Threats to external stability in the pre-crisis period have now been reduced substantially and foreign non-debt creating flows have declined, sufficient to support external stability. The global economic downturn has raised challenges for evaluating the countries’ fiscal stance and fiscal policy focus should be lowering support to debt sustainability, private sector development, and the currency board stability. The two entity pension funds have been under increasing financial pressures. Putting the public pension systems on a sound footing will encompass a number of complementary steps.

Abstract

Threats to external stability in the pre-crisis period have now been reduced substantially and foreign non-debt creating flows have declined, sufficient to support external stability. The global economic downturn has raised challenges for evaluating the countries’ fiscal stance and fiscal policy focus should be lowering support to debt sustainability, private sector development, and the currency board stability. The two entity pension funds have been under increasing financial pressures. Putting the public pension systems on a sound footing will encompass a number of complementary steps.

III. Cyclical Developments And Fiscal Policy Design1

A. Introduction

1. Fiscal policy in Bosnia and Herzegovina (BiH) has been pro-cyclical prior to the 2009 recession, contributing to the build-up of internal and external imbalances. The expansionary fiscal stance during booms has created expenditure dependency and depleted resources that could be used to smooth incomes during downturns. This chapter attempts to enrich the debate on the desirable fiscal stance in BiH by introducing some measures of cyclically adjusted fiscal balance (CAB). A variant of the CAB should anchor fiscal policy decisions going forward to allow counter-cyclical policy conduct and provide a clear signal to the public about the government’s actual fiscal “effort”.

B. Background

2. Indicators of fiscal performance are used regularly to assess government objectives and fiscal policy outcomes. The overall and the primary balances are generally used as measures of sustainability given their close link to borrowing requirements and indicators of debt burden. Various additional indicators of adjusted balance, such as domestic balance or augmented balance, serve different aims including to inform the view on the fiscal impulse, i.e., the impact of fiscal policy on the economy. In natural resource rich countries, structural balance indicators strip the overall balance from resource revenue (or adjusted it for the variation of commodity prices) to assess the underlying fiscal stance and address intergeneration equity concerns.

3. The fiscal policy stance is affected by discretionary policy actions and by the economic cycle, which is outside of policy makers’ control. The evolution in the general government balance is affected by cyclical fluctuations in the economy. In practice, almost all countries take some advantage of the strong cycle to improve their observed fiscal balances. On account of strong revenue performance, usually some fiscal consolidation takes place. At the same time, however, overheating concerns in the private sector call for tighter policies to offset strong demand and create a cushion to support the economy in the event of a downturn. In downturns revenues decline automatically while some expenditures increase (i.e., unemployment benefits). A large part of public expenditure is not sensitive to the cycle, including wages, pensions, other current and capital spending.

4. Balances adjusted for cyclical movements in the economy are used to assess the government’s underlying fiscal stance net of cyclical effects on revenues and expenditures. In seeking to assess the orientation of fiscal policy, it is therefore desirable to adjust the evolution in public finances for the effects of cyclical fluctuations. In doing so it is possible to understand if fiscal policies supports or counteracts the operation of automatic stabilizers. The CAB is obtained by deducting these cyclical components from the observed balance.

5. This paper studies the fiscal stance in BiH over the past several years by disentangling the effects of cyclical movements on the fiscal balance. Three methods are used for constructing the CAB: (i) a standard approach, based on the output gap; (ii) a “disaggregated” balance approach, based on the cyclical position of individual taxes; and (iii) a modified approach that takes into consideration the effect of large external imbalances on revenue collections (IMF, 2007). The fiscal stance in BiH is then examined in light of the results from the CAB and the information on discretionary measures while keeping in mind the shortcomings of indicators. Final considerations are made to incorporate the CAB into medium-term planning in support of counter-cyclicality along with other possible improvements in costing of reforms and policy coordination.

C. Derivation of the CAB for BiH

6. The economy’s position in the cycle is assessed by means of the “output gap” defined as the difference between observed GDP and its potential level. In periods of demand shortfall the output gap is negative; in periods of excess demand, the gap is positive.

The standard (aggregated) approach

7. The standard approach to estimating the CAB isolates the fiscal stance by stripping out the automatic effects of the output gap (YGAP) on the balance (B) using the relationship:

(CAB/YP)t=(B/Y)tαYGAPt,where(1)

B and CAB are the levels of the actual and cyclically adjusted balance, respectively; Y is the actual and YP the potential GDP; and YGAPt = (Yt-YPt)/YPt. The coefficient α measures the automatic effect of a 1 percentage point change in the output gap on the actual fiscal balance in percent of GDP.

8. Following a standard OECD approach, we estimate the automatic response of the budget to the output gap at about 0.45, close to the level of the revenue-GDP ratio. For simplicity, we assume that only the revenue side of the budget exhibits automatic responsiveness, although automatic stabilizers on the expenditure side could also be significant reflecting changes in the size of unemployment benefits that track the cycle.2

The “disaggregated” approach

9. The output gap could be a poor tool for constructing the CAB if revenues and expenditures depend on developments in the tax bases, which in turn move differently from the output gap. In particular, private consumption, employment and wages, as well as profits may be in a different cyclical phase than GDP and/or the magnitudes of their fluctuations may be different (Cajner, 2005).

10. The disaggregated approach addresses these composition effects on the CAB. This approach relies on calculating the cyclical position of the macroeconomic bases as in the output gap estimation applying the HP filter on the bases extended to 2015. Fiscal elasticities that measure the automatic response of individual revenue categories to macroeconomic fluctuations are borrowed from the OECD average 1 for the personal income tax (PIT) and corporate income tax (CIT), 0.9 for indirect revenues, 0.7 for social security contributions (SSC), and multiplied with the corresponding gaps and with the actual values of revenue categories.

CAB=Σi(REVi*BGAPi*α_REViGAPi),where(1)

BGAP is the gap in the revenue bases and α is the elasticity of the revenue category to the respective base.

The “modified” approach

11. In countries with large and unsustainable current account deficits paralleled by booms in absorption, the output gap may not be a good proxy for capturing the automatic effects of changes in the economy on the budget. The reason is that an unsustainable increase in consumption and investment, more often than not financed by remittances and/or external borrowing, can be offset by an increase in imports, leaving gross domestic product at its potential level. Changes in revenue bases, that generally tend to move in parallel with changes in output, may be more pronounced in such endeavors. A simple OLS regression suggests that tax collection have risen in recent years in line with the manifestation of external imbalances.3 A modified CAB could be used that takes into account large external imbalances in evaluating the fiscal stance.

12. The modified CAB separates the automatic effects on indirect tax collections using estimates of the size of the external imbalance (defined as the unsustainable excess/shortfall of actual spending over potential output), while calculating the automatic effects on other revenue collections using estimates of the output gap. The absorption gap estimates are derived based on the analysis of external stability in chapter II of this paper. Following the approach in IMF (2007), the CAB is constructed assuming that indirect taxes respond automatically only to the absorption gap while other revenues respond only to the output gap:

(CAB/YP)t=(B/Y)tβ*YGAPtμ*AGAPt,where(2)

AGAP is the absorption gap, β is the share of indirect revenue in total taxes and μ is the share of direct revenue in total taxes, and β= μ=0.23 since about half of the revenue (excluding grants) consists of indirect taxes and the other half of direct taxes. Non-tax revenue is assumed to be non sensitive to the cycle.

13. Some consideration associated to the construction of CAB:

  • Measurement errors in computing potential GDP may complicate the interpretation of results, in particular when potential GDP is used to present fiscal ratios (Chalk, 2002). Estimates of potential output are subject to uncertainty inasmuch as the framework does not account for the forces driving the business cycle that vary over time.

  • The assumption of constant elasticity has the potential to introduce distortions in the estimates of automatic stabilizers in case of changes in the composition of revenue. For example, Sancak and others (2010) show that long-run revenue elasticities do not hold well during output contractions and expansions. During recessions tax revenue collapses more sharply than what would be expected based on the decline in the tax base contributing to over-estimate revenue. Moreover, discretionary changes affecting tax rates and bases, such as the ones BiH underwent during 2006–09 (see Box 1), may affect long-run elasticities (IMF, 2005).

  • The CAB does not represent the “true” measure of fiscal effort if there are other forces affecting fiscal outcome unrelated to the cycle but beyond government’s control and uncorrelated with the cycle. For instance, commodity prices and interest rates depend largely on world markets, while foreign grants and loans may be affected by the economic situation abroad (Kanda, 2010).4

  • A key issue in the interpretation of the CAB is the choice of scaling variable—nominal GDP or potential GDP. While results differ modestly under general conditions, care is needed in computing and interpreting these indicators, especially in “less general” cases (Fedelino and others, 2009).

D. An Assessment of Fiscal Policy Based on CAB

14. According to the standard CABs, fiscal policy in BIH has been highly pro-cyclical over the past few years (Tables 1 and 2). The discretionary stance has significantly contributed to the deterioration in the surplus in 2007–08.5 While part of the higher deficit in 2009 was due to the operation of the automatic stabilizers, discretionary loosening also occurred. However, a number of reasons could be behind such result including, as we have seen, a different cyclical position of the revenue bases with respect to the actual GDP and large external imbalances not captured by the potential GDP. In fact, during 2007 and 2008, private consumption (i.e., indirect revenue base) has been above potential, while the base for the PIT and the SSC, derived from wages and employment in the economy, was above potential in 2008 and 2009 because of large increases in wages that occurred between 2007 and 2008.

15. A number of discretionary measures have taken place both on the revenue as well as on the expenditure side over the last few years (see Box 1). Recently, there have been increases of excises to align tax rates with EU levels, but discretionary revenue gains were partly offset by declining customs collections owing to EU accession. Direct taxes were lowered and exemptions re-defined. Tax administration has also improved and, contrary to the experience in many countries, collection does not seem to have suffered during the recent crisis. On the expenditure side, progressive increases in wages and social benefits have occurred since 2006 in a pro-cyclical fashion, dampening the effect of the automatic stabilizers. Unemployment benefits, which generally act as a counter-cyclical force on the expenditure side, have increased during good times with the rights awarded to demobilized soldiers in the Federation in 2006.

Discretionary Revenue Measures, 2006–09

ITA

  • 2006 Introduction of VAT at a flat rate of 17 percent.

  • 2008 Association Agreement with EU (lowered import tariffs on EU goods)

  • 2009 Increase in the road fee (+0.10 KM/l), excises on coffee and progressive increase in excises on tobacco

  • 2009 Exemption of IPA funds from VAT

Federation

  • PIT

    • 2008 rate lowered from 30 percent to 10 percent and a number of exemptions redefined;

    • 2009 unification at the Federation level with elimination of cantonal taxes;

  • SSC

    • 2006 introduced new law;

    • 2009 lowered contributions from wages (employees) from 32 percent to 31percent and on wages (employers) from 11,5 percent to 10,5 percent;

RS

  • PIT

    • 2007 new law changed the tax base, introduced exemptions, and progressive rates;

    • 2009 shift in the calculation of tax base (net to gross wage) and lower (single) rate (8 percent) and redefined exemptions

  • CIT

    • 2007 new law eliminating exemption on reinvestments;

  • SSC

    • 2009 shift in the calculation of tax base (net to gross wage) and lower rate (from 42percent to 30.6 percent)

16. The CAB based on the disaggregated approach shows a similar picture compared to the standard approach. Taking into account the effect of the cyclical position of the individual revenues bases and the individual elasticities, the disaggregated CAB also indicates a worsening fiscal stance. The contribution of automatic stabilizers in 2009 this time is weaker that in the standard CAB. Out of the 2 percent deterioration in the overall balance in 2009, 1.1 percent was due to a decline in revenues, while 0.9 percent was due to discretionary expansion (Table 2).

Table 1.

BiH: Measure of Cyclically Adjusted Balance, 2005-09

(Percent of GDP)

article image
Sources: Fund staff estimates.
Table 2.

CABs and the Fiscal Stance, 2006-09

(Differences, in percent)

article image
Sources: Fund staff estimates.

17. The modified CAB approach yields divergent results only in 2009. Because of a more pronounced absorption boom in 2008 revenues were also above potential in that year. Thus, the cyclically adjusted deficit was higher, suggesting a stronger discretionary deterioration (Table 1). As the current account swung closer to its sustainable level sharply in 2009, the cyclical contraction of revenues dominated the movement in the overall balance. Accordingly, the deterioration in the overall balance in 2009 was largely caused by the effect of automatic stabilizers stemming from the absorption bust, while the fiscal stance posted a discretionary contraction of 1.1 percent (Table 2).

E. Incorporating Cyclical Developments into Medium-Term Planning

18. Cyclical developments should be incorporated into medium-term planning and annual budgets should be aligned to medium-term budget frameworks. CABs are essential tools for designing and assessing counter-cyclical policies. Moreover, CABs have the potential for enhancing credibility of medium-term frameworks by separating discretionary policy actions from automatic stabilizers. While anchoring fiscal policy on an indicator corrected for the cycle seems premature at this stage, the CAB should be made integral part of budget documents and discussions involving discretionary measures. By doing so, the authorities could demonstrate the direct effect of policies on the underlying fiscal stance. In addition to providing a signal to markets on policy intentions, thereby affecting the price and availability of lending, inclusion of CAB could help provide support for domestic reforms.

19. The medium-term fiscal policy objective should be aimed at lowering the size of the government and improving spending efficiency. In support of greater inclusion of the informal economy, export competitiveness and currency board sustainability the overall weight of the public sector on the economy should be lowered. The role of the CAB is to provide guidance as to the pace and direction of adjustment that would allow countercyclical policy conduct. In that respect, building up savings in good times will be necessary for sustaining lower balances during crises.

20. Some considerations for the successful implementation of counter-cyclical policy include, but are not confined to:

  • Coordination of sub-national governments’ policies. It is possible that fiscal objectives of central governments are diluted by actions at other levels of government. Here intra-governmental transfers should play a role in supporting counter-cyclicality by avoiding excessive contraction of sub-national spending during downturns or, conversely, contraction of central government spending to offset the action of at sub-national level.

  • Internalizing expected reform costs. The adjustment path, i.e., the medium-term CAB target, has to internalize the cost of contingent liabilities that are likely to materialize in the future either in result of expected reforms (i.e., severance payments related to public sector downsizing, effects of changes in social security contributions, clearance of public enterprise debts) or because of expected increases in non-discretionary expenditure (i.e., pension outlays due to unfavorable demographic trends).

  • Budget flexibility. Unless they take into consideration the cyclical position of the economy, fiscal rules and budget rigidities tend to create obstacles to counter-cyclical policy conduct. Revenue earmarking and indexation of expenditures, typically wages and social benefits, tend to limit flexibility and obstruct counter-cyclicality. These policies should be periodically re-thought and the benefit they create by eliminating uncertainty and strengthening credibility should be weighed against the advantage of greater budgetary flexibility.

  • Financing availability and debt sustainability. Counter-cyclical policy may encounter obstacles when savings are not sufficient to maintain or increase the level of spending during downturns, external financing is not available or is too costly, or debt sustainability is at stake. By enhancing transparency and fiscal policy credibility, the use of CABs can “sooth” the markets during periods of crises and contain borrowing costs, while savings built in good times should provide a cushion in crises. (IMF, 2009)

  • Capacity building. Capacity for analyzing cyclically adjusted indicators has to be accompanied by a good understanding of the cyclical position of the economy. Moreover, capacity for costing of reforms needs to be strengthened.

21. The Fiscal Council (FC) has the task of coordinating fiscal policies among the various level of government with a view to ensuring macro-economic stability and fiscal sustainability. In the exercise of its duties the Fiscal council adopts the Global Framework of Fiscal Balance and Policies in BiH each May which includes medium-term fiscal targets, macroeconomic projections, the distribution of indirect revenue for the coming year, and upper borrowing limits. The FC relies on the Advisory Group, responsible for the preparation and proposal of decisions and other legal acts adopted by the FC. The FC could be put in charge of monitoring fiscal policy conduct.

22. The functions of the FC Advisory Group could be redefined to include ex-post assessment of fiscal policy performance by observing movements in the CAB. The Group could be in charge of elaborating and publishing macroeconomic projections as input into the Budgets, and assessing the governments’ policies against the pre-determined fiscal anchor. The assessment should be done in light of cyclical developments in revenues and expenditures. The Advisory Group could draft a monitoring report annually or semiannually. To enhance transparency and government accountability the report should be available to the public and could be published on the government’s web site.

F. Conclusions

23. The backdrop of global economic downturn has raised challenges for evaluating the countries’ fiscal stance. Deficits have swelled in countries that have experienced revenue shortfalls, and adopted anti-crisis measures, as well as in countries that have maintained regular spending patterns. In such an environment, judgment of fiscal stance has to be modified to take into account developments that are outside of immediate control of authorities. The judgment has to acknowledge the desirability of the manner in which well-designed automatic stabilizers operate to smooth out the effect of cyclical movements on economic subjects.

24. Fiscal policy in BiH has been highly pro-cyclical in the recent past. Over 2007–08, the deterioration in the fiscal stance was actually faster than what is suggested by the traditional balance indicators and only partially related to cyclical factors. Although with somewhat different magnitude, all the three CABs show discretionary expansions in 2007 and 2008. In good times fiscal expansion was based on public wage and war-related entitlement benefits increases. In addition to creating spending expectations for the future, such policies did not allow building savings and, consequently, in 2009, some counter-cyclicality could only be achieved through further external borrowing. However, due to the inadequate composition, geared towards current spending, discretionary expenditure during the crisis year in the Federation did not support economic recovery.

25. Going forward, fiscal policy should focus on lowering the size of the government in support of debt sustainability, private sector development and the currency board stability. Medium-term objectives should be achieved to the extent possible by countervailing the cycle. Fiscal policy should allow the automatic stabilizers to operate by using, when necessary, well-timed, temporary, and targeted fiscal stimulus.

26. The fiscal response should also internalize the cost of reforms and of contingent liabilities stemming from non-discretionary expenditures. Although infrastructure needs would justify greater structural deficits, projections for rising health and pension spending due to unfavorable demographic trends would call for building up savings ahead of increased demand on entitlements. In view of these challenges the capacity to cost reforms should be strengthened and cyclical developments should be made part of medium-term planning.

References

  • Brondolo, John, 2009, “Collecting Taxes During an Economic Crisis: Challenges and Policy Options”, IMF Staff Position Note, SPN/09/17, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Cajner, Tomaz, 2005, “Cyclically Adjusted Budget balances in Slovenia”, Prikazi in Analize XII/2, Bank of Slovenia, Ljubljana

  • Chalk, Nigel, 2002, “Structural Balances and All That: Which Indicators to Use in Assessing Fiscal Policy,” IMF Working Paper 02/101, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Fedelino, Annalisa, and others, 2009, “Computing Cyclically Adjusted Balances and Automatic Stabilizers”, IMF Technical Notes and Manuals, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Girouard, Nathalie and Christophe Andre, 2005, “Measuring Cyclically-Adjusted Budget Balances for OECD Countries,” OECD Economics Department Working Paper 434, OECD Publishing.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2007, Bulgaria: Selected Issues, IMF Staff Country Report No. 07/390 (Washington).

  • International Monetary Fund, 2009, Brazil: Selected Issues, IMF Staff Country Report No. 09/171 (Washington).

  • International Monetary Fund, Fiscal Affairs Department, 2005, “Cyclicality of Fiscal Policy and Cyclically Adjusted Fiscal Balances,” Board Paper (SM/05/393), International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Kanda, Daniel, 2010, “Asset Booms and Structural Fiscal Position: The Case of Ireland,” IMF Working Paper 10/57, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Sancak, Cemile, and others, 2010, “Tax Revenue Response to the Business Cycle,” IMF Working Paper WP/10/71, (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
1

Prepared by Izabela Karpowicz (FAD).

2

While an approximation, these elasticities are close to those estimated for OECD countries. In particular, the OECD (Girouard and André, 2005) elasticities computed for specific tax categories yield an aggregate revenue elasticity close to 1.

3

During severe crises, problems with tax compliance (Brondolo, 2009) may weaken revenues beyond what suggested by historical trends and collection efforts. Similarly, long-run revenue elasticities do not hold well during output contractions and expansions (Sancak and others, 2010).

4

Adjustment for interest rate changes is not deemed necessary for BiH given that borrowing is largely long-term and with fixed interest rates.

5

The discretionary fiscal stance is calculated as the difference in the CABs across years. The effect of automatic stabilizers is calculated as the difference between the change in the overall balance and the discretionary fiscal stance.