How Do Fiscal and Labor Policies in France Affect Inequality?

Contributor Notes

Author’s contact information: respinoza@imf.org; eperezruiz@imf.org

This paper explores the impact of fiscal and labor market policies on efficiency, inequality, and fiscal outcomes in France. We extend the general equilibrium model calibrated for France by Alla and others (2015), with measures of labor and capital income for different groups in the economy (the unemployed, unskilled workers, skilled workers, public servants). For each of these groups we combine data on the income distribution with the outcomes of policy simulations to assess the impact of a suite of stylized policies on output, the fiscal balance, the Gini coefficient, and the shape of the Lorenz curve. We find that most types of fiscal expansions, while adding to the deficit and debt in the near term, generally reduce inequality, the main exception being capital income tax cuts. A reduction of the minimum wage has an ambiguous impact on the income distribution: the Gini coefficient increases, but the lowest income quintile improves its relative position in the income distribution thanks to positive employment effects. The paper also finds scope for “win-win” policy packages that could improve overall efficiency, inequality, and fiscal outcomes, for instance if targeted labor tax reductions are offset by cuts in the public wage bill.

Abstract

This paper explores the impact of fiscal and labor market policies on efficiency, inequality, and fiscal outcomes in France. We extend the general equilibrium model calibrated for France by Alla and others (2015), with measures of labor and capital income for different groups in the economy (the unemployed, unskilled workers, skilled workers, public servants). For each of these groups we combine data on the income distribution with the outcomes of policy simulations to assess the impact of a suite of stylized policies on output, the fiscal balance, the Gini coefficient, and the shape of the Lorenz curve. We find that most types of fiscal expansions, while adding to the deficit and debt in the near term, generally reduce inequality, the main exception being capital income tax cuts. A reduction of the minimum wage has an ambiguous impact on the income distribution: the Gini coefficient increases, but the lowest income quintile improves its relative position in the income distribution thanks to positive employment effects. The paper also finds scope for “win-win” policy packages that could improve overall efficiency, inequality, and fiscal outcomes, for instance if targeted labor tax reductions are offset by cuts in the public wage bill.

I. Introduction1

Context. Income inequality has become a growing concern in Europe, especially as policies adopted during the crisis years may have had adverse effects on the income distribution (see, e.g., Hoeller and others, 2014). Equality is an especially important concept in France. Evaluating the merits of any policy change should therefore consider not only the impact on output, employment, and the fiscal balance, but also on the income distribution. However, the existing literature (see, e.g., Checchi and García-Peñalosa, 2008, and the references therein) lacks a consistent method to systematically assess the impact of policies across all these dimensions in France. This paper attempts to close this gap.

French inequality profile. In France, inequality of disposable income—as measured by the Gini coefficient—increased by 1.7 percentage points during 2007–11, and now broadly matches the OECD average of about 0.3.2 By comparison, pre-policy inequality rose by 2.9 percentage points over the same period to 0.51—compared with an OECD average of 0.47. French redistribution policies therefore play an important role in countering market-driven inequality. This is primarily achieved through a combination of policies such as the minimum wage that compress wage dispersion and a heavy tax and transfer system that reduce inequality of disposable income.

Challenges to the French redistribution model. The traditional equality enhancing policies such as large social services and the minimum wage may have reached their limits as unemployment has become entrenched at around 10 percent and budgets have been severely stretched, with spending rising to 57½ percent of GPD and public debt to 95½ percent of GDP by 2014.

Objective. The aim is to understand better how different policies affect efficiency and inequality. There is an ample literature on the link between France’s tax and transfer system and inequality ex post, but the relationship between labor and fiscal policies and market income inequality has featured less prominently. The aim here is to shed light on two broad trade-offs. First, some policies aimed at lowering inequality come at an efficiency cost, not just budgetary but also possible losses in output and employment (“efficiency”). Second, while some policies may reduce inequality ex post, they may worsen somewhat market income inequality—in other words, part of the policy is self-defeating. To assess these tradeoffs, it is necessary to look at the general equilibrium effects—which is the main contribution of our paper to the literature.

Research questions. How do policies aimed at boosting employment and output affect inequality? How efficient are policies aimed at reducing inequality? Is there room for fiscally neutral (“win-win”) policies that improve output, employment, and the income distribution?

Methodology. We extend the general equilibrium model calibrated for France by Alla, Espinoza, and Pérez Ruiz (2015) with an inequality block, including a Lorenz curve and a Gini index constructed from French data on labor, capital income, and publicly provided goods (health and education). We then shock the model to inspect how a suite of stylized labor and fiscal policies affect output, the fiscal deficit, and inequality through the entire income distribution.

Structure of the paper. Section II briefly discusses French inequality profile in international comparison. Section III presents the general equilibrium model and the empirical Lorenz curves used to gauge the impact of policy shocks on efficiency, inequality, and fiscal outcomes. Section IV presents the simulation results for a suite of stylized policy measures akin to the ones recently adopted to (or being considered by) the authorities. Section V concludes.

II. Inequality in France: A Snapshot

France’s inequality profile. France’s labor income inequality is lower, and capital income inequality is higher, than the OECD average, reflecting a high concentration of capital income at the top. This results in overall market income inequality that is relatively high in France. A heavy tax and transfer system then brings the degree of (post-fiscal policy/disposable) income inequality to around the OECD average.

A01ufig01

Pre- and Post-Policy Household Income Inequality

(In Gini index, 0 to 1; impact in difference in Gini)

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: Joumard and others, 2012.

Labor income inequality. Labor income inequality (among the working-age population) is comparatively low in France. This reflects a number of factors including wage compression (partly due to minimum wage policies) and low part-time employment.3 Among the OECD countries, labor income inequality in 2011 was highest in Mexico, Israel, Chile, the United Kingdom, and the United States, while Greece, Italy, Iceland, Switzerland, and the Czech Republic were the most equal countries.

Market (or pre-fiscalpolicy) income inequality. Significant capital income concentration at the top more than offsets the relatively low labor income inequality in France. Piketty (2008) relates the upward trend in capital income inequality over the last 30 years to the rise in inheritances and inter vivos gifts,4 as capital income and wealth taxes came down. As a result, household market income inequality is above the OECD average.

Disposable income (post-policy) inequality. Relying on a heavy transfer system, disposable (or post-policy) income inequality is brought down to around the OECD average. Among the OECD countries, post-policy inequality in 2011 was highest in Chile, Mexico, Turkey and Italy, while Denmark, Belgium, Finland, and Slovenia were the most equal countries. While France relies on a number of significant large transfers and medium-sized transfers, the overall progressivity of the system remains below the OECD average (Figure 1).5 On the expenditure side, less progressive old-age pensions dominate more progressive cash transfers. On the revenue side, the tax mix relies more heavily on consumption taxes than on personal income taxes, compared with other countries.

Figure 1.
Figure 1.

France: Inequality Indicators

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: OECD.1/ Percentage point reduction in the Gini index scaled by the share of transfers/taxes in households’ disposable income.
A01ufig02

Size and Progressivity of Transfers

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: Joumard and others, 2012.

III. Analytical Framework

A. The Lorenz Curve

Lorenz curve and Gini coefficient. We construct Lorenz curves for labor force participants from actual French data on labor, capital income, and publicly provided health and education services. While the Gini coefficient provides a simple snapshot of overall income dispersion, Lorenz curves allow us to inspect the income redistribution for the whole population. This is particularly relevant when pre- and post-policy Lorenz curves intersect. In such cases, different redistribution configurations may be associated with the same overall change in the Gini coefficient, making the Lorenz curve a more informative way to assess inequality, with “winners” and “losers” from the simulated policy measure.

Labor income distribution. An empirical Lorenz curve is constructed using detailed net of taxes labor income data for different groups of workers (Figure 2, left chart). The labor force is broken down into unemployment benefits recipients (Unedic), workers in private and state-owned enterprises (INSEE), and public servants (INSEE and the 2014 finance law LPF). This dataset is then used to generate the income distribution for the combined group of workers (taking into account the size of each group). Where centile data is not directly available (all groups except workers in private and state-owned enterprises), decile data was linearly interpolated to generate centile data.

Income distribution inclusive of capital and public goods (Figure 2, right chart). Capital income and the income equivalent of the provision of two key public goods (education and health services) were also added to the net labor income distribution to construct a more comprehensive view of inequality in France. Decile data from INSEE (Enquête Revenus fiscaux et sociaux 2012 et séries longues) on capital income distribution and quintile data from Amar and others (2008) on the provision of public services for households were interpolated to complete the income distribution of net market income inclusive of public goods (Figure 2).

Figure 2.
Figure 2.

France Income Distribution, 2011

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: Unedic, INSEE, 2014 finance law, and IMF Staff calculations.

Inequality by labor, capital income, and publicly provided goods. The Gini coefficient for the Lorenz curve on net labor income is estimated at 0.28 (Table 1, Figure 3). Recipients of unemployment benefits have the lowest average net income (EUR 1,200 per month), around half of the most favored group (public servants in the central administration). The average net wage in the private sector is around 10 percent lower than the wage of public servants in the central administration. This group is the most heterogeneous, with a Gini coefficient of 0.29, significantly higher than the Gini coefficients in the public sector of around 0.18. Capital income makes the income distribution more unequal, given capital income concentration on wealthier individuals—the Gini index increases to 0.35, up from a labor income Gini coefficient of 0.28. By contrast, publicly provided education and health services goods make the distribution of income more egalitarian—the Gini index declines to 0.27(see Figure 3 for the Lorenz curves).

Table 1.

Labor Net Income for Different Groups of Workers

(Summary Statistics 2011)

article image
Sources: 2014 finance law LPF, INSEE, and Unedic.
Figure 3.
Figure 3.

Lorenz curves for different groups and different measures, 2011

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Note: Lorenz curves based on French data for unemployment benefits, net-of-tax labor income (for private sector workers, public servants, and workers in state-owned enterprises), net-of-tax capital income, and income equivalent of the provision of health and education services.Source: INSEE, Unédic, 2014 finance law LPF, and Amar and others (2008).

B. The Model

Model’s overview. We develop a one-sector, four-factor general equilibrium model with different wage-setting mechanisms for skilled and unskilled labor, and with a productive role for public spending (Alla, Espinoza and Pérez Ruiz, 2014). The model is solved for the short run, where capital is fixed, and for the long run, where capital is allowed to adjust until its return is equal to the world interest rate. For small variations in the demand for unskilled labor, wages of the unskilled are determined by the minimum wage. Wages of the skilled are determined by a labor supply function which is determined by unemployment and the real (after tax) wage. There are workers willing to take up jobs (involuntary unemployment) at all wage levels, but the amount of unused labor force is highest at the minimum wage level. Both the skilled and unskilled labor forces are constant. The model structure is discussed below.

Output and the labor market. Output is produced by means of a two-stage CES production function using four factors of production: skilled workers, unskilled workers, private capital, and a public good, which is an externality to the private factors of production (see Baxter and King, 1993). The first stage combines effective labor with private capital and the public good to produce output with an elasticity of substitution σ1 = 1/(1-ρ1):

Y=Gγ(δlLρ1+δkKρ1)1ρ1

where Y, K, G, L denote output, private capital, the public good, and effective labor. y is the elasticity of output to the public good. Effective labor is a CES combination of skilled and unskilled workers with elasticity of substitution σ2 = 1/(1-ρ2):

L=(δlsLsρ2+δluLuρ2)1ρ2

where Ls and Lu are skilled labor and unskilled labor. The formulation implies constant returns to scale in the private factors of production and equal degrees of substitution between private capital and the different types of labor. The public good is produced with a constant returns to scale combination of public skilled workers (Lg) and goods (M), with a Cobb-Douglas functional form and parameter ω:

G=M1ωLgω

We assume all public servants are qualified (around 85 percent of new public servants have completed three years of higher education (DGAFP, 2012)). Public workers’ wages are assumed fixed in the short-run, and follows the private skilled workers wages in the long-run.

Firms are assumed to maximize profits, equal to the returns on capital:

RK=π=Y(1+Tp,u)WuLu(1+Tp,s)WsLs

where π, R, Wu, Ws, and Tp,u, Tp,s represent profits, the gross rate of return on capital, unskilled wages, skilled wages, and the employers’ social security tax rate, differentiated for unskilled and skilled workers. The output price, wages, tax rate, and capital stock are taken as given by the firm. The output price is taken as the numeraire and is set equal to one. The first order conditions for profit maximization imply that the labor costs, differentiated by type of worker (LCu, LCs) equalize marginal productivity:

YLu=LCu=(1+Tp,u)WuYLs=LCs=(1+Tp,s)Ws

Substituting the expressions for marginal products from the CES production function and using the chain rule for differentiation one obtains the following labor demand functions:

LCs=(Gγ)ρ1δlδlsY1ρ1Lρ1ρ2Lsρ21LCu=(Gγ)ρ1δlδluY1ρ1Lρ1ρ2Luρ21

Wage formation for skilled workers is featured by a “wage curve” (Blanchflower and Oswald, 1994), which can be rationalized by a variety of models, including bargaining, efficiency wage, and implicit contract models. After-tax wages for skilled workers are assumed to be an increasing function of the reservation wage, which is equal to the (constant) after-tax unemployment benefit and a decreasing function of the unemployment rate of the skilled labor force:

Ws=S(θs,Vs)1TnTpe

where Vs, θS, Tn and Tpe stand for the unemployment rate of skilled workers,6 the after-tax unemployment benefit,7 personal income and employee social security tax rates respectively.8

Skilled wages are not sensitive to the minimum wage in the short run and depend on the minimum wage with an elasticity ηMW of 0.2 in the long run. The (short- and long-run) wage elasticity to unemployment benefits is set at 0.29 (Graafland and Huizinga, 1999).

Unskilled wages formation differs according to the time horizon under consideration. In the short run, after-tax wages for unskilled workers (Wu) are assumed to depend exclusively on the minimum wage W_, with an elasticity of ηMW calibrated at 0.2 (see Aeberhardt and others, 2012, Cette and others, 2012). In the long run, we assume unskilled (real) wages depend on the minimum wage with an elasticity ηMW of 0.4, twice as high as that for skilled10.

Shortrun:Wu=AW_ηMWLongrun:Wu=AW_ηMWS(θu,Vu)1TnTpe

The accumulation of capital is very important for our results. We distinguish two horizons for the results, and a crucial difference is in what is assumed about capital accumulation (see below for more details).

The government. Finally, the budget surplus is given by:

Bs=(Tn+Tk)RK+(Tp,u+Tn+Tpe)WuLu+(Tp,s+Tn+Tpe)WsLsθuUuθsUsLCsLgM

where Tk is the capital income tax rate (which is added to the personal income tax to capture the entire tax burden on capital income), and where it is assumed that the labor cost of public servants is equal to the labor cost of skilled workers employed in the private sector. Other components of the budget are not included as they are not affected by the variables in the model.

We report simulation results in deviations from baseline. To this aim, the model is log-linearized. The log-linearized equations are presented in Appendix B. In the results tables, all variables represent percentage changes, with the exception of the budget surplus, bs, which is expressed in percent of GDP.

C. Short- and Long-Term Solutions

We describe here the model’s behavior over both a short- and a long-term horizon.

Short-term solution. The short-run solution is computed assuming capital does not have time to adjust. Although in the short run investment will be reacting to economic conditions, the impact on the stock of capital of changes in the investment rate will be marginal and can be neglected. Nevertheless, to illustrate the impact of fiscal measures on investment, we present in all simulations the variations in investment (although investment does not have a feedback effect on the other variables). Investment, I, is simply assumed to be a positive function, g, of the ratio of the after-tax rate of return on capital, R, to the foreign rate of return, Rf. This presumes that a larger the rate of return differential is associated with greater incentives for investment:

I=g((1TnTk)RRf)

Employment and output respond to tax reform through the interaction of labor demand and supply. As postulated by conventional tax incidence theory (“Dalton’s law”), the employment effects depend critically on the elasticities of labor demand and supply (Figure 3).11 The more inelastic the labor supply, the more labor appropriates the “room” created by a reduction in social security taxes through higher wages, thus the smaller the employment impact. Simply put, a factor which is provided inelastically bears the full burden of a tax (in this case, it appropriates the full benefit of a tax reduction). Inversely, the more elastic the labor supply, the larger the effect of tax changes on employment. In particular, the assumption that in the short run unskilled wages are fixed to the minimum wage means that the unskilled labor supply is fully elastic and thus employment responses are maximal.

Long-term solution. The long-run solution is computed assuming capital has reached a new steady-state, where the rate of return on capital (net of the capital and personal income taxes) is equal to an exogenous world interest rate (and is thus unchanged by the fiscal measures). The equalization of the net rate of return on capital to this exogenously given world interest rate pins down the new steady-state stock of capital. The baseline assumption that the real rate of return on capital always comes back to its initial value is arguably extreme, and its foundation—the hypothesis that rates of return are equalized across countries—has been found to be violated in empirical data (see, e.g., Frankel 1985). It is natural to thus check the robustness of our results to different assumptions on the convergence of real rates of return (section IV. C).

  • Labor tax reform. In the long run, capital receives positive impetus as a result of a labor tax reform, which amplifies the employment effects from tax wedge reductions. The higher employment generated in response to the labor costs reductions will increase the marginal productivity of/rate of return on capital. Capital will therefore rise to eliminate any excess returns. As capital increases, the marginal product of labor will in turn be given a positive boost, which will stimulate job creation and so on. The potential for job creation ultimately depends on the elasticities of labor and capital supply and demand, which we also assume are different in the long run following EC (1994). Focusing on labor, if labor supply is, e.g., infinitely elastic, as postulated for unskilled workers, wages will remain constant in response to labor tax cuts, employment gains will be large, and so will be the increase in the marginal productivity of capital. By contrast, the labor tax cuts will be fully appropriated by workers with no positive effects for employment if the labor supply is inelastic.

  • Capital tax reform. A higher tax wedge on capital income (coming from specific capital income taxes or from the personal income tax) requires a long-term increase in the gross rate of return on capital (to equalize net returns) and thus on the marginal product of capital, which leads to a fall in the stock of capital and in output. This reduction of output lowers labor demand and eventually wages and employment.

What distinguishes the segmented approach taken in this paper from standard general equilibrium models is the presence of differentiated labor supplies. In the calibration for France (see Section III.D below), we assumed that the labor supply is truncated at a wage level equal to 1.5 times the minimum wage. The labor supply is highly elastic up to that level in the wage distribution, but upward sloping thereafter.

D. Calibration

Calibrated parameters in the model are reported in Table 2.

Table 2.

Model Parameterization for France

article image
Source: sources as quoted and authors’ calculations.

Underlying gross wage distribution was provided by DARES (direction de l’animation de la recherche, des études et des statistiques, France Ministry of Labor)

Nijkam and Poot’s meta-analysis covers a wide range of countries. The mean estimate is -0.07. We nevertheless take a larger elasticity because our focus is on the wage curve for skilled workers. The average elasticity is then -0.07 in the short term (unskilled workers wages are fixed). Studies on French data have not reached a consensus on whether the elasticity in France is lower or higher than in other countries (Montuenga et al., 2003 ; Gianella, 2005; Estevao and Nargis, 2001; Delteil et al, 2004).

Calibration challenge. The calibration challenge is fitting the labor supply into the two “stylized” groups of low skilled (infinite elastic supply) and higher skilled (more inelastic supply). The criterion chosen to divide the population between the two groups is observed wage behavior. Estimates put the breaking point (where wages move from being dominated by the minimum wage to responding to market conditions) at 1.5 times the minimum wage (Aeberhardt and others, 2012; Goarant and Muller, 2012). This cutoff point implies that the share of unskilled workers is 44 percent and the share of unskilled labor costs in total labor costs is 28 percent (DARES).

Wage elasticities to unemployment. The differential wage elasticities to unemployment between the two classes of workers reflect the different market conditions they face. Employment rates are lower for low-wage earners12 and sectors employing unskilled workers tend to face a higher elasticity of demand with respect to prices, thus a smaller proportion of labor cost reductions can be appropriated by employees in the form of wage increases. In addition, the minimum wage acts as an anchor in this segment of the labor force. The lower incidence of unemployment and more prevalent use of open-ended contracts, would lend skilled workers higher power in wage bargaining negotiations, justifying a relatively steep labor supply curve.

Our baseline calibration assumes that this elasticity (ψ) is −0.10 (Table 2) for skilled workers.13 For unskilled workers, we assume that wages: (i) are fixed in the short run (reflecting the constraint of the minimum wage) (ii) adjust in the long run with the same elasticity than skilled workers wages ψu = ψ = −0.1. In the baseline solution, such a calibration implies that half of the tax cut is passed through to higher wages.14

As explained above, we divide the private labor force using the fact that the share of unskilled workers is 44 percent. We also use DARES data to identify the share of unskilled unemployment in total unemployment, based on the reference wage. Finally, the employers’ social security tax rates are separately calibrated for skilled and unskilled labor using the wage and tax distribution’s (DARES).

Elasticities of substitution between production factors. The elasticity of output to the public good (γ) was calibrated using the meta-analysis of Bom and Ligthart (2013), who concluded that growth literature had, on average, found the output elasticity of public capital to be around 0.10. The share ω of public skilled workers (Lg) is then calibrated to match the expenditure shares of public skilled workers and goods in public spending. Using IMF (2013), we calibrate ω to 0.61.

Finally, the elasticity of substitution between skilled and unskilled labor will be crucial in our approach to assess the respective tax cuts impact on skilled and unskilled workers. We then follow Jamet (2005) that gives a recent and robust estimate for this parameter (0.3 in the short run and 1 in the long run).

IV. Policy Simulations: Efficiency-Inequality Tradeoff or Synergy?

A. Impact of Stylized Policies on Output, Inequality, and the Fiscal Deficit

Approach. We simulate the impact of a range of stylized policy measures (Table 3, Figure 4). Specifically, we explore:

  • two variants to the enacted CICE (EUR 20 billion unfinanced labor tax cuts, targeted toward the low wage earners versus broad based);

  • a series of other conventional fiscal measures, including stylized labor and capital income tax cuts, increases in public consumption and in publicly provided health and education services;

  • changes in two central labor market instruments, namely the minimum wage and the generosity of unemployment benefits;

Table 3.

Simulated Results

(In percent change deviation from baseline, unless otherwise stated)

article image
Source: IMF Staff calculations.Note: All simulations have equal ex ante cost of EUR 20 billion except minimum wage, which is a 5 percent increase relative to the baseline, and unemployment benefits, which has an ex ante cost of EUR 5 billion.
Figure 4.
Figure 4.

Policies and the Income Distribution

(Long run impact on Lorenz curves based on labor and capital income; In percentage point difference from the baseline)

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: IMF Staff calculations.Note: Charts show the difference from baseline cumulative income at each decile (a y-axis value of X at the ith decile indicates that the poorer ith decile individuals accumulate X percentage points more of total income relative to baseline). The difference in Lorenz curve is the difference between the post- and the pre-Lorenz curves. A positive shaded area indicates a reduction in inequality and vice-versa.

All simulations have equal ex ante cost of EUR 20 billion, except the minimum wage scenario, which is a 5 percent increase relative to the baseline, and the unemployment benefits simulation, which has an ex ante cost of EUR 5 billion. The discussion below focuses on the intuition underlying the main results, with a focus on the long run, once all variables have fully adjusted to policy shocks (for a comprehensive overview of the model’s results at both short- and long-term horizons, see Appendix C).

Stylized findings. All fiscal expansions improve output, with the response being more pronounced in the long run where capital accumulation responds positively to the shock. While increases in the minimum wage, public employment, and health and education spending worsen the fiscal position in the long run, tax cuts are de facto self-financing. This owes to a very strong response of capital accumulation (hence output and employment) to tax cuts, particularly capital income tax cuts, under the model assumptions of perfect long-run capital mobility and perfect competition.15 A crossing of the Lorenz curve occurs in about three quarters of the policy simulations, implying that value judgments about the impact on inequality impact of the policy depend on social redistribution preferences between segments of the income distribution, and cannot be captured by the Gini coefficient alone.

Simulation 1. EUR 20 bn unfinanced labor tax cuts (employer-paid social security contributions) for workers paid up to 1.5 times the SMIC. This measure cuts down the relative cost of unskilled labor and reduces unskilled unemployment considerably. By fostering profits and capital accumulation, targeted labor taxes can deliver significant output gains that are so significant to make the tax cuts fiscally self-financing in the long run.16 This measure considerably reduces inequality overall, more markedly so at the lower end of the income distribution, given the boost to unskilled employment and the increase in unskilled relative wages (the targeted tax cut is partially pass through to wages in the long run).

Simulation 2. EUR 20 bn unfinanced labor tax cuts (employer-paid social security contributions) for workers paid up to 2.5 times the SMIC. A broad-based tax cut dampens the long-run output, fiscal, and distribution gains relative to the more targeted scheme above. This follows from stronger pass through to wages overall (given higher wage elasticity to employment for skilled workers)17, hence the more muted capital accumulation and output response. Inequality gains are less pronounced too, overall and at the bottom of the distribution, given relatively lower employment creation and falling unskilled relative wages in the long run.

Simulation 3. EUR 20 bn cut in corporate tax revenues (2.2 percentage point cut in dividends income tax rate). By boosting profits and capital accumulation (wages are fixed in the short run), this measure has a strong impact on output in the long run, thus becomes fiscally de facto self-financing. The strong impact of the corporate tax rate on output in the model is due to the assumption of perfect capital mobility over the long run. Inequality is worsened overall (in terms of the Gini coefficient) but the income distribution is improved at the bottom given large employment gains for the unskilled, which more than offset falling unskilled relative wages. Lower unskilled relative wages owe to a relatively higher employment response (relatively lower wage response) than for skilled workers, given higher elasticity to employment for the latter.

Simulation 4. EUR 20 bn (or 1.4 percentage point cut) in average income tax rate.18 The measure reduces the tax wedge on both capital and labor income. The effect of the cut in capital income taxation is similar to that described in simulation 3, with a sizable long-run effect on output, employment, and a worsening of unskilled relative wages and of income distribution. Despite the personal income tax cut being symmetric, it has a differentiated effect on skilled versus unskilled workers: unskilled labor costs are unchanged because of the minimum wage while skilled labor costs are reduced to some extent, thus boosting skilled employment. Overall, this measure boosts output and reduces inequality in the short run, but inequality in the long-run is worsened: unskilled relative employment is higher than in the baseline, but this is more than offset by the reduction in unskilled relative wages and the additional income accruing to capitalists as a result of capital accumulation.

Simulation 5. EUR 20 bn increase in the public wage bill by rising public employment by 390,000 workers. Higher public employment hampers capital accumulation, hence deteriorates output and the fiscal deficit in the long run. It has important redistribution effects for barely unchanged Gini coefficient. The initial positive output response (public employment participates to the provision of public goods) is more than reversed in the long run as rising skilled wages19 increase labor costs in the private sector and profits are shrank. Inequality at the bottom is reduced because of the reduction in unemployment, but inequality at higher percentiles is worsened because of the increase in skilled relative wages (public employment benefits the skilled workers), even though capital owners suffer from the reduction in capital and output.

Simulation 6. EUR 20 bn increase in the public wage bill by rising public wages by 7.1 percent relative to baseline. While both higher employment and higher public wages affect the wage bill, the latter has no general equilibrium effects in the model. Increasing public wages worsens the income distribution at the bottom but improves it toward the top. The overall (Gini) inequality effect is small (positive for labor income only, negative when adding capital to the income distribution). There are no other effects in the model of this measure.20

Simulation 7. EUR 20 bn increase in health and education services. Public goods can raise long-run output and employment markedly, at modest fiscal costs, all the while leaving overall inequality unchanged and benefiting the poor. Extra output and jobs follow from positive externalities from public goods onto total factor productivity. Since the output effect is positive, this measure increases tax collection and decreases unemployment benefits, which partly offsets the initial increase in the fiscal deficit. Raising the provision of public goods benefits the poorer half of the population—those with higher propensity to consume public health and education services.

Simulation 8. A 5 percent increase in the minimum wage. By reducing profits and capital accumulation, a higher minimum wage worsens long-run output, the fiscal position, and inequality, particularly among the poor. Inequality becomes more pronounced within the poorest quintile of the labor force due to upsurge in unskilled unemployment, which outweighs redistributive effects from higher unskilled relative wages. However, the overall Gini coefficient improves as low-wage workers that remain employed earn higher wages.

Simulation 9. EUR 5 bn increase in the unemployment benefits. Higher unemployment benefits boost income for the unemployed and increases the reservation wage for both skilled and unskilled workers. This raises the negotiation power of workers, thereby increasing net wages and labor costs. In the short run, labor income increases benefit all workers and income inequality is reduced, even though some jobs are lost by the increase in labor costs. In the long run, the reduction in output and employment is large because capital accumulation is hampered significantly by the increase in labor costs. Skilled wages suffer (whereas unskilled wages remain supported by the higher reservation wage). The effect on income distribution is a combination of higher unemployment (which hurts income in the lower two deciles), a smaller skill premium, and large losses for capital owners. Overall, in the long run the Gini coefficient is reduced but the employment losses are large.

B. Policy Implications

Efficiency-enhancing policies. Our long run simulation results suggest that, for a given ex ante EUR 20 bn fiscal cost, capital income tax cuts are best for boosting employment and output and improving the fiscal position, followed by flat income tax cuts, targeted labor tax wedge cuts, broad-based labor tax wedge cuts, and increases in public goods (Figure 5, top panel). However, when inequality is taken into account, targeted labor tax wedge cuts appear superior as they deliver sizable output and employment gains, all the while significantly reducing overall inequality and improving income distribution at the bottom. A higher provision of public goods could also have very beneficial effects, provided positive externalities from education and health are substantial. Turning to the short run, there are practically no tradeoffs involved between efficiency and inequality across the various policy instruments, given the tepid output response under fixed capital.

Figure 5:
Figure 5:

Policies Impact on Output and Inequality

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: Authors’ calculations based on model results.

Revisiting conventional inequality instruments. Our simulations suggest that long-standing policies aimed at reducing inequality, in particular the minimum wage, may have unintended consequences on the income distribution. By squeezing profit margins, minimum wage increases discourage investment and depress output. As unskilled labor becomes relatively expensive, more unskilled workers lose their jobs. This worsens the income distribution at the bottom despite the increase in unskilled relative wages. If societal values gave priority to improving the situation of the poor, for example, the first quintile of the distribution, our results suggest the targeted labor tax wedge cuts are the most effective instrument, followed by untargeted labor tax wedge cuts, personal and corporate income tax cuts. Increases in public employment or the minimum wage hurt the poor (through lower unskilled relative wage and employment, respectively) (Figure 5, middle panel; and Figure 6).

Figure 6.
Figure 6.

Policies and the Income Distribution

(Long run impact total income by decile, x-axis; in percentage difference from the baseline, y-axis)

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: IMF Staff.

Fiscally neutral win-win policies. A package of targeted labor tax reductions offset, ex ante, by cuts in public employment (or wages) is a long-term “win-win” strategy in terms of efficiency, inequality, and fiscal outcomes, provided the higher fiscal deficit can be accommodated in the short run (Figure 5, bottom panel). An alternative package could consider cuts in public wages as an offsetting measure, provided those cuts do not hamper public servants motivation and productivity, as is assumed in our model.

C. Robustness of Results

This section discusses the robustness of simulated results, in the long run, to key parameters in the model, including capital mobility, wages elasticity to unemployment (ψ), the elasticity of substitution between effective labor and capital (σ1), the elasticity of output to the public good (γ), and the wages elasticity to the minimum wage (ηWM).21 The results from these robustness checks are reported in table 4.

Table 4.

Robustness to Key Parameters

(Deviation from baseline)

article image
Note: All simulations have equal ex ante cost of EUR 20 bn, except the minimum wage scenario, which is a 5 percent increase relative to the baseline. Central value corresponds to baseline and high and low values pertain to sensitivity analysis. All values are plausible according to the literature.

Based on Lorenz curve on capital and labor income.

ηWM (long run, skilled) is maintained at 0.2.

Capital mobility (εK). Capital market imperfections, risk-return characteristics in the allocation of portfolios, complementarity between the presence of physical capital and the ability to reach local markets, are all likely to reduce the sensitivity of capital accumulation and output to the tax rate on capital.22 In addition, restrictions to competition in product markets may inhibit the capital response in the long run. The extent to which the net rate of return on capital converges back to its initial value (Figure 7, top panel) is found to be important for the ex-post levels of output and the fiscal deficit, but has relatively moderate effects on income inequality (Figure 7, middle and bottom panels). In general, the positive effects of fiscal expansions on the Gini coefficient are strengthened with imperfect capital mobility because the reduction in inequality due to the lower share of income accruing to capital owners dominates the worsening inequality owing to the more muted employment gains obtained when capital accumulation is hampered (Figure 8).

Figure 7.
Figure 7.

Robustness to Imperfect Capital Mobility

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: Authors’ calculations based on model results.
Figure 8.
Figure 8.

Policies and the Income Distribution: Imperfect Capital Mobility

(Long run impact on Lorenz curves based on labor and capital income; In percentage point difference from the baseline)

Citation: IMF Working Papers 2016, 041; 10.5089/9781513515960.001.A001

Source: IMF staff calculations.Note: Charts show the difference from baseline cumulative income at each decile (a y-axis value of X at the ith decile indicates that the poorer ith decile individuals accumulate X percentage points more of total income relative to baseline). The difference in Lorenz curve is the difference between the post- and the pre-Lorenz curves. A positive shaded area indicates a reduction in inequality and vice-versa.

Wage curve elasticity (ψ). The wage curve elasticity is found to be important to assess the long term employment and inequality impact of a targeted labor tax cut, although for realistic calibrations the measure is positive for output, it reduces inequality, and is self-financed. The same elasticity (for all workers) is also a sensitive parameter to assess the impact of an untargeted tax cut: for a high elasticity (in absolute values), untargeted tax cuts can significantly reduce inequality. The higher the elasticity the higher the pass though onto wages and the employment gains. In all, the former effect dominates, and workers gain relative to the owners of capital.

Elasticity of substitution between effective labor and capital (σ1). The effect of a reduction in the corporate tax on inequality is relatively insensitive to the elasticity of substitution between labor and capital.23

Elasticity of output to the public good (γ). The productivity of public goods shapes the model response to spending on public employment and on public goods. We find that while employment and output gains are sensitive to this parameterization, the inequality impact of these measures is not.

Wages elasticity to the minimum wage (ηWM). There is significant uncertainty on the value of ηWM Since the impact of a shock to the minimum wage is a linear function of ηWM, it is straightforward to compute the sensitivity of model results to this parameter. Small changes in the wage elasticity to the minimum wage appear to have large efficiency costs, although the benefits in reducing income inequality are also sizeable.

V. Conclusion

Using a supply-side general equilibrium approach to assess the impact of fiscal and labor market policies in France, we find that fiscal expansions reduce the Gini coefficient in the long run, except for capital and labor income tax cuts (crossings of the Lorenz curve can alter the interpretation of this result). Given its positive impact on employment, a reduction of the minimum wage has an ambiguous impact on inequality: the Gini coefficient is worsened, but the first quintile improves its relative position in the income distribution because of the extra hiring. We also find scope for “win-win” policy packages that could improve overall efficiency, inequality, and fiscal outcomes, for instance if targeted labor tax reductions are offset by cuts in the public wage bill.

While our model is calibrated to France’s economy, there are a number of simplifying assumptions that may affect the actual impact of policies, particularly for the long-term horizon. For instance, the positive impact of labor tax may be weaker if, for instance, social partners end up “using” the tax cut for raising gross wages. Another simplifying assumption is perfect capital mobility over the long term, which contributes to the finding that tax reductions are self-financing over the long term. In practice, capital market imperfections and competition restraining practices may considerably limit investment, even in the long run. If framework conditions are such that capital accumulation is hampered, capital and labor tax bases may barely grow in response to fiscal incentives and tax cuts will add to the government deficit. Notwithstanding these model limitations, the analysis provides an analytical starting point for comparing the output, employment, fiscal, and distributive effects of different policies.

How Do Fiscal and Labor Policies in France Affect Inequality?
Author: Mr. Raphael A Espinoza and Mrs. Esther Perez Ruiz
  • View in gallery

    Pre- and Post-Policy Household Income Inequality

    (In Gini index, 0 to 1; impact in difference in Gini)

  • View in gallery

    France: Inequality Indicators

  • View in gallery

    Size and Progressivity of Transfers

  • View in gallery

    France Income Distribution, 2011

  • View in gallery

    Lorenz curves for different groups and different measures, 2011

  • View in gallery

    Policies and the Income Distribution

    (Long run impact on Lorenz curves based on labor and capital income; In percentage point difference from the baseline)

  • View in gallery

    Policies Impact on Output and Inequality

  • View in gallery

    Policies and the Income Distribution

    (Long run impact total income by decile, x-axis; in percentage difference from the baseline, y-axis)

  • View in gallery

    Robustness to Imperfect Capital Mobility

  • View in gallery

    Policies and the Income Distribution: Imperfect Capital Mobility

    (Long run impact on Lorenz curves based on labor and capital income; In percentage point difference from the baseline)