UK public finances and the financial crisis Carl Emmerson and Gemma Tetlow Presentation given at workshop on “European public finances through the financial crisis”, ZEW Centre for European Economic Research, Mannheim, Germany, 11 June 2014. © Institute for Fiscal Studies Outline • Background: the state in the UK • UK economy before, during and after the crisis • Fiscal policy before the crisis • Fiscal effects of the crisis • Fiscal (and monetary) response to the crisis – Changes to taxation, spending and the fiscal framework – Distributional effect of changes to taxation and welfare spending – Did the tax and spending changes make the system more or less efficient? • Note: – UK fiscal years run from April to March – UK public finance aggregates differ from Maastricht definitions © Institute for Fiscal Studies UK spent around 40% of GDP publicly pre-crisis Total managed expenditure, 1948 to 2007 Percentage of GDP 55 50 45 40 35 Source: Office for Budget Responsibility’s public finances databank. © Institute for Fiscal Studies 2005-06 2002-03 1999-00 1996-97 1993-94 1990-91 1987-88 1984-85 1981-82 1978-79 1975-76 1972-73 1969-70 1966-67 1963-64 1960-61 1957-58 1954-55 1951-52 1948-49 30 Composition of spending over time Total managed expenditure, selected years 100% Share of public spending 90% 23.6 22.2 23.4 80% 7.5 2.2 5.2 7.2 5.4 3.4 5.4 5.7 12.2 13.3 14.4 17.2 15.8 14.4 10.4 13.4 11.7 1989-90 1999-00 2007-08 50% 10.0 3.6 4.9 10.0 40% 12.3 70% 60% 30% 20% 10% 0% © Institute for Fiscal Studies 11.6 13.5 Weak contributory principle: most ishas ondefence meansOffset Healthspending by spending declining become or health-tested benefits and increasingly debt interest important spendingand near-universal pensions Other Debt interest Transport Public order and safety Defence Education Health Pensioner benefits Working age benefits Source: Authors’ calculations based on data from HM Treasury and Department for Work and Pensions. Changing composition of revenues Net taxes and national insurance contributions, selected years 100% Share of total net taxes and NICs 90% 27.7 25.2 23.9 Other 2.7 11.0 3.3 9.8 4.5 7.9 Capital taxes 15.6 16.8 15.6 17.3 16.8 80% 70% 60% 50% 40% 30% VAT 19.5 NICs Income tax 20% 10% Onshore corporation tax 25.7 28.0 28.6 1989–90 1999–00 2007–08 0% © Institute Institute for for Fiscal Fiscal Studies Studies © Notes Source: andFigure sources: 2.5see of IFS Table Green 2.5 of Budget: The IFS February Green Budget: 2014. February 2014. GDP growth had averaged 3.2% a year over the decade up to 2007–08 Real GDP growth rate, 1956 to 2012 8 Growth rate (%) 6 4 2 0 -2 © Institute for Fiscal Studies Source: Office for National Statistics, Quarterly National Accounts (series ABMI). 2010-11 2007-08 2004-05 2001-02 1998-99 1995-96 1992-93 1989-90 1986-87 1983-84 1980-81 1977-78 1974-75 1971-72 1968-69 1965-66 1962-63 1959-60 1956-57 -4 Inflation had been low and stable since the mid1990s but rose during the crisis Growth in prices, 1949 to 2007 25 Retail Prices Index Consumer Prices Index Growth rate (%) 20 15 10 5 0 1949 1952 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 -5 © Institute for Fiscal Studies Source: Office for National Statistics (series DODO, DODP, DODQ, CZVJ, CRAB, CHAW, D7BT, KAB9, KAC4, KAC7). Employment rates had been rising steadily since mid-1990s Employment rate among those aged 16 to 64, 1971 to 2013 100 Men Women Employment rate (%) 90 All 80 70 60 50 Hours worked per worker had been declining over time and continued to do so 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 40 © Institute for Fiscal Studies Source: Office for National Statistics (series LF24, MGSV, LF25). Unemployment peaked at much lower level than seen during previous recessions Unemployment rate among those aged 16 to 64, 1971 to 2013 14 Men Women All Unemployment rate (%) 12 10 8 6 4 2 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 0 © Institute for Fiscal Studies Source: Office for National Statistics (series MGSX, MGSY, MGSZ). Average real earnings have fallen Growth in prices and average weekly earnings, 2007 to 2013 125 Retail Prices Index Consumer Prices Index Whole economy earnings Private sector earnings Public sector earnings Index, 2007=100 120 115 110 105 © Institute for Fiscal Studies Source: Office for National Statistics (series DODO, DODP, DODQ, CZVJ, CRAB, CHAW, D7BT, KAB9, KAC4, KAC7). 2013 2012 2011 2010 2009 2008 2007 100 UK fiscal policy prior to the crisis • Two fiscal rules – Golden rule: current budget must be in balance or surplus over the course of an economic cycle – Sustainable investment rule: debt must not exceed 40% of GDP – No official sanctions for breaching these – Perception that “the goalposts were moved” – redating the cycle • Economic and fiscal forecasts produced by HM Treasury, officially controlled by the Chancellor of the Exchequer • Public service spending totals set in cash terms for 3-year periods – 2007 Comprehensive Spending Review: covered 2008–09 to 2010–11 – Intended to be ‘tight’: cutting public service spending as % of GDP • Default was for (most) tax thresholds and benefit rates to increase in line with Retail Price inflation each year – Few reforms had been announced pre-crisis but not yet implemented: reduction in generosity of disability insurance, increase in pension eligibility age for women © Institute for Fiscal Studies Pre-crisis plan was for fiscal consolidation by 2012–13 Public sector receipts and total managed expenditure, 1997 to 2018 55 % of GDP 50 Revenues Spending Revenue forecast, Mar 2008 Spending forecast, Mar 2008 45 40 35 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 30 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. Borrowing was forecast to fall to 1¼% of GDP... Alternative measures of borrowing, 1948 to 2012 14 Public sector net borrowing Borrowing, forecast Mar 2008 Treaty deficit Borrowing (% GDP) 12 10 8 On eve of the crisis: UK had one of largest structural deficits in OECD and had done less than most to improve this over the preceding decade. 6 4 2 0 -2 -4 1948-49 1951-52 1954-55 1957-58 1960-61 1963-64 1966-67 1969-70 1972-73 1975-76 1978-79 1981-82 1984-85 1987-88 1990-91 1993-94 1996-97 1999-00 2002-03 2005-06 2008-09 2011-12 -6 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. ...and debt was forecast to peak just below 40% Alternative measures of debt, 1948 to 2012 240 Debt (% GDP) 200 160 Public sector net debt PSND, forecast Mar 2008 Treaty debt ratio National debt 120 On eve of the crisis: UK also had one of the highest levels of debt in OECD (although lower than most other G7). 80 40 1948-49 1951-52 1954-55 1957-58 1960-61 1963-64 1966-67 1969-70 1972-73 1975-76 1978-79 1981-82 1984-85 1987-88 1990-91 1993-94 1996-97 1999-00 2002-03 2005-06 2008-09 2011-12 0 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank. Effect of the crisis on UK’s public finances • Level of trend GDP now forecast to be permanently lower than previously expected © Institute for Fiscal Studies A large hit to future potential output? Level of real GDP (index, GDP in 2007–08 = 100) 135 130 125 120 115 110 Trend (March 2008) Actual (March 2008) Actual (March 2014) Trend (March 2014) Real GDP back to precrisis level in 2014, but GDP per capita not set to bounce back until 2016 105 100 95 90 © Institute for Fiscal Studies Source: Authors’ calculations based on HM Treasury and Office for Budget Responsibility forecasts and Office for National Statistics data on outturns. Trend GDP: 17% lower Effect of the crisis on UK’s public finances • Level of trend GDP now forecast to be permanently lower than previously expected • Tax revenues fell as GDP fell in cash/real terms – Small fall as % of GDP due to fiscal drag and compositional changes – Falls in housing prices and transactions reduced stamp duty revenues – Falls in stock prices reduced capital taxes and associated with lower bonus payments – Financial sector contraction reduced corporation tax revenues © Institute for Fiscal Studies Spending and revenues, without action Public sector receipts and total managed expenditure, 1997 to 2018 55 Revenues - without action % of GDP 50 45 40 35 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 30 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. Effect of the crisis on UK’s public finances • Level of trend GDP now forecast to be permanently lower than previously expected • Tax revenues fell as GDP fell in cash/real terms – Small fall as % of GDP due to fiscal drag and compositional changes – Falls in housing prices and transactions reduced stamp duty revenues – Falls in stock prices reduced capital taxes and associated with lower bonus payments – Financial sector contraction reduced corporation tax revenues • Around half of spending set in advance in cash terms and upward pressure on cyclical spending during recession – Spending increased a lot as % of GDP – Policy default would have been for spending to remain high as % GDP without “action” © Institute for Fiscal Studies Spending and revenues, without action Public sector receipts and total managed expenditure, 1997 to 2018 55 Revenues - without action Spending - without action % of GDP 50 45 Borrowing would have been over 10% of GDP 40 35 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 30 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. Borrowing Alternative measures of borrowing, 1948 to 2018 14 Public sector net borrowing Borrowing (without policy action) Borrowing (% GDP) 12 10 Structural borrowing would have been 9.0% of GDP higher than anticipated precrisis 8 6 4 2 0 -2 -4 1948-49 1951-52 1954-55 1957-58 1960-61 1963-64 1966-67 1969-70 1972-73 1975-76 1978-79 1981-82 1984-85 1987-88 1990-91 1993-94 1996-97 1999-00 2002-03 2005-06 2008-09 2011-12 2014-15 2017-18 -6 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. Effect of the crisis on UK’s public finances • Level of trend GDP now forecast to be permanently lower than previously expected • Tax revenues fell as GDP fell in cash/real terms – Small fall as % of GDP due to fiscal drag and compositional changes • Around half of spending set in advance in cash terms and upward pressure on cyclical spending during recession – Spending increased a lot as % of GDP – Policy default would have been for spending to remain high as % GDP without “action” • Without “action” – Borrowing would have remained above 10% of GDP – Unsustainable fiscal position © Institute for Fiscal Studies Fiscal policy response to the crisis (1) • Two new fiscal targets adopted in May 2010 – Fiscal mandate: cyclically-adjusted current budget must be forecast to be in balance or surplus at the end of the rolling five-year forecast horizon • Currently being met – Supplementary target: debt must be falling as a share of GDP between 2014–15 and 2015–16 • Currently on course to be missed © Institute for Fiscal Studies Fiscal policy response to the crisis (2) • The “no policy change” baseline – Tax and benefit rates/thresholds uprated as set out in legislation (mainly RPI-indexed) • Include pre-announced policy changes – Public service spending growth • Pre-announced cash plans up to 2010–11 • 1.8% a year real terms growth from 2011–12 to 2014–15 (pencilled into March 2008 Budget) • Grows in line with GDP thereafter • Short-term fiscal stimulus followed by fiscal tightening – Stimulus in 2008–09 and 2009–10 – Stimulus reversed and tax rises/spending cuts started in April 2010 – Tax rises largely complete by April 2014 – Spending cuts to continue to March 2019 © Institute for Fiscal Studies 12% from tax rises 8% from investment spending cuts 14% from welfare spending cuts 52% from other current spending Cure Composition of the policy response % of GDP 12 Other current spend 10 Debt interest 8 Benefit spend 6 March 2014: 49% done Investment spend Tax 4 2 0 -2 © Institute for Fiscal Studies Source: Authors’ calculations based on HM Treasury and Office for Budget Responsibility figures. Reduce borrowing by 10.3% of GDP Spending and revenues, with action Now aiming for tighter fiscal position than planned pre-crisis Public sector receipts and total managed expenditure, 1997 to 2018 55 % of GDP 50 Revenues - without action Spending - without action Revenues - with action Spending - with action 45 40 35 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 30 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank and authors’ calculations. Debt forecast to peak in 2015–16 Alternative measures of debt, 1948 to 2018 250 Debt (% GDP) 200 Public sector net debt Treaty debt ratio National debt 150 100 50 1948-49 1951-52 1954-55 1957-58 1960-61 1963-64 1966-67 1969-70 1972-73 1975-76 1978-79 1981-82 1984-85 1987-88 1990-91 1993-94 1996-97 1999-00 2002-03 2005-06 2008-09 2011-12 2014-15 2017-18 0 © Institute for Fiscal Studies Source: Office for Budget Responsibility’s public finances databank. Aside: Fiscal institutions after the crisis • Office for Budget Responsibility (OBR) created in May 2010 – Independent fiscal council: accountable to Parliament, not the government – Produces fiscal and economic forecasts based on announced policy – Tasked with assessing compliance with the new fiscal targets • OBR has significantly increased the transparency and credibility of official fiscal and economic forecasts © Institute for Fiscal Studies Aside: Monetary policy response • Bank of England significantly loosened monetary policy – Interest rates: cut from 5.75% in July 2007 to 0.5% by March 2009 – Central bank asset purchases started in March 2009 at £75bn, rising to £375bn by July 2012 • Sterling devalued significantly – By 25% against trade-weighted basket of currencies © Institute for Fiscal Studies Specific measures • Deep cuts to spending on some areas of public services © Institute for Fiscal Studies Planned cuts to public spending Between 2010–11 and 2018–19 and after economy-wide inflation • Total spending cuts of 4.4% • But – debt interest spending rising – social security spending, particularly on pensioners, rising – other non-departmental spending such as on PAYG spending public service pensions and UK contribution to the EU budget rising • Departmental spending on public services cut by 19.9% © Institute for Fiscal Studies Whitehall departments: ‘winners’ Departmental budget in 2015–16 compared to 2010–11, after economy-wide inflation International Development 35.3 Energy and Climate Change 9.4 NHS (Health) 4.3 Transport -7.4 Education -8.2 Defence -10.4 Total DEL -11.0 -80 © Institute for Fiscal Studies -60 -40 -20 0 20 40 Real budget increase 2011–12 to 2015–16 Note: Figures show cumulative change in total DEL after economy-wide inflation. Adjusted for consistency, including for business rate retention policy, movement of cost of operations into the special reserve, financial transactions associated with ‘Right to Buy’ policy, and the Green Investment Bank. Whitehall departments: ‘losers’ Departmental budget in 2015–16 compared to 2010–11, after economy-wide inflation Total DEL Business, Innovation and Skills -11.0 -19.1 CLG Local Government -28.4 Home Office -28.8 Environment, Food and Rural Affairs -30.3 Culture, Media and Sport -33.4 Justice -35.3 Work and Pensions -35.4 CLG Communities -59.5 -80 -60 -40 -20 0 20 Real budget increase 2011–12 to 2015–16 © Institute for Fiscal Studies Note: Figures show cumulative change in total DEL after economy-wide inflation. Adjusted for consistency, including for business rate retention policy, movement of cost of operations into the special reserve, financial transactions associated with ‘Right to Buy’ policy, and the Green Investment Bank. 40 Specific measures • Deep cuts to spending on some areas of public services • Very large tax increases partially offset by some large tax cuts © Institute for Fiscal Studies Decomposing the net tax increases Measures since Budget 2008 estimated to be a £24 billion net takeaway 25 20 15 £ billion, 2014–15 10 5 0 -5 -10 -15 -20 -25 VAT NICs North sea taxes Tax instrument Source: © Institute for Fiscal Studies Authors’ calculations using data from the Office for Budget Responsibility. Estimates for impact in 2018–19 expressed in 2014–15 terms by deflating by nominal GDP growth. Decomposing the net tax increases Measures since Budget 2008 estimated to be a £24 billion net takeaway 25 20 15 £ billion, 2014–15 10 5 0 -5 -10 -15 -20 -25 VAT NICs North sea Income taxes tax Tax instrument Source: © Institute for Fiscal Studies Authors’ calculations using data from the Office for Budget Responsibility. Estimates for impact in 2018–19 expressed in 2014–15 terms by deflating by nominal GDP growth. Decomposing the net tax increases Measures since Budget 2008 estimated to be a £24 billion net takeaway 25 20 15 £ billion, 2014–15 10 5 0 -5 -10 -15 -20 -25 VAT NICs North sea Income taxes tax Fuel duties On-shore CT Other taxes Tax instrument Source: © Institute for Fiscal Studies Authors’ calculations using data from the Office for Budget Responsibility. Estimates for impact in 2018–19 expressed in 2014–15 terms by deflating by nominal GDP growth. Decomposing the net tax increases Measures since Budget 2008 estimated to be a £24 billion net takeaway arising from an £74 billion takeaway and a £50 billion giveaway from 483 measures 25 Takeaway Giveaway Net 20 15 £ billion, 2014–15 10 5 0 -5 -10 -15 -20 -25 VAT NICs North sea Income taxes tax Fuel duties On-shore CT Other taxes Tax instrument Source: © Institute for Fiscal Studies Authors’ calculations using data from the Office for Budget Responsibility. Estimates for impact in 2018–19 expressed in 2014–15 terms by deflating by nominal GDP growth. Changes to the tax system (1/2) • Increasing rates of NICs and main rate of VAT not particularly bad ways to raise large sums of money – VAT increase is in part a windfall tax on those with savings so, if thought to be one-off, could be efficient – but UK VAT base very narrow: increase in main rate increases distortions for both producers and consumers – both weaken work incentives • Cutting main rate of corporation tax and getting rid of low profit rate is a good way to cut taxes • Recent reforms have made direct personal tax schedule less coherent – many have been taken out of income tax at considerable cost, but over one million low earners who don’t pay income tax still pay National Insurance – system of pensions tax relief for those on high incomes has been made less efficient, more complicated and, arguably, unfair © Institute for Fiscal Studies Marginal income tax and employee NICs rate Personal tax schedule 70% 60% Transferable allowance withdrawn “Personal allowance tapered away” 50% 40% 30% 20% 10% 2009–10 2015–16 Income tax allowance and NI threshold moving further apart 0% 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Gross annual income (£1,000s) © Institute for Fiscal Studies Note: 2015–16 system assumes married to a non-income tax paying partner. Changes to the tax system (2/2) • Housing taxation shifted from council tax towards stamp duty – bad as stamp duty strong contender for the UK’s worst tax – lack of reform of council tax: still regressive with respect to property values and, in England, based on 1991 values • Rates of fuel duties have been cut substantially without a long-run strategy – currently 58p/l and falling in real terms – motor vehicles becoming more efficient and congestion worsening – every year the Chancellor cancels the next planned increase © Institute for Fiscal Studies Changing composition of revenues 100% Share of total net taxes and NICs 90% 23.3 27.7 25.2 23.9 2.7 11.0 3.3 9.8 4.5 7.9 5.0 5.7 15.6 17.4 15.6 16.8 17.3 16.8 19.5 19.0 25.7 28.0 28.6 29.6 80% 70% 60% 50% 40% 30% 20% 10% 0% Highest proportion of Other net taxes and NICs from capital Capital taxes taxes since at least 1978 Onshore corporation tax VAT Increasingly coming from a relatively small number of NICs high income people Top 1% Income tax contributed: 11% in 1979 21.3% in 1999-2000 27.5% in 2011-12 1989–90 1999–00 2007–08 2018–19 © Institute for Fiscal Studies and sources: 2.5 of (2014). The IFS Source:Notes Crawford, Keynes see andTable Emmerson Green Budget: February 2014. Specific measures • Deep cuts to spending on some areas of public services • Very large tax increases partially offset by some large tax cuts • Large cuts to benefit spending focussed on working age individuals © Institute for Fiscal Studies Changes to the benefit system • Pensioner benefits largely protected from cuts – most working age benefits now indexed less generously, pensioner benefits indexed more generously – cuts to housing benefit and disability benefits don’t apply to pensioners (and they are little affected by cuts to child-related benefits) • Cuts to health-related benefits involve attempt to restrict benefits to least healthy through more stringent, more frequent, testing • Increase in the earliest age at which individuals can receive a state pension has been brought forward and further increases mooted – coherent response to the public finance challenge of rising longevity • Work incentives, on average, strengthened by benefit reforms © Institute for Fiscal Studies Cure: all in this together? Impact of tax and benefit reforms implemented January 2010 - April 2015 inclusive, no Universal Credit, by income decile 0% Change in net income -1% -2% -3% -2.9% -4% -5% -6% -7% -8% -9% Poorest 2 3 4 5 6 Income decile group 7 8 Note: Assumes full take-up of means-tested benefits and tax credits. Source: Phillips (2014). © Institute for Fiscal Studies 9 Richest All Cure: all in this together? Impact of tax and benefit reforms implemented January 2010 - April 2015 inclusive, no Universal Credit, by family type 4% Change in net income 2% 0% -2% -4% -6% -8% Working-age without children -10% Pensioner households -12% Households with children -14% Poorest 2 3 4 5 6 7 Income decile group 8 Note: Assumes full take-up of means-tested benefits and tax credits. Source: Phillips (2014). © Institute for Fiscal Studies 9 Richest All Cure: all in this together? Impact of tax and benefit reforms implemented January 2010 - April 2015 inclusive, no Universal Credit, by family type 4% Change in net income 2% 0% -2% -4% -6% -8% Working-age without children -10% Pensioner households -12% Households with children -14% Poorest 2 3 4 5 6 7 Income decile group 8 Note: Assumes full take-up of means-tested benefits and tax credits. Source: Phillips (2014). © Institute for Fiscal Studies 9 Richest All Effect on work incentives • Fall in real earnings between 2010 and 2015 would have led to a significant, though not enormous, weakening of work incentives • Tax & benefit reforms have an ambiguous impact on work incentives – strengthened by: cuts to (some) out-of-work benefits and, for most earners, increases in tax allowances – weakened by: increases in tax rates, increases to (some) out-of-work benefits and, for higher earners, cut to tax thresholds – complicated effects of cuts to in-work support • Adam and Browne (2013) find that, on average, the reforms strengthen incentives to be in work and more than offset effects of falling real earnings – strengthened less for those with children than those without • Benefit cuts primarily responsible for that average strengthening – but not dramatic given scale of cuts, partly because of nature of tax credit reforms © Institute for Fiscal Studies Conclusions • Pre-crisis – planned to reduce structural borrowing from 2.7% to 1.2% of GDP • Trend GDP now expected to be substantially lower • Without action, structural borrowing would have risen by 9% of GDP • 9-year fiscal consolidation plan – 10.3% of GDP – 88% from spending cuts • Deep cuts to spending on some areas of public services – share of budget going on NHS continues to grow • Very large tax increases partially offset by some large tax cuts – some changes have improved operation of tax system, but many have worsened it • Large cuts to benefit spending focussed on working age individuals – on average strengthen work incentives © Institute for Fiscal Studies UK public finances and the financial crisis Carl Emmerson and Gemma Tetlow Presentation given at workshop on “European public finances through the financial crisis”, ZEW Centre for European Economic Research, Mannheim, Germany, 11 June 2014. © Institute for Fiscal Studies