This Selected Economic Issues paper for Bosnia and Herzegovina reports that output, exports, and incomes have increased and inflation has stabilized. New modern banking laws have been passed in both entities, and the banking sector has been almost completely privatized, with the majority of assets now under foreign ownership. The reforms to the central bank and to the banking system have been aimed to secure stability and to build an efficient financial system.


This Selected Economic Issues paper for Bosnia and Herzegovina reports that output, exports, and incomes have increased and inflation has stabilized. New modern banking laws have been passed in both entities, and the banking sector has been almost completely privatized, with the majority of assets now under foreign ownership. The reforms to the central bank and to the banking system have been aimed to secure stability and to build an efficient financial system.

II. Market Indicators of Confidence in the km and the Banking System Since 20001

A. Introduction

1.This chapter examines monetary and interest rate data to provide a perspective on the robustness of financial confidence in Bosnia and Herzegovina (BiH).

2.The currency board is almost 8 years old. International reserves stand at 6 months of merchandise imports. And after the 1998–2002 restructuring, banks are booming. These signal confidence in the KM and in the banking system and are major achievements.

3.But confidence can shift quickly. International experience shows a number of countries which enjoyed periods of confidence subsequently witnessed turbulence.2

B. Institutional Background

4. Following the 1992-95 civil war there were a number of currencies circulating in Bosnia and Herzegovina, including the Bosnian dinar, Yugoslav dinar, Croatian Kuna, and the Deutsche Mark (DM). The currency board was set up in 1997 and a new currency, the convertible mark (KM), was introduced, pegged at KM1= 1 Deutsche Mark. It was subsequently pegged at KM 1.9558=Euro 1, upon the abolition of the DM.

5. Since then the KM has largely displaced the other currencies (though the Euro also circulates widely), and the currency board has delivered Euro area levels of inflation. Gross international reserves and net free reserves (net reserves minus reserve money) have risen steadily to Euro 1,768 million and Euro 88 million, respectively, at end-2004. And confidence in the KM was strengthened by the smooth transfer to a new board for the Central Bank of Bosnia and Herzegovina (CBBH) in August 2003, and an uneventful transition from an expatriate to a Bosnian governor at end-2004.

6. Meanwhile, the banking system was weak by war’s end in 1995. Most bank assets were state owned, over 90 percent of loans were nonperforming, and the confidence of depositors was very low.3 Moreover, most of banks’ foreign currency deposits had been seized by the National Bank of Yugoslavia prior to the civil war, thus forcing banks to freeze the foreign currency deposits of their clients. The war had also disrupted links to international financial markets, caused very high inflation, and extensively damaged the physical infrastructure of the country.

7. Since then there has been much progress. New modern banking laws were passed in both entities, and the banking sector has been almost completely privatized, with the majority of assets now under foreign ownership. In the process considerable consolidation has occurred—the number of banks has declined from 72 (55 in the Federation and 17 in the Republika Srpska (RS)) in 1998 to 33 (24 in the Federation and 9 in the RS) at end-2004.4 Of the banks existing at end-2004, five were under provisional administration—four in the Federation and one in the RS.

8. Banking supervision has been entrusted to two entity banking agencies— Federation Banking Agency and the Banking Agency of Republika Srpska, with a coordinating role for the central bank. With some technical assistance as well as Fund program conditionality, supervisory quality has improved significantly in the past two years. Moreover, plans are underway to merge the two banking agencies into the CBBH in order to strengthen the independence of bank supervision. And a deposit insurance scheme has been successfully established—by end-December 2004, 21 banks were members.

9. These reforms—to the central bank and to the banking system—were aimed to secure stability and to build an efficient financial system. And both sets of reforms reinforce each other—a strong monetary arrangement builds confidence in banks, and vice versa. Has that confidence been secured?

C. Confidence in the KM

10. Recent data suggests that confidence by Bosnian citizens in the KM has improved since 2000.

  • Following the introduction of Euro notes and coins, KM currency in circulation doubled between October 2001- December 2001 (Figure 1). This was the result of reverse currency substitution, as people converted their holdings of DM and other Euro area currencies into KM notes and coin in a show of confidence in the KM.

  • Against this, how should the small decline in currency in circulation since then be interpreted? This appears to reflect increasing use of the banking system as a means of settlement and as a store of value; use of cash for these purposes has steadily declined. This interpretation is underscored by the rapid growth in bank deposits since 2001, and the associated increase in the money multiplier (Figure 1). This would suggest that financial deepening, rather than a reversal of confidence in the KM, accounts for trends in currency in circulation after the end-2001 surge. As is also clear from Figure 1, KM deposits in the banking sector have increased at a faster pace than foreign currency deposits until late 2004. This provides further evidence that increasing public confidence in the banking system, rather than a reversal of confidence in the KM, accounts for the trends in currency in circulation since early 2002. The recent pickup in foreign currency deposits appears to be the result of banks’ efforts to meet tighter forex exposure regulations, rather than a reduction in confidence in the KM (¶13).

  • On the credit side, households and companies have also been willing to accept banks demands that the bulk (over 50 percent) of their KM denominated borrowing should be indexed to the Euro, and have not demanded a lower interest rate to compensate them for the currency risk they take on as a result of this indexation. On one interpretation, this suggests that they have strong confidence in the KM and the currency board. However, it is also likely that borrowers are unaware of these technical indexation clauses in their credit contracts. If so, the high degree of indexation signals consumer non-sophistication, rather than confidence in the KM. The truth may reflect some combination of both interpretations.

  • There is also recent anecdotal evidence that some merchants are beginning to be less willing to accept Euro notes and coin for small transactions. This is in contrast with previous years when, for some time and in some places, the KM was not accepted for these purposes, and more recently, when Euro and KM notes and coins were freely interchangeable for all kinds of transactions.

Figure 1.
Figure 1.
Figure 1.

Bosnia and Herzegovina: Currency in Circulation and Commercial Bank Deposits

(in millions of KM)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Source: Central Bank of Bosnia and Herzegovina, and staff calculations.

11. Confidence in the KM is also signaled in the spread between Euro and KM interest rates (Figure 2). Because there is no money market rate for Bosnia and Herzegovina, we proxy this spread using the mid-points of bank demand deposit and short term lending rates in BiH and the Euro area.5 Thus the results should be interpreted with caution, since the spread could reflect a variety of factors influencing the Euro area and Bosnian banking systems, rather than a pure risk premium between the two currencies. Nonetheless, the trend decline in the spread observed in the data since January 2003 is suggestive of increased confidence. This spread is still around 2 percentage points, but it has fallen by ¾ percentage point since January 2003. And—the same issue measured a different way—the spread between Bosnian central rates and 3-month EURIBOR also declined through 2004 and early 2005, after being stable since mid-2002.

Figure 2.
Figure 2.

Spreads between Bosnian and Euro Area Interest Rates

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

12. However, domestic banks do not appear to be as persuaded as the Bosnian public about the KM.

  • Banks have often required that their KM loans are indexed to foreign currencies, typically the Euro (Figure 3). The share of indexed loans is typically higher the longer the maturity of the loan. This behavior by banks suggests some caution on their part about exchange rate risk. Moreover, this behavior has remained firm in the face of several regulatory and other steps taken which have implicitly raised its costs to the banks, suggesting that it is not just an administrative habit, but is a strongly held view (¶13).

  • But, interestingly, banks’ apparent doubts about the KM are not evidently symptomatic of broader doubts on their part about the prospects for the BiH economy. Rather, the strength and persistence of the credit boom since 2002 suggests a significant degree of confidence in broader economic prospects (Figure 4). Banks have responded to the surge in deposits after 2001 by increasing credit, as a result of which credit to the private sector has increased by over 20 percentage points of GDP in the past three years, amidst keen competition by banks for market share. The key qualification to this interpretation is the degree to which bank behavior is dominated by their desire to establish market share, rather than concerns they may have about broader economic prospects.

Figure 3.

Bosnia and Herzegovina. Indexation of Bank Lending

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Sources: Data from the Bosnia and Herzegovina authorities and staff calculations.1/ Adjusted to correct for the structural break due to the takeover by the RS government, in June 2004, of KM 463 million of old bank claims on RS enterprises.
Figure 4.
Figure 4.

Bosnia and Herzegovina: Growth of Credit to the Private Sector 1/

(in percent)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Sources: Central Bank of Bosnia and Herzegovina, and staff estimates.1/ Data on credit to enterprises and credit to the private sector were adjusted to correct for a structural break due to the takeover by the RS government, in June 2004, of KM 463 million of old bank claims on RS enterprises.

13. Importantly, however, the increase in international reserves of the CBBH and reserve money from mid-2003 is not evidence of strengthening confidence in the KM. Instead, it almost entirely reflects banks’ adjustments to tighter prudential regulations. In the context of a currency board and free capital mobility, traditional monetary policy instruments, such as open market operations or changes in policy interest rates, have limited effectiveness. Thus, the authorities have sometimes used supervisory measures to secure prudential and broader monetary goals. In particular, in response to concerns about the credit boom—and the fact that many banks had run foreign exchange exposure positions outside international norms—the authorities took appropriate measures from June 2003 to tighten the regulatory framework:

  • Banks were required to phase in progressively tighter limits on their forex exposure positions, from about 120 percent of capital in mid 2003 to 30 percent of capital, by mid-2004. And from mid-2003, the forex exposure regulations were appropriately amended to include KM assets and liabilities indexed to foreign currencies, in line with international practice.

  • Banks were also required to take steps to come into full compliance with the prudential regulations on liquidity by June 2004. These require that liabilities of maturity up to 90 days should be matched 100 percent by assets of similar maturity. Since previous forbearance in the application of these rules had led to substantial nonobservance of them, most banks were given up to June 2004 to secure compliance, with some allowed to comply by September 2004.

14. These two reforms required banks to increase their deposits at the central bank. Banks steadily increased their borrowing from abroad (largely from their parents), converting the borrowed funds into short term KM deposits held with the CBBH. The currency conversion in this operation helped them to meet the steadily tightening forex exposure rules. And the short term assets created in this operation—the deposits with the CBBH—helped them to meet the liquidity rules. As a result, banks’ reserves at the central bank increased steadily, month-end by month-end, as the forex and liquidity regulations were steadily phased in. And, pari pasu, their NFA declined (Figure 5, lower LHS panel).

Figure 5.
Figure 5.

BiH: Bank Responses to Regulatory Tightening

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Sources: Bosnia and Herzegovina authorities, and Fund staff estimates1/ Data prior to June 2004 is adjusted for classification change. In June 2004, the RS government took over KM 463 million of old bank claims on RS enterprises, and associated foreign liabilities, reducing bank claims on enterprises and bank foreign liabilities accordingly.

15. The increase in bank reserves with the central bank led to corresponding increases in reserve money and international reserves. It was also reflected in the composition of reserve money, with commercial bank reserves rising relative to currency in circulation and cash in vaults (Figure 5, top LHS panel, and Figure 6). And given the currency board structure, which tightly links reserve money to international reserves, these movements in reserve money were matched by increases in international reserves at each month-end from mid-2003 to mid-2004 while the forex and liquidity regulations were phased in. Thereafter, the pace of growth of reserve money began to decline.

Figure 6.
Figure 6.

Gross International Reserves and Reserve Money

(in millions of KM)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Sources: Data from the CBBH and staff calculations.

16. Three further aspects of monetary developments confirm this interpretation of the regulatory—rather than a KM confidence—motive for the increased commercial bank reserves with the CBBH, and therefore in international reserves, from mid-2003.

  • First, in early 2004 the commercial banks began to exploit the fact that the forex and liquidity regulations being phased in from mid-2003 applied only on a month-end basis, not daily. So banks only observed them at end month—borrowing in foreign currencies from external parents at month end, placing these with the CBBH, and reversing these transactions a few days later. This caused substantial intra-month swings in bank and international reserves from January 2004 (Figure 5, top RHS panel), suggesting that in the absence of the new regulations banks would have preferred a significantly lower level of reserves with the CBBH. These swings increased in size, month by month, as banks borrowed more from their parents in order to meet the progressively tighter limits. The within-month swings reached as much as KM 400 million from peak to trough by mid-2004. But their amplitude declined significantly from July 2004, after banks were informed that the banking agencies would be requiring daily reporting of forex exposure positions. The swings have been virtually eliminated since daily forex exposure reporting was implemented in October 2004.

  • Second, this within month behavior suggests that the steady increase in bank reserves with the CBBH at each month end did not reflect the attractiveness of the remuneration offered by the CBBH on those deposits. From 1998 through mid-2003, required reserves were remunerated at market rates, while excess reserves received no remuneration. However, starting in June 2003 all bank reserves with the CBBH (not just required reserves) were remunerated at market rates—the rate earned by the CBBH on its overnight deposits of foreign exchange reserves. This, however, continued only up to December 2003. After that, required reserves continued to be remunerated at market rates, while remuneration of excess reserves was cut from the market rate (2 percent at the time) to a fixed one percent (which happened to be the European Central Bank deposit rate at the time). Despite the cut in remuneration, excess (and total) reserves continued to grow strongly through most of 2004, from month-end to month-end (Figure 5, top LHS panel). But the within month swings in reserves began precisely at this point. This appears to suggest that the reduction in remuneration of excess reserves induced banks to seek a better return on these assets elsewhere, within each month, than was available from the CBBH.

  • Finally, although the reserve requirements were raised, this did not cause the rise in commercial bank reserves with the CBBH either. The base for required reserves was widened in June 2003 to include foreign currency denominated liabilities—which more than doubled the base—and the rate of required reserves was reduced from 10 percent to 5 percent, the net effect of which was to raise required reserves modestly. But the stock of required reserves remained well below the actual level of reserves throughout (Figure 5, top LHS panel).

17. This combined evidence suggests that confidence in the KM has increased significantly since 2000, but that the rise in international reserves after mid-2003 reflected the regulatory reforms, rather than mandated reserve requirement levels or further KM-confidence factors.

D. Confidence in the Banking System

18. Recent trends indicate that confidence by the Bosnian public in the banking system has also increased since 2000.

  • There are clear signs of financial deepening (Figure 7). The share of KM currency in circulation in M1 and broad money has declined as Bosnians use bank deposits for an increasing share of their transactions. However the money multiplier remains on the low side (around 2) and a cross country comparison suggests that there is still room for more financial deepening (Table 1).

  • Correspondingly, the velocity of bank deposits declined significantly in December 2001 and continued to drop thereafter, albeit more gradually, as confidence in the financial system has led to continued strong growth in deposits alongside a steady decline in currency in circulation (Figure 7).6

  • And time and savings deposits have grown faster than demand deposits since December 2001 (Figure 8). However, a pickup in growth of time and savings deposits in late 2004 may in part reflect banks efforts to meet prudential regulations on maturity matching—since they need to lengthen the maturity profile of their liabilities in order to provide longer term loans (¶13, ¶14). Enforcement of these regulations appears to have caused an increase in spreads of time and savings deposit rates over demand deposit rates in 2004. But depositor’s willingness—even if so encouraged— to extend the maturity of their deposits signals confidence in the banking system, even if at a price.

  • The financial deepening has been boosted by increased bank efficiency generated by intense competition for market share, which has led to a substantial decline in spreads between deposit and lending interest rates. Spreads have declined by over two percentage points since January 2002 (Figure 9).

  • It is also noteworthy that despite the substantial changes and consolidation there have been no bank runs. Bank closures have generally occurred as a result of banks being put under provisional administration by the banking agencies, rather than by runs. It is not entirely clear why this is so. One explanation could be that depositors were generally well aware of the weak banks, and avoided them completely. Alternatively, or in addition, the Bosnian public may have believed that there was an implicit government guarantee on deposits, particularly if the banks were state owned. Formal deposit insurance is not likely to have been the cause of the lack of bank runs, because coverage of banks began only in 2001, and has been limited to sound banks, none of which have had any difficulty meeting their obligations to their depositors.

Figure 7.

Bosnia and Herzegovina: Indicators of Financial Deepening

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Source: Central Bank of Bosnia and Herzegovina, and staff calculations.
Table 1.

Cross-Country Indicators of Financial Depth, 2003 1/

article image
Sources: Data from country authorities and IMF staff calculations

Broad money used is M3, except for Czech Republic and Serbia and Montenegro which are M2.

Figure 8.

Bosnia and Herzegovina: Demand and Time and Savings Deposits and their Interest Rates

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Source: Data provided by the Bosnia and Herzegovina authorities and Fund staff calculations.
Figure 9.
Figure 9.

Bosnia and Herzegovina: Spread Between Deposit and Lending Rates

(in Percent)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A002

Sources: Data from the CBBH and staff calculations

E. Prospects

19. Looking forward, there are a number of signals to monitor which could indicate further increases in the confidence in the KM, but these will need to be interpreted with care.

  • If citizens continue to shift out of Euro currency in circulation into KM currency in circulation as a means of settlement and store of value, KM currency will rise and this will be reflected in increased international reserves of the CBBH.

  • If banks reduce indexed lending of their own volition, this would signal a welcome change in their perception of exchange rate risk. It would be reflected in less foreign currency external borrowing and less foreign currency deposit-taking. Note, however, that this aspect of increased confidence in the KM could lower CBBH international reserves. In this favorable context, banks’ need for KM deposits with the CBBH to meet prudential regulations would drop, lowering international reserves. The decline in the pace of growth of CBBH reserves in early 2005 could reflect this, though confirmation of this interpretation of these most recent trends will have to await collection of more data on loan indexation in 2005.

  • If an interbank market develops, this will provide an alternative short-term KM asset for banks, and reduce reserves with the CBBH. However, given that banks already have an excess of short term assets, the market need for interbank transactions is low so one may be unlikely to develop soon.

  • With the European economy continuing to show weakness and the current strength of the Euro against the US dollar, there is a possibility that ECB rates will decline, which will make the rate of remuneration of excess reserves with the CBBH (one percent) more attractive, thus encouraging banks to increase their reserves. This would not reflect a “confidence” effect, but an interest rate differential effect.

20. In the case of banks, the main indicators of confidence will be the money multiplier, velocity, and the maturity of deposits. These have been trending in the right direction, but all have further to go by international standards. The main risks are indirect— through the malaise of the corporate sector. With over 60 percent of state-owned companies lossmaking, and requiring deep restructuring and many others only marginally profitable, banks run the risk of being caught up indirectly in these corporate difficulties.7

21. Moreover, the framework for banks to monitor borrower quality and recover nonperforming loans needs improvement. There have been efforts to set up a collateral and pledge registry, but these have not yet borne fruit. In this situation, banks have improvised by requiring borrowers (for consumer loans) to have personal guarantors who would be liable in case of default. However, many guarantors have guaranteed loans for multiple borrowers, and there is no satisfactory way for banks to track this, thus raising the risk that guarantors cannot meet these multiple contingent liabilities. And the court system is not yet in a position to swiftly adjudicate bank suits against recalcitrant debtors.

F. Concluding Remarks

22. While confidence in the KM is generally much strengthened since 2000 further efforts should be made, notably because banks’ skepticism remains. Such efforts encompass a wide array of policy areas, and include the early implementation of structural reforms to promote private sector growth and competitiveness, particularly corporate restructuring. But as indicated, a drop in international reserves will not necessarily signal a loss of confidence—indeed, if it reflects changes in bank behavior, it could reflect increasing confidence in the KM.

23. Banking system restructuring has progressed at a much faster pace than corporate reforms. As a result, a relatively modern and dynamic banking sector coexists with a weak corporate sector, and it is not clear that the health of the banks can be maintained indefinitely alongside the malaise in the corporate sector. To sustain and strengthen further depositor confidence in banks, continued improvements in the quality of bank supervision, the regulatory environment, and the analytical capacity of the CBBH, alongside a determined effort to strengthen companies, will all be required. Finally, supervisors and the central bank will have to keep a watchful eye on the credit boom and stand ready to act decisively to maintain financial system stability.


Prepared by Daniel Kanda.


See Chapter 3.


See IMF Country Report No. 98/69, September 4, 1998.


See World Bank, Bosnia and Herzegovina Banking Sector Review, June 2004.


Central Euro area interest rate is calculated as the average of the Monetary Financial Institution (MFI) rate on overnight household deposits and the MFI rate on household consumption loans (with floating rates or with rates fixed for one year or less). The central Bosnian interest rate is calculated as the average of BiH demand deposit rate and BiH short term lending rate.


The true velocity is likely higher than presented in Figure 7, since official estimates of nominal GDP exclude the large non-observed economy (see chapter 5 and IMF Country Report No. 04/54, March 9, 2004). However, the gradual decline in reported velocity suggests that the official estimate of the rate of growth of GDP may not be too far off, implying that the grey economy may be growing at a similar rate as the recorded economy.


See Chapter 4.

Bosnia and Herzegovina: Selected Economic Issues
Author: International Monetary Fund