## 2. European Wage Dynamics and Labor Market Integration

## Source details

**Canonical URL:** [2. European Wage Dynamics and Labor Market Integration](https://www.imf.org/-/media/files/publications/reo/eur/2018/may/eur-chapter-2.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/eur/2018/may/eur-chapter-2.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/eur/2018/may/eur-chapter-2.pdf.json)

---

### Recent headline developments
- Employment increased by 12.7 million people since early 2013, exceeding precrisis peaks by 2.2 percent at the end of 2017.
- Unemployment fell from 11 percent in early 2013 to 7.3 percent in late 2017.
- Unemployment dropped faster in the newer EU member states, falling by a total of 5 percentage points on average; the EU15 average rate declined by 3 percentage points.

### Divergent wage dynamics: EU15 versus newer EU members
- EU15 nominal and real wage outcomes
  - Average EU15 nominal wage increases have remained below 2 percent since 2012 and were 1.8 percent in the fourth quarter of 2017.
  - Average nominal wage growth in the EU15 has remained around 1½ percent since 2011, down from just over 3 percent during 2003–09.
  - Average real wage growth in the EU15 slowed by ½ percentage point to about 0.4 percent since 2011.
  - Real product wage growth in the EU15 averaged ½ percent in 2017.
  - The GDP deflator is used to calculate real wages in this analysis.
  - The ratio of real wages to trend labor productivity jumped by 2 percentage points during 2009 and had returned to its 2005–08 average by 2017; at end-2017 unemployment was just ½ percentage point higher than in 2005–08.
  - Inflation in output prices declined by 1 percentage point to about 1¼ percent in recent years.
  - Estimated trend labor productivity growth slowed by ½ percentage point to 0.6 percent.

- Newer EU member states (NMS) wage acceleration
  - Average nominal wage increases in newer member states accelerated to 6¾ percent in the fourth quarter of 2017, up from 3½ percent in 2014.
  - Alternative descriptions note nominal wage growth of about 3½ percent in 2011–15 and a pickup close to 6 percent year over year on average in 2017; another figure cites acceleration to 6½ percent by 2017.
  - Real product wage growth in newer members surged from 3 to 5¼ percent year over year during 2014–17; another passage reports real wage gains of 4.1 percent year over year in 2017 and notes real wage growth picked up to about 4½ percent in 2016–17.
  - Trend labor productivity in newer members averaged about 2 percent in recent years.
  - The average ratio of real wages to trend labor productivity was roughly 2½ percent higher than its historical average in 2017 (a level exceeded only in mid-2008 for two quarters).

- Labor market tightness and timing
  - Slack in newer member states is estimated to have largely disappeared by about mid-2015, consistent with the timing of the wage acceleration.
  - Unemployment in newer members averaged 6 percent by the end of 2017.
  - Correlations between deviations of real wages from trend productivity and unemployment gaps peak after six quarters in newer members and after eight quarters in EU15 countries.
  - Labor shortages exceed precrisis peaks in the newer members and are especially strong in industry.

### Drivers, mechanisms, and interactions
- Domestic drivers and wage moderation
  - Reductions in inflation expectations and trend productivity growth, together with expanded measures of labor market slack, account for much of wage slowdown in advanced economies (Chapter 2, October 2017 WEO).
  - Corrections of postcrisis wage overhangs contributed to EU15 moderation; some countries (for example, Belgium) used wage freezes.
  - In newer members, low domestic slack is consistent with strong wages; spillovers from EU15 labor market recovery could be relevant.

- Evolving employment arrangements and underemployment
  - Shift toward self-employment, temporary contracts, and part-time jobs can reduce employment security and bargaining power, potentially putting downward pressure on wages.
  - Involuntary part-time employment rose starting in the early 2000s and peaked in 2014 at the EU level.
  - The share of people marginally attached to the labor force increased after the global financial crisis but has declined in recent years at the EU level.
  - Cross-country heterogeneity: Germany reduced involuntary part-time employment after the crisis; newer members recorded declines in marginal attachment; temporary contracts rose in newer members before the crisis; self-employment is relatively higher in the Netherlands and the United Kingdom.

- Integration, spillovers, and external linkages
  - Domestic conditions driving wages can have a significant common component; direct spillovers on wage setting across countries are possible.
  - Newer EU members are more flexible relative to domestic slack and more responsive to external conditions, suggesting cross-border spillovers influence wage dynamics.
  - Rising EU15 employment and higher wages attract workers from newer members, affecting labor market tightness and wage pressures across countries.

### Analytical approach and modeling
- Frameworks used
  - Phillips curve framework controlling for inflation expectations and alternative measures of slack beyond unemployment.
  - Error correction model (ECM) to account for wages temporarily away from equilibrium, given nominal wage rigidities and Europe’s double-dip recession.
  - Analyses of spillovers via slack, wages, and migration.

- Slack indicators and empirical inclusion
  - Broader slack indicators: nonemployment index (unemployment, marginal attachment, underemployment), involuntary part-time employment, hours per employee (intensive margin).
  - Inflation expectations proxied by Consensus Forecasts (one year ahead) included with lags.
  - Spillovers considered sequentially: external unemployment, foreign wages, and migration flows.

### Key empirical relationships and timing
- Long-term (country-specific) findings
  - Structural reforms affect long-term real wages; some long-term wage spillovers found in smaller euro area countries.
  - Germany: Hartz reforms (2003–05) lower the equilibrium real wage by 5½ percent; long-term productivity coefficient close to unity (with reform dummy, coefficient 1.027*** (0.0328)).
  - France and Spain: long-term real wage developments depend on domestic productivity trends; Spain: reforms reduce real wages by an estimated 4 percent; France: reform dummy positive due to timing near the global financial crisis.
  - Austria: employment protection reforms reduce real wages by 3 percent; Log Real German Wage coefficient about 0.179*** (0.0509).
  - Belgium: French and German wages have lasting effects (Log Real French Wage 0.494*** (0.165)).
  - Netherlands: Log Real German Wage 0.117* (0.0590).
  - Newer members: in Czech Republic, Estonia, Latvia, Slovak Republic, and Slovenia the long-term coefficient on labor productivity is close to unity; Poland and Hungary are outliers (e.g., Poland 0.696*** (0.0314); Hungary 0.598*** (0.0311)).

- Short-term panel ECM findings
  - Preferred baseline ECM uses the nonemployment index gap and hours per employee.
  - ECM error correction parameter: –0.36 in “other euro area” and –0.40 in newer member states (textual finding).
  - Interpretation: roughly 40 percent of a deviation from long-term equilibrium is corrected each year (64 percent after two years; 78 percent after three years).
  - Selected short-term coefficients (Euro area excluding Germany, panel specifications):
    - Trend Labor productivity, Log, Four-Quarter Change (t–4): coefficients around 0.82*** to 0.57*** across specs (standard errors 0.09–0.12).
    - HICP, Log, Four-Quarter Change (t–5): coefficients around 0.31*** to 0.19***.
    - Expected Inflation, One year Ahead (t–4): coefficients range 0.50** to 0.94***.
    - Nonemployment Gap (t–4): coefficients roughly 2 0.62*** to 2 0.29*** (standard errors 0.05–0.06).
    - Hours Gap (t–4): when included, coefficients 0.31***; 0.16**; 0.44***; 0.50***.
    - Migration variables: Change of Net Migrant Inflow (t–5) 2 0.44*** (0.16); Change of Gross Migrant Inflow (t–5) 2 0.31* (0.17); Change of Gross Migrant Outflow (t–5) 1.23*** (0.37).
    - Error Correction Term (t–4): coefficients around 2 0.36***; 2 0.33***; 2 0.38*** (standard error 0.05).
    - R-squared across specifications: 0.53 to 0.69; Observations vary (e.g., 383, 380, 296); number of countries in panel: 5.

### Spillovers and migration effects
- Spillover magnitudes and timing
  - Spillovers from euro area labor market developments significantly affect wage growth in the “other euro area” panel and especially in newer member states.
  - In “other euro area” countries, changes in euro area unemployment (lagged two quarters) and euro area wage growth (lagged three quarters) are statistically significant.
  - In newer member states, changes in unemployment in the euro area have substantial and significant impacts on wage growth with short lags.

- Migration effects quantified
  - A rise in net migration inflows by 1 percentage point (four-quarter change in net migration inflows in the prior year as a share of the labor force) is estimated to reduce wage growth in the following year by 0.4 percent in the “other euro area” and by 0.5 percent in the newer EU members.
  - The text notes a 1 percentage point rise is “an exceptionally large shock” and that in practice changes in net migration flows are generally small relative to the labor force, so wage effects are small and temporary.
  - When separating inflows and outflows, gross migrant outflows have a larger coefficient than inflows, though the difference is not statistically significant.

### Short-term analysis of Germany
- Germany-specific findings
  - Growth in involuntary part-time employment is a key indicator of slack for Germany.
  - German wage Phillips curve shows high responsiveness to the unemployment gap, trend productivity, and inflation expectations.
  - Error correction term for Germany about –0.6; ECM with changes in involuntary employment is the best baseline model for Germany.
  - German wages appear anchored on domestic conditions and less sensitive to regional spillovers; German wages not responsive to changes in euro area unemployment in certain specifications; changes in net migration flows not statistically significant for German wages.

### Drivers of recent wage movements by region
- Euro area excluding Germany
  - Inflation and inflation expectations, domestic and external slack, and the ECM term explain variations in nominal wage growth.
  - The initial slowing in wage rises by 2.3 percentage points in 2010 is linked to: lower inflation expectations (0.8 percentage point), lower past inflation (0.5 percentage point), domestic labor slack (0.6 percentage point), and the error correction (0.3 percentage point).
  - Error correction continued to weigh on wages by 0.2 to 0.3 percentage point during 2011–15.

- Newer EU members
  - Declines in domestic and foreign slack contributed to the wage pickup in recent years.
  - Diminishing domestic slack in newer members contributed an estimated 2 percentage points to wage growth in recent years.
  - Combined spillovers from euro area unemployment and euro area wage rises accounted for about 1½ percentage points of the wage pickup in newer members during 2015–17.

- Germany
  - The steep initial fall in wage growth by 2.9 percentage points in 2010 is linked to: lower inflation expectations (0.9 percentage point), domestic labor slack (0.3 percentage point), and the error correction (1 percentage point).
  - Wage growth above 3 percent in 2012 led to significant error correction drag during 2012–14, peaking at 0.7 percentage point in 2013.

### Scenario analysis: German wage shock and spillovers
- Illustrative simulation
  - Assumption: hourly wage rates in Germany rise 1 percentage point faster than baseline.
  - Effects:
    - This shock would raise euro area wage growth by about 0.4 percentage point given Germany’s weight.
    - Wage growth in the euro area excluding Germany rises about 0.1 percent after one year and 0.2 percent in the medium term.
    - Impact on newer EU member wages about 0.15 percent after one year but fades to zero over time (long-term real wages in newer members depend on domestic labor productivity).
  - Note: If higher wages lead to higher inflation, medium-term wage impacts would be larger than simulations indicate.

### Inflation expectations, Phillips curve slopes, and pass-through
- Anchoring of inflation expectations
  - Method: πt^e − π̄ = α + β(πt − π̄) + εt with Consensus Forecasts; inflation target π̄ taken as 1.9 percent for euro area members (exceptions: Czech Republic, Hungary, Poland, Slovak Republic use announced targets).
  - Five-year rolling regressions show an increase in the coefficient on actual inflation for the euro area, implying expectations have become more sensitive to actual inflation (some deanchoring).
  - Newer members: inflation expectations appear more anchored over time, with sensitivity to actual inflation quite low.

- Evolution of wage Phillips curve and forward-looking setting
  - Panel with 15-year rolling windows (quarterly data Q1 1996–Q3 2017) estimated separate for euro area (excl. Germany) and newer members.
  - Slope of the wage Phillips curve:
    - Euro area: coefficient on unemployment gap starts at about –0.4 and rises modestly to –0.3.
    - Newer members: coefficient stable at about –1.
  - Forward-looking wage setting:
    - Euro area: coefficient on inflation expectations rose from about 0.3 in the late 1990s to about 0.8 recently; wage response to actual inflation declined modestly from about one-quarter to one-fifth.
    - Newer members: coefficient on expected inflation varied without clear trend; some decline in impact of actual inflation recently.

- Pass-through from wages to prices (panel VAR)
  - Four-variable panel VAR (relative import price inflation, nominal wage growth adjusted for trend productivity, consumer price inflation, unemployment gap) using Q1 1998–Q3 2017; Cholesky ordering: import prices, wages, consumer prices, unemployment gap.
  - Dynamics:
    - Impact of a 1 percentage point wage shock on cumulative price inflation over three years: EU15 -> 0.6 percent in prices and 2.4 percent in wages (implying pass-through of 25 percent).
    - Newer members -> 0.9 percent in prices and 3.6 percent in wages (implying pass-through of 25 percent).
    - Conclusion: medium-term pass-through from wages to prices is positive but less than full in both groups.

### Policy-relevant implications and recommendations
- Measurement and monitoring
  - Supplement unemployment with measures of underemployment (involuntary part-time employment, marginal attachment) and hours worked per person when assessing slack.
  - Use a nonemployment index combining unemployment, marginal attachment, and underemployment as an alternative slack measure.

- Monetary and macro policy
  - Euro area: underpin a firm reanchoring of expectations to the inflation target; ensure the European Central Bank stays committed to its strategy to durably raise inflation to target.
  - Projection implication: provided unemployment continues to fall and inflation gradually increases in line with April 2018 WEO projections, wage rises are projected to pick up in the EU15, but the process will take much longer than in the newer EU members.

- Structural and labor market policies (newer EU members)
  - Mobilize labor supply through measures to reduce skill and locational mismatches to contain wage pressure and support employment and growth.
  - Active labor market policies to increase participation rates and reduce structural unemployment given unfavorable demographics and emigration.
  - Expand retraining courses for the unemployed and apprenticeship systems.
  - Structural reforms to strengthen institutions and improve public sector efficiency to support convergence and reduce emigration pressures.

- Current account and competitiveness
  - Unit labor cost–based real effective exchange rates and an estimated CA/REER elasticity median of 0.33 imply relative unit labor cost changes alone are well short of explaining large current account adjustments.
  - The sharp external balance adjustments were principally driven by a sudden stop in capital inflows; labor market slack and unit labor cost declines supported adjustment but were not the primary driver.
  - Germany and the Netherlands: continued sizable surpluses call for policy adjustments to promote a greater balance of saving and investment; recommended policies include supporting relative unit labor cost gains in Germany and relative increases in the Netherlands.

### Areas for further analysis
- Investigate magnitude and channels of cross-border spillovers (labor flows, wage-setting linkages) between EU15 and newer member states.
- Clarify the role of migration of skilled workers, minimum wage hikes, and public sector wage increases in newer member wage acceleration.
- Monitor evolving employment arrangements (temporary contracts, self-employment, part-time work) as potential sources of persistent slack that can dampen wage growth.

*Source: IMF Regional Economic Outlook: Europe, Chapter 2 (May 2018).*

### 12.7 million people since early 2013, exceeding

### 2. European Wage Dynamics and Labor Market Integration

### Recent headline developments
- Employment increased by 12.7 million people since early 2013, exceeding precrisis peaks by 2.2 percent at the end of 2017.
- Unemployment fell from 11 percent in early 2013 to 7.3 percent in late 2017.
- Unemployment dropped faster in the newer EU member states, falling by a total of 5 percentage points on average; the EU15 average rate declined by 3 percentage points.

### Divergent wage dynamics: EU15 versus newer EU members
- EU15 nominal wage growth and real wage outcomes
  - Average EU15 nominal wage increases have remained below 2 percent since 2012 and were 1.8 percent in the fourth quarter of 2017, about half of precrisis (2001–08) average growth rates.
  - Average nominal wage growth in the EU15 has remained around 1½ percent since 2011, down from just over 3 percent during 2003–09.
  - Average real wage growth in the EU15 slowed by ½ percentage point to about 0.4 percent since 2011.
  - Real product wage growth in the EU15 averaged ½ percent in 2017, little changed from 2014.
  - The GDP deflator is used to calculate real wages in this analysis.
  - The ratio of real wages to trend labor productivity jumped by 2 percentage points during 2009 and had returned to its 2005–08 average by 2017; at end-2017 unemployment was just ½ percentage point higher than in 2005–08.
  - Inflation in output prices declined by 1 percentage point to about 1¼ percent in recent years.
  - Estimated trend labor productivity growth slowed by ½ percentage point to 0.6 percent.

- Newer EU member states (NMS) wage acceleration
  - Average nominal wage increases in newer member states accelerated to 6¾ percent in the fourth quarter of 2017, up from 3½ percent in 2014 (led by the Baltics and southern European countries; central European country wages picked up more recently).
  - Alternative descriptions in the chapter note nominal wage growth of about 3½ percent in 2011–15 and a pickup close to 6 percent year over year on average in 2017; another figure text cites acceleration to 6½ percent by 2017.
  - Real product wage growth in newer members surged from 3 to 5¼ percent year over year during 2014–17; another passage reports real wage gains of 4.1 percent year over year in 2017 and notes real wage growth picked up to about 4½ percent in 2016–17.
  - Trend labor productivity in newer members averaged about 2 percent in recent years.
  - The cumulative effect led the average ratio of real wages to trend labor productivity to be roughly 2½ percent higher than its historical average in 2017 (a level exceeded only in mid-2008 for two quarters).

- Labor market tightness and timing
  - Slack in newer member states is estimated to have largely disappeared by about mid-2015, consistent with the timing of the wage acceleration.
  - Unemployment in newer members averaged 6 percent by the end of 2017, in line with precrisis lows.
  - Correlations between deviations of real wages from trend productivity and unemployment gaps peak after six quarters in newer members and after eight quarters in EU15 countries.
  - Labor shortages exceed precrisis peaks in the newer members and are especially strong in industry.

### Drivers, mechanisms, and interactions analyzed
- Key domestic drivers highlighted
  - Reductions in inflation expectations and trend productivity growth, together with expanded measures of labor market slack, account for much of wage slowdown in advanced economies (as noted in Chapter 2 of the October 2017 WEO).
  - Corrections of postcrisis wage overhangs contributed to EU15 moderation; in some countries (for example, Belgium) wage freezes played a significant role.
  - In newer members, low domestic slack is consistent with strong wages, though other domestic factors and spillovers from EU15 labor market recovery could be relevant.

- Role of evolving employment arrangements and underemployment
  - A shift toward self-employment, temporary contracts, and part-time jobs can reduce employment security and workers’ bargaining power, potentially putting downward pressure on wages.
  - Involuntary part-time employment (OECD and Eurostat measures) rose starting in the early 2000s and peaked in 2014 at the EU level, which could have contributed to wage moderation.
  - The share of people marginally attached to the labor force increased following the global financial crisis but has declined in recent years at the EU level.
  - Cross-country heterogeneity: Germany reduced involuntary part-time employment after the global financial crisis; newer members recorded a decline in marginal attachment consistent with falling unemployment; temporary contracts rose in newer members before the crisis; self-employment is relatively higher in the Netherlands and the United Kingdom.
  - Chapter 2 of the October 2017 WEO finds that involuntary part-time employment increases measured slack; UK wages are noted to be sensitive to cyclical self-employment.

- Integration, spillovers, and external linkages
  - Domestic conditions driving wages can have a significant common component given economic linkages between countries, and direct spillovers on wage setting across countries are possible.
  - Newer EU members are found to be more flexible in relation to domestic labor market slack and more responsive to external labor market conditions, suggesting potential for cross-border spillovers to influence wage dynamics.
  - Rising EU15 employment and appreciably higher wages make the EU15 an attractive destination for workers from newer member states, which can affect labor market tightness and wage pressures across countries.

### Analytical approach and modeling
- The chapter evaluates wage dynamics using:
  - A Phillips curve framework controlling for inflation expectations and alternative measures of slack beyond unemployment.
  - An error correction model (ECM) to account for wages temporarily being away from equilibrium, given nominal wage rigidities and Europe’s double-dip recession.
  - Analyses of spillovers via slack, wages, and migration.

### Key empirical relationships and timing
- Correlation and lag structure between wages and unemployment gaps:
  - Correlations between real wage deviations and unemployment gaps are initially higher in newer EU members and peak after six quarters.
  - In EU15 countries these correlations start lower and peak after eight quarters.
- Real wage behavior:
  - EU15 real wages do not appear out of line with productivity and unemployment after accounting for unwinding of overhangs and lagged wage responses.
  - In newer members, low slack and strong wages are consistent, but the role of cross-border spillovers remains to be clarified.

### Implications and areas for further analysis
- Consider supplementing unemployment with measures of underemployment (involuntary part-time employment, marginal attachment) when assessing labor market slack.
- Adjust wage-inflation analysis for inflation expectations and for differential lags in wage responses across country groups.
- Investigate the magnitude and channels of cross-border spillovers (labor flows, wage-setting linkages) between EU15 and newer member states, given differing responsiveness to slack and external conditions.
- Monitor evolving employment arrangements (temporary contracts, self-employment, part-time work) as potential sources of persistent slack that can dampen wage growth.

*Source: IMF Regional Economic Outlook: Europe, Chapter 2 (May 2018).*

### 1. Employment Structure Indicators2. Alternative Slack Indicators

### eur-chapter-2 - 1. Employment Structure Indicators2. Alternative Slack Indicators

### Employment structure and alternative slack indicators
- Unemployment rate remains the headline slack measure, but alternative indicators capture additional dimensions of labor underutilization, including:
  - involuntary part-time employment,
  - marginally attached to the labor force,
  - temporary contracts,
  - self-employment,
  - a nonemployment index that combines unemployment with marginal attachment and underemployment.
- The nonemployment index:
  - incorporates potential labor input from those currently unemployed, marginally attached to the labor force, and underemployed (Byrne and Conefrey (2017) methodology).
  - across countries, changes in the nonemployment index appear somewhat positively correlated with changes in unemployment rates, but the correlation is not perfect (Figure 2.10).
  - implication: the nonemployment index potentially captures additional information that is not embedded in the unemployment rate.
- Intensive margin (hours worked per person) as a slack indicator:
  - recent literature suggests hours worked per person is relevant for wage growth (Bulligan, Guglielminetti, and Viviano (2017)).
  - findings: the intensive margin of labor utilization is relevant for wage growth; the Phillips curve becomes flatter for lower levels of hours per worker.
  - Germany’s labor market arrangements that reduce hours to limit job losses make the hours-per-worker indicator particularly relevant.
- Simple correlation analysis of slack indicators:
  - highest correlations are between unemployment gaps and the nonemployment index gap (index gaps envisaged as an alternative to unemployment gaps in econometric analysis).
  - other notable correlations:
    - involuntary part-time employment and unemployment gap.
    - involuntary part-time employment and nonemployment index gap.
  - specific country correlations:
    - highest correlation: 0.72 between the unemployment gap and changes in involuntary unemployment in Germany.
    - next highest: 0.5 for the nonemployment index gap in Germany.
  - hours per employee indicator has low correlations with the other indicators.
- Time-series behavior noted in figures:
  - hours worked per employee are shown as percent deviation from trend, with plotted range markers between –3.0 and 2.5 (Figure 2.11 axis labels).
  - labor market indicators (involuntary part-time, marginal attachment, temporary contracts, self-employment) are presented as percent of total employment or percent of working-age population relative to 2000–07 averages in cross plots (Figures 2.8 and 2.9).

### Labor market integration and migration
- EU accession and migration patterns:
  - new accessions to the European Union since 2004 facilitated a rise in migration from these countries to the EU15.
  - between 2000 and 2015, the stock of migrants from new members in the EU15 tripled to 2.1 percent of the EU15 population.
  - migration to EU15 from outside the European Union has also been sizable; migrants to the EU15 came mainly from the former Soviet Union in the early 1990s, from former colonial countries, and in recent years as refugees.
  - migration stocks within new members and from EU15 countries to new members remained broadly stable as shares of the respective populations, and within the EU15 migrant stocks rose gradually to 2.5 percent of the EU15 population.
- Posted workers and cross-border workers:
  - posted workers: employees temporarily sent by their employer to carry out a service in another EU country; they remain temporary and do not integrate into the host country’s labor market but can affect labor market slack.
  - net inflows of posted workers to some EU15 countries—especially Germany—have increased since 2007, with rising net outflows from newer members such as Slovenia and the Slovak Republic (Figure 2.13).
  - cross-border workers across the European Union in 2015 were estimated at 1.3 million (or 0.6 percent of the total employed).
- Emigration from newer EU members:
  - most emigrants from newer EU member countries are highly educated; education levels of emigrants tend to be higher than home-country averages.
  - consequence: skill shortages in home countries, with the brain drain most prominent in the Baltics, potentially contributing to recent hikes in wages in these countries.
  - large-scale emigration may have slowed growth and income convergence with the EU15 and contributed to fiscal burdens from a higher dependency ratio.

### Goods market integration, global value chains (GVCs), and implications for labor
- Pan-European supply chains and intermediate trade:
  - on average, 70 percent of exports of goods and services of EU countries represents trade in intermediate goods that are part of supply chains (Figure 2.14, panel 1).
  - this share is significantly higher than in other parts of the world and has been on the rise in the European Union, particularly over 2000–07.
  - EU linkages are generally stronger than non-EU linkages: on average two-thirds of foreign value added in EU exports originates from within the European Union.
- Germany’s role:
  - Germany plays a key role in EU supply chains.
  - integration of EU countries has increased the most with Germany, followed by China, and to a lesser extent with France and Italy.
  - Germany is the most important source of intermediate inputs for many EU countries, followed by France, the United Kingdom, and Italy (Figure 2.15).
  - Germany acts as a hub, with large inflows and outflows of intermediate goods and services flowing through Germany to various EU countries.
- Foreign direct investment (FDI) and GVCs:
  - as supply chains expanded in newer EU members, the stock of foreign direct investment has been increasing in those countries over time (Figure 2.16, panels 1 and 3).
  - within the European Union:
    - newer members are the largest recipients of foreign direct investment;
    - EU15 countries are generally net foreign direct investment exporters.
  - about 90 percent of inward foreign direct investment to newer EU members is from within Europe and 85 percent from within the European Union.
  - Germany and Austria are major senders of foreign direct investment to central, eastern, and southeastern Europe; Sweden is the largest financier in the Baltics (Figure 2.16, panel 2).
- Implications for labor market sensitivity:
  - supply chain integration increases sensitivity of domestic labor markets to global production cycles and external conditions, contributing to increased business cycle synchronization (IMF 2013).
  - indicators of labor shortages for EU countries are highly correlated with indicators of external demand (for example, aggregate EU industrial production), particularly for more integrated newer member states.
  - the sensitivity of labor shortages to external conditions has increased over time for both the newer members and the EU15, but labor shortages in the newer members are more sensitive to global shocks than in the EU15.

### Exploring drivers of European wage behavior
- Integration channels affecting wages:
  - EU integration can shape wage behavior by changing elasticities of labor demand and supply and by influencing labor market slack via migration, posted workers, and commuting.
  - longer-term integration is expected to promote convergence in productivity through diffusion of technology and management skills.
- Evidence of cross-country wage spillovers:
  - Sweden: IMF (2017a) finds long-term real wages determined by labor productivity; growth in Swedish nominal wages shows a sizable spillover from German wage growth, with the linkage strengthening beginning in the early 2000s. A second spillover operates via changes in euro area unemployment, with a negative effect on Swedish wage growth that increased starting in the early 2000s.
  - United Kingdom: IMF (2018) finds growth in wages negatively affected by the EU unemployment gap.
  - broader indicators of labor market slack (for example, involuntary part-time employment) have significant negative impacts on wages in these country studies; in the United Kingdom, cyclical increases in self-employment also have negative effects on wages.
  - IMF (2016) finds emigration from eastern to western Europe lowered labor supply in the east and was associated with higher remittances, which may have eased budget constraints and contributed to higher wages in the east.
- Integrated analysis approach described in the chapter:
  - the analysis includes broader indicators of slack: unemployment, nonemployment gap, hours per employee gap, and involuntary part-time employment.
  - it includes inflation and inflation expectations: lags of actual inflation and Consensus Forecasts (one year ahead) for inflation as a proxy for inflation expectations.
  - the aim is to provide a formal empirical analysis that includes the various drivers of wage dynamics across European countries.

*Source: IMF Regional Economic Outlook: Europe chapter excerpt (figures and text as provided).*

### 3. Stock of Inward FDI

### 3. Stock of Inward FDI

### Modeling approach and key variables
- Long-term models of real wages are based on trend labor productivity; separate ordinary least squares regressions are estimated for each country.
- Short-term baseline models estimate nominal wage growth (four-quarter growth) using a Phillips curve augmented by:
  - lagged inflation, inflation expectations, trend productivity growth, unemployment gap (level and change), lagged wage growth;
  - the fourth lag of the residual from the long-term equation (error correction term) in the ECM specification.
- Broader slack indicators assessed: nonemployment index gap (alternative to unemployment), hours per employee (intensive margin), involuntary part-time employment (for Germany).
- Spillovers considered sequentially: external unemployment, foreign wages, and migration flows (migration data for 2016–17 limited).

### Long-term analysis: substantive findings and country-specific effects
- General:
  - Structural reforms affect the level of real wages in the long term.
  - Some long-term wage spillovers are found in smaller euro area countries.
- Germany:
  - The Hartz reforms implemented in 2003–05 are found to lower the equilibrium real wage by 5½ percent.
  - Long-term productivity coefficient is close to unity; absent the reform dummy, the coefficient is 0.73.
- France and Spain:
  - Long-term real wage developments depend on domestic productivity trends.
  - Spain: reforms, including unemployment benefits, reduce real wages by an estimated 4 percent.
  - France: a reform dummy has a counterintuitive positive sign due to implementation near the global financial crisis when French real product wages rose about 3 percent relative to productivity.
- Netherlands and Austria:
  - Wages driven principally by domestic productivity.
  - German wages (deflated by each country’s GDP deflator) have a long-term impact with a coefficient of about 0.2.
  - Austria: employment protection reforms reduce real wages by 3 percent.
- Belgium:
  - French wages, in addition to German wages, have a lasting effect on Belgian wages; consistent with a 1996 law linking Belgian wage setting to neighboring countries.

### Long-term coefficients reported (selected values from Tables)
- Table 2.1 (Euro Area: Long-Term Equations for Total Labor Compensation) — selected coefficients (standard errors in parentheses):
  - Germany: Log Trend productivity 1.027*** (0.0328); Hartz Reform Dummy 20.0546*** (0.00541); Observations 919; R-squared 0.967; Dickey-Fuller Test 22.661.
  - France: Log Trend productivity 0.932*** (0.0193); Observations 918; R-squared 0.986; Dickey-Fuller Test 22.648.
  - Austria: Log Trend productivity 0.789*** (0.0300); Log Real German Wage 0.179*** (0.0509); Observations 774; R-squared 0.973; Dickey-Fuller Test 23.880.
  - Belgium: Log Trend productivity 0.437*** (0.152); Log Real French Wage 0.494*** (0.165); Observations 749; R-squared 0.886; Dickey-Fuller Test 21.720.
  - Spain: Log Trend productivity 0.868*** (0.0486); Unemployment benefit Reform Dummy 0.0319*** (0.00272); Observations 918; R-squared 0.802; Dickey-Fuller Test 22.843.
  - Netherlands: Log Trend productivity 0.898*** (0.0229); Log Real German Wage 0.117* (0.0590); Observations 786; R-squared 0.949; Dickey-Fuller Test 23.660.
- Table 2.2 (Newer EU Members: Long-Term Equations for Total Labor Compensation) — Log Trend productivity coefficients (standard errors in parentheses):
  - Czech Republic: 1.074*** (0.0163); Observations 877; R-squared 0.981; Dickey-Fuller Test 23.512.
  - Estonia: 1.187*** (0.0320); Observations 191; R-squared 0.952; Dickey-Fuller Test 22.616.
  - Hungary: 0.598*** (0.0311); Observations 919; R-squared 0.806; Dickey-Fuller Test 21.914.
  - Latvia: 1.012*** (0.0245); Observations 916; R-squared 0.950; Dickey-Fuller Test 22.034.
  - Lithuania: 0.936*** (0.0205); Observations 919; R-squared 0.959; Dickey-Fuller Test 22.609.
  - Poland: 0.696*** (0.0314); Observations 919; R-squared 0.892; Dickey-Fuller Test 22.588.
  - Slovak Republic: 1.031*** (0.0159); Observations 919; R-squared 0.979; Dickey-Fuller Test 22.922.
  - Slovenia: 0.789*** (0.0162); Observations 919; R-squared 0.964; Dickey-Fuller Test 23.038.
- Interpretation for newer members:
  - In Czech Republic, Estonia, Latvia, Slovak Republic, and Slovenia the coefficient on labor productivity is close to unity.
  - Poland and Hungary are outliers: Poland’s low coefficient linked to high initial real wage levels, a post-crisis public sector wage freeze, and surge in temporary foreign migrant workers; Hungary’s low estimate may reflect data issues including discontinuities in hours worked.

### Short-term regional panel analysis and ECM results
- Preferred baseline model: error correction model using the nonemployment index gap and hours per employee as additional slack indicator.
- Error correction parameters (columns (4)–(8) summarized):
  - ECM error correction parameter is sizable and highly significant: –0.36 in “other euro area” and –0.40 in newer member states (textual finding).
  - The ECM term is at the fourth lag: roughly 40 percent of a deviation from long-term equilibrium is corrected each year, or 64 percent after two years and 78 percent after three years.
- Short-term Phillips curve properties (selected coefficients and model metrics from Table 2.3, Euro Area excluding Germany):
  - Wage, Log, Four-Quarter Change (t–4): coefficients across specifications include 0.08*; 2 0.11**; 2 0.04; 0.04; 0.02; 0.05; 0.06; 0.03; 0.10*; 0.10* (standard error 0.05).
  - Trend Labor productivity, Log, Four-Quarter Change (t–4): 0.82***; 0.79***; 0.76***; 0.63***; 0.63***; 0.63***; 0.57***; 0.57***; 0.17; 0.18 (standard errors 0.09 or 0.12).
  - HICP, Log, Four-Quarter Change (t–5): 0.31***; 0.34***; 0.33***; 0.19***; 0.21***; 0.21***; 0.24***; 0.20***; 0.20**; 0.20** (standard errors 0.07 or 0.08).
  - Expected Inflation, One year Ahead (t–4): coefficients range 0.50** to 0.94*** across specs (standard errors 0.19–0.21).
  - Nonemployment Gap (t–4): coefficients 2 0.62***; 2 0.60***; 2 0.45***; 2 0.45***; 2 0.42***; 2 0.39***; 2 0.29***; 2 0.29*** (standard errors 0.05–0.06).
  - Hours Gap (t–4): where included, coefficients 0.31***; 0.16**; 0.16**; 0.14*; 0.44***; 0.50*** (standard errors 0.08–0.11).
  - EA Wage, Log, Four-Quarter Change (t–3): where included, 0.15**; 0.15*; 0.18* (standard errors 0.07–0.09).
  - Migration variables (where included): year-over-year Change of Net Migrant Inflow (t–5) 2 0.44*** (0.16); Change of Gross Migrant Inflow (t–5) 2 0.31* (0.17); Change of Gross Migrant Outflow (t–5) 1.23*** (0.37).
  - Error Correction Term (t–4) in multiple specifications: 2 0.36***; 2 0.36***; 2 0.33***; 2 0.33***; 2 0.32***; 2 0.38***; 2 0.37*** (standard error 0.05).
  - Observations vary by specification: 383, 380, 380, 383, 380, 380, 380, 380, 296, 296.
  - R-squared ranges across columns: 0.53, 0.54, 0.56, 0.60, 0.61, 0.61, 0.61, 0.62, 0.68, 0.69.
  - Number of countries in panel: 5.

### Policy-relevant implications and interpretation
- Error correction dynamics:
  - The sizable and significant ECM term implies substantial mean-reversion of real wages toward long-run productivity-based equilibrium, with roughly 40 percent correction per year.
- Slack measurement:
  - The nonemployment index gap delivers more explanatory power than headline unemployment, especially for newer member states, reflecting broader labor market slack measures (unemployment rate, inactive labor force, involuntary part-time employment).
- Wage flexibility and adjustment:
  - Notably higher coefficients on slack in newer member states indicate greater real wage flexibility, which can aid adjustment to asymmetric shocks.
- Cross-border spillovers:
  - German wages act as a domestic wage-setting anchor for highly interconnected neighbors (Netherlands, Austria) with long-term coefficients around 0.2.
  - Belgium’s wage-setting is influenced by both French and German wages.
- Structural reforms:
  - Reforms (Hartz in Germany; employment protection and unemployment benefit reforms in other countries) have quantifiable long-term impacts on equilibrium real wages (examples: Germany 5½ percent lowering; Spain 4 percent lowering; Austria 3 percent lowering via employment protection reform).

*Italic: IMF staff calculations; figures and tables quoted are from the chapter’s analysis.*

### 2. EUROpEAN WAGE DyNAMICs AND LAbOR MARKET INTEGRATION

### 2. EUROpEAN WAGE DyNAMICs AND LAbOR MARKET INTEGRATION

### Spillovers and Migration Effects
- Spillovers from euro area labor market developments significantly affect wage growth in the “other euro area” panel and especially in the newer member states.
- In the “other euro area” countries, changes in euro area unemployment (lagged two quarters) and the growth rate in euro area wages (lagged three quarters) are both statistically significant (see column 8 in the source table).
- In the newer member states, changes in unemployment in the euro area (columns 8 and 9) have substantial and significant impacts on wage growth with short lags, implying wages must respond quickly to improving job opportunities in euro area countries to help retain workers.
- Migration effects:
  - A rise in net migration inflows by 1 percentage point (measured as the four-quarter change in net migration inflows in the prior year as a share of the labor force) is estimated to reduce wage growth in the following year by 0.4 percent in the “other euro area” and by 0.5 percent in the newer EU members.
  - The text notes that a 1 percentage point rise is “an exceptionally large shock” and that in practice changes in net migration flows are generally small relative to the labor force, so wage effects are small and temporary.
  - When migration is separated into inflows and outflows, gross migrant outflows have a larger coefficient than inflows, though the difference is not statistically significant, suggesting migrants are not perfect substitutes for the domestic workforce in the near term.

### Short-Term Analysis of Germany
- Key indicators and model properties:
  - Growth in involuntary part-time employment is a key indicator of slack for Germany.
  - The wage Phillips curve for Germany shows high responsiveness to the unemployment gap, trend productivity, and inflation expectations.
  - A nonemployment index was significant, but the coefficient on the growth rate in labor productivity became negative in some specifications.
  - Given a longer sample for Germany, changes in involuntary part-time employment were included separately; the sizable coefficient may reflect widespread use of changes in employee hours to avoid layoffs.
  - The error correction term for Germany is particularly high (about –0.6), and contributes substantially to explanatory power, making the error correction model with changes in involuntary employment the best baseline model.
- Spillovers and migration in Germany:
  - German wage setting appears to be anchored on domestic labor market conditions and less sensitive to spillovers than regional panels.
  - German wages are not found to be responsive to changes in euro area unemployment (column 7 in Table 2.5).
  - The estimated coefficient on euro area wage growth has the opposite sign (column 8) from that expected from a spillover.
  - Changes in net migration flows do not have a statistically significant impact on German wages (column 9).

### Factors Driving Wages within Regions
- Euro area excluding Germany:
  - Inflation and inflation expectations have been key factors behind variations in nominal wage growth, together with domestic and external slack and the error correction term.
  - The initial slowing in wage rises by 2.3 percentage points in 2010 is linked to:
    - lower inflation expectations (0.8 percentage point),
    - lower past inflation (0.5 percentage point),
    - domestic labor slack (0.6 percentage point), and
    - the error correction (0.3 percentage point).
  - The error correction continued to weigh on wages by 0.2 to 0.3 percentage point during 2011–15.
  - During 2011–13, inflation expectations and actual inflation recovered, lifting their combined contribution to wage growth by 0.9 percentage point.
  - Domestic and external slack began to weigh more heavily on wages during 2012–14, with a total drag of 0.6 percentage point on wage growth in both 2013 and 2014.
  - Contributions from inflation expectations and inflation fell sharply during 2014–15 by 0.9 percentage point, with only a slight increase by 2017.
- Newer EU members:
  - Declines in both domestic and foreign slack have contributed to wages picking up in recent years (average results from the eight-country panel).
  - The sharp halt in wage rises in 2009 reflected large negative contributions from rising domestic slack and rising euro area unemployment, plus some error correction drag.
  - By 2011 the declines in domestic slack began to support wages; declines in euro area unemployment began to support wages beginning in 2014.
  - By 2016–17, the low level of domestic slack began to reinforce wage rises.
  - The combination of domestic and external pressures does not fully account for the wage acceleration during 2016–17; migration may have played a part, but recent migration data were not yet available for full analysis.
- Germany:
  - Domestic labor market conditions and inflation expectations have been key factors behind variations in nominal wage growth.
  - The steep initial fall in wage growth by 2.9 percentage points in 2010 is linked to:
    - lower inflation expectations (0.9 percentage point),
    - domestic labor slack (0.3 percentage point), and
    - the error correction (1 percentage point).
  - Wage growth higher than 3 percent in 2012 (despite low inflation and productivity growth) led to a significant error correction drag on wages during 2012–14, peaking at 0.7 percentage point in 2013.
  - Contributions from inflation expectations fell during 2014–15 by 0.4 percentage point, with a slight additional decline in 2017.

### Scenario Analysis: German Wage Shock and Spillovers
- Illustrative scenario:
  - Hourly wage rates in Germany are assumed to rise 1 percentage point faster than in a baseline.
  - This shock would raise euro area wage growth by about 0.4 percentage point given Germany’s weight in the euro area.
  - Spillover impacts:
    - Wage growth in the euro area excluding Germany rises about 0.1 percent after one year and 0.2 percent in the medium term.
    - The impact on newer EU member country wages is about 0.15 percent after one year but fades to zero over time, since in the long term real wages in newer member states depend only on domestic labor productivity.
  - The analysis notes that if higher wages lead to higher inflation (via domestic demand or cost channels), the medium-term impact on wages would be higher than these simulations indicate.

### Impacts of Regional Differences in Wage Formation
- Newer EU members:
  - Greater wage sensitivity to domestic labor slack and external developments helps account for faster wage increases.
  - Coefficients on the nonemployment gap and hours gap are about two times larger in newer members (column 8), and the impact of changes in nonemployment is much larger than in the EU15.
  - Domestic slack’s adverse impact on wage growth after the global financial crisis was much larger in newer member states than in the euro area excluding Germany.
  - In recent years, diminishing domestic slack in newer members contributed an estimated 2 percentage points to wage growth.
  - Combined spillovers from euro area unemployment and euro area wage rises accounted for about 1½ percentage points of the wage pickup in the newer members during 2015–17, compared with less than ½ percentage point in the euro area excluding Germany.
- EU15 (including euro area excluding Germany):
  - Wages typically respond very slowly to changes in unemployment and are closely related to inflation and inflation expectations.
  - Expected and lagged inflation are much more important factors driving wage formation in the EU15 than in the newer members; the coefficient on lagged inflation is much more significant and the contribution from expected inflation is more than two times higher in the EU15.

### Expectations, Wages, and Inflation
- Key policy questions addressed:
  - Has sluggish wage growth in the euro area been due to a flattening of the wage Phillips curve or because wages have become more tied to low inflation expectations?
  - Did wage moderation contribute to low inflation in the euro area, and what are implications for returning inflation to the European Central Bank’s target?
  - In the newer member states, would already strong wage growth accelerate further given tight labor markets, and how much of this wage growth (given productivity) would be passed to inflation?
- Approach:
  - The chapter investigates how the role of slack, inflation, and inflation expectations has changed over time by examining:
    1. How anchored inflation expectations have remained over time;
    2. How the parameters of the baseline Phillips curve have evolved over time;
    3. The extent of pass-through from wage growth to inflation.
  - Separate analyses are performed for the EU15 and the newer EU members because of notable differences in wage behavior.

### Anchoring of Inflation Expectations
- Methodology:
  - Anchoring is assessed by estimating πt^e − π̄ = α + β(πt − π̄) + εt, where πt^e denotes inflation expectations, πt actual inflation, and π̄ the inflation target.
  - Inflation expectations are considered firmly anchored in the short term if they are not systematically sensitive to movements in actual inflation (β = 0).
  - Five-year rolling regressions of the equation are estimated to assess how anchoring has evolved over time.
- Data and assumptions:
  - Consensus Forecasts for inflation are used to measure inflation expectations; two-year-ahead forecasts are used in the baseline, with robustness checks using five-year-ahead forecasts.
  - For the euro area and each member country, the inflation target is taken to be 1.9 percent, in line with the European Central Bank’s definition of price stability.
  - For the Czech Republic, Hungary, Poland, and the Slovak Republic, the target is the rate announced by the central bank or the simple average of announced rates at a given point in time, interpolating linearly where necessary.
  - The source notes market-based inflation expectations are potentially biased and are not available or are unreliable for some newer member states; results for the aggregate euro area are robust to market-based measures where available.

*Source: IMF staff calculations.*

### 2. EUROpEAN WAGE DyNAMICs AND LAbOR MARKET INTEGRATION

### 2. EUROpEAN WAGE DyNAMICs AND LAbOR MARKET INTEGRATION

### Inflation Expectations and Deanchoring
- Rolling regressions indicate an increase in the coefficient on actual inflation for the euro area, implying inflation expectations have become more sensitive to movements in actual inflation (Figure 2.24).  
- Robustness: to address potential endogeneity the equation is estimated using lagged actual inflation; headline results remain robust.  
- Similar results are found for one-year-ahead inflation expectations (estimated at the country level) and five-year-ahead inflation expectations (estimated as an aggregate for the euro area).  
- Conclusion: low inflation since the global financial crisis contributed to some deanchoring of inflation expectations in the euro area.  
- Contrast with newer members (NMS): inflation expectations in newer members seem to have become more anchored over time, with expectations less sensitive to actual inflation than before the global financial crisis; some reversal in the past five years but sensitivity remains quite low.

### Role of Slack, Indexation, and Inflation Expectations for Wage Growth over Time
- Model and estimation:
  - Baseline specification: wt = α + β Xt + εt, where wt denotes nominal wage growth and Xt includes actual inflation, expected inflation, unemployment gap, trend productivity growth, and change in the unemployment rate.  
  - Lags of explanatory variables included consistent with earlier analysis.  
  - Estimation: panel with 15-year rolling windows using quarterly data from the first quarter of 1996 to the third quarter of 2017, estimated separately for the euro area (excluding Germany) and the newer EU member states.
- Slope of the wage Phillips curve:
  - Euro area: estimated coefficient on the unemployment gap starts at about –0.4 and rises only modestly to –0.3 (Figure 2.25).  
  - Newer members: coefficient stable at about –1.  
  - Data source for wage measures: hourly total labor compensation (baseline); results using hourly wages and salaries are similar.  
  - Conclusion: the wage Phillips curve has not flattened significantly since the early 2000s.
- Forward-looking wage setting:
  - Euro area: coefficient on inflation expectations increased from about 0.3 in the late 1990s to about 0.8 (and statistically significant) in recent years (Figure 2.26).  
  - At the same time, wage adjustment in response to actual inflation declined modestly from about one-quarter to one-fifth.  
  - Newer members: coefficient on expected inflation has varied without a clear trend; some decline in the impact of actual inflation in recent years.

### Pass-through from Wages to Inflation
- Methodology:
  - Four-variable VAR comprising: relative import price inflation (ratio of import prices over the GDP deflator), nominal wage growth adjusted for trend productivity growth, consumer price inflation, and an unemployment gap (based on OECD estimates of the NAIRU).  
  - Purpose: assess pass-through of wages (adjusted for productivity) to prices while controlling for endogenous feedback effects with prices and slack.  
  - Estimation: panel VAR separately for the EU15 and the newer members using quarterly data from the first quarter of 1998 to the third quarter of 2017.  
  - Identification: Cholesky decomposition with ordering: import prices, wages, consumer prices, unemployment gap.
- Dynamics and magnitude:
  - Immediate impact of a wage shock on price inflation is positive yet small; impact rises over time, peaks around four to six quarters, then dissipates after about three years (Figure 2.27).  
  - EU15: a wage shock that increases wage growth by 1 percentage point on impact is followed by cumulative increases over three years of 0.6 percent in prices, and 2.4 percent in wages, implying a pass-through of 25 percent.  
  - Newer members: the cumulative increase is 0.9 percent in prices, and 3.6 percent in wages, implying a pass-through of 25 percent.  
  - Conclusion: the medium-term pass-through from wages to prices is positive but less than full in both the EU15 and the newer member states.

### Conclusions and Policy Implications
- Key analytical findings:
  - The wage Phillips curve remains broadly alive with stable parameters: a modest slope in EU15 countries and especially strong wage responses to slack in the newer EU member countries. This contributed to much faster deceleration of wages in the newer members after the global financial crisis and to more recent wage acceleration.  
  - Real wages are anchored by labor productivity, but deviations from equilibrium can develop; incorporating such deviations helps explain wage moderation after major shocks in regions with nominal wage inertia (such as the EU15).  
  - Unemployment remains the main indicator of slack, but monitoring involuntary part-time employment and hours worked per person is useful; a nonemployment index can summarize these plus marginally attached workers.  
  - Cross-country spillovers operate through wage and slack developments; spillovers are especially strong in the newer EU members, possibly reflecting positions in global supply chains and migration effects on domestic labor supply. Migration effects on wages appear statistically significant but small and temporary.  
  - Inflation expectations are a key driver of nominal wage developments, especially in the euro area; some deanchoring has appeared recently owing to prolonged low inflation.
- Model performance and recent developments:
  - Models broadly account for wage trends, but wages in the newer EU member countries are growing somewhat faster than models indicate.  
  - Moderation of nominal wages in the EU15 principally reflects low inflation expectations and slower productivity growth, together with a relatively modest and lagged impact of slack on wages.  
  - Projection implication: provided unemployment continues to fall and inflation gradually increases in line with projections in the April 2018 WEO, wage rises are projected to pick up in the EU15, but the process will take much longer than in the newer EU members.  
  - Factors not fully accounted for in newer member wage acceleration: migration of skilled workers from the newer members, hikes in minimum wages, and public sector wage increases—these have lifted wages relative to labor productivity trends.
- Policy recommendations:
  - Euro area: underpin a firm reanchoring of expectations to the inflation target; ensure the European Central Bank stays committed to its strategy to durably raise inflation to target to promote macroeconomic stability.  
  - Current account rebalancing: wage and real exchange rate adjustments helped but do not fully account for large current account adjustments, which were principally the result of swings in domestic demand financed by capital inflows; there remains room for further unit labor cost rises in Germany and perhaps also in the Netherlands.  
  - Newer EU members: urgency for reforms to reduce skill mismatches and support labor force participation given recent wage acceleration:
    - Mobilize labor supply through measures to reduce skill and locational mismatches to contain wage pressure and support employment and growth.  
    - Active labor market policies to increase participation rates and reduce structural unemployment are needed given unfavorable demographics and emigration.  
    - Expand retraining courses for the unemployed and apprenticeship systems to develop necessary skills and alleviate skill mismatches.  
    - Structural reforms to strengthen institutions and improve public sector efficiency would help convergence and encourage potential emigrants to stay.

*Source: IMF staff calculations.*

### Box 2.1. Euro Area Wage Developments and External Rebalancing

### Box 2.1. Euro Area Wage Developments and External Rebalancing

### Unit labor costs and competitiveness
- Unit labor costs are defined as the ratio of wages and salaries per hour (national accounts) to the trend in hourly labor productivity from a Hodrick-Prescott filter.
- The analysis uses unit labor cost–based real effective exchange rates, with the historical average as the base (averaged over 2000–16) to allow for differences in competitiveness at a time of entry to the euro area without implying a long-term equilibrium exchange rate.
- Relative unit labor cost developments are presented both relative to the euro area average and on a multilateral unit labor cost basis.

### Elasticity estimates and implied current account adjustments
- An estimated CA/REER elasticity (current account response to real effective exchange rate changes) with a median of 0.33 for the euro area countries identified is used.
- Applying this elasticity to either (1) the change in relative unit labor costs relative to the euro area average or (2) the change in the multilateral real effective exchange rate on a unit labor cost basis generates implied current account adjustments that are well short of those observed in countries with large adjustments.

### Role of capital inflows and the adjustment process
- The sharp adjustment in external balances within the euro area is consistent with a sudden stop in capital inflows driving the adjustment.
- Pre-global-financial-crisis external deficits were to a large extent the result of capital inflows driving aggregate demand via credit growth and/or fiscal deficits.
- When inflows ended (as a result of the global and euro area crises and the unwinding of distorted asset prices, risk premiums, and private sector expectations), saving and investment shifted in a manner that increased the current account balance consistent with full employment.
- Labor market slack arising during the adjustment supported a decline in wages relative to productivity and hence declines in unit labor costs; these relative declines were supportive of external adjustment but were not the primary driver.

### Country-specific assessments and policy implications
- Greece and Portugal: external balances are now at or above their External Balance Assessment norms.
- Spain: in surplus yet below its External Balance Assessment norm.
- Greece, Portugal, and Spain: have large negative international investment positions, underscoring the importance of aiming for higher productivity growth to boost competitiveness.
- Germany and the Netherlands: continued sizable surpluses call for policy adjustments to promote a greater balance of saving and investment.
  - Recommended policies include supporting continued relative gains in unit labor costs in Germany and a shift toward relative increases in the Netherlands.

*International Monetary Fund | Box 2.1. Euro Area Wage Developments and External Rebalancing (from Regional Economic Outlook: Europe, May 2018).*

---


_Source: https://www.imf.org/-/media/files/publications/reo/eur/2018/may/eur-chapter-2.pdf_