Bosnia and Herzegovina: Selected Economic Issues

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.

VIII. State Building1

A. Introduction

1. The State government is expanding. In addition to hiring new staff across the board, in 2004 key functions shifted from lower levels of government to the State level. Accession to NATO’s partnership for peace and eventually to the EU will require significant further steps in this direction.

2. The IMF supports these efforts. But we have concerns with the nature of the current approach to implementing State-building, notably its implications for fiscal and macroeconomic stability. Accordingly, this note focuses on how to reconcile State-building with macroeconomic objectives—preventing increases in the size of consolidated government and of the consolidated fiscal deficit. A strategy is proposed to accomplish these tasks.

B. Background

3. The diminutive State derives from Dayton. Reflecting the sensitivities which shaped the peace compromise, the Dayton Peace Agreement (DPA) originally confined the State mandate to foreign relations and selected domestic policies. Any mandate not explicitly attributed to the State by the DPA can only be assumed by the State if it is voluntarily relinquished by both Entities.

Text Table 1.

State Government Responsibilities in the DPA

article image

4. The DPA left the State without an independent source of funding. Until 2005, cash transfers were made monthly from the Entities—and at their discretion—to the State to fund its activities. The determination of the annual transfers to the State was entirely ad hoc and the agreed payments were often settled in arrears.

5. Nevertheless, the State has grown. Several functions were shifted to the State, recently, including the Indirect Tax Authority (ITA), the State Border Service (SBS), the Secret Service, and some defense functions. The number of State agencies increased from 17 institutions in 2000 to 40 four years later. In 2004, the administrative expenses of the State were 1.7 percent of GDP, up from less than one percent in 2000 (Text Figure 1).

Text Figure 1.
Text Figure 1.

The Evolution of the State

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A008

6. This largely reflects the aims of the international community. It seeks a single counterpart for negotiation with the outside world, an eventual replacement for the Office of the High Representative (OHR) as a coordinator of domestic and foreign policies, a counter to the centrifugal forces in Bosnia and Herzegovina (BiH), and a government structure which bears a closer resemblance to those in the rest of the world. These aims have their echoes in domestic political debates over State-building. But these debates—pro and con—more profoundly reflect the persistence of the sensitivities which underlay the Dayton agreement in the first place.

7. But the State is still small. Excluding the ITA, which is an unusual State cum Entity hybrid, its share in general government expenditures was less than 5 percent in 2004. At 1.7 percent of GDP, its spending in 2004 was well below that in each entity central government (10 percent of GDP in the Federation and 10½ percent in the Republika Sprska, RS, including municipalities). But the State accounted for about 7 percent of total government employment (excluding Cantons and Extra-budgetary funds) in 2004 (Text Figure 2).

Text Figure 2.
Text Figure 2.

The Size of the State

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A008

8. Looking ahead, further State-building is anticipated. This will reflect expansion in its capacity and staffing for its existing mandates, adoption of new mandates not executed by any level of government yet, and further shifts in functions from lower levels of government. These processes are already underway. Initial State spending goals for the 2005 budget doubled 2004 expenditures, albeit that these were reined back in the face of Entity resistance. Administrative spending is now targeted to rise by about 60 percent from 2004. Further ahead, BiH aspirations to enter NATO and the EU imply additional expansion, well beyond 2005, reflecting integration of the entity police and army forces into two single State-level forces. And in many other spheres, the EU vision of the future State looks very different from its present configuration (Box 1).

The State and EU Stabilization and Association Agreement

The EU has the most comprehensive view in the international community on the evolution of State government operations. Table 1 lists the institutions that are the minimum requirements before the negotiations for a Stabilization and Association Agreement (SAA) can start. They should be established by end-2005. The table also summarizes the institutional priorities of the EU’s medium-term public administration reform program. The development of these institutions is expected to be key to the success of the SAA negotiation process.

These EU-related plans are consistent with those of the OHR. The latter’s institutional priorities are concentrated in the areas of security and home affairs, and are immediate goals. From OHR’s viewpoint, all their priority institutions should ideally be fully operational (staffed, resourced, and in premises) by mid-2005. In December 2004, the OHR added police and defense reforms to their goals.

C. Assessment

9. Relative to some hopes, State-building has been slow. The State remains a small part of government, and given staff shortages, many of its institutions can carry out their functions in name only. Growth of actual State level employment has been well below the “full complement” specified in State institution rule books.

10. The lack of an independent revenue source has played only a minor role in this. While the Entities have resisted increasing annual transfers to the State, these have risen steadily over the years. And even if the State had enjoyed an entirely independent revenue source, concerns to limit the overall tax burden—including taxes levied by other government levels—would have required close coordination with the Entities, implicitly mimicking the coordination that was forced through the administrative transfer system.

11. But State-building has been haphazard. Institutions have tended to be established with scant regard to the overall coherence or cost of the State: the rationale for the target staffing levels in the rule books adopted by each of the new ministries is not apparent; office space for workers is in short supply; and hiring processes are protracted, often reflecting bureaucratic procedures, sensitivity to the ethnic diversity, and other restrictions—for example, the State level police (SIPA) can currently only recruit from other police forces. Further, the political and administrative capacity of the State to build itself is limited, notwithstanding attempts to boost this, such as with the establishment of a State Treasury in 1999. And in practice, State-building initiatives have tended to reflect the uncoordinated priorities of the international community, including in regard to the State Border Service and Judicial functions.

12. Thus, expansion has been costly. The remuneration structures established in the various State ministries are far more generous than those in Entity governments and elsewhere in the economy.

  • Average remuneration in 2004 for a State government employee was almost 50 percent higher than in the Federation government and 180 percent higher than the RS government (Text Figure 3).

  • Even for similar employees, such as in the Police and State Border Service, remuneration discrepancies are large. The SBS enjoys remuneration more than 22 percent above that of the Federation Police force and 130 percent above the RS police. And SIPA employees enjoy a pay premium of some 80 percent and 250 percent over the Federation and RS police respectively (Text Figure 3).

  • In the State judicial system, the average remuneration of employees in the Prosecutors office for 2004 was more than KM 4,500 per month. This is over 16 times the average per capita income in Bosnia and Herzegovina. By comparison, Italian High Court judges earn only 3 times the average per capita income of Italy.

  • Compared to remuneration outside government—with due allowance for data quality and possible underreporting in that sector—State employees’ wages and benefits are estimated to exceed those of workers outside the budgets by 90 percent in the Federation, and some 125 percent in the RS in 2004 (Figure 4).

Text Figure 3.
Text Figure 3.

Average Budget Monthly Gross Remuneration, Including Benefits, 2004 (In KM)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A008

Text Figure 4.
Text Figure 4.

Monthly Gross Remuneration, Including Benefits, 2004 (In KM)

Citation: IMF Staff Country Reports 2005, 198; 10.5089/9781451977783.002.A008

Notwithstanding concerns to ensure an ethnically diverse workforce, there is little evidence that the generosity of remuneration of State employees is warranted.

13. As a result of these remuneration differentials, when functions shift to the State, their costs rise sharply. The consolidated wage bill of customs in 2004, when it was at entity level, was a little under KM 30 million, or 0.2 percent of GDP. The 2005 budget for the ITA (which took over customs administration as a State level function) anticipates a wage bill (excluding severance) of KM 46 million and 0.3 percent of GDP. This is an increase of over 50 percent. In large part, this increase simply reflects the rise in customs officials remuneration—of some 160 percent for former RS customs officers, and 60 percent for former Federation officials—at the point where, doing precisely the same tasks, they changed status from Entity to State employees. There is no evidence that this wage differential with the Entities, or the wage increase as functions shift to the State, is matched by corresponding differentials in the quality of delivery of the affected public services.

14. Nevertheless, consolidated government spending has dropped relative to GDP since 2000. Although State functions and spending have grown, these trends have hitherto been offset by a compression of other spending—at Entity level—relative to GDP. Thus, consolidated expenditure relative to GDP has fallen from above 55 percent of GDP to about 50 percent between 2000 and 2004. But with State spending growing rapidly, this has required a cumulative 11½ percent reduction in real public spending outside of State government (excluding Entity spending on transfers to the State) over this period.

15. Looking ahead, the macroeconomic strains in State-building will increase. With nominal GDP growth slowing from early 2000 rates, and with real State spending set to rise sharply, further State-building will require additional sharp real cuts in Entity spending. This is necessary if consolidated government expenditure is to remain on a downward track, and, given consolidated revenue, if the consolidated fiscal deficit is not to grow. If the latter two conditions are not met, the consequent fiscal impulse would fuel domestic demand, driving imports up further. This, in turn, would widen the current account deficit which is already extremely high by any standard.2

16. Police and defense reforms are the immediate—but not the sole—focus of these concerns (Box 2). These two reforms will transform the State. They alone will increase total employment in the State budget from some 5,600 in 2004, to 33,000, an increase of almost 6 times. In this context, our estimates suggest that far from securing savings, the envisaged police reforms will raise consolidated Entity and State government expenditure on police by 0.6 percent of GDP permanently. And this cost could be higher still when account is taken of the added burdens emanating from the 2002-03 Civil Service Laws which mandate ethnic diversity in government employment. Defense reforms, unless significantly adjusted from current broad indications, will also significantly raise consolidated spending on defense. But in addition, as noted, there are aspirations to significantly expand existing State institutions and to open various new ones. And outside the State-building arena, donors hope to expand municipal spending, deficits, and borrowing to expand other government functions and the authorities are committed to ensuring thorough ethnic diversity at all levels of government. Given total revenue, such expansions in State, municipal, and other activities will raise the size of consolidated government and the overall fiscal deficit significantly, unless accompanied by matching—and draconian—cuts in other government expenditures.

17. And new State funding arrangements introduce further difficulties. From 2005, under the Indirect Tax System law, the process of discretionary monthly cash transfers to the State by the Entities has ended. It has been replaced by the automatic transfer of funds from the ITA single account to the State for its administrative and debt servicing obligations. The annual amount of these funds is determined by the State alone, without any formal reference to the Entity budgets. Thus, while its former reliance on Entity goodwill to make monthly payments is gone, so is the requirement on the State to coordinate its annual spending plans with the Entities. This may ease State-building at the margin. But it also removes all incentive for the State to economize on its already wasteful use of resources. And by removing any need for it to coordinate its consolidated spending with that of the Entities, disorderly macroeconomic outcomes are clearly possible.

Police Reform

The police reform may commence in the second half of 2005. New State level police forces will be established in every region of the country, likely in a phased manner, replacing the Entity police in an orderly sequence. In a numerical exercise here, we assume phased implementation in the RS and the Federation over two years, including the Cantonal Police forces. By end-2007, efficiency gains are assumed to secure a reduction of the police force of 17 percent.

However, as employees move from the Entity to the State, their remuneration is modeled to increase to the average SBS level, thereby more than offsetting the gains from downsizing. With average SBS remuneration of KM 1807 per month in 2004, increasing annually at the rate of GDP growth, the total wage bill will swell by KM 142 million (61 percent) in 2005-07. And the wage bill will rise from 1.8 percent of GDP to 2.4 percent, and stay there afterwards. This is a permanent increase of 0.6 percentage points of GDP. In addition, a variety of transitional costs will be incurred–from severance pay to new equipment, etc. for the new forces.

If, instead of remuneration rising to the current SBS level, police pay in the unified State force was set at the current level of the Federation Central Government police force—a 20 percent cut in assumed pay—the Police reform would raise BiH spending on Police by only 0.2 percent of GDP.

article image

18. State-building has to change course. While the State was small, and increments were modest relative to total government, the disorderly approach, while regrettable, was not destabilizing in a macroeconomic sense. But the scale of future State-building ambitions— notably the police and defense reforms and aspirations to fully staff a raft of other institutions—and the additional pressures arising from the court rulings on domestic claims changes all this.3 And the change in financing arrangements for the State, just at the point when State-building is about to take a quantum leap forward, risks compromising overall macro and fiscal goals. If these matters are not corrected, they risk enflaming the underlying domestic political sensitivities inherent in State-building.

D. A Better Approach

19. Recent experience points to a number of areas where improvements can be made:

20. The EU can assist by clarifying its priorities. With EU accession the fixed point around which plans for developing State institutions are ultimately oriented, the EU can help the authorities to prioritize their efforts to realize its requirements. Prioritization would best not only cover the sequence for the formal establishment of new institutions, but also their de facto establishment through completion of the necessary complement of staff. The detail and coherence of the resultant roadmap would also help to minimize the inevitable uncertainties among civil servants, often at Entity level, who reasonably seek early assurance about the implications of such government reforms for their job security. And assessments of the phasing of this growth of State institutions should be based on frank acknowledgement of the administrative constraints on the State to build itself. The record of ambitious recruiting targets not met—even when funding is fully available—is salutary. And the analytical basis for the target staffing levels encoded in each of the ministries’ rule books should be thoroughly reexamined to ensure a coherent framework for recruitment in each ministry.

21. Cuts in sub-State spending need to be identified. In cases where State-building raises consolidated government spending—where it is not matched by automatic parallel cuts at sub-State level—there is need to identify exactly where and how offsetting reductions can be made. The recent initiative from OHR, the fiscal sustainability group—consisting of a range of Entity, State, and Cantonal representatives tasked to identify savings in budgets—is timely from this perspective. A promising place for it to focus will be war veterans’ benefits. However, finding scope for savings is insufficient. Mechanisms to ensure delivery of the savings are also required. A key possibility here is the use of the coefficients allocating revenue from the single account within the Federation. Appropriately set, these can draw resources from the Cantons as opposed to the Central Government to support efforts at State-building. But if this mechanism is used, there is need for this to be closely coordinated with Cantonal budgeting to ensure that they can make the appropriate adjustments on a timely basis.

22. Remuneration rates at State level need to be lowered. Functions shifted to the State become more expensive because of the elevated remuneration rates there compared with those in the Entities. With large functions due to shift to the State soon, notably Police and Defense, a reduction in State remuneration rates towards Entity rates is essential if such welcome reforms are not to put intolerable pressure on the consolidated budgets. As noted in Box 2 above, the rise in spending due to the police reforms can be lowered from 0.6 to 0.2 percent of GDP with a 20 percent cut in State wage and benefit rates. The authorities have attempted a step in this direction with a proposed 10-percent wage rate cut for State level employees in 2005. While commendable as a sign of intent, this cut has yet to be implemented and is too small, not least because it leaves benefits untouched, and therefore it still leaves large differentials vis-à-vis the Entities and the private sector. Given the wide gaps with the Entity budget remuneration rates and the iconic importance of generous State level pay in wider wage determination, the State needs to go further in this direction. In this context, the recent initiative for a DFID consultant to examine the remuneration structure at State level is timely. That exercise should not simply focus on rationalizing the remuneration structures within the State, but also between the State and Entity governments, and between the State and the non-budget sector.

23. Large reforms should cut, not increase, costs. The scope and phasing of large reforms such as those for Police and Defense requires close examination given their costs. Further, the impact of the combination of all the reforms on total consolidated budget expenditure year by year, and over the medium term needs to inform the pace, sequence, and design of the reforms themselves to ensure that they are consistent with good fiscal order. This will require much closer coordination of the planning for these reforms with each other than has been the case hitherto.

24. The funding arrangements for the State also need to change. Given the single account, the annual budget for administrative expenses of the State should require the formal endorsement of the Entities before the funds can be withdrawn from the single account to fund those expenses. This procedure will institute formal accountability into the State for its absorption of resources under the new Single Account system, and ensure full coordination between the Entity and State budgets. To prevent this negotiation leading to logjam each year in budget preparations, processes to break negotiating deadlocks between the Entities and the State will be needed. These elements form part of the proposed “fiscal architecture” for Bosnia and Herzegovina, including arrangements to be made for the proposed National Fiscal Council.

Table 1.

Bosnia and Herzegovina: Future State Institutions and Functions

article image

Minimum requirements for Stabilization and Association Agreement with the EU. Not all of them need to be fully operational.

Transitional institutions.

Institutional form still to be decided.

Already identified minimum requirements for EU membership negotiations. Not all of them need to be fully operational.


Prepared by Peter Doyle, Iva Petrova, and Joerg Zeuner.


See Chapter 3.


See Chapter 6.

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