## 1. Inequality and Poverty Across Generation ## Source details **Canonical URL:** [1. Inequality and Poverty Across Generation](https://www.imf.org/-/media/files/publications/cr/2017/cr17320.pdf) ## Other formats - [Markdown version](/-/media/files/publications/cr/2017/cr17320.pdf.md) - [Structured JSON version](/-/media/files/publications/cr/2017/cr17320.pdf.json) --- ### A. Introduction and Demographic Context - Structure of pension system: - Earnings-based, mandatory pay-as-you-go contributory scheme and a means-tested, state-financed non-earnings related basic scheme. - State transfers top up low earnings-related pensions to the minimum pension amount. - Life expectancy and fertility: - A 65-year old Spaniard in 2017 can expect to live, on average, to age 86. - Children of current cohorts can expect to live to about 90 once they reach 65. - Fertility was 1.3 children per woman of childbearing age in 2016. - Population and labor-force projections: - INE projects that the population will contract by 12 percent over the next 50 years. - Projected cumulative net immigration of 3.1 million individuals. - The number of potential workers supporting those over 65 will fall from 4 to 1 to less than 2 to 1. - Political and social considerations: - 2011 and 2013 pension reforms (Toledo Pact framework) aim to safeguard financial viability but imply a significant reduction in pension purchasing power. - Key choices for further reform: acceptable retirement income, the public share of retirement income, and affordability alongside other fiscal priorities. ### B. Pension Finances: Recent Developments and Current Balances - Social security budget and contributory pensions: - After years of surpluses, the social security budget began running deficits in 2009. - Contributory pension spending covers about 85–90 percent of social security budget non-financial outlays. - Authorities increased the ceiling on the maximum contribution base by 5 percent in both 2012 and 2013. - Contributions have remained relatively in line with a long-run level of about 9.8 percent of GDP per annum. - Recent deficit and outlook: - Overall pension expenditure, including civil servant pensions, rose to about 12 percent of GDP in 2016. - Authorities foresee the social security deficit narrowing from 1.7 percent of GDP in 2016 to 0.5 percent of GDP in 2020, with GDP projected to grow faster than nominal pension expenditures (3.1 percent annually). - Based on 2017Q1 results, AIReF projects an 84 percent chance that Spain will exceed its social security deficit target of 1.4 percent of GDP in 2017. - Contributors and labor-market effects: - Average number of contributors in 2016 was 17.6 million, up from a local annual minimum average in 2012 of 16.3 million, but below the 2008 peak of 19.1 million. - Post-crisis dynamics included unemployed older workers opting for retirement and a rush to retire before reforms took effect, raising pension expenditures. - Social Security Reserve Fund (SSRF) developments: - SSRF balance peaked at 6.2 percent of GDP in 2011. - Withdrawals beginning in 2009 amounted to 6 percent of GDP. - SSRF balance fell to 1.4 percent of GDP in 2016. - Government planned to borrow EUR 10.2 billion (1 percent of GDP) to cover part of the 2017 deficit. - Financing options highlighted: - Two general approaches to cyclical deficit financing: (i) annual central government coverage as needed; (ii) upfront transfers to build a buffer to avoid depletion. - Emphasis on transparency to distinguish liquidity from sustainability issues. ### C. Ongoing Pension Reforms and Automatic Mechanisms - Parametric reforms (projected fully implemented by 2027): - Statutory retirement age: gradual increase from 65 in 2012 to 67 in 2027. - Pensionable earnings calculation period: increase from last 15 years to 25 years (by 2022). - Partial early retirement: access tightened. - Contribution years for full pension: increase from 35 years in 2012 to 37 by 2027. - 2015 Aging Report projects these reforms will extend the effective retirement age by 3 years to 66.4 years by 2060. - Stability Programme Update 2017–20 states the effective retirement age increased from 63.9 years in 2012 to 64.1 in 2017. - Two automatic mechanisms introduced: - Index for Pension Revaluation (IPR): - Implemented in 2014 to achieve financial balance over the medium term. - Calculated based on past and projected social security income, the number of pensions, and the average pension absent revaluation. - Legally restricted to be within a range of 0.25 percent and the year-on-year percentage change in annual consumer price inflation plus 0.5 percent. - Positive floor guarantees nominal pensions will always increase. - Sustainability Factor (SF): - Introduced in 2019 to link initial pension values to increases in life expectancy. - For 2019–23, SF is adjusted year over year proportional to the ratio of projected life expectancy at 67 in 2012 to life expectancy at 67 in 2017. - SF is a discount factor applied only to the initial value of a retiree’s pension and will be updated every five years using INE life-expectancy projections. - Recent IPR and real pension value effects: - From 2014–16, the annual IPR was set at 0.25 percent. - The calculated indexation amount for 2017 was negative 3 percent under the Stability Programme 2017–2020 macro scenario. - 2017 indexing applied the IPR minimum of 0.25 percent while expected average inflation is around 2 percent, making 2017 the first year since the IPR’s introduction in which the real value of pensions is likely to fall. - Multi-year projections indicate an annual IPR of 0.25 percent will be required at least until the 2050s given current and projected contributions and pension expenditures. - Example impacts on purchasing power: - An individual retiring in 2017 can expect the real value of her pension to decrease by 30 percent over her retirement years under IPR projections and an assumed average annual inflation rate of 2 percent. - A 47–year old today retiring in 2037 at age 67 would see a similar reduction in purchasing power plus an initial pension reduction of about 10 percent due to the SF — a total reduction of 40 percent relative to what the pension would have been without the automatic mechanisms. - Bank-of-Spain estimate: IPR and SF will reduce the public pension deficit by about 0.1 percent of GDP per year, assuming no change in macroeconomic assumptions and no increase in revenue. - Transparency and social acceptance: - Long-term reduction in pension purchasing power may create social and political pressures to deviate from IPR or halt other reforms. - Financial viability is a pre-condition for credible discussions on adequacy and fairness; ad hoc increases to protect purchasing power would undermine reform credibility and shift burdens to younger generations. ### D. Key Findings and Policy Implications - Fiscal and demographic pressures: - Demographic trends (population contraction, low fertility, aging) will exert sustained pressure on contributory pension finances. - SSRF substantially depleted from its 2011 peak, increasing near-term financing needs. - Reform effectiveness and distributional consequences: - 2011 and 2013 parametric reforms and the 2014 IPR plus the 2019 SF are designed to preserve long-run financial viability mainly by incentivizing later retirement and reducing replacement rates. - These measures preserve relative generosity compared to EU averages but imply significant reductions in purchasing power for future retirees (examples: 30 percent over retirement; up to 40 percent when SF applied). - Policy recommendations and considerations: - Maintain the financial viability of the contributory pension system as a prerequisite for any discussion on adequacy. - Ensure transparency about objectives and ramifications of further reforms and clarify intergenerational burden sharing. - Avoid ad hoc measures that would protect current pension purchasing power at the expense of younger generations and reform credibility. - Consider communication and politically feasible adjustments within parameters agreed under the Toledo Pact to facilitate a socially acceptable transition. --- ### 10. Spain’s demographics imply that either contribution revenues must increase or pension benefits must decrease to ensure financial viability of the pension system ### Demographic outlook and accounting identities - Projected changes: - Number of pensions projected to increase by almost 50 percent to 15 million from 2017–50. - Life expectancy at age 65 is projected to increase by about two years. - Eurostat, U.N., and INE project a higher dependency ratio by 2050. - Key identities: - PR/GDP = PE/GDP (pension revenue to GDP equals pension expenditure to GDP). - Contributory pension revenue identity: pension revenue to GDP equals the product of the effective contribution rate and the ratio of labor compensation to GDP plus other revenue items such as state transfers. - Contributory pension expenditure identity: pension expenditure to GDP equals the product of the dependency ratio, the eligibility ratio, the benefit ratio, the inverse of the employment rate, and labor’s share in GDP. - Baseline parameters: - Baseline values: actual values for 2016 and U.N. demographic projections for 2050. - Labor’s share in GDP held constant at 2016 level of 47.2 percent. - Eligibility ratio held constant at 2016 value. - The contributory pension system ran a deficit of about 1.6 percent of GDP in 2016. ### Projected fiscal impact and required tradeoffs - Dependency ratio increase from 29 to 70 would more than double contributory pension spending to 22½ percent of GDP by 2050. - To maintain the current benefit ratio by 2050 would require the effective contribution rate to increase 20 percentage points. - To keep the effective contribution rate in 2050 close to its 2016 level would require a 25-percentage point cut in the benefit ratio. - Plausible tradeoff scenario: - Accept an increase in contributory pension expenditure relative to GDP of 2¼ percentage points in exchange for a decline of 13 percentage points in the benefit ratio to about 35 percent. - This decline would allow retirees who saved resources sufficient to generate 5 percent of the average wage every year in retirement to raise retirement income to 40 percent of average wages (ILO-recommended replacement level). - Assumptions underpinning scenario: - Net migration of 5.5 million individuals by 2050 (compared to INE projections of 2.5 million), with 90 percent of “net” migrants of working age. - Dependency ratio lowered to 63.4 in 2050 under this migration assumption. - Employment rate rises an additional 12 percentage points above baseline to about 80 percent. - Increase in the effective contribution rate to allow more generous indexation of pensions (possible via raising the ceiling on the maximum contribution base). ### Policy options and parametric reforms - Reform package principle: gradual parametric changes plus complementary measures to reduce required parametric tightening; greater generosity in one parameter requires tightening others to maintain viability. - Transparency: provide workers with necessary information about future public pension benefits to enable optimal saving and labor decisions. - Equity considerations: - Intergenerational equity: adjustment burden falls more on future pensioners, current retirees were relatively shielded post-crisis. - Intra-generational equity: a proportionate decline burdens lower income earners more. - Supplementary parametric reforms: - Index statutory retirement age to changes in life expectancy at retirement: - Spain’s 2011 reform raises statutory retirement age to 67 by 2027; leaving it at 67 would lead to higher pension costs as life expectancy increases after 2030. - Bank of Spain estimates an effective retirement age of 70 would reduce pension spending by 2–4 percent of GDP. - Adjust pension accrual rates to encourage longer work lives: - 2011 reform extends required contribution period for a full pension from 35 to 37 years; accruals remain front-loaded (50 percent entitlement in first fifteen years and another 34 percent over next fifteen years). - Alternative: lower initial accrual from 50 to 40 percent and backload annual accrual rates. - Restrict access to early retirement: - About 20 percent of pensions go to individuals less than 65 years old; most are disability pensions. - Review and potentially tighten disability pension eligibility; drop rule allowing early retirement with a full pension. - Extend pensionable earnings reference period toward a contributor’s full career: - 2011 reform extends reference period from 15 to 25 years by 2022. - Review reasons for multiple pensions: - As of 2017, 850,000 pensioners (about 10 percent of all pensioners) received at least two public pensions. ### Revenue-side measures without raising contribution rates - Caution: raising statutory social security contribution rates could negatively affect growth and hiring; Spain’s statutory social security contribution rates are relatively high at 28.3 percent per annum. - Scope to increase social security revenue by raising the minimum and maximum earnings subject to contributions, generating net additional revenue provided the maximum and minimum pensions rise less, respectively. - Bank of Spain estimate: removing the ceiling on wages subject to contributions would increase revenues by about EUR 10 billion, equivalent to a 10 percent increase over current contributions. - Historical precedent: to offset contributor fall in 2012–2013, Spain increased the contribution ceiling by 5 percent while raising the maximum pension amount by 1 percent each year. - Self-employed contributions: - Current minimum contribution level yields a pension value less than the minimum pension, creating an incentive to declare only the minimum. - Recommendation: align the minimum contribution amount for self-employed workers with the amount needed to receive the minimum pension (without the state complement). --- ### 21. General taxation could also provide a source of additional revenue to cover pensions ### General taxation as a revenue option - Spain implicitly decided to use future general tax revenue to fund part of contributory pension expenditure in 2017 by borrowing to fund part of the 2017 social security budget deficit. - Other pension-related expenses already covered from general taxes, including the complement on minimum pensions and non-contributory pension costs (about 0.8 percent of GDP in 2016). - Peer example: Germany has earmarked one point of VAT for the public pension system. - Spain’s VAT policy gap is large, suggesting room to raise additional VAT revenue; earmarking an increase in indirect taxes for contributory pension expenses would: - Spread the burden so that everyone, including pensioners, helps pay. - Have mild redistribution consequences. - Make the additional VAT revenue unavailable for other priority expenditures. - IPR calculation methodology allows inclusion of other revenue sources in annual indexation. ### Transparency and communication - Transparency about demographic pressures and tradeoffs is essential. - Practical measures: - More frequent communication with workers about individual pensions (e.g., annual investment reports documenting past contribution history and projected old-age pension benefit). - Step up promotion of the website "mi pension" for individuals to calculate future pension benefits. - Broader public discourse to explain trade-offs: financial viability vs. pension generosity; intra- and inter-generational equity. - Authorities introduced more transparency by publishing the annual calculation of IPR and underlying components in 2017. ### Productivity, demographics, and pension outcomes - Boosting productivity raises pension values for a given benefit level if productivity gains are reflected in average wages. - Historical and projected figures: - Since 2000, annual productivity growth in Spain has averaged 0.9 percent. - A study finds that a five-percentage point increase in the share of workers aged 55–64 is associated with a decrease in TFP of between two and four percent. - Projected 10-percentage point increase in the share of workers aged 55–64 from 2016 to 2035 implies a decrease in TFP of 4 to 8 percent by 2035. - Projected fall in the working-age population alone would reduce GDP by about 0.2 percent every year. - Implication: Structural reforms to boost productivity are a priority. ### Parametric reforms and NDC systems - Shifting to a notional defined contribution (NDC) system could speed up behavioral changes; peer examples include Finland, Italy, and Sweden. - NDC characteristics: - Contributions recorded into individual “savings account”; balance grows at a notional interest rate; at retirement, account balance divided by life expectancy to derive annual pension. - Fiscal and behavioral effects: - NDC provides automatic fiscal balancing as life expectancy grows. - NDC could incentivize individuals to work longer. - Risks and distributional consequences: - Many workers could end up with minimum pension if they do not contribute for long or have low wages. - NDC generally has less redistributive element compared to defined-benefit systems. ### Private saving, second-pillar options, and other instruments - Private pension fund assets amount to less than 10 percent of GDP; OECD average is 37 percent of GDP. - Policy options to stimulate private savings: - Reconsider tax incentives given limited additional savings generated. - Automatically enroll workers in a government-sponsored private second pillar with opt-out ability; ensure portability. - Promote programs that automatically increase employee savings out of future raises. - Consider introducing and regulating reverse mortgages to unlock housing wealth. - Note: Spain’s existing incentives appear to have generated little additional savings. ### Migration and labor force considerations - Net migration that increases the number of workers raises contributors and allows more gradual parametric changes. - INE projects emigration of 14.1 million over the next 35 years; policies to slow emigration could complement immigration policies. ### Conclusions and recommended policy package - Demographic trends will continue to exert relentless pressure on contributory pension finances; reforms adopted in 2011 and 2013, if fully implemented, will ensure financial viability. - Primary burden of adjustment falls on reducing the real and relative value of pension payments, raising social and political sustainability concerns. - Key decisions required via public discourse: acceptable level of retirement income and share provided by the contributory pension system. - Suggested package elements: - Automatically link retirement age to changes in life expectancy. - Adjust accrual rates and the calculation of pensionable earnings. - Raise the ceiling on maximum earnings subject to contributions and raise the minimum contribution amount paid by the self-employed. - Do not shift survivor and orphan pensions to the central government budget as a means to improve system financial viability. - Avoid ad hoc increases in pensions to protect purchasing power. - Complement parametric reforms with policies that encourage greater private savings and increases in labor productivity. - Consider measures to boost net migration. - Overarching guidance: - Design reforms to maximize benefits and minimize costs, introduced gradually to allow behavioral adjustments, and preserve consensus building (e.g., Toledo Pact) to avoid reform reversals. - Emphasize the contributory pension system was never intended to be a retiree’s sole source of income; Spaniards will need to work longer and save more. *International Monetary Fund, "The Spanish Pension System—Challenges Beyond Financial Sustainability." (cr17320)* ### 1. Inequality and Poverty Across Generation _____________________________________________ 21 ### 1. Inequality and Poverty Across Generation ### A. Introduction and Demographic Context - Spain’s public pension system comprises an earnings-based, mandatory pay-as-you-go contributory scheme and a means-tested, state-financed non-earnings related basic scheme. State transfers top up low earnings-related pensions to the minimum pension amount. - Life expectancy and fertility: - A 65-year old Spaniard in 2017 can expect to live, on average, to age 86. - Children of current cohorts can expect to live to about 90 once they reach 65. - Fertility was 1.3 children per woman of childbearing age in 2016. - Population and labor-force projections: - Spain’s statistical agency (INE) projects that the population will contract by 12 percent over the next 50 years. - Projected cumulative net immigration of 3.1 million individuals. - The number of potential workers supporting those over 65 will fall from 4 to 1 to less than 2 to 1. - Political and social considerations: - The 2011 and 2013 pension reforms (within the Toledo Pact framework) aim to safeguard financial viability but imply a significant reduction in pension purchasing power, generating public debate and calls for reconsideration. - Key choices for further reform center on: acceptable retirement income, the public share of retirement income, and affordability alongside other fiscal priorities. ### B. Pension Finances: Recent Developments and Current Balances - Social security budget and contributory pensions: - After years of surpluses, the social security budget began running deficits in 2009. - Contributory pension spending covers about 85–90 percent of social security budget non-financial outlays. - Authorities increased the ceiling on the maximum contribution base by 5 percent in both 2012 and 2013. - Contributions have remained relatively in line with a long-run level of about 9.8 percent of GDP per annum. - Recent deficit and outlook: - Overall pension expenditure, including civil servant pensions, rose to about 12 percent of GDP in 2016. - Authorities foresee the social security deficit narrowing from 1.7 percent of GDP in 2016 to 0.5 percent of GDP in 2020, with GDP projected to grow faster than nominal pension expenditures (3.1 percent annually). - Based on 2017Q1 results, AIReF projects an 84 percent chance that Spain will exceed its social security deficit target of 1.4 percent of GDP in 2017. - Contributors and labor-market effects: - Average number of contributors in 2016 was 17.6 million, up from a local annual minimum average in 2012 of 16.3 million, but below the 2008 peak of 19.1 million. - Post-crisis dynamics included unemployed older workers opting for retirement and a rush to retire before reforms took effect, raising pension expenditures. - Social Security Reserve Fund (SSRF) developments: - SSRF balance peaked at 6.2 percent of GDP in 2011. - Withdrawals beginning in 2009 amounted to 6 percent of GDP. - SSRF balance fell to 1.4 percent of GDP in 2016. - The 2017 approach includes the government planning to borrow EUR 10.2 billion (1 percent of GDP) to cover part of the deficit. - Financing options highlighted: - Two general approaches to cyclical deficit financing: (i) annual central government coverage as needed; (ii) upfront transfers to build a buffer to avoid depletion. Transparency is emphasized to distinguish liquidity from sustainability issues. ### C. Ongoing Pension Reforms and Automatic Mechanisms - Parametric reforms (projected fully implemented by 2027): - Statutory retirement age: gradual increase from 65 in 2012 to 67 in 2027. - Pensionable earnings calculation period: increase from last 15 years to 25 years (by 2022). - Partial early retirement: access tightened (details by career length). - Contribution years for full pension: increase from 35 years in 2012 to 37 by 2027. - The 2015 Aging Report projects these reforms will extend the effective retirement age by 3 years to 66.4 years by 2060. - Stability Programme Update 2017–20 states the effective retirement age increased from 63.9 years in 2012 to 64.1 in 2017. - Two automatic mechanisms introduced in 2013: - Index for Pension Revaluation (IPR): - Implemented in 2014. - Designed to achieve financial balance of the pension system over the medium term. - Calculated based on past and projected social security income, the number of pensions, and the average pension absent revaluation. - Legally restricted to be within a range of 0.25 percent and the year-on-year percentage change in annual consumer price inflation plus 0.5 percent. - The positive floor guarantees nominal pensions will always increase. - Sustainability Factor (SF): - Introduced in 2019 to link initial pension values to increases in life expectancy. - For 2019–23, SF is adjusted year over year proportional to the ratio of projected life expectancy at 67 in 2012 to life expectancy at 67 in 2017. - SF is a discount factor applied only to the initial value of a retiree’s pension, reducing the replacement rate; it will be updated every five years using INE life-expectancy projections. - Recent IPR and real pension value effects: - From 2014–16, the annual IPR was set at 0.25 percent. - The calculated indexation amount for 2017 by the Ministry of Employment and Social Security, assuming the Stability Programme 2017–2020 macro scenario consistent with medium-term financial balance, was negative 3 percent. - 2017 indexing applied the IPR minimum of 0.25 percent while expected average inflation is around 2 percent, making 2017 the first year since the IPR’s introduction in which the real value of pensions is likely to fall. - Multi-year projections indicate an annual IPR of 0.25 percent will be required at least until the 2050s given current and projected contributions and pension expenditures. - Example impact on purchasing power: - An individual retiring in 2017 can expect the real value of her pension to decrease by 30 percent over her retirement years under IPR projections and an assumed average annual inflation rate of 2 percent. - A 47–year old today retiring in 2037 at age 67 would also see a similar reduction in purchasing power plus an initial pension reduction of about 10 percent due to the SF — a total reduction of 40 percent relative to what the pension would have been without the automatic mechanisms. - Bank-of-Spain estimate: IPR and SF will reduce the public pension deficit by about 0.1 percent of GDP per year, assuming no change in macroeconomic assumptions and no increase in revenue. - Transparency and social acceptance: - The expected long-term reduction in pension purchasing power may create social and political pressures to deviate from IPR or halt other reforms. - Financial viability is a pre-condition for credible discussions on adequacy and fairness; ad hoc increases to protect purchasing power would undermine reform credibility and shift burdens to younger generations. ### D. Key Findings and Policy Implications - Fiscal and demographic pressures: - Demographic trends (population contraction, low fertility, aging) will exert sustained pressure on contributory pension finances. - The SSRF has been substantially depleted from its 2011 peak, increasing near-term financing needs. - Reform effectiveness and distributional consequences: - The 2011 and 2013 parametric reforms and the 2014 IPR plus the 2019 SF are designed to preserve long-run financial viability mainly by incentivizing later retirement and reducing replacement rates. - These measures preserve relative generosity compared to EU averages but imply significant reductions in purchasing power for future retirees (examples: 30 percent over retirement; up to 40 percent when SF applied). - Policy recommendations and considerations (as presented in the source): - Maintain the financial viability of the contributory pension system as a prerequisite for any discussion on adequacy. - Ensure transparency about objectives and ramifications of further reforms and clarify intergenerational burden sharing. - Avoid ad hoc measures that would protect current pension purchasing power at the expense of younger generations and reform credibility. - Consider communication and politically feasible adjustments within parameters agreed under the Toledo Pact to facilitate a socially acceptable transition. *Source: IMF staff paper excerpt — "The Spanish Pension System—Challenges Beyond Financial Sustainability."* ### 10. Spain’s demographics imply that either contribution revenues must increase or ### 10. Spain’s demographics imply that either contribution revenues must increase or pension benefits must decrease to ensure financial viability of the pension system ### Demographic outlook and accounting identities - Number of pensions projected to increase by almost 50 percent to 15 million from 2017–50. - Life expectancy at age 65 is projected to increase by about two years. - Eurostat, U.N., and INE project a higher dependency ratio by 2050. - Key identity used to balance revenues and expenses: PR/GDP = PE/GDP. - Contributory pension revenue identity: pension revenue to GDP equals the product of the effective contribution rate and the ratio of labor compensation to GDP plus other revenue items such as state transfers. (Formula shown in source.) - Contributory pension expenditure identity: pension expenditure to GDP equals the product of the dependency ratio, the eligibility ratio, the benefit ratio, the inverse of the employment rate, and labor’s share in GDP. (Formula shown in source.) ### Baseline assumptions and key parameters used in analysis - Baseline values: actual values for 2016 and U.N. demographic projections for 2050 (red values in source table). - Dependency ratios derived from U.N. population projections. - Employment rate evolution proxied as product of ILO labor market participation projections and IMF staff projections of unemployment, which stabilizes at its natural rate over the long run. - Labor’s share in GDP held constant at 2016 level of 47.2 percent. - Eligibility ratio (number of pensioners relative to individuals 65 and older) held constant at 2016 value. - The contributory pension system ran a deficit of about 1.6 percent of GDP in 2016. ### Projected fiscal impact and required tradeoffs - Dependency ratio increase from 29 to 70 would more than double contributory pension spending to 22½ percent of GDP by 2050. - To maintain the current benefit ratio by 2050 would require the effective contribution rate to increase 20 percentage points (row 2 in source table). - To keep the effective contribution rate in 2050 close to its 2016 level would require a 25-percentage point cut in the benefit ratio. - One plausible tradeoff scenario (row 4 in source table): - Accept an increase in contributory pension expenditure relative to GDP of 2¼ percentage points in exchange for a decline of 13 percentage points in the benefit ratio to about 35 percent. - This decline would allow retirees who saved resources sufficient to generate 5 percent of the average wage every year in retirement to raise retirement income to 40 percent of average wages (ILO-recommended replacement level). - Assumptions underpinning the scenario: - Net migration of 5.5 million individuals by 2050 (compared to INE projections of 2.5 million), with 90 percent of “net” migrants of working age. - The dependency ratio would be lowered to 63.4 in 2050 under this migration assumption. - Employment rate rises an additional 12 percentage points above baseline to about 80 percent. - Increase in the effective contribution rate to allow more generous indexation of pensions (possible via raising the ceiling on the maximum contribution base). ### Policy options and parametric reforms - Reform package principle: gradual parametric changes plus complementary measures to reduce required parametric tightening; there is no free lunch—greater generosity in one parameter requires tightening others to maintain viability. - Transparency: provide workers with necessary information about future public pension benefits to enable optimal saving and labor decisions. - Equity considerations: - Intergenerational equity: adjustment burden falls more on future pensioners, while current retirees were relatively shielded post-crisis. - Intra-generational equity: important to track how benefit ratio changes across income groups; a proportionate decline burdens lower income earners more. - Supplementary parametric reforms to incentivize longer work lives and higher participation (package more effective than single measures): - Index statutory retirement age to changes in life expectancy at retirement: - Spain’s 2011 reform raises statutory retirement age to 67 by 2027; leaving it at 67 would lead to higher pension costs as life expectancy increases after 2030. - Bank of Spain estimates an effective retirement age of 70 would reduce pension spending by 2–4 percent of GDP (Hernandez de Cos et al., 2017). - Note potential intra-generational equity concerns if life expectancy changes vary by socioeconomic group; cap on maximum pension mitigates concern. - Adjust pension accrual rates to encourage longer work lives: - 2011 reform extends required contribution period for a full pension from 35 to 37 years; accruals remain front-loaded (50 percent entitlement in first fifteen years and another 34 percent over next fifteen years per source). - Alternative: lower initial accrual from 50 to 40 percent and backload annual accrual rates. - Restrict access to early retirement: - About 20 percent of pensions go to individuals less than 65 years old; most are disability pensions. - Review and potentially tighten disability pension eligibility; drop rule allowing early retirement with a full pension. - Extend pensionable earnings reference period toward a contributor’s full career: - 2011 reform extends reference period from 15 to 25 years by 2022; Spain remains among few EU countries restricting pensionable period to less than a full career. - Review reasons for multiple pensions: - As of 2017, 850,000 pensioners (about 10 percent of all pensioners) received at least two public pensions. ### Revenue-side measures without raising contribution rates - Caution: raising statutory social security contribution rates on employers or employees could negatively affect growth and hiring; Spain’s statutory social security contribution rates are relatively high at 28.3 percent per annum. - Scope to increase social security revenue by raising the minimum and maximum earnings subject to contributions, generating net additional revenue provided the maximum and minimum pensions rise less, respectively. - Bank of Spain estimate: removing the ceiling on wages subject to contributions would increase revenues by about EUR 10 billion, equivalent to a 10 percent increase over current contributions. - Historical precedent: to offset contributor fall in 2012–2013, Spain increased the contribution ceiling by 5 percent while raising the maximum pension amount by 1 percent each year. - Tradeoff: leaving the cap unchanged preserves intra-generational equity; raising the cap diminishes the linkage between earnings and future pensions for wealthier contributors. - Self-employed contributions: - Current minimum contribution level yields a pension value less than the minimum pension, creating an incentive to declare only the minimum. - Recommendation: align the minimum contribution amount for self-employed workers with the amount needed to receive the minimum pension (without the state complement) to close the loophole. *Source: IMF staff analysis as presented in the chapter titled "10. Spain’s demographics imply that either contribution revenues must increase or pension benefits must decrease to ensure financial viability of the pension system" (cr17320).* ### 21. General taxation could also provide a source of additional revenue to cover pensions ### 21. General taxation could also provide a source of additional revenue to cover pensions ### General taxation as a revenue option - Spain implicitly decided to use future general tax revenue to fund part of contributory pension expenditure in 2017 by borrowing to fund part of the 2017 social security budget deficit. - Other pension-related expenses are already covered from general taxes, including the complement on minimum pensions and non-contributory pension costs (about 0.8 percent of GDP in 2016). - Example from peers: Germany has earmarked one point of VAT for the public pension system (EC, 2016). - Spain’s VAT policy gap is large, suggesting room to raise additional VAT revenue; earmarking an increase in indirect taxes for contributory pension expenses would: - Spread the burden so that everyone, including pensioners, helps pay (unlike increases in social contributions, which only active labor participants pay). - Have mild redistribution consequences. - Make the additional VAT revenue unavailable for other priority expenditures. - Technical note: The IPR calculation methodology allows for the inclusion of other revenue sources in the calculation of the annual indexation. ### Transparency and communication - Transparency about demographic pressures on the contributory pension system and the tradeoffs in safeguarding pension sustainability is essential. - Practical measures: - More frequent communication with workers about individual pensions (e.g., annual investment reports documenting past contribution history and projected old-age pension benefit). - Step up promotion of the website "mi pension" for individuals to calculate future pension benefits. - Broader public discourse to explain trade-offs: financial viability vs. pension generosity; intra- and inter-generational equity. - Transparency supports behavior changes that reduce old-age poverty risk (working longer and saving more) and helps preserve social legitimacy of pension reforms. - Authorities introduced more transparency by publishing the annual calculation of IPR and underlying components in 2017. ### Productivity, demographics, and pension outcomes - Boosting productivity raises pension values for a given benefit level if productivity gains are reflected in average wages; if the maximum pension grows more slowly than productivity, higher productivity increases net social security resources. - Historical and projected productivity and demographic figures: - Since 2000, annual productivity growth in Spain has averaged 0.9 percent. - A study (Aiyar et al, 2016) finds that a five-percentage point increase in the share of workers aged 55–64 is associated with a decrease in total factor productivity (TFP) of somewhere between two and four percent. - For Spain, the projected 10-percentage point increase in the share of workers aged 55–64 from 2016 to 2035 implies a decrease in TFP of 4 to 8 percent by 2035. - The projected fall in the working-age population alone, if it materializes, would reduce GDP by about 0.2 percent every year. - Implication: Structural reforms to boost productivity are a priority. ### Parametric reforms and notional defined contribution (NDC) systems - Shifting to an NDC system could speed up behavioral changes; peer examples include Finland, Italy, and Sweden. - Key characteristics of NDC systems: - Contributions recorded into individual “savings account.” - The savings balance grows at a notional interest rate, usually linked to an economic indicator such as average wage growth. - At retirement, the account balance is divided by life expectancy to derive annual pension. - Pensions in payment are indexed based on some rule. - Fiscal and behavioral effects: - NDC provides automatic fiscal balancing: as life expectancy grows, benefits at a given retirement age tend to fall. - NDC could incentivize individuals to work longer to avoid low pension benefits. - Risks and distributional consequences: - Many workers could end up with minimum pension if they do not contribute for long, earn low returns, or have low wages. - NDC generally has less redistributive element compared to current defined-benefit systems. - In NDC, higher contribution rates automatically translate into higher future benefits, so raising contribution rates to offset permanent shocks does little to restore system balance. ### Private saving, second-pillar options, and other instruments - Spain’s private financial savings and private pension fund assets: - Private pension fund assets amount to less than 10 percent of GDP. - OECD average private pension fund assets: 37 percent of GDP. - Policy options to stimulate private savings: - Reconsider tax incentives given limited additional savings generated and benefits potentially favoring higher-income groups. - Automatically enroll workers in a government-sponsored private second pillar with opt-out ability; contributions could be additional to public pension contributions and administered by the public sector with competing fund managers. Portability of savings is desirable. - Champion a private savings program that automatically increases employee savings out of future raises in wages (Thaler and Benartzi, 2003); such programs have been successful in the United States. - Consider introducing and appropriately regulating reverse mortgages to unlock retirees’ housing wealth as retirement income. - Note: Spain’s existing incentives appear to have generated little additional savings (European Commission, 2008). ### Migration and labor force considerations - Net migration that increases the number of workers raises contributors to the pension system and allows more gradual parametric changes. - Spain experienced high immigration in the early 2000s. - INE projects emigration to be 14.1 million over the next 35 years or so; policies to slow emigration could complement immigration policies, though specific measures are beyond the scope of this paper. ### Conclusions and recommended policy package - Demographic trends will continue to exert relentless pressure on contributory pension finances; reforms adopted in 2011 and 2013, if fully implemented, will ensure financial viability of the contributory pension system. - The primary burden of adjustment falls on reducing the real and relative value of pension payments, which raises social and political sustainability concerns and risks of reform reversal. - Key decisions required via public discourse: acceptable level of retirement income and share provided by the contributory pension system; these determine the replacement rate the system needs to deliver. - Suggested package elements to safeguard financial viability and address concerns: - Automatically link the retirement age to changes in life expectancy. - Adjust accrual rates and the calculation of pensionable earnings. - Raise the ceiling on the maximum earnings subject to contributions and raise the minimum contribution amount paid by the self-employed. - Do not shift survivor and orphan pensions to the central government budget as a means to improve system financial viability. - Avoid ad hoc increases in pensions to protect purchasing power, which risk further ad hoc adjustments and undermine fiscal sustainability. - Complement parametric reforms with policies that encourage greater private savings and increases in labor productivity. - Consider measures to boost net migration. - Overarching guidance: - Pension reforms should be designed to maximize benefits and minimize costs, introduced gradually to allow behavioral adjustments, and preserve Spain’s tradition of consensus building (e.g., via the Toledo Pact Commission) to avoid reform reversals. - Emphasize that the contributory pension system was never intended to be a retiree’s sole source of income; Spaniards will need to work longer and save more. *International Monetary Fund, "Spain — selected chapter: General taxation could also provide a source of additional revenue to cover pensions."* ### 1. Following a deep fall in employment during the crisis, the Spanish labor market has ### 1. Following a deep fall in employment during the crisis, the Spanish labor market has ### Overview of recovery - Between early 2008 and late 2013, Spain lost 3.9 million jobs, equivalent to one out of five. - In early 2014 the labor market turned around and Spanish employment has since increased by 1.4 million jobs (until the first quarter of 2017, based on unadjusted national accounts data). - Overall, at an average rate of 2½ percent per annum, Spanish employment growth has been among the fastest of the recent employment recoveries in euro area countries severely affected by the crisis. - National accounts vs. labor force survey: national accounts show a decline in employment by 3.9 million between early 2008 and 2013:Q4 and an increase by 1.4 million until 2017:Q1; labor force survey shows a decline by 3.7 million between early 2008 and 2014 and an increase by 1.5 million employed until 2017:Q1. ### Unemployment and labor supply - Real GDP has recovered to pre-crisis levels, but total employment remains some 10 percent below its 2008 peak, indicating an upward shift in productivity relative to the pre-crisis period. - Migration increased Spain’s working-age population by 2.3 million people between 2002 and 2008, but migration has been negative since the crisis. - Unemployment stood at 17.3 percent in 2017:Q2. - Youth unemployment is more than double the national average. - Those out of a job for more than a year account for nearly half of all unemployed. - Almost half of the reduction in total employment during the crisis came from the construction sector, which collapsed following the burst of the real estate bubble in 2008. ### Labor market reforms and their effects - Empirical studies find that the 2012 labor market reforms have helped wage moderation and made wages more responsive to overall macroeconomic conditions, though wages remain relatively unresponsive to sectoral developments. - Effects attributed to the 2012 reforms: - Dampened labor-shedding. - Increased hiring. - Made the labor market more resilient to economic shocks. - Some increase in the share of new jobs being permanent, though this effect is limited in size. - Evidence that the duration of unemployment has declined. - The 2012 reform included the Contrato de Emprendedores (full-time permanent contract with an extended trial period up to 12 months) for firms with less than 50 employees, with tax benefits if the employee is kept for at least three years. - Until August 2016 there was also a €500 monthly deduction from social security contributions for new permanent hires. - Recent steps: individual re-integration assistance for long-term unemployed and automatic eligibility of all registered young unemployed for support under the EU’s Youth Guarantee. ### Main features of the 2012 Labor Market Reform (Box 1) - Collective negotiation and firms’ internal flexibility: - Priority to firm-level wage agreements over regional/industry agreements. - Eased/clarified opt-out clauses from collective agreements. - Limited automatic extension of expired collective agreements (ultra-activity) to one year. - Easing of modifications to working conditions. - Promotion of temporary layoffs and reductions in working hours, eliminating prior approval and adopting strong fiscal incentives. - Employment protection: - Reduced compensation for unfair dismissals of permanent workers to 33 days (previously 45 days) per year worked, with a maximum of 24 months (previously 42 months). - Compensation for fair dismissals of permanent workers remained at 20 days per year worked with a maximum of 12 months. - Greater flexibility for fair dismissals due to unfavorable economic conditions. - Lower dismissal costs and increased efficiency via revised procedures. - Hiring: - New permanent contract for small firms (less than 50 workers) with a trial period of one year, with significant fiscal and social security benefits. - Limit to renewal of temporary contracts up to two years. - Increased flexibility in regulation of part-time jobs and teleworking. - Active labor market policies: - Greater incentives for training and apprenticeship contracts. - More professional training by private agencies. - Temporary Employment Agencies authorized to act as private placement agencies. ### Sectoral and regional job growth: broad but uneven - Sectoral shifts during the crisis: - Construction employment declined by two-thirds between early 2008 and end-2013. - Roughly 30 percent of industry jobs were lost during the crisis. - Service-sector employment declines were less pronounced; net employment growth in services has been positive since mid-2013. - Share of services rose from 69 percent of total employment before the crisis to 78 percent in 2013. - Share of construction fell from nearly 13 percent in 2007 to 5½ percent in 2013; industry share fell by 2 percentage points. - Since early-2014 sectoral developments: - Employment has been growing in each main sector (agriculture, industry, construction, services) since mid-2014. - Services account for almost 1.2 of the 1.4 million new jobs (i.e., more than 8 out of 10). - For industry and construction, employment remains far below pre-crisis levels despite similar percentage changes from troughs because they started from a very low base. - In 2016 job growth rates in construction and industry accelerated; construction even outpaced services in that period. - Since 2013, sectoral employment shares have remained broadly stable. ### Employment composition and productivity - Majority of services-sector employment gains are in lower-productivity activities. - Tourism-related activities account for approximately 13 percent of total employment but about one quarter of total net employment creation since end-2013. - Another quarter of new jobs are in professional, scientific, technical and administrative support activities. - 14 percent of new jobs are in public services (education and health care), which have labor productivity below the economy-wide average. - High-productivity sectors like information and communications and real estate services account for only 3½ percent of total employment. - Employment services (a sub-sector of professional services) have expanded by 50 percent since early 2014; these higher-productivity sub-sectors jointly make up another 2 percent of jobs and are trend rising. - The financial sector continues to reduce employment. - Approximate decomposition note: - Approximately two-thirds of the relative increase in service sector employment was mechanical due to the reduction in construction jobs; the other third was due to relatively better performance of service employment. ### Manufacturing branches, exports, and employment - Manufacturing branches linked to construction (wood and cork, metals, machinery and equipment) were among those that reduced employment most during the crisis. - In the upswing, employment has increased significantly in these sectors apart from mineral manufacturing. - Hiring in these areas appears supported by strong improvement in export performance. - Subsectors that increased exports and expanded employment most include other manufacturing, vehicles, food, and footwear and leather. - Employment in building-sector-related manufacturing branches appears to have benefitted from the pick-up in construction activity. - Overall, about half of the new jobs in manufacturing are in subsectors with above-average labor productivity. ### Vacancies and short-term demand - Greatest short-term employment demand, as suggested by vacancies, continues to be in lower-productivity sectors. - Public administration and private administrative services experienced the largest proportional rise in vacancies since 2013H2. - Among high-productivity sectors, information and communications saw the fastest growth in vacancies. - Vacancy statistics caveat: they also capture positions vacated by people moving out of lower-paid low-productivity jobs as demand in better-paid sectors picks up. ### Labor market segmentation and contract types - Spain’s labor market remains segmented; employment protection for permanent workers remains costlier than for temporary ones, including due to uncertain and cumbersome dismissal procedures. - Just over half of all new jobs created since 2014:Q1 have been temporary. - This has brought the share of temporary employees to just above 26 percent of total employees. - Spain still has the highest share of temporary employment in the euro area. - Several studies suggest widespread use of temporary contracts can: - Undermine total factor productivity growth by reducing employees’ effort. - Reduce incentives to acquire firm-specific knowledge and training opportunities. - Favor shorter-duration production projects, which tend to have lower productivity. - Temporary hiring distribution: - Temporary hiring is used across sectors; hires in construction and some services (such as public services, notably the health sector) are particularly likely to be temporary. - High-growth sub-sectors such as professional and technical service activities, communications, and tourism-related activities tend to extend more permanent contracts. - Policy development: - On March 29, 2017, the government agreed with trade unions to convert up to 90 percent of the approximately 300,000 temporary employment positions in public administration into permanent ones over the next three years. ### Data, methodology and scope of analysis - The analysis uses national accounts data for number of jobs, complemented by social security affiliation numbers for sub-sectoral analysis and labor force survey results for information by region, skills and contract terms. - Notes on data comparisons: - For the total economy, national accounts and labor force survey data are very close; differences by sector can reflect sectoral definitions and coverage (resident households vs. resident production units, cross-border workers). - For industry employment, national accounts show a pickup of less than 6 percent between 2014:Q1 and 2017, while labor force survey shows a rebound of nearly 12 percent; social security affiliation records an increase of 7½ percent over the same period. - Sectional focus in the original note: - Section B examines sectoral and regional developments, terms of employment, and productivity characteristics of new jobs. - Section C looks at skill requirements of new jobs and distribution of skills in the labor force. - Section D concludes. *Prepared by Hanni Schoelermann; content drawn from IMF staff analysis in the provided chapter.* ### 13. Involuntary part-time work increased substantially during the crisis and remains high ### 13. Involuntary part-time work increased substantially during the crisis and remains high ### Part-time employment and involuntary part-time work - Overall share of part-time work: - Rose from 11½ percent of total employment in 2007 to almost 16 percent in 2014, and receded to 15 percent in 2016. - Gender breakdown: - Male part-time employment doubled to about 8 percent of total employed men. - Female part-time employment increased by 2 percentage points to 24 percent. - Involuntary part-time employment: - Surged between 2007 and 2014. - Remains currently still above 60 percent, and is among the highest in the euro area. - Policy/background note: - The 2012 labor market reform introduced greater flexibility to use part-time contracts (OECD, 2013). ### Hidden slack in the labor market (U6 and underutilization) - U6 measure and trends: - Overall labor underutilization in Spain came down from its peak of 37 percent in 2013 to 28 percent in early 2017. - The decline in U6 is exclusively due to a drop in unemployment, with underemployment remaining constant. - Labor underutilization in Spain remains among the highest in the euro area. - Definition: - The U6 expresses the sum of unemployed persons, under-employed part-time workers, people seeking work but not immediately available, and those available but not seeking work, as a percentage of the extended labor force. - Demographic movement: - Net emigration of more than 0.7 million 25-to -64-year-olds between 2009 and 2015. ### Regional employment dynamics and sectoral composition - Regional recovery patterns: - Since the employment trough in 2013, employment in the Canary Islands and the South has increased by around 10 percent since 2013. - Employment has picked up by around 4 percent in the Northwest and Madrid. - In absolute terms, most new jobs have been created in the East, which accounts for 30 percent of total Spanish employment. - Employment concentration and sectoral drivers: - The East, Madrid and the Canary Islands have seen employment increasingly concentrated in a few core sectors. - Wholesale (including travel and hospitality) has been the main job creator in the East, the Canary Islands and Madrid (except Madrid where its share differs). - Communications employment gains have been highly concentrated in Madrid. - Northern and central Spain recovery has been driven by wholesale, public administration, and industry. - Regional unemployment and mobility: - Differential employment recovery is partly related to sectoral composition differences. - Large differences in regional unemployment rates point to limited labor mobility. - Widespread use of temporary contracts may reduce incentives to relocate because of transaction costs (housing, schooling). ### Skills mismatches and implications for productivity - Shifts in education composition: - Spain’s working-age population comprises both a high share of lower-educated persons with up to lower secondary education, and a high share of higher-educated persons with tertiary education. - Since the crisis, the share of lower-educated persons has decreased slightly while the share of university-educated persons has increased. - The share of medium-educated persons with upper secondary and non-tertiary post-secondary education has remained constant, at only about half of the euro area average. - Over-skilling and job-skill composition: - Only a little more than one quarter of the jobs created since the employment trough are in high-skill occupations (jobs requiring tertiary as well as post-secondary education). - The vast majority of new jobs involve predominantly routine and service tasks that require fewer skills. - Employment of university-educated workers has increased in absolute numbers, implying many have accepted jobs below their level of education. - OECD/McGowan and Andrews (2015) finding: 27 percent of Spanish workers are over-skilled for their jobs, compared to six percent of those in employment who lack skills for the job they perform. - Reducing the skills mismatch in Spain to the best practice level across OECD economies would increase allocative efficiency by around 10 percent. - Lower-skilled workers and unemployment: - Unemployment rate among lower-educated workers is one of the highest in Europe. - Lower-educated persons constitute more than half of the currently unemployed population in Spain. - The collapse of the construction sector contributes to long-term unemployment among former construction workers with low relocation rates to other sectors. - Evidence shows workers with manual skills are most likely to find jobs in agriculture, domestic service, industry and transport, sectors that have seen comparatively modest employment gains. - Regional qualification disparities: - Economically active persons in the poorer regions (South, Canary Islands and Center) tend to have a significantly lower level of education than in other regions. - The share of low-skilled persons in Spain’s South is almost twice as large as in its capital. - Low qualification levels limit prospects for labor mobility and finding employment, especially when overall unemployment is still high. ### Conclusions and policy recommendations - Nature of the recovery: - The labor market rebound is strong and broad-based but driven predominantly by an expansion of low-productivity employment. - Most new jobs are in wholesale, food and accommodation (tourism-related) activities, and professional and support services—all of which have below-average labor productivity. - Temporary contracts are still widespread and labor mobility remains low, holding back productivity growth. - Large share of involuntary part-time work points to slack in the labor market. - Skills and allocation inefficiencies: - Many new jobs do not match workers’ skill sets, creating inefficient allocation of labor. - Preference for hiring higher-skilled workers for lower-skill jobs may be explained by the relative scarcity of people with professional training. - Policy recommendations: - Address labor-market segmentation and promote on- and out-of-job training to improve employment quality and productivity growth. - Consider introducing a single open-ended contract with no ex-ante time limit and severance payments that increase gradually with tenure to make permanent contracts more attractive for employers. - Alternatively, consider gradually introducing the Austrian ‘backpack’ model so entitlements to severance payments acquired in one job can be carried to subsequent jobs and severance payments grow with tenure regardless of contract type. - Improve apprenticeship and training opportunities, including tailored support for low-skilled long-term unemployed, along the lines of the 2014 employment activation program and the new joint action plan to support long-term unemployed. - Promote labor mobility. - Finish the product market reform agenda to support employment recovery by improving overall allocative efficiency. *Source: IMF staff analysis, "Involuntary part-time work increased substantially during the crisis and remains high."* ### References ### References ### Bibliographic references - Anghel, B., S. de la Rica and A. Lacuesta (2014), “The Impact of the Great recession on Employment Polarization in Spain,” SERIEs, Vol. 5, No. 2–3, pp. 143–171. - Banco de España (2017), Annual Report 2016. Available at: http://www.bde.es/bde/en/secciones/informes/Publicaciones_an/Informe_anual/ - Cabrales, A, Dolado, J., and Mora, R. (2014), Temporary contracts are bad for your cognitive health: Evidence from PIAAC. Available at: http://voxeu.org/article/dual-labour-markets-job-training-andcognitive-skills - Cahuc, P., Charlot, O., and Malherbet, F. (2016), “Explaining the Spread of Temporary Jobs and its Impact on Labor Turnover”, International Economic Review, No. 57, 533–572. - Dolado, J.J., Ortigueira, S., and Stucchi, R. (2016), “Does dual employment protection affect TFP? Evidence from Spanish manufacturing firms”, Journal of the Spanish Economic Association, No. 7, 421–459 - Doménech, R., J.R. Garcia, and C. Ulloa. 2016, “Los efectos de la flexibilidad salarial sobre el crecimiento y el empleo,” BBVA Working Paper 16–05. - European Central Bank (ECB) (2017), “Assessing Labour Market Slack”, ECB Economic Bulletin, No. 3/2017. Available at: https://www.ecb.europa.eu/pub/pdf/other/ebbox201703_03.en.pdf - European Commission (2017), European Semester Country Report Spain 2017, SWD (2017) 74 final. Available at: https://ec.europa.eu/info/sites/info/files/2017-european-semester-country-report-spain-en.pdf - García-Perez, I., 2016, “El efecto de la Reforma Laboral de 2012 sobre la dualidad y el empleo: Cambios en la contratación y el despido por tipo de contrato,” Fedea Policy Papers, 2016/06. Available at: http://goo.gl/48L8kR - Horwitz, Laszlo, and Martin Myant (2015), “Spain’s Labour Market Reforms: The Road to Employment – or to Unemployment?”, Working Paper No. 2015.03, European Trade Union Institute. - Jansen, M., S. Jimenez-Martin, and L. Gorjon (2016), “The Legacy of the Crisis: The Spanish Labour Market in the Aftermath of the Great Recession”, New Skills At Work, JPMorgan Chase & Co and FEDEA, No. 2016/10. - International Labour Organization (ILO) (2014), “Skills Mismatch in Europe”, Statistics Brief, International Labour Office, Geneva. Available at: http://www.ilo.org/wcmsp5/groups/public/---dgreports/---stat/documents/publication/wcms_315623.pdf - International Labour Organization (ILO) (2014), “Spain: Growth with Jobs”, Studies on Growth with Equity, International Labour Office, Geneva. Available at: http://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents/publication/wcms_309955.pdf - International Monetary Fund, 2015, Spain: Selected Issues, IMF Country Report No. 15/233, August, Washington. Available at: https://www.imf.org/external/pubs/ft/scr/2015/cr15233.pdf - International Monetary Fund, 2015, Spain: Staff Report for the 2015 Article IV Consultation, IMF Country Report No. 15/232, August. Available at: http://www.imf.org/external/pubs/ft/scr/2015/cr15232.pdf - International Monetary Fund, forthcoming, Spain: Staff Report for the 2017 Article IV Consultation, IMF Country Report. - Izquierdo, M., and J.F. Jimeno (2015), “Un Análisis de los Resultados de la Última Encuesta de Formación de Salarios de la Empresas Españolas,” Boletín Económico, Banco de España, dated October, 55-64. - Izquierdo, M., S. Puente and P. Font (2013). “Evolución del desajuste educativo entre la oferta y la demanda de trabajo en España,” Boletín Económico, Banco de España, dated June. - Izquierdo, M., A. Lacuesta, and S. Puente (2013), “La Reforma Laboral de 2012: Un Primer Análisis de algunos de sus Efectos,” Boletín Económico, Banco de España, dated September, 55-64. - Lacuesta, A., S. Puente, and E. Villanueva (2012), “Sectoral Change and Implications for Occupational Mismatch in Spain,” Boletín Económico, Banco de España, dated July. - McGowan, Müge Adalet, and Dan Andrews (2015), “Skill Mismatch and Public Policy in OECD Countries,” Economic Department Working Papers No. 1210, Organization for Economic Co-operation and Development. - Organization for Economic Co-operation and Development (OECD) (1998), “Employment Measures in the OECD National Accounts: Comparison between National Accounts and Labour Force Statistics,” STD/NA(98)13, dated September. Available at: http://www.oecd.org/std/na/2682096.pdf - Organization for Economic Co-operation and Development (OECD) (2013), “The 2012 Labour Market Reform in Spain: A Preliminary Assessment,” dated December. Available at: http://www.oecd.org/els/emp/SpainLabourMarketReform-Report.pdf - Organization for Economic Cooperation and Development (OECD), 2014, “The 2012 Labour Market Reform in Spain: A Preliminary Assessment,” dated June. Available at: http://dx.doi.org/10.1787/9789264213586-en - Organization for Economic Cooperation and Development (OECD), 2014, Getting Skills Right: Good Practice in Adapting to Changing Skill Needs: A Perspective on France, Italy, Spain, South Africa, and the United Kingdom, OECD Publishing, Paris. - Puente, S., and A. Casado (2016), “Skills Mismatch Between Labour Supply and Demand in Spain,” Economic Bulletin, Banco de España, dated September. - Sloane, P. J., (2014), “Overeducation, Skill Mismatches, and Labor Market Outcomes for College Graduates”, IZA World of Labor, No. 88, dated November. Available at: https://wol.iza.org/uploads/articles/88/pdfs/overeducation-skill-mismatches-and-labor-market-outcomes-for-college-graduates.pdf. ### Corporate and household balance sheets — The impact from deleveraging - Summary overview: - Spanish corporates and households have substantially reduced their debt since mid-2010, strengthening markedly their balance sheets; pockets of vulnerabilities remain among some corporates (construction and real estate service sectors, as well as small and medium-sized enterprises (SMEs)) and some household segments (low income, unemployed, self-employed, and less wealthy). - Economic activity does not seem to have been constrained by the deleveraging process as household and corporates’ external funding needs have been limited. - A. Credit Development: An Overview - 1. Financing conditions have remained favorable, but total private sector borrowing has continued to decline. - Bank lending rates at low levels comparable to those in the core euro area. - Stock of total borrowing by corporates and households declined through 2017:Q1 (-0.2 percent year-on-year). - Contraction largely driven by decline in bank credit, in part due to write-off of nonperforming loans (NPLs). - Newly extended credit started to outpace repayment in 2017:Q1. - New bank lending remained relatively subdued and stood at around one third of the pre-crisis level. - Recent moderation partly reflects a switch by large corporates from bank borrowing to bond financing. - 2. Continued contraction of bank credit seems to largely reflect weak credit demand. - No evident signs of credit supply constraints in the banking system. - Banks have become financially sounder since the crisis with reduced problem assets, improving profitability, and stronger solvency and liquidity conditions. - Banking system’s capital position can support a moderate credit expansion amid ample liquidity supported by the ECB’s accommodative policies. - Banks now lending at lower interest rates, less stringent lending standards, and more favorable terms of credit. - Weak credit demand reflects deleveraging by corporates and households. - B. Corporate Balance Sheets — Corporate sector deleveraging and macroeconomic implications - 3. Corporate deleveraging magnitude and comparison: - Total borrowing and net accounts payable fell by nearly 30 percent of GDP (from 133 percent of GDP at mid-2010 to 104 percent of GDP at end-2016), mostly due to lower bank borrowing. - This compares with merely 2 percentage points reduction for EU-13 economies. - Spanish corporate sector still about 10 percentage points of GDP more indebted than European peers, but had a lower leverage than comparators (by around 4 percentage points as of end-2016). - 4. Shift in financing sources: - Deleveraging largely achieved through repayment and restructuring of bank loans. - Firms obtained more inter-company loans and nonresident funding; large corporates increasingly turned to the bond market. - Share of financing via the financial system fell from 70 percent in 2007 to 55 percent in 2016. - 5. Sectoral distribution of deleveraging: - Lower bank lending to construction and real estate service firms accounted for almost 80 percent of the reduction in total bank lending in the past year. - This reflects unfinished restructuring efforts in these sectors given still sizeable NPLs and excessive leverage of real estate-related firms. - Excluding write-offs of NPLs, bank lending to corporates expanded in 2017:Q1 for the first time since the crisis. - 6. Investment recovery financing: - Nonresidential investment in machinery and intellectual property recovered strongly; other nonresidential investment, including construction, remained sluggish. - In 2016, corporate sector generated a gross operating surplus and net property income of 19.5 percent GDP, while spending on fixed capital formation amounted to 14.8 percent of GDP. - Post-crisis increase in profit margins underpinned strong gross operating surplus, enabling corporates to rely on retained earnings to finance investment. - New bank lending with maturity of more than one year has grown but remains at just 30 percent of the pre-crisis level. - At the aggregate level, new long-term bank lending is commensurate with the level of nonresidential investment based on the average over the past decade. - Corporate deleveraging does not seem to impede the strong economic recovery. - 7. Changes in working capital management and short-term borrowing: - Corporate short-term borrowing from banks declined significantly; new short-term bank lending (maturity up to one year) now amounts to merely 8 percent of corporate total assets down from about 25 percent pre-crisis. - Corporates scaled down accounts payable and accounts receivable, and increased holdings of liquid assets. - Further analysis needed to fully understand these switches. - 8. Resource allocation and financing constraints: - Weak balance sheets may still constrain some firms from obtaining financing, but efficiency of resource allocation has improved. - Banco de España (2016 Annual Report) analysis shows firms with positive net investment are more profitable and productive, and have stronger balance sheets than firms whose investment is not sufficient to cover depreciation. - More productive firms undertake new investment while financially weak and less profitable firms are less likely to expand productive capacity, supporting a view of more efficient allocation of resources and credit within the economy. - Figures and indicators referenced: - Growth of private sector borrowing: -0.2 percent year-on-year (through 2017:Q1). - Corporate gross operating surplus and net property income: 19.5 percent GDP (2016). - Corporate spending on fixed capital formation: 14.8 percent GDP (2016). - New bank lending with maturity > 1 year: 30 percent of the pre-crisis level. - Short-term new bank lending (maturity up to 1 year): 8 percent of corporate total assets (post-crisis) vs about 25 percent (pre-crisis). - Decline in total borrowing and net accounts payable: nearly 30 percent of GDP (from 133 percent of GDP at mid-2010 to 104 percent of GDP at end-2016). - Share of financing via the financial system: 70 percent in 2007 to 55 percent in 2016. - Contribution of lower bank lending to construction and real estate service firms: almost 80 percent of reduction in total bank lending in the past year. *cr17320 - References* ### 10. Overall financial soundness of the corporate sector has improved, but pockets of ### 10. Overall financial soundness of the corporate sector has improved, but pockets of weaknesses persist ### Corporate leverage and capital structure - Ratio of total equity to total assets increased from 36 percent in 2008 to 45 percent in 2015. - In 2015, about 17 percent of firms had negative equity; their debt, considered to be at risk, accounted for about 7 percent of total corporate debt. - Debt at risk is higher than pre-crisis but well below the 2012 peak of 13.4 percent; the share of firms with negative equity has declined by 3.7 percentage points since the peak. - Highly leveraged firms (bottom quartile) began deleveraging only in 2013; deleveraging progress among more indebted firms appears limited. - The share of debt at risk was much higher for real estate-related firms (construction and real estate service) and SMEs. ### Corporate profitability - Return on assets (ROA) increased to 4.5 percent in 2015, up from the 2012 trough at 3.5 percent, but below the pre-crisis level at 7.4 percent. - In 2015, about 35 percent of firms were loss-making; their debt, considered to be at risk, accounted for roughly 12.6 percent of total corporate debt. - Around 56 percent of firms had a return on assets below 3 percent in 2015, compared with about 42 percent in 2007. - Profitability tends to be higher for large corporations; manufacturing firms show a smaller share of loss-making businesses, while real estate-related firms have a markedly higher share of debt at risk. - Firms with more debt tend to be less profitable; firms in the bottom quartile tend to record negative ROA, though their profitability has improved significantly in recent years. ### Corporate debt servicing capacity - Earnings before interest, tax, depreciation and amortization (EBITDA) relative to interest expense increased from 4 at the 2012 trough to 5.6 in 2015. - In 2015, nearly a quarter of firms were unable to generate sufficient earnings (based on EBITDA) to cover interest expense; these firms accounted for about 20 percent of total corporate debt. - The share of firms lacking capacity to service debt and their debt, while below peak crisis levels, still exceeded 2007 levels. - The portion of firms lacking capacity to service debt was broadly similar across segments by business type and firm size, but the share of debt at risk was much higher for real-estate related firms. - In 2015, about 10 percent of firms were unable to generate sufficient earnings (based on EBITDA) to cover interest expense for three consecutive years (2013–15); their debt accounted for around 9 percent of total corporate debt. For real estate-related firms, the share of similar problem debt exceeded 38 percent. ### Sectoral concentration and nonperforming loans (NPLs) - Construction and real estate service sectors remain weak, with elevated leverage, lower profitability, and higher debt relative to income. - Banks’ NPLs are higher in these sectors: around 25 percent of total bank loans for real estate-related firms as of 2017:Q1. - NPL ratios for corporate lending were 7.7 percent for lending to large corporates and 17.4 percent for lending to SMEs as of mid-2016. - NPL ratios for corporate lending in Spain remain much higher than in most major European economies despite significant declines. ### Financial stability implications and sensitivity analysis - Where borrowers lack capacity to service debt, NPLs may not be recovered; accurate provisioning by banks is critical. - Sensitivity analysis using firm-level data: - A decline in earnings by 10 percent would increase the share of financially weak firms (those with insufficient EBITDA to cover interest expense) by 0.5 percentage points; the respective share of debt owed by these firms would edge up by 0.4 percentage points. - A rise in interest rates by 200 basis points would increase the share of financially weak firms by 2.4 percentage points; the respective share of debt owed by these firms would increase by 4.7 percentage points. - Conclusion: the impact of a sharp increase in funding costs is more prevalent than an earnings decline; banks should focus on completely addressing existing problem assets. ### Household balance sheets and interaction with corporate sector - Total household borrowing and net accounts payable declined from 84 percent of GDP at mid-2010 to 65 percent of GDP at end-2016. - Household deleveraging has been driven primarily by the decline in borrowing from domestic banks; household indebtedness is now broadly comparable with European peers though still relatively high across advanced Europe. - Household borrowing from banks continued to contract owing to strong amortization of existing mortgages. - New bank lending started to increase from a very low base, supported by stronger demand for residential mortgages and consumer credit; however, growth of new bank lending was insufficient to overturn the decline in total bank lending driven by contraction of lending for home purchases. - In 2017:Q1, total bank lending to households declined by 1.4 percent year-on-year, while new bank lending grew by 6.4 percent year-on-year. *Sources: Banco de España; Haver Analytics; and IMF staff calculations.* ### 18. The need for households to obtain financing still appears to be limited (Figure 7). ### 18. The need for households to obtain financing still appears to be limited (Figure 7) ### Household financing need and aggregate flows - In 2016, households earned gross disposable income of 63 percent of GDP. - Household expenditures in 2016 amounted to 58 percent of GDP for consumption and 3 percent for gross fixed capital formation. - At aggregate, the relatively low post-crisis level of residential investment enables households to enjoy consumption (relative to income) at a similar level prevailing pre-crisis without the need to borrow for financing gross fixed capital formation undertaken by households. - Some households still need to borrow for financing both consumption and home purchases. - New bank lending for consumer credit has increased in recent years, with the amount of such new household borrowing appearing broadly in line with the level of consumption based on the behavioral relationship over the past decade. - Households continue to repay existing mortgages, bringing down overall leverage of household investment in housing—based on outstanding borrowing for home purchases relative to housing wealth—in presence of relatively small, albeit increasing, new mortgage borrowing. ### Consumption, savings, and deleveraging - Buoyant consumption has accounted for slightly more than half of real GDP growth that has been more than 3 percent in the past two years. - Household savings have declined to 5 percent of GDP at end-2016, down from 9 percent of GDP at end-2009. - Certain households—low-income, self-employed, jobless, and less wealthy—may face constraints given their excessive indebtedness. - Overall, the amount of household borrowing for financing consumption does not appear low given the current level of consumption. ### Housing market, residential investment, and mortgage demand - Economy-wide house prices were at around 70 percent of the previous peak at end-2016, after having started to recover since 2014. - The crisis affected prices of existing dwellings much more than newly built houses. - Total transactions in the housing market have picked up, led by sales of existing properties. - Residential construction activity remains at a low level and the amortization of existing mortgages continues to outpace new borrowing. - 14 percent of new mortgage loans feature a loan-to-value above 80 percent. - The average loan-to-value for new mortgage lending has gone up to the pre-crisis level. ### Financial soundness of the household sector and vulnerabilities - Household leverage—based on total liabilities to total wealth—declined from 14.4 percent at mid-2013 to 11.5 percent at end-2016, which is slightly above the pre-crisis level. - Macroeconomic improvements—strong employment growth, rising income and wages, and reduced unemployment—have helped bolster household financial soundness. - Spanish households have generally less financial assets than European peers, leaving fewer liquidity buffers to withstand adverse financial situations. - Unemployment remains particularly high at 19 percent as of 2017:Q1. - In 2014, the share of households with debt payment to income above 40 percent amounted to 12 percent for households with income in the bottom quintile and 18 percent for self-employed households, compared with 7 percent for the entire sample. - The share of households with debt exceeding income at least 3 times amounted to 23 percent for employed households, 25 percent for self-employed households, and 29 percent for less wealthy households, compared with 15 percent for the entire sample. ### Nonperforming loans and bank exposure - As of mid-2016, the NPL ratio for household lending was 4.5 percent for Spanish banks. - Three quarters of lending to households in Spain is for home purchases. - Expected losses from household lending are likely to be more limited given borrower liability if foreclosed properties do not cover entire outstanding debt. ### Financially stretched households and debt-at-risk metrics - In 2014, 5.3 percent of households had debt payment to income ratios above 40 percent and debt to income ratios above 3. - Debt considered to be at risk accounted for 19 percent of total household debt in 2014. - Debt at risk not covered by assets amounted to 2 percent of total household debt. - Low-income households had a higher share of debt at risk not covered by assets at 12 percent, though their small borrowing limits the impact on bank balance sheets. ### Sensitivity and scenario analysis (based on 2014 household-level data) - Three shocks considered: (i) decline in income by 10 percent; (ii) rise in interest rates by 200 basis points; and (iii) decline in house prices by 10 percent. - For total households: - The share of financially stretched households would increase by 1.3–1.4 percentage points as a result of each of the first two shocks (declining income or rising interest rates). - The share of corresponding debt at risk would edge up by 4-6 percentage points under those shocks. - The increase for debt at risk not covered by assets would rise only by 0.4 percentage points under the income and interest rate shocks. - A decline in house prices by 10 percent would raise the share of debt at risk not covered by assets by 0.6 percentage points. - The increase in debt at risk not covered by assets could be higher for low-income, unemployed, and less wealthy households depending on shocks. - Despite potentially limited losses, banks may still see a fair amount of NPLs given that the share of total debt at risk is non-negligible (19 percent). ### Deleveraging progress and macro implications - Corporates and households have reduced their debt (based on total borrowing and net accounts payable) by nearly 50 percent of GDP since mid-2010. - In 2016, debt write-offs accounted for two-thirds of the net decline in total lending for banking business in Spain. - Financing need of corporates and households remains subdued due to ample retained earnings for corporates and still limited demand for home purchases for households. ### Conclusions and policy options - Continue accelerating the cleanup of bank balance sheets: NPLs that cannot be recovered should be properly recognized and proactively managed; Banco de España’s analysis on corporate financial soundness and housing market dynamics could support supervisory actions. - Enhance the insolvency framework: strengthen the out-of-court agreements on payments (OCAP) process for SMEs by facilitating coordination and promoting unified approaches among banks; adopt a more efficient class-voting system, including a cram-down mechanism; remove the special treatment of public sector creditors in out-of-court processes. - Bolster the banking system’s capacity to perform credit intermediation: banks should further adjust to profitability challenges and strengthen capital and funding positions; more capital is needed to compensate for the phase-out of exemptions allowed by the Capital Requirements Regulation, to support cleanup of existing impaired assets, and to maintain resilience; banks need to prepare for the ECB’s eventual exit from its accommodative policies. - Expand the eligible collateral pool for SMEs to obtain funding: simplify and expand the registry for security rights over movable assets to lower costs and broaden types of collateral, improving access to finance for SMEs and startups. - Enhance risk monitoring and systemic surveillance: close monitoring is warranted given that about one-sevenths of new mortgage lending features relatively high loan-to-value. *Source: cr17320 - 18. The need for households to obtain financing still appears to be limited (Figure 7).* --- _Source: https://www.imf.org/-/media/files/publications/cr/2017/cr17320.pdf_