Germany
Selected Issues and Statistical Appendix
Author: International Monetary Fund
Publication Date:
28 Oct 1998
eISBN:
9781451810325
Language:
English
Keywords:
ISCR; CR; wage; government; labor market; government level; employment rate; budget balance; real wage growth; skill distribution; Employment; Labor markets; Real wages; Wages; Europe; tax pass-through; wage demand; labor market condition; employment objective; employment situation; Labor taxes; employment effect
Download PDF (2.1 MB)

96. Since unification, the drop in employment has been at the core of Germany’s weak labor market performance. High unemployment boosted social spending while the accompanying decline in the base for income taxes and social contributions posed problems for fiscal revenue raising efforts. This chapter focuses on the interaction between rising social spending, revenue requirements, and labor market institutions which can give rise to a vicious circle of weak labor market performance and deteriorating fiscal positions.

Abstract

96. Since unification, the drop in employment has been at the core of Germany’s weak labor market performance. High unemployment boosted social spending while the accompanying decline in the base for income taxes and social contributions posed problems for fiscal revenue raising efforts. This chapter focuses on the interaction between rising social spending, revenue requirements, and labor market institutions which can give rise to a vicious circle of weak labor market performance and deteriorating fiscal positions.

III. Labor Market Trends and Fiscal Dynamics: Some Simulation Evidence51

A. Introduction and Summary

96. Since unification, the drop in employment has been at the core of Germany’s weak labor market performance. High unemployment boosted social spending while the accompanying decline in the base for income taxes and social contributions posed problems for fiscal revenue raising efforts. This chapter focuses on the interaction between rising social spending, revenue requirements, and labor market institutions which can give rise to a vicious circle of weak labor market performance and deteriorating fiscal positions.

97. In this chapter, a macroeconomic growth model is calibrated for Germany to illustrate the dynamic interplay between labor taxation, employment, and social spending. This technique was selected in order to focus narrowly on key conceptual relationships, which would be clouded in a more realistic, but necessarily more complex structural macroeconomic model. The calibration of the model and the simulation results are designed to be illustrative; they are not intended to be predictive. Moreover, unlike the previous chapter, which focused on a disaggregated characterization of the labor market, this model is an aggregated one. The two chapters should be viewed as complementary, and not competitive, explanations.

98. The simulation results illustrate that in response to fiscal shocks, existing labor market and fiscal structures can interact to produce a vicious circle, with declining employment rates, increasing taxes, and slower output growth. Behavioral changes in the labor market (as captured by parametric changes to the wage-setting equation) can partly offset the effects of the vicious circle. Wage moderation improves employment, output and fiscal performance in the short and long run. Because real wage growth has a strong effect on the overall economy, an unsustainable situation may be reached at relatively low rates of real wage increase. Allowing taxes to affect net wages can mitigate the adverse effects of shocks on employment, output, and the fiscal balance over the medium run. Indeed, improving the downward flexibility of real wages would reduce the employment effects of adverse shocks, while preserving the full benefits from positive shocks. Population aging has effects similar to a series of adverse fiscal shocks under current social insurance arrangements. The simulations suggest that the above-mentioned structural changes would have powerful mitigating effects in the medium run, but long-run trends may remain unsustainable without further pension reforms.

99. The rest of the chapter is structured as follows. Section B describes the model and discusses key parameter values. Section C presents illustrative scenarios to highlight the interaction between labor market characteristics and fiscal policy, and to demonstrate that behavioral changes in the labor market can have beneficial effects for employment and output.

B. Structure of the Model

100. A simple aggregate growth model is calibrated to German specifications to illustrate the interaction between the labor market and the fiscal sector.52 The model has many simplifications (e.g., only the real economy is modeled; the business cycle is not considered; the economy is closed to trade). However, it is these same simplifications that allow us to focus on the dynamic interaction of social spending, labor taxation and employment. The major building blocks are: (i) the production sector; (ii) the labor market; and (iii) the fiscal sector. The main equations of the model are outlined below, while technical details are presented in the Appendix.

Production sector

101. Producers use a Cobb-Douglas technology:

Yt=AtKtαLt1−α,

where Y, K, L stand for real output, real capital stock, and labor input, respectively, and A denotes total factor productivity. As the income share of capital for Germany has been estimated at about 30 percent,53 the parameter α is set at 0.3, which is consistent with estimated values for other industrial countries.

102. The Cobb-Douglas production function, combined with the assumption that productive factors are paid their marginal product,54 implies that the income shares for labor and capital shares are constant in the long run. The wage share is technologically determined, with the trade-off between employment and wages in the hands of the tariff partners—the trade unions and the employers.55 The constant labor income share is consistent with different combinations of wages and employment, which are determined by wage-setting behavior. In particular, higher wages necessitate lower employment, given the assumptions. In a competitive market, wages would be set such that full employment would prevail. If, say, trade unions demand higher wages, employment declines so that the marginal product of labor increases to match the real wage. Structural unemployment emerges as the result.

103. Because investment and disinvestment are assumed to be costly, the capital stock responds sluggishly to its marginal product.56 In addition, government spending is assumed to crowd out private investment; hence capital accumulation will depend negatively on the government expenditure-to-GDP ratio G,57

dKt/Kt = I(MPKt-1,Gt-1/(Pt-1Yt-1)).

104. Social returns to capital are assumed to be larger than private returns. Consistent with the literature on endogenous growth, total factor productivity (TFP) growth depends positively on the growth rate of the capital stock:

dln(At) = θadln(Kt).

105. TFP growth is calibrated at 1.2 percent per annum at the historical capital stock growth rate (3 percent). Faster capital growth would yield higher TFP growth. Long-run output growth would converge to zero without input growth. Technological progress is represented by TFP growth, and thus the possibility of biased technological progress is excluded.

106. Producers are price takers in the capital market, and hence capital is paid its marginal product. Payments to labor and the quantity of labor input are determined in the labor market, described in the next subsection.

Labor market

Labor demand

107. In the absence of adjustment costs, labor demand would be determined such that the gross wage equals the marginal product:

wt=(1−α)AtKtαLt−α.

where Σi λi = 1. Given that the capital share, α, is 0.3, the long-run elasticity of employment with respect to real wages is -3.33, i.e., employment declines by more than 3 percent in response to a 1 percent increase in real wages.58

Wage setting

109. The labor force is assumed to be organized by a trade union. The trade union is a monopolist and sets the wage unilaterally. Real wage demands by trade unions depend on labor market conditions. Slack in the labor market, as indicated by low employment rates in the previous period, depresses real wages. If wage-setting behavior is governed by insiders (e.g., workers who are currently employed), labor market conditions would have a limited effect on real wages. Furthermore, it is assumed that after-tax, rather than gross, real wages are targeted by the trade union. (This is consistent with workers regarding government spending as less utility enhancing than private spending.) In addition, there is a rate of “autonomous wage growth”, which is assumed to capture, inter alia, the impact of trend productivity growth on wage setting. Finally, in the short run sluggish wage adjustment is assumed, reflecting staggered wage contracts. The wage-setting equation takes the form:

dln(wt) = θdln(wt-1)+βdln(et-1)+γdln(τt-1)+ω),

where e denotes the employment rate defined as the ratio of employment to labor force (et=L/Lt*); τ represents the “tax wedge” between net and gross wages which is defined as the ratio of gross to net real wages (τt=wt/wtN); and ω is autonomous wage growth. The labor supply Lt* is given exogenously based on the growth rate and the age structure of the population. From the assumption of a full tax pass-through, γ=1-θ is imposed.

110. Empirical evidence suggests that trade unions in Germany have successfully shifted wage taxes to producers, which supports the assumption of a full tax pass-through. Tyrväinen (1995) finds full long-run tax shifting into real labor costs in Germany. Alesina and Perotti (1997) find near-full tax shifting (about 60-75 percent) into real relative unit labor costs for a group of countries—including Germany—characterized by moderately centralized bargaining practices.59 Other studies, as reviewed in Leibfritz, Thornton, and Bibbee (1997), also indicate that increases in labor taxes are not fully absorbed by net real wages. The extent of tax shifting, however, remains a controversial subject.60 On balance, though, much of the empirical literature places Germany on the list of countries where significant tax shifting occurs. Consequently, and to establish a baseline, a full long-run tax pass-through to real labor costs is assumed, i.e., γ=(1-θ) holds in the wage-setting equation. This restriction will be relaxed later.

111. The elasticity of real wages with respect to the employment rate—the parameter β—indicates the degree of real wage flexibility. The higher the elasticity, the more sensitive real wage demands are to the employment situation. In a competitive labor market, the value of β would be infinite.61 In contrast, in an economy where wages are set based on insider behavior, this parameter value would be near zero. To pick a value for β, Layard, Nickell, and Jackman (1991) were followed, who estimated a wage-setting curve for Germany. Their estimates implied a unitary long-run semielasticity of real wages with respect to the unemployment rate. This translates approximately into a unit elasticity with respect to the employment rate; hence the value was set at β=1-θ. As to the dynamics of the wage-setting equation, θ=0.5 is assumed, which is close to the value estimated by Layard, Nickell, and Jackman. Finally, autonomous wage growth was set at 0.8 percent per annum (ω=0.008), which corresponds to the average residual real wage growth over the 1980s after accounting for lagged real wage growth, the employment rate, and the tax burden.

112. Alesina and Perotti (1997), Scarpetta (1996), and Layard, Nickell, and Jackman (1991) provide some support for the parsimonious approach to modeling labor market behavior adopted here. The approach implicitly assumes that all relevant institutional factors are captured in the wage-setting equation, namely, by the values of β and ω and by the full pass-through of taxes to real labor costs. Alesina and Perotti find a mapping between the institutional setup of wage bargaining and the extent of the tax pass-through. This finding supports the use of the tax pass-through in the wage-setting equation as a proxy for more detailed modeling of the wage bargaining institutions. Further, Scarpetta shows that after controlling for the effect of a richer set of labor market institutions,62 the tax pass-through becomes insignificant in explaining the unemployment rate, again suggesting that the extent of the pass-through could be a valid proxy for labor market institutions. Layard, Nickell, and Jackman find evidence that cross-country differences in β are also related to differences in labor market institutions, as suggested by theoretical considerations.

Fiscal sector

113. A simplified fiscal structure is modeled. Over the medium term (over a four-year period), the deficit is assumed to narrow to ½ percent of GDP via discretionary changes in revenues and expenditures. This fiscal adjustment is assumed to take place in a “symmetric” manner—so that the tax burden is lowered along with the deficit, with both financed by spending cuts.63 The deficit is financed by government borrowing.

114. Over the longer run, a balanced budget is targeted, and the evolution of expenditure and revenue components is determined by corresponding policy reaction functions, which are assumed to remain stable over the simulation period. Four expenditure components are considered: pensions, unemployment benefits, interest on government debt, and other expenditures. Two alternative assumptions are used for pension expenditures. In the baseline, pension expenditures remain fixed as a percentage of GDP at their initial level. In the alternative simulation, pension expenditures increase, influenced by demographic forces.64 Expenditures on unemployment benefits are determined by the level of unemployment and the constant income replacement rate. Public sector interest payments are the product of the interest rate and government debt, which is determined by cumulated deficits. Long-term interest rates are linked to the marginal product of capital. Revenues are collected as taxes on capital and labor. Tax rates on capital are assumed constant. Labor taxes are adjusted so that the deficit converges over the longer run to the desired target (in this case, balance).

C. Simulation Results

115. This section is organized as follows. First, the baseline is presented and the sensitivity of the simulations to parametric changes is examined. Then, the implications of a temporary fiscal shock—an increase of 1 percentage point of GDP in both expenditures and tax revenues—is explored. Finally, pension expenditures are allowed to increase over time in line with prospective demographic changes. It is useful to recall at this point that the scenarios are merely numerical calculations based on calibrated behavioral equations and not projections. The results are not predictive but they do provide an illustration of the dynamic forces at play.

Baseline

116. To establish a baseline, demographic effects are filtered out from government expenditures as pension expenditures are assumed to remain constant as a share of GDP at their initial level. In the labor market, key parameters of the wage setting equations are set at the values discussed in detail earlier. In particular, the long-run elasticity of real wages with respect to the employment rate is assumed to be unity (β=0.5); taxes are assumed to be fully passed on to real wages over the long run (γ=0.5); and autonomous wage growth is assumed to be 0.8 percent per annum (ω=0.008).

117. Under the baseline assumptions, real GDP growth slows to about ½ percent per annum in the long run, while full employment is reached in about ½ periods (Table III-1, Figure III-1). The slowdown in output growth is caused largely by the decline in the labor force (about 1¼ percent per annum) and the assumed absence of labor-augmenting technological change.65 The growth of output per worker, however, remains close to 2 percent in the long run (about its historical average). The government budget turns into surplus after seven periods, and converges to its target of balance over the long run. This relatively benign scenario derives from the joint assumptions of a “symmetric fiscal policy” in the near term, and a stable expenditure-to-GDP ratio over the longer run. The “symmetric fiscal policy” allows the tax burden on labor to fall, slowing the growth in labor costs. Employment expands and output growth remains high, easing the task of deficit reduction, and thus allowing for further decreases in labor taxes.

Figure III-1.
Figure III-1.

Baseline 1/

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Assumes the baseline value for real wage flexibility, full tax pass-through, autonomous wage growth of 0.8 percent per annum; pension expenditures remain constant as a share of GDP.
Table III-1.

Scenario 1—Baseline

article image
Source: Staff calculations.

1996=1.

Annual average growth rates.

Employment as a share of labor force.

Imputed tax rate on labor; 1996=1.

118. Changes in autonomous wage growth have strong effects on the long-run outcome (Figure III-2). From an initial position below full employment, a lower trend increase in real wages—autonomous wage growth (ω)—can put the economy on a trajectory that reaches full employment fester. Halving autonomous wage increases shortens the time required to reach full employment by about five periods, and leads to higher output and faster improvements in the budget deficit (despite lower taxes on labor). Wage moderation can have two origins: (i) voluntary wage moderation to improve the labor market situation;67 and (ii) changes in the generosity of the social benefit system, which would act as a deterrent to excessive wage demands. In the first case, wage moderation is due to a shift in preferences, while in the second, to a change in incentives.

Figure III-2.
Figure III-2.

Baseline Under Alternative Assumptions About Autonomous Wage Growth 1/

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Autonomous wage growth is 0.4, 0.8, 1.0, and 1.2 percent per annum; respectively for the slower, baseline, faster, and unsustainable scenarios.

119. In contrast, faster autonomous wage increases68 lead to a prolonged time (about 15 periods) of stagnant employment. Lower employment depresses the marginal product of capital, discouraging investment, and thus adversely affecting TFP growth. Even over the longer run, the level of output is lower than in the baseline scenario. Increasing autonomous wage growth further to what appears still to be feasible levels69, results in an unsustainable economic situation. Excessive wage demands push up real labor costs, which triggers layoffs to raise the marginal product of labor to the higher level of real wages. More unemployment boosts spending on unemployment benefits and hence raises the tax rate on labor, further increasing labor costs. Although the deteriorating employment situation moderates wage demands, this moderation is insufficient to compensate for the effects of higher taxes and fast autonomous wage growth. As employment losses escalate, the marginal product of capital declines, depressing investment and TFP growth. Output and the tax base shrink further, and the budget deficit balloons despite rapidly rising tax rates.

The effects of a temporary fiscal shock

120. In this section the effects of a deficit-neutral temporary fiscal shock are considered. While maintaining the assumption of no demographic effects on expenditures, in period 4 the expenditure-to-GDP ratio is assumed to increase temporarily by 1 percentage point compared with the baseline. The temporary increase in expenditures is fully financed by higher taxes. Subsequently, the evolution of taxes and expenditures follows the standard dynamics as described above.

121. Tracing the response of the economy to this fiscal impulse reveals the interaction of the labor market and fiscal structures. The initial shock reverberates throughout the economy, and depresses employment for about seven periods (Figure III-3 and Table III-2). As the tax burden on labor rises, employment declines sharply and output levels off. Although the budget deficit is initially unchanged, it widens subsequently, owing to higher spending on unemployment benefits (unemployment expenditures increase by about 1 percent of GDP), and due to lower revenues stemming from worse output performance. The employment rate remains below its pre-shock level for about seven periods after the shock. In the long run, however, the economy reaches full employment, output grows at the same rate as in the baseline scenario, and the government budget remains close to balance. The capital stock and the level of output, however, remain below their baseline paths over the long run. Since employment is lower, the marginal product of capital is smaller, and hence capital accumulation is less rapid.70 The temporary fiscal shock thus has long-term consequences.

Figure III-3.
Figure III-3.

Fiscal Shock 1/

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Government expenditures and revenues increase by 1 percentage point of GDP in period 4.
Table III-2.

Scenario 2—Fiscal Shock

(Deviation from baseline)

article image
Source: Staff calculations.

1996=1.

Annual average growth rates.

Employment as a share of labor force.

Imputed tax rate on labor, 1996=1.

122. The reaction of the economy to the fiscal shock can be influenced by changes in wage-setting behavior. In the present framework the possible changes include the following: (i) greater real wage flexibility represented by an increase in the parameter β; and (ii) a partial tax pass-through represented by a decrease in the parameter value γ. Further, it is also possible to consider asymmetric changes in these parameter values—greater downward real wage flexibility, and partial pass-through of tax increases to real wages. These structural changes in wage-setting behavior—greater downward real wage flexibility, and less tax pass-through—can moderate the adverse effects of the fiscal shock on output and employment by modifying the dynamic interaction between the labor market and the fiscal sector. As a result, the economy could converge back to full employment more quickly, and long-run output costs could be reduced.

123. If wages are more sensitive to labor market conditions, perturbances in the labor market tend to be absorbed more by changes in real wages than by changes in employment. To illustrate how greater wage flexibility helps to stabilize employment, let us consider the effect of an exogenous increase in taxes on labor. On impact, gross real wages rise, and therefore employment declines, independent of the degree of real wage flexibility. Subsequent effects depend on the parameter value β. With more flexible real wages (higher β), real wage growth becomes more subdued in response to the initial deterioration of the employment situation, eventually leading to a smaller decline in employment over longer periods. With employment and output higher, the tax base shrinks less and fiscal pressures decrease, dampening further the after-effects of the initial shock. The mechanism is similar in the case of a favorable fiscal shock caused by a decline in the tax rate on labor. On impact, this leads to slower gross wage growth, and higher employment. Greater real wage flexibility allows real wage growth to pick up as labor market conditions improve. This dampens the initial decline in real labor costs, and partially offsets the initial increase in employment.

124. Greater downward wage flexibility introduces asymmetry into the dynamics. Assume real wage growth reacts more to a deterioration in the labor market, while an improvement in the labor market does not trigger correspondingly faster real wage growth. Consequently, the employment effect of adverse fiscal shocks would be reduced, but the beneficial effects of positive fiscal shocks would remain unchanged. Greater downward wage flexibility is modeled as a higher value in the parameter value β, but only when the employment rate drops: after observing a decline in the employment rate, unions, ceteris paribus, are satisfied with smaller wage increases. This asymmetric behavioral change is the observational equivalent of several institutional changes. For example, trade unions could become more concerned about the number of their employed members. A widening of the wage distribution allowed for by lower minimum wages and supported by less generous social assistance could also produce more downward real wage flexibility. If the alternative to staying employed becomes less appealing, the employment objective might become relatively more important than the wage objective for the trade union, yielding a modified trade-off.

125. Figure III-4 shows the effects of the fiscal shock with varying degrees of downward wage flexibility. If real wages are more flexible downward, the initial employment effect of the fiscal shock subsequently moderates real wage demands. As a result, the employment rate starts increasing rapidly shortly after the shock, and the economy reaches full employment about five periods earlier than with less downward wage flexibility. Adverse implications for output and the budget deficit remain limited.

Figure III-4.
Figure III-4.

Fiscal Shock and Downward Wage Flexibility

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Government expenditures and revenues increase by 1 percentage point of GDP in period 4.2/ Fiscal shock with greater downward wage flexibility.

126. When taxes are only partly passed through to real wages, fiscal shocks have a smaller effect on real wages, and thus on employment. In the case of a tax hike, only a portion of the increase in labor taxes boosts real labor costs because the rest is “absorbed” in lower net real wages, leading to higher employment and real growth (compared with a full tax pass-through). Conversely, in the case of favorable fiscal shocks, a partial tax pass-through mutes the beneficial effects on employment, since tax cuts are partly mopped up by higher after-tax wages instead of being fully reflected in declining labor costs. While an incomplete tax pass-through helps the employment situation in times of increasing labor taxes, it decreases the employment effect of fiscal measures aimed at reducing non-wage labor costs.71 Relaxing the assumption about a full long-run tax pass-through to employers requires γ<(1-θ) in the model. As with the change in β, this can be the reflection of several underlying behavioral or institutional shifts.72 In contrast to the conditional change in β, however, the change in the extent of the pass-through is assumed to be symmetric. With net real wages partly absorbing fiscal shocks, the employment effects of both adverse and beneficial shocks will be dampened.

127. If the increase in taxes is partially absorbed as a cut in net real wages (Figure III-5), the drop in the employment rate following the shock is smaller than under the baseline parameter values. Because net wages partly absorb not only tax increases, but also tax cuts, the subsequent rise in the employment rate is slower than with a more complete tax pass-through, so that the economy reaches full employment about the same time as in the case of a full tax pass-through.

Figure III-5.
Figure III-5.

Fiscal Shock and Partial Tax Pass-Through

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Government expenditures and revenues increase by 1 percentage point of GDP in period 4.2/ Fiscal shock with partial tax pass-through.

The effects of population aging

128. This section examines the implications of a series of fiscal shocks stemming from demographic changes and the characteristics of the German public pension system. Although public pension benefits are linked to previous contributions, the pension system is financed on a pay-as-you-go basis by social security contributions of those currently working. Thus, an increase in public pension expenditures raises the tax burden on labor. Demographic trends73 indicate that over the simulation period, the ratio of the “retirement age” population (those above 65) to the “working age” population (those between 15 and 65) will double (from around 32 percent to 64 percent). Furthermore, the increase in the elderly dependency ratio is expected to speed up considerably after period 15. The resulting steady increase in pension obligations is tantamount to a series of fiscal shocks.

129. Over the short run, major economic indicators are similar to the baseline—the employment rate rises, output grows by more than 2 percent, and the budget deficit improves (Table III-3, and Figure III-6). Shortly after period 5, however, the increasing elderly dependency ratio exerts upward pressure on pension expenditure, and thus on the tax burden on labor. As taxes are passed through to real wages, labor costs increase, and producers lay off workers to keep the marginal product of workers in line with real wages. Output growth slows, which is due in part to the decline in labor input and to slower capital accumulation. Unemployment expenditures increase, triggering a new round of labor tax hikes, higher real wages, and further labor shedding. At the same time, capital accumulation is discouraged as low employment depresses the marginal product of capital, and higher public expenditures crowd out private investment. As population aging continues to push up tax rates, the economy quickly degenerates into a vicious circle of declining employment, shrinking output, increasing taxes, and worsening budget deficits. The process becomes unsustainable by period 10.

Figure III-6.
Figure III-6.

Pension Scenario: Demographic Effects 1/

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ In the baseline scenario, pension expenditures are constant as percent of GDP; in the alternative scenario, pension expenditures include demographic effects.
Table III-3.

Scenario 3—Aging Population

(Deviation from baseline)

article image
Source: Staff calculations.

1996=1.

Annual average growth rates.

Employment as a share of labor force.

Imputed tax rate on labor, 1996=1.

130. As Figures III-7 and III-8 illustrate, structural changes in the labor market (such as greater downward wage flexibility, and a partial tax pass-through) can have powerful effects in the medium run, improving employment, real output, and the budget deficit. Over the long run, however, the sustained adverse fiscal shocks once more trigger a vicious circle. Greater downward wage flexibility (Figure III-7) helps to slow the employment decline after period 5 by increasing the sensitivity of real wages to changes in the employment rate. In response to the rapidly deteriorating labor market conditions, trade unions curtail wage demands. A partial tax pass-through (Figure III-8) slows down the increase in unemployment by buffering the adverse shock to labor costs. Both changes unambiguously improve the labor market outcome over the medium run. Relatively higher employment has beneficial effects for capital accumulation and for output growth. With a larger tax base, the rise in the tax rate on dependent labor necessitated by population aging is smaller, and thus the negative effect of fiscal pressures on employment is more contained. Over the long run, however, the structural changes in the labor market considered here may not be sufficient to combat the effects of population aging without further pension reforms.73 As the elderly dependency ratio rises, a virtuous circle becomes unattainable. Steadily increasing social contribution rates set in motion a vicious circle between the labor market and the fiscal sector.

Figure III-7.
Figure III-7.

Pension Scenario: Demographic Effects and Downward Wage Flexibility

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Pension expenditures rise in line with demographic projections.2/ Pension expenditures rise in line with demographic projections and greater downward wage flexibility is assumed.
Figure III-8.
Figure III-8.

Pension Scenario: Demographic Effects and Partial Tax Pass-Through

Citation: IMF Staff Country Reports 1998, 111; 10.5089/9781451810325.002.A003

1/ Pension expenditures rise in iine with demographic projections.2/ Pension expenditures rise in line with demographic projections and only partial tax pass-through is assumed.

APPENDIXI III-1: A Model for Simulating Labor Market and Fiscal Dynamics

131. This describes the structure of the model and its calibration. The focus of the model is to provide a framework that tracks the interaction between fiscal and labor market dynamics.

Real economy

132. The real economy is modeled according to a neoclassical production function. For simplicity and tractability, a standard Cobb-Douglas production function with neutral technological progress is assumed. The factor inputs, capital (K) and labor (L) are combined to yield output (Y) in the following manner (Tables III-A1 and III-A2):

Yt=AtKtαLt1−α,

133. The capital stock is assumed to grow as a positive function of the marginal product of (private-sector) capital in the previous period, and a negative function of the share of government expenditure to GDP, G,

dln(Kt) = θkαYt-1/Kt-1(Φk-Gt-1/(Yt-1Pt-1))

134. Total factor productivity (TFP) growth depends on capital stock growth:

dln(At) = θadln(Yt-1)

135. As this model simplifies the economy by assuming away financial assets, the marginal product of capital is the main determinent of the real interest rate. Nominal interest rates are equal to the real interest rate plus the rate of inflation. Simulated real interest rates are calibrated to follow historical data by applying a multiplicative constant.

it = (1+πt)(1+θrαYt/Kt)-1

Labor market

136. Labor demand is determined by the marginal product of labor, which in equilibrium is assumed to adjust to a weighted average of present and past real wages (see below).

Lt=Lt−1(Σ2i=0λtwt−i/wt−1−1)−1/α(AtAt−1)1/αKtKt−1
article image
Table III-A2.

List of Baseline Parameter Values

article image

137. The wage-setting process is a dynamic one where wage growth is assumed to be determined by labor market conditions, changes in t, and an autonomous trend (ω) which proxies for, inter alia, productivity growth.

dln(wt) = θdln(wt-1) + βdln(et-1) + γdln(τt-1)+ω

138. In calibrating the model, the effects of lagged employment and wage rates are included to reflect the effects of staggered wage contracting. The simulations replace et-1 and τt-1 in the above equation with a weighted average (with weights of 2/3 and 1/3, respectively) of their respective values one and two periods back in time.

139. If the economy is operating at full employment, wage growth is determined as the larger of the extent of wage growth implied by the wage setting equation, or the wage growth implied by the increase in labor’s marginal product under maintained full employment.

Fiscal sector

140. The imputed tax rate on the employed is calculated under the assumption of a given level of revenues and a constant tax rate on capital.

τt = Revt/(YtPt)/St-1 - θτ(1-st-1)/st-1

141. Total government expenditures are the sum of unemployment benefits, pension expenditures, other primary expenditures, and interest expenditures.

Gt = Gut + Gpt + Got + Git

142. Unemployment expenditures are equal to the number of unemployed times the benefit rate. The benefit rate is calculated as the product of gross nominal wages and the replacement ratio.

Gut = (L*t-Lt)wtPtθu

143. Pension expenditures for a given population cohort are determined as the number of pensioners in the cohort times the level of pensions at retirement times an indexation factor. Expenditures for all cohorts are added to obtain total pension expenditure. The number of pensioners for a given cohort c at a given time t (Nt, c) is calculated on the basis of assumed demographic projections. Pensions at retirement (κc) are calculated as gross nominal wages times the pension replacement ratio. The replacement ratio is assumed to decline over time, corresponding to the recently passed pension reforms. The indexation factor allows pensions to rise with net nominal wages (index factor It, c = (1 - τt)wtPt/(1-τc)wcPc).

Gpt = ΣcNt,cκcIt,c

144. Other primary expenditures are adjusted to bring the government fiscal balance (as a ratio to GDP) closer to its target value.

d(Got/(PtYt)) = θGo(GBt-1/(Pt-1Yt-1)-GB*)

145. Interest expenditures in a give period are equal to the nominal interest rate times the average stock of government debt in that period.

Git = it(Dt-1 + Dt)/2

146. Revenues as a share of GDP are also assumed to be adjusted to bring the government deficit closer to its target value.

d(Revt/(PtYt)) = θRev(GBt-1/(Pt-1Yt-1)-GB*)

147. Government debt is determined as cumulated government deficits.

Demographic assumptions

148. Demographic assumption are based on the World Bank’s population projections for Germany, which assume zero net immigration after 2005.74

Table III-A3.

Demographic Assumptions

(In percent)

article image

References

  • Alesina, Alberto and Roberto Perotti, 1997, “The Welfare State and Competitiveness,” American Economic Review, Vol. 87 No. 5, pp. 921–939.

    • Search Google Scholar
    • Export Citation

  • Börsch-Supan, Axel and Reinhold Schnabel, 1998, “Social Security and Declining Labor-Force Participation in Germany,” American Economic Review Papers and Proceedings, Vol. 88 No. 2, pp. 173–178.

    • Search Google Scholar
    • Export Citation

  • Habermeier, Karl and Fabrice Lenseigne, 1998, “French Public Finances: Modeling Long-Term Prospects and Reform Options,” IMF Working Paper WP/98/12 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, 1996, “Germany, Receny Economic Developments and Selected Issues,” IMF Staff Country Report, No. 96/111 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, 1997, “Germany, Selected Issues,” IMF Staff Country Report, No. 97/101 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Layard, Richard, Stephen Nickell, and Richard Jackman, 1991, Unemployment, Macroeconomic Performance, and the Labor Market, (Oxford: Oxford University Press).

    • Search Google Scholar
    • Export Citation

  • Leibfritz, Willi, John Thornton, and Alexandra Bibbee, 1997, “Taxation and economic performance,” OECD Economics Department Working Paper No. 176, (Paris: OECD).

    • Search Google Scholar
    • Export Citation

  • Organization for Economic Cooperation and Development, 1994, OECD Jobs Study, (Paris: OECD).

  • Organization for Economic Cooperation and Development, OECD Employment Outlook, various years (Paris: OECD).

  • Scarpetta, Stefano, 1996, “Assessing the role of labor market policies and institutional settings on unemployment: A cross-country study,” OECD Economic Studies, No. 26/1, pp. 43–98.

    • Search Google Scholar
    • Export Citation

  • Steiner, Viktor, 1996, “Finanzierungsalternativen der Sozialen Sicherung und ihre Beschäftigungswirkungen,” Soziale Sicherung und Arbeitsmarkt—Empirische Analyse und Reformansätze, edited by Steiner, Viktor and Klaus Zimmermann (Baden-Baden: Nomos Verlagsgesellschaft).

    • Search Google Scholar
    • Export Citation

  • Tyrväinen, Timo, 1995, “Wage determination in the long run, real wage resistance and unemployment: multivariate analysis of cointegrating relations in 10 OECD economies,” Bank of Finland Discussion Paper, 12/1995 (Helsinki: Bank of Finland).

    • Search Google Scholar
    • Export Citation

STATISTICAL APPENDIX

Table A1.

Germany: Key Data on Output, Income and Demand

article image
Sources: Statistisches Bundesamt, Volkswirtschaftliche Gesamtrechnungen; Deutsche Bundesbank, Monthly Report.

According to place of residence.

According to place of work.

Excludes social security contributions paid by employers.

Table A2.

Germany: Aggregate Demand

(Percentage changes at 1991 prices)

article image
Sources: Statistisches Bundesamt, Volkswirtschaftliche Gesamtrechnungen.

Change in percent of previous year’s GDP.

Table A3.

Germany: Household Income, Consumption, and Saving

(Percentage changes)

article image
Sources: Statistisches Bundesamt, Volkswirtschaftliche Gesamtrechnungen.

Disposable income in the official national accounts in 1994 is understated because of underrecording of net investment income inflows from abroad.

Deflated by private consumption deflator.

Table A4.

Germany: Labor Harket

(In thousands, unless otherwise indicated)

article image
Sources: Deutsche Bundesbank, and data provided by authorities.

According to place of work.

Labor force calculated from employment and unemployment data.

Table A5.

Germany: Wages and Prices

(Percentage changes)

article image
Sources: Statistisches Bundesamt, Volkswirtschaftliche Gesamtrechnungen; Deutsche Bundesbank.

Percentage change from a year ago.

Computed from seasonally adjusted data.

Table A6.

Germany: General Government Finances 1/

(Tn billions of deutsche mark; national accounts basis)

article image
Source: Federal Ministry of Finance.

Including the German Unity Fund

Excluding the assumption of Treuhand debt

Interim technical projections provided by the authorities.

Table A7.

Germany: Territorial Authorities’ Finances

(Administrative basis; in billions of deutsche mark)

article image
Source: Federal Ministry of Finance.

Interim technical projections provided by the authorities; from 1998 onward without public hospitals.

Including Berlin (west).

Including Berlin (east).

European Recovery Program (ERP), Burden Equalization Fund (LAF), European Community accounts, Credit Repayment Fund (KAF) (until 1994), Bundeseisenbahnvermögen (BEV) (1994), Entschädigungsfonds (from 1994), Steinkohlefonds (from 1996).

Table A8.

Germany: Federal Government Finances

(Administrative basis; in billions of deutsche mark)

article image
Source: Federal Ministry of Finance.

As approved by the Cabinet.

For 1996, includes an approximate DM 20 billion reduction due to reclassification of child allowances from an expenditure to a tax deduction.

Table A9.

Germany: Länder Government Finances

(Administrative basis; in billions of deutsche mark)

article image
Source: Federal Ministry of Finance.

Interim technical projections provided by the authorities.

Including Berlin (west).

Including Berlin (east).

Table A10.

Germany: Municipalities’ Finances

(Administrative basis; in billions of deutsche mark)

article image
Source: Federal Ministry of Finance.

Interim technical projections provided by the authorities.