## _wp14114 - APPENDIX A

## Source details

**Canonical URL:** [_wp14114 - APPENDIX A](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14114.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14114.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14114.pdf.json)

---

### I. Introduction
- Context and policy:
  - France has a high labor tax wedge historically offset by subsidies and tax breaks (Balladur and Juppé reforms in the 1990s; Aubry and Fillon tax breaks in the 2000s; more recent tax credits proposed by President Hollande).
  - Latest initiative: a cut in the tax wedge worth EUR 30 billion over 2013-17 aimed at reversing deteriorating employment and competitiveness.
  - Major shares are channeled through the Crédit d’impôt pour la compétitivité et l’emploi (CICE) and the Responsibility and Solidarity Pact (RSP); intended to reduce labor costs for wages up to 2.5 to 3.5 times the minimum wage.
- Purpose of the analysis:
  - Calibrate a basic supply-side general equilibrium model for the French economy.
  - Abstract from demand-side effects to focus on effects on potential output and potential employment induced by these measures.
  - Emphasis on skilled versus unskilled trade-offs and alternative financing schemes.
  - Simulations reported for short run (prior to any capital response) and long run (after full capital adjustment). No specific time frame provided.
- Headline model results (reported):
  - Short run: around 200,000 jobs could be created by a EUR 30bn reduction in employers’ social security contributions.
  - Long run: gains could reach 600,000 jobs.
- Targeting summary:
  - Tax cuts targeted to low wage workers have 2.3 times as large an effect on employment as untargeted reductions for equal budgetary cost (1.5 times as large in the long run).

### II. History of labor tax cuts in France
- Minimum wage role and offsets:
  - Minimum wages ensure a living wage, reduce social exclusion, act as efficiency wages, and reduce income inequality.
  - Since 1993, France has reduced social security contributions especially at the minimum wage to offset high labor costs.
- Major episodes and targeting evolution:
  - 1993—1998: Balladur and Juppé reforms reduced labor costs for wages up to 1.2—1.33 times the SMIC.
  - 1998—2002 and 2003—2005: Aubry and Fillon tax breaks increased reductions from 18.2 to 26 percent of gross wages for the SMIC, converging toward zero for wages at 1.6 times the SMIC.
  - 2007—2012: marginal adjustments; July 2007 tax cuts at the SMIC level increased to 28.1 percent of gross wages for firms with fewer than 20 employees.
  - 2013 onwards: President Hollande proposed EUR 30 billion over 2013-2017; two thirds via CICE to reduce labor compensation for wages up to 2.5 times the SMIC (80 percent of workers) by 2015; remaining EUR 10 billion to target workers up to around 3.5 times the minimum wage.
- Fiscal cost and progressivity:
  - Fiscal cost in 2009: around EUR 22.2 billion (1.2 percent of GDP); almost half due to measures to counter the cost of moving to the 35-hour week.
  - Recent labor cost reductions committed under the CICE and the RSP imply EUR 30 billion (1.5 percent of GDP) of additional labor tax cuts over 2013-17.
  - Tax wedge progressivity: social charges start at 8.2 points for the SMIC and reach 42.3 at 2.5 times the SMIC.

### III. Analytical framework — model overview
- Structure and key assumptions:
  - One-sector, four-factor general equilibrium model with capital, skilled and unskilled labor, involuntary unemployment, and a productive public good.
  - Open, price-taking economy with full capital mobility. Goods and capital markets perfectly competitive; labor market subject to frictions.
  - Unskilled labor supply infinitely elastic at the minimum wage (wages fixed at SMIC for small changes); skilled wages follow a wage curve negatively related to unemployment.
  - Public good G produced with constant returns from public skilled workers (Lg) and goods (M) with elasticity ω.
- Production:
  - Two-stage CES production function: first-stage elasticity σ1 = 1/(1-ρ1); effective labor is CES of skilled and unskilled with elasticity σ2 = 1/(1-ρ2).
- Factor and wage determination:
  - Firms maximize profits; labor demand derived from marginal products of CES function.
  - Skilled wages: W_s = S(V_s, θ_s) (wage curve). Unskilled after-tax wages fixed at minimum wage W = F(L_u) for small changes.
  - Investment I = g( (1+T_n) R / R_f ); capital adjusts in long run so return equals world interest rate.
- Solution concepts:
  - Short run: capital fixed; employment and output respond via labor demand and supply.
  - Long run: capital adjusts to new steady state; tax reform raises marginal productivity/return on capital, inducing capital accumulation.

### IV. Calibration and key modeling assumptions
- Wage-curve elasticity for skilled workers (ψ): -0.15.
- Elasticity of output to public good (θ): 0.1.
- Labor's share in total costs (λl): 0.6.
- Share of unskilled workers: 44 percent.
- Share of unskilled labor costs in total labor costs: 28 percent.
- Employment-to-unemployment ratios: EURs = 21.5 (skilled); EURu = 5.3 (unskilled).
- Elasticities of substitution (Table 1 entries):
  - σ1 short term = 0.14; σ1 long term = 0.4.
  - σ2 short term = 0.43; σ2 long term = 1.4.
- Minimum-wage cutoff: 1.5 times the minimum wage divides behavior where market conditions begin to dominate; calibration sets unskilled group as wages driven by minimum wage up to that cutoff.
- Long-run pass-through calibration: long-run response calibrated with a 5th degree polynomial implying that for 400,000 jobs created, half of the tax cut is passed through to higher wages.
- Interpretation note: ψ = -0.15 implies a labor supply elasticity of 0.67 for a baseline unemployment rate of 10 percent.

### V. Short-run simulation results (fixed capital, no offsetting spending cuts)
- Policy scenarios:
  - CICE: EUR 20 billion; 3 ppt reduction in employers' SSC for wages up to 2.5 times the SMIC.
  - Untargeted RSP: EUR 10 billion; 1.5 ppt reduction in employers' SSC for wages up to 2.5 times the SMIC (82 percent of workers).
  - Targeted RSP: EUR 10 billion; 5.1 ppt reduction in employers' SSC for wages up to 1.5 times the SMIC (44 percent of workers).
- Selected quantitative short-run outcomes (percent changes unless otherwise indicated):
  - CICE (EUR 20 billion; 3 ppt reduction):
    - Output: 0.27
    - Total employment: 0.45
    - Employment, skilled: 0.33
    - Employment, unskilled: 0.78
    - Total employment (thousands): 134.4
    - Employment, skilled (thousands): 46.7
    - Employment, unskilled (thousands): 87.6
    - Labor costs: -1.30
    - Investment: 1.95
  - Untargeted EUR 10 billion (1.5 ppt reduction):
    - Output: 0.14
    - Total employment: 0.23
    - Employment, skilled: 0.16
    - Employment, unskilled: 0.39
    - Total employment (thousands): 67.2
    - Employment, skilled (thousands): 23.6
    - Employment, unskilled (thousands): 44.3
    - Labor costs: -0.66
    - Investment: 0.97
  - Targeted EUR 10 billion (5.1 ppt reduction for low-wage earners):
    - Labor costs fall by 3.5 percent.
    - Unskilled employment: +1.4 percent (around 161.0 thousand jobs).
    - Profits (rate of return): +1.61 percent.
    - Output: +0.23 percent.
    - Labor costs, unskilled: -3.54
- Combined and sensitivity:
  - Sum of CICE (EUR 20bn) and untargeted RSP (EUR 10bn) creates 200,000 jobs in the short run.
  - Employment impacts sensitive to σ1: with σ1 = 0.4, 350,000 jobs; with σ1 = 0.75, a EUR 30bn untargeted cut would generate 400,000 jobs.
  - Targeting low-skilled multiplies short-run employment impact by 2.3 and raises output effect by 1.6 relative to untargeted cuts.

### VI. Short-run results: budgetary-neutral reductions (tax cuts offset by spending cuts)
- Fiscal context:
  - An unfinanced EUR 10 billion tax cut worsens the short-term fiscal balance by 0.5 percent of GDP ex post.
- Spending cut types modeled:
  - Cuts in public goods (excluding public employment).
  - Cuts in public employment (assumed upfront; in some scenarios calibrated as destruction of 246,000 public service jobs).
- Selected targeted EUR 10bn short-run scenarios (selected entries):
  - No spending cut (baseline):
    - Fiscal balance (percent of GDP): -0.50
    - Output: 0.23
    - Total employment (thousands): 156.1
    - Private employment, unskilled (thousands): 161.3
    - Labor costs (private sector): -1.08
    - Investment: 1.62
  - Spending cut on public goods (γ = 0.1):
    - Fiscal balance: -0.18
    - Output: -0.46
    - Total employment (thousands): 117.0
    - Private employment, unskilled (thousands): 136.0
  - Cut in public employment (γ = 0.1; destroy 246,000 public jobs):
    - Fiscal balance: -1.02
    - Output: 0.29
    - Total employment (thousands): -34.0
    - Private employment, skilled (thousands): 87.7
    - Private employment, unskilled (thousands): 124.7
    - Public employment (thousands): -246.4
    - Skilled unemployment increases by 24 percent on impact; skilled wages fall by 3.6 percent given ψ = -0.015 in that scenario.
    - Net short-run effect: private hiring does not compensate upfront public job losses; overall 33,000 more workers unemployed and ex post fiscal deficit worsens to -1.0 percent of GDP in that scenario.
- Untargeted EUR 10bn short-run scenarios (selected entries):
  - No spending cut (baseline):
    - Fiscal balance: -0.50
    - Output: 0.14
    - Total employment (thousands): 67.9
    - Private employment, skilled (thousands): 23.6
    - Private employment, unskilled (thousands): 44.3
    - Investment: 0.99
  - Spending cut on public goods (γ = 0.1):
    - Fiscal balance: -0.18
    - Output: -0.55
    - Total employment (thousands): 28.1
  - Cut in public employment (γ = 0.1; destroy 246,000 public jobs):
    - Fiscal balance: -1.02
    - Output: 0.23
    - Total employment (thousands): -132.0
    - Private employment, skilled (thousands): 108.6
    - Private employment, unskilled (thousands): 5.8
    - Public employment (thousands): -246.4
- Memorandum: pass-through to unskilled wages in baseline budget-neutral short-run simulations reported as 0.0.

### VII. Long-run results (capital adjusts)
- Mechanism:
  - Over 10–15 years, capital adjusts to higher profits; return anchored by world rate; capital accumulation amplifies employment effects.
- Selected long-run outcomes (targeted EUR 10bn, no spending cuts — baseline):
  - Output: +1.3 percent.
  - Unskilled employment: +4 percent.
  - Unskilled wages: +1.6 percent.
  - Unskilled labor costs: small decline (text indicates a small fall).
- Combined EUR 30bn (CICE + untargeted RSP), unfinanced by spending cuts:
  - Combined employment effect: around 600,000 jobs (Table 6, column (d)).
  - Sensitivity: If only one-third of the labor cost reduction is passed into higher wages, employment increase reaches 720,000 (Table 6, column (e)).
  - Upper bound with infinitely elastic unskilled labor supply: one million jobs (column (f)).
- Table 5 (Targeted employer SSC reduction) — selected long-run entries (percent change unless otherwise indicated):
  - Fiscal balance (percent of GDP) columns (a)-(f): 0.69, 0.91, 1.60, 0.69, 1.22, 1.97
  - Output: 1.31, 1.58, 2.38, 1.31, 0.94, 4.31
  - Total employment (percent): 1.31, 1.58, 2.38, 1.31, 1.31, 4.54
  - Total employment (thousands): 489.2, 589.0, 889.1, 489.2, 489.2, 823.8
  - Private employment, unskilled (thousands): 455.3, 548.6, 827.6, 455.3, 455.3, 792.6
  - Labor costs, unskilled (private sector): -1.97, -2.37, -3.58, -1.97, -1.97, -1.81
  - Wages, unskilled (private sector): 1.61, 1.21, 0.00, 1.61, 1.61, 1.77
  - Memorandum pass-through to unskilled wages: 0.5, 0.3, 0.0, 0.5, 0.5, 0.5
- Table 6 (Untargeted employer SSC reduction) — selected long-run entries (percent change unless otherwise indicated):
  - Fiscal balance (percent of GDP) columns (a)-(f): 0.40, 0.85, 1.61, 0.28, 0.51, 1.19
  - Output: 1.01, 0.55, 3.94, 1.97, 2.24, 3.03
  - Total employment (percent): 1.01, 0.92, 4.17, 1.97, 2.24, 3.03
  - Total employment (thousands): 340.8, 304.8, 610.6, 420.4, 720.9, 1018.5
  - Private employment, unskilled (thousands): 276.9, 243.3, 554.9, 456.3, 549.9, 826.9
  - Labor costs, unskilled (private sector): -1.04, -0.90, -0.56, -1.50, -1.91, -3.10
  - Wages, unskilled (private sector): 0.00, 0.15, 0.48, 1.62, 1.22, 0.02
  - Memorandum pass-through to unskilled wages: 0.0, 0.1, 0.5, 0.5, 0.4, 0.0

### VIII. Policy implications and caveats
- Main policy messages:
  - Targeting employer SSC reductions at low-wage/unskilled workers substantially increases short-run and long-run employment impacts relative to broad-based cuts.
    - Short-run example: targeted EUR 10bn yields about 161,000 unskilled jobs vs. 44,300 under untargeted EUR 10bn.
    - Targeting short-run multiplier: about 2.3 relative to untargeted.
  - Financing matters:
    - Unfinanced cuts worsen fiscal balance by 0.5 percent of GDP ex post in the short run.
    - Offsetting tax cuts with cuts in public goods reduces or reverses employment and output gains.
    - Cuts in public employment produce complex reallocation: private hiring increases but typically does not offset public job losses immediately and can worsen fiscal balance due to automatic stabilizers.
- Caveats and limitations:
  - Results highly sensitive to σ1 (elasticity of substitution between labor and capital) and pass-through assumptions for unskilled wages.
  - Model abstracts from Keynesian demand-side fiscal effects in the short run.
  - Potential unintended side-effects not captured: low-wage traps, timing and delivery of CICE as a tax credit, use of funds to repair firm balance sheets rather than hiring.
  - Employment outcomes depend on interaction with social benefits, unemployment insurance, and minimum wage policies.

### IX. Sensitivity and comparative evidence
- Sensitivity dimensions:
  - Targeting scheme (low wages vs. across-the-board).
  - Financing scheme (no cut, cut public goods, cut public employment).
  - Pass-through to unskilled wages and slope of wage curve for skilled workers.
- Comparative results with other studies:
  - Nouveau and Ourliac (2012): reductions in social security contributions 1993–2012 would have created or preserved between 400,000 and 800,000 jobs for an ex ante cost of around 1.2 percent of GDP (EUR 22 billion).
  - Haute Conseil du financement de la protection sociale (March 2014): EUR 10 billion uniform reductions deliver between 134,000 and 214,000 additional jobs after 5 years depending on model; concentrated relief yields greater job creation in many models; ex-ante funding reduces effects to between 43,000 and 81,000 jobs.
  - Authorities’ simulations (INSEE 2014) estimate CICE and RSP create around 500,000 jobs altogether.

### X. Model equations and solution structure (brief)
- Model solved for variables: bs, y, k, r, l, l_s, l_u, lc_s, lc_u, w_s, w_u, i with an additional equation for short-run/long-run closure and a budget constraint.
- Key relations:
  - CES production, factor price frontier, labor demand as functions of elasticities and labor costs.
  - Labor costs as functions of wages and payroll tax rates.
  - Investment depends on relative return to foreign rate; skilled wage follows wage curve with elasticity ψ; unskilled wages anchored at minimum wage for small changes.

### XI. Appendix B — sensitivity on key parameters
- Wage curve elasticity for skilled workers (ψ):
  - Baseline ψ = -0.15; literature range between -0.05 and -0.25.
  - Comparative statics: more negative ψ → larger reduction in skilled wages in equilibrium, benefiting skilled employment; output slightly higher for more sensitive wage curve; budget impact mostly insensitive to ψ.
- Share of workers with wages driven by minimum wage (λs):
  - Baseline λs implied by calibration: share of skilled workers 0.56 → λs (share of labor costs for skilled) = 0.72.
  - Employment in subgroups insensitive to λs; total employment and output decrease with higher λs because unskilled share (1-λs) shrinks.
- Footnote calibration reference: 5.1 ppt reduction in employers' SSC for unskilled workers is equivalent to 0.5 percent of GDP ex post in the simulations reported.

### XII. Key statistics and numeric facts (verbatim)
- EUR 30 billion over 2013-17 (tax wedge cut).
- Wages targeted up to 2.5 to 3.5 times the minimum wage.
- Short run job creation estimate: around 200,000 jobs from a EUR 30bn reduction in employers’ social security contributions.
- Long run job creation estimate: gains could reach 600,000 jobs.
- Targeting effectiveness: targeted tax cuts to low wage workers: 2.3 times employment effect versus untargeted for equal budgetary cost (1.5 times in the long run).
- Aubry and Fillon tax breaks increased reductions from 18.2 to 26 percent of gross wages for the SMIC.
- July 2007 tax cuts at the SMIC level increased to 28.1 percent of gross wages for firms with fewer than 20 employees.
- Fiscal cost in 2009: around EUR 22.2 billion (1.2 percent of GDP).
- Recent labor cost reductions under CICE and RSP: EUR 30 billion (1.5 percent of GDP) over 2013-17.
- Tax wedge progressivity: social charges start at 8.2 points for the SMIC and reach 42.3 at 2.5 times the SMIC.
- Cutoff for wage behavior where market conditions dominate: 1.5 times the minimum wage.
- Share of unskilled workers: 44 percent.
- Share of unskilled labor costs in total labor costs: 28 percent.
- Elasticities and calibration entries: σ1 short term = 0.14; σ1 long term = 0.4; σ2 short term = 0.43; σ2 long term = 1.4.
- Wage-curve elasticity for skilled workers (ψ): -0.15.
- Employment-to-unemployment ratios: EURs = 21.5 (skilled); EURu = 5.3 (unskilled).
- Long-run illustrative pass-through memorandum rates: 0.0, 0.1, 0.3, 0.5 (reported across scenarios).

*Source: APPENDIX A of the IMF working paper (_wp14114).*

### APPENDIX A .............................................................................................................

### _wp14114 - APPENDIX A .............................................................................................................

### I. INTRODUCTION
- France has a high labor tax wedge and a long tradition of offsetting it with subsidies and tax breaks (Balladur and Juppé reforms in the 1990s; Aubry and Fillon tax breaks in the 2000s; more recent tax credits proposed by President Hollande).
- The latest initiative is a cut in the tax wedge worth EUR 30 billion over 2013-17 aimed at reversing deteriorating employment and competitiveness.
- Major shares of the amounts are channeled through the Crédit d’impôt pour la compétitivité et l’emploi (CICE) and the Responsibility and Solidarity Pact (RSP); these are intended to reduce labor costs for wages up to 2.5 to 3.5 times the minimum wage.
- Purpose of the analysis:
  - Calibrate a basic supply-side general equilibrium model for the French economy.
  - Abstract from demand-side effects to focus on effects on potential output and potential employment induced by these measures.
  - Emphasis on skilled versus unskilled trade-offs and alternative financing schemes.
  - Simulations reported for short run (prior to any capital response) and long run (after full capital adjustment). No specific time frame is provided.
- Model features (summary):
  - Open, price-taking economy with capital, skilled and unskilled labor, involuntary unemployment, and full capital mobility.
  - Goods and capital markets are perfectly competitive; labor market subject to frictions.
  - Unskilled labor supply is infinitely elastic at the minimum wage; skilled wages are negatively related to unemployment (upward sloping labor supply).
  - Labor incidence principle invoked: the more inelastic the labor supply, the more labor appropriates the room created by reducing the tax wedge through higher wages (Dalton’s law).
- Key simulation headline results (from model):
  - Short run: around 200,000 jobs could be created by a EUR 30bn reduction in employers’ social security contributions.
  - Long run: gains could reach 600,000 jobs.
- Targeting and effectiveness:
  - Tax cuts targeted to low wage workers have 2.3 times as large an effect on employment as untargeted reductions for equal budgetary cost (1.5 times as large in the long run).
  - Rationale: base effects (larger labor cost reduction for wages close to the minimum wage for equal budgetary cost) and higher empirical labor supply elasticity for unskilled workers.
  - Untargeted cuts generate lower employment impact because labor cost reductions for high-wage earners are partly appropriated through higher wages; this offsetting effect is weaker at the low end due to high incidence of unemployment among unskilled workers.
- Paper structure overview:
  - Section II reviews past episodes of tax wedge reductions in France.
  - Section III describes model features and baseline calibration (detailed in Appendix A).
  - Section IV presents simulation results for alternative targeting and financing schemes and sensitivity analyses.
  - Section V contrasts results with other models of the French economy.
  - Section VI concludes.

### II. A WALK THROUGH MEMORY LANE (History of labor tax cuts in France)
- Minimum wage role and trade-offs:
  - Minimum wages ensure a living wage, reduce social exclusion, act as efficiency wages, and reduce income inequality.
  - High minimum wages increase unemployment for workers whose productivity falls below total labor cost; France has offset this by reducing social security contributions, especially at the minimum wage, since 1993.
- Major episodes of targeted labor tax cuts:
  - 1993—1998: Balladur and Juppé reforms reduced labor costs for (gross) wages up to 1.2—1.33 times the minimum wage (SMIC) for all enterprises.
  - 1998—2002 and 2003—2005: Aubry and Fillon tax breaks increased reductions from 18.2 to 26 percent of gross wages for the SMIC, converging toward zero for wages at 1.6 times the SMIC.
  - 2007—2012: Marginal adjustments; July 2007 tax cuts at the SMIC level increased to 28.1 percent of gross wages for firms with fewer than 20 employees; October 2007 formula change to avoid discouraging overtime; January 2011 adjustment to factor exceptional bonuses into exemption calculation.
  - 2013 onwards: President Hollande proposed further tax wedge cuts worth EUR 30 billion over 2013-2017; two thirds channeled through the CICE to reduce labor compensation for wages up to 2.5 times the SMIC (80 percent of workers) by 2015; remaining EUR 10 billion to target workers earning up to around 3.5 times the minimum wage.
- Evolution of targeting and progressivity:
  - Exit threshold from tax reliefs increased from wages close to the SMIC in the early 1990s up to wages at 2.5 times the SMIC under the CICE.
  - Tax wedge structure remains highly progressive: social charges start at 8.2 points for the SMIC and reach 42.3 at 2.5 times the SMIC.
- Fiscal costs:
  - Fiscal cost in 2009 amounted to around EUR 22.2 billion (1.2 percent of GDP); almost half due to measures to counter the cost of moving to the 35-hour week.
  - Recent labor cost reductions committed under the CICE and the RSP imply EUR 30 billion (1.5 percent of GDP) of additional labor tax cuts over 2013-17.
- Observed impacts noted by economists:
  - Since 1993, cost of labor at SMIC relative to median wage decreased sharply; share of unskilled workers in total employment improved quickly in 1994-95 despite economic slowdown.

### III. ANALYTICAL FRAMEWORK
H3: A. The Model
- Overview:
  - One-sector, four-factor general equilibrium model with distinct wage-setting for skilled and unskilled labor and a productive role for public spending.
  - Simulates reduced employers’ social security contributions compensated with expenditure measures.
  - Solutions computed for short run (fixed capital) and long run (capital adjusts so return equals world interest rate).
  - For small variations in demand for unskilled labor, unskilled wages determined by the minimum wage; skilled wages determined by a labor supply function tied to unemployment and the real after-tax wage.
  - Workers willing to take up jobs at all wage levels (involuntary unemployment), but unused labor highest at minimum wage level; skilled and unskilled labor forces are constant.
- Production structure:
  - Output produced by two-stage CES production function using four factors: skilled workers, unskilled workers, private capital, and a public good (externality to private factors).
  - First-stage elasticity of substitution σ1 = 1/(1-ρ1); production function: (notation from source)
    - Y = [ γ (δ_k K)^{ρ1} + (1-γ) (δ_L L)^{ρ1} ]^{1/ρ1}  (presented in source as KLGY ... with parameters)
  - Effective labor is CES of skilled and unskilled with elasticity σ2 = 1/(1-ρ2).
  - Public good G produced with constant returns from public skilled workers (Lg) and goods (M) with production elasticity ω: G = Lg^{ω} M^{1-ω} (formulation presented in source).
- Firm profit maximization and first-order conditions:
  - Profits π = RK + (1- T_p)(W_u L_u + W_s L_s) - Y (presented in source as ss... expression).
  - Labor demand functions derived from marginal products of CES function (expressions provided in source).
- Wage formation:
  - Skilled wages follow a wage curve: W_s = S(V_s, θ_s) where after-tax wages positive in reservation wage (after-tax unemployment benefit) and negative in skilled unemployment rate.
  - Skilled unemployment rate defined with notation V_s and unemployment U_s, with N_s exogenous labor supply and Lg exogenous public employment.
  - Unskilled after-tax wages fixed at the minimum wage W, except for large increases in employment rate: W = F(L_u).
- Investment and budget:
  - Investment I positive function g of ratio of after-tax return on capital R to foreign rate R_f: I = g( (1+T_n) R / R_f ) (source notation).
  - Budget surplus B specified with components including payroll taxes, unemployment benefits, pensions, social charges; labor cost of public servants equal to labor cost of skilled private sector workers (full expression in source).
- Log-linearization:
  - Simulations reported in deviations from baseline; model log-linearized with equations in Appendix A.

H3: B. Short- and Long-Term Solutions
- Short run:
  - Capital fixed; employment and output respond via labor demand and supply interaction.
  - Employment effects depend critically on elasticities of labor demand and supply (Dalton’s law): more inelastic labor supply implies larger appropriation of tax cut by higher wages and smaller employment impact.
- Long run:
  - Capital reaches new steady-state where return equals exogenous world interest rate.
  - Tax reform raises marginal productivity/return on capital, inducing capital accumulation which amplifies employment effects.
  - Potential job creation depends on elasticities of labor and capital supply and demand; assumed different in long run following EC (1994).
  - If labor supply is infinitely elastic (as for unskilled), wages remain constant, employment gains large, and marginal productivity of capital increases substantially; if labor supply inelastic, tax cuts are appropriated by workers with no employment effects.
- Threshold effects and segmented labor supply:
  - Labor supply is truncated at wage level equal to 1.5 times the minimum wage in calibration: highly elastic up to that level, upward sloping thereafter.

H3: C. Calibration
- Division between unskilled and skilled based on observed wage behavior; breaking point estimated at 1.5 times the minimum wage (Aeberhardt and others, 2012; Goarant and Muller, 2012).
- Implications of cutoff:
  - Share of unskilled workers is 44 percent.
  - Share of unskilled labor costs in total labor costs is 28 percent (DARES).
- Differential elasticities:
  - Low-wage earners have lower employment rates and sectors employing unskilled workers tend to face higher elasticity of demand with respect to prices; a smaller proportion of labor cost reductions can be appropriated by employees in the form of wage increases.
- Note on unemployment benefits:
  - For calibration, focus on replacement rate dimension of unemployment benefit system (source discussion referencing Stovicek and Turrini, 2012).

### Key statistics and numeric facts (verbatim from source)
- EUR 30 billion over 2013-17 (tax wedge cut).
- Wages targeted up to 2.5 to 3.5 times the minimum wage.
- Short run job creation estimate: around 200,000 jobs from a EUR 30bn reduction in employers’ social security contributions.
- Long run job creation estimate: gains could reach 600,000 jobs.
- Targeting effectiveness:
  - Targeted tax cuts to low wage workers: 2.3 times employment effect versus untargeted for equal budgetary cost (1.5 times in the long run).
- Historical percentages and figures:
  - Aubry and Fillon tax breaks increased reductions from 18.2 to 26 percent of gross wages for the SMIC.
  - July 2007 tax cuts at the SMIC level increased to 28.1 percent of gross wages for firms with fewer than 20 employees.
  - Fiscal cost in 2009: around EUR 22.2 billion (1.2 percent of GDP).
  - Recent labor cost reductions under CICE and RSP: EUR 30 billion (1.5 percent of GDP) over 2013-17.
  - Tax wedge progressivity: social charges start at 8.2 points for the SMIC and reach 42.3 at 2.5 times the SMIC.
  - Cutoff for wage behavior where market conditions dominate: 1.5 times the minimum wage.
  - Share of unskilled workers: 44 percent.
  - Share of unskilled labor costs in total labor costs: 28 percent.

*Source: APPENDIX A of the IMF working paper (_wp14114).*

### 16.9 percent in 20

### _wp14114 - 16.9 percent in 20

### Calibration and key modeling assumptions
- Wage-curve elasticity for skilled workers (ψ): -0.15.
- Elasticity of output to public good (θ): 0.1 (based on Bom and Ligthart (2013) meta-analysis).
- Labor's share in total costs (λl): 0.6.
- Share of skilled labor costs in total labor costs (λs): (1-0.28)*λl.
- Employment-to-unemployment ratios: EURs = 21.5 (skilled); EURu = 5.3 (unskilled).
- Elasticity of substitution between effective labor and capital and between skilled and unskilled labor appear in Table 1 (σ1 short term = 0.14; σ1 long term = 0.4; σ2 short term = 0.43; σ2 long term = 1.4).
- Minimum-wage constraint for unskilled: for small changes in unskilled employment, wages unchanged; long-run response calibrated with a 5th degree polynomial implying that for 400,000 jobs created, half of the tax cut is passed through to higher wages.
- Note on interpretation: wage-curve elasticity of -0.15 relating unemployment rate and the real wage implies a labor supply elasticity of 0.67 for a baseline unemployment rate of 10 percent.

### Short-run results: Social security reductions without offsetting spending cuts
- Policy scenarios modeled: CICE (EUR 20 billion; 3 ppt reduction in employers' SSC for wages up to 2.5 times the SMIC), Responsibility Pact (RSP) EUR 10 billion under two targeting assumptions:
  - Untargeted RSP: 1.5 ppt reduction in employers' SSC for wages up to 2.5 times the SMIC (82 percent of workers).
  - Targeted RSP: 5.1 ppt reduction in employers' SSC for wages up to 1.5 times the SMIC (44 percent of workers).
- Selected short-run quantitative outcomes (percent changes unless otherwise indicated):
  - CICE (EUR 20 billion; 3 ppt reduction):
    - Output: 0.27
    - Total employment: 0.45
    - Employment, skilled: 0.33
    - Employment, unskilled: 0.78
    - Total employment (thousands): 134.4
    - Employment, skilled (thousands): 46.7
    - Employment, unskilled (thousands): 87.6
    - Labor costs: -1.30
    - Investment: 1.95
  - Untargeted EUR 10 billion (1.5 ppt reduction):
    - Output: 0.14
    - Total employment: 0.23
    - Employment, skilled: 0.16
    - Employment, unskilled: 0.39
    - Total employment (thousands): 67.2
    - Employment, skilled (thousands): 23.6
    - Employment, unskilled (thousands): 44.3
    - Labor costs: -0.66
    - Investment: 0.97
  - Targeted EUR 10 billion (5.1 ppt reduction for low-wage earners):
    - Tax wedge for low-wage earners reduced by 5.1 percentage points.
    - Labor costs fall by 3.5 percent.
    - Unskilled employment: +1.4 percent (around 161.0 thousand jobs).
    - Profits (rate of return): +1.61 percent.
    - Output: +0.23 percent.
    - Labor costs, unskilled: -3.54
- Combined effects and sensitivity:
  - Sum of CICE (EUR 20bn) and untargeted RSP (EUR 10bn) would create 200,000 jobs in the short run.
  - Employment impacts sensitive to elasticity of substitution (σ1): with σ1 = 0.4, 350,000 jobs would be created; with σ1 = 0.75, a EUR 30bn untargeted cut would generate 400,000 jobs.
  - Targeting the tax cuts to low-skilled could have multiplied the short-run employment impact by 2.3 and raised the output effect by 1.6.

### Short-run results: Budgetary-neutral reductions (tax cuts offset by spending cuts)
- Fiscal context: An unfinanced EUR 10 billion tax cut worsens the short-term fiscal balance by 0.5 percent of GDP ex post; simulations examine financing via spending cuts.
- Types of spending cuts modeled:
  - Cuts in public goods (excluding public employment).
  - Cuts in public employment (assumed upfront; calibrated destruction of 246,000 public service jobs in some scenarios).
  - Cuts in transfers (not explicitly modeled for labor-supply effects).
- Targeted EUR 10bn (5.1 ppt reduction) — selected short-run scenarios (Table 3):
  - No spending cut (baseline):
    - Fiscal balance (percent of GDP): -0.50
    - Output: 0.23
    - Total employment (thousands): 156.1
    - Private employment, unskilled (thousands): 161.3
    - Labor costs (private sector): -1.08
    - Investment: 1.62
  - Spending cut on public goods (γ = 0.1):
    - Fiscal balance: -0.18
    - Output: -0.46
    - Total employment (thousands): 117.0
    - Private employment, unskilled (thousands): 136.0
  - Cut in public employment (γ = 0.1; destroy 246,000 public jobs):
    - Fiscal balance: -1.02
    - Output: 0.29
    - Total employment (thousands): -34.0
    - Private employment, skilled (thousands): 87.7
    - Private employment, unskilled (thousands): 124.7
    - Public employment (thousands): -246.4
    - Skilled unemployment increases by 24 percent on impact; skilled wages fall by 3.6 percent given ψ = -0.015 in the scenario described.
    - Net short-run effect: increase in private hiring does not compensate upfront public job losses; overall 33,000 more workers unemployed in that scenario and ex post fiscal deficit worsens to -1.0 percent of GDP due to lower revenues.
  - If public employment is highly productive (higher γ), cuts in public employment can be more harmful, turning GDP and employment effects negative.
- Untargeted EUR 10bn — selected short-run scenarios (Table 4):
  - No spending cut (baseline):
    - Fiscal balance: -0.50
    - Output: 0.14
    - Total employment (thousands): 67.9
    - Private employment, skilled (thousands): 23.6
    - Private employment, unskilled (thousands): 44.3
    - Investment: 0.99
  - Spending cut on public goods (γ = 0.1):
    - Fiscal balance: -0.18
    - Output: -0.55
    - Total employment (thousands): 28.1
  - Cut in public employment (γ = 0.1; destroy 246,000 public jobs):
    - Fiscal balance: -1.02
    - Output: 0.23
    - Total employment (thousands): -132.0
    - Private employment, skilled (thousands): 108.6
    - Private employment, unskilled (thousands): 5.8
    - Public employment (thousands): -246.4
- Memorandum: pass-through to unskilled wages in these short-run scenarios is reported as 0.0 in the baseline budget-neutral simulations.

### Long-run results
- Mechanism: over 10–15 years, capital adjusts to higher profits and equilibrium returns to the world rate; small changes in labor costs can yield large effects on capital stock, output, and employment.
- Targeted EUR 10 billion (no spending cuts) — baseline long-run outcomes:
  - Output: +1.3 percent.
  - Unskilled employment: +4 percent.
  - Unskilled wages: +1.6 percent.
  - Unskilled labor costs: small decline (text indicates "unskilled labor costs fall by just" then continues in source).
- Long-run results are sensitive to:
  - Assumptions on pass-through of tax cuts to unskilled wages.
  - Financing assumptions (whether tax cuts are offset by spending cuts and which spending categories are cut).
  - Alternative calibrations of elasticities and substitution parameters.

### Policy implications highlighted by simulations
- Targeting employer SSC reductions at low-wage/unskilled workers substantially increases short-run and long-run employment impacts relative to broad-based (untargeted) cuts:
  - Short-run: targeted EUR 10bn yields about 161,000 unskilled jobs vs. 44,300 under untargeted EUR 10bn.
  - Short-run multiplier: targeting could raise employment impact by a factor of about 2.3 relative to untargeted cuts.
  - Long-run: targeted cuts produce larger increases in output and unskilled employment.
- Financing matters:
  - Unfinanced cuts worsen the fiscal balance by 0.5 percent of GDP ex post in the short run.
  - Offsetting tax cuts with cuts in public goods reduces or reverses employment and output gains; cuts in public employment yield complex reallocation effects (private hiring partially increases but typically does not offset public job losses immediately and can worsen fiscal balance via automatic stabilizers).
- Sensitivities and caveats:
  - Results are highly sensitive to the elasticity of substitution between labor and capital (σ1) and to assumptions on the pass-through of tax cuts to wages for unskilled workers.
  - The model abstracts from Keynesian demand-side fiscal effects in the short run; results focus on supply-side and long-run capital adjustment channels.

*Source: IMF staff estimates and model simulations reported in the chapter PDF.*

### 1.9 percent). The employment of skilled labor increases marginally (by 0.2 percent), since

### _wp14114 - 1.9 percent). The employment of skilled labor increases marginally (by 0.2 percent), since

### Main findings
- Employment effects of targeted reductions in employer social security contributions:
  - The employment of skilled labor increases marginally (by 0.2 percent), since increased demand also raises skilled labor costs by 0.8 percent.
  - In the long run, the increase in skilled workers’ wages limits initial gains in external competitiveness and labor costs are unaffected by tax policy.
- Reduction in social security taxes leads to budgetary savings in the long run due to:
  - Increases in the tax base.
  - Savings on outlays for unemployment insurance (substantial for countries with high replacement ratios in the first year of unemployment, e.g., France).
- A self-financed targeted tax cut appears robust to different wage-curve assumptions, but budgetary savings are lower if the tax cut is untargeted.
- Financing assumptions matter:
  - If EUR 10 billion tax cut is combined with a reduction in spending on public goods, growth effect falls from +1.3 percent to +0.9 percent (Table 5 column (d) versus column (e)).
  - Employment is not affected by the cut in public spending in that scenario because relative cost of labor to capital is constant in the long run (cost of capital anchored by world interest rate).
  - If the tax-wedge cut is combined with a cut in public employment, large wage moderation from public dismissals reduces firms’ labor costs, prompting capital accumulation and output effects; private skilled employment rises roughly as much as the initial cut in public employment, leaving total skilled employment unchanged, while demand for unskilled workers rises by around 800,000.

### Quantitative results (selected long-run responses)
- Combined CICE and (untargeted) RSP, unfinanced by spending cuts:
  - Combined employment effect: around 600,000 jobs (Table 6, column (d)).
  - Sensitivity: If only one-third of the labor cost reduction is passed into higher wages, employment increase reaches 720,000 (Table 6, column (e)).
  - Upper bound with infinitely elastic labor supply for unskilled workers: one million jobs (column (f)).
- Table 5 (Targeted employer social security tax reduction) — selected entries (percent change unless otherwise indicated):
  - Fiscal balance (percent of GDP): columns (a)-(f): 0.69, 0.91, 1.60, 0.69, 1.22, 1.97
  - Output: 1.31, 1.58, 2.38, 1.31, 0.94, 4.31
  - Total employment (percent): 1.31, 1.58, 2.38, 1.31, 1.31, 4.54
  - Total employment (in thousands): 489.2, 589.0, 889.1, 489.2, 489.2, 823.8
  - Private employment, unskilled (in thousands): 455.3, 548.6, 827.6, 455.3, 455.3, 792.6
  - Public employment (in thousands): 0.0, 0.0, 0.0, 0.0, 0.0, -246.4
  - Labor costs, unskilled (private sector): -1.97, -2.37, -3.58, -1.97, -1.97, -1.81
  - Wages, unskilled (private sector): 1.61, 1.21, 0.00, 1.61, 1.61, 1.77
  - Memorandum item: pass-through to unskilled wages: 0.5, 0.3, 0.0, 0.5, 0.5, 0.5
- Table 6 (Untargeted employer social security tax reduction) — selected entries (percent change unless otherwise indicated):
  - Fiscal balance (percent of GDP): columns (a)-(f): 0.40, 0.85, 1.61, 0.28, 0.51, 1.19
  - Output: 1.01, 0.55, 3.94, 1.97, 2.24, 3.03
  - Total employment (percent): 1.01, 0.92, 4.17, 1.97, 2.24, 3.03
  - Total employment (in thousands): 340.8, 304.8, 610.6, 420.4, 720.9, 1018.5
  - Private employment, skilled (in thousands): 64.0, 61.5, 302.1, 64.1, 71.0, 191.6
  - Private employment, unskilled (in thousands): 276.9, 243.3, 554.9, 456.3, 549.9, 826.9
  - Public employment (in thousands): 0.0, 0.0, -246.4, 0.0, 0.0, 0.0
  - Labor costs, unskilled (private sector): -1.04, -0.90, -0.56, -1.50, -1.91, -3.10
  - Wages, unskilled (private sector): 0.00, 0.15, 0.48, 1.62, 1.22, 0.02
  - Memorandum item: pass-through to unskilled wages: 0.0, 0.1, 0.5, 0.5, 0.4, 0.0

### Sensitivity and scenario analysis
- Results sensitive to:
  - Targeting scheme (targeted on low wages versus untargeted across-the-board).
  - Financing scheme (no spending cut, spending cut on public goods, cut in public employment).
  - Pass-through from labor cost reductions to unskilled wages (different pass-through rates reported across columns).
  - Slope of the wage curve for unskilled workers.
- Nonlinearities:
  - Under baseline supply curve for unskilled workers, model is not linear; results for EUR 10bn RSP are not simply a third of results for combined EUR 30bn CICE and RSP.

### Comparison with other studies and simulations
- Nouveau and Ourliac (2012) summary: reductions in social security contributions between 1993 and 2012 would have created or preserved between 400,000 and 800,000 jobs for an ex ante cost of around 1.2 percent of GDP or EUR 22 billion.
- Haute Conseil du financement de la protection sociale (March 2014) simulations for EUR 10 billion tax cuts (0.5 percent of GDP) show:
  - Uniform labor cost reductions deliver between 134,000 and 214,000 additional jobs after 5 years depending on model/assumptions.
  - Tax relief concentrated on low wages yields greater job creation, but less so in models with a tradable sector (Nemesis).
  - Ex-ante funding of tax cuts reduces employment effects to between 43,000 and 81,000 jobs depending on model.
  - Simulations show improvement in government deficit if there are positive employment effects.
- Authorities’ simulations (INSEE 2014, Programme national de réforme, 2014) estimate CICE and RSP create around 500,000 jobs altogether, consistent with the paper’s baseline long-run results.
- Explanation for lower effectiveness relative to past measures: current schemes are less focused on low wages and also aim to benefit exporting firms which employ higher-skilled labor.

### Policy implications and caveats
- EUR 30 billion of labor tax cuts impact employment but effects depend strongly on targeting, financing, and wage dynamics; uncertainty is significant.
- Policy levers within reach:
  - Targeting schemes.
  - Compensatory measures.
  - Wage-moderation agreements (responsibility element under RSP).
- Potential unintended side-effects not captured by model:
  - Higher progressivity of labor taxation could create “low wage traps” by raising the cost of training and rewarding higher productivity, possibly biasing specialization toward low-skilled intensive activities.
  - CICE design as a tax credit (received as part of annual corporate tax settlement) may weaken immediate hiring incentives compared with direct labor-tax cuts, despite economic equivalence in the model.
  - Implementation coincides with low profit margins, possibly leading firms to use CICE/RSP to improve financial situations rather than increase hiring.
- Employment outcomes also depend on interaction with other policies: social benefits, unemployment insurance, and the minimum wage (SMIC). A broader welfare analysis would be required to compare alternative policy mixes.

### Model overview (brief)
- A 12-equation model solved for 14 variables: bs, y, k, r, l, l_s, l_u, lc_s, lc_u, w_s, w_u, and i, with a 13th equation (i=0 in the long run or k=0 in the short run) and a 14th equation (budget constraint bs=0 or an assumption for other tax rate).
- Key relationships:
  - Production function, factor price frontier, labor demand as functions of elasticities of substitution and labor costs.
  - Labor costs expressed as functions of wages and payroll tax rates.
  - Investment equation assumes constant foreign rate of return.
  - Wage determination for skilled workers includes unemployment elasticity ψ; minimum wage assumed constant for unskilled workers.
- Appendix contains equations expressed in percent change deviations and definitions of parameters (λ’s, σ’s, ψ, ω, EUR ratio of employment to unemployment).

*Source: _wp14114 - 1.9 percent). The employment of skilled labor increases marginally (by 0.2 percent), since (staff estimates and model results presented in the supplied PDF content).*

### APPENDIX B

### APPENDIX B

### Wage curve elasticity for skilled workers (ψ)
- Baseline calibration assumes ψ = -0.15 (Table 1).
- Literature range: most studies estimate this elasticity between -0.05 and -0.25 (meta-analysis of Nijkamp and Poot, 2005).
- Comparative statics over this range show:
  - In equilibrium, the reduction in skilled wages is higher the more negative ψ (the higher the sensitivity of wages to unemployment for skilled workers).
  - Larger wage loss for skilled workers (and thus reduced labor cost) benefits skilled employment.
  - Output is slightly higher for a more sensitive wage curve.
  - The budget impact of the tax measure is mostly insensitive to the wage curve elasticity.
- Figures and metrics shown (short-run response, EUR 10 bn targeted tax cut in employers' social security contributions 1/):
  - Employment (In percent, deviation from baseline) plotted across ψ from -0.25 to -0.05 for Skilled and Unskilled.
  - Net Wage (In percent, deviation from baseline) plotted across ψ from -0.25 to -0.05 for Skilled and Unskilled.
  - Output (In percent, deviation from baseline) plotted across ψ from -0.25 to -0.05.
  - Government Deficit (In percent of GDP, deviation from baseline) plotted across ψ from -0.25 to -0.05.
- Footnote: 1/ Equivalent to 5.1 ppt reduction in employers' SSC for unskilled workers or 0.5 percent of GDP ex post.

### Share of workers for which wages are driven by the minimum wage (λs)
- Baseline calibration sets the share of workers for which wages were sensitive to unemployment (skilled workers) to 0.56, implying λs, the share of labor costs for skilled workers, would be 0.72.
- Calibration rationale:
  - Predicated on empirical findings that salaries of workers earning up to 1.5 times the minimum wages are affected by the minimum wage.
  - Salaries for workers earning 1.2 or 1.3 times the minimum wage are also partly affected by labor market tightness.
- Comparative statics results:
  - Employment in the different subgroups is insensitive to λs.
  - Impact on total employment is a decreasing function of λs, since the size of the unskilled group (for which the employment response to the tax cut is positive) is increasing in 1- λs.
  - Impact on output is a decreasing function of λs.
- Figures and metrics shown (short-run response, EUR 10 bn targeted tax cut in employers' social security contributions 1/):
  - Employment (In percent, deviation from baseline) plotted across λs from 0.70 to 0.90 for Skilled and Unskilled.
  - Output (In percent, deviation from baseline) plotted across λs from 0.70 to 0.90.
  - Government Deficit (In percent of GDP, deviation from baseline) plotted across λs from 0.70 to 0.90.
  - Net Wage (In percent, deviation from baseline) plotted across λs from 0.70 to 0.90 for Skilled and Unskilled.
- Footnote: 1/ Equivalent to 5.1 ppt reduction in employers' SSC for unskilled workers or 0.5 percent of GDP ex post.

*Source: Staff simulations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp14114.pdf_