Czech Republic: Selected Issues

International Monetary Fund. European Dept.

doi:10.5089/9781484363751.002.A001

Jump to Content Jump to Main Navigation

Czech Republic: Selected Issues

Author: International Monetary Fund. European Dept.

Publication Date:: 26 Jun 2018

eISBN:: 9781484363751

Language:: English

Keywords:: ISCR; CR; productivity growth; child care; observed pay gap; average pay gap; participation of young women; older worker; Women; Aging; Labor force participation; labor force decomposition

Download PDF (1.1 MB)

Selected Issues

Abstract

Selected Issues

Labor Supply in the Czech Republic: Stocktaking and Policies

The Czech working age population is projected to decline. This has important implications for labor supply and long-term growth. Our paper analyzes recent developments in labor force participation, and assesses perspectives for labor supply and policies that could affect it. The first part of the paper provides a baseline projection for labor supply in the medium and long term under existing policies. It proceeds to answer the question to what extent policies could help raise the effective labor supply. We find that policies aimed at increasing the participation of young women and older workers are important and could help mitigate the decline in the labor force, but are unlikely to stop it. The second part of the paper focuses on female labor force participation and its determinants. We examine the scope for raising female labor force participation by reducing the relative tax rate on the second earner. We find that removing the non-working spouse tax credit could boost female labor force participation by 6 percentage points.

A. Motivation

1. The demographic outlook is poor. Although the Czech population is younger than that of western Europe, it is expected to shrink in the coming years. Moreover, the working-age (15–64) population is projected to decrease more quickly than the total population. According to the United Nations medium fertility scenario, the working age population in the Czech Republic is projected to decline by 6 percent by 2030 and 21 percent by 2050.¹ Eurostat projections are more optimistic, but the differences are not large with 5 and 17 percent declines, correspondingly. At the same time, the share of population aged 65+ is projected to double and the dependency ratio will increase from 28 percent currently to 40 percent by 2030 (and nearly 60 percent in 2050).

2. Increases in participation rates have offset the decline in the working age population so far. The Czech working age population (15–64) has been declining since 2009, with a cumulative decline of 7 percent from 2009 to 2017; however, the labor force has stayed largely unchanged, as significant increases in participation rates, from 70 to 76 percent, have compensated for the decline in population. In line with statutory retirement age increases, older workers have contributed the most to the labor force growth, with the participation of workers 55–64 years old increasing from 50 to 64 percent. (Similar trends are common in many advanced economies (IMF 2018). A surprising development in the Czech Republic is that while contribution of prime-age men has increased, that of women in the same age category (25-54 years old) has not changed.

3. Participation rates have reached the EU maximum levels in many age cohorts, except for those of young women and older workers:

Participation rates of men aged 25–55 are now at the EU maximum of 96 percent, while for young women (25–45 years old) there is still a gap with the best performers. (The current participation rate is 78 percent versus the EU maximum of 91 percent.)
Notwithstanding the recent progress in the participation of older workers, further improvements could be made to reach the level of best EU performers. Namely, the participation of men aged 55–64 is at 73 percent versus the EU maximum of 83 percent, and the participation of women is at 55 percent versus the EU maximum of 78 percent.

4. The paper aims to assess labor force prospects and policies. The paper is structured as follows. The first section assesses long-term prospects for labor supply under current policies and analyzes what improvements in effective labor supply could be gained from plausible increases in female labor force participation, participation of older workers, and increases in retirement age. The second section focuses on the factors and policy distortions affecting female labor participation, using a heterogeneous agent model to assess how removing the non-working spouse tax credit (STC) could affect female labor force participation.

B. Labor Supply Prospects: Baseline and Policy Scenarios

5. Labor force developments are projected by 5-year age and gender cohorts. For the purpose of our analysis, we use the following labor force decomposition:

L a b o r f o r c e = \underset{j}{Σ} P o p u l a t i o n_{j} * P a r t i c i p a t i o n r a t e_{j}

where j is a 5-year cohort of men or women from 16 to 80 years old. We use the United Nations population projections (2017 vintage, medium fertility scenario) and historical labor force participation rates from Eurostat. To assess the labor force prospects overall, we calibrate the assumptions for the participation rates of certain groups, e.g. older workers or women, depending on various policy scenarios. Population projections remain the same in all scenarios.

6. Under current policies the labor force is projected to decline by 5 percent by 2030 (21 percent by 2050).² The baseline scenario under unchanged policies assumes unchanged age and gender-specific participation rates for all cohorts, except seniors. The participation rates for senior cohorts reflect the envisaged gradual increase in the statutory retirement age in the Czech Republic. To calibrate the increase in participation rates of older workers, we assume that they will increase to the average level of participation rates in countries with a similar statutory retirement age. Namely, the participation rate for men (for those 55–69 years old) is projected to increase from 55 to 59 percent, and for women (those 55–69 years old) from 39 to 45 percent. Under the baseline scenario, the share of older workers will increase from the current 16 percent to 22 percent in 2030.

7. Increasing participation of young women (25–45) could reduce the labor force shortfall by up to 2 percentage points. This policy scenario assumes female labor force participation is increased to the EU maximum levels of the corresponding age-gender cohort. The rate of increase is assumed to be constant across years and countries, but different across age-gender cohorts and calibrated based on historical data. A rapid female labor force participation (FLFP) increase scenario assumes an average annual increase in the FLFP of 1.3 percentage points (calibrated based on the data for 1995–2016 for the best performer in the EU, Spain). A moderate FLFP increase scenario assumes an average annual increase in the FLFP of 0.5 percentage points, corresponding to the average annual increase in the EU-15 countries over the last 20 years.

8. Increasing the participation of older (55+) workers without changing the retirement age could contribute another 2 percentage points. This scenario assumes increases in participation rates of older men and women to the EU maximum of countries with a similar retirement age. Changes in participation rates start from the first projection period, and the target participation rate changes with the projected increases in retirement age. The average participation rate for men aged 55–69 increases to 63 percent by 2030, and to 48 percent for women.

9. Raising the statutory retirement age would have the biggest impact on the labor force. An ambitious scenario that assumes the retirement age increases to 67 for both men and women by 2030 (baseline scenario for advanced Europe) and links the increase in retirement age to increases in life expectancy in the subsequent period (2030–50) would have the largest impact, of 3 percentage points. A more moderate increase in retirement age—linking the increase in the statutory retirement age to changes in life expectancy till reaching the ceiling of 67 for both men and women—would increase the labor force by 1 percentage point. Both scenarios assume participation rates at the average of the EU countries with a similar statutory retirement age.

10. Reforms could mitigate the fall in labor force, but are unlikely to completely offset it. Under a combined moderate policy improvement scenario (a moderate rate of increase in female labor force participation, retirement age increases in line with life expectancy, but not higher than 67, and participation rates for older workers at EU maximum level), the labor force decline is estimated at 2 percent in 2030 and 14 percent in 2050. Under the very optimistic (hence less likely) scenario (a rapid increase in female labor force participation, retirement age increases to 67 by 2030, and participation rates for older workers at EU maximum level), the labor force would increase by 3 percentage points by 2030, but then start to decline later with a gap of 6 percent in 2050.

C. An Analysis of Distortions Affecting Female Labor Force Participation

11. Labor force participation of young women in the Czech Republic remains relatively low. Female labor force participation has increased from 62 percent in 2009 to 69 percent in 2017 and is now around the EU average, though the developments differ significantly across age cohorts. The increase was driven primarily by higher participation of women aged 45+. A comparison with European peers shows that the participation rates of these cohorts is at the EU maximum level. On the other hand, the participation rates of younger women remain below the EU average.

12. Several factors can affect female labor force participation. A large literature provides cross-country evidence that better access to childcare and greater flexibility in work arrangements are associated with higher attachment of women to the labor market.³ Additionally, EC (2017, 2018) documents the link between the gender employment gap in the Czech Republic and low availability of affordable childcare, low use of flexible work arrangements, and a lack of long-term care facilities.

13. There are signs of insufficient supply of public childcare. Despite the increased provision of child care facilities financed by the EU structural funds, there is evidence of large unmet demand. The Ministry of Labor and Social Affairs estimates that there were 32,000 rejected applications for publicly provided childcare places due to lack of space.⁴ Furthermore, according to the Czech Ministry of Labor and Social Affairs, there are currently approximately 300,000 beneficiaries of the Czech Republic’s relatively long parental leave allowance. This suggests that the latent demand for public childcare facilities could be even higher.

14. The gender pay gap is second highest in the EU and could negatively affect female labor force participation. The observed pay gap is higher in the private sector, and highest in financial services and in managerial and professional occupations. While the age profile of the pay gap in the Czech Republic is similar to that of other European countries—the gap is most pronounced for women between the ages of 35 and 44 years, which in turn could reflect lower participation of women the earnings pay gap in the Czech Republic is also higher for women below the age of 25.⁵

15. Analysis suggests that the earnings gap is driven mostly by differences in hourly pay. Although males do have higher employment rates and work more hours on average, these differences are relatively small compared with those in regional peers and other European countries.

The contribution of the hours gap is negligible. This accords with the observed low share of part-time workers for both men and women in the Czech Republic.
The sectoral composition of employment does not explain the gender pay gap. The pay gap would be 4.5 percent larger were women equally represented across industries (NACE 2-digit industry level). The pay gap is computed as the ratio of weighted average of female annual earnings to the weighted average of male annual earnings,⁶ where weights represent the industry-gender share. In the counterfactual scenario, the share of women in each industry is the same and equal to the share of women in the labor force.
The share of women in different occupations explains some portion of the gender pay gap. Specifically, the pay gap would be 2.1 percent smaller were women equally represented across occupations. In the counterfactual scenario, the share of women in each occupation is the same and equal to the share of women in the labor force. The difference in the pay gaps indicates that the current occupational allocation increases the gender pay gap.
Women in the Czech Republic have higher educational attainment than men. The age profile shows a V-shape, with women having higher educational attainment through the cohort aged 35–44. For the cohort aged 55–64, males have higher educational attainment. The educational attainment age profile of the EU differs slightly in that females in the EU have higher educational attainment than males at all ages.

16. The Czech Republic has a high relative tax rate on second earner income that is likely to discourage women’s participation in the labor force. Christiansen et al. (2016) provide a cross-country comparison of relative tax rates, defined as the tax rate of the second earner divided by tax rate of the first earner in a childless household. This relative tax rate is high in the Czech Republic when compared with neighbors and other advanced economy peers. Furthermore, this relative tax rate is higher if the second earner earns 33 percent of the mean wage than if the second earner earns 67 percent of the mean wage.

17. The non-working spouse tax credit (STC) contributes significantly to the relative tax rate on the second earner. The non-working spouse tax credit acts as a participation tax for second-earners who are primarily women. We investigate how removing this distortion could affect female participation. Box 1 provides a numerical example of the implied participation tax that results from the provision of this tax credit.⁷

Box 1:

An Illustration of the Effect of the Non-Working Spouse Credit on Participation Incentives

This box aims to illustrate the way by which the non-working spouse credit can affect incentives for second earners to participate. To keep the analysis as simple as possible, not all taxes, credits and government assistance are considered—the analysis should be regarded as an illustration of the direction of effects, and not a comprehensive numerical evaluation.

For an indicative example, consider Jakub and Tereza, a married couple, and their young son Pavel. Suppose Jakub earns CZK 360,000 per year and Tereza does not work. Jakub’s “super-gross wage” is defined as his gross wage plus 25 percent from his employer’s social security contribution and 9 percent health insurance—that is, 1.34 × 360,000 = CZK 482,400.

Jakub pays 15 percent income tax. He receives a basic income tax credit, roughly CZK 25,000, and the non-working spouse credit, also roughly CZK 25,000, which the family receives because Teresa is not working. We assume also that Pavel is enrolled in preschool childcare and hence the family receives a preschool childcare tax credit (CZK 11,000). This yields a tax burden of 11,360 CZK and implies a household budget constraint of CZK 348,640.

Now suppose Tereza works and earns about 22.8 percent less than Jakub (per the average pay gap faced by women in the Czech private sector). This would imply annual earnings of CZK 277,920 for Tereza. In this scenario, the net income tax due for Jakub becomes 0.15 × (482,400) – 25,000 – 11,000 = CZK 36,360, and the net income tax due for Tereza is 0.15 × (372,413) – 25,000 = CZK 30,862, because there is individual taxation and both spouses work. Now the household budget constraint becomes 360,000 + 277,920 – 36360 – 30,862 = CZK 570,698.

With these assumptions, Tereza’s marginal income is the difference in total household income between the scenarios, 570,698 – 348,640= CZK 222,058, and Tereza’s participation tax is the difference in total household tax, 36,360+ 30,862 – 11,360 = CZK 55,861. Thus, Tereza’s decision to work is taxed at 25.2 percent.

The participation tax rate would be somewhat higher were Tereza working part-time and earning less. Suppose Tereza earns 33 percent of average annual earnings (CZK 120,000). In this scenario, the tax on her income is below the value of the basic income tax credit so her individual tax obligation would be zero. The household budget constraint becomes 360,000 + 120,000 – 36,360 = CZK 443,640. With these assumptions, Tereza’s marginal income is 443,640 – 348,640 = CZK 95,000 and Tereza’s participation tax is 36,360 – 11,360 = CZK 25,000. Thus, Tereza’s decision to work would be taxed at 26.3 percent.

There are other family benefits not considered in the example above, such as child tax credits, a tax base deduction, and parental benefits. Including these would alter the numerical results, but the main conclusion—that Tereza faces a non-zero participation tax rate—would remain the same. For example, taking into account the first-child tax credit (worth CZK 13,404) shows that while Tereza’s participation tax rate would be slightly lower, at 24 percent, the difference with Jakub’s tax rate would be even higher, as his participation tax rate goes to zero under this example.

Marginal Tax Rates for Average Earners

Participation Tax Rates for Low-Income Second Earner

18. To study the effect of STC on female participation we build a heterogenous agent model calibrated to match key characteristics of the Czech economy (Annex 1). Households in the model are comprised of male and female workers who jointly decide on consumption, savings, and labor supply. Male workers are assumed to supply labor inelastically, but each household decides whether the female will work or not. Hours supplied, conditional on working, are fixed. There is a representative competitive firm that employs workers and operates the economy’s capital and is owned by households. The government taxes labor income and redistributes all proceeds as a lump-sum transfer to households. Lump-sum transfers could be viewed as a proxy for government’s provision of goods and services such as public childcare facilities. Households optimize by choosing consumption and leisure, taking wages and interest rates as given. The combination of preference, technology, and asset market specifications implies that accumulated household wealth will also influence households’ female worker labor supply decision.

19. Eliminating the non-working spouse tax credit could increase female labor force participation by 6 percentage points and the freed-up resources, and extra tax revenue, could be used to increase the supply of childcare facilities. We simulate removing the non-working spouse tax credit. A comparison of the steady states between the baseline and policy scenarios shows that more women choose to work responding to the increased the incentives for the second-earner. This raises household earnings, particularly for lower-income workers. The government budget gets higher tax revenue from eliminating the tax credit and taxing income of additional women employed. This implies higher lump-sum transfers to households which could proxy for an increased supply of public childcare facilities. Overall welfare is higher under the policy scenario.

20. Robustness exercises confirm the main simulation results. An exercise in which lumpsum transfers are held fixed and the increased proceeds from taxes are spent on non-productive government consumption displays only marginally different results. Consumption and wealth are slightly lower but female labor force participation is fractionally higher in this exercise.

Table 1.

Czech Republic: Results from Policy Experiment

^{^*}Welfare is calculated as the average present-discounted value of lifetime utility.

Table 1.

Czech Republic: Results from Policy Experiment

Variable	Model
	Baseline	Remove Non-Working Spouse Credit	Change
Female Participation Rate	66.4	72.6	+6.2pp
Saving Rate	0.10	0.11	+1.0pp
Consumption	1.49	1.52	+2.0%
Assets	5.09	5.28	+3.7%
Labor Supply	0.53	0.55	+2.4%
Output	1.65	1.70	+3.1%
Female Earnings	0.52	0.56	+6.4%
Welfare^*	−8.31	−8.01	+3.6%
Transfers	0.10	0.12	+12.5%

^{^*}Welfare is calculated as the average present-discounted value of lifetime utility.

Table 1.

Czech Republic: Results from Policy Experiment

Variable	Model
	Baseline	Remove Non-Working Spouse Credit	Change
Female Participation Rate	66.4	72.6	+6.2pp
Saving Rate	0.10	0.11	+1.0pp
Consumption	1.49	1.52	+2.0%
Assets	5.09	5.28	+3.7%
Labor Supply	0.53	0.55	+2.4%
Output	1.65	1.70	+3.1%
Female Earnings	0.52	0.56	+6.4%
Welfare^*	−8.31	−8.01	+3.6%
Transfers	0.10	0.12	+12.5%

^{^*}Welfare is calculated as the average present-discounted value of lifetime utility.

D. Conclusions

21. Policies to increase participation rates and retirement age are important and can mitigate the decline in labor force, but are unlikely to offset it. Under a combined moderate policy improvement scenario, the labor force is expected to decline by 3 percent in 2030 and 15 percent in 2050. Under the very optimistic (hence less likely) scenario the labor force would increase by 3 percentage points by 2030, but then start to decline later with a gap of 8 percent by 2050.

22. Reducing the relative tax facing the second earner could boost female labor force participation. Our results indicate that removing the non-working spouse tax credit would increase female labor force participation by up to 6 percentage points.

Annex I. Model Specifications

The model features heterogenous agents with incomplete asset markets and indivisible labor and all households are comprised of male and female workers that face idiosyncratic earnings risk. Preferences are additively log-separable in household consumption and male and female labor, with separate disutility of labor parameters for men and women, and a common Frisch elasticity. There is a representative competitive firm that employs workers and operates economy’s capital while the government taxes labor income and redistributes as lump-sum transfer. The basic tax credit and non-working spouse credit are explicitly included in the model.¹ It is assumed that the government cannot transfer resources across periods i.e., the government budget has an overall balance of zero every period.

At the start of a model period, uncertainty is realized—earnings for female and male workers become known—and households begin with some wealth. During the model period, the male worker supplies labor inelastically, earning income for the household subject to taxes and tax credits. The household decides if the female worker will work. If the woman works, she also earns income for the household subject to taxes and tax credits. If the woman does not work, she derives some utility from staying at home. This preference for non-market activity could represent a comparative advantage in home production. There is complete risk-sharing within the household and the household decides how much to save and how much to consume.

The female worker participation decision will depend on the tradeoff between potential market earnings² and non-participation. It will also depend on household wealth through its effect on the consumption-leisure tradeoff. In equilibrium, the combination of earnings uncertainty and incomplete asset markets results in an ergodic distribution of wealth.

A recursive formulation of the household optimization problem is as follows. Let V, V^E, and V^N denote the value functions for a household, a household with an employed female, and a household with an unemployed female. Then a household solves:

V (x_{m}, x_{f}, a; μ) = \max_{h \in {0, \bar{h}}} {V^{E} (x_{m}, x_{f}, a; μ), V^{N} (x_{m}, x_{f}, a; μ)}

Where x_m is male productivity, x_f is female productivity, a is beginning of period family wealth, μ is the distribution of households over wealth and productivity, h is the choice of hours worked and $\bar{h}$ represents full-time hours, V, V^E, and V^N denote the value functions for a household, a household with an employed female, and a household with an unemployed female.

The value to the household of the female worker participating, V^E, is given by the following equation:

V^{E} (x_{m}, x_{f}, a; μ) = \max_{a^{'}} {\ln c - B \frac{{\bar{h}}^{1 + 1 / γ}}{1 + 1 / γ} + β E [\max {V^{E} (x_{m}^{'}, x_{f}^{'}, a^{'}; μ^{'}), V^{N} (x_{m}^{'}, x_{f}^{'}, a^{'}; μ^{'})}]}

subject to

\begin{matrix} c + a^{'} - w x_{m} \bar{h} (1 - {\tilde{τ}}_{m}) + w x_{f} \bar{h} (1 - {\tilde{τ}}_{f}) + (1 + r) a + T \\ a^{'} \geq \bar{a} \\ {\begin{matrix} μ \end{matrix}}^{'} = ▿ (μ) \end{matrix}

where c is household consumption, a^′ is household savings, w is the market wage, r is the real rate of return on assets, ${\tilde{τ}}_{m}$ and ${\tilde{τ}}_{f}$ are net tax rates of male and female workers, T is lump-sum transfers, a is the borrowing limit, and ▿ is the transition operator for the law of motion of the distribution of households. The preference parameters are given by β, the discount factor, B, the preference for leisure parameter, and γ, the Frisch elasticity of labor supply.

The value to the household of the female worker staying at home, V^N, is given by the following equation:

V^{N} (x_{m}, x_{f}, a; μ) = \max_{a^{'}} {\ln c - B \frac{0^{1 + 1 / γ}}{1 + 1 / γ} + β E [\max {V^{E} (x_{m}^{'}, x_{f}^{'}, a^{'}; μ^{'}), V^{N} (x_{m}^{'}, x_{f}^{'}, a^{'}; μ^{'})}]}

subject to,

\begin{matrix} c + a^{'} - w x_{m} \bar{h} (1 - {\tilde{τ}}_{m, f}) + (1 + r) a + T \\ a^{'} \geq \bar{a} \\ {\begin{matrix} μ \end{matrix}}^{'} = ▿ (μ) \end{matrix}

where ${\tilde{τ}}_{m, f}$ is net tax-rate when female worker is unemployed.

It is assumed that the government budget is always in balance and factor markets clear:

\int_{A, X} w (x_{m} \bar{h} {\tilde{τ}}_{m} + x_{f} h {\tilde{τ}}_{f} + x_{m} \bar{h} {\tilde{τ}}_{m, f}) 1_{{a^{'} = a^{'} (x, a, μ), h^{*}}} d μ = \int T d μ

The measure, μ, is defined over a σ-algebra of A and X, where A and X represent the sets of all possible realizations of assets a and productivities x_m and x_f.

In equilibrium, wages adjust to clear the labor market and an equilibrium reservation wage distribution emerges for female labor supply which governs the participation decision for different asset levels. The tax code is one determinant of this distribution so changes in the tax code will also affect female labor force participation. The structure of the problem precludes the use of local perturbation methods and requires global iterative methods for a solution. The model is solved using the algorithm employed by Chang and Kim (2007).

The baseline calibration of the model matches reasonably well the Czech earnings distribution and the female labor force participation rates. The parameters that govern the Markov earnings process are borrowed from Chang et al. (2011) who use U.S. individual household level data from the Panel Study of Income Dynamics (PSID) to estimate the wage process. The preference for leisure parameter is calibrated to match the Czech female labor force participation rate and yields a baseline of 66.4 percent. Unfortunately, data unavailability precludes a comparison of the baseline wealth distribution. The gender earnings gap is exogenously imposed to reflect on average the most recent data of the unadjusted observed earnings gap in the Czech private sector, 22.8 percent.³

References

Blau, Francine. and Lawrence Kahn, 2013, “Female Labor Supply: Why is the US Falling Behind?”, American Economic Review, Vol 103, No. 3, pp.251–56.
- Search Google Scholar
- Export Citation
Chang, Yongsung and Sun-Bin Kim, 2007, “Heterogeneity and Aggregation: Implications for Labor-Market Fluctuations,” American Economic Review Vol 97, No. 5, pp.1939–1956.
- Search Google Scholar
- Export Citation
Chang, Yongsung, Sun-Bin Kim, Kyooho Kwon, and Richard Rogerson, 2011, “Interpreting Labor Supply Regressions in a Model with Full and Part-Time Work,” American Economic Review Papers and Proceedings, Vol.101, No. 3, pp.476–481.
- Search Google Scholar
- Export Citation
Christiansen, Lone, Huidan Lin, Joana Pereira, Petia Topalova, and Rima Turk, 2016, “Individual Choice or Policies? Drivers of Female Employment in Europe,” IMF Working Paper, 16/49.
- Search Google Scholar
- Export Citation
Cipollone, Angela, Eleonora Patacchini, and Giovanna Vallanti, 2013, “Women Labor Market Participation in Europe: Novel Evidence on Trends and Shaping Factors,” IZA Discussion Papers, No. 7710.
- Search Google Scholar
- Export Citation
Dao, Mai, Davide Furceri, Jisoo Hwang, Meeyeon Kim, and Tae-Jeong Kim, 2014, “Strategies for Reforming Korea’s Labor Market to Foster Growth,” IMF Working Paper, 14/137.
- Search Google Scholar
- Export Citation
Genre, Véronique, Ramon Gómez-Salvador, and Ana Lamo, 2010, “European Women: Why Do(n’t) They Work?” Applied Economics, Vol. 42, Issue 12.
- Search Google Scholar
- Export Citation
Henn, Christian, 2017, “Gender at the Frontier: Policies to Underpin High-Quality Labor Supply in Norway,” IMF Country Report, No.17/181.
- Search Google Scholar
- Export Citation
Hsieh, Chang-Tai., Erik Hurst, Charles Jones, and Peter Klenow, 2013, “The Allocation of Talent and U.S. Economic Growth,” NBER Working Paper, No.18693.
- Search Google Scholar
- Export Citation
IMF, 2016, “Time for a Supply-Side Boost? Macroeconomic Effects of Labor and Product Market Reforms in Advanced Economies“, Staff Discussion Note, 17/03.
- Search Google Scholar
- Export Citation
Jaumotte, Florence, 2003, “Female Labour Force Participation: Past Trends and Main Determinants in OECD Countries,” OECD Working Paper, No.376.
- Search Google Scholar
- Export Citation
Jobspin.cz, 2017, “Salaries in the Czech Republic by Regions, Sectors, and Jobs in 2017,” JobSpin.cz.
- Search Google Scholar
- Export Citation
Jobspin.cz., 2018, “Labor Force Participation in Advanced Economies”, World Economic Outlook 2018.
- Search Google Scholar
- Export Citation
Olivetti, Claudia, and Barbara Petrongolo, 2017, “The Economic Consequences of Family Policies: Lessons from a Century of Legislation in High-Income Countries”, Journal of Economic Perspectives, Vol 31, Issue 1.
- Search Google Scholar
- Export Citation
Thévenon, Olivier, 2013, “Drivers of Female Labour Force Participation in the OECD”, OECD Social, Employment and Migration Working Papers, No.145.
- Search Google Scholar
- Export Citation
United Nations, 2017, “World Population Prospects: The 2017 Revision”, United Nations.
- Search Google Scholar
- Export Citation

This assumes a positive contribution from net inward migration.

We use the UN population projections. The UN data are at five-year frequency, with some interpolated to annual frequency. The latest historical observations are at 2015. The participation rate in the Czech Republic has increased since 2015, driven by the increased participation of older workers. The simulation takes into account the increase in participation rates from 2016 to 2017 by assuming that participation in 2020 is at the level observed in 2017.

See WEO (2018), Olivetti and Petronglo (2017), Jaumotte (2003), Genre et al. (2010), Blau and Kahn (2013), Cipollone et al. (2013), Thévenon (2013), Dao et al. (2014), Christiansen et al. (2016b), and WEO (2016).

Some of these rejected applicants may be subsequently placed in private childcare facilities albeit at higher cost.

Research shows that women leaving labor force early to raise children typically do not recover earnings when they return.

We obtain qualitatively similar results when hourly pay is used instead of annual earnings pay. This is consistent with evidence provided above that differences in hours worked contribute little to the overall gap in earnings.

We abstract from parental benefits, such as parental allowance received, as income that would apply regardless of the household’s work decision for the second-earner. Including these in our numerical example would not change the qualitative results.

The demand and supply of childcare services is not explicitly modeled thus also omitted is the childcare tax credit (See Box 1).

An unadjusted gender pay gap is assumed so that the average difference in potential market earnings between male and females is 22.8 percent, the 2016 value of the gender pay gap in the Czech private sector.

The model is unable to reproduce a comparable equilibrium pay gap due to selection effects. That is, the preference and technology parameters are such that only high-earning women participate in the market which reduces the equilibrium gender pay gap. To explain the contribution to TFP growth of reduced misallocation of labor due to gender and race discrimination, Hsieh et al. 2013 estimate an occupational choice model in which firms earn rents via gender and race discrimination.

Annex I. Model Specifications

Sectoral misallocation (Aoki 2012): For each sector i, we can measure how capital and labor deviates from the optimal allocation. TFP relative to benchmark (i.e. efficient) economy is then

\ln (T E P_{c z} / T E P_{B M}) ≃ Σ_{i} {\bar{σ}}_{i} \ln \frac{A_{i}^{C Z}}{A_{i}^{U S}} + Σ_{i} {\bar{σ}}_{i} {α_{i} \ln \frac{λ_{K i}^{C Z}}{λ_{K i}^{B M}} + (1 + α_{i}) \ln \frac{λ_{L i}^{C Z}}{λ_{L i}^{B M}}}

where λ_Li/Ki is deviation of capital/labor from optimal allocation, ${\bar{σ}}_{i}$ is sector’s share of value added. λ_Li/Ki = 1 when capital or labor is optimally allocated to the sector i. Since U.S. is relatively distortion-free economy, its’ capital intensity of each sector is used to set capital intensity α_i. Results show how reallocating capital and labor across sectors to match the U.S. level distortion would affect country’s aggregate productivity without changing total amount of capital and labor present in the economy.

Within sector misallocation (Hsieh and Klenow 2009): We assume industry output is a CES aggregate of differentiated products. The production function for each differentiated product is given by a Cobb-Douglas function of firm TFP, capital, and labor: Y_{S_i} = K_{S_i}^α_sL_{S_i}^(1−α_s). “Physical” productivity in industry s, firm i is defined as ${A S}_{i} = \frac{Y S_{i}}{K {S_{i}}^{α_{s}} (w L S_{i})^{(1 - α_{S})}}$ , where Y is output, K is capital, and wL is labor compensation. “Revenue productivity” is defined as ${T F P R S}_{i} = \frac{P S_{i} Y S_{i}}{K {S_{i}}^{α_{s}} (w L S_{i})^{(1 - α_{S})}}$ , where P_{S_i} is the price of output produced by firm i. TFPRS_i is a function of marginal revenue products of labor and capital of the firm i. Assuming joint log-normal distribution of A_{S_i} and TFPRS_i, industry TFP is related to firm-level TFP and distortions in the following way:

\log T E P_{s} = \frac{1}{σ} \log (Σ_{i = 1}^{M_{s}} A_{S i}^{σ - 1}) - \frac{σ}{2} var (\log T F P R_{S_{i}})

where σ is elasticity of substitution between firm value added.¹

In an economy with no distortions, TFPR should be equalized across firms within industry. Large dispersion of TFPR (marginal products) in a given industry indicates poor allocative efficiency within that industry,² as the efficient allocation of inputs requires marginal products to be equalized across firms in a given industry. If Y is actual output, let Y_e be efficient level of output, then $100 (\frac{Y_{e}}{Y} - 1)$ shows how much TFP could be increased by reallocating capital and labor within industries.

References

Aoki, Shuhei, 2012, “A simple accounting framework for the effect of resource misallocation on aggregate productivity,” Journal of the Japanese and International Economies, Vol. 26, Issue 4.
- Search Google Scholar
- Export Citation
Dabla-Norris, Era, Si Guo, Vikram Haksar, Minsuk Kim, Kalpana Kochhar, Kevin Wiseman, Aleksandra Zdzienicka, 2015, “The new normal: a sectoral-level perspective on productivity trends in advanced economies,” IMF Staff Discussion Note, No. 15/03.
- Search Google Scholar
- Export Citation
Haldane, Andy, 2017, “Productivity puzzles,” Bank of England.
- Search Google Scholar
- Export Citation
Hsieh, Chang-Tai and Peter Klenow, 2009, “Misallocation and manufacturing TFP in China and India”, Quarterly Journal of Economics, Vol. 124, Issue 4.
- Search Google Scholar
- Export Citation
IMF, 2015, “The New Normal: A Sector-Level Perspective on Productivity Trends in Advanced Economies”, IMF Staff Discussion Note, No. 15/3
- Search Google Scholar
- Export Citation
IMF, 2017, “Upgrading the tax system to boost productivity”, Fiscal Monitor, Ch. 2.
- Search Google Scholar
- Export Citation
Jones, Charles, 2016, “The Facts of Economic Growth”, Handbook of Macroeconomics, Vol. 2.
- Search Google Scholar
- Export Citation
Jones, Charles. 2018, “Manufacturing jobs: implications for productivity and inequality”, World Economic Outlook, Ch. 3.
- Search Google Scholar
- Export Citation
Jones, Charles. 2018, “Trends in Firm Productivity and Growth: The Role of Structural Factors”, IMF Working Paper, forthcoming.
- Search Google Scholar
- Export Citation
Restuccia, Diego and Richard Rogerson, 2007, “Policy Distortions and Aggregate Productivity with Heterogeneous Plants,” NBER Working Paper, No. 13018
- Search Google Scholar
- Export Citation
Restuccia, Diego and Richard Rogerson. 2017, “The Causes and Costs of Misallocation”, Journal of Economic Perspectives, Vol. 31, No. 3.
- Search Google Scholar
- Export Citation

See United Nations World Population Projections, 2017. Projections are available at five-year intervals, with the latest historical observation in 2015 and the projections from 2020 onward.

TFP data used for analysis is available only until 2014 and does not take into account growth developments in 2015–17.

Data are sourced from OECD “Contributions to labor productivity growth” http://stats.oecd.org/index.aspx?queryid=66347

Note that the TFP data used for this analysis is available only until 2014, and therefore does not account for growth developments from 2015 onwards.

These estimates are from the Groningen Growth and Development data on productivity levels.

For example, more could be produced by simply reallocating existing resources from less productive companies to more productive ones without any changes in individual technologies used.

To estimate α, we need to set a benchmark country that is relatively distortion-free. Here we use U.S. as our benchmark economy because it is known to be relatively-distortion free with reliable data.

These results are calculated using EU KLEMS data, September 2017 release.

The result is the average over 1987 to 2015. Averaging over different periods (e.g., 2010 to 2015) does not affect the result qualitatively.

However, benchmark U.S. is not perfect in allocation of resources itself. TFP gains from reallocating labor across sectors is 1.9% and TFP gains from reallocating capital across sectors is 10.6% for the U.S. TFP gains from both adjustments is 12.7% (averaged from 1998 to 2015).

The result is robust for different ranges of α (calculated with KLEMS 2017) and averaging over different periods.

Data are from the Orbis database from the Bureau van Dijk.

The result is robust to other measures of dispersion such as ratio of top 25^th percentile to 75^th percentile.

In Slovakia, the difference for the manufacturing sector is 2. The quality of the Orbis data for Austria and Germany do not allow for a similar comparison.

Export orientation and R&D expenditures were also considered, but there were too few observations to draw robust inferences.

Note that this regression shows simple association not causation. As typical in these sorts of regressions, only small amount of variation is explained by the variables, notwithstanding that they are highly significant.

See Feyrer (2007) and Aiyar et al. (2016)

We set σ = 3. Note that gains from liberalization are increasing in σ and estimates of the substitutability of competing manufacturing firms in the trade and industrial organization range from 3 to 10 (Hsieh and Klenow 2009).

Industry level elasticity of output to capital: α_s is calculated from U.S. manufacturing industry database, averaged over (1958-2011) (http://www.nber.org/nberces/)

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/002/2018/188/article-A001-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Czech Republic: Selected Issues

Author: International Monetary Fund. European Dept.

Volume/Issue:: Volume 2018: Issue 188

Publisher:: International Monetary Fund

ISBN:: 9781484363751

ISSN:: 1934-7685

Pages:: 31

DOI:: https://doi.org/10.5089/9781484363751.002

Keywords
Labor Supply in the Czech Republic: Stocktaking and Policies
References
- Annex I. Model Specifications
References

View raw image

Working age (16-64) population

(Index, 2015=100)

View raw image

Labor Force

(cumulative change, thousands)

View raw image

Active Population by Age and Gender, 2017

(Contributions to cumulative change in percent of 2000 total level)

View raw image

Labor Force Participation Rate by Age Group Total, 2017

View raw image

Labor force, index (2015=100)

View raw image

Labor force, index (2015=100)

View raw image

Labor force, index (2015=100)

View raw image

Labor force, index (2015=100)

View raw image

Labor force, index

(2015=100)

View raw image

Female Labor Force Participation

(Percent)

View raw image

Female Labor Force Participation Rate by Age Group, 2017

(Percent)

View raw image

Maternal Employment Rates by Age of Youngest Child

(Percent, 2013)

View raw image

Gender pay gap by age cohort

(Percent)

View raw image

Gender pay gap, unadjusted ^1/

(percent, 2016)

View raw image

Gap in Female versus Male Earnings by Components, 2014

(Percent of male earnings)

View raw image

Czech Republic: Gender pay gap by industry

(Percent, mean annual earnings)

View raw image

Czech Republic: Gender pay gap by occupation

(Percent, mean annual earnings)

View raw image

Gender Gap in Educational Attainment by Age, 2017

(Index)

View raw image

Relative Tax Rate by Second Earner Income, 2014

(Fraction of First Earner’s Tax Rate)

Table 1.

Czech Republic: Results from Policy Experiment

Variable	Model
	Baseline	Remove Non-Working Spouse Credit	Change
Female Participation Rate	66.4	72.6	+6.2pp
Saving Rate	0.10	0.11	+1.0pp
Consumption	1.49	1.52	+2.0%
Assets	5.09	5.28	+3.7%
Labor Supply	0.53	0.55	+2.4%
Output	1.65	1.70	+3.1%
Female Earnings	0.52	0.56	+6.4%
Welfare^*	−8.31	−8.01	+3.6%
Transfers	0.10	0.12	+12.5%

^{^*}Welfare is calculated as the average present-discounted value of lifetime utility.

Topics

Countries

Series

About

Resources

Abstract

Labor Supply in the Czech Republic: Stocktaking and Policies

A. Motivation

B. Labor Supply Prospects: Baseline and Policy Scenarios

C. An Analysis of Distortions Affecting Female Labor Force Participation

D. Conclusions

Annex I. Model Specifications

References

Annex I. Model Specifications

References

Same Series

Other IMF Content

Other Publishers

Asian Development Bank

Inter-American Development Bank

International Labour Organization

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics