## wp18283

## Source details

**Canonical URL:** [wp18283](https://www.imf.org/-/media/files/publications/wp/2018/wp18283.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18283.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18283.pdf.json)

---

### I. Introduction — summary findings
- Exports of goods and services increased from 26 percent of GDP in 2007 to 34 percent of GDP in 2017.
- Spain’s current account improved from a deficit of 10 percent of GDP in 2007 to an average surplus of around 1.5 percent of GDP in 2013–17.
- Spain’s net international investment position was -84 percent of GDP at end-2017.
- Regression analysis indicates external demand substantially boosted exports; the contribution of competitiveness to export growth turned positive after the crisis.
- Counterfactual exercises based on OECD (2014) and Doménech et al. (forthcoming) suggest the 2010 and 2012 labor market reforms significantly contributed to competitiveness gains.
- The paper estimates that the 2010 and 2012 labor market reforms account for nearly one-tenth to above one-quarter of the real export growth rate from 2010 to 2013 for Spain.
- Panel analysis implies that relaxing employment protection from “highly stringent” to “slightly stringent” in advanced countries is associated with an improvement in the real export growth rate in a range of approximately 2 to 7 percentage points.

### II. Recent evolution of Spanish exports — patterns and sectoral detail
- Real export growth was almost 5 percent since 2013.
- Average annual growth rates:
  - Goods exports: 4.5 percent in 2001–07; 5.4 percent in 2010–17.
  - Services: 2.9 percent in 2001–07; 4.3 percent in 2010–17.
- Product-group growth (average annual rates): Consumption, Intermediate, and Capital goods all exhibited higher growth in 2010–17 than in 2001–07; capital goods’ annual growth rate roughly doubled.
- Sectoral shifts (2007–2017):
  - Overperformers: foods, beverages, and tobacco; chemicals; non-durable consumption goods (textiles).
  - Shares declined for durable consumption goods, vehicles, and (to a lesser extent) capital goods.
- Automobiles consolidated as Spain’s top export product; Spain became the second-biggest car exporter in continental Europe despite a reduction in vehicle export share.
- Export quality metrics:
  - Henn et al. (2017) measure: Spain’s export quality declined between 2007 and 2014, particularly across manufactures (index referenced where "World frontier" = 1).
  - European Commission (2018a): Spain ranked low compared to euro-area peers over 2005–2016 on another export-quality measure.
  - Economic Complexity Index corroborates limited improvement in export quality; technological content remained stable and concentrated in medium-high technology goods (Grinberg, 2017).
- Global value chain (GVC) participation: Spain’s participation is relatively low; exports tend to be close to final consumption rather than intermediate stages (Prades and Villanueva, 2017).
- Services:
  - Spain accounts for less than 3 percent of the world’s total services exports; that share started to increase since 2016 after prior declines.
  - Exports of non-travel services increased from 4 percent of GDP in 2007 to 5.4 percent of GDP in 2017.
  - Tourism receipts’ average annual growth accelerated in 2010–17 compared to 2001–07; Spain ranked first in the two most recent WEF Travel and Tourism Competitiveness Reports referenced.
- Export trends and geographic diversification:
  - Spain's exports to the EU accounted for "52.7 percent in 2017."
  - Over 2010–2017, the annual real growth rate of goods exports to the EU was "3.9 percent", compared to "8.6 percent" for exports outside the EU.
  - The number of exporters active for at least four consecutive years rose by "nearly one-third" between 2007 and 2017.
  - Almunia et al. (2018) document that, on average, Spanish manufacturing firms replaced around "one-third" of their lost domestic sales with sales in foreign markets over 2009–13.
  - Negative slopes in partner-growth versus export-share plots indicate exports were concentrated in trading partners with relatively low growth rates of real import demand; that slope becomes less negative when using Spain’s export shares as of 2017 than those of 2006.
  - Five largest export recipients: France, Germany, Italy, the United Kingdom, and Portugal.

### III. Assessing the drivers of export growth — regression results and decomposition
- Empirical approach: Standard export equation linking growth of total export volumes of goods and services to ΔLn(Foreign Demand) and ΔLn(ULC-Based REER) with quarterly data 1995:Q1–2017:Q4; includes dummy for 2008:Q4 outlier; variables in log differences.
- Table 1 estimates (Sample: 1995Q1-2017Q4; Dependent variable: ΔLn(Real Exportst)):
  - ΔLn(Foreign Demandt): 2.560*** (0.320) ; 2.870*** (0.303)
  - ΔLn(ULC-Based REER t-1): -0.302** (0.145) ; -0.248* (0.138)
  - ΔLn(Private Consumption t-1): -0.363* (0.189) (column 2)
  - Observations: 90
  - Adj. R2: 0.567 (column 1); 0.577 (column 2)
  - Note: Statistical significance denoted by ***, **, * for 1%, 5%, and 10%; robust standard errors in parentheses; constant and 2008Q4 dummy coefficient omitted from table.
- Decomposition results:
  - Foreign demand explains the bulk of export growth in most periods.
  - Contribution of cost competitiveness (ULC-based REER) switched from negative before the crisis to positive afterwards, reflecting a generally declining ULC-based REER since mid-2008.
  - The adjustment in Spain’s ULC reflects wage moderation since 2010, declining employment until 2013, and strong recovery in real GDP more recently.
  - Improved competitiveness accounted for about 15 percent of the average annual growth rate of real exports over 2010–13 (based on the simple decomposition framework in the paper).

### IV. Labor market reforms, counterfactuals, and quantified impacts
- Labor market reforms in 2010 and 2012 aimed to facilitate wage adjustment in high-unemployment context.
- OECD (2014) estimates: the 2012 reforms and the 2012–14 Agreement induced a drop in the growth of Spain’s business-sector ULC of between 1.2 percentage points and 1.9 percentage points from 2011:Q4 to 2013:Q2 (average quarterly drop between 0.2 percentage points to above 0.3 percentage points).
- Counterfactual experiments using the regression decomposition and OECD estimates:
  - If OECD’s estimated average quarterly effects of the 2012 reforms on ULC remained until 2017, the contribution of competitiveness gains to export growth over 2012–17 would have been reduced by around one-half relative to the baseline.
  - If OECD’s estimated effects lasted only until 2013, the contribution of competitiveness to export growth over 2012–13 would have been reduced by at least one-quarter compared to baseline.
- Box 1 (Doménech et al. forthcoming Structural VAR counterfactual):
  - Shutting down wage shocks for 2008–11 yields an alternative ULC path: Spain’s ULC would have declined relative to actual series by more than 2 percentage points on impact, and by around 4 percentage points per quarter over the medium term.
  - Under that alternative ULC path, the contribution of competitiveness (ULC-based REER) to average annual export growth over 2008–2011 would have been two to nearly four times as large as in the baseline scenario (range reflects confidence bands).
- Spain-specific quantified impacts:
  - For Spain, the 2010 and 2012 labor market reforms are estimated to be associated with faster growth of real exports over 2010–13 in the order of "around 1 to 4 percentage points".
  - This implies the reforms accounted for "nearly one-tenth to above one-quarter" of Spain’s total export growth rate from 2010 to 2013.
- Counterfactual ULC scenarios:
  - A counterfactual scenario for unit labor costs considers greater wage flexibilization starting in 2008 rather than 2012; deviations from actual unit labor costs are reported in percentage points across quarters since 2008Q1 with baseline and lower/central/upper bound wage-flexibility scenarios.
- Mechanisms:
  - Cost-competitiveness channel is central: including contemporaneous log change of the ULC-based REER as an additional control largely reduces the significance of labor market flexibility coefficients.
  - Productivity channel likely less important for near-term export gains because productivity gains typically take time to materialize.
  - Product market reforms can also matter; firm-level evidence indicates better service regulation positively affects export volumes of large firms through an input cost channel.

### V. Cross-country panel evidence on labor market flexibility and exports
- Sample and variables:
  - Annual data for a sample of 26 advanced countries (sample period varies depending on indicator availability; in most cases it covers 1985-2016).
  - Key regressor: labor market flexibility indicator (four OECD employment protection indicators and a WEF “hiring and firing practices” indicator).
  - Baseline controls include log change of trading partners’ real import demand, lagged log change of the ULC-based REER, lagged economic growth, lagged downturn indicator, lag of labor market reform indicator, country and year fixed effects.
  - Endogeneity caveat: estimated coefficients (훽 and 훿) may not be interpreted as causal effects.
- Regression findings (summary of Table 2):
  - Less stringent employment protection is associated with faster export growth.
  - Effects mostly statistically significant for regulations governing individual and/or collective dismissals of workers with permanent contracts.
  - Regulations on temporary contracts are not associated with export growth in a significant way.
  - Positive effects of less stringent employment protection on exports strengthen with low economic growth (statistically significant interaction terms with LowGrowth).
- Selected coefficient estimates (preserved exactly as reported):
  - OECD Employment protection: individual and collective dismissals (permanent contracts): "-0.062**", "-0.072**".
  - OECD Employment protection: individual and collective dismissals (permanent contracts) * LowGrowth: "-0.007**", "-0.007**".
  - OECD Employment protection: individual dismissals (permanent contracts): "-0.049*", "-0.061**".
  - OECD Employment protection: individual dismissals (permanent contracts) * LowGrowth: "-0.007**", "-0.006*".
  - OECD Employment protection: average notice for no-fault individual dismissal (permanent contracts): "-0.010", "-0.017**".
  - OECD Employment protection: average notice for no-fault individual dismissal (permanent contracts) * LowGrowth: "-0.007***", "-0.007**".
  - OECD Employment protection: temporary employment: "-0.007", "-0.013".
  - OECD Employment protection: temporary employment * LowGrowth: "-0.005", "-0.004".
  - WEF's Global Competitiveness Index: Hiring and firing practices: "0.033**", "0.028".
  - WEF's Global Competitiveness Index: Hiring and firing practices * LowGrowth: "-0.002", "-0.002".
  - Lagged ULC-Based REER (log change): "-0.119**", "-0.128***", "-0.133***", "-0.126***", "-0.070", "-0.088", "-0.117*", "-0.122**", "-0.115*", "-0.023".
  - Trading partners' real import demand (log change): "0.843***", "0.714***", "0.691***", "0.688***", "0.988***", "0.708***", "0.708***", "0.676***", "0.690***", "0.990***".
- Observations and fit (as reported across columns):
  - Observations: "448", "496", "496", "496", "225", "448", "496", "496", "496", "225".
  - Adj. R-squared: "0.599", "0.593", "0.595", "0.586", "0.634", "0.529", "0.523", "0.524", "0.523", "0.542".
- Indicator notes:
  - OECD indicators vary from 0 (least restrictive) to 6 (most restrictive).
  - WEF indicator varies from 0 (worst) to 7 (best).
  - OECD's "average notice" indicator is the average of REG3A, REG3B, and REG3C indicators from the OECD EPL database.
  - LowGrowth is a dummy variable that takes the value of 1 in years of below-average real GDP growth, and 0 otherwise.
  - Statistical significance denoted by "***" for 1%, "**" for 5%, and "*" for 10%.

### VI. Quantitative amplification, robustness, and mechanisms
- Amplification effects:
  - Relaxing hiring and firing regulations (from 75th percentile of OECD indicators / 25th percentile of WEF to 25th percentile of OECD / 75th percentile of WEF) are between "nearly 2 percentage points to around 7 percentage points".
- Robustness experiments:
  - Considered dropping lagged log change of the ULC-based REER, using NEER controls, adding lagged interaction terms, controlling for product market reform indicators.
  - Main baseline results largely unchanged; in some cases magnitudes of ββ coefficients declined somewhat (in absolute value).
  - When all labor market flexibility indicators were lagged one period, coefficients on key regressors became statistically insignificant.
  - Alternative regressions including Duval et al.’s (2018) labor market regulation indicators produced expected sign but statistically insignificant coefficients for their employment protection indicator.
- Mechanisms:
  - Cost competitiveness channel is primary: including contemporaneous ULC-based REER changes diminishes labor flexibility coefficients’ significance.
  - Productivity channel likely slower to affect exports.
  - Product market reform and service regulation can affect export volumes through input-cost and competition channels.

### VII. Policy conclusions and recommendations
- Spain’s improved export performance was broad-based and driven by multiple factors: diversified destinations, greater export orientation of firms, depressed domestic demand encouraging exports, and regained cost competitiveness aided by labor market reforms.
- Outlook: positive, with room for further improvement underpinned by policy action.
- Recommended policy priorities (as implied by the analysis and literature cited):
  - Further upgrades in productivity growth and functioning of labor markets, building on past reforms.
  - Enhance competition in product markets to support export performance.
  - Strengthen human capital and institutions to improve export quality (including government efficiency and regulatory quality).
  - Upgrade worker mobility across industries, regions, and skills via enhanced formal education, improved opportunities for vocational training and life-long learning, and targeted housing assistance.

*Source: IMF staff compilation from the content of wp18283.pdf (References section and accompanying paper excerpts).*

### References ________________________________________________________________18

### References

### I. Introduction — summary findings
- Exports of goods and services increased from 26 percent of GDP in 2007 to 34 percent of GDP in 2017.
- Spain’s current account improved from a deficit of 10 percent of GDP in 2007 to an average surplus of around 1.5 percent of GDP in 2013–17.
- Spain’s net international investment position was -84 percent of GDP at end-2017.
- Regression analysis indicates external demand substantially boosted exports; the contribution of competitiveness to export growth turned positive after the crisis.
- Counterfactual exercises based on OECD (2014) and Doménech et al. (forthcoming) suggest the 2010 and 2012 labor market reforms significantly contributed to competitiveness gains.
- The paper estimates that the 2010 and 2012 labor market reforms account for nearly one-tenth to above one-quarter of the real export growth rate from 2010 to 2013 for Spain.
- Panel analysis implies that relaxing employment protection from “highly stringent” to “slightly stringent” in advanced countries is associated with an improvement in the real export growth rate in a range of approximately 2 to 7 percentage points.

### II. Recent evolution of Spanish exports — patterns and sectoral detail
- Real export growth was almost 5 percent since 2013 (real export growth remained healthy and remarkably stable, at almost 5 percent since 2013).
- Average annual growth rates:
  - Goods exports: increased from 4.5 percent in 2001–07 to 5.4 percent in 2010–17.
  - Services: increased from 2.9 percent in 2001–07 to 4.3 percent in 2010–17.
- Product-group growth (average annual rates): Consumption, Intermediate, and Capital goods all exhibited higher growth in 2010–17 than in 2001–07; capital goods’ annual growth rate roughly doubled.
- Sectoral shifts (2007–2017): notable overperformers included foods, beverages, and tobacco; chemicals; and non-durable consumption goods (textiles). Shares declined for durable consumption goods, vehicles, and (to a lesser extent) capital goods.
- Automobiles consolidated as Spain’s top export product; Spain became the second-biggest car exporter in continental Europe despite a reduction in vehicle export share.
- Export quality metrics:
  - Henn et al. (2017) measure: Spain’s export quality declined between 2007 and 2014, particularly across manufactures (index referenced where "World frontier" = 1).
  - European Commission (2018a): Spain ranked low compared to euro-area peers over 2005–2016 on another export-quality measure.
  - Economic Complexity Index corroborates limited improvement in export quality; technological content remained stable and concentrated in medium-high technology goods (Grinberg, 2017).
- Global value chain (GVC) participation: Spain’s participation is relatively low; exports tend to be close to final consumption rather than intermediate stages (Prades and Villanueva, 2017).
- Services:
  - Spain accounts for less than 3 percent of the world’s total services exports; that share started to increase since 2016 after prior declines.
  - Exports of non-travel services increased from 4 percent of GDP in 2007 to 5.4 percent of GDP in 2017.
  - Tourism receipts’ average annual growth accelerated in 2010–17 compared to 2001–07; Spain ranked first in the two most recent WEF Travel and Tourism Competitiveness Reports referenced.

### III. Assessing the drivers of export growth — regression results and decomposition
- Empirical approach: Standard export equation linking growth of total export volumes of goods and services to ΔLn(Foreign Demand) and ΔLn(ULC-Based REER) with quarterly data 1995:Q1–2017:Q4; includes dummy for 2008:Q4 outlier; variables in log differences.
- Table 1 estimates (Sample: 1995Q1-2017Q4; Dependent variable: ΔLn(Real Exportst)):
  - Coefficient on ΔLn(Foreign Demandt): 2.560*** (0.320) in column (1); 2.870*** (0.303) in column (2).
  - Coefficient on ΔLn(ULC-Based REER t-1): -0.302** (0.145) in column (1); -0.248* (0.138) in column (2).
  - Coefficient on ΔLn(Private Consumption t-1): -0.363* (0.189) in column (2) (only column (2) includes this).
  - Observations: 90; Adj. R2: 0.567 (column 1) and 0.577 (column 2).
  - Note: Statistical significance denoted by ***, **, * for 1%, 5%, and 10%; robust standard errors in parentheses; constant and 2008Q4 dummy coefficient omitted from table.
- Decomposition results:
  - Foreign demand explains the bulk of export growth in most periods.
  - Contribution of cost competitiveness (ULC-based REER) switched from negative before the crisis to positive afterwards, reflecting a generally declining ULC-based REER since mid-2008.
  - The adjustment in Spain’s ULC reflects wage moderation since 2010, declining employment until 2013, and strong recovery in real GDP more recently.
  - Improved competitiveness accounted for about 15 percent of the average annual growth rate of real exports over 2010–13 (based on the simple decomposition framework in the paper).

### IV. Labor market reforms, counterfactuals, and quantified impacts
- Labor market reforms in 2010 and 2012 aimed to facilitate wage adjustment in high-unemployment context.
- OECD (2014) estimates: the 2012 reforms and the 2012–14 Agreement induced a drop in the growth of Spain’s business-sector ULC of between 1.2 percentage points and 1.9 percentage points from 2011:Q4 to 2013:Q2 (average quarterly drop between 0.2 percentage points to above 0.3 percentage points).
- Counterfactual experiments using the regression decomposition and OECD estimates:
  - If OECD’s estimated average quarterly effects of the 2012 reforms on ULC remained until 2017, the contribution of competitiveness gains to export growth over 2012–17 would have been reduced by around one-half relative to the baseline.
  - If OECD’s estimated effects lasted only until 2013, the contribution of competitiveness to export growth over 2012–13 would have been reduced by at least one-quarter compared to baseline.
- Box 1 (Doménech et al. forthcoming Structural VAR counterfactual):
  - Shutting down wage shocks for 2008–11 yields an alternative ULC path: Spain’s ULC would have declined relative to actual series by more than 2 percentage points on impact, and by around 4 percentage points per quarter over the medium term.
  - Under that alternative ULC path, the contribution of competitiveness (ULC-based REER) to average annual export growth over 2008–2011 would have been two to nearly four times as large as in the baseline scenario (range reflects confidence bands).
- Cross-country panel regressions on employment protection:
  - Relaxing employment protection legislations in advanced countries from “highly stringent” to “slightly stringent” is associated with an improvement in the real export growth rate in a range of approximately 2 to 7 percentage points (panel estimate reported in summary).

### V. Key statistics and quantitative points (preserved exactly)
- Exports of goods and services: 26 percent of GDP in 2007; 34 percent of GDP in 2017.
- Current account: deficit of 10 percent of GDP in 2007; average surplus of around 1.5 percent of GDP in 2013–17.
- Net international investment position: -84 percent of GDP at end-2017.
- Real export growth: almost 5 percent since 2013.
- Average annual growth rates:
  - Goods exports: 4.5 percent in 2001–07; 5.4 percent in 2010–17.
  - Services: 2.9 percent in 2001–07; 4.3 percent in 2010–17.
- OECD (2014) estimated ULC growth effects from 2012 reforms: between 1.2 percentage points and 1.9 percentage points (2011:Q4 to 2013:Q2); average quarterly drop between 0.2 percentage points to above 0.3 percentage points.
- Panel estimate: relaxing employment protection associated with export growth improvement of approximately 2 to 7 percentage points.
- Regression table (Table 1) exact coefficients and standard errors:
  - ΔLn(Foreign Demandt): 2.560*** (0.320) ; 2.870*** (0.303)
  - ΔLn(ULC-Based REER t-1): -0.302** (0.145) ; -0.248* (0.138)
  - ΔLn(Private Consumption t-1): -0.363* (0.189) (column 2)
  - Observations: 90; Adj. R2: 0.567 (column 1), 0.577 (column 2).

*Source: IMF staff compilation from the content of wp18283.pdf (References section and accompanying paper excerpts).*

### 52.7 percent in 2017. Over 2010–2017, the annual real growth rate of goods exports to the

### wp18283 - 52.7 percent in 2017. Over 2010–2017, the annual real growth rate of goods exports to the

### Export trends and geographic diversification
- Spain's exports to the EU accounted for "52.7 percent in 2017."
- Over 2010–2017, the annual real growth rate of goods exports to the EU was "3.9 percent", compared to "8.6 percent" for exports outside the EU.
- These trends have allowed Spain to narrow its geographic diversification gap with France, as reflected by a market concentration index, though it still lags behind Italy and Germany.
- The negative slopes in partner-growth versus export-share plots indicate exports were concentrated in trading partners with relatively low growth rates of real import demand; that slope becomes less negative when using Spain’s export shares as of 2017 than those of 2006.
- The five largest export recipients (France, Germany, Italy, the United Kingdom, and Portugal) are highlighted as major destinations.

### Drivers of post-crisis export growth
- External demand was a key driver; the impulse from external demand benefited from geographic diversification toward faster-growing countries (emerging and developing markets in Asia, Latin America, North Africa, and the Middle East).
- The slump in domestic demand and greater export orientation of firms contributed to export growth:
  - Almunia et al. (2018) document that, on average, Spanish manufacturing firms replaced around "one-third" of their lost domestic sales with sales in foreign markets over 2009–13.
  - The number of exporters active for at least four consecutive years rose by "nearly one-third" between 2007 and 2017.
- Contribution of cost competitiveness turned positive after the crisis, helped by labor market reforms.
- Export growth was broad-based across sectors, but the export boom has not been accompanied by improvements in export quality, and Spain’s participation in global value chains remains below the international average.

### Cross-country panel evidence on labor market flexibility and exports
- The analysis uses annual data for a sample of "26 advanced countries" (sample period varies depending on indicator availability; in most cases it covers 1985-2016).
- Key independent variable: labor market flexibility measured by four OECD employment protection indicators and a WEF Global Competitiveness Report survey-based indicator of “hiring and firing practices.”
- Baseline estimating equation includes:
  - Dependent variable: log change of total real exports or real exports of goods only;
  - Key regressor: labor market flexibility indicator (LLL);
  - Interaction term: LLL * LowGrowth dummy (LowGrowth = 1 if real GDP growth falls one-standard-deviation below in-sample average);
  - Controls: log change of trading partners’ real import demand, lagged log change of the ULC-based REER, lagged economic growth, lagged downturn indicator, lag of labor market reform indicator, country and year fixed effects.
- Endogeneity caveat: estimated coefficients (훽 and 훿) may not be interpreted as causal effects.

### Regression findings (summary of Table 2)
- Less stringent employment protection is associated with faster export growth.
- OECD employment protection indicators (individual and collective dismissals for permanent contracts) have negative coefficients; WEF hiring and firing practices indicator has positive coefficients.
- Main regularities:
  - (i) Effects mostly statistically significant for regulations governing individual and/or collective dismissals of workers with permanent contracts.
  - (ii) Regulations on temporary contracts are not associated with export growth in a significant way.
  - (iii) Positive effects of less stringent employment protection on exports strengthen with low economic growth (statistically significant interaction terms with LowGrowth).
- Selected coefficient estimates from Table 2 (preserved exactly as reported):
  - OECD Employment protection: individual and collective dismissals (permanent contracts): "-0.062**", "-0.072**" (columns shown with standard errors in parentheses).
  - OECD Employment protection: individual and collective dismissals (permanent contracts) * LowGrowth: "-0.007**", "-0.007**" (standard errors in parentheses).
  - OECD Employment protection: individual dismissals (permanent contracts): "-0.049*", "-0.061**".
  - OECD Employment protection: individual dismissals (permanent contracts) * LowGrowth: "-0.007**", "-0.006*".
  - OECD Employment protection: average notice for no-fault individual dismissal (permanent contracts): "-0.010", "-0.017**".
  - OECD Employment protection: average notice for no-fault individual dismissal (permanent contracts) * LowGrowth: "-0.007***", "-0.007**".
  - OECD Employment protection: temporary employment: "-0.007", "-0.013".
  - OECD Employment protection: temporary employment * LowGrowth: "-0.005", "-0.004".
  - WEF's Global Competitiveness Index: Hiring and firing practices: "0.033**", "0.028".
  - WEF's Global Competitiveness Index: Hiring and firing practices * LowGrowth: "-0.002", "-0.002".
  - Lagged ULC-Based REER (log change): "-0.119**", "-0.128***", "-0.133***", "-0.126***", "-0.070", "-0.088", "-0.117*", "-0.122**", "-0.115*", "-0.023" (with respective standard errors shown in Table 2).
  - Trading partners' real import demand (log change): "0.843***", "0.714***", "0.691***", "0.688***", "0.988***", "0.708***", "0.708***", "0.676***", "0.690***", "0.990***".
- Observations and fit (as reported):
  - Observations: "448", "496", "496", "496", "225", "448", "496", "496", "496", "225" (across columns).
  - Adj. R-squared: "0.599", "0.593", "0.595", "0.586", "0.634", "0.529", "0.523", "0.524", "0.523", "0.542".
- Notes on indicators:
  - OECD indicators vary from "0 (least restrictive) to 6 (most restrictive)"; WEF indicator varies from "0 (worst) to 7 (best)".
  - OECD's "average notice" indicator is the average of REG3A, REG3B, and REG3C indicators from the OECD EPL database.
  - LowGrowth is a dummy variable that takes the value of "1" in years of below-average real GDP growth, and "0" otherwise.
  - Statistical significance denoted by "***" for 1%, "**" for 5%, and "*" for 10%.

### Quantitative amplification and Spain-specific counterfactuals
- Amplification effects of relaxing hiring and firing regulations (from 75th percentile of OECD indicators / 25th percentile of WEF to 25th percentile of OECD / 75th percentile of WEF) are between "nearly 2 percentage points to around 7 percentage points".
- For Spain, the 2010 and 2012 labor market reforms are estimated to be associated with faster growth of real exports over 2010–13 in the order of "around 1 to 4 percentage points".
  - This implies the reforms accounted for "nearly one-tenth to above one-quarter" of Spain’s total export growth rate from 2010 to 2013.
- Counterfactual analysis and ULC scenarios:
  - A counterfactual scenario for unit labor costs considers greater wage flexibilization starting in 2008 rather than 2012; deviations from actual unit labor costs are reported in percentage points across quarters since "2008Q1" with baseline and lower/central/upper bound wage-flexibility scenarios (figures and confidence intervals described in the source).
  - Incorporating contemporaneous log change of the ULC-based REER as an additional control largely reduces the significance of labor market flexibility coefficients, suggesting a cost-competitiveness channel.

### Mechanisms, robustness, and complementary reforms
- Cost competitiveness is a key mechanism: including contemporaneous ULC-based REER changes diminishes the labor flexibility coefficients’ significance.
- Productivity channel likely less important for near-term export gains because productivity gains typically take time to materialize.
- Product market reforms can also matter: regressions augmented with OECD product market regulation indicators reveal in some cases statistically significant roles for PMR, suggesting enhanced competition in product markets could strengthen export performance.
- Firm-level evidence: Correa-López and Doménech (2017) find better service regulation positively affects export volumes of large firms through an input cost channel.

### Policy conclusions and recommendations
- Spain’s improved export performance was broad-based and driven by multiple factors: diversified destinations, greater export orientation of firms, depressed domestic demand encouraging exports, and regained cost competitiveness aided by labor market reforms.
- Outlook: positive, with room for further improvement underpinned by policy action.
- Recommended policy priorities (as implied by the analysis and literature cited):
  - Further upgrades in productivity growth and functioning of labor markets, building on past reforms.
  - Enhance competition in product markets to support export performance.
  - Strengthen human capital and institutions to improve export quality (including government efficiency and regulatory quality).
  - Upgrade worker mobility across industries, regions, and skills via enhanced formal education, improved opportunities for vocational training and life-long learning, and targeted housing assistance.

*Source: IMF staff calculations and analysis, wp18283 (excerpt provided).*

### REFERENCES

### REFERENCES

### Key bibliographic entries
- Almunia, M., P. Antràs, D. Lopez-Rodriguez, and E. Morales, 2018, “Venting Out: Exports During a Domestic Slump,” Mimeo.
- Alvarez-López, M.E. and R. Myro, 2018, “Diversificación de Mercados y Crecimiento de la Exportación Española,” Mimeo.
- Andrés, J., O. Arce, and C. Thomas, 2017, “Structural Reforms in a Debt Overhang,” Journal of Monetary Economics, 88: 15–34.
- Bank of Spain, 2016, “Competitive Adjustment and Recovery in the Spanish Economy,” Annual Report 2015.
- Bank of Spain, 2017, “Current Account Adjustment,” Annual Report 2016.
- Bank of Spain, 2018, Economic Bulletin 1/2018.
- Bassanini, A., L. Nunzieta, and D. Venn, 2009, “Job Protection Legislation and Productivity Growth in OECD Countries,” Economic Policy, 24(58): 349–402.
- Bayoumi, T., R. Harmsen, and J. Turunen, 2011, “Euro Area Export Performance and Competitiveness,” IMF Working Paper 11/140, Washington: International Monetary Fund.
- Bentolila, S. and G. Bertola, 1990, “Firing Costs and Labour Demand: How Bad is Eurosclerosis?” Review of Economic Studies, 57(3): 381–402.
- Bluedorn, J. and H. Lin, 2017, “External Adjustment in Europe: Competitiveness, The Real Exchange Rate, and the Trade Balance,” Euro Area Policies: Selected Issues, IMF Country Report No. 17/236, Washington: International Monetary Fund.
- Bobeica, E., O. Tkacevs, and S. Christodoulopoulou, 2016, “The Role of Price and Cost Competitiveness for Intra- and Extra-Euro Area Trade of Euro Area Countries,” European Central Bank Working Paper Series 1941.
- Buck, T., 2017, “Spain: Boom to Bust and Back Again,” Financial Times, April 6.
- Cacciatore, M., R. Duval, G. Fiori, and F. Ghironi, 2016, “Market Reforms in the Time of Imbalance,” Journal of Economic Dynamics and Control, 72: 66–93.
- Correa-López, M. and R. Doménech, 2017, “Service Regulations, Input Prices and Export Volumes: Evidence from a Panel of Manufacturing Firms,” Banco de España Working Paper No. 1707, Madrid.
- Cuñat, A. and M. Melitz, 2012, “Volatility, Labor Market Flexibility, and the Pattern of Comparative Advantage,” Journal of the European Economic Association, 10(2): 225–254.
- Dao, M., 2015, “Recent Labor Market Reforms: A Preliminary Assessment,” Spain: Selected Issues, IMF Country Report No. 15/233, Washington: International Monetary Fund.
- De Lucio, J., R. Mínguez, A. Minondo, and F. Requena, 2017, “Los Márgenes del Crecimiento de las Exportaciones Españolas Antes y Después de la Gran Recesión,” Estudios de Economía Aplicada, 35(1): 43–62.
- De Lucio, J., R. Mínguez, A. Minondo, and F. Requena, 2018, “¿Se ha Producido un Milagro Exportador en España?,” Mimeo.
- Doménech, R., J.M. García, and C. Ulloa, forthcoming, “The Effects of Wage Flexibility on Activity and Employment in the Spanish Economy,” Journal of Policy Modeling.
- Duval, R. and D. Furceri, 2018, “The Effects of Labor and Product Market Reforms: The Role of Macroeconomic Conditions and Policies,” IMF Economic Review, 66(1): 31–69, Washington: International Monetary Fund.
- Duval, R., D. Furceri, B. Hu, J. Tovar Jalles, and H. Nguyen, 2018, “A Narrative Database of Major Labor and Product Market Reforms in Advanced Economies,” IMF Working Paper 18/19, Washington: International Monetary Fund.
- Eppinger, P., N. Meythaler, M. Sindlinger, and M. Smolka, 2018, „The Great Trade Collapse and the Spanish Export Miracle: Firm-level Evidence from the Crisis,” The World Economy, 41: 457–493.
- European Commission, 2018, European Semester Country Report Spain 2018.
- European Commission, 2018a, “The Determinants of Export Quality in the Euro Area,” Quarterly Report of the Euro Area, Vol. 17, No. 1.
- Garicano, L., C. Cruz, 2018, “El Milagro Exportador Español,” El País, May 25.
- González Pandiella, A., 2015, “A Constant Market Share Analysis of Spanish Goods Exports,” OECD Economics Department Working Papers (2015)4.
- Gonzalez Sanz, M.J. and C. Machuca, 2015, “The Internationalisation of Spanish SMEs: Main Developments and Their Determinants,” Banco de España, Economic Bulletin, December.
- Grinberg, F., 2017, “Assessing Spain’s Competitiveness,” Spain: Selected Issues, IMF Country Report No. 17/24, Washington: International Monetary Fund.
- Henn, C., C. Papageorgiou, J.M. Romero, and N. Spatafora, 2017, “Export Quality in Advanced and Developing Economies: Evidence from a New Data Set,” World Bank Policy Research Working Paper 8196, Washington: World Bank Group.
- International Monetary Fund, 2016, “Time for a Supply-Side Boost? Macroeconomic Effects of Labor and Product Market Reforms in Advanced Economies,” World Economic Outlook (April 2016), Chapter 3, Washington.
- International Monetary Fund, 2018, “2018 External Sector Report –Technical Supplement,” Washington.
- International Monetary Fund, World Bank, and World Trade Organization, 2018, “Reinvigorating Trade and Inclusive Growth.”
- Izquierdo, M., A. Lacuesta, and S. Puente, 2013, “The 2012 Labour Reform: An Initial Analysis of Some of its Effects on the Labour Market,” Banco de España, Economic Bulletin, September.
- Myro, R., 2015, “España en la Economía Global: Claves del Éxito de las Exportaciones Españolas,” RBA Libros.
- Myro, R., 2018, “Crecimiento Económico con Equilibrio Exterior. Un Nuevo Escenario para la Economía Española,” EuropeG, Policy Brief No.13.
- OECD, 2014, “The 2012 Labour Market Reform in Spain: A Preliminary Assessment. OECD Publishing.”
- Prades, E. and C. García, 2015, “Actualización de la Función de Exportaciones Españolas de Bienes,” Banco de España, Boletín Económico, Abril.
- Prades, E. and P. Villanueva, 2017, “Spain in the Global Value Chains,” Banco de España, Economic Bulletin 3/2017, Analytical Articles.
- Stothard, M., 2018, “Spain’s Exporters Drive Robust Economic Recovery,” Financial Times, April 23.
- The Economist, 2017, “Bullish Again: Spain’s Reforms Point the Way for Southern Europe,” June 15.

### Appendix: Data sources and employment protection indicators used in cross-country panel regressions
- Data sources:
  - IMF’s WEO database
  - OECD’s Employment Protection Database
  - WEF’s Global Competitiveness Report Database
- Note on OECD indicators: reflect employment protection regulations in force on January 1st of a given year; therefore they embody some degree of lagged reform effects.
- Other labor market-related indicators from OECD and the WEF-GCI databases were included in alternative regressions (not reported) but did not produce statistically significant results.

### Employment Protection Indicators included in the reported regressions
- Regulation of regular contracts (individual dismissals) (“EPR_V1”): 1985-2013; Source: OECD – Employment Protection database
- Regulation of regular contracts (individual and collective dismissals) (“EPRC_V2”): 1998-2013; Source: OECD – Employment Protection database
- Regulation of temporary contracts (“EPT_V1”): 1985-2013; Source: OECD – Employment Protection database
- Average length of notice period (“REG3” = simple average of ‘REG3a’, ‘REG 3b’, and ‘REG3c’): 1985-2013; Source: OECD – Employment Protection database
- Hiring and firing practices (“Hiring/Firing”): 2006-2015; Source: World Economic Forum – Global Competitiveness Index Historical Dataset

### Spain: indicator evolution and measurement notes
- Chart notes:
  - OECD indicators vary from 0 (least restrictive) to 6 (most restrictive).
  - WEF's "Hiring/Firing" indicator varies from 0 (worst) to 7 (best).
  - The chart illustrates the evolution of EPR_V1, EPRC_V2, REG3, EPT_V1, and Hiring/Firing for Spain; most show improvements coinciding with the implementation of the labor market reforms of 2010 and 2012.
- Extension for regressions:
  - To gain observations, the OECD indicators were extended until 2016 under the conservative assumption of no-changes in employment protection legislations since 2013.
  - The results are robust to estimating the regressions only until 2013.

### Robustness experiments and key methodological findings
- Robustness experiments considered (in addition to those reported in the main text):
  - (i) Dropping the lagged log change of the ULC-based REER as control, and adding instead either the lagged log change of the nominal effective exchange rate (NEER) or both the lagged and contemporaneous log change of the NEER.
  - (ii) Adding a lagged interaction term of the low-growth dummy with the labor market flexibility indicator.
  - (iii) Controlling for product market reform indicators taken from Duval et al.’s (2018) database.
- Outcomes:
  - In these robustness experiments, the main baseline results were largely unchanged, though in some cases the magnitudes of the ββ coefficients declined somewhat (in absolute value).
  - When all the labor market flexibility indicators were lagged one period, the coefficients on the key regressors of interest became statistically insignificant.
  - In alternative regressions that included labor market regulation indicators from Duval et al.’s (2018) database, the estimated coefficient attached to their indicator on “employment protection legislation for regular workers” had the expected sign (i.e., implying that less stringent employment protection has positive effects on exports) but was statistically insignificant.

*Source: wp18283 - REFERENCES (IMF working paper content).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18283.pdf_