This paper analyzes the macroeconomic impact of targeted labor market reforms aimed at boosting employment and labor productivity and the price responsiveness of German residential investment. Germany's population is getting older, and potential growth is set to decline. Demographic projections suggest that labor force will start declining around 2020, and will drop at an accelerating pace once immigration flows normalize. After years of stagnation, German housing prices and new residential rents have increased more steeply since 2009, especially in large cities. This paper provides econometric evidence that supply response to changes in housing prices has declined over the past several years and discusses how various housing policies can foster this response.

Abstract

This paper analyzes the macroeconomic impact of targeted labor market reforms aimed at boosting employment and labor productivity and the price responsiveness of German residential investment. Germany's population is getting older, and potential growth is set to decline. Demographic projections suggest that labor force will start declining around 2020, and will drop at an accelerating pace once immigration flows normalize. After years of stagnation, German housing prices and new residential rents have increased more steeply since 2009, especially in large cities. This paper provides econometric evidence that supply response to changes in housing prices has declined over the past several years and discusses how various housing policies can foster this response.

Macroeconomic Effects of Labor Supply Policies1

A. Introduction

1. Germany’s population is getting older and potential growth is set to decline. Demographic projections suggest that the labor force will start declining around 2020, and will drop at an accelerating pace once immigration flows normalize. In addition, as in other advanced countries, TFP growth has declined relative to previous decades, and is expected to remain subdued. Potential output growth is therefore expected to decline, creating a long-term challenge for the sustainability of the social security system and public finances in general. On the other hand, there is scope to offset these adverse trends by expanding labor supply by women and older workers, and by upgrading the skills of immigrants (including refugees).

2. This paper analyzes the macroeconomic impact of targeted labor market reforms aimed at boosting employment and labor productivity. A large scale multi-country DSGE model (the Global Integrated Monetary and Fiscal model, GIMF) is used to simulate the implication of policies that aim to i) enhance women’s participation in the labor force through the provision of full-time childcare and after-school programs; ii) promote longer working lives through indexing the retirement age to life-expectancy; and iii) improve the labor market integration of low-skilled immigrants through increased training. On top of durably increasing employment and output, these policies also help shore-up public pensions and reduce old-age poverty while providing positive (if small) spillovers to the rest of the euro area.

3. The analysis also shows that the reforms have short-term positive effects on domestic demand. Because agents are forward-looking they front-load part of the expected long-term income gains from the reforms, raising current consumption and investment even though the reforms are budget-neutral by construction. This last point is important as it relates to the recent debate on the potential adverse short-term effects of some structural reforms.2 To the extent that the policies considered are credible, the paper shows that they have both short-term and long-term positive effects. Furthermore, the short-term expansionary demand effect lowers the current account surplus, supporting external rebalancing; in the case of the pension reform, the reduction of the current account surplus is permanent, as the reform reduces household savings also in the long-run.

B. Population Aging and its Fiscal Effects

4. Germany’s population is aging fast. Demographic projections point to a sharp decline in working age population starting around 2020, which optimistic assumptions on immigration flows can only slow down, not reverse. The associated decline in the labor force is expected to weigh heavily on potential output. For instance, the European Commission estimates the negative contribution of aging to potential output growth at about -0.7pp per year from 2025 to 2035; see Figure 1). On the fiscal front, projections by the Federal Ministry of Finance3 point to an increase in age-related spending between 3.1 and 6.7 percent of GDP by 2060 under current rules (Figure 2), with a correspondent sustainability gap—the measure of the necessary average improvement in primary balances to meet all increase in spending—of 1.2 to 3.8 percent of GDP.

Figure 1.
Figure 1.

Aging Population and Potential Output Growth, 2013–30

(Contributions in Percent)

Citation: IMF Staff Country Reports 2016, 203; 10.5089/9781498334020.002.A001

Source: 2015 EC Ageing Report.
Figure 2.
Figure 2.

Demographically sensitive expenditures, 2000–60

(Percent of GDP)

Citation: IMF Staff Country Reports 2016, 203; 10.5089/9781498334020.002.A001

Source: Fourth Report on the Sustainability of Public Finances, 2016 and Ministry of Finance.

C. The Policy Levers

5. What measures should be considered to mitigate the effect of aging on potential output and the public finances? Drawing on recommendations from past consultations as well as ongoing discussions at the national level, we identified three areas where reforms could help alleviate the upcoming pressure on public finances and growth. Because these measures boost the long-term potential of the economy, they also improve agents’ perceptions of long-term wealth. If credible, these policies encourage households to frontload part of the long-term gains into current consumption, and firms to invest in future productive capacities. To isolate their macroeconomic effects, the policy experiments are designed to be budget-neutral.

6. First, women’s participation in the labor market could be enhanced. While women’s participation rate is relatively high in Germany (73 percent in 2014), women tend to work fewer hours than men (30.5 hours per week on average, or 9 hours less than men). The labor supply gap is thus mostly along the intensive margin. While cultural preferences are likely to play an important role, this gap can also be explained by the high effective marginal tax wedge on the second earner in a household.4 Moreover, in spite of increased efforts to expand the provision of affordable and high quality full time child care, demand still exceeds supply (only one third of 0–3 year old children are currently enrolled) and after-school programs are still insufficient. Relieving these barriers is expected to increase women’s labor supply.5

7. Second, retirement age could be extended beyond the current goal of 67 years in 2029. Labor force participation is still relatively low in Germany for the 64+ wage group, and the effective retirement age, estimated by the OECD at 62.7 years in 2014, is below that of several other OECD countries. Longer working lives can be incentivized either by indexing statutory retirement ages to life expectancy and/or by removing existing disincentives to remain in the labor force beyond pensionable age.6 In Germany, the sustainability of the public pension system is ensured by a combination of adjustments to the contribution rate and the replacement rate. But, this strategy has important limitations going forward as social security contributions are already very high (contributing to the high labor tax wedge) and the replacement rate in the first pillar is low. The authorities have tried to overcome the limitation of the first pillar by introducing the second and third pillar pensions—occupational pensions and private pension schemes (including the subsidized Riester pensions)—but so far these schemes have limited population coverage. Moreover, all three pillars are challenged as potential growth, the real interest rate, and more generally the returns on assets are low due to aging.

8. Third, investing in the training of (the current and future waves of) immigrants is key to improve their integration into the labor market. More training would allow immigrants to contribute further to the country’s production capacities. While Germany has played a key role in hosting and integrating successive waves of immigrant populations, which have in turn improved Germany’s demographic outlook, survey evidence shows that migrants have lower labor market participation, higher unemployment rates, and earn lower wages than natives with similar characteristics. These gaps narrow but are not eliminated even many years after arrival (see Beyer, 2016), though immigrants with German writing skills and/or a German degree usually perform better.

D. Why GIMF?

9. GIMF is a global (6 regions) dynamic stochastic general equilibrium model widely used inside and outside the IMF to analyze a vast range of policies and their implications for growth, inflation, and the public and external accounts. Its multi-country structure (Germany, euro area ex-Germany, the U.S., Japan, rest of Asia, and remaining countries) allows a general equilibrium analysis of global interdependences and spillover effects of alternative policies, including through financial spillovers associated with the effect of debt on global interest rates.

10. Due to its non-Ricardian features, GIMF is particularly well suited to analyze the kind of labor market reforms described above and their fiscal implications. GIMF’s underlying overlapping generations and finite horizons structure captures important life-cycle income patterns. Moreover, the presence of liquidity-constrained households (LIQ) allows realistic responses of private consumption to temporary changes in labor income and taxation.

E. The Simulations

11. To better isolate the effects of the labor market reforms from those of the fiscal costs of the reforms, we analyze the policy experiments under the condition of budget neutrality. Any discretionary fiscal cost (gain) implied by the reform is counter-balanced by a reduction (increase) in lump-sum, untargeted, transfers that keeps the structural balance unchanged. This allows us to disentangle the pure effect of the reform from the associated fiscal policy impulse. Note that in the short run automatic stabilizers are allowed to operate and the fiscal balance is affected by the cycle, though variations are very small.

12. In all experiments we assume that monetary policy remains accommodative. The policy rate lifts off only after 3 years and even then only gradually, reverting back to the model’s inflation targeting rule by the fifth year. This is to better capture the current macroeconomic environment in which euro area inflation is expected to remain below target for a number of years.

  • Policy #1: Expanding the availability of childcare services. The government currently spends about ½ percent of GDP in the provision of childcare services, against the OECD average of ¾ percent of GDP. The simulated experiment assumes that German spending on childcare services increases permanently by ¼ percent of GDP, to the OECD average in year 1. Relying on elasticities estimated in Christiansen and others (2016b), such a change is expected to lead to a 5 percent increase in female labor supply7 (2½ percent increase in total labor supply), which we assume materializes gradually over ten years.

  • The simulation shows that increasing women’s labor supply has sizable effects on economic activity in the short-term, as the expected long-term increase in income leads households and firms to frontload consumption and investment (Figure 3). Real GDP climbs by ¾ percent in the short run and by 2½ percent in the long run. Real wages increase in the short-term as labor demand increases by more than labor supply. Inflation and inflation expectations increase and the real interest rate falls further, stimulating consumption and investment. Households draw down their savings and the trade and current account balances deteriorate during the transition to the new steady-state. As labor supply gradually increases, real wages revert back down, and so do the trade balance and the current account. In the very long-term, real wages decline with respect to the baseline to accommodate the higher labor share in production.

Figure 3.
Figure 3.

Macroeconomic Effects of Policy #1

(Expanding the Provision of Childcare Services)

Citation: IMF Staff Country Reports 2016, 203; 10.5089/9781498334020.002.A001

  • Policy #2: Increasing working lives. We consider the effect of lengthening individuals’ working lives by 1 year, phased in over 25 years. This is a relative modest change, considering the sizable gaps between the average retirement age in Germany and in some other OECD countries.8 For simplicity we assume that the participation rate in the extra year of work is the same as in the earlier years.9

  • In the model, this translates both into an expansion of labor supply and a lengthening of individual agents’ labor income profile. Because labor supply increases and agents need lower savings to finance old-age consumption, consumption and investment rise in the short run and the long run (see Policy #1), and the current account balance deteriorates permanently. Real GDP rises slowly to a level that is 1¾ percent above the baseline by year 25 (when the labor supply expansion is fully phased in). Note that, since the model’s fiscal rule stabilizes the debt-to-GDP ratio, general transfers rise to offset additional tax revenues from higher employment and lifetime income. Alternatively, the additional revenues could be used to reduce social security contribution rates, which would provide a further boost to output by reducing the labor tax wedge and encouraging labor supply.

  • Policy #3: Better integration of low-skill immigrants into the labor market. Lastly, we look at the effects of government provision of specialized education and job training to low-skill immigrants. The initial skill gap between immigrants and natives is assumed to be 30 percent, in line with the wage gap estimated by Beyer (2016). Low-skilled immigrants are assumed to represent 2 percent of the labor force (about 820 thousand individuals). Thus, the yearly additional fiscal cost of the additional training would amount to about EUR 5.5 thousand per person, close to the per pupil public spending on vocational (post-secondary non-tertiary) education in 2013 (Eurostat). The cost is assumed to amount to 0.15 percent of GDP annually over 5 years and training is assumed to gradually (but permanently) boost immigrants’ productivity (concentrated in the non-tradable sector) such that the skill gap with natives is cut by half.

  • As expected, the policy has positive effects on output, real wages, consumption, and investment in the short-run and long-run. Higher labor productivity induces an increase in domestic investment and a deterioration in the current account balance in the transition to the new steady-state. Compared with the two previous policies, the effect of this policy on GDP is small, but persistent (real GDP is estimated to permanently increase by 0.17 percent of GDP by year 10). The calibration may, however, underestimate the full impact since the model does not have a mechanism to capture how better skill matching in the labor market might reduce the unemployment rate of immigrants.10

13. Spillovers to the rest of the euro area. All three policies boost consumption, investment, output and therefore imports in the short and long run. Spillovers to the rest of the world are positive but small, commensurate to the demand effect of each of the policies and the importance of trade linkages. In particular, for the rest of the Euro Area, the short-term net trade spillover both from extended childcare provision and training of refugees is roughly 5 percent of the impact on Germany’s real GDP. For the prospective increase in retirement age, it rises to 7.5–10 percent (in years 1 and 2). Spillovers to other economies are also positive but smaller, reflecting weaker trade linkages. Combining the different policies would magnify the effect of each individual policy. Likewise, for policies #1 and #3, spillovers would be larger if the fiscal cost of the policy was financed through a widening of the deficit.

F. Conclusions

14. Targeted policies to expand the labor supply of women, older workers, and immigrants can boost potential output, shore up social security and stimulate consumption and investment in the short run. These reforms would also entail positive (though small) spillovers to the rest of the world, in particular the rest of the euro area, and contribute to external rebalancing by reducing the current account surplus. Because the reforms have complementary effects on labor supply, permanent income, inflation expectations and the real interest rate, there is an added value to implementing them together to reap the benefit from positive feedback loops.

Figure 4.
Figure 4.

Macroeconomic Effects of Policy #2

(Increasing the Effective Retirement Age)

Citation: IMF Staff Country Reports 2016, 203; 10.5089/9781498334020.002.A001

Figure 5.
Figure 5.

Macroeconomic Effects of Policy #3

(Better Integration of Low-Skill Immigrants in the Labor Market)

Citation: IMF Staff Country Reports 2016, 203; 10.5089/9781498334020.002.A001

References

  • Beyer, R., 2016, “The Labor Market Performance of Immigrants in Germany”, IMF Working Paper WP/16/6

  • Christiansen, L., Lin, H., Pereira, J., Topalova, P., and R. Turk, 2016a, “Unlocking Female Employment Potential in Europe: Drivers and Benefits”, IMF, European Departmental Paper 16/1

    • Search Google Scholar
    • Export Citation
  • Christiansen, L., Lin, H., Pereira, J., Topalova, P., and R. Turk, 2016b, “Individual choice or Policies? Drivers of Female Employment in Europe”, IMF Working Paper WP/16/49

    • Search Google Scholar
    • Export Citation
  • Duval, R., and D. Furceri, 2016, “Time for a Supply Boost, Macroeconomic Effects of Labor and Product Market Reforms in Advanced Economies”, IMF April 2016 World Economic Outlook, Chapter 3

    • Search Google Scholar
    • Export Citation
  • Karam, P., Muir, D., Pereira, J., and A. Tuladhar, 2010, “The Macroeconomic Implications of Public Pension Reforms”, IMF Working Paper WP/10/297

    • Search Google Scholar
    • Export Citation
  • Kumhof, M., Laxton, D., Muir, D., and S. Mursula, 2010, “The Global Integrated Monetary and Fiscal Model (GIMF)—Theoretical Structure“, IMF Working Paper WP/10/34

    • Search Google Scholar
    • Export Citation
  • OECD, 2016, Taxing Wages 2016, Special Feature: Measuring the Tax wedge on Second Earners, OECD Publishing, Paris

  • Pereira, J., 2015, “Women in the Labor Market and the Demographic Challenge”, Germany: Selected Issues, IMF Country Report No. 15/188

    • Search Google Scholar
    • Export Citation
1

Prepared by Jean-Marc Natal and Joana Pereira (all EUR). The authors would like to thank Dirk Muir (RES) for precious help with model simulations, and Benjamin Carton (RES) for very useful discussions. The chapter also benefitted from valuable comments by seminar participants at the Bundesbank.

2

For a more in depth analysis of structural reforms, see Chapter 3 of the April 2016 WEO, “Time for a Supply Boost, Macroeconomic Effects of Labor and Product Market Reforms in Advanced Economies”.

3

See “Fourth Sustainability Report on the Sustainability of Public Finances”, 2016.

6

A recent cabinet agreement was reached to remove disincentives to remain in employment after retirement age, in particular the deduction of earnings from pensions for those who can retire at age 63 but choose to work longer, and the ineligibility to further pension increases from further pension contributions.

7

GIMF does not differentiate between extensive and intensive margins in labor supply.

8

For instance, the average effective retirement age was estimated at 66.1 in Switzerland vs. 62.7 in Germany in 2014, while life expectancy is only 2 years higher in Switzerland.

9

Alternatively, the reader may interpret the experiment as mimicking an increase in the statutory retirement age of more than one year, but with reduced participation in the additional years, so that the effective retirement age rises by 1 year.

10

Beyer (2016) estimates that unemployment among immigrants falls by up to 2½ percentage points (versus an initial gap with natives of 7 percentage points) as a result of language skills acquisition and other education.

References

  • Bundesbank (2003), “Price Indicators for the Housing Marketxs”, Monthly Report, September, 4558.

  • Caldera, Aida and Asa Johansson (2013), “The Price Responsiveness of housing Supply in OECD Countries”, Journal of Housing Economics, 22, 231249.

    • Search Google Scholar
    • Export Citation
  • Federal Institute for Building, Urban Affairs, and Spatial Development (Bundesinstitut fur Bau-, Stadt-und Raumforschung, BBSR) (2015), Wohnungsmarktprognose 2030.

    • Search Google Scholar
    • Export Citation
  • Gattini, Luca, and Ioannis Ganoulis (2012), “House price responsiveness of housing investments across major European economies”, ECB WP No 1461.

    • Search Google Scholar
    • Export Citation
  • Hug, Leonie, and Jochen Möbert (2016), “Residential Construction: Policy is Contributing Only to a Gradual Reduction in Excess Demand”, Focus Germany, Deutsche Bank Research, April.

    • Search Google Scholar
    • Export Citation
  • Kholodilin, Konstantin A. (2015), “Fifty Shades of State: Quantifying Housing Market Regulations in Germany”, DIW Discussion Paper 1530.

    • Search Google Scholar
    • Export Citation
  • Montalvo, Jose (1995), “Comparing Cointegrating Regression Estimators: Some Additional Monte Carlo Results”, Economic Letters, 48(3-4), 229234

    • Search Google Scholar
    • Export Citation
  • Pestel Institut (2015), “Modellrechnungenzum sozialen Wohnungsbau”.

  • Stock, James H., and Mark W. Watson (1993), “A Simple Estimator of Cointegrating Vectors in Higher Order Integrated Systems”, Econometrica, 61, 783820.

    • Search Google Scholar
    • Export Citation
  • Tobin, James (1969), “A general equilibrium approach to monetary theory”, Journal of Money, Credit and Banking, Vol. 1, No. 1, 1529.

    • Search Google Scholar
    • Export Citation
  • Voigtländer, Michael, and Tobias Hentze (2015), “Bedeutung der Grunderwerbsteuer für das Wohnungsangebot”, Institut der deutschen Wirtschaft Köln paper.

    • Search Google Scholar
    • Export Citation
1

Prepared by Jérôme Vandenbussche (EUR). The chapter benefitted from valuable comments by seminar participants at the Bundesbank.

2

As discussed in the 2016 Staff Report and the 2014 Selected Issues Paper “Recent Housing Market Developments”, demand for housing has also been stimulated in recent years by increasing disposable income, a strong labor market, declining interest rates, and greater interest of international investors, while mortgage lending standards have remained stable.

3

Residential investment captures investment not only in new buildings but also in existing buildings (e.g. renovations, or extensions). Investment in existing buildings should also follow the logic of Tobin’s Q. Data on residential investment in new dwellings is only available at the annual frequency, so we can’t analyze separately the two types of residential investment in our econometric framework.

4

According to the World Bank’s Doing Business database, it takes 96 days to obtain a building permit in Germany. This time has been constant over the past 10 years.

5

The nominal long-term lending rate series starts in 2003q1. We extend it backward by assuming a constant spread between the lending rate and the 10-year nominal Bund yield, which we source from the OECD.

6

The decline of the LR elasticity is monotonic. Therefore the fact that data for two large German states are included in the housing price index series from 2015 only cannot cast a doubt on our main result.

7

See also Hug and Möbert (2016) for a comprehensive discussion.

Germany: Selected Issues
Author: International Monetary Fund. European Dept.
  • View in gallery

    Aging Population and Potential Output Growth, 2013–30

    (Contributions in Percent)

  • View in gallery

    Demographically sensitive expenditures, 2000–60

    (Percent of GDP)

  • View in gallery

    Macroeconomic Effects of Policy #1

    (Expanding the Provision of Childcare Services)

  • View in gallery

    Macroeconomic Effects of Policy #2

    (Increasing the Effective Retirement Age)

  • View in gallery

    Macroeconomic Effects of Policy #3

    (Better Integration of Low-Skill Immigrants in the Labor Market)