IMAD Working Paper Series http://www.umar.gov.si/en/publications/working_papers Gonzalo C. Caprirolo Slovenia: Enhancing the Fiscal Framework in Delivering Fiscal Stability Working Paper No. 4/2011, Vol. XX Abstract: This paper aims at contributing to enhance the fiscal framework with a view on fiscal stability. The paper reviews the conduct of fiscal policy over the past decade in order to draw lessons for enhancing the existing framework. Attention is given to policy formulation, outcomes and their assessment. The relative strengths of the existing policy framework are evaluated. The paper provides concrete proposals for enhancing the existing fiscal framework which might be useful at the time when changes to the Law on Public Finance are considered. It proposes a clear framework for ownership, accountability and transparency covering the roles of the government and parliament with regard to fiscal stability and policy making. Key words: fiscal stability, transparency, accountability, fiscal strategy, debt target, age related expenditure, parliament, fiscal rules, fiscal council, EU fiscal surveillance, policy in real time, countercyclical policy, Law on Public Finance The Working Paper Series is intended for the pubiication of the findings of research work stiii in progress, the analysis of data series, a^nd the presentation of methodologies in pa^rticuia^r resea^rch a^reas. The aim of the series is to encoura^ge the exchange of ideas about economic and development issues and to publish findings quickly, even if they are not fuiiy conclusive. The opinions, findings, and conclusions expressed are entirely those of the authors a^nd do not necessarily represent the views of the Institute of Ma^croeconomic Ane^iysis e^nd Development. The contents of this pubiication may be reproduced in whole or in part provided that the source is acknowledged. IMAD Working Paper Series Publisher: Institute of Macroeconomic Analysis and Development Gregorčičeva 27 SI-1000 Ljubljana Slovenia Phone: (+386) 1 478 1012 Fax: (+386) 1 478 1070 E-mail: gp.umar@gov.si Editor in Chief: Barbara Ferk, MSc (barbara.ferk@gov.si) Working paper: Slovenia: Enhancing the Fiscal Framework in Delivering Fiscal Stability Author: Gonzalo Carlos Caprirolo Cattoretti, MSc (gonzalo.caprirolo@gov.si) Not language edited. Peer reviewed. Ljubljana, June 2011 CIP - Kataložni zapis o publikaciji Narodna in univerzitetna knjižnica, Ljubljana 336.22(497.4)(0.034.2) CAPRIROLO, Gonzalo Slovenia [Elektronski vir] : enhancing the fiscal framework in delivering fiscal stability / Gonzalo C. Caprirolo. - El. knjiga. -Ljubljana : Institute of Macroeconomic Analysis and Development, 2011. - (Working paper series / IMAD ; 2011, 4) Način dostopa (URL): http://www.umar.gov.si/fileadmin/user upload/publikaciie/dz/2011/dz04-11.pdf ISBN 978-961-6839-08-2 (pdf) 256656128 CONTENTS 1 INTRODUCTION............................................................................................................................................1 2 FISCAL POLICY LESSONS AND CONSIDERATIONS................................................................................2 2.1 The conduct of fiscal policy............................................................................................................................................2 2.1.1 Fiscal policy and business cycle..........................................................................................................................4 2.1.1.1 Using real time and ex-post data in the assessment of fiscal policy......................................................................................4 2.1.2 Assessing policy developments with regard to fiscal stance...............................................................................9 2.1.3 Benchmarking fiscal policy against the concept of a prudent fiscal policy making............................................10 2.1.4 The strength of countercyclical fiscal policy.......................................................................................................15 2.1.5 Taxation policy and fiscal performance.............................................................................................................16 2.1.6 Expenditure policy and fiscal performance........................................................................................................19 2.1.7 EU funds and fiscal stance................................................................................................................................21 2.1.8 Behaviour of expenditure entailments during the cycle.....................................................................................24 2.1.9 Debt developments and fiscal stability...............................................................................................................25 2.2 Fiscal consolidation......................................................................................................................................................28 2.3 Long term sustainability of public finances...................................................................................................................29 3 ASSESSMENT OF THE EXISTING FISCAL FRAMEWORK......................................................................32 3.1 The policy track record.................................................................................................................................................32 3.2 The existing fiscal framework.......................................................................................................................................38 3.2.1 Budget planning.................................................................................................................................................40 3.2.1.1 Macroeconomic forecasting.................................................................................................................................................40 3.2.1.2 Fiscal projections.................................................................................................................................................................40 3.2.2 Fiscal strategy and targets.................................................................................................................................41 3.2.3 Medium-term budget framework........................................................................................................................44 3.2.4 Accountability for fiscal stability.........................................................................................................................49 3.2.5 The EU fiscal surveillance dimension................................................................................................................53 4 TOWARDS ENHANCING THE DOMESTIC FISCAL FRAMEWORK..........................................................55 4.1 Accounting and statistics..............................................................................................................................................55 4.2 Macroeconomic and fiscal forecasting.........................................................................................................................55 4.3 Medium-term budgetary framework.............................................................................................................................56 4.3.1 Setting the targets in line with fiscal stability objectives.....................................................................................56 4.3.2 Criteria for assessing consistency of fiscal targets with fiscal stability objectives.............................................56 4.3.3 Assessment of adequacy of fiscal targets with fiscal stability objectives...........................................................59 4.3.4 Approval of the fiscal strategy and reporting on its compliance.........................................................................60 4.3.5 Multi-annual planning.........................................................................................................................................60 4.3.6 Numerical fiscal rules.........................................................................................................................................63 4.3.7 Procedural rule on revenue neutral tax reforms................................................................................................64 4.3.8 Performance orientation....................................................................................................................................64 4.3.9 Monitoring government's adherence to fiscal stability objective........................................................................64 4.3.10 Enhancing countercyclical role of fiscal policy.................................................................................................65 5 CONCLUSIONS...........................................................................................................................................66 List of figures, tables and boxes Figure 1: General government balance and debt, in % of GDP.................................................................................................................2 Figure 2: General government revenue and expenditure, in Bn. EUR.......................................................................................................2 Figure 3: General government balance (actual and planned), in % of GDP..............................................................................................3 Figure 4: European Commission's estimates of output gap for Slovenia, deviation of actual output from potential output as % of potential GDP.............................................................................................................................................................................................5 Figure 5: Spring forecast of real GDP growth rate, in %............................................................................................................................5 Figure 6: Spring forecast of GDP, at current prices...................................................................................................................................5 Figure 7: State budget balance (outturn minus budget) and output gap....................................................................................................6 Figure 8: State budget. Expenditure and revenue (difference between outturn and adopted budget, in mio EUR...................................7 Figure 9: Government accounts (outturn minus budget), in mio EUR.......................................................................................................7 Figure 10: Slovenia: Change in structural deficit ESA 95, April 2008........................................................................................................9 Figure 11: Slovenia: Change in structural deficit ESA 95, April 2010........................................................................................................9 Figure 12: Change structural balance with respect to MTO (-1% of GDP)..............................................................................................10 Figure 13: Expenditure growth and nominal potential GDP growth - Average annual inflation..............................................................12 Figure 14: Expenditure growth and nominal potential GDP growth - 2% inflation...................................................................................12 Figure 15: Expenditure growth and nominal potential GDP growth - Average potential growth 2002-2008 and 2% inflation................12 Figure 16: Expenditure growth and nominal potential GDP growth - Average potential growth 2002-2008 and change in GDP deflator ..................................................................................................................................................................................................................12 Figure 17: Expenditure to GDP ratio, in %...............................................................................................................................................13 Figure 18: Government expenditure (% change), output gap and GPD growth (%)................................................................................14 Figure 19: Government balance's actual (% GDP) and structural (% potential growth)..........................................................................14 Figure 20: Structural balance's composition (% GDP).............................................................................................................................14 Figure 21: Expenditure to GDP ratio, in %...............................................................................................................................................14 Figure 22: Effect of one-off change (1% of GDP) in policy instrument (tightening)..................................................................................16 Figure 23: Tax revenue and GDP growth and output gap........................................................................................................................18 Figure 24: Contribution to tax revenue growth (p.p.) and GDP growth (%).............................................................................................18 Figure 25: Tax revenue under tax reform and no reform scenario, in bn. EUR.......................................................................................18 Figure 26: Government balance under tax reform and no reform scenario, in % of GDP.......................................................................18 Figure 27: Government expenditure, annual change and contribution to change, in %..........................................................................20 Figure 28: Government expenditure, structure, in % of GDP...................................................................................................................20 Figure 29: Government expenditure, structure, in % of GDP...................................................................................................................20 Figure 30: Government investment, in bn. EUR.......................................................................................................................................20 Figure 31: Total investment, in mio EUR..................................................................................................................................................22 Figure 32: Total investment, structure in %..............................................................................................................................................22 Figure 33: Investment financing, in mio EUR...........................................................................................................................................23 Figure 34: Investment financing, structure in %.......................................................................................................................................23 Figure 35: Capital expenditure, in mio EUR.............................................................................................................................................23 Figure 36: Capital expenditure, structure in %.........................................................................................................................................23 Figure 37: Capital transfers, in mio EUR..................................................................................................................................................23 Figure 38: Capital transfers, structure in %..............................................................................................................................................23 Figure 39: Output gap and total government expenditure growth rate.....................................................................................................25 Figure 40: Output gap and compensation of employees' growth rate......................................................................................................25 Figure 41: Output gap and gross fixed capital formation's growth rate....................................................................................................25 Figure 42: Output gap and social transfers growth rate...........................................................................................................................25 Figure 43: Spring forecast of GDP real growth rate, in %........................................................................................................................26 Figure 44: General government debt, in % of GDP..................................................................................................................................26 Figure 45: 10-year bond spread against benchmark German bund........................................................................................................27 Figure 46: Government balance including 33% of frontloading contingent liabilities, in % of GDP.........................................................30 Figure 47: Required expenditure effort to meet MTO target, in % of GDP..............................................................................................30 Figure 48: Required revenue effort to meet MTO target, in % of GDP....................................................................................................30 Figure 49: Pension expenditure, in mio EUR...........................................................................................................................................34 Figure 50: State budget: Expenditure and revenue - difference between figures in supplementary (REB) and revised budgets (SSP) and the adopted budget (SP), mio. EUR..................................................................................................................................................36 Figure 51: Planning expenditures and tax reform while revenue buoyancy, bn. EUR.............................................................................36 Figure 52: Government debt to GDP ratio under structural deficit target of 1% of GDP and 5% nominal GDP growth..........................39 Figure 53: State budget, revenue and expenditure..................................................................................................................................45 Figure 54: General government expenditure growth, in %.......................................................................................................................48 Figure 55: General government's expenditure, in % of GDP...................................................................................................................48 Figure 56: Fiscal variables under a debt-to-GDP target ratio of 18% (scenario of 4.5% nominal GDP growth from 2016 onwards)......58 Figure 57: Fiscal variables under scenario of expenditure growth (4.5%) lagging behind nominal GDP growth (5%) from 2016 onwards ..................................................................................................................................................................................................................63 Table 1: Structural balance estimates in % of potential output..................................................................................................................8 Table 2: Structural balance estimates, in % of potential output.................................................................................................................8 Table 3: Difference between variables estimated in real time and with ex-post data................................................................................8 Table 4: Indicators local government finances.........................................................................................................................................27 Table 5: Fiscal targets and their evolution................................................................................................................................................42 Table 6: Dates of adoption of budgets, revision and supplementary budgets.........................................................................................44 Table 7: Aggregate expenditure ceilings in a sample of countries...........................................................................................................61 Box 1: A Framework for determining government expenditure ceilings "the fiscal rule"..........................................................................46 Box 2: The Fiscal Council and its mandate..............................................................................................................................................51 Box 3: The Central Planning Bureau (CPB) in the Netherlands...............................................................................................................59 Summary The paper aims at contributing to enhance the fiscal framework by elaborating on the incentive structure which would facilitate the delivery of fiscal policy in line with fiscal stability objectives. The proposal is based on assessment of fiscal policy outcomes and conduct of policy on the run to the crisis and takes into account recent policy changes mainly the introduction of a framework to determine government expenditure ceilings and a Fiscal Council to monitor government adherence to fiscal stability objectives. Main findings concerning fiscal policy and policy outcomes include: i) government balances that were close or lower deficits than those planned in Convergence and Stability programs (the main exception was the year 2008); ii) government expenditure growth followed de facto a close or more stringent course than the one implied by a prudent fiscal policy making principle in which expenditure growth does not exceed a projected medium-term rate of potential output growth (exception 2001 and 2008); iii) to some extent actual government's expenditure dynamics were anchored on forecasts of macroeconomic variables (e.g. GDP) whose outcome was higher than forecasted; iv) there was a degree of deficit bias in policy making but the main factors underpinning budget dynamics (slippages and compression) were wage and tax policies which require sustainable policies; v) the assessment of the delivering on the MTO depends whether it is done with real-time or ex-post data, primarily concerning volatile potential growth and output gap estimates); vi) prudent expenditure growth is desirable but to achieve fiscal targets it is necessary a consistent tax policy; vii) post crisis conduct of fiscal policy (consolidation) indicates that delivering on fiscal targets requires primarily ownership as adherence to instruments cannot deliver on targets but require discretion and political will. Recent changes to the policy framework, i.e. setting expenditure ceilings consistent with debt and primary balance target and the establishing of a Fiscal Council, have made explicit the objective of fiscal stability in the conduct of fiscal policy. Recent policy development also shows that these changes have not prevented the shifting of policy targets (Stability program 2011 and 2010). Yet, they can be regarded as steps towards a framework for fiscal accountability within a democratic policy setting. The paper proposes an incentive structure aiming at delivering fiscal stability (consolidation and post consolidation periods) within a democratic setting in which delegated official and institutions fulfil their prime roles and responsibilities. The paper proposes the following amendments to Act on Public Finance: 1) making explicit the fiscal stability objective in the legislation and; 2) Enhancing accountability in planning, approving, delivering and monitoring the objective of fiscal stability. Accordingly, the government should be entrusted with the design of a fiscal strategy and the parliament with the responsibilities of approving, monitor its implementation with respect to agreed principles of fiscal stability and demanding correction if targets are not met. The paper proposes a set of principles of fiscal stability to benchmark government strategies. It suggests the formulation of 4 year fiscal strategies, including targets, to be aligned with the political cycle. The framework should introduce a procedural rule on tax reform financing to ensure the delivery on targets (expenditure policy is not enough) and full report on tax expenditure. The fiscal surveillance role of the parliament should be enhanced by establishing a Parliamentary Budget Office or requesting independent fiscal institutions to provide reports to the parliament to monitoring government fiscal strategies. The paper also proposes enhancing accountability of the Fiscal Council by streamlining its role in line with its own mandate as this institution is financed with tax payers' money. Povzetek Cilj delovnega zvezka je prispevati k izboljšanju fiskalnega okvira. Pri tem se avtor osredotoca predvsem na strukturo spodbud, ki bi omogočile izvajanje fiskalne politike v skladu s cilji fiskalne stabilnosti. Predlog temelji na oceni rezultatov fiskalne politike in vodenju politike pred časom krize ter upošteva nedavne spremembe politik, predvsem uvedbo okvira za določanje zgornje meje javnofinančnih izdatkov in uvedbo Fiskalnega sveta, ki nadzoruje delo vlade pri zasledovanju ciljev fiskalne stabilnosti. Glavne ugotovitve glede vodenja fiskalne politike in rezultatov le-te zajemajo: i) javnofinančni primanjkljaji so bili blizu ali nižji od načrtovanih v konvergenčnih programih in programih stabilnosti (glavna izjema je bilo leto 2008); ii) rast javnofinančnih odhodkov je bila dejansko nižja ali v skladu z rastjo, ki je ustrezna pri vodenju preudarne fiskalne politike (angl. prudent policy making principle), kier rast odhodkov naj ne bi presegala napovedane srednjeročne potencialne rasti BDP (izjemi sta bili leti 2001 in 2008); iii) do določene mere je bila dejanska dinamika javnofinančnih izdatkov določena z napovedmi makroekonomskih agregatov (npr. BDP), ki so bili na koncu višji od napovedanih; iv) do neke mere je bilo vodenje politike nagnjeno k deficitu, vendar sta bila glavna dejavnika proračunske dinamike (odmiki od načrtovane dinamike in krčenje proračuna) politika plač in davčna politika, ki zahtevata vzdržno politiko; v) ocena izpolnjevanja srednjeročnega fiskalnega cilja (angl. MTO) je odvisna od tega ali temelji na trenutnih podatkih ali ex-post podatkih, predvsem zaradi nihanj ocen potencialnega proizvoda in proizvodne vrzeli ; vi) zaželena je sicer zmerna rast izdatkov, vendar je za dosego fiskalnih ciljev potrebna konsistentna davčna politika; vii) vodenje fiskalne politike po krizi (konsolidacija) kaže na to, da je za doseganje fiskalnih ciljev v prvi vrsti potrebno lastništvo, saj doseganje ciljev ni možno le z uporabo instrumentov, temveč zahteva diskrecijo in politično voljo. Nedavne spremembe okvira politik, kot je določitev zgornje meje izdatkov v skladu s cilji dolga in primarnega salda ter ustanovitev Fiskalnega sveta, so izpostavile cilj fiskalne stabilnosti pri vodenju fiskalne politike. Nedavni razvoj politik prav tako kaže, da te spremembe niso preprečile sprememb v fiskalnih ciljih (Program stabilnosti 2011 in 2010). Lahko pa jih štejemo kot korak naprej k okviru fiskalne odgovornosti v sklopu demokratičnega določanja politike. Delovni zvezek podaja strukturo spodbud za zagotavljanje fiskalne stabilnosti (za obdobje konsolidacije in po njem) v demokratičnem okolju, kjer pristojne osebe in institucije izpolnjujejo svoje primarne vloge in odgovornosti. Avtor predlaga naslednje spremembe Zakona o javnih financah: 1) v zakon je potrebno jasno zapisati cilj fiskalne stabilnosti in 2) okrepiti je potrebno odgovornost pri načrtovanju, odobravanju, izvajanju in nadzorovanju cilja fiskalne stabilnosti. V skladu s tem bi morala vlada pripraviti fiskalno strategijo, državni zbor pa bi imel odgovornost odobravanja in nadzora nad implementacijo glede na dogovorjena načela fiskalne stabilnosti, zahteval pa bi lahko tudi popravek v primeru, da cilji ne bi bili doseženi. Delovni zvezek predlaga sklop načel fiskalne stabilnosti za primerjavo z vladnimi strategijami. Predlaga štiriletne fiskalne strategije, vključno s cilji, ki so usklajeni s političnim ciklom. Okvir bi moral uvesti proceduralno pravilo financiranja davčne reforme, da bi se s tem zagotovilo doseganje ciljev (politika izdatkov ne zadošča) in popolno poročilo o davčnih izdatkih. Vloga parlamenta kot fiskalnega nadzornika bi morala biti okrepljena z vzpostavitvijo posebne parlamentarne proračunske službe ali s pozivom neodvisnim fiskalnim institucijam, ki bi pripravila poročila kot podporo za nadzor nad vladno fiskalno strategijo. Delovni zvezek predlaga tudi okrepitev odgovornosti Fiskalnega sveta, z uskladitvijo njegove vloge z njegovim mandatom, saj je ta institucija financirana z davkoplačevalskim denarjem. 1 INTRODUCTION In the last two years important policy changes to fiscal policy frameworks have or are been made both at the national and the EU levels. In Slovenia these have included the setting up of a fiscal council, the introduction of a framework to determine government expenditure ceilings and the introduction of a new program budget classification used in the elaboration and execution of the State budget. Other important change concerns the government possibility when preparing the budget to rely on alternative forecast of macroeconomic variables and not only on a single forecast as it has been the case in the past. At the international level given the looming fiscal deficits there has been a strong push in some institutions (EU Commission, IMF, OECD) for enhancing fiscal frameworks and changing institutional settings. One key institutional change in this regard is the introduction of the so-called EU semester which in fact provides the EU Council (supported by the Commission) with the possibility of influencing ex-ante budget developments in member states. Fiscal frameworks are rooted in traditions and history and institutional changes require rationale as they can have implications that go beyond fiscal policy making. Given that some changes to the Slovenian fiscal framework are still being settling, this paper assesses the existing fiscal framework with the view of delivering fiscal stability and provides suggestions as to how this dimension could be enhanced taking into account the democratic institutional set up. The paper in the second section assesses the conduct of fiscal policy over the past decade and the fiscal challenges ahead in order to draw rationale for improving the fiscal framework. Section three assesses the existing fiscal framework which underpinned past fiscal performance and its most recent changes. In section four, the paper provides concrete suggestions on how to enhance the fiscal framework centered around the issue of fiscal stability. Last section concludes. 2 FISCAL POLICY LESSONS AND CONSIDERATIONS 2.1 The conduct of fiscal policy The conduct of fiscal policy over the past decade can be broadly divided in two phases. The first phase can be called of a gradual fiscal consolidation covering the period 2001-2007 and a second phase in which consolidation reversed and the economy was hit by the international financial crisis. In the first phase the general government deficit as a percentage of GDP declined from 3.6% in 2002 to a balance position in 2007. In this phase the general government debt reached a peak in 2002 (27.9% of GDP) and then declined to the lowest level in the decade in 2008 (21.9% of GDP) (Figure 1). In 2008 fiscal consolidation reversed and since then public finances have worsened significantly on the back of the international financial crisis and its transmission to the domestic economy which was reflected primarily in the collapse of government revenue. In 2009 the general government deficit was 5.8% of GDP and the debt-to-GDP ratio jumped to 35.2% at the end of that year. The deficit and debt figures as percentage of GDP for 2010 are 5.5% and 38% respectively. Given the substantial widening of the fiscal gap the tasks ahead are to ensure fiscal consolidation and sustainability of public finances. Figure 1: General government balance and debt, in % of GDP Figure 2: General government revenue and expenditure, in Bn. EUR 45,0 40,0 35,0 30,0 25,0 20,0 15,0 10,0 5,0 0,0 -5,0 -10,0 1 1 1 1 1 Government debt f / z/ ...... i > / y Source: SORS, Ministry of Finance: Report on Excessive Deficit Procedure 2010 and Stability Program 2010 11,0 9,0 7,0 Looking at government's revenue and expenditure dynamics over the two phases (Figure 2), it can be argued that despite the trend decline in deficits over the period 2003-2007, the reduction could have been faster and as a result there was a degree of "tolerance" for the existence of deficit or what can be called an in-built deficit bias in policy makers1. Data indicates that actual balances outperformed planned deficit reduction in the period 2004-2007 (Figure 3) suggesting conservative planning while planned balances in 2007 for the years 2008-2009 foresaw deficits below 1% of GDP (Figure 3). Such a policy was broadly in line with a medium-term target estimated in real time and set in structural terms at the level of 1% of GDP. The envisaged fiscal stance in the year 2007 for the period 2008-2009 was also consistent with a foreseen decline in the debt-to-GDP ratio (SP 2007). Nevertheless, in 2008 the actual level of deficit was well above the planed level in 2007 (Figure 3) reflecting expenditure increase and the effect of the crisis on government revenue that exposed the impact of the 2006-2007 tax reform. While expenditure expansion in 2008 points out to a degree of tolerance for the existence of a deficit (discretionary increase in some expenditure categories in an election year) it also points out to unsustainability of wage policy (stop and go policies) which after a long period of wage compression it had to be adjusted at some point in time. The strong fiscal deterioration after the widening of the deficit in 2008 cannot be explained primarily by deficit bias (i.e. policy attitude), but mainly by the effect of the crisis on public finances and its negative interaction with policy changes affecting tax revenue and expenditure. Figure 2 also highlights two policy issues which demand different policy response: a degree of deficit tolerance until 2008 (persistence or low size fiscal deficits) and; fiscal adjustment since 2009 defined by the need to reduce a sizable fiscal gap. While a robust fiscal framework might contribute to mitigate deficit bias and enhance sustainability of public finances, fiscal consolidation requires strong political determination and careful choice of policies (quality aspects) and speed of adjustment to ensure its sustainability. Figure 3: General government balance (actual and planned), in % of GDP 1,0 0,0 -1,0 -2,0 Q. Q (D S? -3,0 -4,0 -5,0 -6,0 -7,0 Actual -Convergence program (May 2004) —A— Convergence program (Jan 2005) —0— Convergence program (Dec 2005) -----------Stability Program (Dec 2006) ----Stability Program (Dec 2007) —m— Stability Program (April 2009) -Stability Program (Jan 2010) Source: SORS, Ministry of Finance (Convergence and Stability programs various years) 1 According to literature (e.g. Debrun, et al., 2009) policymakers act rationally and respond to incentives, including reelection concerns, pressures from interest groups and constituencies. 2.1.1 Fiscal policy and business cycle Fiscal policy outcomes (government balance and debt) in the period under consideration were influenced by the effect of business cycle and by policy measures. Therefore, the issue is to determine the impact of business cycle on fiscal outcome and the impact/attitude of fiscal authority with respect to the cycle or the fiscal stance. The latter is related to the issue of active use of fiscal policy as a macroeconomic stabilizing tool and to the goals, targets or rules governing fiscal policy. 2.1.1.1 Using real time and ex-post data in the assessment of fiscal policy Assessing the impact of the business cycle on government finances and of the fiscal stance with regard to the economy is contingent to a large extent to whether fiscal stance is assessed with real time or ex-post data. This is also related to the issue of the use of fiscal policy for short-term economic stabilization. Real time data pertains to information actually available to policy-makers at the when time their decisions are/were taken. Expost data, or "realized" data refers to revised data not available at the time when decisions are/were made. This distinction becomes critical when assessing the behaviour "attitude" of fiscal authorities with regard to the design of the fiscal framework and compliance with policy commitments.2 Different set of data "ex-ante" or "ex-post" can also provide different information as to the perceived effect of economic developments on budgetary outcome. One important variable used in assessing/formulating fiscal developments is the output gap. This variable is subject to great uncertainty and revisions. In particular real-time estimates of the output gap are characterized by a high degree of uncertainty. According to Marcellino (2010), the magnitude and the sign of the real-time estimates of the euro area output gap are very uncertain and the uncertainty in output gap estimates tends to be more significant than estimates of real GDP growth. Such uncertainty in the case of the euro area derives mostly due to parameter instability, while data revisions seem to play a minor role. In the case of Slovenia and catching up countries where data tends to be more volatile the estimates of output gap are also more uncertain. Figure 4 shows subsequent estimates of the output gap for Slovenia made by the EU Commission. It shows that the output gap estimates are also quite volatile in the case of Slovenia. In particular the output gap estimated for the year 2007 ranges from 0 to 6.3 percent of potential GDP. Notice in Figure 4 that in Autumn 2006 the real time estimate of the output gap for the year 2007 which was used to assess cyclical conditions of the economy for the year in which the proposed budget was going to be executed and theoretically used to decide a discretionary policy response (if such would have been the policy approach of the fiscal authority) indicated that the output gap of the economy was closed. Subsequent revisions of the output gap for that year suggest not only that it was positive but it was probably one of the largest positive gaps in recent history. 2 According to Agnello and Cimadomo (2009) the assessment of fiscal stance of industrialized countries overthe1994-2006 based on revised data (ex-post data) points to a pro-cyclical fiscal policy stance whereas based on real-time data indicate that fiscal policy was intended to be counter-cyclical. They refer to new strand in the literature focusing on fiscal policy plans, or "intentions", rather than actual policies which is relevant to assess policy formulation. In particular, Cimadomo (2007) proposes to estimate fiscal policy rules based on information sets similar to those available to fiscal policy makers at the time of budget policy formulation (budgetary decisions and real-time macro indicators). The main finding of this literature is that ex-ante fiscal plans are counter-cyclical, especially in expansions, whereas ex-post data point to a-cyclicality (Beetsma and Giuliodori, 2008) and Beetsma and Giuliodori, 2007). The literature also focus on actual policies, and incorporate revised information as regards fiscal indicators and real-time data for the output gap and other explanatory variables (Forni and Momigliano, 2005 and Golinelli and Momigliano, 2006). Figure 4: European Commission's estimates of output gap for Slovenia, deviation of actual output from potential output as % of potential GDP Spring 2009 ■ Autumn 2008 ■ Autumn 2007 ■ Autumn 2006 5,0 4,0 3,0 2,0 Q- ^^ 1,0 sS 0,0 -1,0 -2,0 -3,0 2004 2005 Source: European Commission (Spring and Autumn forecast) Figure 5: Spring forecast of real GDP growth rate, in % 2006 2007 2008 Figure 6: Spring forecast of GDP, at current prices i^33,0 Source: SORS and IMAD. In the case of Slovenia the forecast of real GDP growth rate and nominal GDP at current prices tends to be less uncertain than that of the output gap. Notice that the forecasts of real GDP growth rate (Figure 5) and of GDP at current prices (Figure 6) made at the time of budget preparation in April each year (Spring Forecast -SF) do not exhibit large dispersion. For the period 2004-2008 what is important to highlight is that the outturn figures of real GDP growth rate and GDP at current prices were higher than those in the forecast used in budget preparation. This impinged a countercyclical bias to budget formulation which is also reflected in budget outcomes during 2004-2007. Figure 3 shows that the outturn figures for general government budget balances outperformed those estimated based on adopted budgets3. 3 In the case of Slovenia the better fiscal outcomes than planned during 2004-2007 resulted also from a non deliberate interaction between budget planning and forecast which for example it was the case in Netherlands in the past. Until 2006 governments in the see next page 41,0 39,0 35,0 31,0 29,0 27,0 25,0 Given the difference in output gap estimates using real time and ex-post data one way of assessing authorities' adherence to fiscal targets in light of business cycle developments can be to look at the difference between budget outturn (deficit/surplus) and adopted budget figures in light of ex-post estimates of output gap (2010).4 In the case of the State budget, which is under the direct control of the government, the comparison indicates that public finances exhibited in general a degree of anti-cyclicality in the period 2001-2007 (Figure 7). When the output gap was negative the budgetary outcome was a higher deficit than planned and when the output gap turned positive the budgetary outcome was a lower deficit/higher surplus than planned. If attention is given only to the difference between figures in budget outturn and in adopted budgets it would suggest that during 2002-2004 there were budget slippages and during 2005-2008 positive surprises or conservative planning/assumptions (Figure 7). Figure 7: State budget balance (outturn minus budget) and output gap 200 100 - CC -100 3 E -200 - -300 -400 -500 7,0 6,0 5,0 4,0 3,0 2,0 1,0 0,0 -1,0 -2,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 Source: Ministry of Finance. A more detailed look at the State budget figures and to the differences between expenditure and revenue outturn and budgeted (Figures 8 and 9) indicates that in the years 2001-2003 when revenue outturn felt short of that budgeted, expenditure was corrected downwards (exception in 2003) pointing out to a degree of optimism in revenue forecast. In the period 2004 to 2008 when revenue outturn exceeded budgeted revenue, expenditure was mostly corrected downwards (exception in 2004 and 2008). Among the reasons that might contribute to explain why expenditure was adjusted downwards in the period 2005-2007 would be that the financing of some expenditure categories is co-financed with domestic sources which might not have been used due to low withdrawal of EU funds or simply due to non realization of discretionary expenditure. Netherlands followed a cautious budget planning based on the so-called ^trend-based budget policy". The system consisted of planning the budget based on a cautious stance regarding the forecast damping revenue projections. The growth forecast used in planning the budget was based on the neutral economic scenario adjusted by a margin of caution. The margin of caution in the growth forecasts was 0.5% of GDP since 1993-1994 to 2001 and then reduced to 0.25% of GDP (Beetsma et al., 2010). 4 Notice that the budget framework in the case of the State budget demands exact deficit/surplus targeting which is in turn is achieved by means of revised or supplementary budgets and in some years by budget execution law that allowed for a small margin of deviation regarding the figure in the adopted budget. 0 Figure 8: State budget. Expenditure and revenue (difference Figure 9: Government accounts (outturn minus budget), in mio EUR between outturn and adopted budget, in mio EUR 7,0 6,0 5,0 4,0 3,0 2,0 1,0 0,0 -1,0 -2,0 -3,0 -4,0 Expenditure Revenue Revenue minus EU receipts OLrtpurt gap % potertiä oLrtpurt (LHS) 1.400,0 1.200,0 1.000,0 800,0 600,0 400,0 200,0 0,0 -200,0 -400,0 -600,0 -800,0 200 1 2003 ♦ 20 i 08 > ■ )0 -4 50 -2 0 2001 # ■ 2006 0 4 00 60 0 80 200 JJBSfP0""' ♦ ■ —-200"— 2007 200S Rl (Out-:b2) ■ nnn ♦ Re venue minu s EU funds ■ -500 ■ Id al revenue Revenue (outturn minus budgeted) Source: Ministry of Finance, own calculation and European Commission (output gap) When assessing the fiscal stance with regard to the business cycle or the fiscal stance in itself the concept of structural balance and its change are often used.5 This concept has also pre-eminence in the context of EU fiscal surveillance despite the high degree of uncertainty associated with estimates of potential output and output gap (which is also the case in Slovenia; see Figure 4) and the divergence between estimates using real time and ex-post data. The uncertainty of output gap estimates is also related to the underlying macroeconomic forecast, revised GDP data and the length of the forecast which highlights the importance and unbiased nature of the forecast and of the potential output estimates. As a consequence, the assessment of fiscal policy (structural balance) based on estimates of potential output on real time and expost realized data can provide different result. In particular, the "realized" stance of fiscal policy can differ from the "true", or intentional, policy stance. This consideration is not only relevant for assessing fiscal stance but also when determining the appropriate fiscal targets, objectives and policy framework. The issue of reliability of estimates of structural balances for assessing fiscal stance, its position and the policy response to the economic developments (measured by changes in structural balance) can be illustrated in the case of Slovenia by looking at estimates of structural balances on real time in 2007 and using ex-post data. Table 1 shows three consecutive estimates of structural balance on real time made by the EU Commission (EC) on September 2007, April and July 2008 all made based on the government balances provided in the 2007 Stability Program (SP). The structural balances for the years 2007-2009 in all cases with the exception of the forecast for 2008 made in Autumn 2007 range as a percent of potential GDP between -0.7% and -1.1%. Notice in particular that the structural balance reported in the Stability Program for 2007 and assessed in July 2008 was 0.8%. Based on the EC assessments, the broad picture indicates that the structural balance estimated on real time declined in 2006 and would stabilized in the period 2007-2009 around 1% of potential GDP. Such a structural balance would have been consistent with a constant debt-to- 5 The concept of structural balance refers to the budget outcome excluding the effects of the business cycle on revenue and expenditure arising from the divergence between actual and potential output estimates or the so-called output gap. GDP ratio close to 30% of GDP in the long run (assuming that contingent liabilities are tackled by pension reform and structural policy). With a structural deficit of 0.8% of GDP the debt-to-GDP ratio would have remained constant at the level of 25% of GDP which was slightly higher than the level reached in 2007 (23.1%). Table 1: Structural balance estimates in % of potential output Structural balance 2006 2007 2008 2009 2010 Stability program (Dec 2007) -1.4 -0.8 -1.0 -0.8 -0.2 European Commission (Autumn 2007) -1.1 -1.1 -1.4 -1.0 European Commission (April 2008) -1.3 -0.7 -1.1 -0.7 European Commission (July 2008) -1.1 -0.8 -1.0 -0.7 -0.1 Source: European Commission: Public finances in EMU 2008 (July 2008) The same exercise made with ex-post data for the period 2006-2008 using EU Commission and the Ministry of Finance estimates of output gap provide a complete different picture to that just described based on Table 1. For similar government balances in the period 2006-2008 the structural balances are completely different. Notice in particular that the structural deficit in 2007 jumps from 0.8% as assessed in July 2008 to 2.9% (Table 2). More broadly, estimates of structural balance made with realized data indicate deficits well above 2 percentage points of potential GDP and the trend shows a substantial worsening in the period 2006-2007. Table 2: Structural balance estimates, in % of potential output Structural balance 2006 2007 2008 2009 2010 European Commission (April 2009) -2.1 -1.7 -2.5 -4.9 -5.2 European Commission (April 2010) -2.6 -2.9 -4.8 -3.8 -4.4 Stability Program (April 2009) -2.9 -4.1 -2.3 Stability Program (January 2010) -4.3 -4.2 -4,0 Source: European Commission: Public finances in EMU 2009 and 2010 Based on structural balances estimated in real time it can be said that the fiscal stance in 2006-2007 was closed to the government fiscal target of 1% structural deficit (Table 1). Estimates based on ex-post data indicate that the structural deficit was close to 3% of GDP and far away to the fiscal target (Table 2). The reason for this discrepancy during 2005-2008 can be traced, as mentioned before, to the extreme large variability of output gap estimates which in turn is explained by a larger discrepancy in estimates of potential output growth than discrepancy in estimates of real GDP growth rates (Table 3). Table 3: Difference between variables estimated in real time and with ex-post data Real GDP growth, % Potential output growth,% Output gap Nov.07 Apr.10 Difference1 Nov.07 Apr.10 Difference1 Nov.07 Apr.10 Difference1 2005 4.1 4.5 10 4.0 3.3 -18 -1.6 0.4 -125 2006 5.7 5.8 2 4.2 3.4 -19 -0.2 2.7 -1450 2007 6.0 f 6.8 13 4.9 3.2 -35 0.9 6.3 600 2008 4.6 f 3.5 -24 4.7 3.2 -32 0.8 6.6 725 2009 4.0 f -7.8 -295 4.5 2.1 -53 0.3 -3.7 -1333 Source: European Commission 1 - As a percentage of the variable estimated in November 2007 (real time); f - forecast. Notwithstanding the uncertainty of using structural balance in guiding policy in real time, governments in the EU are bound to set policy targets in terms of structural balances. In addition, changes in structural balances mapped out to the output gap are used to withdrawn conclusions regarding policy intentions against the business cycle and the changes are interpreted as discretionary actions of a government. While theoretically yearly changes in structural balances and more importantly changes in structural balances regarding a determined structural balance level set as fiscal target could be interpreted as discretionary actions or policy intention, in practice this is difficult to pinpoint as the calculation of structural balance depends on output gap estimates. Furthermore, the issue is even more complicated because the impact of policy changes on the policy stance might be masked by the state of the economy and difficulty to assess changes in tax elasticities (e.g. the effect o tax reduction at the time when the economy is booming). The fiscal stance in Slovenia in the period 2006-2009 illustrates this issue when looking at changes in structural balances measured with two different output gap estimates using real time (April 2008) and ex-post data (April 2010). When using real time data (Figure 10), the estimations indicate that the fiscal stance was tightened in a countercyclical manner in the year 2007 and was aimed to be tightened in 2009 meanwhile it was loosened pro-cyclically in the years 2006 and 2008. The same fiscal stance using ex-post data (Figure 11) indicates pro-cyclical loosening during 2006-2008 and pro-cyclical tightening in 2009. Figure 10: Slovenia: Change in structural deficit ESA 95, April 2008 Figure 11: Slovenia: Change in structural deficit ESA 95, April 2010 0,8 üS 0,2 20 07 20 )9 / / 7 ............. \ / / / > / / r 20 r------------ \ 2008 / ▼ Pro cyclical oiosening : ounter-cyc ical tighte 009 A , ning \ \ > \ 007 \ / \ \ \ 2008 1 'ro-cyclica loosening 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1, 6 change in structurä deficit, % Source: SORS, own calculation -6,0 -4,0 -2,0 0,0 2, change in structurä 0 4,0 6,0 8,0 deficit, % 2.1.2 Assessing policy developments with regard to fiscal stance In a policy framework in which governments are mandated to set fiscal targets in terms of structural balance (i.e. balance net of cyclical elements), like the one in place in the EU since 2006, fiscal policy relies mainly automatic stabilizers to offset business cycle fluctuations and discretionary policy is ruled out. In such a framework, the evaluation of the fiscal stance against the cycle seems to be made more appropriately based on deviations of structural balances with respect to a benchmark policy objective than simply by estimates of changes in structural balance. Also in this case the fiscal stance can be evaluated using real time and exposed data. Taking into account the fact that since 2005 Slovenia had to and formally set its medium term objective (MTO) in terms of structural deficit of 1% of GDP, deviations from that target could indicate the direction of the policy intention with regard to the cycle. Figure 12 show deviations from the MTO mapped out against the output gap using real time data (2008 (08)) and ex-post data (2010 (10)). Figure 12 indicates that when using real time data, the fiscal stance does not deviate significantly from the medium term target. The 1,0 0,5 -0,5 -1,0 -0,2 -1,5 0,4 -2,0 -2,5 -0,6 2007(10) deviation averages 0.1% of potential GDP in the period 2006-2009, suggesting a fairly neutral stance around the target. Using ex-post data (Figure 12) indicates that the structural balance diverges from the medium term objective significantly (the average deviation is -2.5% percentage points of potential GDP). This would indicate a very loose fiscal policy in both good (2006-2008) and bad times (2009). Figure 12: Change structural balance with respect to MTO (-1% of GDP) 8,e- 6,0 4,0 - 2,0 -2008(8) 2006(8) ' I 1,5 -1,0 -0,5 0 -2,0 - -4,0 - Q. Q CJ3 2009(10) Change in cyclically adjusted balance, % 6r-fi- Counter-cyclical tightening 0,5 Pro-cyclical tightening Source: SORS, own calculation 2.1.3 Benchmarking fiscal policy against the concept of a prudent fiscal policy making To assess the fiscal stance with respect to the business cycle, besides using the concept of structural balance, whose estimation is contingent to the set of data used, expenditure developments can be benchmarked against the notion of expenditure growth consistent with prudent fiscal policy making. In fact government expenditure dynamics, in absence of tax changes, determine fiscal aggregates (i.e. Deficit and debt). The concept of prudent fiscal policy making with regard to government expenditure can be operationalized in different ways. For example, the following measures can be regarded as a proxy of a prudent rate of government's expenditure growth: i) the estimated potential growth of the economy in real time; ii) the future average potential growth estimated in real time for a given period; iii) the average forecast of real GDP growth for a future given period which, due to economic convergence, should be lower than current growth in normal times; iv) the forecast of real GDP growth rate adjusted by a margin of prudence in policy making as for example was the case in the Netherlands until recently (Beetsma 2010) and; vi) the average long-term growth of GDP as estimated by the Working Group of the Economic Policy Committee and the European Commission every three years. In general benchmarking policy based on the concept of prudent rate of economic growth has clear advantages than for example when based on the concept of structural balance. It simplifies the assessment of policy in real time and it is more transparent. Nevertheless, choosing the "right" measure used as a proxy of prudent expenditure growth is subject to discretion. With the purpose of assessing past policy against the principle of prudent fiscal policy making the following measures are used: i) actual nominal rates of GDP growth; ii) nominal potential growth estimated using the average annual inflation rate and potential output based on real time data (2007) and ex-post data (2010); iii) nominal potential growth estimated using a 2% average inflation rate and potential growth rate based on real time data (2007) and ex-post data (2010); iv) nominal potential growth estimated using a 2% average inflation rate and potential growth rate based on real time data (2007) and ex-post data (2010); v) nominal potential growth estimated using a 2% average inflation rate and the average potential growth rate for the period 2000-2008 based on real time data (2007) and expost data (2010) and; vi) nominal potential growth estimated using the change in GDP deflator and the average potential growth rate for the period 2000-2008 based on real time data (2007) and ex-post data (2010). The comparison between the nominal growth rates of general government expenditure and GDP in the period 2000-2008 indicates that the expenditure growth rate was lower than that of GDP with the exception of the years 2000-2001 and 2008 (Figure 13). As a consequence, the expenditure-to-GDP ratio declined in the period 2002-2007. Using the nominal growth of potential output, calculated based on the average annual inflation rate in the period under consideration and the annual potential output growth rate (based on real time data (Potential 1 (Apr.07)) and ex-post data (Potential 1 (Apr.10)) as a proxy of prudent rate of expenditure growth, and comparing it with actual expenditure growth indicates that the actual expenditure growth rate lagged behind nominal potential growth rate in all years with the exception of 2001 and 2008 (Figure 13)6. The strong expenditure growth in 2001 and 2008 was due to large wage increases. Based on these two estimates it could be concluded that with the exception of 2001and 2008 policy was broadly prudent resulting in a decline of expenditure-to-GDP ratio and gradual closing of the deficit. Another issue is whether fiscal policy should have been even more prudent which in turn depends on policy objectives pursued by incumbent government at that time. Some of the proposed indicators used as a proxy of prudent expenditure growth are more demanding (Figure 14). This is the case of using a nominal potential growth rate (Potential 2) estimated by adding to the annual potential growth rate a 2% average inflation rate (the ECB inflation target) or by using the growth rate of nominal potential growth (Potential 3) estimated by adding to the average potential growth rate (2000-2008) a 2% annual average inflation rate (the ECB inflation target) in Figure 15. A less stringent indicators of the notion of a prudent expenditure growth include the growth rate of a nominal potential output (Potential 4) estimated by adding to the average potential growth rate (2000-2008) the annual change in GDP deflator (Figure 16). When expenditure growth is benchmarked against the nominal potential growth using the change in annual GDP deflator and average potential growth rate (Potential 4), the comparison indicates (Figure 17) that expenditure growth rate was broadly above the prudent rate of growth only in the years 2001 and 2008 and also depending whether real time of ex-post data is used. Notice that Figures 13a trough 13d show the various measures of rate of prudent expenditure growth estimates with real and ex-post data and actual expenditure dynamics. In general the prudent expenditure growth rate estimated based on ex-post data is more demanding. Figures 14 and 15 provide a similar conclusion. The years in which expenditure growth was above a prudent rate of growth were during 2000-2003 and 2008 and this depends upon whether potential growth is estimated with real time or ex-post data. The comparison indicates that when an arbitrary rate of inflation is used in estimating the prudent expenditure growth (Figures 14 and 15) and for example not a rate that has economic meaning, such as the inflation rate corresponding to real convergence (Balassa-Samuelson) or one which includes the effects of external shocks (oil, food and taxes), the resulting prudent growth rate is more stringent 6 Potential output rate growth estimates correspond to those of EU Commission made in April 2007 and April 2010. and if used in policy formulation it can have important consequences in terms of distorting budget allocations as discussed below. Figure 13: Expenditure growth and nominal potential GDP growth - Figure 14: Expenditure growth and nominal potential GDP growth ■ Average annual inflation 2% inflation Expenditure Growth Nominä GDP growth Pctentiä 1 (Ap-.07) 1 (Apr. 10) Expenditure Growth Nominä GDP growth Potentiä 2 (Apr.07) Potentiä 2 (Apf.10) 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2001 2002 2003 2004 2005 2006 2007 2008 Figure 15: Expenditure growth and nominal potential GDP growth - Figure 16: Expenditure growth and nominal potential GDP growth ■ Average potential growth 2002-2008 and 2% inflation Average potential growth 2002-2008 and change in GDP deflator 2000 2001 2002 2003 2004 2005 2006 2007 Source: SORS and Ministry of Finance, own calculation 2000 2001 2002 2003 2004 2005 2006 2007 One issue is to assess past policy against a benchmark of what can be regarded as a prudent policy a different one is to use the benchmark for policy formulation as for example in the case of a fiscal rule. Figure 17 shows the different impact on the expenditure-to-GDP ratio if the various alternative measures of prudent rate of expenditure growth (Potential 1 to 4) would have been used as fiscal rules in the past. Figure 14 shows that in the case in which the average inflation rate of 2% would have been used as part of potential growth estimates (Potential 2) then the expenditure ratio would have fallen significantly. The ratio in 2008 would have ranged between 36% and 38% which is lower than the actual ratios in 2007 (42%) or in 2008 (44%). The reduction of the expenditure ratio to the level between 36% and 38% would have implied a reduction in real terms of some entitlements and/or the cut of certain expenditures highlighting the importance of determining 12,0 10,0 policy priorities and expenditure level before determining any expenditure rule. Alternatively, such an approach if not accompanied by clearly defined policies can lead to credibility losses. Figure 17: Expenditure to GDP ratio, in % I I Actual ratio -----------Potential 2 (Apr.07) -------- Potential 3 (Avg.Apr 10) 50,0 = 48,0 46,0 44,0 42,0 40,0 38,0 36,0 34,0 32,0 30,0 2000 2001 2002 Source: SORS and MF, own calculation .......... Potential 1 (Apr.07) t— Potential 2 (Apr.10) ■ Potential 4 (Avg.Apr 07) ..—— Potential 1 (Apr.10) -■-•■•-■ Potential 3 (Avg.Apr 07) Potential 4 (Avg.Apr 10) 2003 2004 2005 2006 2007 2008 Figure 17 also shows the effect on the expenditure ratio if expenditure dynamics would have been determined by expenditure policy rules based on the alternative measures of potential growth discussed above. In the case where the expenditure rule would have been calculated based on potential growth estimated using annual inflation (Potential 1) and annual change in GDP deflator (Potential 4) and when these estimates would have been made in real time (April 2007) the results indicate that the expenditure ratio would have been higher than the actual one. In the case where such rules are estimated with ex-post data the actual expenditure ratio would be also lower with the exception of 2008. These simulations suggest that expenditure policy did not compensate for GDP growth particularly when economic growth was high (2005-2007) and the policy and its interaction with economic developments delivered a declining expenditure ratio from 2002 until 2007. Combining information of structural balance estimates in real time (Table 1) and expenditure ratios (Figure 17) suggests that the actual expenditure growth rate was consistent with a prudent expenditure growth rate estimated based on GDP deflator and potential output (Potential 4). Thus, based on real time data information indicates that there was consistency between the targeted structural balance and expenditure dynamics. The expenditure reduction contributed to meet the MTO in real time. However, if account is taken of recent output gap estimates (April 2010), which due to the crisis have been revised significantly (Figure 4), it follows that the expenditure growth rate needed to keep a fiscal stance close to a structural balance target of a deficit of 1% of GDP (Figure 18 and 19) should have been lower than actual rates in the years 2001 and 2006-2008. Notice in Figure 18 that during the period with relatively low negative output gaps (2001-2004) and with the exception of 2001 the actual expenditure growth rate was similar than that consistent to keep a 1% of GDP structural deficit target. In 2005 when the output gap turned positive and when the output gap became widely open (2006-2008) the expenditure growth exceeded that of the rate of growth required to keep the structural deficit close to 1% of GDP. Figure 18: Government expenditure (% change), output gap and Figure 19: Government balance's actual (% GDP) and structural (% GPD growth (%) potential growth) 16,0 14,0 _ 2 Expenditure (actual) ■ Expenditure (required) - GDP growth - Output GAP (RHS) 2001 2002 2003 2004 2005 2006 2007 2008 Source: Ministry of Finance, SORS, own calculation Figure 20: Structural balance's composition (% GDP) ■ Balance (Required) ■ Balance (actual) •Structural balance (Apr.10) ^Structural balance (required) 59 -1,0 ..... • ..... J ......... 2001 2002 2003 2004 2005 2006 2007 2008 Figure 21: Expenditure to GDP ratio, in % • Expenditure (Actual ratio) • Expenditure (Required ratio) J_I_^_I_L 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2001 2002 2003 2004 2005 2006 2007 2008 Source: Ministry of Finance, SORS, own calculation Figure 21 also shows that under ex-post estimates of output gap the policy required to meet the MTO by means of expenditure reduction alone would have resulted in lowering the expenditure-to-GDP ratio from 46% of GDP in 2000 to 40% in 2008 or it should have required that the ratio in 2008 would have been lower than the actual one by 4 percentage points of GDP. The strongest expenditure reduction would have to have taken place in 2006 and 2007 (i.e. to offset cyclical fluctuations (Figure 21)). Engineering such a strong expenditure reduction (4% of GDP) in the short-term beyond that planned in the budget with a lower forecast of GDP growth rates used in the budget formulation than actual rates (Figure 5 and 6) seems quite challenging. This suggests that given the uncertainty of output gap estimates targeting a structural balance might require an ex-ante relative large surplus. The comparison between fiscal stance based on real-time and ex-post data also 2,0 12,0 1,0 10,0 0,0 -2,0 -5,0 2,0 -6,0 highlights that in order to avoid distortions in terms of expenditure compression in case of sudden and deep changes in business cycle it is required that expenditure priorities are determined on a long-term perspective and on a sustainable basis. 2.1.4 The strength of countercyclical fiscal policy The economic developments propelling and resulting in the grate recession and the discrepancy in structural balances estimated with real time and ex-post data highlights the challenge of adherence to a neutral policy stance of 1% of GDP structural deficit. This challenge is also significant under the assumption of perfect foresight as it is difficult to enact a reduction of expenditure-to-GDP ratio in the magnitude of 4 percentage point of GDP in two years. The underlying issue is whether policy can react in the short run to developments in economic activity in order of not only keeping a neutral fiscal stance (defined in structural terms against the cycle) but also to effectively counter act the effect of the economic cycle and shocks. Answering this question requires understanding the availability of information on economic activity in real time and how it is internalized in policy formulation by following the process of preparation, approval and changes to the adopted budget. In Slovenia every year on a rolling over basis two-year ahead budgets are approved. In autumn of year t-1 fiscal authorities approve the budget for year t and t+1. Two-year budgets are designed on the basis of ex-ante projections on the future state of the economy. The forecast used in planning budgets correspond to spring forecast elaborated in year t-1. The budgets can be changed by means of supplementary budget contingent to changes in macroeconomic conditions or when the government incurs in new commitments7. Before the introduction of a new framework determining expenditure ceilings (See section 3) changes to the t+1 budgets were allowed before budget execution by means of budget revision. Relevant information pertaining quarterly GDP developments in the current year where the budget is implemented and which can be used to justify changes in the fiscal course are released by the Statistical Office (SURS) as follows: for the first quarter on May 31s'; for the second quarter on August 31s'; for the third quarter on November 30'h and; for the fourth quarter on February 28'h. This schedule of release was changed in 2009 and the dates of release were shortened by 10 days. This implies that unless there are strong signals that demand revision in the forecast like those in early 2009 due to the crisis, changes to the budget based on latest GDP data for the first quarter of the year in which the budget is executed can be made staring only in June when half of a year of budget execution has elapsed8. At that point the issue is whether fiscal policy can effectively counteract business cycle movements or shocks in the very short run as realization of fiscal measures depends importantly on implementations lags. Therefore, when assessing fiscal stance and its possibilities it is important to look at information available in real time and how this information can be used in policy formulation or to change course of action. The interplay between availability of information in real time, budget formulation process and the likely effect of fiscal policy to offset business cycle fluctuation, the later subject to uncertainty due openness of the economy, monetary integration and behaviour of households, highlights the importance of two issues: i) the quality of information in real time and of the unbiased nature of forecast used in policy formulation (addressed 7 According to existing Law on Public Finance. 8 On average in past years about 48% of budget appropriations are executed by the month of June. The year 2010 has been the exception where 67% of general government budget was executed until the end of June (cash basis). in Section 2) and; ii) the need of pursuing a prudent fiscal policy (expenditure and revenue) which is anchored in set of stable expenditure policy priorities over the medium term. The assessment of the strength of fiscal policy in counteracting business cycle fluctuations and affecting economic activity, performed using the EU Commission DSGM model QUESTIII calibrated for Slovenia (IMAD), suggest that it is not significant (Figure 22). Simulations of countercyclical policy in good times involving a one-off adjustment in three fiscal instruments (reduction in government consumption, reduction in government investment, and increase in VAT rate) equivalent to 1% of GDP on GDP indicate that the strongest impact the year when the measure is implemented results from a reduction in investment (0.31% of GDP) followed by a reduction in government consumption (0.26% of GDP) and increase in VAT (0.06% of GDP). The effect of the measures last the following years and reverses in the case of government consumption (fifth year) and VAT rate (10 year) while in the case of investment tends to be more permanent. Figure 19 also shows the effect of an exogenous increase in capital cost of the economy by 50 basis points which have long-lasting effects. In an institutional set up like the one in Slovenia, where there is not independent monetary policy, the issue is not only the fiscal impact but the net effect on economic activity arising from the interaction of different macroeconomic variables. For example, when exogenous changes in interest rates move in different direction than desired fiscal policy intentions (the issue becomes more complicated if account is taken of the reaction of households and private agents). Figure 22: Effect of one-off change (1% of GDP) in policy instrument (tightening) 0,10 0,05 0,00 -0,05 -0,10 -0,15 -0,20 -0,25 -0,30 -0,35 -0,40 Source: own calculation Government consumption Government investment Increase in VAT Exogenous increase in cost of capital (0.5 p.p.) 5 years 6 7 8 9 10 2.1.5 Taxation policy and fiscal performance Government expenditure is a key variable in explaining and underpinning fiscal policy developments the other one is government revenue, particularly tax policy. Even when the fiscal policy aim is the adherence to given expenditure ceilings, if there is no consistent tax policy with those ceilings and underlying fiscal targets or tax changes (carried out or planned) are no revenue neutral or accompanied by expenditure reduction in the case of tax burden reduction, then the fiscal position and its sustainability is affected. In particular there is evidence on the pro-cyclical nature of discretionary measures affecting tax revenues whereby governments tend to 0 2 3 4 implement tax cuts during expansionary phases while resorting to tax increases during slowdowns (Barrios and Fargnoli, 2010). In Slovenia the revenue dynamics from tax and social security contributions has been underpinned by that of the business cycle but also by discretionary policy changes. Figure 20 shows that total revenue from tax and social security contributions has moved closer to the annual change in GDP (correlation 96%) than to the output gap as for example calculated on April 2010 (correlation 58%). The fact that output gap estimates are subject to great uncertainty also affect the estimates of tax revenue adjusted by business cycle (cyclically adjusted tax revenue). Figure 24 shows that during the period 2005-2008 there was a strong revenue growth in VAT, personal and corporate income tax (2005-2007) associated mainly with strong economic performance. However, tax revenue was also affected by policy changes. Over the past ten years there were incessant policy changes affecting primarily the tax system. Ranking the effect of tax changes according to their magnitude on fiscal revenue indicates that among the most important tax changes ordered chronologically were: i) the increase in VAT rate of 2002; ii) the transfer of income revenue from custom duties to the EU budget in 2004; iii) change to corporate income tax in 2004/2005; iv) changes to personal income tax (introduction of five brackets and lowering tax burden on low income earners in 2005); v) the tax reform package of 2006-2007 (including the elimination of payroll tax, introduction of a dual personal income tax system and changes to CIT) and; vi) increases of excise duties in 2004 and 2009 (the first related to change in income duty taxation due to EU membership and the latter to offset elimination of the last tranche of payroll tax). These changes had different rationale and their impact on the size of government revenue has been either positive or negative. Tax policy changes can be classified with regard to the purpose they were made between those aiming at structural change and those at offsetting adverse revenue developments (usually change in tax rates are not used as discretionary countercyclical tool in good times). With regard to those changes pursuing structural impact the issues are its timing and financing. Among tax changes targeted to offset negative business cycle development (2002 and 2009) or the effect of structural changes in revenue (2004 transferring custom duty revenue to EU budget) are the increase in VAT rate in 2002 and increase in excise duties rates in 2004 and 2009. Among the important changes in tax system aiming at changes in its structure are the changes in personal income tax and corporate income tax that took effect in 2005 and the reform in 2006-2007 which included the elimination of the so-called payroll tax, transformed the personal income tax in a dual system and introduced changes to the corporate income tax. Taking into account output gap estimates of April 2010 (Figure 4), it can be said that the 2005-2007 tax reform took place at the time when the economy underwent a strong expansion (starting in 2005) culminating in 2007. The reform of personal income tax in 2005 was financed by changes in corporate income tax and from changes in accounting standards affecting corporations and was revenue neutral (Figure 24). This was not the case of the reform in 2006-2007 which coincided with strong revenue performance due to the cycle and underlying developments in the economy (strong expansion financial and construction sectors). Although the possibility of raising the VAT rate was conceived to finance the tax reform in 2006, this was not the case due to revenue buoyancy that offset temporarily the revenue loss and as a consequence the reform was not structurally revenue neutral (tax reductions offset by increase in other taxes). Since expenditure was neither reduced in 2008, but increased at the time when economy entered into recession (last quarter of 2008), the adverse effect of the tax reform on revenue and fiscal position was not offset and was magnified by the fall in economic activity. Figures 25 and 26 illustrate the impact of the 2006/2007 reform on tax revenue and government balance assuming that expenditure would have followed the expenditure scenario envisaged in the Stability Program in 2007 (SP2007). Figure 25 shows that under no reform scenario revenue would have been higher by 2% of GDP in 2010. Figure 26 shows that in absence of tax reform and the crisis the budget would have been in surplus in 2009 and; when taking into account the crisis, the budget would have registered a lower deficit around 4% of GDP. More importantly, the recent experience indicates that when planning revenue without adjusting for the effect of cyclical fluctuations this can lead to pro-cyclical action in terms of tax reform and increase or plan high level of expenditure (See Figure 2). Such developments highlight the importance of ensuring in advance the financing of tax changes/reform and to take into account, to the possible extent, cyclical conditions in planning tax reform and expenditure. In particular, a planned revenue decrease in the magnitude of 1% of GDP due tax changes in good times tends to result in higher loss of revenue in bad times because the overall revenue is affected by cyclical conditions. Higher planned revenue can also lead to higher expenditure level. Therefore, when planning revenue attention should be given to the effect of business cycle and proceed prudently with the revenue forecast. Figure 23: Tax revenue and GDP growth and output gap —!—!—r • Tax revenue and Soc. Sec. • GDP • Output gap (April 10) Figure 24: Contribution to tax revenue growth (p.p.) and GDP growth (%) Value added taxes (VAT) Taxes and duties on impotts Personal income tax Corporate income tax Excse duties 2000 2001 2002 2003 2004 2005 2006 2007 2008 2001 2002 2003 2004 2005 2006 2007 2008 2009 Figure 25: Tax revenue under tax reform and no reform scenario, in Figure 26: Government balance under tax reform and no reform bn. EUR ' Tax revenue (SP 07) no reform Tax revenue (SP 07) Tax revenue actual (no reform) Tax revenue actual scenario, in % of GDP \m IT □ Balance SP 07 (no reform) ■ Balance SP 07 ■ Actual Balance □ Actual balance (no reform) n 1 F 2005 2006 2007 2008 2009 2010 2008 2009 Source: Ministry of Finance, own calculation 20,0 15,0 10,0 L 5,0 0,0 -5,0 -10,0 11,0 1,0 10,0 1,0 9,0 -2,0 -3,0 8,0 4,0 5,0 7,0 -6,0 -7,0 6,0 While elected governments decisions with regard tax changes and their timing should be not constrained such changes should be framed within a medium term strategy to reduce the adverse affect on sustainability of public finances. 2.1.6 Expenditure policy and fiscal performance With respect to government expenditure, a more detailed analysis of its composition and the policies underpinning its dynamics also contributes to draw lessons for shaping the appropriate policy framework. Figure 27 shows the annual change of total expenditure and the contribution of various categories based on economic classification. It shows that the decrease in total expenditure growth in the period 2001 to 2005 and its stabilization until 2007 was underpinned by wage and social transfers' policies. The wage contribution to total expenditure growth declined from 4 p.p. in 2001 to 1.7 p.p. in 2007 and the contribution of social transfers was reduced from 4.6 p.p. in 2001 to 1.8 p.p. in 2007. The total contribution of both types of expenditures to total expenditure decreased from 8.6 p.p. in 2001 to 3.5 p.p. in 2007. As a consequence of these developments and lower interest payments, the ratio of expenditure-to-GDP declined in 7 years by 5.2 p.p. from 47.6% in 2001 to 42.4% in 2007 (Figure 26). In 2008 the trend decline reversed and the ratio sharply increased to 49% of GDP in 2009 due primarily to drop in GDP activity. The wage dynamics underscoring expenditure reduction resulted primarily from a reform policy aiming at eliminating wage disparities among public sector employees (2002-2008) but also by a deliberate effort to reduce its dynamics (2004). As a consequence of the lengthy process of implementation of the law to correct wage disparities in the public sector, the real wage growth per employee in the public sector was lower by 273% than that in the private sector or the average rate was lower by 2.1 p.p. in the period 2002-2007. The wage difference increased substantially in 2004 when wage growth in public sector turned negative in real terms. In this period private sector wage growth lagged productivity growth by 32%. As regard to the fiscal implications of the law, its effect would have been neutral on public finances under the assumption of a permanent nominal wage reduction in the salaries of high wage earners (i.e. wage compression). This did not happen because the delayed implementation of the law exacerbated the need of wage adjustment (7 years). The delayed implementation of the law contributed to fiscal consolidation during 2002-2007. Its negative impact on wage developments in different sectors was uneven like the health sector which registered positive real wage growth only in 2003 and 2006. In the same period the number of employees in public sector increased. The combination of wage policy and the increase in number of employees resulted in a lower wage per employee as percentage of GDP. As a result of the implementation of the Law an upward level shift in wage bill took place in 2008 and partly in 2009. Despite these increases, the share of wages in total government expenditure remained constant even after the increase in wages in 2008 (Figure 28). Going forward it is crucial for whatever fiscal framework to maintain a wage policy which is sustainable and does not result in shocks and efficiency losses. This implies that after the wage freeze envisaged during the current process of fiscal consolidation the government or policy makers should have to agree on a wage policy which at least keep real wage constant to avoid shocking public finances like in 2001 and 2008. The longer a correction in wages takes place due to a wage freeze or contention the probability of a shock becomes higher. Figure 27: Government expenditure, annual change contribution to change, in % and Figure 28: Government expenditure, structure, in % of GDP 3 Other expendture 3 Subsidies 14,0 12,0 10,0 8,0 ° 6,0 4,0 2,0 3 Intermedate ccnsumpticn s fxed capita formation and K transfers I Social transfers (Total_ • Social transfers (Total) •Compensation of employees • Intermediate consumption - Gross fixed capital formation and K transfers Other expenditure m....... 2001 2002 2003 2004 2005 2006 2007 2008 2009 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Figure 29: Government expenditure, structure, in % of GDP ■ Gross fixed capital formation I Capital transfers • Total % GDP (rhs) Figure 30: Government investment, in bn. EUR ■Other expenditure ainterest expenditure ■Subsidies □Gross fixed capital formation and K transfers ■ Intermediate consumption ■ Com pensati on of empi oyees ■Social transfers (Total) 0,^ 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 55,0 50,0 45,0 40,0 ^35,0 30,0 25,0 20,0 15,0 10,0 5,0 0,0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: Ministry of Finance The other important component contributing to expenditure reduction were social transfers (Figures 27 and 28). The dynamic of transfers was influenced by lowering inflation, improving in economic conditions, changes in indexation of social transfers (temporary in the case of the pension formula) and unification of their indexation. In 2008 the trend reduction in growth of social transfers reversed by the granting of additional transfers to families which resulted in a level increase in social transfers. Permanent changes in social rights without being accompanied by permanent expenditure offsetting measures or by increase in taxes and similarly of granting tax allowances without offsetting measures also called for attention in the design of a fiscal framework. Another important contributor to the decrease in expenditure was the lowering of interest payments in nominal terms in the period 2002-2008 resulting in a decrease in debt service as percentage of GDP from 2.4% in 16,0 0,0 2,0 5,0 ^,0 0,0 60,0 1,5 1,0 1,0 2001 to 1.1 % in 2008. This was facilitated by the change in composition of debt portfolio away from indexed debt and the repayment of debt in 2007. Among the elements that dampened the reduction in government expenditure on the other hand, was the trend increase in gross capital formation in the period 2006-2008 (Figure 27 and 29). The expenditure in gross fixed capital formation and capital transfers was higher by 1.8% of GDP in 2008 than the average expenditure in investment-to-GDP ratio during the period 2000-2006. Such a level shift in the ratio of investments-to-GDP raises the issues of why the ratio was not higher in the past and whether the increase reflects deficit bias as it is always possible to justify investment expenditure on the grounds of its effect on economic growth. While this argument might be valid depending on investment allocation the issue is the investment's financing and the effect on fiscal variables. For example, if the investment-to-GDP ratio in 2008 would have remained at the average level of 2000-2007, then the budget instead of registering a deficit in that year would have shown a balance position and the debt level would have been lower. A similar argument could be made with regard to investment levels envisaged in the period 2010-2012. In particular deficits on average could be lower by 1.5% of GDP than those envisaged and, if a substitution of sources of investment financing from domestic sources by EU funds takes place, then the level of investment could even remain constant. Justifying high investment levels when the economy is undergoing economic recovery is easier. However, its level has to be assessed in light of impact on risks the economy faces (e.g. risk of downgrading the sovereign rating) and alternative EU sources of financing. Therefore, the issue of investment is important when considering the design of a fiscal framework as it touches upon the decisions of democratic elected governments regarding its priorities and its medium term impact on public finances. For example, whether investment should be part of an expenditure rule constraining the overall expenditure level or should be excluded as for example the so-called golden rule. In any case given the fact that investment financing affects deficit and debt dynamics its level should be consistent with fiscal targets and macroeconomic priorities which are better addressed within a medium term strategy. 2.1.7 EU funds and fiscal stance An important issue concerning government investment (capital expenditure and investment transfers) is the possibility of its financing with EU budget sources which depends on the institutional capacity for withdrawing funds and for which a framework of accountability should be enhanced. From the point of view of its fiscal impact, investment financed with EU funds does not affect directly public finances' outcomes (i.e. deficit and debt) as expenditure is offset by revenue transfers from the EU budget. However, it does it indirectly to the extent that co-financing is required to withdraw funds and to the extent that substitution of investment financing of domestic sources by EU funding is pursued. It also affects indirectly government revenue by means of the effect of investment in potential growth of the economy (Varga and in't Veld, 2010). In Slovenia, the share of investment financed with EU sources in total general government investment at the end of 2009 was 20% and the 2010 supplementary budget (September 2010) and adopted budgets for 20112012 (November 2010) envisaged an increase in the share of EU financing particularly in 2010 (36% of total investment). Given the uncertainty in the dynamics of withdrawing EU funds or over-estimation of funds withdrawal capacity, adopted budgets tend to overestimate investment expenditure and as such the total expenditure level. Only in 2010 the supplementary budget adopted in September reduced total investment expenditure by €390 millions (mainly capital expenditure) and the estimated outturn (January 2011) of EU funds withdrawn in 2010 was even lower than envisaged in the supplementary budget by €314 millions. This implies total lower investment expenditure with respect to that in the adopted budgets of 1.9% of GDP. This raises issues of accountability and the need to address this issue within a robust policy framework (i.e. enhancing planning or capacity of designing projects). To underline the interrelation between investment dynamics, it's financing and impact on public finances' outcomes, figures 31 and 32 show the evolution and estimation of investment at the level of the central government for the period 2003-2013 (2003-2009 (outturn), 2010 (supplementary budget), 2011-2012 (adopted budget) and 2013 (forecast)). Figure 31 shows that investment has and it is planned to increase. While both capital expenditure and transfers have increased, the increase in capital transfers has been larger and it share in total investment is planned to remain above 50%. Capital expenditure dynamics seem to be broadly following the political cycle (2005-2008 and 2008-2012). Investment outlays are low at the beginning of a new government period and the raise towards the end of a government mandate (Figure 33). The share of EU financing in total investment financing of the central government has increased in the past years (Figure 34). It reached about 30% in 2009 (20% of total general government investment) and it is planned to be close to 50% in the future (30% of total general government investment). This implies that the share of domestic sources' co-financing has increased and would have to do so in the future. To accommodate increases in co-financing needs without affecting public finances, the share of investment priorities directly financed with domestic sources would have to decrease or alternative sources of financing should have to be provided. A more detailed look at the components of investment (i.e. capital expenditure and transfers) and their financing indicates that while the level of capital expenditure financed with domestic sources has doubled since 2007 (Figures 35 and 37) and broadly follows the political cycle (Figure 35), this has not been the case of capital transfers that remained around € 300 million since 2003 (Figure 37). The practically doubling of the increase in size of capital transfers since 2008 has been financed with EU funds (Figure 38). Figure 31: Total investment, in mio EUR Figure 32: Total investment, structure in % ■ CAPITAL TRANSFERS ICAPITAL EXPENDITURE 2003 2004 2005 2006 2007 2008 2009 2010B2011 B 2012B 2013F 2003 2004 2005 2006 2007 2009 2010B 2011 B 2012B 2013F Figure 33: Investment financing, in mio EUR 1.600 Figure 34: Investment financing, structure in % lEU sources iDomestic sources 2003 2004 2005 2006 2007 2008 2009 2010B2011 B2012B 2013F Figure 35: Capital expenditure, in mio EUR 2003 2004 2005 2006 2007 2008 2009 2010B2011 B2012B 2013F Figure 36: Capital expenditure, structure in % 2003 2004 2005 2006 2007 2008 2009 2010B2011 B2012B 2013F 2003 2004 2005 2006 2007 ■ 2010B 2011 B 2012B 2013F Figure 37: Capital transfers, in mio EUR Figure 38: Capital transfers, structure in % ■ EU sources ■ Domestic sources 2009 2010B2011 B2012B2013F 2003 2004 2005 2006 2007 2008 2009 2010B2011 B2012B 2013F Source: Ministry of Finance 2.1.8 Behaviour of expenditure entailments during the cycle A related topic to investment dynamics is the overall cyclical behaviour of government expenditure (political and economic). This issue is relevant since many arguments to constrain policy makers' decisions by means for example of a fiscal rule is that expenditure in election years or within a political cycle (e.g. strong investment expenditure) tends to be politically motivated and pro-cyclical in economic sense. As mentioned before, the government expenditure growth rate with the exception of 2001 and 2008 was below of that of GDP or to what can be regarded as a prudent rate of economic growth (Figures 13 and 16) and, as a consequence, the deficit and debt as percentage of GDP declined. Expenditure growth surpassed that of GDP in 2001 and 2008 which both were election years, but in both years a key driver was the increase in public sector wage. On the other hand, this was not the case in the 2004 which was also an election year partly explained by the two-year budget approval system that anchored expenditure dynamics and because wage policy that culminated in wage adjustment in 2008 was still under negotiation at that time. While in 2008 social transfers increased (Figures 27 and 49), an important element that contributed to that was the indexation of pensions due to also increase in wages of public sector. Thus, while investment tends to follow the political cycle and one-off type of increase in social transfers tends to happen in election years, the main source of instability and stability affecting expenditure is wage policy. Therefore, improvements to the fiscal framework should deal primarily with sustainable wage policy. Attempts to curve expenditure based on expenditure rule would be ineffective without a long-term and sustainable wage policy. This argument also applies to other expenditure entitlements. Without structural reform of entitlements and only temporary reliance of expenditure compression such policy would lead to shocks to public finances and breaching of any ceiling or constraint. The assessment of role of fiscal stance with regard to the business cycle as discussed above indicates that to a large extent depends on the set of information used for analysis (ex-ante or ex-post information). It was also pointed out that expenditure dynamics followed a prudent rate of growth mainly with the exception of 2001 and 2008. The analysis can be complemented by benchmarking expenditure developments with respect to the output gap. In particular, to determine whether expenditure policy is pro or counter-cyclical, the nominal growth of government expenditure is benchmarked against the estimates of output gap (European Commission) available to policy makers in real time (relevant for policy formulation) and also based on expost data. Figure 39 shows the output gap estimated in November 2006 (OG 2006)) which coincided with budget approval for the years 2007 and 2008 and which could have been used for deciding on countercyclical expenditure policy and ex-post output gap estimates made in April 2010 (OG 2010). The first observation is that the output gap estimated with real time data is substantially smaller than that with ex-post data (20062009). In both cases in 2001 and 2008 the policy is pro-cyclical particularly in 2008. In the period 2002-2005 expenditure appears counter-cyclical (neutral until 2007 based on real time information) and during 2006-2008 pro-cyclical based on ex-post information. Combining the dynamics of some key expenditure categories and ex-post estimates of output gap (OG 2010), it is possible to observe that the economic and political cycles interact as follows: i) wage policy adjustments coincided with positive output gaps particularly in 2008, but was also pro-cyclical in 2004-2007 (Fiscal consolidation policy free rode on wage policy; Figure 40); ii) investment tends to follow a political cycle which its effect on the cycle can or not be counter-cyclical (Figure 41) and; iii) social transfers tend to be associated with political cycle although low inflation dynamics in 2004-2007 and positive economic performance in 20062007 contributed in lowering its relative size (Figure 42). These observations highlight the importance of a medium-term strategy framing annual budgets as discussed in Section 4. Figure 39: Output gap and total government expenditure growth Figure 40: Output gap and compensation of employees' growth rate rate -OG 2010 (LHS) - OG 2006 (LHS) total expenditure growth -5,0 0,0 -10,0 -2,0 -15,0 -4,0 OG 2010 (LHS) -Compensation of employees r r - - - - 1996 1998 2000 2002 2004 2006 2000 2002 2004 2006 2008 Figure 41: Output gap and gross fixed capital formation's growth Figure 42: Output gap and social transfers growth rate rate 10,0 ü 2000 2002 2004 2006 2008 2000 2002 2004 2006 Source: European Commission and Ministry of Finance, own calculation 2.1.9 Debt developments and fiscal stability As mentioned before the possibility of fiscal policy to influence macroeconomic developments and its impact in the short run in Slovenia is constrained not only due to the structural characteristics of the economy (open, small and without independent monetary policy) but also due to availability of information to policy makers in real time and implementation lags. Also in the recent past, with exception of 2009, there were no major revisions in GDP or macroeconomic forecast used for budget preparation that would have required in advance sizable fiscal interventions to offset macro developments (Figure 43). Notwithstanding, these forecasts an episode of inflation outburst and current account deficit widening took place in 2007 and 2008 which might have argued for fiscal intervention. However, the issue is whether fiscal policy would have been able to offset such developments under the constraints mentioned before beyond the use of automatic stabilizers. 8,0 6,0 5,0 8,0 £ ^ 2,0 6,0 2,0 2,0 0,0 8,0 30,0 12,0 25,0 6,0 10,0 20,0 4,0 15,0 2,0 0,0 -2,0 -2,0 4,0 10,0 Figure 43: Spring forecast of GDP real growth rate, in % Figure 44: General government debt, in % of GDP 5,0 4,0 3,0 2,0 1,0 üS 0,0 -1,0 -2,0 -3,0 -4,0 -5,0 ■ 2004SF ■ 2005SF • 2006SF • 2007SF 2008SF 2009SF ■ 2010SF _L_ ■■■■■i..... R 2009 2010 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: IMAD, SORS While the relative strength of government expenditure or revenue to offset macroeconomic developments is debatable, the role of government debt in mitigating adverse shocks to the economy and minimizing risk is less questionable as it has been also underscored by the recent debt crisis affecting severely some euro area countries. In particular, higher debt is associated with higher risk premiums which in turn have spillovers to the overall borrowing costs of the economy. On the contrary, low levels of debt provide room for full operation of automatic stabilizers and discretionary policy in the case of unprecedented shocks. Before the crisis the government debt ratio in Slovenia fell from 27.9% in 2002 to 21.9% in 2008 (Figure 44). This was due to the interaction of debt management policy (changes in indexation of debt, debt repayments (2008)), strong economic activity influencing the size of the government balance and GDP and low interest rate environment. Such a relative low level of debt without doubt provided the economy in 2009 with the most powerful macroeconomic shock absorber. In fact the debt ratio is estimated to have jumped by 16.1 p.p. of GDP in two years 2009-2010 (Figure 44). In 2009 the debt ratio catered for the collapse in government revenue (1.4% of GDP) and tax revenue (1.6% of GDP), discretionary policy and injection of government deposits to domestic owned banks at the time when the EU interbank market practically closed. In absence of the relative low level of debt ratio the possibility of policy response to the crisis would have been smaller and the effect of the crisis more severe. The increase of government deposits alone with banks accounted for a rise in the debt ratio equivalent of 7.1% p.p. of GDP in 2009. The relative low debt at the time of independence in 1991 and again in 2008 provided the economy with the possibility of financing bank rehabilitation and enterprise restructuring in the post independence period and again to preserve the stability of the banking system in the aftermath of the crisis in 2009. The general government debt as percentage of GDP at the end of 2010 is estimated in 38% (SORS) which is so far the highest historical ratio but still ranks among the lowest among the EU.9 9 The lowest available figure for debt ratio was in 1994 (18.7% of GDP). 40,0 6,0 35,0 30,0 25,0 20,0 15,0 10,0 The relative low government debt-to-GDP ratio has been reflected in the cost of borrowing of the government and of the economy at large. This has been particularly important in the aftermath of the crisis where the Slovenia's government debt spread with respect to the benchmark (Bund) has not widened significantly and remained fairly stable (Figure 42). The low spread in turn underpins the borrowing cost of domestic banks and the economy. Therefore, based on this considerations and assessment of possibilities of effective use of fiscal policy as countercyclical tool, it can be concluded that the debt-to-GDP ratio should gain the status as the key policy target anchoring fiscal developments and macroeconomic stability as further discussed in Section 4. Figure 45: 10-year bond spread against benchmark German bund 12,0 10,0 8,0 6,0 4,0 2,0 0,0 ■ Greece ■ Portugal - Irland - Spain - Slovenia Source: Eurostat The relative size of the debt is important but also its structure, particularly the maturity profile given the fact that most of debt is denominated in national currency. Going forward, given the increase in government debt not only in Slovenia but also in the EU countries, the challenge is to maintain the relatively long-term debt maturity profile reached at the end of 2009 in order to avoid refinancing risk and temptation of issuing short-term debt to take temporary advantage of low interest cost. Table 4: Indicators local government finances 2006 2007 2008 2009 Government revenue (% GDP) 5.1 5.0 5.0 5.8 Deficit (% GDP) 0.1 0.1 0.6 0.6 Share in general government debt 2.8 3.2 4.2 4.1 Debt-to-GDP ratio (%) 0.8 0.7 0.9 1.5 Source: SORS and Ministry of Finance In the context of debt dynamics, a recent development that deserves attention is the increase in local government debt which has taken place despite of debt rule limiting local government indebtedness (Section 3) and the increase in local government revenue in real terms which contrast with the decrease in overall general government revenue. Table 4 shows that the share of local government debt in total general government debt is not very large (4.1%) but it has kept its relative size when the general government debt increased substantially in 2009. An important spillover effect of the local governments' increase in debt and interest payments is that in pursuing further increases in indebtedness capacity or reducing the debt burden they can increase local government revenue (e.g. taxes or increasing the price of services) affecting competitiveness of the economy. In the post crisis period, one of the most important challenges facing public finances is the low GDP growth rate, which is lower than the interest rate at which the government borrows implying an unsustainable debt accumulation process. Provided that the cost of borrowing, which is underpinned by fiscal policy adherence to consolidation objectives, remains at current levels and the economy rebounds as expected by 2012 then the debt sustainability condition will be met again. This highlights the importance of adherence to fiscal consolidation objectives. 2.2 Fiscal consolidation The immediate challenge fiscal policy faces is fiscal consolidation which can be defined as a sizable reduction in the deficit. From the policy point of view fiscal consolidation is different than the issue of deficit persistence (deficit bias). Fiscal consolidation requirement arises in the aftermath of a crisis and when reduction in expenditure or correction of the fiscal imbalance is inevitable and has to be internalized by the policymaker. Deficit persistence on the other hand takes place in good or normal times and implies policy makers' tolerance for the existence of deficit whose correction is not perceived as urgent. In the case of Slovenia, the rationale for the reduction of the deficit which is estimated in 5.6% of GDP at the end of 2010 does not only arise from domestic macroeconomic concerns (i.e. deficit persistence at high levels could jeopardized macroeconomic stability which under current international environment is highly penalized by the financial market) but also from legal commitment under the Stability and Growth Pact, which demands a reduction of the deficit below 3% of GDP by 2013 and, beyond by the requirement of reaching a self imposed medium term objective set formally in terms of structural deficit at the level of 1% of GDP. This implies that the fiscal consolidation during the next three years is also firmly anchored by the need to lower the government deficit below 3% of GDP. Furthermore, an additional incentive is the recent decision of Standard&Poor's credit rating agency (December 2010) to place the credit rating of Republic of Slovenia under negative watch pending its revision to the government implementation of its fiscal consolidation strategy as detailed in the Stability Program 2010 and in particularly of stabilizing the debt-to-GDP ratio. Thus fiscal consolidation in Slovenia is demanded by objective conditions (macroeconomic stability), legally bound by international commitments and aligned by the incentives or punitive action of an eventual downgrading of government debt with adverse macroeconomic implications for the whole economy. Such a set of policy constrains without doubt makes the need for additional constraints quite superfluous. What might be required is to place the envisaged adjustment process within a robust medium term policy strategy and enhance the policy framework to ensure that fiscal policy is designed with regard to stability principles. The fiscal strategy should include: a) fiscal targets and policy mix that hedge against a major fiscal risk (i.e. increase in borrowing conditions of the government) and mechanisms that ensure that they are met; b) a policy mix that redistributes the cost of adjustment equitable; c) structural measures that ensure long lasting consolidation; d) keep consistency with the macroeconomic stabilizing role that debt has played during the crisis; e) minimize past policy mistakes (wage and tax policy). The fiscal framework should be strengthened with regard to the principle of fiscal stability and of democratic accountability of the government and the parliament with respect to policy making as discussed in Section 4. Ensuring ownership, transparency and accountability is of outmost importance also at the time when important reforms to the EU fiscal framework are taking place. 2.3 Long term sustainability of public finances The issue of sustainability of public finances over the long-term implies dealing with contingent liabilities arising from population ageing and the formal obligation of the State to cover pension and health insurance of the population in the future. In dealing with these issues, EU countries adopted in 2001 the so-called three prong strategy consisting in reforming pension and health systems, reducing government debt (front loading future liabilities) and increasing employment and productivity. More recently with the purpose of enforcing the reduction of future contingent liabilities by means of debt reduction a methodology for frontloading the budgetary cost of ageing was developed and included in the Code of Conduct for the preparation of the Stability Program despite the fact that such an approach was not formally adopted by the ministries of finance in the ECOFIN. According to the methodology, the medium term objective, which in principle is set by each country according to its conditions, preferences and taking into account a safety margin for not breaching the 3% Maastricht deficit criterion, should include a fraction of the adjustment needed to cover the present value of the future increase in age-related government expenditure which is set at the minimum of 33%. This implies for Slovenia a more demanding medium term objective than the one currently set (structural deficit of 1% of GDP) and also a pre-defined policy mix between the elements of the three pronged strategy. The MTO calculated by the EU Commission for Slovenia taking into account 33% of frontloading of contingent liabilities, before the pension reform was adopted by the parliament in 2010, was a structural surplus of 0.75% of GDP. Assessing compliance with the new methodology in determining the MTO will not result in imposition of sanctions which is the case of non compliance with other elements of the fiscal surveillance framework (e.g. deficit criterion). Non-compliance will result in policy recommendation by the European Council regarding the degree of ambition with respect to the minimum benchmark of front loading contingent liabilities chosen by the Commission and a large number of EFC member countries. For Slovenia in particular, non compliance with the 33% frontloading target given the fact that the debt-to-GDP ratio is still below 60% will result in an assessment that will state that the MTO does not appear to take sufficiently into account the implicit liabilities related to ageing. The controversial issue in setting a minimum frontloading parameter of contingent liabilities is that it pre-empts government choices as regard to the combination of elements of the three-pronged strategy and implies imposing distributive choices by implicitly raising the level of taxation or reducing the level of expenditure.10 What is also important to highlight in the context of the debt crisis affecting some euro area countries is that markets tolerance for financing large fiscal gaps has narrowed significantly which in effect implies that it would be quite impossible to financing future liabilities arising from ageing of the population and thus structural reform will be inevitable. The issue at the end is that incumbent governments while addressing the problem they should be able to decide on the policy mix they will follow during its term in office, which underline the issue of ownership. To have an idea of the implications of the new framework for setting the MTO for the Slovenia's fiscal stance, the general government balance compatible with the MTO including a 33% front loading parameter of long-term contingent liabilities was estimated for the previous decade (Figure 46). The estimation indicates that the government balance would have to have registered surpluses for most of the years with the exception of 2009 and 2010. The surplus would have to have been around 4% of GDP in 2007 and 2008. In order to achieve 10 A complication also arises concerning frontloading as it might entail large surplus during economic upturns that could be challenged on political economic grounds. these targets the government would have to have chosen among the options of restraining expenditure, increasing revenue or combination of both policies. Figures 47 and 48 show the expenditure and revenue levels required to achieve the modified MTO (Figure 46). Figure 47 shows that under the assumption that government revenue would have been fixed (actual revenue) the expenditure ratio should not have exceeded 43% of GDP and in 2006 and 2007 the expenditure structure should have been flexible enough for allowing a large expenditure reduction in the magnitude of 1.7 p.p. of GDP in 2006 and 2.5 p.p. in 2007. Such large size of expenditure reductions over a short period of time are quite demanding to implement. Figure 46: Government balance including 33% of frontloading contingent liabilities, in % of GDP §§ ^^ Figure 47: Required expenditure effort to meet MTO target, in % of Figure 48: Required revenue effort to meet MTO target, in % of GDP GDP %43,0 ■ Actuä expenditure ■ Actuä revenue Required expenditure -I-i-i-12000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Actual expenditure Actual revenue Required revenue -1-1- 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: Ministry of finance, own calculation Figure 48 shows the required revenue-to-GDP ratio compatible with the actual expenditure-to-GDP ratio that would have been needed to meet the MTO including 33% frontloading parameter of contingent liabilities. It indicates that on average the revenue ratio should have been 47% of GDP and in 2007 and 2008 it should have had increased around 48% of GDP. This implies that the revenue ratio on average should have been higher by 3.7 p.p. of GDP in the whole period. 39,0 37,0 35,0 The policy implications that fallow from the previous exercise and that could be taken into account when enhancing the existing fiscal framework are the following: i) whatever medium term target chosen in terms of government structural balance as percentage of GDP it will be achieved on a more reliable basis when the revenue is the key variable anchoring the target and it should be ensure that its level is stable; ii) expenditure level should reflect stable priorities which should not be compressed or shocked by unsustainable policies or abrupt changes; iii) if a decision on frontloading a share of future contingent liabilities is taken by means of setting a more demanding MTO than that at current level, this would imply to have lower structural deficit or surplus which in both cases would demand higher level of revenue than expenditure or relative higher level of taxation which should be assessed taking into account current levels and the impact of fiscal consolidation: iv) deciding on front-loading contingent liabilities implies the choice of relative higher level of taxation (distributive choice) and on the policy mix among the elements of the three pronged strategy. Such decisions seem difficult to be outsourced from elected officials accountable to its constituency to a rule or to an independent institution deciding on these choices implying the need to enhance the existing policy framework (Section 4) and; v) while there is not optimal level of debt and thus optimal fiscal policy then a robust fiscal framework in Slovenia should contribute to fiscal policy in delivering macroeconomic stability (i.e. not affecting adversely the cost of funds and provide room for offsetting unexpected shocks including contingent liabilities) and thus contributing to sustainable growth.11 11 Notice that beyond macroeconomic stability, fiscal policy has other key objectives such as equity (intragenerational and intergenerational) and thus any fiscal target (debt and deficit) also influences this dimension. Notice in particular that the bulk of government expenditure in EU countries can be classified as social expenditure. 3 ASSESSMENT OF THE EXISTING FISCAL FRAMEWORK Designing a suitable policy framework that would contribute to deliver macroeconomic and fiscal stability requires considering the policy track record which contributes to determine it vulnerabilities and strengths and assessing the role that institutions and institutional arrangements play in shaping fiscal policy. A robust policy framework should be consistent with the principles of democratic accountability which underpin sustainability of fiscal policy and with EU and euro area fiscal policy requirements in line with membership obligations. Such a framework should ensure consistency between fiscal discipline and fiscal sovereignty. These issues are discussed in this section. 3.1 The policy track record As discussed in Section 1 in the period leading to the crisis the outturn of government balances was close or lower deficits than those planned in Convergence and Stability programs (Figure 3). The main exception was the year 2008 when the deficit outturn was higher than planned in 2007 (1 percentage point of GDP). In the pre-crisis period, with also the exception of 2008, expenditure growth followed a more stringent course than the one implied by prudent fiscal policy making or a fiscal rule in which expenditure growth does not exceed a projected medium-term rate of potential output growth (Figures 13 and 16). A stricter expenditure rule (Figure 14 and 15) if followed, particularly during the first 4 years of the decade, would have resulted in compressing the expenditure-to-GDP ratio by 10 p.p. of GDP in the period 2000-2008 (Figure 17). To some extent actual government's expenditure dynamics were anchored on forecasts of macroeconomic variables (e.g. GDP) whose outcome was higher than forecasted (Figure 43). The non deliberate interaction between forecasted macroeconomic variables whose value was lower than actual outcome and their use in fiscal projections could be explored for policy formulation in the future (i.e. Fiscal projections could be elaborated deliberately with a margin of prudence). Prior to the crisis there was a degree of deficit bias in policy making but the main factors underpinning budget dynamics were wage and tax policies which should be addressed directly with appropriate policies. The assessment of the government compliance with the medium-term objective set officially in the Convergence Program (2005) in terms of a structural deficit of 1% of GDP depends on the set of data used for measuring compliance. The assessment based on data available on real time to policy makers (European Commission 2008) indicates that the fiscal target was met (Table 1). The same assessment based on ex-post data (European Commission 2010) indicates that the government was quite far away from the target (Table 2). This raises serious doubts regarding formulation of policy based on estimates of structural balances in real time and its assessment based on ex-post data. Nevertheless, this does not question the issue that Slovenia had and followed a fiscal target or whether there was unwillingness to comply with it. The issue is whether the target should get more pre-eminence and ownership within the fiscal framework and to define its status within the policy framework. This is particularly the case since the introduction of the so-called "fiscal rule" for setting government expenditure ceilings introduced in 2010 since it blurs its relevance or does not establish a clear and transparent link between the MTO and the expenditure ceilings. Assessing the fiscal policy stance measured by the change in the structural balance against the position of the economy in the business cycle (i.e. output gap) based on real time and ex-post data also provides conflicting and different views on the fiscal stance (Figures 10 and 11). Thus, measuring the fiscal stance based on change in structural balance is questionable unless the fiscal outcome is assessed with original macroeconomic forecast. The potential role that fiscal policy can play as countercyclical tool and its possibilities to offset fluctuations and shocks in the case of Slovenia is subject to strong constraints, not only deriving from structural reasons, such as the size of the economy, its openness and the absence of monetary policy, but also from reasons related to information and implementation lags and the very possibility to implement a sizable contraction or expansion in the budget balance. The fiscal multiplier in the case of a fiscal contraction engineered by investment reduction in the magnitude of 1% of GDP is estimated to lower GDP growth rate by 0.3% in the first year (Figure 22). This means that for a desired reduction/increase in GDP growth in the magnitude of 1 percentage point it will be necessary to reduce/increase investment as percentage of GDP in the magnitude of 3.3 percent. The strength of fiscal policy to offset economic fluctuations depends upon monetary conditions, the credit channel of monetary transmission and behavioural response of households and enterprises. With regard to information that could have been used in deciding whether to enact countercyclical policy, the overview of past GDP forecasts used in budget formulation indicates that with exception of 2009, when the economy underwent a deep recession and the forecast required various revisions, they did not provide strong signal for fiscal intervention as the dispersion of forecasts was narrow around a 4% real growth rate (Figure 5). Similarly, output gaps estimated in real time and used at the time of budget formulation were not widely spread before the crisis. Based on most recent experience (i.e. the grate recession), the effectiveness of countercyclical policy seems to be related mainly to its impact on mitigating unemployment in the short-run and depends on a level of government debt that allows the operation of fiscal stabilizers without jeopardizing macroeconomic stability. Nevertheless, there are important institutional changes that could enhance the strength of countercyclical fiscal policy such as the ability and speed of withdrawing EU funds and taking into account the impact of investment projects carried out by non-financial public enterprises. Unless the existing EU fiscal framework that relies on automatic stabilizers with respect to a chosen MTO is changed, despite the fact that the calculation of structural balance is controversial (i.e. real time or ex-post data), it seems that a more appropriate benchmark for assessing fiscal policy changes with respect to business cycle is by means of the difference and sign between structural balance and the MTO than simply the change in structural balance. This is because the difference between the structural balance and the MTO provide a view of the stance with respect to the chosen official neutral fiscal position. Subject to the same caveats as regard to the measure of structural balance in real time, the analysis with regard to the fiscal attitude towards business cycle based on real time data indicates that the fiscal stance vis-a-vis the MTO was fairly neutral before the crisis (the difference against the MTO was very small). However, if the same assessment is made based on ex-post data it indicates that fiscal policy was pro-cyclically loose during 20062008, while in 2009 was counter-cyclically loose (Figure 12). While the assessment of the degree of pro-cyclicality of the fiscal stance depends on which data is used (real time or ex-post data), the impact of approved policies on public finances whose impact take time to materialize and extends beyond a government's mandate is less controversial. Looking in perspective, key policies underpinning fiscal developments in the past include wage, investment and tax policies. Other policies, such as subsidies, granting of additional social transfers and one-off increases in social transfers (e.g. granting child allowance and transfers to pensioners in 2008) coinciding with political cycle, although affecting the fiscal stance were less significant with regard to their fiscal impact. In particular, public sector wage policy also influenced by the political cycle after the increase in 2001 was instrumental in contributing to curb wage and expenditure dynamics. While on theoretical ground the wage reform that begun in 2002 could have been implemented in a fiscally neutral manner, due to its implementation delay (2002-2008) and political economy considerations (likely opposition to a reduction in nominal wages of relative high wage earners in public sector), it generated and resulted in a discrete increase in public sector wage costs. Public wages after growing below inflation jumped in 2008 at the time of an election year and coincided with the beginning of the recession. The 2001 and 2008 wage adjustments and the wage policy between them highlight the importance of sustainable wage policy in the public sector as a corner stone for a resilient fiscal framework. Other important expenditure policy whose effect extends beyond a government cycle is investment. While government investment influences potential growth it also contributes to debt accumulation unless is financed with EU funds. Over the past years the investment cycle has broadly followed the political cycle. Its level increased in the last two years before the crisis and since then has remained at a relatively high level. In good times financing a high level of investments is less of an issue and probably less required. Under the current scenario where investment might be required, there is a need for fiscal consolidation and there is an impending risk of the government credit rating being downgraded the level of investments should be assessed in light of the risks and policy objectives. In particular, investment priorities should be decided carefully and the framework for accountability of EU funds withdrawal should be enhanced with a view of improving the speed and amount of resources withdrawn. Besides policy changes influencing finances there are shocks that affect government expenditure such as those affecting the indexation dynamics of social transfers. While there has been progress in unifying indexation of social transfers in terms of inflation, pensions are still indexed to wages. Shocks to these variables change the level of social transfers permanently. This has been particularly the case of pension's expenditure in 2008 where its level increased permanently mainly due to public sector's wage increase resulting from the implementation of the Wage Act (Figure 49). Thus to avoid distortions undermining sustainability the fiscal framework should mitigate the possibility that shocks become permanently imbedded in expenditure outlays. Not addressing these developments by targeted policies, even in the case where the fiscal framework relies on binding expenditure constrains, can create distortions. In particular, a wage shock like the one observed in 2008 and transmitted to expenditure through higher level of pension outlays could lead to compression of other expenditure priorities. Figure 49: Pension expenditure, in mio EUR 350 300 - 250 - I Change in expenditure (LHS) -Total pension expenditure (RHS) 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Ministry of Finance Tax revenue is another important driver of public finances and the effect of changes to its structure extend beyond a government's mandate. Regardless of any limit or binding ceiling on expenditure if government revenue falls, due to tax reform including granting of tax expenditure, public finances can be affected. The timing of changes in taxes is also relevant as it can affect macroeconomic conditions and its effect on public finances can be blurred depending on which phase of the cycle they are made. The tax system has undergone intense changes in the past and changes are still being planned (e.g. introducing a cap on social security contributions). While tax burden reduction can be justified on various grounds (e.g. economic efficiency gains), to ensure that tax changes in the future are consistent with stability of public finances there seems to be a need for a procedural rule that would constrain tax reforms to those that are revenue-neutral. In particular a procedural rule should be introduced that ensures that tax revenue reduction proposals (including through tax expenditures) should spell out in the same proposal offsetting commensurate expenditure cuts (permanent) or revenue increase to ensure neutrality on the fiscal stance. Another important considerations regarding tax revenue is the pro-cyclicality associated with the timing when tax reform measures take place and the issue of whether in budget planning and execution expenditure adjusts to revenue developments in a pro-cyclical manner. Regarding tax reform measures, these have been done incessantly in the past years and most measures have been aimed at reducing taxes but not all (e.g. Introduction of VAT in 1999 and adjustment in 2001). If attention is given to the most recent tax changes and to ex-post estimates of output gap it can be inferred that the 2006-2007 tax reform took place when the output-gap was wide opened or contributed to it. A different conclusion would be arrived by using real time estimates of output gap. This makes the timing of a reform a controversial issue. In any case, even under the assumption of a perfect foresight regarding the actual output gap (size and sign) it can be argued that tax reform measures are part of policy agenda of elected governments which cannot wait for the optimal cyclical conditions to be implemented. While this argument is valid, the other issue is whether changes to tax system that reduce revenues permanently should be made without compensating measures. In a framework in which a medium term objective should be achieved and preserved this does not seem to be the case. Thus the argument in favour of a procedural rule ensuring revenue neutral tax reforms is not only consistent with the requirement of meeting a medium term objective and of maintaining a neutral fiscal stance with respect to the cycle (which stage is difficult to pinpoint in real time) but also with the requirement of ensuring fiscal responsibility of incumbent governments which extend beyond the political cycle. The other issue is whether in budget planning and execution expenditure follows revenue in a pro-cyclical manner. Evidence regarding State budget figures indicates that during 2000-2003 supplementary budgets (REB) reduced revenues below the level set in adopted budgets (SP) (Figure 50). With the exception of 2003 the expenditure level in supplementary budgets was also lower than that in the adopted budget. In the period 2004-2008, the revenue levels in supplementary budgets (REB) or revised budget (SSP) were higher than those in adopted budgets (SP). In this period with the exception of 2008 expenditures were adjusted downwards in supplementary (REB) or revised budgets (SSP) (Figure 50). Looking at this behaviour in light of ex-post output gap estimates suggests that in budgets that are actually executed (supplementary (REB) or revised budgets (SSP) revenues were revised downwards in bad times and adjusted upwards in good times. Notice also that in 2004 and 2008 the adjustments were particularly large. This suggests that there is a certain degree of cyclicality when planning and executing the budget as revenues are adjusted in line with business cycle developments. Furthermore, a careful look at the dynamic of total and individual tax revenue during 2006-2008 highlights the issue that the effects of the business cycle were imbedded in planned budget and accommodated increase in expenditure (Figure 51). When the cycle reversed abruptly in 2009 expenditure remained high (Figure 2). Thus, procedural rules preventing upward adjustment in revenue estimates during budget execution by means of revised and supplementary budgets could be introduced and revenue planning should adjust for the effects of cyclical fluctuation in good times. This procedural rule should link revenue estimates for budget preparation to a prudent rate of medium-term economic growth. Alternatively, revenues should not be upward adjusted in good times beyond the level already determined in an adopted budget. Given that the existing fiscal framework demands the State budget target to be achieved in nominal terms, a related issue concerns whether during budget execution in reaching the target higher-than-budgeted revenues (supplementary or revised budget) are matched by increase in expenditure. Evidence regarding State budget figures (approved and outturn budgets) indicates that in the period 2001-2008 there was not such a cyclical behaviour (Figure 8). During 2001-2003 when revenue outturn felt short of that budgeted, expenditure was corrected downwards (exception in 2003) pointing out to a degree of optimism in revenue forecast. In the period 2004 to 2008 when revenue outturn exceeded that envisaged in the adopted budget (surprises) expenditure was mostly corrected downwards (exception in 2004 and 2008). In light of what was previously discussed it seems that pro-cyclicality, particularly in good times, comes mainly from budget planning (revised and supplementary budgets) than from ex-post adjustment of expenditure to revenue in budget execution process. In bad times (2001-2003) pro-cyclicality was also imbedded in budget execution as expenditure was reduced to match lower revenues. Therefore, careful attention should be given to the macroeconomic forecast and to a prudent planning of government revenue which in turn restrains the overall policy stance. Thus, there is a case for a procedural provision specifying that higher-than-expected revenues should be allocated to debt reduction. Figure 50: State budget: Expenditure and revenue - difference between figures in supplementary (REB) and revised budgets (SSP) and the adopted budget (SP), mio. EUR Figure 51: Planning expenditures and tax reform while revenue buoyancy, bn. EUR 2000 2001 2002 2003 2004 2005 2006 2007 2008 0,0 REB-SP REB-SP REB-SP SSP- SP SSP-SP REB-SP REB-SP REB-SP SSP-SP • Sociä security contributions - Domestic taxes on goods and services • Taxes on income and profit - Taxes on payroll and workforce trend Taxes on property • Taxes on irternationä trade and transactions Other taxes —--— No payrolltax reform * \ 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Ministry of Finance, own calculation Undoubtedly the trend decline in debt-to-GDP ratio and the low ratio before the crisis (21.9% in 2008) provided the Slovenian economy with the most powerful macroeconomic stabilizer. It did not only allow for covering sizable deficits resulting from higher expenditure and collapse of revenues but also contributed to mitigate the effect of the crisis on the domestic banking system without resulting in an increase of interest rates for both the government and the economy at large. In an economy where there is not independent monetary policy, the price of government debt in the secondary market provides the reference rate for borrowing for all domestic agents. In turn, the relative size of the debt has critical importance for overall 6,0 5,0 3,0 2,0 1,0 borrowing costs of the economy. The debt crisis in the euro area and imminent risk of downgrading of Slovenia's sovereign credit clearly highlight the importance of the debt as key policy variable to anchor the policy stance. While the policy framework should reflect the importance of the debt ratio as the key policy variable, meeting a debt target cannot be achieved only by constraining deficits and below the line operations but also require encompassing policies that mitigate fiscal risk such as the stability of the financial system, limiting government guarantees, minimizing the government involvement in solving too-big-to-fail institutions and sectors.12 Among these risks and as highlighted by the current crisis the stability of the domestic banking system clearly shows to be an overriding priority. This is because a weak banking system feeds back on government debt service and debt level. In the case of Slovenia this is more important because of the government's majority ownership of the two largest banks which currently face the challenge of low capital ratios and negative credit rating outlooks. While many new EU member countries that have fully privatized their banking systems (primarily to foreign banks) did not need to intervene to support financially domestic banks this was not the case in Slovenia. The issue is of key important since the crisis is still feeding through the banks' balance sheets and to sustained growth and lower debt ratios it is necessary healthy and well capitalized banking system. This is even clearer by the fact that all countries in the EU, under adverse cyclical conditions, have increased the capital adequacy of their respective banking systems during 2009-2010 while Slovenia lagged behind. A related issue that deserves attention in shaping the fiscal policy framework but which is different in terms of its policy implication is the issue of contingent liabilities associated with ageing. According to an agreement backed by most members of the Economic and Financial Committee under the ECOFIN, the European Commission will start assessing periodically, in the context of the evaluation of Stability and Convergence programs, the extent to which countries incorporate in their fiscal stance (MTO) the frontloading of contingent liabilities. In particular, the assessment would consist of countries' compliance with and MTO that frontloads 33% of contingent liabilities associated with ageing. This for the case of Slovenia, as originally estimated by the Commission would imply that the MTO would have to be revised upwards. The MTO instead of the current level set at 1% of GDP deficit in structural terms should be set at the level of 0.75% of GDP structural surplus (i.e. upward correction of 1.75 p.p. of GDP). Based on recent past fiscal performance this would have implied running current government surpluses up to the level of 4% of GDP in good times. Since running surpluses implies making distributional choices (tax level above expenditure) than cannot be outsourced to a rule, but to be determined by democratically elected governments, the fiscal framework should ensure that elected authorities determine the policy mix between structural reform and fiscal restraint to address the issue of contingent liabilities associated with ageing and ensure that they do not become explicit. For this purpose, at the start of a new government's mandate an assessment of the risk of the fiscal stance breaching the minimum benchmark (safety margin) of observing the 3% Maastricht deficit criterion13 over the medium term (next 10 years) or a given debt-to-GDP target should be made (See Section 3). In the event that fiscal policy would fail to deliver on this target, the incumbent government should have to decide on the policy mix to mitigate the medium term fiscal risk and incorporate the relevant policy measures in the fiscal strategy during the period of its' mandate. 12 The outstanding amount of government guaranteed debt as percentage of GDP at the end of 2010 is estimated in 20.6% (MF 2011). 13 Formally, = " ^ ', where MB is the minim the representative output gap for weak cyclical conditions. Formally, 3 e • rog , where MB is the minimum benchmark, z is the budgetary sensitivity to the output gap and ROG is 3.2 The existing fiscal framework As a result of the economic crisis with roots in the financial sector fiscal and budget institutions are being strengthened at the EU level and in many countries. Slovenia is not the exception and major changes concern the establishment of a fiscal council in 2009, the introduction of a "fiscal rule" in 2009 which has been redefined in 2010 and the introduction of a budget planning procedure based on programs. This section looks at these changes and to the overall policy framework with the aim of enhancing the fiscal setting and incentives for policymakers to deliver sound fiscal policy based on the principles of accountability, transparency and stability. By fiscal stability is meant a policy and fiscal position that is consistent with macroeconomic stability (i.e. does not affect adversely the level of interest rates in the economy, that provides public finances' resiliency to shocks and ensures its long-term sustainability). The Slovenian fiscal framework encompasses a system of arrangements, procedures and institutions that underlie the planning, implementation, reporting and assessment of budgetary policies. The planning within a medium-term budget framework was until recently based on single independent forecast of macroeconomic variables prepared by IMAD which was regarded as strength or best practice of the Slovenian fiscal framework (EC 2010). Budget planning is based on a top-down approach and recently expenditure limits are derived from a framework "fiscal rule" as discussed below. Budget is subject to parliamentary approval and the parliament reviews budget execution. More recently a Fiscal Council was established with the purpose of ex-post assessment of fiscal stability. In general the framework is robust with regard to fiscal reporting and budget execution but could be improved with regard to three dimensions: i) bringing forward the objective of fiscal stability in fiscal planning; ii) enhancing accountability in delivering and monitoring the objective of fiscal stability by defining the roles of the government with regard to policy formulation (design fiscal strategy) and the role of the parliament in approving and monitoring the fiscal strategy with respect to agreed principles of fiscal stability and; iii) improving transparency by specifying the task and content of the respective responsibilities of the government and parliament with regard to fiscal stability. According to Law on Public Finance, the responsibilities with respect to fiscal policy (budget preparation, adoption execution and monitoring) between the government and the parliament are as follows: The government is in charge of planning, reporting and executing fiscal policy as imbedded in budgets and; the parliament is in charge of approving the budgets and final accounts and follows budget execution twice per year. To the extent that the parliament approves the budget proposed by the government and given formal constraints on the parliament to change the size of expenditure (i.e. subject to a reduction of other expenditure in equal size or increase revenue in the same extent (Article 30 of the Law on Public Finance) both institutions are co-responsible for fiscal policy. Nevertheless, while the government elaborates and endorses its fiscal strategy in the Budget Memorandum the parliament only acknowledges its existence every year as part of the documents accompanying the budget proposal. The existing legislation, both the Constitution and the Law on Public Finance do not elaborate in great detail on the objective of fiscal stability or fiscal responsibility. As mentioned before by fiscal stability it is meant a policy guided by principles that foster the observance of a medium-term budget position consistent with stable cost of funds for the whole economy, maintains enough fiscal space for addressing major shocks to the economy without creating major distortions and avoiding unsustainable policies including the timely addressing of contingent liabilities. The Law on Public Finance has an explicit reference to the need to strive for macroeconomic stability in the context of budget preparation and execution (Article 2). However, it does not elaborate on the procedures or provides orientation on how to pursue such objective and the respective responsibilities of the government and the parliament on delivering on the objective of macroeconomic stability. The Law on Public Finance requires the government to report to the parliament on the achieving of results and objectives of adopted budgets (Article 62), but this is understood with regard to content of programs and appropriations not on the issue of fiscal stability or macroeconomic stability. More recently, with the establishment of the Fiscal Council, this body has given the responsibility to evaluate stability and sustainability of fiscal policy as well as its quality. However, as a result this new arrangement the accountability of the parliament with regard to monitoring the government responsibility in delivering fiscal stability has been blurred. Under the existing arrangements both the government in charge of designing the fiscal strategy and the parliament which approves annual budgets and underlying fiscal policy (both composed by democratically elected individuals) are de facto accountable for the principle of macroeconomic stability to a body of independent non elected individuals. Slovenia has formally communicated to the EU commission (2005) in the context of the fiscal surveillance procedure (convergence and stability programs) that it follows a medium term fiscal objective defined as a 1% of GDP deficit in structural terms. This target was not formally changed. Such a target under the assumptions that the economy would grow at a nominal steady state growth rate of 5% per year and that ageing related expenditure would be contained by means of pension and health reforms it would deliver a long term debt-to-GDP ratio of 21% (Figure 52). Thus, the current MTO which was formally determined but not inscribed in the legislation or other document than the convergence and stability programs if strictly followed under the assumptions mentioned could deliver stability of public finances. Figure 52: Government debt to GDP ratio under structural deficit target of 1% of GDP and 5% nominal GDP growth 50,0 45,0 40,0 35,0 ..... 30,0 25,0 20,0 15,0 ..... 10,0 5,0 0,0 Source: own calculation Notwithstanding, the existence of formal medium-term target, more recently in 2010 the government in the Decree of Development Planning and Procedures for the Preparation of the National Budget (DDP) introduced the notion of "fiscal rule" without a direct link to the MTO. The rule as discuss later is in fact a framework to determine the rate of growth of general government expenditure (expenditure ceilings) based on desired targets concerning general government primary fiscal balance and debt-to-GDP ratio, but do not relate to the MTO. This means that there is a degree of overlapping among targets that needs to be clarified to enhance accountability and transparency. While targeting a structural balanced position, a constant expenditure growth rate or a given level of debt-to-GDP ratio can contribute to bring forward the objective of fiscal stability, having many variables which are mutually linked tend to over determine the objective their pursue and blur accountability. If the purpose is to institutionalize the planning and delivery of prudent fiscal policy then the key issue is to set the appropriate framework for commitment with clear objective, procedures and monitoring arrangements. In reviewing the existing fiscal framework these issues are addressed below. Account is also taken of EU fiscal surveillance requirements which at present are not fully incorporated in the Law on Public Finance (e.g. The Maastricht numerical constraints (fiscal rules) on deficit and debt). 3.2.1 Budget planning 3.2.1.1 Macroeconomic forecasting Until recently, before the amendments to the Decree of Development Planning (DDP) in 2010, the budget macroeconomic forecast was based solely on the forecast of the Institute of Macroeconomic Analysis (IMAD), which has been regarded as a realistic and independent forecast both by the public an international institutions (OECD 2011 and EC 2010). Now with the amendments to the Decree, the government can use alternative forecast (EU commission, Bank of Slovenia and OECD) to plan the budget which blurs transparency and accountability. The issue is the potential arbitrage among forecasts in the elaboration of the fiscal projections. Thus returning to previous practice of relying on a single forecast which includes all relevant variables needed to elaborate fiscal projections should enhance transparency. This requirement could be explicitly incorporated in the Law on Public Finance. Alternatively, to ensure accountability and transparency the Law could determine the way various forecast are used (e.g. simple average, weighted average). 3.2.1.2 Fiscal projections Fiscal projections for budget planning are elaborated by the Ministry of Finance which until recently was based only on the IMAD's macroeconomic forecast. While the official estimates of the Ministry of Finance should be used for budget preparation to contribute to accountability, alternative estimates of government tax revenue could be elaborated to benchmark those of the government. This could be done as follows: a) instructing the IMAD to produce and publish forecast of tax revenue; b) requesting Bank of Slovenia to publish its own forecast or; c) commending this task to a new office that could be establish under Parliament responsibility to monitor fiscal developments (see Section 4). Currently the government's budget planning figures and the government fiscal strategy as presented in the Budget Memorandum are based on the national methodology. This methodology is based on cash principles and covers the budgets of the State, local communities, Health and Pension funds. This methodology seems appropriate to be followed for budget preparation and execution while a transition to accrual principles for budget preparation is not done. However, the government of Slovenia due to EU membership has to report its fiscal accounts in accrual basis (ESA 95 methodology) which is used for fiscal surveillance. Under this methodology the scope of the general government is wider and the size of the general government aggregates due to also accrual principles is different (e.g. government expenditure, revenue and deficit). To ensure responsibility and accountability it would seem appropriate that the Budget Memorandum or the government fiscal strategy should include in the same document the estimates of budget aggregates and their projection based on both methodologies. This would also enhance the responsibility of the parliament with respect to the fiscal policy and its accountability. 3.2.2 Fiscal strategy and targets According the Law on Public Finance the fiscal strategy for current and next three years as described in the Budget Memorandum is an act of the government that is presented for information to the parliament every year. The parliament approves two year budgets on a rolling basis which are part of the government's fiscal strategy. However, it does not approve formally the government strategy. This implies a degree of inconsistency in the way fiscal responsibility is framed. This is exacerbated by the fact that the law constrains substantially the power of the parliament to modify the structure of the budget (Article 30). To amend this anomaly the parliament could also approve the government strategy as presented in the Budget Memorandum given the fact that it approves the budget figures that are part of the strategy. Alternatively, which is the case in some countries (e.g. Netherland, Finland and Denmark) coalition agreements at the beginning of a government mandate could set the fiscal strategy for the period of the mandate including the targets and variables to be monitor by the parliament and public. Currently the State budget is approved for the next two years on a rolling over basis and the budget memorandum includes fiscal forecast for the year t+3. Local government budgets and social security funds are elaborated based on the forecast decided by the government and they are approved by their respective constituencies. Overall macroeconomic consistency is ensured by regular coordination with the social security funds and local governments. General Government accounts are presented in a consolidated (National methodology) when budget adopted by the government and then presented to parliament in October each year. Formally, the Law on Public Finance does not demand from a government in defining fiscal policy (budgets for the period ahead t+3 years) to set a medium term objective or fiscal targets but in preparing the budget to strive for macroeconomic stability. Nevertheless, as a result of the process of EU membership Slovenia had to formally set fiscal targets guiding fiscal policy or underpinned it. Table 5 shows fiscal targets and their evolution in different documents prepared in the context of EU fiscal surveillance. Formally, a structural deficit of 1% is currently the medium term objective. However, the formal link between the fiscal target defined in the EU surveillance context (Stability Program) and those spell out in the Budget Memorandum is not very clear. The government when reporting on the achievement of the medium-term objective in the stability programs (2007-2010) bases the information on adopted budgets and projections of fiscal variables set in the Budget Memorandum. However, it does not spell out in the Budget Memorandum how the adopted budgets for next two years and the projection for the third year are consistent with the medium-term objective reported in the Stability Program. Neither the Budget Memorandum spells out how the medium term objective is going to be achieved and/or how divergences are going to be reconciled by means of policy changes. Period Fiscal target 2001-2002 (Pre-accession Economic Program) Balanced budget position 2002-2004 (Pre-accession Economic Program) Close to structural balance position 2005- (Convergence and Stability programs) Structural deficit of 1% of GDP Since the adoption of the Law on Public Finance in 1999 budget figures presented in the Budget Memorandum for the next two years and projections for the third year were in fact fiscal targets. Reconciliation of these figures with the MTO was implicitly done in convergence and stability programs. More recently in the Budget Memorandum 2011-2012 the budget targets are derived or are consistent with the requirement of Slovenia reducing the deficit below Maastricht criterion (3% of GDP) by 2013 and a targeted level for government debt not exceeding 45% of GDP until 2015. Yet, the relation of these budget targets with respect to the medium-term objective is not addressed. With regard to the pace of consolidation it is argued that the targets chosen respond to the expenditure-based consolidation approach and the need to preserve the development function of fiscal policy. The choice of a fiscal target anchoring fiscal consolidation of a deficit below the 3% of GDP threshold is obvious given requirement of compliance with Stability and Growth pact. The issue is whether a faster debt reduction dynamics than the one envisaged might be required given the risk of sovereign credit downgrading and the adverse effect on borrowing cost for the whole economy and economic recovery. More broadly and beyond the fiscal consolidation period the issue is what should be the variable and its target level anchoring fiscal policy over the medium term or the key variable to target within the fiscal framework. In Section 1 it was argued that the various fiscal aggregates such as deficit, debt or expenditure are mutually consistent and that a rationale for setting their targeted level should be derived as well as the order of precedence among them. Based on the recent experience with the financial crisis, the debt-to-GDP ratio seems to be the main fiscal variable to be targeted over the medium term. In Section 3 modalities for making the debt level the main medium-term fiscal target are discussed and its compatibility with a MTO. Setting fiscal targets for framing budgets in a medium term perspective raise the issue of credibility and accountability. Meeting fiscal targets depends primarily on political commitment. Fiscal targets to be consistent with the objective of macroeconomic stability have to be derived over the medium term. Fiscal targets can be enforced over elected policy makers by formally imposing upon them the requirement of adherence to a pre-determined level of a given fiscal aggregate (e.g. structural balanced position or debt-to-GDP ratio not exceeding 25% of GDP) which implies that their responsibility for fiscal stability is outsourced or delegated to a constraint. Alternatively, it can be demanded that they set the targets in conformity with a set of macroeconomic stability principles and criteria for assessing it which implies that their responsibility with regard to fiscal stability is made explicit or transparent and it is not outsourced neither delegated. Nevertheless, in both cases the main issue is setting an ex-ante framework for the delivery of a sound medium-term fiscal policy strategy and in both cases the key challenge is the delivery on the strategy's targets. In the case when a fiscal target is imposed to elected policy makers (for example by a numerical rule) the additional complication is that there is difficulty in determining which target or the optimal target level to be enforced. In principle a target can always be improved by a better or superior target whose rationale can be always justified if not by economic rationale by the very issue of being pursued by an elected government.14 In the second case, when the policy maker decides on the fiscal target to pursue, the issue of optimality is less controversial as elected policymakers can define and make explicit their fiscal policy intentions with respect to fiscal stability principles and criteria (e.g. observing a minimum safety margin with regard to breaching the 3% Maastricht deficit criterion). Thus for the sake of accountability in pursuing and delivering targets the option of enforcing on elected policy makers the requirement to make explicit their fiscal strategy with regard to fiscal stability objectives seems incentive compatible as the ownership of the targets is ensured. Until recently, proposed budgets for the upcoming two years and budget projections for the third year were fiscal targets in themselves of a not very explicit medium-term fiscal strategy. In addition, the policy framework allowed and still formally does to use the possibility of supplementary budgets to amend approved budgets when macroeconomic conditions change or when the government incurs in new commitments (Table 6). While it can be argued that this is an appropriate exit clause to cope with uncertainty, the issue is whether objective circumstances demanded changes and how the changes fit in the government medium-term fiscal strategy. More recently in the Budget Memorandum (2011-2012) the proposed two year budgets and budget projection for the third year are part of the strategy to reduce deficit below Maastricht criterion by 2013. This reflects and improvement as the fiscal target is explicit. However, despite the fact this target is implicitly anchored on an expenditure growth path derived as discussed below from a new framework ("fiscal rule") the issue is still the delivery on the strategy and accountability for deviations from it. This is because all the parameters used in deriving the expenditure path can be revised after two years and there is no requirement demanding correction and justification for deviations. Furthermore, under uncertain economic circumstances relying only on expenditure containment might not deliver the appropriate pace of debt or deficit reduction. Thus the fiscal framework could be improved by imposing the requirement on newly elected governments to set a medium-term fiscal policy with explicit fiscal targets and reporting on their delivery. In particular, newly elected governments could frame the fiscal strategy in the budget memorandum at the beginning of their mandate. The budget memorandum could be re-named for this purpose as fiscal strategy. Updates of the Fiscal strategy (budget memorandum) should describe the fulfilment of the strategy and, as appropriate, explain reasons for possible deviations and measures to rectify eventual deviations. One important issue regarding the current framework that simultaneously targets debt and primary balance is the link between the developments in the general government budget as defined according to national methodology and the evolution of debt. This is because data indicates (Stability Program 2011) that there is a relatively larger difference between the envisaged change in government debt and the budget deficit in 2011 measured based on national methodology (1.2% of GDP) than when the differential is between debt and the deficit measured based on ESA 95 basis (0.5%). This questions the reliability of the new framework in actually delivering on the debt target. Probably the difference might arise from a different definition of the scope of 14 While there is a consensus on the need to control the accumulation of government debt so as to avoid unsustainable debt dynamics, such a consensus does not exist with regard to the criteria for deciding what the level of debt should be or, if governments should run surpluses, what level and rate of financial asset accumulation is fiscally sustainable or desirable (Price, et al. 2008). While the consideration of future commitments is relevant (ageing expenditure) the issue is how better this issue is address (pension reform or accumulation of assets (running budget surpluses). Other issues relevant in deciding targets include considerations regarding adequacy of savings and the role of fiscal policy which might imply distortions in terms of taxes, transfers and government regulation. The financing of growth enhancing expenditure and using surpluses to reduce taxes are also relevant. general government between the national methodology and ESA methodology. To ensure accountability and transparency when presenting the fiscal strategy and targets projections of government balance and debt should be submitted simultaneously for approval and discrepancies explained. Table 6: Dates of adoption of budgets, revision and supplementary budgets Budget year Adopted Revised Supplementary budget 2002 Dec. 01 Oct. 02 2003 Dec. 01 Oct. 03 2004 Dec. 02 Dec. 03 2005 Dec .03 Jun. 05 2006 Dec. 05 2007 Dec. 05 Nov. 06 Jul. 07 2008 Nov. 06 Nov. 07 May 08 2009 Nov. 07 Mar. and Jul. 09 2010 Nov. 09 Jul. 10 2011 Nov. 09 Sep. 10 Source: Ministry of Finance 3.2.3 Medium-term budget framework A medium-term budgetary framework (MTBF) sets the budget procedure in a medium-term perspective. It includes arrangements and procedures that enable extending fiscal policy beyond the annual budgetary calendar as many decisions have or can have effect over the next years (e.g. In the case of Slovenia (20022008) the reform of wage policy or tax reform). Given the MTBF medium term orientation it should contribute to facilitate the delivery of fiscal strategy on the targets imposed or decided by a government. Broadly the Slovenian MTBF complies with standard desired features such as realistic economic assumptions, multiannual budgetary planning, budgetary centralization at the planning and approval stages, top-down budgeting, rules constraining the fiscal performance of local government and pension fund. In addition, the two-years budgeting process and the requirement that the budget outcome should be met on nominal terms as approved have resulted in a relatively stable and more predictable budgetary outcome. In particular, with the exception of the years 2003, 2004 and 2008 upward revisions of government expenditure figures in revised or supplementary budgets did not diverge significantly from those in adopted budgets (Figure 50). The framework also contributed to a gradual fiscal consolidation until 2007 and resulted in a general government expenditure growth rate below that of estimated potential output with the exception of 2001 and 2008. One of the perceived weaknesses of the framework might be that expenditure ceilings, which are instrumental for the delivery of fiscal targets within a MTBF were not binding. A step forward in this direction has been made recently with the introduction of a framework "fiscal rule" which determines a fix ceiling for general government expenditure for the current and following year and an indicative ceilings for the two subsequent years (see Box 1). Beyond the issues involved in the new framework for determining expenditure ceilings as discussed below or an alternative framework for defining them, the relevant issue is the extent to which they can contribute to achieve fiscal targets over the period where they are applied and binding nature in good times. Expenditure ceilings derived either with a formula or following a principle of sound prudent policy making (e.g. expenditure growth not exceeding medium term economic growth) might not be enough to deliver fiscal targets if they are not consistent with government revenue developments and fiscal targets, which makes policy making in real time a more complex process than simply constraining expenditure growth. This highlights the importance of accountability in meeting fiscal targets rather than of abiding to a constraint that becomes non-binding or not enough to deliver the targets. This is illustrated by looking at how the new framework performed at the time when it was first applied in the elaboration of the adopted budgets for 2011-2012. Figure 53 shows that the expenditure ceiling for the 2010 adopted State budget and derived from the new framework (€10.4 Bn.) was first adjusted downwards in the supplementary budget adopted in July 2010 (€9.8 Bn.) and then the 2010 budget outturn was even lower expenditure (€9.2 Bn.).15 The expenditure level regardless the ceiling imposed had to be adjusted in accordance to lower revenue than budgeted in order to meet the deficit target. Therefore, in order to deliver a fiscal target, expenditure has to be adjusted in line with revenue developments. Thus, in a consolidation episode what anchors the process are targets and political commitment rather than ceilings that become not binding. Ceilings become more binding in good times but in those times, if there are not well defined targets or accompanied by policies that result in stable government revenue then they become useless. Notice also that revenue planning during bad times tends to be optimistic while in good times pessimistic (Figures 8 and 9) suggesting the notion of a "prudent planning" during bad times and particularly during a fiscal consolidation episode. Figure 53: State budget, revenue and expenditure 11,0 10,5 10,0 9,5 Dj 9,0 8,5 8,0 7,5 ■ 7,0 2009 2010 -Adopted 2010 -Suplem. 2010 -Outturn 2011-Adopted 2012-Adopted (Dec. 09) (Jul. 10) (Jan. 11) Source: Ministry of Finance 15 In the budget 2010 the expenditure ceiling strictly applied to expenditure financed with own resources. Nevertheless the revision downwards was in similar magnitude than that of the overall expenditure. Box 1: A Framework for determining government expenditure ceilings "the fiscal rule" The so called "fiscal rule" introduced in 2009 was formally defined in the Decree on the Documents of Development Planning Bases and Procedures for the Preparation of the Central and Local Government budgets (Official Gazette of the Republic of Slovenia, No. 54/2010). It cannot be considered as a rule in the sense commonly understood of being a permanent constraint on fiscal aggregates in terms of numerical limit (e.g. maximum deficit 3% of GDP) as most of their parameters can be changed including the estimates of potential output. It can be conceived more as a procedural framework to derive expenditure ceiling based on targets defined by incumbent policy makers. It is composed of two parts: one that sets expenditure in line with trend growth (semi automatic depending on the estimates of potential growth and forecast of GDP deflator) and the second a corrective term (discretionary) to reduce growth (increase) in line with given targets. Formally the rule in the Official Gazette is defined as follows: The "fiscal rule" sets the upper limit of general government expenditure and is determined using the following formula: Gt+1 = Gt X (1 + g*) Nominal growth of general government expenditure (g*) is determined as follows: g* = gtre"d - ux(bt - b*) - vx(ft - f*) trend growth Corrective part of the formula where Gt+1 Forecast for general government expenditure in the following year (in EUR), Gt Estimate of general government expenditure in the current year (in EUR), gtre"d Arithmetic average of the past three years, the current year and forecasts for nominal growth of potential gross domestic product in the next three years (%) bt Estimate of consolidated gross general government debt for the current year (share of GDP), b* Target level of consolidated gross general government debt (share of GDP), ft Estimate of general government primary balance for the current year (share of GDP), f* Target level of general government primary balance (share of GDP), u Speed of reaching the target level of consolidated general government gross debt (a value between and 1), v Speed of reaching the target level of general government primary balance (a value between 0 and 1). Potential gross domestic product is estimated using the production function method, which is also the official method used by the European Commission in calculating potential gross domestic product. The b*, f*, u and v parameters are determined for the period of two years. If fiscal consolidation is called for due to the deterioration of macroeconomic indicators and a consequently lower estimate of potential gross domestic product, the government of the RS can change the fiscal rule parameters and, consequently, the upper limit of general government expenditure._ The so-called "fiscal rule" as presented in Box 1 can be best described as a framework for deriving expenditure ceilings. In the literature by fiscal rule is understood "a permanent constraint on fiscal policy through simple numerical limits on budgetary aggregates" (Kopits and Symansky, 1998; (see also IMF, 2009). In principle those ceilings derived from the framework together with procedural rules concerning tax revenue reform and prudent forecast could deliver on fiscal targets. However, the framework as currently defined has some drawbacks. They relate to the coverage and uncertainty involved in the calculation of the ceilings. The framework covers total general government expenditure and this conflict with the Law on Public Finance that gives local communities the right to set the level of their expenditure. A particular problem concerns the inclusion of expenditure financed with EU funds. This type of expenditure should not be constrained by a ceiling as the dynamic of withdrawing EU funds is not yet fully predictable as shown by frequent revisions of these entitlements in the revised and supplementary budgets. In addition, this type of expenditure to a large extent does not have large impact on the budget outcome as they are financed with revenue from the EU. Another issue concerning the coverage of the rule is that it is applied to government expenditure estimated on national methodology and not on ESA 95 with different coverage of government. To enhance transparency the Budget Memorandum or the fiscal strategy of the government should include estimates of the ceilings based on both methodologies (cash and accrual). With regard to expenditure of local communities, given the fact that the level of their indebtedness is subject a numerical fiscal rule (i.e. total payment of principal and interest in an individual year should not exceed 8 % of revenues generated in previous year), the issue is more of compliance with the rule and its' monitoring. Nevertheless, the rule should be assessed in light of the possibility that a heavily indebted municipality could raise tax and non tax revenue as a consequence of indebtedness dynamics with potential distortions. Looking more in detailed at the framework for determining the ceilings on growth of general government expenditure as described in Box 1 it can be observed that it has two components. The first component works as a procedural rule which sets the growth in line with nominal potential GDP growth rate. The second component is a discretionary feature aimed at correcting downwards the expenditure dynamics in line with government targets in terms of primary expenditure and debt. While the link with fiscal targets is explicit in the second component, despite the fact that all parameters can change, this is not the case in the first component. The first component is based on estimates of potential growth which are quite volatile and depend on who does the estimates (see Section 2 and alternative estimates of potential growth in the Slovenian Stability Program and European Commission's publications). Given the uncertainty of the estimates and to ensure transparency the estimates of trend growth could be done independently as it is also the case of the macroeconomic forecast.16 One additional complication regarding the estimation of nominal potential trend of GDP comes from the forecast of the GDP deflator. Given the uncertainty of the various parameters involved in deriving the expenditure growth and ceilings it seems that a better option in terms of accountability and transparency would be to simply announce fixed nominal expenditure ceilings with an unambiguous and direct link with the pursued ultimate fiscal target. With regard to the targets involved in the corrective part of the framework, it seems puzzling the targeting of both the primary deficit and the debt while the two variables are related. Nevertheless, the option of leaving unspecified the levels of the ultimate targets underscoring the expenditure ceiling seems appropriate, as they give room for incumbent governments (ownership) to set their own targets within a more predictable framework. In principle the option of allowing an incumbent government to sets its fiscal targets in accordance with fiscal stability principles seems a useful feature to keep in an enhanced policy framework as discussed in Section 4. An additional issue not properly address in the existing modified framework is how the so-called rule relates to the existing MTO and the status of the MTO. The Decree on the Documents of Development Planning only states that the "rule" allows reaching in the medium term a balance budget in structural terms. However, this is not necessarily the case as reaching given targets depend on revenue developments (e.g. tax reform can reduce revenue permanently) and cannot be binding even in normal times. Furthermore, the "rule" if mechanically applied could lead to expenditure compression below desired levels. 16 This raise the issues of which institution should be in charge of estimating the trend growth and what underlying forecast of GDP should be used to estimate it. This is particularly relevant taking into account recent changes to the policy framework allowing to use alternative forecast in preparation of the budget. To illustrate the effects of what the "rule" could have delivered if applied in the past in terms of expenditure dynamics and government balance, expenditure dynamics were estimated for the period 2000-2009 based on EU Commission estimates of potential output (available since 2004 on a rolling over basis). To derive the nominal output growth (trend growth) actual numbers of GDP deflator were used. Figures 54 and 55 show the annual expenditure growth and expenditure-to-GDP ratio resulting from estimating expenditure growth trend with the formula described in Box 1 (i.e. keeping the expenditure growth rate constant for 2 years and revising it for the third year). Figure 54 shows that actual expenditure growth lagged behind that of expenditure derived from the formula with two major exceptions resulting from wage adjustments in 2001 and 2008 again underscoring the importance of sustainable wage policy. Figure 51 shows that if government expenditure would have grown with the pace defined with the formula (trend growth) the expenditure-to-GDP ratio would have been higher than the actual ratio with the exception in 2009. As a consequence the deficit incurred by the general government would have been higher in most of the years as well as the debt-to-GDP ratio. This indicates that the formula underpinning the growth trend would have not delivered a better fiscal outcome than the actual. For this to have happened a different approach (discretionary policy) or formula would have been required. Going forward it is clear that under the current consolidation process (conceived solely based on expenditure compression) the required expenditure growth rate that ensures that the targets are met is lower than expenditure growth derived from the trend growth (i.e. discretionary policy). This points out to the redundancy of the formula in the times of consolidation and again highlighting the importance of fiscal commitment and well defined targets. Beyond the consolidation process the key question is whether the formula, due to lower estimates of real potential growth in the future, if applied mechanically would lead to expenditure compression beyond of what is politically acceptable or desirable. Another important consideration in this regard is inflation developments and how the "rule" with cope with them in light of expenditure priorities. Figure 54: General government expenditure growth, in % Figure 55: General government's expenditure, in % of GDP 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2001 2002 2003 2004 2005 2006 2007 2008 Source: Ministry of Finance, own estimation Looking at the fiscal outcome resulting from the estimates of expenditure simulated with the "rule" (Figures 54 and 55) in light of the impact of the 2006-2007 tax reform on revenue indicates that the fiscal developments would have gone out of hand highlighting the importance of procedural rules underpinning tax policy. This is particularly important taking into account existing proposals to reduce social security contributions in the future (e.g. capping contribution of high income earners and reduction of corporate income tax rate). 50,0 48,0 46,0 8,0 44,0 6,0 42,0 4,0 40,0 2,0 0,0 0 Expenditure ceilings either derived from complex formulas or not (e.g. Expenditure growth set at 2% during next three years) and procedural rules covering also revenues are only instruments for achieving fiscal targets within a MTBF that are either imposed or determined by elected officials. The important issue is whether fiscal targets are consistent with macroeconomic stability over the medium term and its delivery. This is an important dimension that is still formally missing in the existing framework that could be enhanced as suggested in Section 4. 3.2.4 Accountability for fiscal stability Under current legal arrangements where the parliament although approving annual budgets has limited powers to modify the fiscal targets, it is clear that the primary responsibility for fiscal planning and strategy rests in the executive branch and on the parliament should lay the responsibility for monitoring its implementation. However, while the parliament is formally entrusted to approve budgets and annual financial statements of previous budget years its role with regard to monitoring the government's achieving targets and results pursued in the budget document is vague and limited to getting acquainted with them (Article 62 of Law on Public Finance). Furthermore, the share responsibility with regard to macro-fiscal stability is not defined. In particular the parliament does not approve the government fiscal strategy as presented in the Budget Memorandum but only gets acquainted with it. The existing framework does not mandate the parliament to monitor the consistency of annual budgets with a prudent medium term fiscal strategy during the duration of a government on office. It does not require the parliament to monitor the achievement of fiscal targets within a medium term strategy, neither to demand accountability of the executive branch with regard to fulfilling fiscal stability objectives nor to demand correction of deviations from given fiscal targets. Thus, the parliament is not mandated to exercise its share responsibility with respect to fiscal stability objectives beyond approving annual budgets and revising final accounts. Therefore, the role of the parliament with regard to fiscal stability objectives could be enhanced by entrusting it with the explicit mandate to approve and monitor the government fiscal strategy and the consistency of proposed two-year budgets with the medium-term fiscal objectives and targets imbedded in the fiscal policy strategy. For this purpose, the Law on Public Finance could be amended and require incumbent governments to specify at the beginning of their mandate a medium term fiscal strategy consistent with principles and set criteria on fiscal stability. The strategy should include medium-term fiscal targets and annual strategy to attain them. The parliament could be also entrusted with the authority to demand from the government correction in case of deviations of fiscal targets. The parliament could perform such a role with the help of a parliamentary budget office to be established for this purpose. Alternatively, it could rely on specially tailored reports made for this purpose by existing independent fiscal institutions. The parliamentary budget office could be similar to that of the Congressional Budget Office (CBO) in the United States. In the case of Slovenia such an office could assist the parliament with independent (positive and nonpartisan) analyses on the incumbent government fiscal strategy's compliance with fiscal stability objectives (see section 4.3.1) and its execution. In particular it could assess ex-ante (i.e. when a new government presents its strategy) whether a fiscal strategy before is implemented is consistent with fiscal strategy objectives. Then the office can ex-post assess whether proposed budgets are consistent with fiscal targets and government fiscal strategy. Such an office of a nonpartisan nature and at the service of the parliament could enjoy the appropriate legitimacy for the task. The task of the parliament with regard to its enhanced fiscal surveillance mandate can be facilitated also by IMAD which is the case of the CPB in Netherlands. The IMAD, at the beginning of a new government's mandate could assess the compliance of a coalition's fiscal strategy before it is implemented with or a new government's fiscal strategy fiscal stability objectives according to predetermined criteria. The assessment could be sent to the parliament or made public. In addition, for the parliament's regular assessment of the implementation of the government's strategy the IMAD and Bank of Slovenia could publish budgetary forecasts. Currently the IMAD publishes once per year an independent assessment of government fiscal policy, this publication could be used also as an input in the parliament when discussing and approving the implementation of the government's fiscal strategy.17 Providing the parliament with the mandate of approving and monitoring the fulfilment of the government's fiscal strategy with principles of fiscal stability seems even more appropriate at the times when fiscal surveillance framework in the EU is enhanced and is becoming closer to the process of formulation of budget targets in the context of the so-called EU semester (see section 3.2.5). In particular, such a role for the parliament could enhance its legitimacy with regard to the fiscal strategy followed by the government in Slovenia. Until recently, besides the parliament and before the establishment of the Fiscal Council, monitoring fiscal developments was carried out at national level by various official institutions with different emphasis and degrees of exposure to the public. One of such institutions is the Court of Audit. It facilitates the task of the parliament with regard to the assessment of government programs' fulfilment with objectives and provides audited government financial statements ensuring their integrity. However, the Court of Audit does not have an explicit mandate to audit fiscal stability objectives and targets but can also audit some policies of the government such as the degree of success of government policy in mitigating the impact of the financial crisis in the economy.18 Its current mandate covers the supervision of state accounts, the state budget and all public spending in Slovenia and may audit any act on past or planned operations. An important issue is the macroeconomic expertise. Bank of Slovenia also monitors fiscal developments in Slovenia as part of its policy mandate within the Euro area with emphasis in macroeconomic stability. However, these reports are not publicly available. The third institution is IMAD which regularly publishes a comprehensive assessment of the fiscal stance including its consistency with macroeconomic objectives in the publication Economic Issues.19 More recently in 2009 a Fiscal Council was established with the task of ex-post evaluation of fiscal and structural policy of the government. While the monitoring tasks with regard to fiscal policy are well defined in particular with respect to fiscal stability objectives (see Box 2) the key issues concern the Council's legitimacy within the democratic setting to carry out this task, the delivery on its mandate and its technical capacity. 17 The Court of Audit as independent institution reporting to the parliament could be also given the task of auditing government's fiscal stability objectives and report on those to the parliament. However, assessing fiscal stance requires specialized technical capacity involving macroeconomic and fiscal forecasting which is not abundant but could be developed with the appropriate resources. Nevertheless, in view that the IMAD has already such a capacity including regular elaboration of independent forecast and assessment of fiscal policy the IMAD could be more suited for such task. 18 http://www. rs-rs. si/rsrs/rsrs.nsf 19 http://www.umar.gov.si/en/publications/ Box 2: The Fiscal Council and its mandate According to the Law on Public Finance the tasks of the Fiscal Council are as follows: - ex-post evaluation of the stability and sustainability of fiscal policy as set out in the annual Budget Memorandum and Stability Program and compliance with the rules of the Stability and Growth Pact; - assessment of the adequacy of fiscal targets with the medium-term macro-fiscal framework; - annual assessment of the effectiveness of public spending, including EU funds, on the basis of annual accounts; - assessment of the trends of individual categories of government revenue and expenditure in terms of their impact on the sustainability of public finances, based on annual accounts and adopted budgets for the coming years; - assessment of compliance of government fiscal policy with the long term sustainability of public finances due to population ageing; - assessment of the transparency of public finances and the quality of economic forecasts used in the process of budget preparation; - evaluation of the effectiveness of implementation of structural policies in terms of providing long-term sustainability of public finances, economic growth and employment; - assessment of general government debt management and guarantees schemes in terms of sustainability and stability of public finances._ The main issue concerning the assessment of fiscal policy is accountability: to whom the government is or should be accountable for delivering on fiscal stability objective? Clearly in the context of the democratic set up the answer is the parliament which is co-responsible for fiscal stability. In fact the parliament already monitors annual budget execution and approves budgets. However, with regard to the fiscal stability objective its powers are limited as it cannot alter significantly budget appropriations and does not frame fiscal targets. Furthermore, notwithstanding its responsibility in approving budgets and monitoring final accounts, its responsibility for monitoring fiscal stability objectives or fiscal strategies is not explicitly defined. Thus once the fiscal responsibility objective is made explicit by requiring the government to be abided by it then the issue is whether the responsibility for monitoring government's compliance should be outsourced to an independent institution or delivered directly by the parliament. When the objective of fiscal stability is formally introduced in policy making it implies the need for the government to define fiscal strategies incorporating this objective and for the parliament (the highest policy and political body within a framework of democratic accountability) to scrutinize the government adherence to the objective. Outsourcing the responsibility for surveying compliance with the objective to an independent institution can be considered as a contradiction since the most important roles of the parliament is scrutiny of the government policies or, in other words, "holding the government to account". The parliament responsibilities cannot be selectively determined with regard to which issues it is going to hold the government to account and in which issues not as this would contradict its own role. Outsourcing the parliament responsibility to fiscal surveillance to an independent body will imply that both the executive branch that is in charge of designing the fiscal strategy and the parliament that is in charge of approving the annual budget (elected official and representatives) and monitoring the government become accountable to a body of non-elected officials without democratic legitimacy. When the government is compelled to deliver on the fiscal stability objective, then the appropriate set up (governance framework) would be one in which the parliament discharges the fiscal surveillance function directly by monitoring government's compliance with principles of fiscal responsibility. Under this arrangement, the parliament in addition to its current mandate of monitoring budget execution and government's fulfilment of objectives of budget programs would also monitor the government adherence to fiscal stability objectives. This task could be facilitated by the parliament's setting up a Parliamentary Budget Office similar to the CBO in the United States. The office could provide the parliament with nonpartisan analyses on the government setting and delivering fiscal stability strategy. To this end the existing Fiscal Council could be transformed into a Parliamentary Budget Office in charge of preparing reports and analysis to facilitate the parliament task to deliver on the new mandate of monitoring government adherence to its medium term strategy and as appropriate to demand to the government rectification for deviations from the strategy's targets. An alternative approach would be to mandate the Fiscal Council with the task to prepare yearly reports for the Parliament consideration on government's strategy fulfilling fiscal stability objectives and adherence to fiscal targets endorsed by the parliament. These reports could be considered by the parliament at the time of yearly budget approval. Such an approach could be also considered to be consistent with a governance structure of the fiscal framework in which the roles of the Parliament (approving and monitoring fiscal strategies consistent with fiscal stability objective) and government (elaboration and execution of fiscal strategy) are preserved and aligned with democratic set up based on the principle of delegation. Two other issues concerning the functioning of the Fiscal Council as it currently exists deserve attention. One is the effective delivery on its own mandate and who monitors that this is the case (i.e. accountability) and the other is the technical capacity to carry its function which hinge on the overall credibility of the current fiscal framework. The Fiscal Council according to its mandate has a role of an independent institution in charge of ex-post monitoring government policy adherence to fiscal stability objectives and observance of quality aspects in conducting fiscal policy (i.e. quality of public finances). The role of the Council as stated in its mandate (Box 2) is to provide positive assessment of government fulfilment of fiscal policy and not normative assessment (i.e. to state "what is" and "what will happen if" not "what should or ought to be"). This is because fiscal policy decisions entail income redistribution and decision on taxation/transfers comes from elected representatives ("no taxation without representation"). While it is argued that monetary policy also affects income redistribution and it is carried by an independent institution (Calmfors and Wren-Lewis, 2011), it can be said that the objective of monetary policy is price stability while the objective of fiscal policy is primarily of sustainable redistribution and mitigating effects of monetary policy and shocks in fulfilling its objective.20 On average changes in interest rates across business cycle should cancel each other while fiscal policy decision on average has redistributive bias. A similar issue relates to the size of the government that reflects preferences of electorate, tradition and values. Therefore, according to the Fiscal Council's mandate it could be expected than when scrutinizing fiscal policy of the government to draw findings such as "the government fiscal targets in the budgets 2011 and 2012 are in line with the government medium-term fiscal strategy" or "the fiscal targets have been postponed" but not statements such as "the government should set a different fiscal target" or "the government should rise such tax" which reflects Council's preferences. In a democratic setting where policy makers are accountable to the electorate, their task is the design of fiscal policy including targets and as appropriate to outsource needed advisory services to deliver on their mandate. Having an institution besides the executive branch dictating policy (i.e. non-elected individuals formulating policy) would imply another layer of power in the constitutional set up in addition to those of the executive, legislative and judiciary powers. The key issue is that fiscal policy underpins political preferences regarding a given social model and policy decisions in pursuance of policy targets entails distributive and efficiency tradeoffs in terms of taxes, social transfers and provision of services etc. While elected officials are accountable for policies, non-elected officials are not-accountable for their policy suggestions. Furthermore, there is no consensus as to what is an optimal policy in a particular domain. This is the case of structural policies and 20 The EU-27 average share of expenditure in social protection in total expenditure was 65% in 2008 (IMAD, 2010). fiscal policies which have distributional and efficiency aspects. Even in the case where an independent fiscal council would only set a given fiscal target (e.g. government budget surplus of 0.5% of GDP or expenditure-to-GDP ratio at 38%) such a decision would entail trade-offs in terms of distribution and efficiency. Under the existing fiscal framework it is not clear to whom the Fiscal Council is held account to in delivering its tasks and whether its mandate is appropriately fulfilled. One of the key issues in this regard is whether the Council is providing positive or normative assessment on fiscal policy. Whether the Council is setting the fiscal framework or formulating policy in different fields such as taxes. So far based on the output generated by the Council it would seem appropriate to separate the advisory function of the Council, which could be directly assumed by an advisory body to the Ministry of Finance or to the government (e.g. Strategic Economic Council), and the strict monitoring function that could be assumed directly by the parliament supported by a Parliamentary Budget Office.21 Alternatively the mandate of the Council should be streamlined by demanding the Fiscal Council to present a yearly independent report on government's adherence to fiscal strategy and fiscal stability objectives. Such a report could also serve as a basis for regular monitoring of Councils adherence to its mandate which is currently lacking within the existing framework.22 The last issue concerns the Fiscal Council's institutional capacity to deliver on its mandate and thus its credibility. While the Council can outsource analyses for its policy evaluation, in order to deliver its mandate effectively it would require enhancing its capacity (e.g. full time employees) which in the context of scare capacity and resources would have financial implications and entail duplicity of tasks performed by other institutions. While resources allocated for analysis could be assigned to wages of full time employees, in the context of existing institutional set up it would seem more appropriate to use existing institutions (e.g. IMAD) to provide technical assessment on the government's fulfilment of fiscal stability objectives. Alternatively, the government could report directly to the parliament on the tasks included in the Fiscal Council's mandate at the time when the budget is approved (e.g. reporting on fulfilling fiscal targets and on corrective measures to be introduced in the case of deviation). The parliament in turn when assessing the government's report on fiscal stability could resort to the support of a Parliamentary Budget Office that could be established for this purpose or to a tailored report from IMAD on government's compliance with fiscal stability targets. 3.2.5 The EU fiscal surveillance dimension Another important dimension concerning fiscal policy making and monitoring is the compliance with requirements under the Stability Growth Pact and its fiscal surveillance framework. This framework is often updated and currently this is the case. One of the most prominent changes concerns the introduction of the so-called EU semester. The most important feature in terms of policy formulation concerns an ex-ante economic and budgetary surveillance by the EU Council and Commission. It consists of two phases. In the first phase, the EU Council (supported by the EU Commission) provides horizontal policy guidance (Spring) to be incorporated in the elaboration of Stability Program by a all members and; in the second phase, the EU Council provides as appropriate country's specific recommendations on macro-structural policies to be taken into account by government and parliaments in the preparation and approval of the annual budgets (June/July). Later in autumn national parliaments adopt budgets according to each country's fiscal and legal framework. 21 See 2010 Report of Fiscal Council and Opinion: http://www.gsv.gov.si/si/fiskalni_svet/ . 22 The Council activities are financed with tax payers money and therefore it is expected that its activities reflect its mandate. Another important policy change under consideration is the introduction of an EU Council directive on requirements for budgetary frameworks endorsing the introduction in EU countries of numerical fiscal rules to foster compliance with SGP in the area of budgetary policy23. Furthermore, and what was already approved is that EU Council as part of the EU semester will provide, as appropriate, concrete recommendations on national fiscal rules and institutions. With regard to the eventual new requirement on introducing numerical fiscal rules in line with SGP, concretely to comply with the debt (60% of GDP) and deficit criterion (3% of GDP), it would seem natural to update the existing national fiscal framework to account for these two constraints which are binding as part of the international commitments. While the existing framework should be aligned with EU legal requirements, the ownership of the policy should be firmly anchored in the domestic framework and in the roles of the government (elaboration of fiscal strategy) and parliament (approving and monitoring government's compliance with the strategy) to ensure accountability and credibility. While EU fiscal surveillance requirements, now with and ex-ante dimension (EU semester), can stir fiscal responsibility, the domestic fiscal framework in itself should be the base for account stable policy. The next section elaborates on concrete proposals on this regard. 23 http://ec.europa.eu/economy_finance/een/019/article_88106_en.htm 4 TOWARDS ENHANCING THE DOMESTIC FISCAL FRAMEWORK Previous sections addressed policy implementation and the fiscal framework with regard to the issue of fiscal stability and identified weaknesses and strengths in policy delivery and in policy formulation. This section elaborates on proposals that can improve the existing fiscal framework with regard to the issues of fiscal stability, accountability and transparency. Under the proposed enhanced policy framework the objective of pursuing fiscal stability over the medium term should be made explicit in the legislation (Act on Public Finance). The cornerstone of the new fiscal framework should be a clear legislative and regulatory structure defining the responsibilities of the executive and parliament with regard to fiscal stability. The government should be in charge of defining the medium-term fiscal strategy and the parliament should be in charge of adopting the strategy and monitoring its implementation. There are various dimensions of the fiscal framework that can be enhanced in order to facilitate that the share responsibility of the government and parliament with regard to stability of public finances is delivered. These involve the existing set of arrangements, procedures, rules and institutions that underlie the conduct of budgetary policies of general government. In particular it is important to improve the way they are set with regard to planning, approval and monitoring of fiscal strategy. 4.1 Accounting and statistics Currently the fiscal reporting which is important to understand the fiscal position is comprehensive and timely, although reporting on budget execution on ESA 95 methodology lags behind reporting on cash methodology which could be improved. This can be done by requiring the Ministry of Finance to provide monthly estimates of budget figures based on ESA 95 methodology. An important category over which information is not available is on tax expenditure.24 The law on Public Finance could be amended requiring the Ministry of Finance to make information on tax expenditures regularly available. 4.2 Macroeconomic and fiscal forecasting Until recently (Official Gazzete No. 54/2010) the fiscal forecasting was based on a single forecast done by IMAD, which is regarded as a realistic macroeconomic forecast using the most up-to-date information. The obligation of the use of this forecast in the past contributed to anchor budget planning and deliver a rate of government expenditure growth that can be considered as a prudent rate. Furthermore, the macroeconomic forecast made by IMAD was regarded as independent forecast by the EU Commission and as a successful feature of the Slovenian macroeconomic framework25. The fiscal framework could be improved by amending 24 Revenue that a government foregoes through the provisions of tax laws that allow: (1) deductions, exclusions, or exemptions from the taxpayers' taxable expenditure, income, or investment; (2) deferral of tax liability, or; (3) preferential tax rates. 25 EU Commission (2009). "Domestic Fiscal Frameworks: How National Fiscal Governance Can Contribute to Budgetary Consolidation Over the Medium term" the Law on Public Finance and re-establishing the previous practice of using the IMAD forecast as the solely source of macroeconomic forecast for budget preparation. Under the existing framework the responsibility for fiscal forecast lies on the Ministry of Finance which seems appropriate in view of information and expertise on the matter. It is also consistent with a clear division of labour. Nevertheless, to enhance transparency it can be required that the fiscal forecasts for budget planning to be subject to regular and comprehensive evaluation. Similarly, IMAD and Bank of Slovenia could publish their fiscal forecast to benchmark that of the government. 4.3 Medium-term budgetary framework The orientation of the medium-term framework with regard to the objective of macro-fiscal stability could be enhanced in various ways including: the setting of fiscal targets; the way fiscal targets are reflected in annual budgets; the procedural rules ensuring the delivery of targets; the quality aspects of policies in reaching the targets and; the type of information required for monitoring the targets. 4.3.1 Setting the targets in line with fiscal stability objectives One of the key elements missing in the current framework is the explicit requirement for a government to set fiscal targets in relation to macro-fiscal objectives. While recent changes to the national policy framework to derive expenditure ceilings (Box 1) implies the need to set debt and primary balance targets, in the existing framework (Law on Public Finance) there is no reference to them, on how they should be set or their validity. Furthermore the EU fiscal legislation requires that countries set and abide to a medium term objective (MTO) which is not formally addressed in the existing framework and neither there is a link between the fiscal targets used in setting the ceilings and the MTO. The Decree of Budget Planning only mentions the idea that with the new framework for deriving expenditure ceilings a balance position in structural terms can be reached. Yet, this is not necessarily the case if account is not taken of revenue developments and tax policy. The Law on Public Finance could include a reference with regard to setting fiscal targets and its link to the MTO. A requirement for a new government in the first three months to specify its policy objectives and targets for the duration of its term in office could be introduced. Targets should be incorporated in the macro-fiscal strategy of the government (budget memorandum) for a four-year period. The fiscal strategy could be drawn in a document that should replace the current budget memorandum and should be sent by the government for parliamentary approval and follow up. The government should be abided by the strategy and the law should specify the requirement for the government to rectify the strategy in the case the targets are missed. Temporary departure from targets could be allowed in the case of natural disasters or acute economic slowdown under the requirement for the government to set policy to correct the departure from targets. Alternatively, binding agreements on coalition partners could be drawn specifying macro-fiscal objectives and targets for the duration of the government (Lienert, 2010). Such objectives and targets should be in turn part of the fiscal strategy of the government to be approved and monitored by the parliament. 4.3.2 Criteria for assessing consistency of fiscal targets with fiscal stability objectives Requiring the government to set targets with regard to macro-fiscal stability objectives is a step forward in terms of accountability for fiscal stability. The issue is to determine the extent to which the targets meet these objectives. In particular, no consensus exists as to the optimal level of debt or size of net government assets. Reality makes these choices difficult as for example arguments for higher investment or lowering taxes can be made in view of their effect on the economy but which could in turn increase debt. Thus, changes to the Law on Public Finance while providing governments with the ownership of setting targets and following them during their term in office they should also provide guidance (i.e. criteria) for assessing what can be understood to be a stable fiscal policy. Such a criteria could be derived taking into account: a) the recent crisis, its effect on Slovenian economy and public finances which has highlighted the critical importance of the debt and its relative size as the main stabilizer and macro-economic shock absorber; b) the need for the fiscal balance to observe a safety margin not to breach the 3% of GDP deficit threshold according to international obligations (which in the case of Slovenia before the crisis was estimated in a structural deficit of -1.1% of GDP) and; c) the need to ensure debt sustainability and observance of the safety margin including the accounting of contingent liabilities over the medium term. In setting medium term targets, given the fact that different fiscal variables are interlinked, an order of precedence among them should be decided as it would seem conflicting to choose two targets simultaneously. Based on recent experience and the rationale put forward in Section 2 (i.e. effect of indebtedness on cost of funds for the economy and macroeconomic shock absorber), it would seem appropriate to anchor the fiscal stability objective in terms of a debt-to-GDP ratio target. Such a choice seems to be consistent also with an optimal welfare maximizing fiscal constraint.26 However, since there is not optimal level for a given debt target and reaching it depends also on economic growth which under current circumstances is low, government debt targets could be benchmarked against a trend decline consistent with reaching a debt level ratio already experienced in the past; the level observed prior to the crisis below 23% of GDP. Although this is an arbitrary debt level, it provides a good reference about the type of fiscal space necessary to observe to cope with an unprecedented shock as the one experienced in 2009. Thus, the first criterion to assess the alignment of given government fiscal targets with macro-stability objectives would be to determine whether they are consistent with a sustained trend decrease in the debt-to-GDP ratio towards a medium to long-term debt target (e.g. 20% of GDP).27 Assessment of compliance with this criterion (e.g. speed of reducing the ratio) should take into account the face of the cycle and macroeconomic risk (e.g. downgrading of credit rating and exceptional events). Defining a long-term target for the debt-to-GDP ratio implicitly determines the government balance (and structural balance) consistent with it and vice versa. Furthermore, assuming that expenditure grows in line with nominal GDP the debt target is also consistent with a given or desired expenditure-to-GDP ratio which covers desired policy priorities. Figure 52 shows that after a process of consolidation until 2016, if the long-term target for the debt is chosen to be 18% of GDP and if the economy grows at an annual nominal average rate of 4.5% the long-term structural deficit compatible with that debt target will be 0.8% of GDP. Figure 56 also shows that if the expenditure grows in line with GDP from 2016 onwards the expenditure-to-GDP ratio will remain constant at the level reached in 2016 (44.3% of GDP). 26 According to Bi and Michael Kumhof (2009) the optimal constraint or rule in terms of welfare is far more aggressive than simple structural surplus rules or automatic stabilizers. 27 This could be feasible starting 2012 under an alternative consolidation policy scenario (Caprirolo 2010). Figure 56: Fiscal variables under a debt-to-GDP target ratio of 18% (scenario of 4.5% nominal GDP growth from 2016 onwards) 60,0 50,0 -40,0 ■ 30,0 20,0 ■ 10,0 0,0 -10,0 50,4 ..................... I Government balance % GDP (-0.8% target) ■ Debt-to-GDP ratio ■Expenditure% GDP (RHS) 51,0 50,0 49,0 48,0 47,0 46,0 45,0 44,0 43,0 42,0 41,0 o o CM CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM O LO O CM Source: own estimation A second criterion for benchmarking a given government policy target with the principle of fiscal stability would be to assess if the resulting structural balance is consistent with the requirement of observing a safety margin for not breaching the 3% of GDP deficit and with a maximum structural deficit for euro area countries of 1% of GDP. In the case of Slovenia the structural deficit satisfying that level is of 1.1% of GDP. Thus governments could chose to run fiscal policies with a maximum structural deficit of 1% of GDP provided that contingent liabilities are addressed by means of structural reform. Notice in Figure 48 that a structural balance of 1% under similar GDP growth assumption is also consistent with a long-run debt target of 20% of GDP. Benchmarking government policies against the maximum structural deficit for not breaching the 3% of GDP deficit threshold assures consistency with the EU obligations with respect to setting and adhering to an MTO. It also makes transparent the link between a long-term debt fiscal target derived from economic rationale and the MTO, a link which is currently missing in the existing framework. While the Law on Public Finance could mandate governments to set the level of the MTO and adherence to them, guiding policy base on structural balances is very uncertain as real time estimates of output gap vary enormously with respect to ex-posts estimates (Section 2). Such an arrangement would entail the risk of credibility loss. For a government to cater for the uncertainty of output gap estimates if following policy based on a structural balance target this will require running substantial surpluses (e.g. Sweden before the crisis) which in turn implies high tolerance for high taxation. A better solution would be to let the MTO at current level and deliver a better outcome depending on the degree of policy ambition of incumbent governments. Furthermore, under current conditions where a process of fiscal consolidation has to take place and the distance to the MTO is still large and its measurement uncertain, changing the level of existing MTO will be simply irrelevant as reaching it requires relative longer process of consolidation or an alternative consolidation path to that currently traced. In any case, credibility should be earned by first delivering on existing targets. Notice also that under current process of fiscal consolidation assessing compliance with the minimum safety margin is not the key priority but making sustained and sustainable progress towards reaching it. The third criteria for benchmarking government targets with stability objectives concern addressing future liabilities associated with population ageing. Meeting the objective of keeping existing debt ratio in line with its long term level and not breaching the safety margin consistent with the 3% deficit requires addressing the issue of contingent liabilities. As mentioned in Section 2 this in the context of EU countries is made by means of the so-called three pronged strategy of increasing employment, reducing debt and reforming pension systems. Furthermore, the EU Commission will assess the degree to which countries' fiscal stance would frontload contingent liabilities according to a chosen frontloading parameter. Since the estimation of long-term liabilities is subject to a large degree of uncertainty and addressing the issue requires ownership of policy including the decision among the components of the policy mix, a criterion for benchmarking the degree of consistency between the government's strategy with the stability of long term public finance could be derived. The proposed criterion would consist in assessing, on a rolling over basis, each time a new government is formed whether structural balance projections over the next 10-year period are consistent with the criterion of observing a structural deficit position below the safety margin. If this is not the case, then governments would have to determine the degree of frontloading liabilities to cater for the liabilities above the safety margin in the 10-year horizon or enact further pension reform measures or carry related structural reform. 4.3.3 Assessment of adequacy of fiscal targets with fiscal stability objectives Once a newly elected government sets the fiscal targets to be part of the fiscal strategy during its time in office, a newly established Parliamentary Budget Office or the IMAD could assess the consistency of the targets with the objective of fiscal stability following the criteria set above (4.3.2) and present its results in a report to the parliament for consideration. The report could be included with the government fiscal strategy and budget documents sent to the parliament for approval. Such an approach would be similar to that performed by the CPB in the Netherlands when assessing fiscal policy in provisional coalition agreements (see Box 3). Box 3: The Central Planning Bureau (CPB) in the Netherlands CPB is independent expert institute not involved directly in economic and fiscal decision-making process (Bos and Teulings 2010). Nevertheless, it has access to confidential information and its forecasts and analysis are used in the decision-making process. The role of the CEPB is comparable to that of the Congressional Budget Office in the US but performs wider analysis than budgetary analysis. Formally the CPB is a branch of the civil service within the Ministry of Economic Affairs. But the informal rules of conduct, procedures and track record have resulted in CPB having a strong independence. The CPB tasks related to fiscal policy making include the following: - The provision of the official estimates on economic growth, purchasing power and unemployment (used by the Ministry of Finance in the elaboration of fiscal forecast); - Forecasts and analysis of Dutch public finance in short and medium term; - Analysis of the sustainability of Dutch public finance; - Analysis of the economic effects of the election plans of Dutch political parties (stipulating plans for next four years) and (provisional) coalition agreements carried out before the start of a new government which are regarded as a critical benchmark for fixing the medium-term fiscal framework. Other related tasks include: - Cost-benefit analysis of major infrastructure projects; - Studies on the best way to organize major areas of government revenue and expenditure, e.g. social security, health care, education, housing and taxation; - Long run forward looking studies and many background studies; - Participation in government advisory groups. 4.3.4 Approval of the fiscal strategy and reporting on its compliance The government fiscal strategy including medium-term fiscal targets and the annual budgetary strategy for attaining the targets should be submitted to the parliament for debate and approval. The fiscal strategy could replace the current budget memorandum or its content could be aligned with the new required structure. The strategy should state the targets and include consistent projections of the general government expenditure, revenue (including tax expenditures), deficit, structural balance and debt. Projections should be presented in terms of both cash and ESA 95 methodologies and explaining their consistency. The strategy should include a description of medium-term policies with an impact on general government finances broken down by major revenue and expenditure item. The strategy would present alternative forecast scenarios and corresponding budgetary projections highlighting how fiscal variables would evolve under different economic assumptions. The Strategy would also present estimation of old age related expenditure for the next 50 years and reform scenario with impact on expenditure and replacement rate. A table reconciling changes in debt and deficit specifying the reasons of discrepancies should be included. The stock of guaranteed debt, called guarantees and planned granting of guarantees should be reported in the same document. The report should include a detailed account of tax expenditures. Every year as part of the budgetary process the government would update the strategy by reporting on fulfilment and deviations of fiscal targets. Reporting should be done on cash and ESA 95 methodologies. Deviations from targets should be explained in terms of macroeconomic developments and the effect on revenue and expenditure. The reasons from the deviation should be broken down into changes in forecast, policy and accounting changes. In the case of deviations the strategy should include measures to rectify them. By providing the parliament with the mandate of approving the government fiscal strategy the ownership of fiscal policy making is also enhanced vis-a-vis the recent changes in the EU framework which provided an ex-ante role to the EU in the fiscal policy planning. 4.3.5 Multi-annual planning One of the cornerstones of the current medium-term budgetary framework is the top-down budgeting process setting the ceilings on expenditure limiting the total amount of State budget resources which is subsequently allocated in expenditure areas and programs. While expenditure ceilings until recently were set only for the State budget (Official Gazzete No. 54/2010), coordination between the Ministry of Finance with the social security fund and local governments allowed for consistency for planning the overall general government expenditure and balance. The framework was and still underpinned by two fiscal rules (i.e. establishing a permanent constraint) on Pension fund (balance budget) and local government budgets (debt limit set at 20 % of the previous year revenue). Until the approval of the Decree on Development Planning in 2010 the budgetary procedure did not maintain the expenditure ceiling determined for two years regardless of macroeconomic circumstances, but provided for the option of adjusting them in light of changing macro-economic circumstances and commitments adopted. As discussed in Sections 2 and 3 despite this option there were no major upward revision of expenditure in revised or adopted budgets with the exception of 2003-2004 and 2008. The Decree on Development Planning brought changes to the scope and the way expenditure ceilings are set. With regard to the coverage the ceilings do not concern only the State budget but the whole general government budget as defined under the national methodology and does not only cover domestic financed expenditure but also expenditure financed with EU funds. With regard to how the ceilings are set the Decree introduced a framework linking desired targets (primary balance and debt) to the expenditure ceilings and determined that they should be set for four years and to be fixed for current and next year. The existing framework could be enhanced with regard to the definition of the coverage of the ceilings, their nature, how they are determined and supporting arrangements (rules). Setting a ceiling on local government expenditure raises legal issues that can question the ceilings themselves. Thus the ceilings should exclude local government expenditure which can be indirectly controlled by applying effectively the existing debt rule and by means of budget transfers. Expenditure financed with EU budget sources should be excluded from the ceilings as their dynamic is not fully predictable. Besides, it is on tax payers' interest to ensure the full withdrawal of EU resources. Similarly, expenditure ceiling could exclude interest payments on debt as this category is not under discretion of the government. Expenditure ceilings should apply to domestic financed State budget expenditure but overall consistency at the level of general government should be ensured by enforcement of rules applied to local governments and Pension Fund and their coordination (see Table 7 for international comparison). A balanced budget rule applied to Health Fund could be also introduced similar to that applied to the Pension Fund by which deficit of Health Fund will be covered by means of State Budget's transfers. Since the government strategy and its planning will be elaborated and reported in cash and ESA95 basis and the coverage is different, consistency should be ensured in a way that expenditure ceilings are consistent with targets set in both methodologies. Given the fact that ESA 95 methodology is the international standard for public accounts, careful planning should ensure that the link between the expenditure ceilings, total expenditure aggregates and fiscal targets is ensured and disclosed. Table 7: Aggregate expenditure ceilings in a sample of countries Coverage Features Country Level of detail Time horizon Social Debt Local % of public spending Rolling or Frequency of Sec. Interest Gov. fixed revision Total 2 fixed + 1 Sweden Yes No Transfers 64 Spending 3 rolling Every year Total Finland Some No No 36 Spending 4 4 fixed Every 4 years Netherlands Yes No Transfers 80 4 Sectors 4 4 fixed Every 4 years Source: Cangiano (2010) Since 2010 expenditure ceilings are determined on April of the current year for a four year time horizon (the current year and the following three years). The Decree is not clear as to whether the time horizon of the validity of the ceilings exceeds the time of a government in office. It mentions that revision of the ceilings, when they are deemed to be fixed, is possible only when macroeconomic conditions require further consolidation effort. The ceilings are set to be fixed for the first two years (current and next year), and the other two are indicative. Currently this implies that for 2010 and 2011 ceilings are fixed and indicative for 2012 and 2013. In 2012 election year the ceilings will be set fixed for 2012 and 2013 and indicative for 2014 and 2015. This entails that the ceiling set by a previous government is binding for the new government and new legislature during the first year after the election and indicative for two consecutive years. This means that previous government policy is imbedded on the new government policy rising concerns on accountability. Under the budget procedure before the adoption of the Decree on Development Planning the Parliament under proposal of the government approved two year budgets which also extended beyond the mandate of the incumbent government. However, the budget procedure allowed for budget revision or supplementary budget in case of forecast revision and new policies. Under the new procedure, which might become legislation the decision of a previous government concerning fiscal strategy will become binding for the new one and legislature blurring accountability. Since the aim of a robust policy framework is ownership and accountability, the responsibility of setting fiscal ceilings should be aligned with political responsibility and they should be determined taking into account that their purpose is to facilitate the achievement of fiscal targets. Accordingly, expenditure ceilings could be determined following two alternative options linking their time horizon with the term of a government in office. The first option would consist in setting expenditure ceilings in a 6-year time horizon consisting of a four-year fixed ceilings period (coinciding with political cycle) and two-year indicative ceiling (coinciding with obligation of preparing two years budgets in the last year of a government in office). Once a newly elected government is in office, in the month of April will set ceilings for the next four year period. The ceilings will be fixed for the current and next year. In the second half of its mandate the government in the month of April will set ceilings for the next four years. The ceilings will be fixed for the last two years of its mandate and it will be indicative for the two years following the end of its political mandate. Four year fixed expenditure ceilings coinciding with the validity of a government fiscal strategy will be applied. The second option could consist in re-establishing the budget procedure followed before the change in the Decree on Development Planning by which budget expenditure was set in a two-year framework on rolling basis with the option of adjusting the ceilings in light of changing macro-economic circumstances and commitments adopted but now with the legal obligation for a government's adherence to the fiscal targets set for a four-year period at the beginning of its mandate in its fiscal strategy (i.e. including projection of government expenditure, revenue and balance). Regarding the issue of how expenditure ceilings are determined what matters is that they underscore the fiscal targets set for a four year period by the government in line with the objective of fiscal stability and criteria of compliance (i.e. Decreasing debt ratio towards the long term target and observance of safety margin with regard to 3% of GDP deficit also including long-term contingent liabilities). If the existing framework introduced in the Decree on Development Planning is still used to derive the ceilings, then it should be corrected to take into account the observations made above regarding the definition of scope of expenditure covered and providing pre-eminence to the debt-to-GDP target. Alternatively, in setting the ceilings the government could follow a prudent principle of fiscal policy making by restricting expenditure growth below forecasted average GDP growth over next five year period as for example shown in Figure 57. Similarly, ceilings could be derived and adjusted in line with fiscal targets and priorities for example by following a simple rule of maximum nominal growth rate of expenditure of 3% per year. The bottom line would be for the government to set and abide to fiscal targets rather than to instruments. Figure 57: Fiscal variables under scenario of expenditure growth (4.5%) lagging behind nominal GDP growth (5%) from 2016 onwards 60,0 50,0 40,0 ..... 30,0 ..... -rn—rn—rn—rn—rn—rn—rn—rn—rn—n—[—r I Government balance % GDP (-0.8% target) ■Expenditure% GDP (RHS) -Debt-to-GDP ratio Q- Q (D sS 20,0 ..... 10,0 ..... 0,0 -10,0 60,0 40,0 20,0 0,0 s? -20,0 -40,0 (D -60,0 .to -80,0 -100,0 -120,0 -140,0 ocg^cDcoocg^cDcoocg^cDcoocg^cDcoo -.-■.-■.-■.-■.-cgcgcgcgcgcococococo^^^^^LO Source: own estimation 4.3.6 Numerical fiscal rules The numerical rules applied to the Pension Fund and local governments contribute to achieve fiscal targets defined for the general government. An additional rule applied to the Health Fund similar to that applied to the Pension Fund would also facilitate the predictability of planning and delivering fiscal targets. With regard to the municipalities while the current rule limits their indebtedness in the recent years the overall indebtedness of municipalities has increased. Since the limit on total debt is set against government revenue, heavily indebted municipalities have the incentive to increase tax and non-tax revenue to both pay obligations but also increase the indebtedness capacity. To enhance transparency and reduce incentives of increasing debt and tax burden, the Ministry of Finance should publish regularly the indebtedness position of individual municipalities, the debt to revenue ratio and the change in tax and non tax revenue. An annual report should be submitted to the parliament on the municipalities' fulfilment of the debt rule. To ensure consistency with the EU fiscal framework the Law on Public Finance could be amended and include two numerical rules with regard to the deficit (3% of GDP) and debt (60% of GDP) (i.e. The Maastricht criteria). Consideration could be given also to a rule linking revenue developments to a prudent rate of medium-term economic growth. Such a rule could serve as an anchor the policy aiming at meeting the MTO in view of the uncertainty surrounding structural balances estimates and the importance of maintaining a stable revenue structure. Such an approach would also cater for adjusting structural budget balances for asset price cycles. 4.3.7 Procedural rule on revenue neutral tax reforms As discussed in Section 2, to deliver fiscal policy consistent with the objective of fiscal stability is not enough to place a ceiling on expenditure. Fiscal policy is made also in relation to revenue developments and there is evidence that tax reform reducing revenues is pro-cyclical in good times. This has been also the case of recent experience in Slovenia with the 2006-2007 tax reform that was not revenue neutral. In addition, most recent experience (in the year 2010) as discussed in Section 3 shows that in bad times or in periods of fiscal consolidation expenditure ceilings are not binding or not enough to ensure consolidation and what matters is the political commitment to achieve given fiscal targets. In particular in 2010 government revenue outturn was lower than that originally budgeted and that planned in the supplementary budget. If the government would have simply adhered to the ceiling then the deficit would have been higher. This highlights the issue of discretion rather than of rules at the time of fiscal consolidation. A pragmatic approach had to deliver a deficit that was close to that originally targeted. In view of the pro-cyclicality of tax reforms and their implementation in a non-revenue neutral manner a procedural rule concerning revenue neutral tax reform should be introduced in the Law on Public Finance. The rule should state that any proposal that would result in a reduction of government tax revenue (tax rate or granting tax expenditure) should be at the same time fully offset by a measure that either increase alternative source of revenue or reduce permanently expenditure. 4.3.8 Performance orientation While the link between program budgeting including performance orientation and fiscal stability might not be so clear, in reality such an approach could contribute to deliver fiscal outcomes that are more sustainable as it allows for an in-depth scrutiny of expenditure programs and priorities. The recent change in planning the budget, now based on programs and not on economic classification is a good step forward that needs consolidation as yet expenditure based on economic classification still underpin budget planning or setting the expenditure ceilings. Such a procedure should continue and performance orientation should be effectively introduced (i.e. non-financial performance targets associated with each program with regular reporting on progress). The Article 62 of the Law on Public finance which requires the government to report to the parliament on the achievement of results and objectives of adopted programs could be adjusted to take into account performance orientation. Similarly, the Law on Public Finance could require the Ministry of Finance to carry out periodic expenditure reviews to assess program performance and to report to parliament. 4.3.9 Monitoring government's adherence to fiscal stability objective To ensure accountability for fiscal stability in a credible, predictable and transparent manner within the democratic setting the parliament should be given explicit mandate to ensure that the budgets are consistent with long-term fiscal stability objectives. Given the parliaments' limited mandate to modify budgets but its' co-responsibility for the underlying fiscal strategy by approving annual budgets, it should discharge its responsibility with regard to fiscal stability by approving and ensuring that the government's fiscal strategy as mentioned above (4.3.4) is implemented. Entrusting the government and the parliament with a more explicitly mandate with regard to fiscal stability objective provides the policy framework solid and sustainable foundation within the democratic framework. It aligns incentives for delivering fiscal policy objectives with political commitment. Alternatively, similar aim could be reached by outsourcing the government's responsibility of designing fiscal policy to a rule (e.g. requirement of structural balance position) and the monitoring of fiscal policy in line with fiscal stability principles to an independent institution (e.g. fiscal council monitoring compliance with a rule). However, incentives for political commitment and its foundations are weaker since do not fit the democratic setting by which politically elected individuals are in charge of defining and delivering on key policy aspects and are accountable to the electorate. A key document accompanying the strategy for parliament consideration should be the assessment of compliance of the strategy with the criteria on fiscal stability as discussed before. The parliament during the implementation of the strategy could receive from the government at the time of budget approval a yearly update of the implementation of the strategy. The content of the Budget Memorandum could be modified for this purpose and the government would explaining on the fulfilment of its objectives. In case that the targets would not be met it would specify the policy for restoring adherence to the targets. The report would be adopted by the parliament. In addition, at the time of budget approval an independent annual report assessing the implementation of the strategy's objectives and targets could be provided for consideration to the parliament and elaborated to this purpose by a newly created Parliamentary Budget Office (nonpartisian) that could replace the existing Fiscal Council or by IMAD. In the case that the existing Fiscal could be transformed in a Parliamentary Budget Office it would provide reports required for parliamentary discussion and orientation. The advisory function of the Fiscal Council could be allocated directly to the Ministry of Finance or to the Government. 4.3.10 Enhancing countercyclical role of fiscal policy Lowering the debt-to-GDP ratio in accordance with fiscal targets can provide the economy with room to offset economic shocks but also institutional features can be improved to enhance the strength of countercyclical fiscal policy. In particular, the ability and speed of withdrawing EU funds could be improved. Government targets with regard to EU funds withdrawal should be carefully monitored and as relevant institutional capacity could be strengthened taking into account expertise of relevant institutions (e.g. EIB). Also attention could be given to the impact of investment carried out by non-financial government enterprises. 5 CONCLUSIONS The paper looked at fiscal policy performance over the past years to draw lessons for enhancing the fiscal framework. The assessment of past fiscal developments indicates that there was not a tendency excessive debt accumulation, the debt burden as percentage of GDP remained fairly constant and declined before the crisis. Until 2007 there was a process of fiscal consolidation where deficits declined and disappeared in 2007. During that time expenditure grew below what can be considered a prudent rate of growth (i.e. below nominal average potential growth). Revenue planning for budget preparation exhibited an optimistic bias in bad times and a pessimistic bias in good times. Looking at data pertaining State budget planning and execution indicates that there were no major upward expenditure revision with exception of those in the years 2003, 2004 and 2008. During 2005-2007 total expenditure in revised and supplementary budgets was lower than originally approved in adopted budgets. The assessment of compliance with the MTO depends on whether this assessment is made with real time or ex-post data. Evidence regarding output gap estimates indicates that are quite volatile and revisions can shift the output gap both in size and sign suggesting that targeting fiscal policy in real time based on structural balances or assessing it is subject to high degree of uncertainty and can lead to credibility loss. Furthermore, given the uncertainty surrounding output gap estimates setting ambitious targets in terms of structural balance might require high tax burden to compensate for the uncertainty associated with output gap calculations. On the conduct of fiscal policy, evidence indicates that there was a degree of deficit bias beyond that deliberately chosen by the medium term fiscal target (i.e. Structural deficit of 1% of GDP) which is not explained only by discretionary policies enacted in the year where the budget was adopted but primarily by the delayed effect of past adopted policy (wage and tax policies). In fact, key policies influencing fiscal policy developments during the past decade are wage policy (compression and shocks in 2001 and 2008) and non revenue neutral tax reform whose effects in public finance take time to completely unfold. These aspects highlight the importance of sustainable policies to underpin any fiscal framework and policies including the current process of fiscal consolidation and of a strategy that takes into account the potential effect of unsustainable policies. Evidence regarding the use of fiscal policy and its strength to offset business cycle developments and shocks indicates that it is not only constrained by structural factors (e.g. the degree of openness of the economy and no-independent monetary policy) but also by institutional aspects such as availability of information (e.g. size of output gap in real time and quarterly GDP data) and budget procedure limiting sizable fiscal contraction or expansion in short run. This highlights the importance of the debt-to-GDP ratio as the prime shock absorber and of targeting its level over the medium term. Recent experience indicates that the debt ratio is a key macroeconomic stabilizer as it underpins the cost of funds for the whole economy but also the key fiscal absorber for unprecedented shocks such as the one recently experienced. The independent forecast of macroeconomic variables played important role in fiscal developments over the past decade as they anchored fiscal developments in good times (e.g. forecast of GDP growth figures used for budget preparation were lower than their respective outturn). The elaboration of fiscal projections based on projection of macroeconomic variables whose outcome was lower than forecast contributed to planned more conservatively government revenue and thus expenditure. Until 2010 the elaboration of the macroeconomic forecast was based only on a single forecast which ensured transparency and accountability, returning to such a practice regarded as best practice can be warranted. An assessment of the existing framework indicates that there is room for improvement. In particular, the objective of fiscal stability is not explicit within the existing framework and the tasks between the executive branch and parliament with this regard not well defined. While there is a robust medium term fiscal framework it does not spell out the requirement for the government to declare and commit to a monitoreable fiscal strategy that pursues fiscal stability nor there is a requirement for the parliament to exercise control on the fulfilment of medium term fiscal targets or the alignment of bi-annual approved budgets with medium-term fiscal targets. Recently a framework for determining binding expenditure ceilings with regard to optional targets in terms of primary balance and debt has been introduced. However, the framework is weak with respect to the coverage of the ceilings (e.g. EU funds), the procedure and prioritization of targets and overlaps mandates of different sectors of the general governments. In particular it abides local governments which are legally autonomous. The experience with the use of ceilings so far in 2010 indicates that there are not a binding instrument to ensure fiscal consolidation (i.e. the ceiling was revised downwards in the supplementary budget and the expenditure outturn was even lower) and that political commitment is the critical dimension in delivering targets as well as a cautious revenue forecast. The framework does not clearly align the time horizon of ceilings and the government policy mandate. To fill out the institutional vacuum with regard to monitoring the government's accountability for fiscal stability a Fiscal Council was established. Despite of its mandate (i.e. ex-post positive assessment of government compliance of fiscal targets), the institution carries out normative and positive assessment of government fiscal policy. Once the concept of fiscal stability becomes explicit in government policy formulation it implies the need for the parliament to discharge its responsibility with respect to the scrutiny of this aspect of the government's fiscal policy. The paper provides concrete measures to improve the fiscal framework with respect to the issue of fiscal stability. It provides a forward looking incentive structure to deliver prudent fiscal policy. In particular, it proposes a clear framework for accountability and transparency covering the roles of the government and parliament with regard to fiscal stability and policy making. Among the key proposal that could be incorporated in the Law on Public Finance is the requirement for an incumbent new government to spell out and commit to a medium-term fiscal strategy for the duration of its mandate in line with objective (measurable) criteria regarding fiscal stability. The parliament in turn is required to approve and monitor the medium term fiscal strategy of the government. To deliver the new task the parliament should be equipped with the appropriate technical support. For this purpose the Fiscal Council could be transformed in a nonpartisan technical budget office of the parliament or alternatively IMAD could provide independent assessment of government's strategy compliance with define criteria on fiscal stability. If the Fiscal Council is not transformed into a parliament budget office, it could also provide the parliament with a yearly assessment of government compliance with its fiscal strategy. Such a report should adhere strictly to the Council mandate and the current normative function should be transferred to an advisory body to the Ministry of Finance. To ensure overall consistency of the framework the paper also suggests additional measures including the introduction of a procedural rule ensuring revenue neutral tax reforms and of a prudent revenue forecast. The proposed policy framework not only improves democratic accountability for fiscal stability but by providing the parliament with the mandate of approving the government fiscal strategy it enhances ownership of the fiscal strategy vis-a-vis the recent changes in the EU framework which provided an ex-ante role to the EU in the fiscal policy planning. REFERENCES 1. Agnello, L. and Cimadomo, J. 2009, "Discretionary Fiscal Policies over the Cycle: New Evidence based on the ESCB Disaggregated Approach" ECB. Working paper series No 1118. 2. Barrios,S and Fargnoli,R. 2010, "Discretionary measures and tax revenues in the run-up to the financial crisis" European Economy, Economic Papers 419. European Commission. 3. Beetsma, R. and Giuliodori M. 2008, "Fiscal adjustment to cyclical developments in the OECD: An empirical analysis based on real-time data", CEPR working paper n.6692. 4. Beetsma, R., Giuliodori, M., Walschot, M. and Wierts, P. 2010, "Fifty Years of Fiscal Planning and Implementation in the Netherlands", DNB Working Paper No 260. 5. Bi, H. and Kumhof, M., 2009, "Jointly Optimal Monetary and Fiscal Policy Rules under Borrowing Constraints", IMF Working paper WP/09/286. 6. Bos, F. and Teulings, C., 2010, "CPB and Dutch fiscal policy in view of the financial crisis and ageing." CPB Document No 218. 7. Calmfors, L. and Wren-Lewis, S., 2011, "What Should Fiscal Councils Do?", CESifo Economic Studies 49. 8. Cangiano,M., 2010, "Restoring Fiscal Sustainability in Central and Eastern Europe: The Role of Budget Institutions", 6th Annual Meeting of OECD-CESEE Senior Budget Officials. 9. Caprirolo, C., 2010, "Slovenia: Alternative fiscal consolidation Policy", Zveza ekonomistov Slovenije, 12 izobaževalni seminar o javnih financah in državnemu revidiranju, zbornik referatov. 10. Cimadomo, J., 2007, "Fiscal Policy in Real Time". CEPII Working Paper, No. 2007-10. 11. Debrun, X., Hauner, D. and Kumar, M., 2009, "Independent fiscal agencies", Journal of Economic Surveys. Vol. 23, No. 1, pp. 44-81. 12. European Commission, 2007, "European economic forecast - spring 2007". 13. European Commission, 2008, "Public finances in EMU, 2008" . 14. European Commission, 2010, "European economic forecast - spring 2010". 15. Fiscal Council, 2010, Report of Fiscal Council and Opinions, http://www.gsv.gov.si/si/fiskalni_svet/ . 16. Forni, L. and Momigliano, S., 2005, "Cyclical sensitivity of fiscal policies based on real-time data", Applied Economics Quarterly 50. 17. Giuliodori, M. and Beetsma, R. 2008, "On the Relationship between Fiscal Plans in the European Union: An Empirical Analysis Based on Real-Time Data", Journal of Comparative Economics, vol .2 no. 36. 18. Golinelli, R. and Momigliano, S., 2006, "Real-Time Determinants of Fiscal Policies in the Euro Area", Journal of Policy Modeling, 28(9). 19. IMF - International Monetary Fund, 2009, "Fiscal Rules - Anchoring Expectations for Sustainable Public Finance". 20. Kopits, G. and Symansky, S., 1998, "Fiscal policy rules", Occasional Paper 162, International Monetary Fund. 21. Lienert, I. ,2010, "Should Advanced Countries Adopt a Fiscal Responsibility Law?", IMF. Working Paper WP/10/254. 22. Marcellino, M. and Musso, A., 2010, "Real time estimates of the euro area output gap: reliability and forecasting performance", ECB Working paper series No. 1157. 23. Ministry of Finance, Republic of Slovenia, Convergence and Stability programs, http://www.mf.gov.si/ . 24. Price, R., Joumard, I., Andre, C. and Minegishi, M., 2008, "Strategies for Countries with Favourable Fiscal Positions", OECD Economics Department Working Papers, No. 655. 25. UMAR, 2010, "Ekonomski izzivi 2010". 26. Varga, J. and in 't Veld, J., 2010, "The Potential Impact of EU Cohesion Policy Spending in the 2007-13 Programming Period: A Model-Based Analysis", European Economy, Economic Papers. 422.