## Appendix 1. Model Predictions

## Source details

**Canonical URL:** [Appendix 1. Model Predictions](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11109.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11109.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11109.pdf.json)

---

### Key findings and research questions
- Paper investigates dispersion in RULC growth rates across euro area countries with a focus on the EMU period.
- Three central questions:
  - How big are RULC growth differentials in EMU by historical perspective?
  - What are the reasons underlying such differentials and why does growth dispersion remain after the start of EMU?
  - How can policy narrow such differentials?
- Empirical answer highlighted: RULC growth divergence persisted despite EMU and, in the run-up to the crisis, differentials persisted or even widened.
- Normative assumption: RULC should display no growth under equilibrium conditions (the "zero-growth rule"); deviations reflect imperfect competition and adjustment dynamics.

### Empirical facts on prices, NULC and RULC
- Convergence in GDP deflator inflation halted during the EMU decade.
- Cross-country standard deviation of 5-year cumulated changes in relative GDP deflator:
  - fell from around 30 percent in the mid-1980s to 9 percent in 1999 and 3.5 percent in 2009.
- NULC dispersion:
  - decreased until 2003 but rose in the run up to the crisis.
  - Germany: growth rates in NULC remained well below the euro area average.
  - Spain, Greece and Ireland: NULC experienced significant increases relative to the average.
- RULC:
  - expected to be stable in the long run but display significant medium-run deviations from zero;
  - dispersion in RULC growth differentials has not been markedly reduced since the launch of the euro.
- Adjustment under a single currency and low inflation can be painful (example: 2009 downward adjustment in RULC in GIIPS associated with public wage cuts and abrupt unemployment increases).

### Theoretical model — structure and drivers
- Technology and structure:
  - CES production technology with capital K, labor L, total factor productivity A, share parameter α, elasticity of substitution η (0 < η < ∞).
  - Under perfect competition, RULC relate monotonically to capital-output ratio Y/K (equation (2)).
- Key model implications:
  - If factor elasticity of substitution η < 1, capital deepening (higher K/Y) is associated with increasing RULC.
  - Introduction of intermediate inputs alters the RULC — capital-output relationship through the relative price of intermediates pI/p̃ (equation (5)).
  - Imperfect competition: firms set prices with a markup μ; RULC depend on μ and bargaining power β (equations (6)–(9)).
  - Labor adjustment costs (hiring/firing restrictions) raise the marginal cost of labor and depress equilibrium employment and RULC (equations (10)–(11)).
- General rule:
  - RULC in presence of multiple factors given in equation (12), combining capital-output ratio, intermediate input prices, markup μ, bargaining power β, and adjustment cost terms δ(·).

### Model predictions (explicit propositions)
- Proposition 1:
  - Capital accumulation increases RULC if η < 1 (i.e., (13) 0 if 1/ RULC KY η ∂ > < ∂ ).
- Proposition 2:
  - RULC increases when the relative price of intermediates rises if η < 1 (i.e., (14) 0 if 1 I RULC pp η ∂ > < ∂ ).
- Proposition 3 (institutional factors):
  - Direction of change of RULC in response to institutional changes is independent of η.
  - Proposition 3.1: RULC are a decreasing function of firms’ markup μ and an increasing function of workers’ bargaining power β (statement (15): 0,0, for any 0 RULC RULC η μ β ∂∂ < > > ∂∂ ).
  - Proposition 3.2: RULC are a decreasing function of labor marginal adjustment costs under convex adjustment costs (statement (16): conditions on δ' and δ'' with sign 0).
- Graphical characterization illustrated in Figure 3 (Real Unit Labor Costs and the Capital-Output Ratio).

### Empirical strategy and data
- Empirical specification:
  - log-linear panel model (equation (17)) where first differences of relative RULC are regressed on:
    - Technological drivers: relative capital-output ratio (K/Y) and relative price of intermediate inputs (pI/p̃) from EU KLEMS.
    - External drivers: nominal effective exchange rates (NEER) vis-à-vis intra-area and extra-area groups (NEER_EA, NEER_12, NEER_29 depending on sample).
    - Institutional drivers: bargaining centralization (barcent), replacement rate of unemployment benefits (rr), openness (imports+exports/GDP), employment protection legislation (eplreg), product market regulation (pmr), and output volatility (std dev of output growth).
    - Controls: relative unemployment gap (based on NAWRU from AMECO) and the lagged level of relative RULC to test convergence (expect negative coefficient).
  - All variables expressed in (log) first-difference form and relative to a weighted average of remaining euro area partners (emu-i); country fixed effects included.
  - Institutional variables standardized (zero mean, unit standard deviation) so coefficients interpreted as effect of a one-standard-deviation change.
  - Trade-weighted emu-i averages use time-varying weights reflecting each country’s trade composition.

### Empirical evidence linking technology and RULC
- Relative capital-output ratios are strongly and positively correlated with relative RULC (Figure 4): co-movements point to a value of η lower than unity (consistent with Proposition 1 and 2).

### Estimation approach and robustness checks
- Equation (17) initially estimated using Generalized Least Squares (GLS) on a balanced panel of 11 euro-area members (Austria, Belgium, Germany, Greece, Finland, France, the Netherlands, Ireland, Italy, Portugal and Spain).
- Sensitivity analyses:
  - Exclusion of individual countries (Table 2, Appendix 3): technological, external factors, product market regulation and employment protection legislation largely robust; some institutional factors show less stability.
  - Split-sample to assess shift to EMU: separate estimates for 1996-2008.
  - Re-estimation using Generalized Instrumental Variables (GIV): macroeconomic variables instrumented with a minimum of two lags, institutional variables with one- and two-period lags.

### Main empirical findings
- Dynamics and persistence:
  - RULC growth differentials converge to zero over the very long run; convergence is relatively slow (lagged endogenous coefficient small in absolute value).
  - Short-term response: relative RULC decrease in response to a depressing labor market; estimates imply a reduction of about 0.3 percent in relative RULC for each point of additional unemployment gap differential with EMU.
  - Example interpretation: a value of -0.14 for the lag coefficient implies that it takes about five years for RULC to absorb half the size of a shock.
- Technological and external determinants:
  - Higher capital accumulation and increases in the price of intermediates both lead to relatively higher growth in RULC.
  - Implied elasticity of substitution from estimates: around 0.7; a one-percent relative increase in the capital-output ratio implies a rise of 0.4 percent in RULC relative to euro area partners.
  - External factors (degree of openness and nominal value of the euro) have a non-negligible impact.
- Institutional determinants and channels:
  - Stringent product market regulation is associated with relatively lower RULC.
  - Wage formation indicators:
    - Higher replacement rates increase relative RULC.
    - Higher bargaining centralization tends to deliver relatively higher employment and thus higher RULC.
    - Higher degree of openness (tradable sector) puts downward pressure on RULC.
  - Employment protection legislation (EPL): more rigid employment regulations shrink employment bringing down RULC; negative effect on job creation magnified by uncertainty (interaction EPL × volatility).

### Key regression coefficients (Dependent variable: Dlog RULC)
- GLS 1980-2008 (Total panel (balanced) observations: 319; S.E. of regression: 1.01; R-squared (weighted): 0.48; R-squared (unweighted): 0.33)
  - l og RULC(-1): -0.136 (5.64)***
  - Unemployment gap: -0.336 (4.49)***
  - ∆log NEER_EA: 0.104 (5.37)***
  - ∆log NEER12: 0.009 (3.31)***
  - ∆ Degree of openness: -0.048 (2.57)**
  - ∆log Capital-output ratio: 0.446 (10.68)***
  - ∆log Relative price of intermediate inputs: 0.084 (2.30)**
  - ∆ Product market regulation(-1): -0.463 (3.17)***
  - ∆ Employment protection legislation: -0.814 (1.83)*
  - ∆ Unemployment replacement rate(-1): 0.900 (2.19)**
  - ∆ Bargaining centralization: 1.271 (1.87)*
  - Interaction ∆ EPL and volatility of output growth: -0.119 (1.82)*
  - AR(1): 0.205 (3.35)***

- GLS 1996-2008 (Total panel (balanced) observations: 143; S.E. of regression: 1.06; R-squared (weighted): 0.96; R-squared (unweighted): 0.50)
  - l og RULC(-1): -0.183 (15.54)***
  - Unemployment gap: -0.186 (6.08)***
  - ∆log NEER_EA: 0.223 (8.08)***
  - ∆log NEER12: 0.095 (5.09)***
  - ∆ Degree of openness: 0.060 (6.21)***
  - ∆log Capital-output ratio: 0.771 (28.70)***
  - ∆log Relative price of intermediate inputs: 0.142 (5.70)***
  - ∆ Product market regulation(-1): -0.714 (8.32)***
  - ∆ Employment protection legislation: 0.421 (1.81)*
  - ∆ Unemployment replacement rate(-1): 0.350 (1.74)*
  - ∆ Bargaining centralization: 5.841 (12.28)***
  - Interaction ∆ EPL and volatility of output growth: -0.519 (8.43)***
  - AR(1): -0.107 (2.81)**

- GIV 1980-2008 (Total panel (balanced) observations: 319; S.E. of regression: 1.02; R-squared (weighted): 0.48; R-squared (unweighted): 0.13)
  - l og RULC(-1): -0.125 (5.86)***
  - Unemployment gap: -0.338 (2.46)**
  - ∆log NEER_EA: 0.199 (4.28)***
  - ∆log NEER12: 0.047 (10.49)***
  - ∆ Degree of openness: -0.137 (2.62)***
  - ∆log Capital-output ratio: 0.648 (7.09)***
  - ∆log Relative price of intermediate inputs: 0.528 (4.34)***
  - ∆ Product market regulation(-1): -0.725 (2.76)***
  - ∆ Employment protection legislation: 0.676 (0.62)
  - ∆ Unemployment replacement rate(-1): 1.266 (2.42)**
  - ∆ Bargaining centralization: 3.664 (1.64)*
  - Interaction ∆ EPL and volatility of output growth: -0.719 (3.50)***
  - AR(1): 0.0 (AR restricted to zero; AR insignificant)

(Note: absolute value of t-statistics in parentheses; superscripts *, **, *** indicate significance at 10, 5 and 1 percent levels, respectively. GLS corrects for cross-section heteroskedasticity, contemporaneous correlation, and assumes an AR(1) error autocorrelation structure. In GIV estimation institutional variables instrumented with one and two-period lags; macro variables with minimum two lags.)

### Contribution of factors to RULC differentials (period 2000-08; coefficients from 1996-2008 estimates)
- Technological factors (capital-output ratio and relative price of intermediate inputs): 46 percent of the total deviation explained by the model.
- Lagged dependent variable: 24 percent.
- External factors (degree of openness and nominal value of the euro): 9 percent.
- Relative unemployment gap: 7 percent.
- Product and labor market regulations: 14 percent.

### Robustness to EMU period and sensitivity
- Shift to EMU generally preserves sign of coefficients but increases sensitivity of differentials to fundamentals.
- Degree of inertia slightly reduced during EMU years; capital-output ratio and price of materials more influential; sensitivity to extra-NEER increased.
- Some institutional coefficients (product market regulation, bargaining centralization, EPL×volatility) larger post-EMU; unemployment-gap sensitivity reduced post-EMU.
- Appendix III (Table 2) reports country-dropping GLS estimates: coefficient signs and significance generally robust; extra-area NEER and some institutional terms show greater instability across reduced samples.

### Policy-relevant interpretation and policy implications
- Continued dispersion in RULC differentials is associated with:
  - Increased sensitivity of RULC to fundamentals after the shift in monetary regime (EMU).
  - Divergent evolutions of fundamentals across countries, especially capital-output ratios, nominal effective exchange rates, and country-specific institutional features.
- Policy implications:
  - Giving up national monetary policy makes a country’s RULC growth gap with EMU more sensitive to country-specific structural features.
  - Devise reforms to limit pro-cyclicality of credit supply; ensure bank capital requirements reflect country differences in asset price overvaluation; review structural housing market features that increase bubble risk (e.g., tax incentives for mortgages).
  - Openness and specialization upgrades in peripheral countries should help reduce dispersion in the medium term.
  - Greater coordination or convergence of labor and product market policies can contribute to narrowing RULC differentials; institutional heterogeneity in product and labor market regulations amplifies RULC dispersion.

### Appendix highlights
- Appendix I: Model comparative statics derive signs of derivatives:
  - ∂RULC/∂(Y/K) and ∂RULC/∂(relative price of intermediates) depend on elasticity of substitution parameters; under η < 1, capital accumulation raises RULC.
  - ∂RULC/∂(markup) negative under stated conditions; ∂RULC/∂(bargaining power) positive.
  - Labor adjustment costs enter negatively under convexity assumptions.
- Appendix II: Computation linking relative RULC, relative profit margins, and the factor elasticity of substitution; empirical computation of factor elasticity of substitution feasible via augmented regressions.
- Appendix III (Table 2): country-dropping GLS estimates — robustness of coefficient signs and significance; extra-area NEER and some institutional terms show greater instability across reduced samples.

*Source: _wp11109 - Appendix 1. Model Predictions (PDF chapter/section)*

### Appendix 1. Model Predictions ..........................................................................................

### Appendix 1. Model Predictions ..........................................................................................

### Appendix listings and locations
- Appendix 1. Model Predictions .......................................................................................... 21
- Appendix 2. Computing the Factor Elasticity of Substitution from Estimation Results .... 23
- Appendix 3. Sensitivity vis-à-vis the Country Sample ....................................................... 24

*Source: _wp11109 - Appendix 1. Model Predictions (PDF chapter/section)*

### References .............................................................................................................

### Real Unit Labor Costs in the Euro Area (excerpt)

### Key findings and questions
- Paper investigates dispersion in RULC growth rates across euro area countries with a focus on the EMU period.
- Three central questions:
  - How big are RULC growth differentials in EMU by historical perspective?
  - What are the reasons underlying such differentials and why does growth dispersion remain after the start of EMU?
  - How can policy narrow such differentials?
- Empirical answer highlighted: RULC growth divergence persisted despite EMU and, in the run-up to the crisis, differentials persisted or even widened.
- Normative assumption: RULC should display no growth under equilibrium conditions (the "zero-growth rule"); deviations reflect imperfect competition and adjustment dynamics.

### Empirical facts on prices, NULC and RULC
- Convergence in GDP deflator inflation halted during the EMU decade.
- Cross-country standard deviation of 5-year cumulated changes in relative GDP deflator:
  - fell from around 30 percent in the mid-1980s to 9 percent in 1999 and 3.5 percent in 2009.
- NULC dispersion decreased until 2003 but rose in the run up to the crisis:
  - Germany: growth rates in NULC remained well below the euro area average.
  - Spain, Greece and Ireland: NULC experienced significant increases relative to the average.
- RULC: expected to be stable in the long run but display significant medium-run deviations from zero; dispersion in RULC growth differentials has not been markedly reduced since the launch of the euro.
- Adjustment under a single currency and low inflation can be painful (example: 2009 downward adjustment in RULC in GIIPS associated with public wage cuts and abrupt unemployment increases).

### Theoretical model — structure and drivers
- Starting assumptions:
  - CES production technology with capital K, labor L, total factor productivity A, share parameter α, elasticity of substitution η (0 < η < ∞).
  - Under perfect competition, RULC relate monotonically to capital-output ratio Y/K (equation (2)).
- Key model implications:
  - If factor elasticity of substitution η < 1, capital deepening (higher K/Y) is associated with increasing RULC.
  - Introduction of intermediate inputs alters the RULC — capital-output relationship through the relative price of intermediates pI/p̃ (equation (5)).
  - Imperfect competition: firms set prices with a markup μ; RULC depend on μ and bargaining power β (equations (6)–(9)).
  - Labor adjustment costs (hiring/firing restrictions) raise the marginal cost of labor and depress equilibrium employment and RULC (equations (10)–(11)).
- General rule for RULC in presence of multiple factors given in equation (12), combining capital-output ratio, intermediate input prices, markup μ, bargaining power β, and adjustment cost terms δ(·).

### Model predictions (explicit propositions)
- Proposition 1:
  - Capital accumulation increases RULC if η < 1 (i.e., (13) 0 if 1/ RULC KY η ∂ > < ∂ ).
- Proposition 2:
  - RULC increases when the relative price of intermediates rises if η < 1 (i.e., (14) 0 if 1 I RULC pp η ∂ > < ∂ ).
- Proposition 3 (institutional factors):
  - Direction of change of RULC in response to institutional changes is independent of η.
  - Proposition 3.1: RULC are a decreasing function of firms’ markup μ and an increasing function of workers’ bargaining power β (statement (15): 0,0, for any 0 RULC RULC η μ β ∂∂ < > > ∂∂ ).
  - Proposition 3.2: RULC are a decreasing function of labor marginal adjustment costs under convex adjustment costs (statement (16): conditions on δ' and δ'' with sign 0).
- Graphical characterization illustrated in Figure 3 (Real Unit Labor Costs and the Capital-Output Ratio).

### Empirical strategy and data
- Empirical specification is a log-linear panel model (equation (17)) where first differences of relative RULC are regressed on:
  - Technological drivers: relative capital-output ratio (K/Y) and relative price of intermediate inputs (pI/p̃) from EU KLEMS.
  - External drivers: nominal effective exchange rates (NEER) vis-à-vis intra-area and extra-area groups (NEER_EA, NEER_12, NEER_29 depending on sample).
  - Institutional drivers: bargaining centralization (barcent), replacement rate of unemployment benefits (rr), openness (imports+exports/GDP), employment protection legislation (eplreg), product market regulation (pmr), and output volatility (std dev of output growth).
  - Controls: relative unemployment gap (based on NAWRU from AMECO) and the lagged level of relative RULC to test convergence (expect negative coefficient).
- All variables are expressed in (log) first-difference form and relative to a weighted average of remaining euro area partners (emu-i); country fixed effects included.
- Institutional variables are standardized (zero mean, unit standard deviation) so coefficients can be interpreted as the effect of a one-standard-deviation change.
- Trade-weighted emu-i averages use time-varying weights reflecting each country’s trade composition.

### Empirical evidence linking technology and RULC
- Relative capital-output ratios are strongly and positively correlated with relative RULC (Figure 4): co-movements point to a value of η lower than unity (consistent with Proposition 1 and 2).

### Policy-relevant interpretation
- Continued dispersion in RULC differentials is associated with:
  - Increased sensitivity of RULC to fundamentals after the shift in monetary regime (EMU).
  - Divergent evolutions of fundamentals across countries, especially capital-output ratios, nominal effective exchange rates, and country-specific institutional features.
- Policy implication highlighted:
  - Giving up national monetary policy makes a country’s RULC growth gap with EMU more sensitive to country-specific structural features.
  - Since divergences are partly related to heterogeneous regulatory options, policy action seeking more homogenous regulation (labor and product market regulations, wage-setting institutions) can contribute to narrowing RULC differentials.

*Source: IMF Working Paper (excerpt). *

### 2008. Equation (17) is initially estimated using Generalized Least Squares (GLS). To check

### _wp11109 - 2008. Equation (17) is initially estimated using Generalized Least Squares (GLS). To check

### Estimation approach and robustness checks
- Equation (17) is initially estimated using Generalized Least Squares (GLS) on a balanced panel of 11 euro-area members (Austria, Belgium, Germany, Greece, Finland, France, the Netherlands, Ireland, Italy, Portugal and Spain).
- Sensitivity analyses:
  - Exclusion of individual countries (see Table 2, Appendix 3) — technological, external factors, product market regulation and employment protection legislation are largely robust; some institutional factors show less stability.
  - Split-sample to assess shift to EMU: separate estimates for 1996-2008 (second column of Table 1).
  - Re-estimation using Generalized Instrumental Variables (GIV) to address potential inconsistency from GLS and endogeneity; macroeconomic variables instrumented with a minimum of two lags, institutional variables with one- and two-period lags.

### Main empirical findings
- Dynamics and persistence:
  - RULC growth differentials converge to zero over the very long run; convergence is relatively slow (lagged endogenous coefficient small in absolute value).
  - Short-term response: relative RULC decrease in response to a depressing labor market; estimates imply a reduction of about 0.3 percent in relative RULC for each point of additional unemployment gap differential with EMU.
  - Example interpretation: a value of -0.14 for the lag coefficient implies that it takes about five years for RULC to absorb half the size of a shock.
- Technological and external determinants:
  - Higher capital accumulation and increases in the price of intermediates both lead to relatively higher growth in RULC.
  - Implied elasticity of substitution from estimates: around 0.7; a one-percent relative increase in the capital-output ratio implies a rise of 0.4 percent in RULC relative to euro area partners.
  - External factors (degree of openness and nominal value of the euro) have a non-negligible impact.
- Institutional determinants and channels:
  - Stringent product market regulation is associated with relatively lower RULC.
  - Wage formation indicators:
    - Higher replacement rates increase relative RULC.
    - Higher bargaining centralization tends to deliver relatively higher employment and thus higher RULC.
    - Higher degree of openness (tradable sector) puts downward pressure on RULC.
  - Employment protection legislation (EPL): more rigid employment regulations shrink employment bringing down RULC; negative effect on job creation magnified by uncertainty (interaction EPL × volatility).
- Robustness to EMU period:
  - Shift to EMU generally preserves sign of coefficients but increases sensitivity of differentials to fundamentals.
  - Degree of inertia slightly reduced during EMU years; capital-output ratio and price of materials more influential; sensitivity to extra-NEER increased.
  - Some institutional coefficients (product market regulation, bargaining centralization, EPL×volatility) larger post-EMU; unemployment-gap sensitivity reduced post-EMU.

### Key regression coefficients (Table 1) — Dependent variable: Dlog RULC
- Column: GLS 1980-2008
  - l og RULC(-1): -0.136 (5.64)***
  - Unemployment gap: -0.336 (4.49)***
  - ∆log NEER_EA: 0.104 (5.37)***
  - ∆log NEER12: 0.009 (3.31)***
  - ∆ Degree of openness: -0.048 (2.57)**
  - ∆log Capital-output ratio: 0.446 (10.68)***
  - ∆log Relative price of intermediate inputs: 0.084 (2.30)**
  - ∆ Product market regulation(-1): -0.463 (3.17)***
  - ∆ Employment protection legislation: -0.814 (1.83)*
  - ∆ Unemployment replacement rate(-1): 0.900 (2.19)**
  - ∆ Bargaining centralization: 1.271 (1.87)*
  - Interaction ∆ EPL and volatility of output growth: -0.119 (1.82)*
  - AR(1): 0.205 (3.35)***
  - S.E. of regression: 1.01
  - R-squared (weighted): 0.48
  - R-squared (unweighted): 0.33
  - Total panel (balanced) observations: 319
- Column: GLS 1996-2008
  - l og RULC(-1): -0.183 (15.54)***
  - Unemployment gap: -0.186 (6.08)***
  - ∆log NEER_EA: 0.223 (8.08)***
  - ∆log NEER12: 0.095 (5.09)***
  - ∆ Degree of openness: 0.060 (6.21)***
  - ∆log Capital-output ratio: 0.771 (28.70)***
  - ∆log Relative price of intermediate inputs: 0.142 (5.70)***
  - ∆ Product market regulation(-1): -0.714 (8.32)***
  - ∆ Employment protection legislation: 0.421 (1.81)*
  - ∆ Unemployment replacement rate(-1): 0.350 (1.74)*
  - ∆ Bargaining centralization: 5.841 (12.28)***
  - Interaction ∆ EPL and volatility of output growth: -0.519 (8.43)***
  - AR(1): -0.107 (2.81)**
  - S.E. of regression: 1.06
  - R-squared (weighted): 0.96
  - R-squared (unweighted): 0.50
  - Total panel (balanced) observations: 143
- Column: GIV 1980-2008
  - l og RULC(-1): -0.125 (5.86)***
  - Unemployment gap: -0.338 (2.46)**
  - ∆log NEER_EA: 0.199 (4.28)***
  - ∆log NEER12: 0.047 (10.49)***
  - ∆ Degree of openness: -0.137 (2.62)***
  - ∆log Capital-output ratio: 0.648 (7.09)***
  - ∆log Relative price of intermediate inputs: 0.528 (4.34)***
  - ∆ Product market regulation(-1): -0.725 (2.76)***
  - ∆ Employment protection legislation: 0.676 (0.62)
  - ∆ Unemployment replacement rate(-1): 1.266 (2.42)**
  - ∆ Bargaining centralization: 3.664 (1.64)*
  - Interaction ∆ EPL and volatility of output growth: -0.719 (3.50)***
  - AR(1): 0.0 (AR restricted to zero; AR insignificant)
  - S.E. of regression: 1.02
  - R-squared (weighted): 0.48
  - R-squared (unweighted): 0.13
  - Total panel (balanced) observations: 319

(Note: absolute value of t-statistics in parentheses; superscripts *, **, *** indicate significance at 10, 5 and 1 percent levels, respectively. GLS corrects for cross-section heteroskedasticity, contemporaneous correlation, and assumes an AR(1) error autocorrelation structure. In GIV estimation institutional variables instrumented with one and two-period lags; macro variables with minimum two lags.)

### Contribution of factors to RULC differentials (period 2000-08; coefficients from 1996-2008 estimates)
- Technological factors (capital-output ratio and relative price of intermediate inputs): 46 percent of the total deviation explained by the model.
- Lagged dependent variable: 24 percent.
- External factors (degree of openness and nominal value of the euro): 9 percent.
- Relative unemployment gap: 7 percent.
- Product and labor market regulations: 14 percent (relatively modest but potentially persistent absent institutional change).

### Conclusions and policy implications
- EMU launch did not reduce dispersion in RULC growth differentials; persistent divergences in firms’ profitability across euro area members.
- Capital accumulation is the most important driver of divergences in RULC; capital accumulation financed by loose credit conditions (especially in construction) contributed to higher relative RULC.
  - Policy implication: devise reforms to limit pro-cyclicality of credit supply; ensure bank capital requirements reflect country differences in asset price overvaluation; review structural housing market features that increase bubble risk (e.g., tax incentives for mortgages).
- Increased responsiveness of RULC differentials to nominal competitiveness with the rest of the world (extra-area NEER) implies that openness and specialization upgrades in peripheral countries should help reduce dispersion in the medium term.
- Institutional heterogeneity in product and labor market regulations amplifies RULC dispersion; without greater coordination of labor and product market policies, institutional disparities will continue to impair RULC convergence.
- Giving up national monetary policy (EMU) makes a country’s RULC growth gap with EMU more sensitive to country-specific structural features.

### Appendix highlights
- Model comparative statics (Appendix I) derive signs of derivatives:
  - ∂RULC/∂(Y/K) and ∂RULC/∂(relative price of intermediates) depend on the elasticity of substitution parameters; under η < 1, capital accumulation raises RULC.
  - ∂RULC/∂(markup) negative under stated conditions; ∂RULC/∂(bargaining power) positive.
  - Labor adjustment costs enter negatively under convexity assumptions.
- Appendix II shows computation linking relative RULC, relative profit margins, and the factor elasticity of substitution; empirical computation of factor elasticity of substitution is feasible via augmented regressions.
- Appendix III (Table 2) reports country-dropping GLS estimates: coefficient signs and significance generally robust; extra-area NEER and some institutional terms show greater instability across reduced samples.

*Italic source: _wp11109 - 2008. Equation (17) is initially estimated using Generalized Least Squares (GLS). To check*

### REFERENCES

### _wp11109 - REFERENCES

### Labour income share and labour market dynamics
- Arpaia, A. Pérez, E., and K. Pichelmann, 2009, ―Understanding Labour Income Share Dynamics in Europe,‖ European Economy, Economic papers No. 379, European Commission, Brussels.  
- Bentolila, S., and G. Saint-Paul, 2003, ―Explaining Movements in the Labor Share,‖ Contributions to Macroeconomics, Berkeley Electronic Press, Vol. 3, No. 1, pp. 1103-1103.  
- Bassanini, A., and R. Duval, 2006, ―Employment Patterns in OECD Countries: Reassessing the Role of Policies and Institutions‖, OECD Social, Employment and Migration Working Papers No. 35, OECD.  
- Garibaldi P., and P. Mauro, 2002, ―Anatomy of Employment Growth,‖ Economic Policy, Vol. 17(34), pp. 67–114.  
- Guichard S., and E. Rusticelli, 2010. ―Assessing the Impact of the Financial Crisis on Structural Unemployment in OECD Countries,‖ Economics Department Working Paper, No. 767, Paris: OECD.  
- Furceri D., and A. Mourougane, 2009, ―How Do Institutions Affect Structural Unemployment in Times of Crisis?‖ Economics Department Working Paper, No. 730, Paris: OECD.  
- Nickell, S., and L. Nunziata, 2001. ―Labour Market Institutions Database.‖  
- Visser, J., 2007, ―Data Base on Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts in 26 countries between1960 and 2006 (ICTWSS),‖ University of Amsterdam.

### Wage bargaining, institutions, and policy
- Calmfors, L. and J. Driffil, 1988, ―Centralization of Wage Bargaining and Macroeconomic Performance,‖ Economic Policy, No. 6, pp. 13-61.  
- Sapir A., 2006, ―Globalization and the Reform of European Social Models,‖ Journal of Common Market Studies, Vol. 44(2), pp 369–390.  
- Martin J., 1996, ―Measures of Replacement Rates for the Purpose of International Comparisons: A Note,‖ OECD Economic Studies, No. 26.  
- OECD, 2004, OECD Employment Outlook: Reassessing the OECD Jobs Strategy, Paris.  
- OECD, 2009, OECD Employment Outlook: Tackling the job crisis, Paris.

### Macroeconomic theory, markups, inflation, and stagflation
- Blanchard, O., 1997, ―The Medium Run,‖ Brookings Papers on Economic Activity, Economic Studies Program, The Brookings Institution, Vol. 28(1997-2), pp. 89-158.  
- Rotemberg J.J., and M. Woodford, 1991, ―Markups and the Business Cycle,‖ NBER Chapters, in: NBER Macroeconomics Annual 1991, Vol. 6, pp. 63-140 National Bureau of Economic Research, Inc.  
- Bruno, M., and J. Sachs, 1985, Economics of Worldwide Stagflation, Cambridge, Mass., Harvard University Press.  
- Honohan, P. and P.R. Lane, 2003. ―Divergent Inflation Rates in EMU,‖ Economic Policy, Vol.37, 359-394.  

### Euro area, external imbalances, and industry productivity
- Chen, R., Milesi-Ferreti, G.M and Tressel, T., 2011, ―Euro Area    Debtor Countries: External Imbalances in the Euro Area,‖ IMF Working Papers, forthcoming.  
- O’Mahony, M., and M. P. Timmer, 2009, ―Output, Input and Productivity Measures at the Industry Level: the EU KLEMS Database,‖ Economic Journal, 119(538), pp. F374-F403.

*Source: _wp11109 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11109.pdf_