 million) in the same year.

Table 5.2 – Insurance Market Indicators, 2000-2003 Indicator 2000 2001 2002 2003 Insurance Companies registered in Serbia 35 33 36 36+3 Incl. Life Insurance Companies 17 16 18 18 Companies with foreign capital 5 5 8 8 Companies majority foreign owned 2 4 6 6 Reinsurance companies registered in Serbia 3 3 3 3 Agents/brokers 126 131 145 147 Number of insurance companies liquidated 2 5 0 0 Total premiums, mln US$ 125.9 256.4 365 427.1 Total premiums, mln Dinars 7,953 17,350 21,545 12,336 of which premiums collected by state companies 6,331 12,033 15,147 16,112 premiums collected by foreign companies 265 644 551 836 premiums collected by life-insurers 47 156 271 942 Total claims, mln Dinars 6,122 9,942 11,417 12,904 Total paid claims, mln Dinars 3,892 7,007 7,767 8,532 Total technical reserves, mln Dinars 3,274 6,223 6,881 9,010 Total reinsurance with local companies, mln Dinars 565 1,279 1,692 2,416 Total capital, mln Dinars 9,921 15,392 18,627 20,519 Reported operating costs , mln Dinars 1,467 4,716 6,575 8,287 Insurance density (premiums per capita), Dinars 743 1,621 2,032 3,111 Insurance density (premiums per capita), US$ 11.7 23.9 34 56.95 Insurance penetration (premiums in % of GDP) 2.52 2.9 3.07 2.22

Source: Ministry of Economy, Department of Insurance Supervision The claims paid by the industry in 2002 stood at 68 percent of total filed claims and 36 percent of collected premiums, while technical reserves were lower than the amounts of claims paid. This suggests that majority of insurance companies underestimate their potential costs and so report inflated profits. Such practice leads in turn to a serious erosion of the capital of insurance companies and undermines their capacity to meet future claims. At the same time, the quality of services remains low and the costs are relatively high, with operating costs at 30 percent of total premiums.

The majority of insurance companies are small and institutionally underdeveloped. They are also characterized by poor corporate governance, weak internal procedures, poor quality of accounting and inadequate risk management and assets diversification procedures and systems. Institutional weaknesses typical for the sector include the following:

47

• Reported financial data is unreliable, inconsistent and is subject to manipulation by the companies;

• Serious underestimation of technical reserves and thus under-provisioning which impacts the solvency and potential survival of many insurance companies, including some of the largest institutions;

• Poor accounting and audit practices with no internal controls in place; • Poor diversification of assets, with more than 20 percent and sometimes up to 50

percent invested in fixed assets; • No transparent data on reinsurance to allow a proper assessment of the risks

reinsured both within the country and transferred abroad. • No requirement on the companies to publish their annual audited reports which

raises major concern with respect to protection of policy holders and investors rights.

Regulatory and Supervisory framework

The existing legal framework, based on the 1996 Insurance Law, has proven inadequate for dealing with the apparent inefficiencies of the insurance sector. Drafted on the basis of international best practice, the new Law on Insurance, adopted in summer 2004, seeks to correct many of the gaps identified by local and international experts (see, e.g., the Bank’s report dated November 2002). Most importantly, the new law introduces a EU compliant classification of insurance products by life and non-life groups. Moreover, the new law establishes new, much more stringent capital requirements to be effective from January 1, 2005 which most likely will lead to significant market consolidation. Other improvements include stricter requirements for establishment and licensing of an insurance company, reporting and pre-approval of qualified share, requirements for “fit and proper” testing of company owners and managers, limitations on insider activities, provisions for registration and regulation of insurance intermediaries, detailed provisions for calculation of reserves, settlement of claims and risks management, and updated auditing and accounting provisions.

Even with the passage of a very good law, however, the ultimate success of insurance sector restructuring will depend on the powers and quality of the insurance regulator. The existing regulatory authority is critically under-resourced and needs to be immediately reconfigured. While even the old (1996) law allowed on-site examinations, there is still no Inspections manual, or adequate staffing and training for supervisors to launch this work. At the same time, the enforcement powers and capacity of the supervisor are extremely limited; the practices and procedures for risks assessment do not exist and a traditional supervisory cycle incorporating results of on-site, off-site supervision, market intelligence and other types of analysis and sources of information is not established. Reports from companies are still received on an annual basis in paper form.

In recent months, there has been some progress on the long-debated issue about the best path for reform of the regulator and its positioning among other government institutions. The new law on insurance transfers insurance supervision responsibilities to the NBS. While this move represent a logical immediate solution since bank supervision has definitely stronger powers and better trained staff for prudential supervision, it has a

48

number of risks as well. The most obvious risk, which has certainly been considered by the authorities, is the danger of stretching the bank supervision’s already limited capacity into new areas of financial intermediation. This could leave banking sector regulation without the ability to fill the crucial gaps in its own work (as identified in Chapter 4) and slow down the process of overall regulation and supervision at this critical stage of its development.

More importantly, although they share many common features, banking and insurance remain different industries, and thus require specialized education and experience for the proper assessment of risks and the identification of corrective actions. In this respect, the activities of insurance and pension funds as well as investment funds have more in common, and thus higher synergy can be achieved there. This is turn raises a broader question about the integration of supervisory structures, related to the development of supervisory capacity for the entire NBFI sector. The authorities are encouraged to weigh the benefits and costs of integrated supervision presented in Box 5.1, keeping in mind that temporary solutions often last longer than expected and they often cost more than anticipated.

Box 5.1 – Integration of Supervisory Structures

The main arguments for full integration of the supervisors include:

• For small financial systems like Serbia, a unified approach permits significant economies of scale and enables pooling of the existing regulatory expertise.

• In small financial systems with high degree of industry concentration and cross ownership between institutions, it can help enhance the effectiveness of supervision and reduce the scope for regulatory arbitrage.

• An unified approach allows financial intermediaries to conduct business on the basis of a single authority’s permission (license).

• Integration of supervision over homogeneous risks allows regulators to achieve synergy and thus reduce transaction costs.

• There is a greater degree of flexibility in deployment/rotation of the staff and larger career opportunities.

At the same time, the following arguments against unified supervision have been raised:

• An integrated approach appears to offer limited opportunities for economies of scope as opposed to economies of scale.

• Tension and problems can arise from existence of different culture and interpretation of the regulatory role and powers stemming from the distinctive nature of regulators for example in such area as banking and securities.

• There may be potential risks of diseconomies of size and increased cost for the industry and tax payers. • If the personnel is limited and not adequately trained, a single regulator may end up focusing on the

priority areas of supervision, thus neglecting the importance of other segments of the supervised financial services market.

• Large agencies tend to grow into more powerful and independent bodies and can become less accountable and responsive to the industry institutions.

The experiences of countries which undertook financial supervision reform in the recent decade include a wide variety of approaches aimed at achieving synergy and increasing effectiveness of regulation without unjustified expansion of the immediate and future costs of the reform. The economies with smaller and/or underdeveloped financial markets tend to choose the integrated supervision approach, which can be implemented in one (full integration) or two (partial integration) phases. Great Britain's Financial Services Authority is a good example of the integrated supervisor. This path was recently followed by a number of countries – Hungary, Germany, Korea, the three Baltic states, Kazakhstan, etc. – that decided to integrate all of the existing supervisory agencies into a single body. Australia, Bulgaria, Ukraine, Poland have been enhancing and consolidating their supervisory

49

function; however, they still have several regulators in the system. Before integration of all the supervisors, Hungary merged banking supervision and securities market regulation as a first phase of the reform. It should be noted though that majority of countries which adopted an integrated supervision approach and established a single supervisor have started their reforms in the environment where individual supervisory agencies regulating the crucial and largest financial services industries (banking, insurance, pension funds) were already well established and far more developed than in Serbia. It is also important to note that, on its own, integrated supervision does not compensate for the lack of professional knowledge and unresolved policy issues such as the degree of independence of a regulator, compensation and motivation of civil servants, staffing and professional development.

3. Leasing

The passage of the Leasing Law in May 2002 became a major impetus for rapid development of the leasing industry in Serbia, with seven new companies registered in the course of 2003 (see Table 5.3).

Table 5.3 – Activities of Leasing Companies in Serbia, December 2003

Name of a Company Country of Origin Registration date Assets, EUR mln*

Raiffeisen Leasing Austria Feb. 2003 11 HypoLeasing Austria June 2002 20 Lubljanska Banka Leasing Slovenia Sept. 2003 N/a VolksLeasing Austria 2003 N/o S-Leasing Austria Oct. 2003 N/a Delta Serbia Nov.2003 N/o HVB Leasing Austria 2003 N/a TOTAL 31

* Experts estimate; n/a – not available; n/o - not operational yet At least four new leasing companies became fully operational in 2003, including Austria’s HypoLeasing and Raiffeisen Leasing that managed to create leasing portfolios of Euro 20 million and Euro 11 million, respectively. Virtually all the existing companies are direct subsidiaries of reputable Western banks. This phenomenon is attributed to the fact the current law prohibits banks from doing direct leasing. Moreover, the minimum start up capital requirement for leasing companies of Euro 100,000 sets rather a high barrier to market entry by local capital constrained institutions, thus leaving the business to larger international investors. Leasing companies are registered and regulated by the general Company law. They are not subject to any financial regulation and submit annual reports to the Solvency Center at NBS.

To mitigate some of the existing risks and coordinate the efforts, the existing leasing companies have united in a Leasing Committee established under the auspices of the Bankers Association. In the absence of a credit bureau and given the costly enforcement procedures, they have also established a “black list” of non-performing borrowers.

Experience from other CEE economies suggests that Serbian leasing industry can grow very fast into an important segment of the financial market. Local analysts forecast an increase of at least four times in outstanding finance lease assets in 2004 to a new total of Euro 170 million.

50

Notwithstanding the existing problems with evaluation of collateral, registration of security interest and repossession of assets, leasing is proving to be an attractive instrument for medium term capital investments. Many SME businesses are availing themselves of this new financing opportunity as they often do not possess adequate collateral to access bank lending. “Finance leases” now account for almost 85 percent of transactions, while several years ago only the “operating lease” was practiced in Serbia. More favorable tax treatment of corporate purchases of equipment under finance lease agreements encourages leasing companies to invest in productive assets and transport equipment. As would be expected at this early stage of development, up to 70 percent of equipment purchased on finance lease terms are vehicles, another 20 percent are equipment and machinery.

The main remaining obstacle in developing leasing services in Serbia lies in the area of lease registration. The law on leasing foresaw the establishment of a lease registry; however, its establishment has been delayed. It seems that the ambiguity in interpretation of the existing tax legislation (specifically, sales and services tax for corporate and retail clients) has been resolved by passing amendments to the respective legislation in June 2003. At the same time, the uncertainty about the potential impact from the expected introduction of the VAT law in January 2005 remains and may hinder leasing development in the second half of 2004 if the issue is not properly and rapidly addressed. Changes are also needed in the legislation on execution service to allow faster repossession of assets and recovery of bad debts.

4. Capital Markets

The Serbian securities market is relatively young and unshaped; its infrastructure and instruments are still at the early stages of development. Total market capitalization at the end-2002 equaled Dinar 40 billion (Euro 634 million). By March 2004, the market capitalization grew to Dinar 203 billion which equaled 28 percent of GDP. Securities are traded at the Belgrade Stock Exchange (est. in 1989) and on the OTC. NBS is selling NBS bills and Foreign Currency Savings Bonds (RS Bonds) in the primary market.

BSE is the main trading floor in Serbia. Its turnover reached Dinar 93 billion at the end of 2003, with shares of privatized companies and corporate bonds accounting for the largest portion of trades at BSE (see Figure 5.1). BSE has 57 shareholders and 79 members, including 72 brokerage and dealing houses and 7 banks. Securities are traded at BSE via electronic trading system. BSE organizes two types of trading, i.e., simple price and open prime auction. Up to October 2003, NBS bills were offered at BSE and represented about 30 percent of total trading volume. After moving NBS bills trading to the central bank, the overall turnover at BSE declined despite the increase in the trading of shares and corporate bonds.

Foreign Savings Currency Bonds (RS Bonds) are considered to be the most liquid and attractive securities both for local and foreign investors. While RS Bonds are not convertible, they can be used to buy shares in the socially-owned companies undergoing auction privatization. Since their introduction in September 2002 (12 series, maturity till 2016, total issue – Euro 4.2 billion), they are traded on the continuous basis.

51

The market for corporate bonds and shares of privately owned enterprises is growing fast, although it remains less liquid and transparent. In 2003, 178 issues of private short-term corporate bonds were registered by SEC for the total amount of Dinars 14.3 billion. There were no IPOs registered and despite the existing rules and procedures, BSE did not list a single company yet. There are no portfolio investors and in the absence of the special legislation on investment funds (which was drafted but not endorsed by the Government), investment and equity funds are not yet operating in the market.

Figure 5.1 – Belgrade Stock Exchange Turnover in 2003

Shares 32.43%

Corporate bonds24.42%

Government bonds0.96%

Foreign Currency Savings Bonds

10.80%

Corporate commercial notes

13.36%

Giro Funds3.35%

NBS bills 14.68%

Source: Belgrade Stock Exchange

Similar to experiences in many other ECA countries, Serbia’s capital market still has not been able to establish effective mechanisms for corporate sector financing. The market remains shallow, non-transparent and the real market capitalization is probably lower than the reported data. This is partially explained by the inefficiencies of the enterprise sector and the relatively small size of the financially more sound companies. Most large companies are often overburdened with old debts and efficiency problems (see also Chapter 6).

The existing regulatory and institutional framework for the securities markets is in need of substantial changes. While the basic regulations are in place, they do not seem to have sufficient detail for encouraging sound market conduct and its further deepening. It may be useful to issue the Corporate Governance Code based on the respective OECD principles for corporate governance. The major legislative gaps include poor protection of minority investors’ rights, inadequate principles of corporate governance and weak requirements for disclosure of information. In addition to the above-mentioned deficiencies, Serbia still does not have functioning legislation for portfolio investors and assets managers. Similarly, while the basic institutions already exist, such as one stock exchange, several clearing and settlement systems, OTC and several registries, and independent regulator, their activities need further enhancement and streamlining. This includes creation of a single securities registry, introduction of custodians and asset-management licensed activities.

52

The reputation of the securities regulator, the effectiveness of its regulations and strong enforcement capacity are the major prerequisites for the future development of sound capital markets. The Serbian Securities Commission (SSC) (est. 1990) has a broad mandate for regulation of capital markets established by the 1994 Law on Securities and Capital Markets. Currently it has a staff of approximately 30 experts. A potential major issue for policy discussion relates to the recent initiative to reconsider the powers and status of the SSC. The proposed revisions to legislation call for revoking the SSC status as an agency independent from the Government due to the alleged conflict between the current text of the legislation and the Constitution. Made effective, these revisions will significantly curtail the SSC’s independence and powers, which would contradict the IOSCO principles and best international practice.

5. Proposed reform agenda for NBFIs

To summarize, despite the potential and growing demand for products and services of various NBFIs, their current level of development is not satisfactory and future market prospects are not clear. The immediate benefits of NBFI intermediation will be realized only after the following priority reforms in the legal, regulatory and institutional areas are undertaken:

• Insurance market reform agenda should include: (i) the strengthening of the existing legislation through enactment of the new Insurance Law and related by-laws; (ii) considerable enhancement of the regulatory and supervision framework according to the best international practices and IAIS principles, (iii) diligent examination of the existing practices of insurance companies to ensure their compliance with the regulatory requirements; and (iv) managed exit from the market of the non-compliant and poorly performing companies.

• Leasing will grow fast and provide much needed resources for SME finance and assets modernization, provided that: (i) clarifications and if necessary, proper legal amendments, are introduced with respect to VAT treatment of leasing transactions to make it a more competitive and attractive source of funding as compared to bank lending; (ii) a streamlined, publicly accessible registration system for secured interests and lease transactions is established; (iii) repossession of leased assets in the case of a lessee’s default is significantly simplified and allows out-of-court dispute resolution.

• Capital markets cannot be expected to grow fast in the absence of institutional and portfolio investors. Thus, the authorities need to focus on establishing a sound legal framework for the emergence and operations of investment funds. In parallel, the powers of the Serbian Securities Commission need to be redefined in order to ensure proper oversight of the securities market, introduce adequate reporting according to IFRS and IOSCO principles, enforce information disclosure and protect the interests of minority shareholders. Special legislation on mortgage securities is needed to allow the issue of lower risk secured mortgage bonds and mortgage backed securities. To increase access to publicly traded shares and diversify investment portfolio, Serbia may eventually want to participate in the current discussions on establishment of the regional trading system/stock exchange.

53

• Development of capital markets and accessibility of long-term finance will largely depend on the success of pension reform and the emergence of private pension funds (voluntary pension schemes). Special legislation and adequate supervisory authority for pension funds and asset managers will need to be established to ensure a sound development of contractual savings schemes.

• Factoring and other types of assets-based finance can grow into an important segment of the financial market, provided that: (i) existing institutional weaknesses (such as limited knowledge and skills of financial intermediaries, inefficient executive service); and (ii) legal impediments (such as ambiguous legislation for assets-based finance, costly and lengthy enforcement of collateral, specifically, procedures for collection of assigned receivables) will be addressed. Serbia may consider ratification of the UNIDROIT Convention on International Factoring and the most recent UNICITRAL Convention on Assignment of Receivables to encourage development of factoring and facilitate trade finance.

• The non-bank credit institutions, which are currently very small and weak, may potentially increase their market share and the role in SME finance. While this process may take a long time, it can be accelerated by directing some of the existing state-supported schemes (e.g., Guarantee Fund, Development Fund or other special purpose facilities) via non-bank creditors, provided that the National Bank of Serbia will introduce better reporting and accounting requirements as well as ensure proper supervision of the non-bank credit institutions.

Many of these areas outlined for reform require specific, more detailed action programs and more importantly, a dedicated constituency which is willing and capable to deliver these reforms in a consistent, transparent and comprehensive manner. For this purpose, the authorities are also encouraged to undertake the following steps in reforming the NBFI sector:

(a) develop and adopt a long term comprehensive strategy for financial sector development in Serbia, highlighting the objectives and expected results of the reforms in the individual segments of the financial market (banking, non-bank financial intermediaries, capital markets) as well as the cross-linkages and synergies in the proposed reforms and their benefits to the economy of Serbia. This work will require the creation of a special working group with participation of various stakeholders. Similar work was undertaken in many countries in recent years (Canada, US, UK, Ukraine, Russia, Kazakhstan) and served as a sound foundation for market development and regulatory reform;

(b) develop a strategy for financial regulation reform in Serbia, incorporating best

practices and lessons learnt, especially from the countries in transition. This should include the pressing as well as potential needs stemming from the pension reform and development of new sophisticated financial instruments and institutions. The reform should not result in undermining the role or reducing the powers of the existing regulators as it may send a wrong signal to the market and international community. It should instead build on the existing strengths, by adding synergies and reducing the costs of the reform to the tax-payers and market participants. It

54

should also consider introduction of market levies to support financing of the financial regulator(s) activities.

(c) establish an integrated, harmonized legal framework for NBFI activities. Depending

on the existing practices, this can include: (i) framework legislation; (ii) special legislation; and (iii) by-laws and regulatory acts (see Figure 5.2 for an example). To that effect, authorities will need to review the distribution of powers and responsibilities between various ministries and agencies in developing financial sector related legislation and rationalize these efforts by identifying core professional agencies to be responsible for such work on a consistent basis;

(d) consider joining global professional associations, such as International Association

of Insurance Supervisors (IAIS) and International Network for Pension Insurers. Membership of financial regulators in these networks not only opens a great opportunity for knowledge sharing and eases access to the international experience, but also provides a channel for continuous knowledge enhancement and training for supervisors;

(e) encourage development of professional public associations and strengthen those

already existing (Bankers’ association, Association of Brokers and Dealers, Insurance Association). Experience shows, that such public professional organizations can play a crucial role in: (i) enhancing market discipline and improving business conduct and ethics of professional participants; (ii) offering training for market participants; (iii) representing the interests of the market participants in policy discussions and legislative consultations; (iv) increasing awareness and building public trust in financial institutions by promoting the knowledge about the industry to the customers and various stakeholders.

55

Fi

gure

5.2

– H

olis

tic L

egal

Fra

mew

ork

for

activ

ities

of N

on-B

ank

Fina

ncia

l Ins

titut

ions

So

cial

Pol

icy

Dev

elop

men

t St

rate

gy

Fina

ncia

l Sec

tor D

evel

opm

ent

Stra

tegy

Ec

onom

ic D

evel

opm

ent

Polic

y/St

rate

gy

Judi

cial

Sys

tem

and

Pr

actic

es

Nor

mat

ive

acts

on

regu

lato

ry a

nd

supe

rvis

ory

fram

ewor

k fo

r NBF

Is (t

o be

dev

elop

ed b

y a

Reg

ulat

or fo

r N

BFIs

)

Nor

mat

ive

acts

of o

ther

min

istri

es o

n de

velo

pmen

t of i

ndiv

idua

l sec

tors

of e

cono

my

that

rela

te to

act

iviti

es o

f NBF

Is (a

gric

ultu

ral

sect

or, S

MEs

, tra

nspo

rt, h

ousi

ng, e

tc.)

Etc.

Law

on

priv

ate

pens

ion

fund

sLa

w o

n In

sura

nce

Law

on

Mut

ual I

nves

tmen

t In

stitu

tions

(Inv

estm

ent

Fund

s)

Law

on

Leas

ing

Law

on

Trus

t/ As

sets

M

anag

emen

t

Law

on

Mor

tgag

e

Legi

slat

ion

that

re

gula

tes

spec

ific

aspe

cts

of a

ctiv

ities

of

NBF

I: La

w o

n D

eriv

ativ

es;

Law

on

Paym

ents

Sy

stem

; La

w o

n C

urre

ncy

Re g

ulat

ion

Legi

slat

ion

on

Con

tract

ual R

ight

s

Civ

il an

d C

omm

erci

al

Cod

es

La

w o

n Se

cure

d In

tere

sts

Legi

slat

ion

on R

ules

and

Pr

oced

ures

of A

ccou

ntin

g,

Audi

t, D

iscl

osur

e of

In

form

atio

n an

d R

epor

ting

La

w o

n Ac

coun

ting;

Va

rious

nor

mat

ive

acts

of

MO

F, N

BS, T

ax

Adm

inis

tratio

n, S

EC,

etc.

Taxa

tion

Legi

slat

ion

for

NBF

I: La

w o

n En

terp

rise

Inco

me

Tax;

La

w o

n VA

T;

Law

on

Pers

onal

Ta

x;

etc.

Legi

slat

ion

on

esta

blis

hmen

t and

le

gal a

spec

ts o

f ac

tiviti

es o

f NBF

I: La

w o

n Ec

onom

ic

Entit

ies;

La

w o

n En

terp

rise;

La

w o

n C

oope

rativ

es

(und

er d

evel

opm

ent);

etc.

Law

on

Secu

ritie

s an

d St

ock

Exch

ange

La

w o

n Ba

nks

and

Oth

er

Fina

ncia

l Ins

titut

ions

Cap

ital M

arke

t D

evel

opm

ent S

trate

gy

Bank

ing

Sect

or

Dev

elop

men

t Stra

tegy

Law

on

Nat

iona

l Ba

nk

Stra

tegy

for D

evel

opm

ent o

f Non

-Ba

nk F

inan

cial

Inst

itutio

ns

Con

cept

ual L

evel

Stra

tegi

c Le

vel

Fram

ewor

k La

ws

Supp

ortin

g Le

gisl

atio

n (d

escr

ibes

spe

cific

ar

ea o

f act

iviti

es o

f N

BFIs

)

Spec

ial L

egis

latio

n on

Indi

vidu

al N

BFIs

Nor

mat

ive

Acts

on

NBF

Is

56

Chapter 6: Finance and the Real Economy

Notwithstanding the initial steps on banking reforms in the past three years, Serbia’s international benchmarking presented in Chapter 2 suggests that the current levels of financial intermediation are not sufficient to support the needs of a growing economy.

Serbia is considered to be a “late riser” in the sense that it has only recently experienced a rise in the volumes of bank credit to the real (private) sector.23 As a consequence the share of all credits from banks to the enterprise sector amounts to only 19 percent of GDP24. This is quite low compared with what one sees in more advanced transition countries or in comparison with levels from the past. Although some forms of NBFI support to the private sector are growing quite fast (especially leasing), this is from a very low base.

As noted by Cottarelli et al., banking sector reform in transition countries can be divided into three phases: (i) the recognition that a large share of the loans extended by public banks, mostly to state enterprises, have to be written off and the losses shifted to the government; (ii) the sale of banks, primarily to foreign investors; and (iii) the beginning of more standard banking operations, including increased lending to truly private enterprises. As Serbia has made only limited progress along this path over the past three years (see Chapters 2 and 4), the country’s banking system remains poorly equipped to deliver an adequate range of credit and other financial services to the enterprise sector and especially to its expanding SME component.

Today’s status quo is one is which many large, mostly state-owned banks are seriously fettered by their backlogs of non-performing loans. Other banks, including the new foreign banks, are expanding rapidly but find it easier to base this expansion on multi-national companies and on consumer and mortgage lending. The alternative of servicing enterprise sector is much less attractive in the present legal and institutional environment. The high reliance of the credit system on foreign currency loans presents considerable risks especially to prospective SME borrowers.

In this chapter we first provide a summary of the patterns of financing used by various parts of enterprise sector at the moment. Then we attempt to structure the problem and look selectively at the main dimensions of the task of enhancing the delivery (volumes and terms) of financial services to the growing private sector. Separate sections are provided on the possible catalytic role of donor-supported credit schemes, and on benefits and risks of state-sponsored institutions such as the Development Fund (DF). The Investment Climate Assessment (ICA) and Private Sector Note (PSN), currently also under preparation, examine the issues of access to and cost of finance from the real sector perspective in greater detail.

23 Cottarelli, Carlo, Giovanni Dell’Ariccia and Ivanna Vladkova-Hollar, “Early Birds, Late Risers, and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and the Balkans”, IMF Working Paper 03/213. 24 GDP at current prices in year 2002 was 701.4 billion Dinars, while the total amount of credit to enterprise sector (both short term and long term) was 133.4 billion Dinars (end of period data). The total is the sum of lines 2 (short term credit), 8 (short term securities), 17 (short term forex credits), 22 (long term loans) and 28 (long term forex loans). (Source: NBS Statistical Review).

57

1. Patterns of Enterprise Financing in Serbia

Different patterns of enterprise financing have been associated traditionally with the different forms of productive organizations characteristic for the Serbian economy. At the risk of some over-simplification it is useful to differentiate between: (i) the large socially-owned enterprises; and (ii) the more numerous and growing small and medium enterprises (this includes smaller socially-owned enterprises).

Large socially-owned enterprises

The bulk of the enterprise sector in Serbia has been organized traditionally as socially-owned enterprises.25 As the transition process got seriously underway in 2001, there were around 4,600 SOEs in Serbia, ranging from small firms with less than 50 employees to a number of large conglomerate organizations each having many subsidiaries. By official record, the SOE sector still employs more than half of the total labor force in the enterprise sector, although this estimate is certainly inflated taking into account the large informal sector.

To release the assets held by the SOEs for productive use, the authorities, with support from the Bank and other donors, embarked in 2001 on an ambitious, multi-track privatization program. This goal has been pursued by the Serbian Privatization Agency (PA) with reasonable success through two main methods of privatization, namely: (i) tenders of large enterprises where early privatization has been judged to be possible; and (ii) auction sales of small and medium SOEs (more details on progress to date can be found in the PSN). A third approach involves the large loss-making enterprises that need substantial restructuring prior to their privatization. Similar to experiences in other transition economies, this process is proceeding much more slowly, and it is here that the ongoing burden on banks is mostly felt.

As of April 2004, around 70 large problematic industrial conglomerates have been selected by the authorities to undergo organizational and/or financial restructuring as they cannot be sold in their current condition. The financing of these problem cases represents a major burden on banking sector resources – especially in the state-owned banks. Historically, bank credit, either in its direct form or in the form of state-guaranteed obligations channeled through the banks to enterprises, was the major source of funds to the SOE sector. But as its availability dwindled through the 1990s and early 2000s, it was progressively replaced by inter-enterprise arrears and direct subsidies as the primary financing mechanisms. Nonetheless an estimate made it 2002 suggested that around 50 percent of the liabilities of the large SOEs is owed to the commercial banks26. Much of this takes the form of accumulated interest and penalties built up during the 1990s. Almost the whole amount of the bank liabilities owed by large SOEs are overdue. They are also heavily concentrated in the four largest state banks that are now in receivership.

25 In addition to the SOEs, the state directly owns a number of large utility companies (electricity, post and telecommunications, water) and transportation companies (including the railways). 26 Source: World Bank, A Strategy for Restructuring the Enterprise and Financial sectors in the Republic of Serbia, June 2002.

58

Some initial indication of the scale of this problem for particular banks can be seen from the preliminary results of the study led by Prof. Mark Schaffer on the restructuring of enterprise debt. Based on the in-depth diagnostic reports for eight banks (seven majority state-owned + Komercijalna Banka) completed in mid-2003, the team came up with the following findings:

• Of some 70 large SOEs currently in the PA restructuring portfolio, 23 different enterprises are mentioned in the 8 bank diagnostic reports. Seven firms appear in more than one diagnostic report, i.e., are they are debtors of more than one bank. 16 enterprises are mentioned in the diagnostic reports as having Paris and London Club (PLC) loans.

• The 23 firms in restructuring account for 14 percent of the total loans of the 8 banks, and 37 percent of their PLC loans.

• 25 percent of the total provisioning in these 8 banks relates to PA firms in restructuring, and almost half of the total provisioning in these 8 banks relates to the PLC loans.

Consistent with the similar stages of transition experience elsewhere (notably in Ukraine and Russia), Serbia’s large SOEs have a pattern of financing dominated in terms of stocks (but not equally in terms of new flows) by bank credit27; by suppliers credits including large amounts of non-sustainable arrears to key public utilities and by heavy direct subsidies by the state. This pattern of financing is clearly problematic and equally clearly is connected intimately to the balance-sheet problems of the main state banks. Because the early privatization of most of these enterprises is out of the question, and bankruptcy is precluded for political reasons, the burdens and uncertainties involved in the present patterns of financing will continue for some time to come. This will constitute an ongoing (but hopefully declining) source of stress both for the budget and for some state banks that extend financial support to these SOEs.

The good news here seems to be that the 2001-02 interventions and especially the closure of the four largest banks have already removed a large part of the backlog of large bad enterprise loans from the active part of the banking system. Hence the remaining problem banks have less than a quarter of their portfolios afflicted by the remaining exposures to the large problematic enterprises. The bad news is that the part of bank balance sheets still afflicted in this way is still too large for effective operations.

Smaller SOEs and Other Small and Medium Companies

The greater part of SOEs, employing roughly 500,000 people (80 percent of total SOE employment) in effect belongs to Serbia’s SME sector.28 It is significant that fully 50 percent of these SOEs employ less than 50 employees each. In general, these other SOEs are in better financial condition than the large conglomerates in restructuring, with less than 27 SOEs and state enterprises still enjoy direct and indirect subsidies and access to bank credit. The bank credit element is complicated by the use of compensating budgetary deposits in some cases to match part of the new credits. 28 Employment of 50 persons or less is one of the main criteria for classifying an enterprise in Serbia as “small”. This is based on the Law on Accounting and Auditing of December 2002. A “medium” enterprise is one employing between 50 and 250 persons.

59

10 percent of the enterprises reported to be insolvent in 2002 (i.e., have a negative net equity position) in Serbian accounting terms.

As of end 2003, there were almost 240,000 registered SMEs of which around 67,000 are active (see Table 6.1). The bulk of all registered enterprises by number are SME enterprises in the private sector The total SME employment was around 670,000. However, there is reported to be a far larger number than this – up to 1 million persons - engaged in various gray economy activities many of which could be “legalized” given an appropriate reform of incentives.

Table 6.1 – Structure of Serbian Enterprise Sector

2000 2001 2002 2003

Num. Empl. Num. Empl. Num. Empl. Num. Empl.

Total no. of Enterprises 220305 227667 236630 241932 No. of registered SMEs 216780 223796 234027 239270 No. of active SMEs 61722 610619 63985 581193 66219 654768 67703 669442 No. of shops 176724 353448 188812 377624 195186 390372 207596 415192

Source: SME Agency and Republican Statistical Bureau Given the weaknesses of the large SOE enterprises and their ongoing shedding of labor as part of privatization and restructuring process, the SME sector is clearly a critical source of future employment opportunities. The government strategy for SME development is targeting an increase in the total numbers of SMEs to 400,000 by 2007, and the creation of over one million new jobs in the sector. However, today’s levels of credit and other financial services available to this large part of the enterprise sector are wholly inadequate to meet the needs of significant expansion.

According to the survey data, the majority of the Serbian SMEs are highly dependent on internal funds/retained earnings as their main source of financing. The PICS survey29 indicates that dependence on retained earnings is well over 80 percent (see Figure 6.1). By implication the enterprises do not make much use of external sources of finance, including bank credit. It is not unusual – even in the developed economies -- for retained earnings to show up as the major source of enterprise finance. What is unusual is the high absolute level of the ratio in Serbia: substantially higher than the corresponding figures based on the BEEPS survey in Croatia (52 percent), Hungary (63 percent), Bulgaria (66 percent), Poland (68 percent), Slovakia (69 percent) and Romania (70 percent).

The available evidence indicates that Serbian enterprises make only a very limited use of two main forms of external financing at the present time, namely: (i) bank credit; and, (ii) trade credit. The former accounts for 4-5 percent of the total working capital financing of enterprises and for about 6 percent of their investment capital financing. Trade credit accounts for a similar percentage of working capital finance – 4-5 percent - but for only around 1 percent of investment capital. These data say nothing about an important third source of finance for many enterprises namely arrears of payments although this is thought to be far less important for SMEs than for the large SOEs. The markets are not yet

29 Productivity and Investment Climate Survey (PICS) 2003

60

sufficiently developed to enable institutions like stock markets and bond markets to play a major role (see also Chapter 5).

Figure 6.1 – Sources of finance for private enterprises

Informal sources Banks and other financial institutions

83.5%

Retained earnings Equity

All other

84.9%

% share of financing working capital

% of financing new investments

Trade credit Source: PICS 2003

Trade credit is particularly wide-spread in Serbia, with between 43 percent (BEEPS 2) and 67 percent (PICS 2) of all enterprise making some use of such financing and between 21 percent (BEEPS) and 35 percent (PICS2) of all enterprise sales benefiting from this method of finance. Recent empirical evidence suggests that a relatively high reliance on trade credit occurs when a country’s legal system is inefficient30. Weak legal arrangements raise the costs, and lower the advantages to formal financial institutions such as banks. Trade creditors it is argued can steer around the legal problems more successfully for several reasons, namely:

• information acquisition - suppliers have a cost advantage over banks in acquisition of information about the financial health of the buyers and the risk of default;

• the credit negotiation/liquidation process is easier when a supplier is operating a regular commercial relationship with a borrower; and

• trade creditors may be able to punish debtors without resort to the legal system by withholding further deliveries.

International practice shows that in countries with higher creditor rights protection, firms borrow relatively more from banks then from their suppliers. This in turn suggests that the enforcement of trade credit contracts relies more on informal mechanisms, and that its relative importance in Serbia vis-a-vis bank credit will decline as some of the institutional/legal constraints are addressed.

2. Structuring the Problem and Proposed Solutions

It is well established from the experience of other countries in the region that the early transition years for banks are characterized not only by poor availability but also by very high costs of banking services to enterprises. This is due to several factors:

30 Demirguc-Kunt, A. and V. Maksimovic ”Firms as Financial Intermediaries: Evidence from Trade Credit Data” The World Bank Working Paper (2002);

61

• The high overhead costs of banks associated with various internal operating inefficiencies (these were quantified for Serbian banks in Chapter 3).

• A large carry over of non-performing loans that further raise the costs, and reduces the scope of providing new and hopefully more successful services

• Significant gaps in the institutional arrangements for banking that raise the costs that banks need to incur in order to offer loans and other services safely (examples are the weak arrangements from valuing and enforcing collateral, onerous tax arrangements, a weak customer base and poor technical skills in, for example, developing and assessing business plans of prospective clients).

• Significant incentives for enterprises – potential clients of banks – to operate in the parallel economy thereby reducing the population of “good” bankable clients and projects available to banks31.

These factors together have two significant effects that impact negatively on productive activity. First, they raise the effective costs of banking services to all clients of banks (see the vertical axis of Figure 6.2 below). Second, they reduce the volume of credits and other services that banks can provide (horizontal axis). The linkages between these two effects are shown schematically in the simple demand/supply diagram in Figure 6.2.

Figure 6.2 – Costs and Volumes in Banking

Interest rate

Quantity of Banking Credit and Other Services

Supply 2

Supply 1

Demand 1

Demand 2

Shift due to Higher Costs

Shift due to move to Shadow Economy

R1

Q1

R2

This brief assessment of the situation suggests that the task of encouraging the Serbian banks to provide a larger and more efficient range of banking services to the productive economy can be thought about in two main parts. It is convenient to refer to these two parts of the task as the “debt workout” and the “service innovation” components respectively.

First, there is the task of helping some banks – the state banks mainly – reverse out from under their heavy burden of deadweight loans to SOEs through some type of debt workout mechanism. This will not only lower banks’ costs but also clean up their balance-sheets as

31 It is estimated that Serbia’s gray economy may account for as much as one third of GDP and could employ as many as 1 million persons. This is far more than in the registered SME sector.

62

the necessary but not sufficient condition) for greater provision of credit and other financial services to new emerging productive sector. But this first task obviously needs to include the parallel restructuring/privatization of a substantial part of the enterprise client base. The present report does not dwell on the challenges of debt workout, as the issue is covered in great detail in parallel PSN.

Second, there is the task of service innovation, aimed at creating a more customer friendly and market-oriented culture in all surviving banks so as to extend the breadth and depth of coverage of banking services to newer private forms of productive activity. Service innovation can reduce many of the other costs associated with an inadequate legal framework and other aspects of a creditor unfriendly institutional environment. It is emphasized that “innovation” in this context refers not only to efforts internal to individual banks. In addition it embraces the general institutional reforms and provisions of new public goods of collateral, contract enforcement etc. that can serve to lower the costs of banking delivery. Both can help to shift the supply curve in Figure 6.2 downwards and to the right.

3. Issues related to Service Innovation

As was indicated earlier, the innovation required relates to both: (i) changes in the internal operational efficiencies of the banks themselves and also in the way in which they regard private sector lending; and (ii) the provision of new institutional arrangements (public goods) that can facilitate lower cost banking and make the banks more willing to offer improved services to private sector clients. We consider these two aspects of the problem separately in what follows.

A. Changes internal to the banks

(i) Costs and Charges. The key objective here is to enhance competition between banks and achieve a gradual reduction of their overhead costs and interest spreads to levels more closely convergent on good international practice, and affordable by more clients. This will require the structural changes in the sector that were examined in greater detail in Chapter 4. These include the closure of the weakest banks that are likely to remain high-cost because of their weak balance sheets; the fast privatization of the viable state banks to bring in new strategic investors willing to compete more actively with the established private banks; and greater competition for the better banks in the system that presently find it a bit too easy to make profits in spite of relatively high costs.

Chapter 3 analysis suggests that high overhead costs of banks are associated with various internal operating inefficiencies. These inefficiencies together with the lack of funds, low creditor rights, and risks have as a consequence high interest rates32. The average real interest rate is in the 6-12 percent range. It should be noted also that virtually all local Dinar 32 Corporate lending presently seems highly concentrated in six banks (three national and three foreign banks) that together account for over 50 percent of the aggregate amount of corporate credits. The reasons for this need to be considered in conjunction with the more general assessment of bank competitiveness as defined in Chapter 3. However, the differentials in amounts charged as between the six leading banks and other banks do suggest that some competition exists and that interest rates are likely to be driven further downwards in part at least because of intensifying competition.

63

denominated loans in Serbia include a foreign exchange clause, setting repayment rates in foreign currency to be converted to a local currency on the day of the repayment. Although interest margin and real interest rate are decreasing (see the Figure 6.3 below), Serbia still has significantly higher interest rates compared to neighboring countries. Equally, the foreign exchange clause is transferring large amounts of risk to enterprise borrowers who in many cases will have no means to hedge such risks.

(ii) Bank Procedures. Most lending to enterprises and especially to SMEs in Serbia is backed by collateral.33 The PICS 2003 survey showed that in 81 percent of cases creditors required collateral or deposits for the credits, compared to 50 percent in 2001. However, the amount of collateral requested by banks varies widely, depending on the risk assessment of the client. It sometimes exceeds 150 percent of the loan principal. More banks need to gain the ability to conduct good quality credit risk analysis and generally to move the criteria for extending credits away from collateral-based arrangements and towards an objective assessment of the quality of different potential borrowers. This will not eliminate the need for reform of Serbia’s system of registering and enforcing collateral (see the discussion below) but it will take some pressure off of that system.

Figure 6.3 – Interest rate dynamics

Interest Margin

0

5

10

15

20

25

30

Dec-01 Jun-02 Dec-02 Jun-03 Dec-03

Interest rate - Short Term Credits - foreign currency based

0.00

5.00

10.00

15.00

Serbia Croatia Bulgaria

Source: NBS Source: Central banks (end-2003 data) (iii) Availability of Funds. Survey results indicate that the banking sector has a cautious attitude towards expanding loans for the longer maturities that most private businesses need at some stage. Short-term (up to 90 days) foreign-currency deposits still dominate in Serbia. Long-term credit is exceedingly scarce even for larger enterprises. This is partly because the banks themselves lack access to long-term sources of funds. Gradual lowering of currently very high mandatory deposit requirements for commercial banks by the NBS could increase their credit potential. Ultimately, further financial deepening based on stronger lower-cost banks with more capital seems to be a crucial pre-condition for improving this situation. 33 Banks also apply a variety of preventive measures in order to secure their credit portfolios. In some cases banks add a small insurance fee against potential losses, which provides sufficient protection for the bank on a portfolio of hundreds of SMEs.

64

(iv) Bank Attitudes. Regional experience suggests that changes in bank culture and attitudes play no small role in increasing the lending to private sector. Existing local banks primarily focus on servicing larger, often state- or socially-owned, enterprises. Most show a lack of interest in SMEs and have little tradition in this area. At the same time, experience from other countries shows that the more competitive and lower cost banks will be more successful more quickly in establishing SME business areas. This is because, although SME lending can provide higher margins than many other types of business, it is also more costly in several respects. The banks best able to absorb these costs are those investing in methodology and training to achieve economies of scale. This in turn requires the long-term strategies and resources that are more likely to be found in the most competitive banks. So the conclusion here is that the enhanced competition that reforms discussed elsewhere in this report should achieve, are also a part of the solution to achieving greater bank involvement with SME lending.

It would normally be expected that new foreign banks can bring a different and fresher attitude to the problem of signing on new forms of private clients and offering a broader range of banking service. However, thus far this demonstration effect seems to have been relatively weak in Serbia. It is true that the new foreign banks find it easier to mobilise local deposits because of their reputational advantages. However, they have naturally focused their attentions on their own traditional business with majority foreign-owned businesses and international trade finance. In so far as they have engaged with local Serbian clients, their activities have been relatively narrowly concentrated on consumer lending, leasing and mortgages. This is not surprising as a part of an initial business strategy. It could lead to a broader product focus once the banks begin to face more serious competition in a few years’ time.

B. Issues External to the Banks

(i) Collateral and Related Legal Constraints. Despite significant improvements since 2001, supported by the Bank and other donors, the legal and institutional framework for enterprise lending still has significant gaps. These gaps result in higher costs of banking and deter new lending in ways that the banks themslevs can do little to influence.

The central collateral registry together with the credit registry has been repeatedly quoted by bankers as a critical condition for improving access to finance. Although the Law on Secured Transactions was adopted in mid-2003, a centralized collateral registry has not yet been established. There is in addition the practical point concerning the true value and usefulness of the collateral that the banks hold. The market demand for some of the collateral goods held by banks is questionable. This explains why in some cases banks use two or three items of collaterals for one single credit. Finally, due to the weaknesses of capacity of the judiciary, banks encounter numerous legal and practical difficulties of taking possession of collateral goods in case of default.

Mortgages on buildings represents the most frequently used type of collateral (45 percent of all collaterals). Even though there is a legal framework for granting a loan that is secured by a mortgage, the practice in the market is quite different. This is because a large part of Serbia’s real estate is not appropriately registered in public books. Even in those cases

65

where real estate is registered properly discrepancies between the real facts and the data contained in the public register are still common. These institutional and legal weaknesses need urgent reform if a mortgage-based system of lending is going to have a real chance to develop.

(ii) The Shadow Economy. The volume of potential corporate banking business is held back by the operations of many SMEs in the shadow economy. The tax evasion that is one of the principal motivations for black-economy activity also constitutes one of the major sources of financing for many potential banking clients. As a result, both the budget and the banking industry lose out because of the perpetuation of a large parallel economy. This report makes no attempt to review the agenda of reform needed to make more SMEs compliant with registration and other formal requirments of “legal” operation. Suffice it to say that such an agenda is an important part of the program for enhancing the attractiveness to banks of private sector activities in general and SME lending in particular.

(iii) Stimulus from External Credit Lines. Experience from other ECA countries shows that external credit lines can be a powerful catalyst to encourage local banks to move more quickly into SME activities. IFIs have a significant role to play here, both through providing finance through participating banks, as well as through technical assistance to improve the banks’ lending skills. The EBRD has already invested both in a micro-finance bank and in credit lines for SMEs channelled through local financial intermediaries. EIB, EAR and KfW have also established special facilities to promote access to finance for SMEs (see Table 6.2). Existing credit lines already seem to have a positive influence on lowering the cost of financing, with the interest rates (including the currency clause element) ranging from 8 percent in the case of IFI-sponsored facilities to 12 percent or more in the case of credits from banks’ own resources.

Table 6.2 – SME Finance Facilities and Credit Lines in Serbia Bank Source of

Lend. Limit / amount Validity

(Tenor) Grace period

Interest rate p.a.

Collateral brief details

IFIs Credit Lines NBS FMU Eksim banka Novosadska banka Zepter banka Čačanska banka Komercijalna banka

EAR, own resources €20.000-200.000 up to 5 yrs up to 12 mo.

LIEUR3MD+6%

(4% Banks + 2% NBS FMU)

8.07% Va 27.02.2004

mortgage on premises - pledge line (equipments)

NBS FMU Eksim banka Komercijalna banka

EIB €200.000 - up to 50%

of the final beneficiary’s project cost

up to 12 yrs up to 3 yrs 7.88% mortgage on

premises - pledge line (equip.)

Banks with special programme - funds for financing SMEs

Zepter banka KfW € 50,000 up to 3 yrs up to 6 mo. 10% - 11,5% mortgage on premises

Komercijalna banka KfW € 50,000 up to 3 yrs up to 6 mo. 9.5% - 12% mortgage on premises

Eksimbanka KfW € 50,000 up to 5 yrs up to 12 mo. 10% - 12% mortgage on premises

KfW,own resources up to 4 yrs 10% - 12%

Kulska banka ING BANK

Neth.

€ 1.000 - € 50.000 up to 5 yrs

up to 12 mo. 12%

mortgage on premises - surety

€ 1.000 - € 10.000 up to 2 yrs up to 1 mo. 24% € 10.000 - € 50.000 up to 3 yrs up to 6 mo. 16,80% - 21% Procredit bank

KfW, IFC, EBRD, own resources € 50.000 - € 500.000 up to 5 yrs up to 6 mo. 9,6% -18%

mortgage on premises - pledge line (equipments)

Government Programs

66

Development Fund Budget up to 40% of project up to 5 yrs up to 12 mo.

1-5% (plus fee paid to com. bank for the

issuance of the bill of exchange)

bill of exchange guaranteed by the

bank

However, the scope of the donor involvement is still low compared to the regional levels. Data on the number of requests for financing and approved credits shed light on the strong unmet demand for credit. Credit line managers state that most of the requests they have received are compromised by the lack of transparency of underlying financial statements, inadequate security (collateral), incomplete business plans or unacceptable debt to equity ratios. On the other side, lengthy delays in disbursement are caused by the problems related to establishing acceptable mortgages on premises to be used for collateral.

4. Role of Government-Sponsored Programs

In the embryonic situation we see today, Government-sponsored programs such as the Development Fund and Guarantee Fund are potentially useful instruments to leverage private finance for private enterprises. They can also be helpful in supporting start-up companies in the difficult early phase where financing problems are likely to be most acute. Centralized schemes, if properly implemented, can provide timely bridge financing on accessible terms through loans, guarantees, and venture capital.

Box 6.1 – Serbian Development Fund

Since it was established in 1992 as the successor of the previous similar state fund, the DF’s main role has been that of distributing state budgetary resources to support the needs of the state and socially-owned enterprises. Numerous debt-to-equity swaps with enterprises in arrears resulted in a significant share of the DF in ownership of some of the companies. In 2003 alone, the DF issued US$160 million in loans to more than 1,200 enterprises which represented 0.75 percent of GDP. In the same year, the DF received direct budgetary support of US$137 million. In the post-Milosevic era, the authorities tried to reconfigure the role of the DF by targeting some of its activities to support SME development in Serbia. Yet the major part of the DF’s loans still appears to be used to prop up large, loss-making SOEs. According to the official plan for 2004, the DF will grant loans totaling Dinar 11.2 billion, out of which Dinar 6.5 billion will be used for “reviving production” (read – subsidies), and 3.7 billion Dinars will be invested in the development of SMEs. The activities of DF raise significant concerns with regards to the governance of this state institution, the transparency of its operations and the principles that should be driving the distribution of state resources. The experience of other countries in the region, as well as in Latin America and Africa suggests that performance of such state institutions without proper oversight from the Government may result in a significant budget drain and a chronic deterioration of the financial discipline of the recipients of the subsidies. Consequently, the purpose, scope, and mechanisms of DF operations should be revisited by the Serbian authorities as a matter of priority. The use of funds needs careful monitoring as do their targeting to the most needy segments of the economy, and their timely repayment. The continued existence of this institution in its present form may have a negative impact not only on the future fiscal position of the state but also inhibit the establishment of a market-based, transparent and competitive financial system. However, as illustrated by the story of the Development Fund (see Box 6.1) there are important preconditions for this potential to be realized. Any public support schemes must be carefully designed so as not to distort markets through displacing other forms of financing. Leverage is most likely to be achieved if any government intervention into SME financing can be channeled through commercial banks, to ensure maximum compliance

67

68

with market mechanisms. As spontaneous bank financing begins to expand (because of the package of reforms already discussed) the need for the pump-priming of the state should be allowed to decline. Finally, as shown by ample examples in the transition region and worldwide, special government funds too often become an instrument for exerting political influence over credit flows. Therefore, the authorities need to establish and maintain rigorous mechanisms for controlling the funds’ activities.

Serbia – Financial Sector Notesiteresources.worldbank.org/INTSERBIA/Resources/SerbiaFSN.pdf · The team is grateful for the kind assistance of the Serbian authorities in the preparation

Documents