## Measuring Size-Related CIT Tax Incentives (cr1724)

## Source details

**Canonical URL:** [Measuring Size-Related CIT Tax Incentives (cr1724)](https://www.imf.org/-/media/files/publications/cr/2017/cr1724.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2017/cr1724.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2017/cr1724.pdf.json)

---

### Background and motivation
- Spain exhibits a sizeable productivity gap relative to Germany and France; the gap widened since the 1990s and narrowed somewhat during the global financial crisis, but remains large relative to the 1970s–1990s.
- Weak productivity performance linked to:
  - Dominance of many low productive small firms.
  - Inefficient allocation of resources (high resource misallocation among advanced European countries, though it declined post-crisis).
  - Firms that are smaller, less productive, less innovative, grow more slowly, and are less export-oriented than European peers.
- Chapter focus:
  - Regulatory and business environment.
  - Size-related public policies, including CIT tax incentives for small firms.
  - Innovation activities.
  - Access to finance and firm-specific balance sheet factors.
- High-level policy implication: largest productivity gains from lowering regulatory barriers to competition and the cost of doing business (including at the regional level). Complementary gains from improving access to equity and credit for innovative start-ups, addressing disincentive effects of size-contingent rules, and improving efficiency of R&D incentives.

### Data and empirical strategy
- Sample: about 500,000 Spanish companies for the 2000–13 period; ORBIS firm-level data compiled by IMF Research Department.
- Coverage: firms included account for about half of corporate gross value added and two thirds of total cost of employees; micro firms somewhat underrepresented.
- Primary regression framework:
  - Y_ist = β * Exp_s * Reg_t + γ * (EMTR_nt_i,s,t − EMTR_i,s,t) + δ * X_ist + α_a + α_i + α_t + ε_ist
  - Y_ist measures TFP level, TFP growth or value added growth for firm i in sector s at time t.
  - Exp_s * Reg_t captures sectoral exposure interacted with regulation (β).
  - (EMTR_nt_i,s,t − EMTR_i,s,t) captures size-related tax incentives (γ).
  - X_ist includes firm innovation activity, access to finance, and firm size.
  - Controls: year fixed effects α_t, five age-category dummies α_a, firm fixed effects α_i; robust standard errors clustered at the firm level.

### Explanatory variables and measurement
- Product market regulation:
  - Proxy: OECD Product Market Regulation (PMR) index (varies in the (0,6) range; higher = more stringent).
  - Interaction: PMR index interacted with sectors’ exposure to regulation (proxied by U.S. firm turnover at two-digit NACE level).
  - Interpretation: negative β on interaction → easing regulation disproportionately boosts performance in sectors highly exposed to regulation.
- Size-related tax incentives:
  - Measured as EMTR (standard CIT rate) − EMTR (targeted lower CIT rate for SMEs).
  - EMTRs simulated using firm-level data combined with tax code information; forward-looking using a hypothetical marginal investment project.
  - Interpretation: negative γ on EMTR differential → lowering the differential between standard and SME-targeted EMTRs would boost TFP level and growth and value added growth.
- Access to finance:
  - Proxy: debt-to-asset ratio (debt = short-term financial debt + long-term financial debt, as in ECB (2014)).
  - Interaction: debt-to-assets interacted with SME dummy to test differential effects by firm size.
- Innovation:
  - Proxy: share of intangibles to total assets (following IMF, 2016c).
- Firm size and age:
  - Firm size proxied by lagged value added growth; lagged firm size an important determinant.

### Box 1 — EMTR method and key assumptions
- Method:
  - Uses forward-looking effective marginal tax rates (Devereux and Griffith, 1999) to evaluate tax impact on firm cost of capital for a hypothetical marginal investment project.
  - EMTR defined as the difference between the cost of capital and the after-tax rate of return of an alternative asset, relative to the cost of capital.
  - Size-related tax incentive = EMTR (standard CIT rate) − EMTR (targeted lower CIT rate for SMEs).
- Key assumptions used in EMTR simulations:
  - (i) real rate of return of 5 percent, annual inflation of 2 percent and a pre-tax real rate of return on capital of 20 percent;
  - (ii) economic depreciation rates for four categories: machinery, buildings, intangibles, and land and the share of assets by these categories, resulting in tax saving on depreciation allowance;
  - (iii) firm-level data used to compute the share of debt financing.
- Source data for statutory and SME-specific CIT rates and thresholds: OECD database. Country-specific depreciation rules taken from ZEW (2012). Where firm parameters unavailable, assumptions based on ZEW (2012) following Egger et al. (2009).

### Stylized facts (TFP, firm size, sectors, and regulation)
- Firm TFP varies across firm size, sectors, and time.
- 2013 firm composition:
  - Micro and small firms (with less than 20 employees) comprised 90 and 95 percent of the total number of non-financial firms in Spain, respectively.
- Sectoral TFP:
  - Lowest productivity and largest productivity differentials: market services, agriculture, and energy.
  - Highest productivity: manufacturing.
- Time trends:
  - Firm TFP declined well before the global financial crisis.
  - Post-crisis: real TFP stabilized and started to recover among larger firms; decline persists among micro firms.
  - Firm growth (real value added growth) lost dynamism for larger firms post-crisis; real value added of micro firms declined.
- Regulation and market entry:
  - OECD PMR shows elevated “de jure” regulation relative to peers despite recent improvements.
  - Barriers: barriers to entrepreneurship, barriers to entry in services, cumbersome license and permit systems.
  - De facto regulation and regional fragmentation important; time to open an industrial SME varies greatly across regions, with some southern regions taking twice the time compared to some northern regions.

### Size-related tax incentives, CIT rates and rules (selected details)
- EMTR increases with firm size in the sample; within SMEs, smaller firms get higher benefits.
- CIT rates and reforms:
  - During 2010–14 the CIT rate varied between 20 and 30 percent.
  - Smallest firms (less than 25 employees and turnover below €5 million) taxed at the lowest end of the 20–30 percent range.
  - Larger SMEs taxed at 25 percent for the first €300,000 of profit and at 30 percent for profits exceeding this threshold.
  - Large firms taxed at the upper end of the 20–30 percent range.
  - 2015 tax reform: CIT rate lowered to 25 percent for all firms and the lower CIT rate for small firms was replaced with a 15 percent CIT rate for new firms.
- Sectoral and age differences:
  - Agricultural, service and energy sectors benefit from relatively larger tax incentives.
  - Current incentive structure does not favor young companies; younger firms have lower tax benefits compared to older firms.
- Size-dependent rules and thresholds:
  - Audit, accounting, or labor regulations apply at employment thresholds (e.g., more than 10, 30, and 50 employees).
  - Firm density drops around employment thresholds, including at 50 workers.
  - Turnover thresholds of €450,000 and €6 million affect simplified VAT regime and CIT advance payments (recent reforms reduced some burdens).

### Access to finance and crisis patterns
- Debt-to-asset ratios relatively high and increased for SMEs during the crisis.
- Excessive reliance on bank financing and very low availability of venture capital constrain R&D investment and productivity growth (Aghion, 2014).
- Financing constraints proxied by debt-to-asset ratio prevent firms from investing and growing, leading to inefficient resource allocation and lower productivity level and growth.

### Empirical results: determinants of firm productivity and growth (selected estimates)
- Panel data 2000–13; fixed effects regressions (Table A3).
- Selected coefficients (preserved values; coefficients scaled by standard deviations of respective variables):
  - Lagged total factor productivity: -0.84*** (columns a–c); -0.29*** (columns d–f)
  - Tax incentive (EMTR): -5.57***; -5.58***; -5.91***; -5.95*** (various specifications)
  - Debt-to-assets: -1.05***; -0.76***; -1.09***; -0.85***; -0.02 (selected columns)
  - Debt-to-assets*sme: -0.29***; -0.23***; -0.84***; -0.85*** (selected columns)
  - Innovation (share of intangibles): 0.11***; 0.00*; 0.18***; 0.01; 0.01
  - Innovation*Regulation: -0.00***; -0.00***; -0.00***
  - Size (lagged value added): -0.06***; 0.06***; -0.58***; -0.59***
  - Regulation (Exposure*PMR): -0.02***; -0.03***
  - Age categories (2–5 years to Over 20 years): coefficients range from -0.01** to -0.09*** across specifications
  - Time dummies: Yes
  - Constant: 7.05***; 6.97***; 6.49***; 11.21*** (selected)
  - Number of observations: 2,051,207; 2,051,352; 2,089,036 (across columns)
  - R-sq.: 0.47; 0.11 (across different dependent variables)
- Interpretation of key empirical findings:
  - Regulation: entry barriers, administrative burden, and regional fragmentation are detrimental to productivity and productivity growth, especially in sectors exposed to regulation (negative and significant Exposure*PMR).
  - EMTR (size-related tax incentives): negative and significant effect on TFP level, TFP growth, and value added growth—evidence of disincentives to grow and a “small business trap.”
  - Debt-to-assets: high ratios (credit constraints) negatively and significantly correlated with TFP level and growth; effect stronger for SMEs (negative interaction with SME dummy).
  - Innovation: higher intangibles share associated with higher productivity and growth but direct positive effect relatively small; interaction innovation x regulation negative, implying stringent regulation can reduce productivity benefits of innovation.
  - Firm size and age: smaller firms less productive; productivity growth tends to be negatively related with firm size and age—larger firms reap economies of scale but are less dynamic.

### Policy implications and recommendations (firm productivity, taxation, and finance)
- Priority reforms:
  - Foster competition in product and service markets: reduce regulatory strictness and administrative burdens impeding entry and creative destruction; press ahead with Market Unity Law implementation and liberalization of professional services.
  - Improve access to equity and credit financing, particularly for small and innovative firms; reduce reliance on bank financing and expand venture capital.
  - Eliminate size-related CIT tax incentives (progress made by 2015 reform) and further reduce other size-related rules and regulations (reporting, auditing, labor thresholds) that create small business traps.
  - Increase efficiency of R&D incentives and enhance private R&D investment to generate positive spillovers.
  - Address labor market duality and improve quality of labor input via education reforms.
- Overall: addressing product and service market regulation, size-related tax incentives, and financing constraints should help raise TFP and firm growth.

### External sector, trade elasticities, and DSA scenarios (selected)
- Trade and external balances:
  - Latest estimates put the income elasticity of imports at around 1½.
  - Fall in the oil price (about 47 percent) improved the trade balance by about €20 billion (2 percent of GDP).
  - NIIP of about 90 percent of GDP at end-2015; net external debt ratio of 88 percent of GDP at end-2015.
- Baseline and adverse scenarios (medium run projections):
  - Baseline: trade surplus projected at 2.9-3.1 percent of GDP (current account surplus of 1.8-2.0 percent of GDP) and external debt at 142 percent of GDP.
  - Euro appreciation scenario: Euro appreciates against the USD by 30 percent → REER-ULC appreciates by 8 percent → trade balance reduces to about ½ percent of GDP; external debt 149 percent of GDP in 2021 (8 pps above baseline).
  - Oil price reversion scenario: oil price reverts to its 2001–15 average of US$ 66 in 2017 → trade balance reduced to 1.4 percent of GDP; external debt 152 percent of GDP by 2021.
  - External demand slowdown scenario: external demand decelerates by 1 percentage point every year → trade balance reduced to 1.7 percent of GDP; external debt 152 percent of GDP.
  - Combined negative shocks (Euro appreciation, oil price increase, external demand deceleration): trade balance could turn into a deficit of about 1½ percent of GDP; external debt 157 percent of GDP in 2021; combined scenario would add 13 percent of GDP to external debt in 2021 compared to baseline.
- Export concentration and firm-level dynamics:
  - In 2015, largest 100 exporters (about 0.2 percent of exporting firms) accounted for 42 percent of total exports.
  - Largest 5,000 exporters (about 10 percent of exporting firms) accounted for about 90 percent of total exports.
  - Number of exporting firms rose by about 50 percent between 2011 and 2015; about 70 percent of total export growth since 2011 explained by firms exporting more than €50 million per year (an increase of 107 companies).

### Banking system: structure, profitability, asset quality, and capacity to support growth
- System structure and size:
  - Banking system assets about 250 percent of GDP (accounted for 70 percent of aggregated assets) as of end-2015.
  - Insurance, pension funds, and investment funds accounted for 17 percent of aggregated assets.
  - Shadow banking around 20 percent of GDP as of end-2014.
  - Securitization-related financial vehicle corporation assets about 20 percent of GDP.
  - Outstanding private sector debt securities issued domestically and internationally amounted to 68 percent of GDP as of end-2015.
  - Stock market capitalization stood at 62 percent of GDP as of end-2015.
  - Number of deposit-taking institutions decreased by about a third during 2008–15 to 134.
  - Herfindahl index of banking system concentration doubled between 2007 and 2015.
  - Fourteen significant banks under the SSM account for about 93 percent of banking system assets on a consolidated operations basis.
- Profitability and low interest environment:
  - As of 2016Q2, return on assets for banking business in Spain over the past year was about 0.7 percentage points lower than in 2007.
  - Declines in profitability owing to lower net interest income and other operating income of 0.27 and 0.35 percentage points, respectively.
  - To maintain net interest income given compressed margins, credit would need to grow by about 5.5 percent annually, compared with IMF staff projections of annual credit growth of slightly above 1 percent over the next five years.
- Asset quality and provisioning:
  - Consolidated NPL ratio fell to 5.8 percent in June 2016 from 6.9 percent a year earlier; peak in 2013 was 9.4 percentage points higher for banking business in Spain.
  - NPLs for banking business in Spain were €123 billion by mid-2016.
  - NPLs remain highest in construction and real estate: NPLs amount to 28 percent of outstanding lending to the sector and account for 38 percent of total NPLs.
  - Provisioning coverage of nonperforming exposures (with provisions and collaterals accounted) was 95 percent for the Spanish banking system as of end-June 2016 (EU-wide average 86 percent).
  - Aggregated uncovered NPEs amounted to 4.1 percent of CET1 capital for significant Spanish banks overall (EU-wide level 9.5 percent).
  - If problem assets decline at about €30 billion a year, it could take on average more than five years to completely resolve legacy assets.
- Capital adequacy and capacity to support credit growth:
  - Total capital ratio: 14.6 percent at end-June 2016, up from 13.7 percent at end-2014.
  - CET1 capital: Spanish banks’ holding exceeds regulatory minima but lags European peers.
  - Banking system capacity estimates (Figure 12 summary):
    - Previous year’s profits could support system-wide credit growth of about 6.5–8.5 percent per year if banks used annual profits to fulfill regulatory capital requirement, keeping capital ratio unchanged (baseline profits based on 2015H1).
    - If profits drop by 25 percent over the next three years, capacity still at 6.5 percent per year.
    - If profits halve compared to the 2015H1 level, capacity drops to 4.4 percent annually.
    - IMF staff projects credit to grow about 1 percent annually over the next three years.
  - Footnote assumptions: dividend payout ratio 25 percent; tax rate 25 percent; assumes no additional cleanup or capital build-up efforts in baseline; funding access via TLTRO-II not an issue.
- Policy implications for banking:
  - Continue NPL reduction, adequate provisioning, fuller use of enhanced insolvency regime, and bank-specific time-bound NPL reduction plans.
  - Improve efficiency (branch rationalization, technology) and raise non-interest income.
  - Bolster high-quality capital through retained earnings to ensure capacity to support growth and meet new regulatory requirements (MREL, NSFR).

### Fiscal non-compliance among regions: determinants and policy issues
- Context: weak compliance among regional governments hampered Spain’s fiscal consolidation; non-compliance increased following the sovereign debt crisis and accounted for bulk of general government target deviation.
- Definitions:
  - Voluntary non-compliance: region has capacity but not incentives to comply.
  - Involuntary non-compliance: region willing and ex-ante capable but lacks capacity ex-post due to shocks rendering targets unfeasible.
- Key empirical findings (dynamic panel, Arellano-Bond first-difference GMM, regions 2002–15, Basque Country excluded):
  - Common shocks and adjustment needs contributed to involuntary non-compliance; non-compliance frequencies moved with nominal GDP growth forecast errors.
  - Fiscal non-compliance displayed inertia (positive autocorrelation), reflecting budget rigidities.
  - Adjustment efforts (difference between fiscal deficit target in year t and t-1) increased non-compliance; uniform targets across heterogeneous regional starting positions amplified difficulty.
  - Regions with limited fiscal autonomy exhibit larger non-compliance frequencies and smaller compliance margins; vertical fiscal imbalances negatively affect compliance.
  - Political factors: non-compliance frequencies increase during national and regional election years, among regions politically aligned with the center, and in regions with strong pro-autonomy preferences.
- Selected regression highlights (exact reported estimates):
  - Lagged compliance margin: model estimates include 0.74*, 1.09**, 0.91**, 2.42***, 0.31* across models.
  - Growth forecast errors: 0.09*; 0.12***; 0.12*** across models.
  - Region-National growth differential: -2.09*; -2.67*; -2.32*; -2.17* across models.
  - Fiscal Target Adjustment: -0.80**; -1.13**; -0.82***; -0.51*** across models.
  - Fiscal Rule Index X Lagged compliance margin: -0.06** (model 8).
  - Lagged Annual Change in Implicit interest rates: 1.03*** (model 10).
  - Regression diagnostics: Hansen test p-values reported (instrument validity) and m1/m2 serial correlation p-values presented; m2 p-values ≥ 0.05 across models reported.
- Policy issues for discussion:
  - Improve macro-fiscal forecasting at central and subnational level to reduce unanticipated common shocks.
  - Assess reforms to regions financing system and spending mandates to increase fiscal autonomy and reduce vertical fiscal imbalances.
  - Adopt differentiated regional fiscal targets to improve feasibility without undermining incentives.
  - Implement rules-based automatic enforcement mechanisms to ensure compliance during election years and in good times.

*Italic: Source — IMF staff chapter "Measuring Size-Related CIT Tax Incentives" and related country report content (cr1724), extracted from the provided PDF content.*

### 1. Measuring Size-Related CIT Tax Incentives ___________________________________________  8

### 1. Measuring Size-Related CIT Tax Incentives

### A. Background and Motivation
- Spain exhibits a sizeable productivity gap relative to Germany and France; the gap widened since the 1990s and narrowed somewhat during the global financial crisis, but remains large relative to the 1970s–1990s.
- Spain’s weak productivity performance has been linked to:
  - Dominance of many low productive small firms.
  - Inefficient allocation of resources (high resource misallocation among advanced European countries, though it declined post-crisis).
  - Firms that are smaller, less productive, less innovative, grow more slowly, and are less export-oriented than European peers.
- The chapter assesses the relative importance of policy and firm-specific drivers of firm productivity and growth, focusing on:
  - Regulatory and business environment.
  - Size-related public policies, including CIT tax incentives for small firms.
  - Innovation activities.
  - Access to finance and firm-specific balance sheet factors.
- Main high-level policy implication noted up front: the biggest productivity gains can be expected from lowering regulatory barriers to competition and the cost of doing business, including at the regional level. Further improvements in access to equity and credit financing for innovative start-ups, addressing disincentive effects of size-contingent rules, and improving the efficiency of R&D incentives to enhance private R&D investment are also recommended.

### B. Data and Empirical Strategy
- Data:
  - Sample covers about 500,000 Spanish companies for the 2000–13 period.
  - Firm-level data from the ORBIS database of Bureau Van Dijk (BvD), compiled by the IMF Research Department.
  - Firms included account for about half of corporate gross value added and two thirds of total cost of employees.
  - ORBIS contains balance sheets, income statements, several productivity variables, location and sectoral classification over 2000-13; micro firms are somewhat underrepresented.
- Empirical strategy:
  - The primary regression framework:
    - Y_ist = β * Exp_s * Reg_t + γ * (EMTR_nt_i,s,t − EMTR_i,s,t) + δ * X_ist + α_a + α_i + α_t + ε_ist
    - Where Y_ist measures TFP level, TFP growth or value added growth for firm i in sector s at time t.
    - Exp_s * Reg_t captures sectoral exposure interacted with regulation (coefficient β).
    - (EMTR_nt_i,s,t − EMTR_i,s,t) captures size-related tax incentives (coefficient γ).
    - X_ist is a vector including firm innovation activity, access to finance, and firm size.
    - Year fixed effects α_t, five age-category dummies α_a, firm fixed effects α_i; robust standard errors clustered at the firm level.

### C. Explanatory Variables and Measurement
- Product market regulation:
  - Proxy: OECD Product Market Regulation (PMR) index.
  - PMR index varies in the (0,6) range, with higher values denoting more stringent regulation.
  - Interaction: PMR index interacted with sectors’ exposure to regulation, proxied by U.S. firm turnover at the two-digit NACE level.
  - Interpretation: A negative β on the interaction implies easing regulation would disproportionately boost performance in sectors highly exposed to regulation.
- Size-related tax incentives:
  - Measured as the difference between effective marginal tax rates (EMTR) under the standard CIT rate and EMTR under targeted (lower) CIT rates for smaller firms.
  - EMTRs are simulated using firm-level data combined with tax code information; the approach is forward-looking using a hypothetical marginal investment project.
  - Interpretation: A negative γ on the EMTR differential implies that lowering the differential between standard and SME-targeted EMTRs would boost TFP level and growth and value added growth.

Box 1 — Measuring Size-Related CIT Tax Incentives (method and key assumptions)
- Method summary:
  - Uses forward-looking effective marginal tax rates to evaluate tax impact on the firm cost of capital for a hypothetical marginal investment project (Devereux and Griffith, 1999).
  - EMTR defined as the difference between the cost of capital and the after-tax rate of return of an alternative asset, relative to the cost of capital.
  - Size-related tax incentive = EMTR (standard CIT rate) − EMTR (targeted lower CIT rate for SMEs).
- Key assumptions used in EMTR simulations:
  - (i) real rate of return of 5 percent, annual inflation of 2 percent and a pre-tax real rate of return on capital of 20 percent;
  - (ii) economic depreciation rates for four categories: machinery, buildings, intangibles, and land and the share of assets by these categories, resulting in tax saving on depreciation allowance;
  - (iii) firm-level data used to compute the share of debt financing.
- Source data for statutory and SME-specific CIT rates and thresholds: OECD database. Country-specific depreciation rules taken from ZEW (2012). Where firm parameters unavailable, assumptions based on ZEW (2012) following Egger et al. (2009).

### D. Literature and Mechanisms Identified
- Key mechanisms from the literature affecting firm productivity dynamics:
  - Labor market duality (share of temporary workers) and sector-level collective bargaining agreements have negatively affected firm TFP in Spain.
  - Stringent labor market regulations reduce allocation efficiency.
  - Financial frictions from size-dependent borrowing constraints and weak firm balance sheets have been important, especially post-crisis.
  - Product market regulation exacerbates misallocation and is detrimental to productivity dynamics in sectors exposed to regulation.
  - Size-related policies, including tax incentives for small firms, can create a “small business trap,” constraining firm and productivity growth.
  - Innovation, particularly by young and innovative firms, is positively linked to productivity growth; such firms contribute disproportionately more to productivity growth than small firms.

### E. Key Findings (empirical focus noted in chapter summary)
- The chapter’s empirical analysis investigates:
  - Stylized facts of TFP level and growth and value added growth over 2000–13.
  - The roles of regulation, size-related tax incentives, innovation, access to finance, and firm-specific factors for productivity and growth.
  - The relative importance of these factors to inform policy design.
- High-level empirical conclusion emphasized in the introduction:
  - The largest productivity gains are expected from lowering regulatory barriers to competition and the cost of doing business (including at the regional level).
  - Further contributions to productivity growth are expected from:
    - Improving access to equity and credit financing, particularly for innovative start-up companies.
    - Addressing potential disincentive effects of size-contingent rules (including size-related CIT incentives).
    - Increasing the efficiency of R&D incentives and enhancing private R&D investment, which should generate positive spillovers that are hard to capture empirically.

_Italic: Source — IMF staff chapter "Measuring Size-Related CIT Tax Incentives" (cr1724), extracted from the PDF content provided._

### 10.      Access to finance. The analysis uses debt-to-asset ratio as a proxy for the strength of firm

### 10.      Access to finance. The analysis uses debt-to-asset ratio as a proxy for the strength of firm 

### Access to finance (proxy and measurement)
- Debt-to-asset ratio is used as a proxy for the strength of firm balance sheets and access to finance.  
- Debt is defined as the sum of short-term financial debt (loans) and long-term financial debt (long-term liabilities), as in ECB (2014).
- Rationale: Excessive reliance on bank financing and very low availability of venture capital in Spain constrain R&D investment and productivity growth (Aghion, 2014). Access to finance is expected to be especially binding for small companies with relatively smaller tangible assets used as collateral.
- Empirical approach: access to finance proxy is also interacted with a dummy variable for SMEs to test differential effects by firm size.

### Other factors included in the analysis
- Innovation proxied by the share of intangibles to total assets (following IMF, 2016c).
- Firm-specific factors: firm size and firm age.  
- Firm size proxied by lagged value added growth; lagged firm size found to be an important determinant of firm productivity and growth.
- Note: labor market distortions were not proxied due to data limitations; time and firm fixed effects implicitly account for omitted factors.

### Stylized facts (TFP, firm size, sectors, and regulation)
- Firm TFP varies across firm size, sectors, and time.
- 2013 firm composition:
  - Micro and small firms (with less than 20 employees) comprised 90 and 95 percent of the total number of non-financial firms in Spain, respectively.
- Sectoral TFP patterns:
  - Lowest productivity and largest productivity differentials: market services, agriculture, and energy.
  - Highest productivity: manufacturing.
- Time trends:
  - Firm TFP declined well before the global financial crisis.
  - Post-crisis: real TFP stabilized and started to recover among larger firms; decline persists among micro firms.
  - Firm growth (real value added growth) lost dynamism for larger firms post-crisis; real value added of micro firms declined during the same period.
- Regulation and market entry:
  - OECD index of product market regulation shows elevated “de jure” regulation relative to peers despite recent improvements.
  - Barriers identified: barriers to entrepreneurship, barriers to entry in the service sector, and cumbersome license and permit systems.
  - De facto regulation and regional market fragmentation have important effects; time to open an industrial SME varies greatly across regions, with some southern regions taking twice the time compared to some northern regions.

### Size-related tax incentives and rules
- Effective marginal tax rate (EMTR) increases with firm size in the sample; within SMEs, smaller firms get higher benefits.
- CIT rates and reforms:
  - During 2010–14 the CIT rate varied between 20 and 30 percent.
  - Smallest firms (those with less than 25 employees and with turnover below €5 million) taxed at the lowest end of the 20–30 percent range.
  - Larger SMEs were taxed at 25 percent for the first €300,000 of their profit and at 30 percent for profits exceeding this threshold.
  - Large firms taxed at the upper end of the 20–30 percent range.
  - 2015 tax reform: CIT rate lowered to 25 percent for all firms and the lower CIT rate for small firms was replaced with a 15 percent CIT rate for new firms.
- Sectoral and age differences:
  - Firms in agricultural, service and energy sectors benefit from relatively larger tax incentives compared to other sectors.
  - Current incentive structure does not favor young companies; younger firms have lower tax benefits compared to older firms.
- Size-dependent rules and thresholds:
  - Certain audit, accounting, or labor regulations apply at employment thresholds (e.g., more than 10, 30, and 50 employees).
  - Firm density drops around employment thresholds, including at 50 workers.
  - Thresholds on firm turnover of €450,000 and €6 million affect application of simplified VAT regime and CIT advance payments (recent reforms have reduced some burdens).

### Access to credit: patterns during the crisis
- Debt-to-asset ratios have been relatively high and increased for SMEs further during the crisis.
- Excessive reliance on bank financing made financing constraints particularly strong during the crisis and constrained R&D investment.
- Financing constraints proxied by debt-to-asset ratio prevent firms from investing and growing, leading to inefficient resource allocation and lower productivity level and growth.

### Empirical results: determinants of firm productivity and growth
- Data and method:
  - Panel data over 2000–13, fixed effects regressions (Table A3).
- Key empirical findings (coefficients scaled by standard deviations of respective variables):
  - Regulation: regulatory entry barriers, administrative burden (licensing requirements), and market fragmentation across regions are detrimental to productivity and productivity growth, especially in sectors more exposed to regulation.
  - Debt-to-assets: high debt-to-asset ratio (credit constraints) is negatively and significantly correlated with TFP level and growth. Effect is stronger for SMEs.
  - Size-related tax incentives (EMTR): negatively and significantly affect TFP level, TFP growth, and growth in value added—creating disincentives for firms to grow and contributing to a “small business trap.”
  - Innovation (intangibles share): more innovative firms are more productive and grow faster; the direct positive effect is relatively small, possibly due to low R&D investment in Spain and potential offsetting negative interaction with regulation.
    - Interaction: innovation x regulation shows a negative effect, implying stringent regulation can reduce the productivity benefits of innovation.
  - Firm size and age:
    - Smaller firms are generally less productive than larger ones.
    - Productivity growth tends to be negatively related with firm size and firm age—larger firms reap economies of scale but are less dynamic and grow at a relatively slower pace compared to smaller firms.

### Policy implications and recommendations
- Reforms should continue to focus on:
  - Fostering competition in product and service markets (reducing regulatory strictness and administrative burdens that impede entry and creative destruction).
  - Improving access to equity and credit financing, particularly for small and innovative firms.
  - Eliminating size-related CIT tax incentives (already advanced by the 2015 reform) and further reducing other size-related rules and regulations to stimulate firm growth and productivity.
  - Implementing labor market and education reforms to improve the quality of labor input and boost firm productivity.
- Overall: policy factors—product and service market regulation, size-related tax incentives, and financing constraints—have played an important negative role for firm productivity level and growth; addressing these areas should help raise TFP and firm growth.

*IMF staff analysis based on firm-level data (ORBIS) and econometric estimation over 2000–13.*

### 21.      High productivity growth requires frontier innovations and efficient allocation of

### 21.      High productivity growth requires frontier innovations and efficient allocation of resources

### Key findings on productivity and firm dynamics
- High productivity growth requires frontier innovations and efficient allocation of resources.
- Firm readiness to innovate and the efficiency of within-sector allocation of resources are central determinants of firm TFP dynamics.
- Vulnerabilities and barriers include: barriers to competition, disincentives for firm growth, limited ability to innovate, and insufficient access to financing.
- Spain has achieved three years of current account surpluses alongside positive GDP growth, but vulnerabilities remain, including a large net international investment position (NIIP) of 90 percent of GDP.
- Since 2014 the current account showed surpluses sustained by regained competitiveness, export growth, low oil prices, and low interest rates.
- The staff’s baseline forecast envisages current account surpluses in the order of 1.8-2.0 percent of GDP over the medium term.
- The euro depreciated by about 20 percent against the US dollar since 2014, with direct and indirect positive effects on exports.
- The EU was Spain’s largest trade partner, accounting for 65 percent of Spain’s exports in 2015.
- About half of Spain’s nominal export growth in 2014 and 2015 is explained by exports to the EU.
- Exports to the UK amounted to about 3½ percent of GDP, of which about a quarter is tourism.
- Exports are sensitive to relative prices and external demand; imports have a high income elasticity of about 1½ and are sensitive to domestic demand and exports.
- Imports expanded by over 5 percent in real terms in 2015, well above real GDP growth.
- The export technological content is concentrated in medium-high technology goods (about 45 percent), with export composition roughly unchanged over the last decade.
- Investment in intangibles (R&D, training, advertising) was weak; latest available data for investment in intangibles is from 2010 and show Spain was one of the poorest performers in the EU-15.

### Policy recommendations to enhance firm TFP dynamics and innovation
- Foster competition in product and service markets by:
  - Addressing delays in implementation of the Market Unity Law.
  - Lowering regulatory barriers and administrative burden faced by firms from the three government layers.
  - Pressing ahead with the long delayed liberalization of professional services.
- Reduce and eliminate size-related rules and regulations (in reporting, auditing, and labor-related regulation) that create small business traps and hurt productivity and growth.
  - Note: Some disincentive effects were reduced with the 2015 tax reform that equalized the CIT rate across firms of different sizes to 25 percent and replaced lower CIT rates for small firms with a 15 percent CIT rate for new firms.
- Improve access to finance, in particular equity financing for small firms, to maintain robust investment growth and sustained growth in R&D investment.
- Enhance firms’ innovation capacity by addressing weaknesses in public R&D spending efficiency and weak public-private sector cooperation.
- Address labor market duality through further labor market reforms and improve the quality of labor input through education reforms to sustain TFP growth and encourage innovation.

### Empirical evidence and estimation results (selected)
- Sample spans the years 2000-13.
- Table A1: Spain: Average SME share and ORBIS sample size by year (selected totals)
  - Total average SME share: 0.990
  - Total sample size: 3,413,080
- Table A2: Spain: Average SME share and ORBIS sample size by sector (selected)
  - Total average SME share: 0.99
  - Total sample size: 2,844,192
  - Manufacturing sample size: 538,263; Av. SME share: 0.98
  - Wholesale and retail trade; repair sample size: 795,370; Av. SME share: 0.99
  - Construction sample size: 543,596; Av. SME share: 0.99
- Table A3: Spain: Effects of Policy and Firm Factors on Firm Productivity and Growth (OLS with Fixed Effects, Robust Standard Errors) — selected coefficient estimates (preserved values)
  - Lagged total factor productivity: -0.84*** (columns a–c); -0.29*** (columns d–f)
  - Tax incentive (EMTR): -5.57***; -5.58***; -5.91***; -5.95*** (various specifications)
  - Debt-to-assets: -1.05***; -0.76***; -1.09***; -0.85***; -0.02 (selected columns)
  - Debt-to-assets*sme: -0.29***; -0.23***; -0.84***; -0.85*** (selected columns)
  - Innovation (share of intangibles): 0.11***; 0.00*; 0.18***; 0.01; 0.01
  - Innovation*Regulation: -0.00***; -0.00***; -0.00***
  - Size (lagged value added): -0.06***; 0.06***; -0.58***; -0.59***
  - Regulation (Exposure*PMR): -0.02***; -0.03***
  - Age categories (2-5 years to Over 20 years): coefficients range from -0.01** to -0.09*** across specifications
  - Time dummies: Yes
  - Constant: 7.05***; 6.97***; 6.49***; 11.21*** (selected)
  - Number of observations: 2,051,207; 2,051,352; 2,089,036 (across columns)
  - R-sq.: 0.47; 0.11 (across different dependent variables)
- Notes: EMTR = Effective Marginal Tax Rate; EATR = Effective Average Tax Rate; PMR = Product Market Regulation.

### External sector and competitiveness insights
- Spain’s negative NIIP is still among the largest in Europe at 90 percent of GDP; low interest environment and long debt maturity mitigate some risks.
- Structural, cyclical, and external factors contributed to current account surpluses: stronger export dynamics, a large positive terms-of-trade shock from the fall in oil prices, and ECB monetary policy.
- Export market share growth (2010–15) was strong relative to peers, supported by wage moderation, firm internationalization, and ECB policy.
- Box 1 empirical elasticities (error correction model; estimation periods and selected elasticity estimates preserved):
  - Exports — Long term: Foreign demand 0.91 (non-oil goods), 0.74 (services); REER_ULC -0.63 (non-oil goods), -0.42 (services).
  - Exports — Short term: Δ Foreign demand 0.70 (non-oil goods), 1.96 (services); Δ REER_ULC -0.24 (non-oil goods), -0.07 (services).
  - Imports — Long term: Domestic demand 1.44 (non-oil goods), 1.70 (services); Exports 0.45 (non-oil goods), 0.36 (services).
  - Imports — Short term: Δ Domestic demand 0.46 (non-oil goods), 0.35 (services); Δ Exports 0.15 (non-oil goods), 0.17 (services).
  - Estimation period: Imports 1995Q1-2015Q4; Exports (period included within text).
- Spain’s export composition remained roughly unchanged over the last decade; about 45 percent of goods exports are medium-high technological content.

*Source: Chapter 21, “High productivity growth requires frontier innovations and efficient allocation of resources,” Country Report (Spain) — IMF.*

### 3.2 percent. Latest estimates put the income elasticity

### cr1724 - 3.2 percent. Latest estimates put the income elasticity

### Trade balance, import income elasticity, and oil price effects
- Latest estimates put the income elasticity of imports at around 1½.
- The continued high import income elasticity could weigh on the size and sign of future trade balances despite a baseline medium-term projection of sustained trade surpluses and gradual external sustainability improvements.
- Spain’s trade balance benefited significantly from lower oil prices:
  - The fall in the oil price (of about 47 percent) gave an impulse to economic activity and improved the trade balance by about €20 billion (2 percent of GDP).
  - Oil price volatility and uncertainty remain a risk factor for the future trade balance.

### Net International Investment Position (NIIP) and valuation effects
- Spain has one of the largest negative NIIP positions in Europe:
  - NIIP of about 90 percent of GDP at end-2015.
  - Net external debt ratio of 88 percent of GDP at end-2015.
- Structure and maturity:
  - Direct investment (equity) component is small compared to peers.
  - Only 9 percent of the portfolio investment falls due in the next twelve months.
  - ECB purchases of Spanish assets act as an additional mitigating factor against rollover and interest rate risks.
- Sectoral NIIP shifts between 2010 and 2015:
  - Consolidated public sector NIIP went from -17 to -54 percent of GDP.
  - Private sector reduced its NIIP from -71 to -36 percent of GDP.
- Valuation effects:
  - Valuation effects have been large and volatile, averaging about of -3 percent of GDP annually when excluding the 2012 crisis year.
  - From 2013-15, valuation effects worsened the NIIP by more than 5 percent of GDP per year.
  - Without valuation effects, the negative NIIP position would have been 17.5 percent of GDP smaller in 2015.
  - In 2015 the NIIP also suffered from a lower valuation of Spain’s foreign assets, especially direct investment, amounting to a valuation loss of about 3 percent of GDP.

### Extended external Debt Sustainability Analysis (DSA) — scenarios and projections
- Methodology:
  - Uses estimated export and import elasticities from error correction mechanism equations to simulate behavioral responses of the trade balance to shocks, allowing scenarios for oil prices, REER, domestic or external demand, etc.
  - Exports and imports are disaggregated into goods (energy and non-energy) and services to account for different elasticities.
- Baseline and scenario projections (medium run):
  - Baseline: trade surplus projected at 2.9-3.1 percent of GDP (current account surplus of 1.8-2.0 percent of GDP) and external debt at 142 percent of GDP.
  - Euro appreciation scenario:
    - Euro appreciates against the USD by 30 percent → REER-ULC appreciates by 8 percent.
    - Trade balance reduces to about ½ percent of GDP in the medium term.
    - External debt would be 149 percent of GDP in 2021 (8 pps above the baseline).
  - Oil price reversion scenario:
    - Oil price reverts to its 2001–15 average of US$ 66 in 2017 → trade balance reduced to 1.4 percent of GDP, external debt to 152 percent of GDP by 2021.
  - External demand slowdown scenario:
    - External demand decelerates by 1 percentage point every year → trade balance reduced to 1.7 percent of GDP and external debt of 152 percent of GDP in the medium term.
  - Combined negative shocks (Euro appreciation, oil price increase, external demand deceleration):
    - Trade balance would be hit much more severely, resulting in a deficit of about 1½ percent of GDP and external debt of 157 percent of GDP.
    - Combined scenario would add 13 percent of GDP to Spain’s external debt in 2021 compared to the baseline.

### Export performance, firm concentration, and firm-level dynamics
- Export concentration and firm shares:
  - In 2015, the largest 100 exporters (about 0.2 percent of all exporting firms) accounted for 42 percent of total exports.
  - The largest 5,000 exporters (about 10 percent of all exporting firms) accounted for about 90 percent of total exports.
  - High export concentration is explained by high atomization in Spanish firms: average Spanish firm size is a fraction compared to many other advanced economies.
- Destination concentration:
  - Export concentration varies significantly between destinations; non-EU (farther) markets are typically served by fewer and more productive firms.
- Export reliance and productivity:
  - On average, Spanish firms are not particularly reliant on selling abroad; both large and SME exporting firms have a relatively low turnover from exports compared to non-exporting peers.
  - Larger exporters have been more productive and resilient to macroeconomic shocks; unit labor costs typically rise less for larger, more productive firms.
  - The “Spanish Paradox”: resilient export shares despite REER appreciation in unit labor cost terms.
- Post-crisis dynamics:
  - The number of exporting firms rose by some 50 percent between 2011 and 2015.
  - About 70 percent of total export growth since 2011 is explained by the number of firms that export more than €50 million per year (an increase of 107 companies in this category).
  - Firms that started exporting after the crisis tended to remain exporting; the increase in regular exporters (exported for at least four consecutive years) reflects structural improvement.
  - As domestic demand recovers, incentives for new entrants to export may diminish.

### Key empirical findings on trade elasticities and risks
- Exports are sensitive to relative prices and external demand.
- Imports are very sensitive to domestic demand and exports.
- Regained competitiveness contributed to post-crisis export dynamism.
- Import growth associated with the economic recovery is a source of risk for the external surplus.

### Policy recommendations and conclusions
- Priority: preserve the current account surplus while supporting a dynamic economy to gradually reduce external vulnerabilities.
- Fiscal policy:
  - Strengthen the public sector balance sheet and gradually reduce the fiscal deficit to increase external resilience.
- Structural reforms:
  - Further enhance competitiveness, productivity growth, and employment.
  - Pursue ambitious labor and product market reforms to increase firm competitiveness and ease entry into exporting activities by removing obstacles to firm growth.
  - Implement policies to foster R&D and innovation to raise firms’ international competitiveness and continue reducing domestic and external imbalances.
- Note on mitigating factors:
  - Low interest rates have helped the income balance by reducing external interest payments on gross external debt while the remuneration on increasingly diversified external asset portfolios fell less sharply.
  - The ECB Public Sector Purchase Program (PSPP) is a mitigating factor as it has provided stable demand for Spain’s public sector liabilities.

*Source: IMF staff report.*

### References

### cr1724 - References

### References
- Altomonte, C., T. Sono, and H. Vandenbussche, 2013, "Firm-level Productivity and Exporting: Diagnosing the Role of Financial Constraints," Product Market Review (2013).
- Antràs, P., R. Segura-Cayuela, and D. Rodriguez-Rodriguez, 2010, "Firms in International Trade (with an application to Spain)," SERIEs Invited Lecture at the XXXV Simposio de la Asociación Española de Economía.
- Bank of Spain, 2014, “The Recent Behavior of Imports and their Determinants,” Economic Bulletin, April 2014 Quarterly Reporto on the Spanish Economy (Madrid).
- Crespo Rodríguez, A., G. Pérez-Quirós, and R. Segura Cayuela, 2012, "Competitiveness Indicators: The Importance of an Efficient Allocation of Resources," Economic Bulletin JAN (2012).
- Crespo, Aranzazu, and R. Segura-Cayuela, 2014, "Understanding Competitiveness," European University Institute Working Papers 2014/20 (Italy).
- Correa-López, M., and R. Doménech, 2012, “La Internacionalición de la Empresa Española,” BBVA Research WP 12/30. Available via the Internet: http://goo.gl/F4OybC.
- Hospido, L., and E. Moreno-Galbis, 2015, "The Spanish Productivity Puzzle in the Great Recession," Bank of Spain Working Paper 1501 (Madrid: Bank of Spain).
- International Monetary Fund (IMF), 2015, Spain—Article IV Consultation, IMF Country Report 15/232 (Washington).
- Verschelde, M., M. Dumont, G. Rayp, and B. Merleved, 2014, "European Competitiveness–A Semiparametric Stochastic Metafrontier Analysis at the Firm Level," ECB Working Paper 1701 (Frankfurt: European Central Bank).
- Prades, E., and C. Garcia, 2015, Actualizacion de la function de las exportaciones espanolas de bienes,” Boletin Economico, Banco de Espana, April 2015.
- Vidon, E., 2011, “Spain’s External Sustainability,” in Spain—Selected Issues Paper, IMF Country Report 11/216 (Washington: International Monetary Fund).
- Xifré, Ramon, 2014, “The Competitiveness of the Spanish Ecoomy, a Bird’s Eye View of the Four Largest Euro Area Economies,” IESE Research Paper D/1088 (Barcelona: IESE Business School).

### A. Introduction
- Chapter scope: discusses new macro-financial challenges amid ongoing balance-sheet adjustments and explores four macro-financial questions: (i) impact of low interest rates on banks’ profitability and future implications, (ii) spillovers from more difficult macro-environments in emerging economies, (iii) challenges from low profitability for ongoing bank balance-sheet adjustment, and (iv) how low profitability could affect banks’ capacity to support medium-term economic growth once credit growth picks up.

### B. Current Landscape
- Spain’s financial system structure and size:
  - Banking system assets amounted to about 250 percent of GDP, which accounted for 70 percent of aggregated assets, as of end-2015.
  - Insurance companies, pension funds, and investment funds accounted for 17 percent of aggregated assets.
  - Shadow banking amounted to around 20 percent of GDP as of end-2014.
  - Securitization-related financial vehicle corporations’ assets amounted to about 20 percent of GDP.
  - Outstanding private sector debt securities issued domestically and internationally amounted to 68 percent of GDP as of end-2015.
  - Stock market capitalization stood at 62 percent of GDP as of end-2015.
- System concentration and structure:
  - Number of deposit-taking institutions decreased by about a third during 2008–15 to 134.
  - Only two savings banks remain, down from 47 in 2007.
  - Herfindahl index of banking system concentration doubled between 2007 and 2015.
  - Fourteen significant banks under the Single Supervisory Mechanism (SSM) account for about 93 percent of banking system assets on a consolidated operations basis.
  - Two largest banks—Banco Santander and BBVA—have overseas operations; operations outside Spain contribute around 85 percent of group-wide earnings for these banks, and their combined assets account for about 57 percent of banking system assets.
  - Foreign subsidiaries and bank branches in Spain number about 100 but assets are relatively small (about 4 percent of banking system assets).

### C. Profitability and Low Interest Environment
- Profitability trends and drivers:
  - As of 2016Q2, return on assets for banking business in Spain over the past year was about 0.7 percentage points lower than in 2007.
  - Declines in profitability largely owing to lower net interest income and other operating income of 0.27 and 0.35 percentage points, respectively, with broadly similar impairment costs.
  - Spanish banks had higher return on assets than EU peers due to stronger pre-loss income despite larger impairment costs.
  - Spanish banks enjoy markedly larger net interest income, benefiting from a business model oriented to retail banking.
  - Cost-to-income ratio kept at a favorable level, though falling trend in operating expenses has reversed recently.
- Cross-bank variation:
  - In 2015, the two large international banks and three domestic-oriented banks registered relatively high return on assets by the European benchmark, benefiting from overseas subsidiaries, operating cost efficiency, and limited asset impairment.
- Key challenges and sensitivities:
  - Repricing of loans is sensitive to money market rates; interest rates for new lending have fallen toward core euro area levels.
  - Average interest deposit and lending rates based on outstanding amounts have declined more rapidly in Spain than in other parts of the euro area.
  - Repricing for SME and house-purchase lending is particularly sensitive to EURIBOR in Spain.
  - Housing loans account for a larger share of banks’ total assets in Spain.
  - Scenario: a further reduction of net interest margins by 10 basis points could lower significant banks’ net income before taxes by about 10 percent, with differing sensitivities across banks.
  - Banks have limited room to further reduce funding costs: resident deposits account for about 55 percent of total funding and the average interest rate on outstanding term deposits is at a very low level.
  - Lesser use of interbank funding reduces benefits from negative interbank interest rates.
  - Continued deleveraging: deleveraging by corporates and households continues and bank lending has continued to fall.
  - To maintain net interest income from lending given compressed margins, credit would need to grow by about 5.5 percent annually, compared with IMF staff projections of annual credit growth of slightly above 1 percent over the next five years.
- Efficiency, costs, and policy implications:
  - Post-crisis restructuring reduced offices and employees significantly since 2008; cost-to-income ratio remains competitive on a consolidated operations basis.
  - Staff and administrative expenses have steadily increased from the trough in early 2013.
  - Low profitability implications:
    - Reduced organic capital generation (retained earnings underpinned most of the system’s capital buffer buildup in 2015).
    - Lower profits can hinder NPL reduction via sales or debt relief.
    - When return on equity is below the cost of capital (currently the case), raising new capital is more difficult.
  - Potential responses to compressed net interest income:
    - Boost relatively low non-interest income.
    - Further improve efficiency and reduce operating costs via greater use of technology, consolidation of smaller banks for economies of scale, and rationalization of branch networks.
- Risk monitoring:
  - Recovery of housing market and construction sector is at an early stage; deleveraging is ongoing.
  - Some signs of increasing leverage in certain segments: share of new mortgage loans with high loan-to-value (LTV) ratios and average LTV of new housing loans have both increased, though levels remain relatively low.

*Prepared by Phakawa Jeasakul (MCM); content as provided in the source PDF*

### 12.      The Spanish banking system has relatively strong cross-border linkages, particularly

### The Spanish banking system has relatively strong cross-border linkages, particularly with Latin America and the United Kingdom

### Cross-border linkages and channels of spillovers
- As of March 2016, Spanish banks’ exposures to:
  - Latin America: 12.2 percent of total exposures
  - United Kingdom: 11.4 percent of total exposures
  - Spain: 54.6 percent of total exposures
- Business model note:
  - Spain’s two large international banks have been pursuing international expansion largely based on the standalone subsidiary model.
  - Key channel of cross-border spillovers: distribution of profits from subsidiaries to parent banks, especially because Latin America and the United Kingdom are among the most important sources of earnings for these two global Spanish banks.

### Macroeconomic environment and regional vulnerabilities
- Regional macro developments:
  - Brazil: experienced an output contraction in recent years, with growth prospects starting to improve.
  - Many Latin America economies: slowed down; macroeconomic uncertainties increased markedly, with rising trade protectionism risks (e.g., for Mexico).
  - United Kingdom: outlook increasingly uncertain in the aftermath of the ‘Brexit’ referendum.
  - Problem assets have been on a rising trend in Brazil and Turkey.

### Subsidiary profitability and shock scenarios
- Latin America subsidiaries:
  - Exhibit relatively strong financial performance compared with other major banks in the region.
  - Given relatively strong profitability and relatively high provisioning, Spanish subsidiaries should be able to handle additional losses.
  - Their annual pre-impairment net income could potentially absorb about up to twice the currently still moderate NPL level.
  - Scenario: if impairment costs by subsidiaries increased by a 25 percent, this would reduce the contribution of profits to group-wide capital by about 25 percent.
- Other regions:
  - Scenario: a 50 percent increase in impairment costs by subsidiaries in Turkey and the United Kingdom would reduce the contribution of profits to group-wide capital of these two banks by less than 5 percent.

### Balance sheet adjustments and credit conditions
- Private sector deleveraging and credit:
  - New bank lending has picked up, notably consumer credit for households and lending to SMEs and certain corporate sectors (agriculture, manufacturing, non-real-estate services).
  - Total credit growth was still negative in October 2016.
  - Private debt-to-GDP is about 64 percentage points below its 2007 peak level.
  - Excess leverage now concentrated mostly in a few corporate sectors (construction and real estate) and in households.
  - Going forward, deleveraging is projected to moderate and remain concentrated in a few sectors.

### Asset quality: recent improvements and remaining legacy
- Nonperforming loans (NPLs) and trends:
  - Consolidated NPL ratio fell to 5.8 percent in June 2016 from 6.9 percent a year earlier.
  - This is 3.6 percentage points below its peak in 2013.
  - NPL ratio for banking business in Spain declined from 11.0 to 9.4 percent over the same period.
  - This is 4.2 percentage points below its peak in 2013 (almost 30 percent reduction over three years).
  - NPL ratio for overseas operations: broadly stable at around 2.3 percent.
- Stock of problem assets:
  - NPLs for banking business in Spain were €123 billion by mid-2016.
  - The amount of foreclosed real estate assets has remained broadly stable.
- Sectoral concentration:
  - Housing market: house prices are up by about 8 percent from the trough.
  - NPLs remain highest in construction and real estate:
    - NPLs amount to 28 percent of outstanding lending to the sector
    - NPLs account for 38 percent of total NPLs
- Provisioning and buffers:
  - Provisioning coverage of nonperforming exposures (with provisions and collaterals accounted) was 95 percent for the Spanish banking system as of end-June 2016, compared to the EU-wide average of 86 percent.
  - Aggregated uncovered NPEs amounted to 4.1 percent of CET1 capital for significant Spanish banks overall, compared with the EU-wide level at 9.5 percent.
  - The ECB’s 2014 Comprehensive Assessment identified only small capital adjustments for Spanish banks (the lowest among euro area members).
- Time to complete NPL clean-up:
  - If the reduction of problem assets continues at about €30 billion a year, it could take on average more than five years to completely resolve legacy assets.
  - The ECB Guidance to Banks on NPLs and time-bound, realistic and ambitious NPL reduction plans are welcomed.
- Two vulnerabilities linked to remaining NPLs:
  - Interest rate sensitivity: a 1 percentage point increase in interest rates could reduce the return on assets by about 0.07 percentage points based on the current level of NPLs and foreclosed real estate assets.
  - Stress-test exposure: in the adverse scenario of the EBA’s 2016 stress testing exercise, estimated credit losses increase for Spanish banks by 0.82 percent of risk weighted assets relative to the baseline scenario, compared with the EU-wide level of 0.58 percent of risk weighted assets.

### Capital adequacy and leverage
- Capital position:
  - Total capital ratio: 14.6 percent at end-June 2016, up from 13.7 percent at end-2014.
  - CET1 capital: Spanish banks’ holding of CET1 exceeds regulatory and supervisory minima but lags behind European peers.
  - EBA 2016 Transparency Exercise: CET1 capital ratio of significant Spanish banks on a fully-loaded basis was 1.6 percentage points below the transitional basis (lower position largely related to intangible assets).
- Stress-test and leverage observations:
  - The EBA’s 2016 Stress Testing Exercise (six largest Spanish banks) showed Spanish banks display similar resilience to adverse shocks as EU peers when measured by reduction in CET1 capital ratios, though most banks would end up with somewhat lower leverage ratios than the EU-wide average.
  - Spanish banks are generally less leveraged than European peers: the ratio of tangible equity to tangible assets of the Spanish banking system is broadly in line with the EU-wide average despite the lower level of CET1 capital.
  - Main driver: greater risk weight intensity due to more prevalent use of the standardized approach.
  - The risk weight intensity of the two Spanish global banks is higher than most global systemically important banks.
  - Based on the EBA’s 2015 Transparency Exercise, most significant Spanish banks outperformed European peers in terms of the leverage ratio (tier-1 capital to total exposures) despite somewhat lower CET1 capital ratios.
- Sources of capital build-up in 2015:
  - Net interest income boosted the capital ratio by 4.3 percentage points.
  - Other net operating income boosted the capital ratio by 2.1 percentage points.
  - Impairment costs reduced the capital ratio by 1.6 percentage points (down from 5 percentage points in 2012).
  - Net capital injection and change in risk weighted assets had minimal effect.
  - Implication: retained earnings were an important organic source of capital; sustained low profitability could challenge further capital buffer building.

*Source: IMF staff report (cr1724).*

### 5.0 percent

### 5.0 percent

### F. Capacity to Support Growth over the Medium Term

- Spanish corporates and households have continued to deleverage, resulting in declines in debt-to-GDP ratios by 49 and 18 percentage points, respectively, from the peaks.
- Banks’ lending has continued to fall as amortization has outpaced new lending.
- Weak credit demand drivers listed in the source:
  - debt overhangs and NPLs in some corporate segments (construction and real estate),
  - still depressed disposable income levels,
  - high unemployment,
  - concentrated indebtedness among low-income households,
  - general need to further build up financial wealth (traditionally well below European peers).
- Key question: Once credit demand picks up over the medium term, would banks have sufficient capital position to expand balance sheets and finance expansion?
- Banking system capacity estimates (Figure 12):
  - Previous year’s profits could support system-wide credit growth of about 6.5–8.5 percent per year if banks used their annual profits to fulfill the regulatory capital requirement, keeping the capital ratio unchanged.
  - Baseline estimation based on banks’ profits during 2015H1; profitability has declined since then.
  - Under reduced profitability assumptions (sensitivity analysis):
    - If profits drop by 25 percent over the next three years, system-wide capacity to provide additional credit would still be at 6.5 percent per year, above nominal GDP growth projected through 2021.
    - If profits halve compared to the 2015H1 level, capacity would drop to 4.4 percent annually.
  - Some banks would have less room for credit expansion than others in these scenarios.
- Footnotes and assumptions for the credit provision estimate:
  - Assumes banks use additional profits to expand balance sheets.
  - Baseline profits are based on the 2015H1 level adjusted for additional credit.
  - Assumes no additional cleanup efforts in the baseline, with no additional provisions.
  - Assumes no additional efforts to build up capital buffers.
  - Dividend payout ratio assumed at 25 percent and tax rate at 25 percent.
  - Reduced profitability scenarios assume declining profits by 25 and 50 percent over 2016-18 relative to the 2015H1 level.
  - Note: funding should not be an issue given banks could obtain funding from the ECB under the new Targeted Long-term Refinancing Operations (TLTRO-II).
  - IMF staff projects credit to grow about 1 percent annually over the next three years.

### G. Conclusions and Policy Implications (Banking)

- Overall assessment:
  - The banking system has gained further strength amid new challenges.
  - Better asset quality, stronger capital and funding positions, and reduced debt overhangs mean the system is closer to putting most crisis legacies behind it.
  - Banks have progressed at different speeds; overall NPLs and foreclosed assets remain sizeable, though much lower than in some European banking systems.
  - Banks face challenges from the low interest rate environment and new regulatory initiatives.
  - Spillovers from a less favorable macroeconomic environment in some economies where Spanish banks have a strong footprint would be limited to potentially lower profits and weaker contributions to parents’ capital buffers.
- Profitability pressures:
  - Banks’ profitability in Spain is currently well below the pre-crisis level, with the return on equity lower than the cost of capital.
  - Profitability has fallen in the past year: reduced net interest and other income offset by falling impairment costs.
  - Low interest rates and difficult operating conditions will pressure cost structures and business models.
  - Key strategies to address profitability:
    - Achieve greater efficiency (Spanish banks still rely on a larger branch network than European peers).
    - Further reduce operating expenses.
    - Raise non-interest income.
- Impaired assets and insolvency:
  - Efforts to reduce impaired assets should continue.
  - Reduction in NPLs has generally proceeded well, but at different speeds across banks; adequate provisioning and fuller use of the enhanced insolvency regime are encouraged.
  - Bank-specific time-bound, realistic and ambitious NPL reduction plans as foreseen in the ECB Guidance to Banks on NPLs are welcomed.
  - Insolvency reform supports more efficient debt restructuring and gives a “fresh start” to individuals; usage has been relatively limited so far.
  - A stock taking exercise of the insolvency framework’s functioning could identify design changes to help deleveraging.
- Capital and funding:
  - Continued efforts to strengthen banks’ capital and funding positions will enhance resilience and capacity to support growth over the medium run.
  - Encourage banks to increase high-quality capital through retained earnings.
  - Bolstering capital is prudent to safeguard financial stability and ensure adequate capital in light of new regulatory initiatives.
  - Additional capital would help ensure sufficient credit provision to financially-sound corporates and households as credit demand picks up.
  - Banks may need to adjust liability structures to fulfill new regulatory requirements, such as Minimum Requirements for Own Funds and Eligible Liabilities (MREL) and Net Stable Funding Ratio (NFSR).

### Fiscal Non-Compliance among Spanish Regions — A. Background and Framework

- Context:
  - Weak compliance among regional governments has hampered Spain’s fiscal consolidation efforts.
  - Regions have been subject to annual nominal budget balance targets over the last two decades under different rule-based fiscal frameworks.
  - Fiscal compliance was weak and varied markedly across regions; non-compliance increased following Europe’s sovereign debt crisis and accounted for the bulk of general government target deviation in Spain’s Stability Program.
- Definition and objective:
  - Fiscal non-compliance is defined as the inability to remain within the limits set under these targets (Lledó, 2015).
  - The note gauges how frequently regions missed the target, by how much, and what factors contributed, based on a conceptual framework developed in Delgado-Téllez and others (forthcoming).

### Fiscal Non-Compliance — B. What Explains Involuntary Non-Compliance by the Regions?

- Key findings:
  - Common shocks and adjustment needs contributed to involuntary fiscal non-compliance among regions.
  - Non-compliance frequencies moved in tandem with nominal GDP growth forecast errors.
  - Compliance margins also moved with forecast errors; positive and statistically significant estimates under various models support this.
  - Idiosyncratic shocks did not show a systematic impact on non-compliance.
  - Fiscal non-compliance displayed inertial patterns in the form of positive auto-correlated margins, possibly reflecting budget rigidities from incremental budgeting or multi-year expenditure commitments.
  - Fiscal non-compliance increased with the required adjustment effort—proxied by differences between the fiscal deficit target in year t and t-1.
  - Adjustment efforts have been heterogeneous across regions because fiscal deficit targets have been set uniformly across regions in most years despite different starting positions.
  - Adjustment efforts are found to have a negative and statistically significant impact on fiscal compliance margins in most specifications.

### Fiscal Non-Compliance — Box 1: Framework for Voluntary vs Involuntary Non-Compliance

- Definitions:
  - Voluntary fiscal non-compliance: regional government (RG) has capacity but not incentives to comply.
  - Involuntary fiscal non-compliance: RG is willing and ex-ante capable to comply, but lacks capacity ex-post because the target becomes unfeasible after shocks.
- Determinants of voluntary non-compliance:
  - Bailout incentives depend on factors increasing central government’s non-bailout costs:
    - regions’ economic size (“too big or too small to fail”),
    - political influence (“too connected to fail”),
    - lack of fiscal autonomy (“too dependent to fail”).
  - Overspending incentives can rise from weak fiscal rules and insufficient market discipline.
- Involuntary non-compliance more likely in fiscal stress periods with large negative fiscal shocks and ambitious center-imposed targets that become unfeasible.
- Empirical approach:
  - Analysis covers Spain’s autonomous regions during 2002–15 (Basque Country excluded).
  - Uses dynamic panel regression with Arellano-Bond first-difference GMM to allow for inertia and endogeneity.

### Fiscal Non-Compliance — C. What Factors Have Impacted Voluntary Non-Compliance?

- Regions’ size, fiscal autonomy, and politics influence voluntary non-compliance:
  - Impact of region size is not clear-cut: non-compliance frequencies tend to be higher among larger regions, while margins are inconclusive.
  - Fiscal non-compliance frequencies tend to be larger among regions with more limited autonomy to raise own revenues.
  - Non-compliance frequencies increase and compliance margins decrease among regions with less autonomy to cut spending and larger vertical fiscal imbalances (VFI).
    - Vertical fiscal imbalances defined as [1−Own Revenue/Own Spending]; Own revenue (spending) corresponds to region’s total revenue (spending) minus transfers.
  - Non-compliance frequencies are not necessarily larger among regions in top expenditure and VFI quartiles.
  - Political factors:
    - Non-compliance frequencies seem to increase during national and regional election years,
    - Increase among regions politically aligned with the center,
    - Increase in regions with strong pro-autonomy preferences,
    - Increase with large parliamentary representations.
  - Apart from election years, political alignment, pro-autonomy inclinations, and parliamentary representation did not systematically affect compliance margins in a statistically significant way.

*International Monetary Fund — Spain: Selected findings, projections, and policy implications as presented in the source content.*

### 4.      Stronger fiscal rules have not necessarily helped improve fiscal compliance, while the

### cr1724 - 4.      Stronger fiscal rules have not necessarily helped improve fiscal compliance, while the

### Main findings on fiscal rules and compliance
- Subnational fiscal rules in Spain were significantly strengthened after the Euro Area sovereign debt crisis with the adoption of formal monitoring and enforcement procedures (Figure 6).
- Such procedures have not been fully implemented, undermining fiscal compliance (Lledó, 2015).
- Regression results indicate:
  - Stronger fiscal rules do not show any direct impact on fiscal compliance margins.
  - Stronger fiscal rules appear to have an indirect impact on compliance margins by helping reduce inertial patterns (Table 1, models 8 and 9).
  - In Table 1, model 8: Fiscal Rule Index X Lagged compliance margin = -0.06**.

### Impact of financial markets on fiscal compliance
- Financial markets affect fiscal non-compliance through two channels:
  - Regions more reliant on market-issued securities than on bank loans exhibit less fiscal non-compliance, as regions internalized the impact of fiscal non-compliance on credit ratings and market-financing costs (Figure 7).
  - Regions with lower (poorer) credit ratings and facing higher implicit interest rates experienced larger frequencies of fiscal non-compliance, suggesting that financial markets may have weakened fiscal compliance by raising financing costs for less creditworthy regions.
- Note on measures:
  - Region ratings are based on a numerical scale from 0 to 21 derived from Fitch, Standard and Poor, and Moody region ratings (the lower the ratings, the poorer the creditworthiness).
  - Security to loans ratio defined as ratio of regions' public debt in government securities to banking loans.
  - Implicit interest rates estimated as region's interest payments in percent of end-of-year region public debt stock.

### Regression evidence (highlights from Table 1: First-Difference GMM Estimates of Fiscal Compliance Margins)
- Dependent variable: difference between regions’ fiscal deficit outturns and fiscal deficit targets (larger difference = larger fiscal compliance margin).
- Selected coefficient estimates and significance (exact values as reported in Table 1):
  - Lagged compliance margin:
    - Model (1): 0.74*
    - Model (2): 1.09**
    - Model (8): 0.91**
    - Model (9): 2.42***
    - Model (10): 0.31*
  - Growth forecast errors:
    - Model (1): 0.09*
    - Model (6): 0.12***
    - Model (12): 0.12***
  - Region-National growth differential:
    - Model (1): -2.09*
    - Model (2): -2.67*
    - Model (8): -2.32*
    - Model (11): -2.17*
  - Region-National inflation differential:
    - Model (7): -2.41**
    - Model (10): -2.83*
  - Fiscal Target Adjustment:
    - Model (1): -0.80**
    - Model (2): -1.13**
    - Model (8): -0.82***
    - Model (10): -0.51***
  - Social spending share in regional government spending:
    - Model (6): -0.33**
  - Investment share in total regional spending:
    - Model (6): -0.24***
  - Vertical fiscal imbalances:
    - Model (6): -0.15***
    - Model (7): -0.13**
  - Fiscal Rule Index:
    - Model (7): 0.05
    - Model (8): 0.02
    - Model (9): 0.01
  - Fiscal Rule Index X Lagged compliance margin:
    - Model (8): -0.06**
  - Region Ratings:
    - Model (10): (not directly listed as a coefficient in the excerpted columns; see related rows)
  - Lagged Annual Change in Implicit interest rates:
    - Model (10): 1.03***
  - Ratio of security to loans:
    - Model (10): 0.00
  - National election dummy:
    - Model (10): -0.60***
  - Regional election dummy:
    - Model (3): -0.36*
    - Model (10): -0.50*
  - Party congruence dummy:
    - Model (11): -0.41***
  - Number of observations across models: values include 176, 160, 145, 144 (see Table 1 for model-specific counts).
  - Number of regions across models: mostly 16, with model (10) using 15 regions.
- Regression diagnostics (selected p-values as reported):
  - Hansen test p-values across models: 0.60, 0.80, 0.44, 0.87, 0.66, 0.28, 0.37, 0.82, 0.65, 0.14, 0.68, 0.15, 0.05 (instrument sets are valid where p-value ≥ 0.05).
  - m1 p-values (serial correlation order 1): 0.12, 0.11, 0.13, 0.15, 0.23, 0.02, 0.09, 0.10, 0.06, 0.14, 0.12, 0.20, 0.00.
  - m2 p-values (serial correlation order 2): 0.39, 0.61, 0.43, 0.49, 0.42, 0.48, 0.61, 0.66, 0.77, 0.16, 0.28, 0.22, 0.67 (observed p-values higher than 0.05 under the m2 test for all models).

### Interpretation and caveats
- The evidence suggests an ambiguous role for financial markets:
  - They can discipline compliance where reliance on market-issued securities makes regions internalize market consequences.
  - They can weaken compliance for regions with poor creditworthiness by raising financing costs.
- Regression analysis corroborates a disciplinary effect of financial markets by showing fiscal non-compliance to decline among regions facing larger financing costs (Table 1, model 10).
- A caveat noted: one cannot rule out reverse causality with fiscal non-compliance leading to poorer credit ratings.

### Policy issues for discussion (listed in source)
- Improve macro-fiscal forecasting at the central and subnational level to minimize unanticipated common shocks.
- Assess whether and how reforms on regions financing system and spending mandates, including minimum spending standards, may increase fiscal autonomy and reduce vertical fiscal imbalances.
- Adopt differentiated regional fiscal targets to improve the feasibility of fiscal adjustment plans, without undermining incentives to implement such plans.
- Implement rules-based automatic enforcement mechanisms to ensure fiscal compliance during election years and, especially, in good times.

*Source: cr1724 - 4.      Stronger fiscal rules have not necessarily helped improve fiscal compliance, while the (IMF country report content).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2017/cr1724.pdf_