I. Tax and welfare reform in the Czech Republic—structural implications and challenges1
1. The Czech Republic is facing substantial fiscal challenges in coming years and decades. The key short-term priority is to meet the obligation under EU’s Stability and Growth Pact to reduce the government deficit below 3 percent of GDP. The longer term challenge is to ensure a sustainable fiscal position as demographic developments imply rapid and significant increases in pension and health care spending. Meeting these challenges will require major reform of government spending and taxation, while also taking into account structural implications – such as labor market participation, the level and structure of unemployment, tax competition, and income distribution. The underlying question is how to combine broad public and political support for a comprehensive welfare state with the inescapable requirement to maintain healthy government finances.
2. A number of tax and welfare measures have been implemented in recent years, most of which have tended to raise the fiscal deficit (such as the 2006 tax cuts and the 2006/07 benefit increases). Spending reforms, in particular for health and pensions, have not been implemented as envisaged, and other attempts to curb government deficits, such as the 2004 Medium Term Expenditure Framework and planned cuts in government staff, have had little impact. Most recently, an ambitious and comprehensive reform package has been passed (starting in 2008), aiming both at improving government finances, strengthening labor supply and mitigating the effects from tax competition.
3. This paper takes a closer look at the reform package, as well as measures introduced in 2007. Section B below briefly reviews the backdrop for fiscal reform, section C outlines the key reform measures, section D analyses their structural implications (impact on incentives and distribution). Section E proposes possible measures going forward and section F concludes.
B. The Backdrop for Reform
4. The government deficit is projected at just above the 3 percent EU-threshold in 2007, further declining below the threshold in the following years. Maintaining the deficit below 3 percent will remain a key near-term challenge, especially if cyclical developments turn less benign than projected. However, more substantial deficit reduction is required to meet the long term spending pressures from the ageing population. The Ministry of Finance estimates the so-called fiscal gap (the permanent improvement of government balances required for long run fiscal sustainability) at about 8 percent of GDP.2 The long-term spending pressures in the Czech Republic are more substantial than in most other EU and OECD countries (The European Commission, 2006).
5. The serious fiscal circumstances requires ongoing and firm prioritization and scrutinizing of spending and taxation. In particular, new spending plans or revenue losing tax cuts should be carefully considered and generally be avoided unless there are highly compelling reasons to undertake such initiatives.
6. The Czech tax/GDP ratio of about 38 percent is slightly lower than the EU-15 average, but higher than the tax ratios of e.g. Poland and Slovakia (Figure 1).3 The tax structure stands out from the EU-15 average and other countries in the region by relying more heavily on social security contributions (which are among the highest in the world) and the corporate income tax, and less on taxation of personal income, consumption and property. This suggests that revenue enhancing tax reform could focus on increased taxation of consumption and property as well as shifting some of the burden from social security contributions onto the personal income tax or other taxes, as discussed below.
Figure 1.Tax Structure in Selected Countries, 2004
Source: OECD, Revenue Statistics, 2006.
7. Labor participation is at par with the average EU-15 level, and higher than elsewhere in the region, both for men and women (Figure 2). The United Kingdom and Germany (as well as the Nordic countries) stand out as having somewhat higher overall participation rates. On the other hand, hours worked in the Czech Republic and other New Member States (NME’s) tend to be substantially higher—by up to 30 percent—than in EU-15 (Figure 3).
Figure 2.Labor Participation Rates in Selected Countries, 2006
Source: OECD (2007a).
Figure 3.Hours Worked in Selected Countries, 2006
Source: OECD (2007a). Note: these are actual annual hours worked, including employed and the self-employed.
8. Participation among the older generations on the labor market (55-64 years of age) is broadly in line with the EU-15 average and somewhat above other countries in the region. Youth participation (15-24 years of age), while in line with regional levels, is much lower than the EU-15 average and has declined over the past four years (Figure 4), reflecting to a large extent higher inflow to secondary and tertiary education.
Figure 4.Participation Rates Among the Younger and Older Generations, 2006
Source: OECD (2007a).
9. The unemployment rate has declined substantially over the past few years, at least in part reflecting a strong cyclical upswing. The unemployment rate, based on labor force surveys, was about 7 percent in 2006 and has declined further to below 6 percent in 2007, lower than both the EU-15 average and that of other countries in the region (Figure 5).4 The EU-Commission calculates that long term unemployment makes up for about 50-55 percent of unemployment, a number that remains stubbornly high even as the overall unemployment rate has been substantially reduced (Figure 6). The relatively high and persistent incidence of long-term unemployment is a characteristic the Czech Republic shares with other NME’s (Schiff and others, 2006).
Figure 5.Unemployment Rates, 2006
Figure 6.Long-Term Unemployment in Percent of Total Unemployment
10. Overall, the Czech labor market compares relatively well with its neighbors in the region and the EU-15 average. It is, however, lagging somewhat behind the best performers in the EU—including the UK and the Nordic countries—basically because of low participation rates for the younger and older generations, women and a relatively high and persistent incidence of long term unemployment.
11. Factors influencing participation and long-term unemployment are complex and often hard to measure. There are, however, indications that the post-communist era industrial restructuring and associated skill mis-matches in specific regions are among the key reasons for the high rate of long-term unemployment in the Czech Republic5, and likely to be more important than overall labor market institutions such as unemployment benefits, minimum wages, and employment protection legislation (IMF, 2004; Juradja and Munich, 2002).6 The latter factors, on the other hand, might be more pertinent in explaining why participation rates are still somewhat below best international performers.7 Broader skill mismatches, in particular a rapidly increasing demand for degree-level education, also seem to have played a role (OECD, 2006). Tax and social benefit systems exert important influences on labor market performance, but as discussed below, are unlikely to cause excessive across the board disincentives to work. Rather, these systems may interact to generate substantial disincentives for limited groups of individuals and families.
12. The Czech Republic is a highly egalitarian society. The distribution of income is more compressed and the incidence of poverty lower than in almost any other OECD country —despite a tendency for increasing inequality since the mid-1990s (Burniaux and others, 2006; World Bank, 2007). The recent increases in the minimum wage and the minimum living standards are likely to further bolster this position. Poverty and distributional issues are hence likely to be concentrated on smaller groups, often situated in geographical and ethnic pockets.
13. The tax and benefit systems are each progressive, implying a highly progressive—and hence redistributive—effect from these two systems in combination (Schneider, 2004a). Some studies find that the progressivity, in particular the means testing of social benefits, creates significant disincentives to work, or to move up the earnings ladder, at low earning levels—i.e. the presence of both unemployment and poverty traps (Schneider, 20004b). On the other hand, IMF (2006) finds that while the social transfer system in general tends to be effective in reducing inequality in earnings and the risk of poverty, there is still ample scope for expanding the means testing of social transfers and enhancing spending efficiency (a point amplified by the relatively broad coverage of the benefit increases introduced in 2006 and 2007). This points to the key distributional challenge in designing tax and welfare systems, between narrowly targeting benefits and income tax relief’s to those in need (in order to contain government spending) and limiting the disincentives to work or move up the income ladder created by that same targeting.
C. The Reform Package—Key Features
14. The reform adopted by Parliament in August 2007 and subsequently approved by the Senate and the President is a comprehensive overhaul of the welfare and tax systems. It intends to strengthen work incentives, partly by compressing real replacement rates, partly by cutting personal income taxes, in particular at the lower and higher end of the earnings scale. The reform implies an overall shift of the tax burden from income to consumption, thus aiming at spurring savings and investment, while also reducing pollution and carbon dioxide emissions through new excises on energy consumption. A reduction of the corporate income tax rate will strengthen the competitiveness of the Czech Republic in terms of attracting investment, and several measures are applied in the health and social area in an attempt to contain spending increases.
15. Key features of the reform package are:8
Personal income tax: The centerpiece of the tax reform is a 15 percent flat tax on personal income. Since the tax will be levied on gross earnings plus the employer’s social security contributions, the effective flat rate is 23 percent. Like the recent wave of flat tax reforms in Central- and Eastern Europe, this is not a classical flat expenditure tax with one marginal rate (Hall and Rabushka, 1985), but a one-rate tax on personal income (Box 1).9 In practice, however, there will still be multiple marginal effective tax rates due to the tax credits, the social security contributions and the means-tested social benefits. The reform involves a substantial scaling up of various tax credits, and, importantly, abolition of the recently introduced joint taxation for couples with children. The final withholding tax on dividends and interest income will be aligned at 15 percent, and the tax free status of mortgage-backed bonds eliminated. Personal capital gains on financial and real assets will continue to be tax exempt after 6 months’ ownership.
Social security contributions: There will be a cap on social security contributions for wage earners, kicking in at 4 times the average wage (about CZK 1,000,0000 annually). The already existing cap for the self-employed will be raised to the same level.
The corporate income tax will be lowered gradually from 24 percent to 19 percent in 2010. There will be some modest expansion of the tax base.
The lower VAT rate, currently at 5 percent, will be raised to 9 percent. The standard 19 percent rate will be maintained.
Box 1.Experience with Flat Tax Reforms in Central and Eastern Europe
Flat taxes on personal income have been introduced in nine countries in Central and Eastern Europe (year of introduction):
- Estonia: 26 percent (1994)
- Lithuania: 33 percent (1994)
- Latvia: 25 percent (1995)
- Russia: 13 percent (2001)
- Serbia: 14 percent (2003)
- Slovakia: 19 percent (2004)
- Ukraine: 13 percent (2004)
- Georgia: 12 percent (2005)
- Romania: 16 percent (2005)
Some of these countries—Estonia, Slovakia, Romania and Serbia—have a flat CIT rate at the same level as the PIT rate.
While there are substantial differences in design, a common feature is that the shift to flat taxes in these countries have generally been associated with tax cuts at the bottom and the top of the income distribution, while taxation of middle income levels have either been unchanged (implying lower overall revenue) or increased (to achieve revenue neutrality). The effects of these reforms in terms of efficiency and simplicity are ambiguous, especially where combined with base narrowing measures, and there are generally no sign of Laffer type behavioral responses (generating revenue increases from cutting taxes). Moreover, most of these countries continue to rely heavily on social security contributions and means-tested social benefits, which in turn imply continued multiple, and for some earnings ranges significant, marginal effective income tax rates. Distributional effects of the flat taxes are not unambiguously regressive, and in some cases may have increased progressivity, including through the impact on compliance (Keen and others, 2006).
The Czech reform proposal have significant similarities with the successful 2004 reform in Slovakia, although with important exceptions – for instance, Slovakia merged the two VAT rates into one single rate, implemented a wide ranging pension reform, cut social assistance programs more substantially and closed a number of corporate loopholes (for a discussion of Slovakia’s reform, see Brook and Leibfritz (2005) and Moore (2005)).
Benefits: All social benefits, except for pensions, will cease to be automatically indexed to cost of living measures. The unemployment benefit will be tightened, some benefits are eliminated and others streamlined. The regular child benefit will no longer be graduated according to income level (but will be discontinued above an income threshold which is substantially lowered) and the additional child benefit will be phased out earlier. For analytical purposes, this paper also includes as part of the new regime the change to the accommodation allowance that was enacted from January 2007. This change has important implications for marginal and average effective tax rates at low income levels, as discussed below.
Health: To curb expenses for sickness, financing of sickness benefits will, to a larger extent, be borne by employers and no benefits will be paid for the first three days of sickness. Limited co-payments–in the order of 1 to 3 euros–are introduced for medical services (doctor’s visits; short-term hospital stays; emergency services; prescriptions).
Pension: The reform leaves out the important question of reform to the pension system, except for proposing a further general increase in the retirement age to 65 (an increase to 63 years for men is already being phased in towards 2013) and extending the qualifying period from 25 to 35 years of work. A more comprehensive overhaul of the pension system, based on the recommendations of a 2005 expert group and each of the political parties, is expected within the next couple of years (See Hemmings and Whitehouse (2006) for a thorough review of the five proposals currently on the table).
D. Analysis and Assessment of the Reform Package
Personal income tax
16. The reform package will result in rather substantive tax cuts for many income earners. Effects on labor market participation and unemployment depends on the path of average effective tax rates (influencing mainly the decision to participate in the labor market) and marginal effective tax rates (influencing mainly the decision to work more hours).10
17. To grasp the importance of changes to the benefit system implemented from 2007, the analysis below focuses on changes over the years 2006 to 2008.11 The main features driving those changes are summarized in Table 1.
|Tax rates, percent||12; 19; 25; 32||12; 19; 25; 32||15|
|Minimum Living Standard||Incl.|
|Regular child benefit,|
|1.1; 1.8; and 3.0|
|1.5; 2.4; and 4.0|
|2.4 times MLS|
|Additional child benefits,|
|1.6 times MLS||2.2 times MLS||2.0 times MLS|
|Cap on social security|
18. Between 2006 and 2008, the average effective tax rate (ATR) is reduced substantially at the bottom and the top of the earnings scale, reflecting primarily the expanded tax credits (from 2008) and the significantly enhanced accommodation allowance and minimum living standard transfers (from 2007) for the former and the lower marginal tax rate and the cap on social security contributions for the latter. The small reduction in the regular child benefits from 2007 to 2008 is thus more than offset by larger tax credits for working, low income groups, and higher social transfers.12 Mid-range groups (CZK 250,000-500,000) are much less affected by the reform since the base increase to “super gross,” i.e. including social security contributions, basically outweighs the lower marginal tax rate (Figure 7). There will be small tax increases for some mid-income families, where the loss of the option for joint taxation and the earlier phase out of child benefits dominate the effect of lower marginal tax rates and the higher basic and child tax credits. The tax cuts for high earning individuals imply that the overall tax system ceases to be progressive from 2008, since the average effective tax rate actually falls for earning levels above CZK one million.
Figure 7.Average Effective Tax Rates
Source: Staff calculations.
19. The story is less clear-cut on marginal effective rates (MTR). At the bottom of the wage distribution, the marginal tax rate has increased from 12 to15 percent. For the very lowest wages, this is offset by the increased basic tax credit which keeps more people entirely out of the tax net.13 However, the benefit withdrawals keep marginal effective tax rates in 2008 higher for most low- to mid-income earners than they were in 2006 (Figure 8). In particular, the high MTR’s—70 to 80 percent—kicking in at or just above the minimum wage (in the CZK 100,000-300,000 earnings range) act as a barrier to labor supply among single wage earners with children as well as low income families. Further up the wage distribution the MTR’s drop as the effect of the flat 15 percent rate compared with the existing progressive scale kicks in. The very high incomes (those earning wages above four times the average, or about CZK one million per year) receive a substantial reduction in the MTR (from 40 to 15 percent) as the social security contribution is capped nominally beyond that wage level.
Figure 8.Marginal Effective Tax Rates
Source: Staff calculations.
20. The problem of interaction of the tax and benefit systems in creating high effective marginal tax rates can be illustrated by the example of a single wage earner with two children. At low earnings, her marginal tax rate (before benefits) would be 12½ percent (i.e. her own social security contribution), rising to 33 percent a bit further up the wage scale. But, when taking into account the phasing out of child benefits and the accommodation allowance, the effective marginal tax rate is much higher—up to 80 percent at low earnings (Figure 9).14 The new regime for the regular child benefit, with just one cut-off income level, is an improvement over the old system, where the benefit was reduced at three different points on the income scale, creating large spikes in the marginal effective tax rates at each point. Evidently, such spikes, in turn, imply strong disincentives for efforts to increase earnings (since around these income levels, it takes a substantial increase in gross earning to keep take home pay from falling).
Figure 9.Marginal Effectives Tax Rates Before and After Transfers (2008)
Source: Staff calculations. Note. The spike at around CZK 260,000 is caused by the cut-off of the regular child benefit.
21. Considering marginal effective tax rates alone, however, might not give the full picture for assessing labor market incentives, since they—by definition—consider only a marginal increase in earnings. What is often more relevant, especially at the bottom of the earnings scale, is the increase in take home pay when going from very low earnings to a level that is, say, CZK 100,000 higher (for instance, shifting from a part time job to a full time job; attaining further education to gain a higher salary; moving city for a better paid job).
22. It turns out that the 2008 regime for many groups implies higher tax takes on additional earnings when moving up the income ladder from low- and mid-income earnings (Table 2).15 Conversely, at the higher end of the income scale, the tax take on additional income will be smaller than before, thus spurring work incentives.
|Initial level of gross earnings||Single, no children||Single, no children||Single, 2 children||Spouse unemployed||Spouse unemployed||Spouse unemployed||Spouse low income||Spouse low income|
23. Another key measure for work efforts at the bottom of the earnings scale is the gain in take home pay from moving from unemployment benefits or social transfer income into earnings from work. The 2008 regime generally implies a larger reward from work (compared with the old regime) when moving into low paid jobs and a smaller reward when going into average paid jobs (Table 3).16
|Implied tax rate (% gross earnings)||Single, no children||Single, no children||Single, 2 children||Single, 2 children||Spouse, inactive||Spouse, inactive||Spouse low income||Spouse low income|
|From UI (APW)to APW||65||67||75||74||64||67||68||80|
|From UI (MW)to APW||47||56||63||70||64||66||55||76|
|From UI (APW)to MW||115||109||100||72||99||50||103||86|
|From UI (MW)to MW||70||83||70||60||99||46||70||76|
|From social transfers to APW||46||46||63||61||64||57||42||67|
|From social transfers to MW||67||58||70||38||99||24||37||54|
24. The 2008 system is thus, in general, better at encouraging participation at low wages, while it is worse at encouraging participation at average wages (which in turn could dampen incentives for the unemployed to undertake further education or training, or to apply for better paid jobs in other regions). And, although participation incentives have improved for some groups, the overall impression is that the tax and benefit systems still combine to create rather high effective tax rates and thus significant unemployment traps.
25. In sum, labor supply effects from the reforms in 2007/08 are ambiguous as there are offsetting substitution and income effects, depending on earnings level. At the lower end of the earnings scale there is a combination of lower aggregate and higher marginal tax rates, both of which tend to discourage work effort (such as working more hours, seeking better paid jobs, or undertaking education). Although the enhanced tax credits helps to increase the income gains from employment (over unemployment benefits or social transfers, which are not taxable income), thus spurring entry into low paid jobs, the high effective marginal rates at low earnings—caused primarily by the withdrawal of the additional child benefit, the accommodation benefit and the minimum living standard transfer - are highly detrimental to further work effort. Moreover, the tax credits are not targeted, making them less cost-effective measures for encouraging participation.
26. At mid-income levels, tax changes are rather small, but tend to combine higher average rates and lower marginal rates, both of which increase labor supply (as the income and substitution effects works in the same direction). The high-wage earners face substantially lower average and marginal rates and their labor response is hence ambiguous (lower ATR’s reduce supply while lower MTR’s increase it). Since these groups already tend to work long hours, however, it is likely that the income effect will dominate, hence reducing supply. One exception is that there will be less salary driven emigration (and “brain drain”) among high income earners and it would be easier to attract high skilled foreign labor. However, migration decisions do not exclusively depend on net income prospects and those prospects, in turn, do not depend exclusively on taxation.17
27. For spouses, there will be enhanced incentives to participate through the abolition of joint taxation and the higher personal tax credit.18 This is, however, to a considerable extent offset by the significant increase in the spouse tax credit, which—through the income effect—discourage participation. From a labor supply viewpoint, this is unfortunate since the supply of labor of the second earner in the household is normally much more sensitive to tax changes than the labor supply of the primary earner.
28. For the older generations on the labor market, the new tax credit for retirees who continue to work would substantially strengthen incentives to work. Given the relatively low effective retirement age (IMF, 2006), the deadweight loss associated with the credit is likely to be limited. It needs to be ensured, though, that the full credit is limited to those working full time. One drawback of the credit is that it exacerbates, by giving a tax credit to the elderly, generational inequalities already found to be rather substantial in the Czech Republic (IMF, 2005).
29. On the labor demand side, the proposed tax rates will reduce tax wedges at the bottom and the top of the earnings scale, along the lines observed for the average effective tax rates. One noteworthy feature of the proposal is that tax wedges for married couples with children (unemployed spouse) are substantially lower than those facing single wage earners with children, because of the large spouse credit for the former (Figure 10). This is not a new feature of the tax and benefit system, but one that might be worthwhile re-considering since there is little economic reason for upholding the inequality between couples and singles. And, as discussed above, the spouse credit acts as a substantial barrier for labor participation for spouses.
Figure 10.Average Tax Wedges (2008)
Source: Staff calculations.
The international context
30. Tax wedges in the current system are already at the lower end of the EU-15 and regional spectrum, especially for families (Figure 11). Hence, there is no indication that the tax system act as a stronger brake on labor demand in the Czech Republic than in other EU countries or among its neighbors. After the reform, this feature will be even more pronounced.
Figure 11.Average Tax Wedges in Selected Countries
Source: OECD (2007) and staff calculations. Note. The average effective tax wedge measures total tax, social security contributions and benefits as a percent of gross labor costs (including employers social security contribution).
31. Average effective tax rates are also lower than in most other EU countries and the reform exacerbates this feature (Figure 12). Low income families pay negative income taxes/receives a tax subsidy—in contrast to most other countries, except for some of those applying earned income tax credits (such as the United States and New Zealand)—thanks to the generous tax credits and the various benefits.
Figure 12.Average Effective Tax Rates in Selected Countries
Source: OECD (2007) and staff calculations. Note. The ATR measures total tax, employee social security contributions and benefits as a percent of gross earnings.
32. Marginal effective tax rates are relatively low when excluding benefits and relatively high when including benefits (Figure 13). Marginal tax rates will increase for some family types under the new regime, especially in the low- to mid-income range, where rates will continue to significantly exceed EU-15 and regional standards. The top marginal rate, on the other hand, will now be reduced from 40 to 15 percent and hence become, by far, the lowest in the EU and indeed the OECD (Figure 14).
Figure 13.Marginal Effective Tax Rates in Selected Countries
Source: OECD (2007) and staff calculations. Note. The MTR measures marginal tax, employee social security contributions and benefits as a percent of marginal gross earnings.
Figure 14.Top Marginal (all-in) Tax Rate on Personal Income
Source: The OECD tax database.
The corporate income tax
33. The proposal to cut the corporate income tax rate comes after a series of rate cuts in recent years, not only in the Czech Republic, but across the EU and most other parts of the world. The current rate in the Czech republic, at 24 percent, is somewhat below the EU-15 average and the rates in countries like Germany and the United Kingdom, but above rates found in some other countries in the region, including Hungary, Poland and Slovakia (Figure 15). However, looking at the wider taxation of dividend income (CIT and PIT combined), the Czech Republic is faring better compared with the EU-15 average and its regional competitors, with a combined rate currently at 35 percent, the same level as Poland (Figure 16). Among EU-25 countries, only Slovakia and Greece have lower combined rates.
Figure 15.Statutory Corporate Income Tax Rates in Selected Countries
Source: The OECD tax database.
Figure 16.All-in Rates on Dividend Income in Selected Countries
Source.: The OECD tax database.
34. The corporate income tax base seems to be less eroded by tax preferences and incentives than many other EU and OECD countries. Although a relatively generous investment tax credit is offered for new investment (subject to conditions of scale, sector, location etc.) and significant tax incentives apply to spending on research and development, the base appears to be relatively broad and depreciation rules overall broadly in line with other countries in the EU.19
35. A rough indication of the broadness of the corporate income tax base is the so-called CIT effectiveness indicator, which measures how much corporate income tax revenue is raised (in percent of GDP) per percentage point tax rate. The CIT effectiveness in the Czech republic, at almost 0.220, is about twice the size of the average found in large industrial countries and also substantially higher than those of neighboring countries. The relatively broad base combined with a CIT rate close to the EU-15 average implies that revenue from the corporate income tax is much more significant in the Czech Republic than in most other EU- and OECD countries.
36. The key issue is whether there are pressures on the corporate income tax base or rate substantial enough to warrant adjustments. Such pressures could stem, for instance, from increased international mobility of capital and/or corporate tax cuts among regional competitors, but could also arise if the wider taxation of capital income is seen as excessively discouraging savings and investment. Although answering these questions is not straightforward (since they depend not only on the headline tax rate but also on its base, which in turn is a function of complex rules for capital allowances, deduction for financial expenses etc.), it is instructive to look at two central summary statistics: the average effective tax rate (AETR) and the marginal effective tax rate (METR).21 The AETR basically affects the location of an investment, whereas the METR affects how much an investor will choose to invest once he has decided in which country to locate the investment.22
37. At the current CIT rate, the AETR’s in the Czech Republic are slightly higher than those of Hungary and Slovakia (for equity—but not for debt—financed investment), but substantially below the OECD average (Figure 17).23 This indicates that the CIT does not act as a major barrier for choosing the Czech Republic as investment location (either by inward FDI or by domestic firms). In addition, although taxation is an important variable driving FDI patterns, several studies have shown that it is only one parameter out of many determining investment locations of multinational businesses.24 Evidently, lowering the CIT rate to 19 percent would further improve the Czech position, but at the cost of a substantial loss in revenue.
Figure 17.AETR’s in Selected Countries
Source: Staff calculations; Devereux and others (2002).
38. The METR for investment in machinery and equipment, financed fully by new equity or retained earnings, again, is slightly higher than rates in Hungary and Slovakia, but below rates found in most other OECD countries (Figure 18). For debt financed investment, where marginal effective tax rates are often negative because of the deductibility of interest costs (and the tax system hence provides a subsidy to the marginal investment), the Czech rates are lower (i.e. the marginal tax subsidy is higher) than for most other countries, including Hungary, Poland and Slovakia. Overall, this indicates that marginal investment is not excessively discouraged in the Czech Republic at current CIT rates and rules. Interestingly, lowering the CIT rate to 19 percent would increase the METR for debt financed investment (i.e. reducing the tax subsidy).
Figure 18.METR’s in Selected Countries
Source: Staff calculations; Devereux and other (2002).
39. In conclusion, there seems to be no urgent need to lower the corporate tax rate from its current level of 24 percent or provide other incentives for investors. On the other hand, the trend in EU and globally points to continued further reductions of CIT rates, and it would therefore be prudent for the Czech Republic to prepare for lower CIT rates. The proposed 19 percent rate constitutes a good aiming point, but there is nothing in the analysis pointing to a need to advance the change—indeed, it could be even more back loaded than what is proposed (i.e. reaching the 19 percent later than 2010). The CIT base is relatively broad and there seems to be limited scope for offsetting revenue losses from the CIT rate cuts by base broadening measures (except for reducing or eliminating the investment tax credit and the incentives for spending on R&D).25
The value added tax
40. The proposed increase in the low VAT rate from 5 to 9 percent is a welcome step to not only reduce distortions associated with the dual rate structure, but also to shift more of the tax burden from income to consumption.
41. Unsurprisingly, the subsidy implied by the low rate (i.e. consumers save the difference between the low rate of 5 percent and the standard rate of 19 percent) is relatively evenly distributed across income deciles when measured as percent of net income (Figure 19). This mirrors international experience with multiple VAT rates, almost always showing a non-or only slightly progressive pattern of the subsidy. In terms of the distribution of the absolute amounts of the subsidy, the highest income decile reaps almost one-fourth of the total subsidy while the lowest income deciles only gets about 5 percent. One item, in particular, prevents the subsidy from being more progressive: the low tax rate on the first transfer of housing, which in particular benefits households in the 10th decile.26 Without this item, the subsidy, as a percent of net income, would gradually decline across the income range (Figure 19).
Figure 19.Distribution of the Subsidy Implied by the Low VAT Rate
Source: Staff calculations, based on the 2006 household survey.
42. It follows that increasing the low rate to 9 percent would not be very regressive. In fact, all income groups would face an increase in costs of between 1-1½ percent of their net income (Figure 20).27 If transfer of housing were to become subject to the standard rate, the incremental costs would be much more substantial for the highest income decile, but basically unchanged for the lowest incomes.
Figure 20.Distribution of Increase in Expenditures from Raising Low Rate from 5 to 9 Percent
Source: Staff calculations and the 2006 household survey.
43. A key cost containing measure is the abolition of automatic indexation to the minimum living standard. Over time this will reduce effective replacement rates (except for pensions), implying government cost savings and added incentive to seek employment for those out of work.28 However, it is questionable whether it is politically feasible, or socially acceptable, to allow relative living standards of those of out work to persistently decline in comparison with those in work. Hence, there is a probability that these measures, de facto, will be fully or partly clawed back over time, thus providing temporary effects only. Most of the other benefit measures (except for the reduction in the regular and additional child benefits) are of minor importance for labor market outcomes and their cost reducing impact modest (see Appendix 1 for details).
44. The measures to limit health spending would substantially reduce perverse incentives for firms and workers. They will reduce financial incentives for firms to use the sickness benefit system as a vehicle for short-term off-loading unneeded workers and encourage businesses to verify that sickness claims are genuine (by shifting the burden of financing to the employer) as well as discourage workers from shirking (by imposing a three day quarantine on sickness benefits). These are bold measures, likely to be effective in curbing costs related to short term sickness. The issue is to what extent specific industries or groups in the labor market, with high “true” exposure to short term sickness, will be excessively burdened by the new regime. For wage earners, this could in principle be solved through wage negotiations, but for firms this would mean a “double” competitive disadvantage (financing both the sickness benefits and higher wages to compensate for the quarantine). The extent and scale of this issue remains to be seen.
45. The introduction of user fees is also likely to be highly effective in curbing public spending. Evidence from other countries suggest a substantial sensitivity of spending to the level of user co-payments, especially if these are low to begin with, and especially for pharmaceutical drugs and doctor’s visits (Docteur and Oxley, 2003). The cap on total annual own-payments (the cap will apply to prescription fees and doctors visits, but not hospitalization and emergency services) will shield most individuals and families from excessive additional costs and the tax relief’s to low income working individuals would, in most cases, be expected to more than compensate for the combined effects of the increase in the low VAT rate and the introduction of user fees. The user fees will also constitute a revenue source for the government, although the fiscal effect from this is likely to be far less than the cost savings as demand for treatment is curbed.
46. Containing public sector payroll increases below 1.5 percent per year until 2010 might be achievable, but is not a sustainable long term strategy. Public sector wages would normally be expected to grow more or less in line with private sector wages, otherwise a wage gap will arise or increase, which will eventually drain the public sector of labor (with the highest quality/most mobile individuals to go first). Hence, there will be a need to reduce government employment (a 3 percent annual reduction is planned), but it remains an open question whether such reductions are achievable.
47. The key distributional feature of the 2008 reform package is a substantial tax cut for low income working individuals and an even larger cut for high income earners. The unemployed and recipients of social transfers will not benefit from the tax cuts (since these sources of income are not taxed) and hence be relatively worse off. This is exacerbated by the increase in the low VAT rate and the introduction of user fees in the health sector. On the other hand, social benefits have increased substantially through the 2007 reform of the accommodation allowance and minimum living standard transfer for most low income households, including those out of work.29
48. The benefits have thus become somewhat more narrowly targeted (Figure 21). Depending on the exact circumstances, families with annual net incomes up to CZK 200-300,000 will receive larger benefits than in 2006 (in some cases substantial increases), while families with net incomes above this range, i.e. mid- to upper income families, will receive lower or no benefits.
Figure 21.Benefits (children and accommodation), 2006 and 2008
Source: Staff calculations. Note: Calculations shown for families with two and four children, respectively, all aged 10-15 years. Accommodation allowance based on normative costs in a rented apartment in a large city outside of Prague. Additional accommodation allowance and minimum living standard transfers not included.
49. The 2007/08 changes to benefits have moved against multi-children families and single parents: the difference in benefits between 2 and 4 children is now smaller than before (Figure 21) and the difference between couples and single parents larger (Figure 22).
Figure 22.Benefits (children and accommodation), Singles and Couples, 2006 and 2008
Source: Staff calculations. Note. Calculations shown for a single parent with two children and a couple with two children aged 10-15 years. Accommodation allowance based on normative costs in a rented apartment in a large city outside of Prague. Additional accommodation allowance and minimum living standard transfers not included.
50. The increase of the lower VAT rate is likely to be more or less neutral across the income distribution, with slightly higher impact on the low and high income groups, as discussed above. For all practical purposes, the reductions in the personal income tax will more than compensate low earning households. Since social transfers are no longer indexed to living costs, there will be no automatic compensation for the higher VAT rate for those receiving these transfers. However, many of those are likely to be part of households that receive income tax cuts.
51. High income earners will benefit not only from the lower marginal tax rate (15 versus 32 percent), but in particular from the cap on social security contributions (Figure 23). These features alone, notwithstanding the higher tax credits, will save high earners up to 20-25 percent of their gross earnings. The absence of individual earnings data for workers prevents an accurate estimate of the number of income earners which will benefit from the tax cuts. Based on 2004 data of those filing tax returns, the share of those with earnings above the cap on social security contributions is about 3 percent.
Figure 23.Overall Reduction in Marginal Taxation for High Income Earners
Source: Staff calculations, based on 2004 taxpayer data (those filing tax forms only).
52. The changes to the health care system could have major distributional implications. Although the distribution of health services in a universal coverage system like that of the Czech Republic is uniform ex ante, experience from other countries suggests that low income households consume more of these services than high income households (because of higher physical work-load and less healthy lifestyles). Hence, there is likely to be a regressive effect associated with the introduction of the user fee and the reduction in sickness benefits. For working individuals this is likely to be more than compensated by the tax cuts. For social benefit recipients, however, any excessive burden arising from the user fees would need to be addressed from the spending side, through targeted measures.30
Fiscal implications of the package
53. The fiscal package is likely to improve the government balance by about ½ percent of GDP in 2008 and ¼ percent of GDP in 2009, whereas it will worsen it from 2010 absent new measures (IMF, 2008). The short-term gains are mostly related to one-off measures, such as the postponement of the lower health care contribution from employers, and the gradual phasing in of the corporate tax cut. Once fully phased in, the reduction in employer health care contributions more than offset the additional costs for employers, resulting in a net loss for the government sector (Ministry of Finance, 2007b). Moreover, the non-indexation of benefits and the low wage increase envisaged for the public sector are likely to be short lived measures, as discussed above. Hence, the consequences of the reform package on fiscal sustainability are likely to be negative.31
54. Some factors could potentially render the budgetary impact more positive than projected by the Ministry of Finance. For instance, the personal income tax cuts will have second round effects on consumption and employment, which in turn may imply increased revenues from consumption and income taxes. However, the timing and magnitude of such effects are highly uncertain and depends on the exact design of the tax cuts—how many are affected at the margin, what are their elasticity of labor supply and their marginal propensity to consume. As discussed above, labor supply effects from the reform are not unambiguously positive and dynamic effects are thus likely to be modest, at best. The cut in the corporate tax rate may also lead to higher investment (again depending on elasticities to after tax cost of capital), economic activity and hence tax revenue—but again, the timing and scale are highly uncertain (and, as pointed out above, debt financed investment will become more expensive at the margin).
55. The reform package includes welcome steps to move the tax burden to indirect taxation, including environmental taxes, and lowering taxation on capital income, while also reducing health spending and the generosity of some social benefit schemes. However, it does not seem to fully deliver on its objectives, and certainly not in a cost efficient way.
56. First, the package will not help reducing the substantial fiscal gap—in fact, chances are that it will add to long term fiscal pressures since the tax cuts are well-defined and permanent while the offsetting spending measures are not; second, considering the labor market structure—with disincentives to work concentrated at low- and mid level wages—the personal income tax cuts do not appear to be well-targeted and the issue of unemployment and low wage traps created by the interaction of taxes and benefits remains unresolved; third, the introduction of a cap on social security contributions limits overall revenue efficiency in a system where these contributions to some extent replace taxes as a means of financing overall government operations. Going forward, there is a need to maintain the reform momentum while combining it with careful fiscal prioritization and policies that could achieve sustainable improvements in government balances and improved labor market outcomes.
E. A Way Forward
57. The Czech Republic maintains government spending levels that are closer to the EU-15 average than to it’s regional peers. Government revenues, therefore, needs to be higher than among countries with lower spending levels. And, revenues are not even sufficient to ensure healthy government finances at current spending levels.
58. This leaves little or no room for tax cuts, since the burden of reducing spending would be further aggravated if taxes were reduced. For the Czech Republic going forward, the difficult—and delicate—policy task is to find an appropriate balance between tax increases and spending cuts that result in the required long-term improvement of the government balance.
Strategies to boost employment
59. The 2008 reduction in the personal income tax, aimed at stimulating labor demand and supply, comes at the costs of untenable losses in government revenues. It might also be a relatively inefficient instrument in addressing the problems in the Czech labor market. Going forward, several alternative routes might thus be worthwhile considering.
60. First, there are ways to alter the tax wedge to make it less detrimental to work effort while still raising the same, or even more, revenue.32 Broadening the tax base is the first best solution. In the case of the Czech Republic, this could be done by shifting the tax burden from social security contributions to income tax and/or the value added tax for the part that is used for financing broader government operations (Box 2). This would spread the burden of taxation wider, i.e. not only to those working, but to people in unemployment or outside the labor market (in inactivity, studying or retirement), who also benefits from the welfare services. To be fully effective, such a base shift should be accompanied by a decision to make unemployment benefits and social benefits taxable.33
Box 2.Replacing Social Security Contributions with PIT or VAT?
The choice between using the personal income tax (PIT) or the VAT as substitute for social security contributions evidently involves a number of rather complex issues. It seems, however, that a strong case can be made for the Czech Republic to rely mostly on the latter:
- The traditional distributional argument - that the VAT is less progressive than the PIT – is less convincing when there is a flat income tax.
- Since social benefits are no longer automatically indexed to living costs, increases in the VAT rate will no longer automatically be compensated for recipients of social benefits, hence there is a reduction in the replacement rate that would otherwise not be there.
- The VAT does not distort savings and investment.
- Indirect taxes are a smaller part of total tax revenue than in other EU or OECD countries.
- VAT increases would help to alleviate intergenerational inequities as it taxes both past and present earnings.
61. A broader base would imply that the effective wedge on labor income would decline, and that effective replacement incomes would be reduced (if not instantly, then at least if future increases in the tax rate were to implemented at some point). This, in turn, would stimulate both labor demand and supply (OECD, 2007b).
62. Second, for those contributions that are more closely linked to labor market activity (such as pensions), the burden of the tax wedge could be reduced by establishing a closer individual link between the contributions and the benefits—hence making contributions feel less like a tax, being less distortionary. Some of the pension reform proposals currently under consideration will achieve this, but evidently would also require a rethinking of the distributional effects of the pension system.
63. Third, targeted measures to certain groups in and out of the labor market might be more effective (and certainly more cost efficient) than the sweeping across the board tax cuts that are now being implemented. In particular, since the problems of participation and long-term unemployment in the Czech Republic seem to be contained to limited and reasonably well-defined groups of the population, targeted active labor market measures would appear to be a workable solution in practice (further education; tighter and better enforced requirements for activation; limited and temporary subsidy schemes for on the job training; deregulation of labor and product markets).34
64. Fourth, labor market policies may need to target the young. Although educational attainment appears to explain most of the level and trend of youth inactivity, there might still be a need to strengthening educational incentives and opportunities as well as applying active labor market policies more effectively towards this age group (since there is a relatively high incidence of unemployment among the 20-24 years old). Moreover, the tax credits and child benefits should apply only for children under the age of 18 years. The current age threshold, at 26 years, is high by international standards and does not seem to correspond to neither the actual burden on parents nor the earnings capacity of the children. Reducing it to 18 years would reduce transfer spending, increase tax revenue and boost youth labor supply. A further relaxation of the employment protection of those on permanent contracts (especially rules of dismissal) would also help labor market performance in general and youth participation in particular, since these are the, by definition, outsiders on the labor market. Moreover, avoiding further substantial discretionary increases in the minimum wage (as those implemented over the past few years) would reduce the risk of further discouraging demand for younger individuals (as well as the demand for other exposed groups, such as low-skilled workers and long-term unemployed).35
65. Fifth, if more needs to be done to spur labor supply and demand at low earnings, there are better targeted instruments than lowering the tax rate or increase tax credits for all those in work. The two classical instruments are reductions in the employers non-wage costs (i.e. their social security contributions) for those with low earnings to spur labor demand, while supply can be stimulated through an earned income tax credit for low earning individuals or households.36 Both measures would, of course, have to be carefully designed (including their interaction with benefit systems), so as maximize their impact on employment at low earnings, while minimizing disincentives created further up the earnings scale as the tax preference is phased out, including not only the labor supply decision but also the incentive for undertaking education (Mullins, 2006).37 At the current cyclical juncture, with widespread shortages of labor supply, it would appear appropriate to focus more on labor supply incentives and less on stimulating labor demand.
66. Sixth, the spouse tax credit should be substantially reduced or eliminated as it now act as a significant barrier to spouse participation. It could also be considered whether part-time jobs to a larger extent could be used to promote participation of spouses and elderly individuals (in particular women), for instance by extending (part of) the new tax credit for working retirees to individuals working part time. The Czech Republic stands out among OECD countries as having one of the lowest incidences of part time jobs, a feature that might prevent flexibility and leave a non-negligible part of the labor pool untapped. The reductions in parental benefits introduced by the fiscal reform package will contribute to a faster return of mothers to the labor market, but it should be accompanied by enhanced childcare facilities.
67. Seventh, a strategy for the labor market would need to also consider broader measures such as further relaxation of labor market regulations and employment protection; improving overall educational attainment and implementing long-term strategies to reduce skill mismatches; and enhancing labor mobility (reducing rent controls, improving public transportation and infrastructure, allowing portability of pensions etc.). These are all areas that are beyond the scope of this chapter.
Strategies for reducing costs
68. OECD (2006) gauges that the overall employment rate could potentially be raised by up to 10 percentage point by increasing participation and reducing long term unemployment. This would not only help closing a large chunk of the income gap to EU-15, but also significantly bolster public finances. Hence, structural measures to boost employment along the lines listed above could dramatically improve the fiscal stance of the Czech Republic (by limiting spending on unemployment benefits and social transfers and raising more tax revenues from earnings and consumption).
69. However, even if such beneficial effects were to materialize, there would likely still be a need to cut government spending. Prime candidates are the health care and pension systems, but other areas could be subject to spending cuts as well, such as the entire system of social transfers, business and agricultural subsidies, the education system and other parts of government.38
70. One particular issue is the potential for devolving more spending responsibility to municipalities, or the newly established regional authorities. While evidently a non-trivial political and economic exercise, such an initiative could nonetheless be a stepping stone for a gradual increase in the taxation of real property, which is particularly low in the Czech Republic compared with other EU and OECD countries. It could thus form part of a strategy for tapping a relatively low distortion source of revenue, while also bringing spending decisions closer to local needs.
Strategies for increasing revenue
71. Given the substantial fiscal challenges, even structural improvements and spending compression might not be sufficient to ensure long-term fiscal sustainability. Hence, tax increases also need to be considered. While taxation of capital is constrained by the high and increasing international mobility of capital, there is still potential to increase most other taxes without excessive distortions to the economy.39
72. The most promising route, and even one that could potentially reduce distortions, would be to eliminate the low VAT rate altogether. There is much evidence as to the superiority of one single VAT rate in terms of efficiency and administration (Ebrill and others, 2001) and, as discussed above, the re-distributional properties of the low rate are questionable. Hence, gradually raising the low rate towards the 19 percent standard rate (or a slightly lowered unified rate) over the next, say, 10 years, could add considerably to reducing the fiscal gap.40 Some compensation would have to be provided to the most exposed households, but targeted spending measures would be a better and cheaper route to follow. An alternative, possibly less contentious (but also less effective) way of raising revenue would be to move some of the items currently taxed under the low rate to the standard rate (such as package tours, transfer of housing etc.).
73. The personal income tax could also contribute to raising revenue, both by reducing the value of the tax credits (for instance by not indexing them for a while), and by slightly raising the 15 percent flat rate (alternatively re-introduce progressivity, for instance a second rate for high earnings).
74. In a system, where social security contributions de facto replace taxes as a way of financing overall government operations, these contributions should not be capped. The cap does not effectively serve its traditional purpose of strengthening the link between contributions and benefits (which in any case could be pursued through pension and health care reform) and it is detrimental to overall revenue efficiency.41 The contributions of the self-employed should also be aligned with those of employees. This would not only require removing the cap on those contributions as well, but also expanding the base on which the contributions from the self-employed is calculated: currently, this base is 50 percent of net operating income, implicitly assuming that labor’s share in total value added is 50 percent, rather than the 65 percent normally assumed at the aggregate level (Bronchi and Burns, 2000). The self-employed should also be obliged to pay the sickness contribution.
75. Closer alignment of the tax burden of the self-employed with those in employment would increase revenue and reduce horizontal inequities—in particular, the widespread practice of “pseudo” self-employment among high income earners, by which people are hired under contract to work as self-employed rather than under standard employment contracts, so as to avoid paying the full social security contributions levied on employees (OECD, 2006). Importantly, the relatively low marginal tax rate on personal income means that eliminating the cap on social security contributions for the self-employed could be done without expecting massive incentives to convert employment income into capital income (to avoid paying the social security contributions): the marginal tax rate on capital income is 35.4 percent, while it is 32.8 percent on employment income.42 While the “pseudo” self-employed would generally not have the alternative to work in the informal sector, others might be tempted to cross the line, and administrative vigilance would be required.
76. Environmental taxes could be increased further. The Czech Republic has relatively low taxes on fuels and electricity and these have even declined as a percent of energy consumption over the past ten years (OECD, 2007b). Gasoline and motor vehicle ownership taxes would be among the least regressive (or even progressive) environmental taxes to increase.
77. Taxes on real estate could be good candidates for increase, since these are lower than in most other EU and OECD countries. However, in line with most other countries, these taxes are levied at the sub national level, hence it would require a re-design on central-local government relations to implement an across the board surge in these taxes for broader financing purposes.
78. The reforms to the tax and welfare systems, while leading to short-term budgetary improvements, do not deliver on the key objective of reducing the fiscal gap. They might even add to the medium-to long-term fiscal pressures. The most troubling part is the reform to the personal income tax and social security contributions, which besides losing substantial revenue appear not to be well-targeted, considering the level and structure of labor market issues. The spending measures will help to streamline, and to some extent reduce, public spending, but they are insufficient and some of them are likely to be unsustainable.
79. The most promising part of the reform is the increase in the low VAT rate, the new excises on energy consumption, and the initiatives to control health care spending and reduce abuse in the sickness benefit system.
80. Going forward, it is imperative to flesh out and implement substantial spending reducing reform in the pension system and possibly also in the health care system and elsewhere in the government sector. Tax increases are also likely to be necessary to fill the fiscal gap, with the key options being increasing the low VAT rate further and scaling back personal tax credits or raising the rate of the personal income tax. Raising excises and broadening the corporate tax base could also be part of a revenue raising strategy.
81. Structural reform to enhance employment should be addressed urgently, but refocused from across the board tax measures to better targeted instrument. If successful, such reforms, by themselves, would alleviate much of the pressure to reduce spending and increase taxes.
Box 3.Summary of Recommendations
- Reduce social security contribution rates, while increasing the personal income tax and/or the VAT to achieve revenue neutrality;
- Get rid of cap on social security contributions, incl. for the self employed;
- Move gradually to one single VAT rate;
- Explore the scope for changing the child tax credit into a targeted earned income tax credit;
- Redesign child benefits and accommodation allowance such that, in combination with the tax system and social security contributions, barriers to move up income ladder are reduced;
- Reduce or eliminate the spouse tax credit;
- Introduce more and better targeted active labor market measures, in particular effective enforcement of activity requirements;
- Promote labor market flexibility, e.g. part time employment and relaxation of employment protection legislation;
- Phase out corporate tax loopholes;
- Spending restraint (and flexibility) reform: pension; health care; social transfers; other areas;
- Fill remaining fiscal gap by tax on consumption, pollution, personal income, and possibly strengthening the property tax.
- Search Google Scholar
- Export Citation
- Search Google Scholar
- Export Citation
Annett, Anthony, 2007, “Lessons from Successful Labor Market Reformers in Europe,” IMF Policy Discussion Paper no. 01/07.
Bassanini, Andrea and RomainDuval, 2006, “Employment Patterns in OECD Countries: Reassessing the Role of Policies and Institutions,” OECD Social, Employment and Migration Working Paper No. 35.
Benes, Jaromir and PapaN’Diaye, 2004, “A Multivariate Filter for Measuring Potential Output and the NAIRU: Application to the Czech Republic,” IMF Working Paper No. 04/45.
Bronchi, Chiara and AndrewBurns, 2000, “The Tax System in the Czech Republic,” OECD Economics Department Working Paper No. 245.
Brook, Anne-Marie and WilliLeibfritz, 2005, “Slovakia’s Introduction of a Flat Tax as Part of Wider Economic Reforms,” OECD Economics Department Working Papers No. 448.
Burniaux, Jean-Marc, FlavioPadrini and NicolaBrandt, “Labor Market Performance, Income Inequality and Poverty in OECD Countries,” OECD Economic Department Working Paper No. 500.
Calmfors, Lars and JohnDriffill, 1988, “Bargaining Structure, Corporativism and Macroeconomic Performance,” Economic Policy, vol. 6.
Caliendo, Marco, LudovicaGambaro and PeterHaan, 2007, “The Impact of Income Taxation on the Ratio between Reservation and Market Wages and the Incentives for Labor Supply,” Applied Economics Letters, forthcoming.
De Mooij, Ruud A. and SjefVederveen(2001), “Taxation and Foreign Direct Investment: A Synthesis of Empirical Research,” CESifo Working Paper No. 588, October2001.
Devereux, Michael P., RachelGriffith and AlexanderKlemm, 2002, “Corporate Income Tax Reforms and International Tax Competition,” Economic Policy, Vol. 35 (Oct).
Devereux, Michael and RachelGriffiths, 2003, ‘Evaluating Tax Policy for Location Decisions,’ International Tax and Public Finance10(2): 107–26.
Docteur, Elizabeth and HowardOxley, 2003, “Health Care Systems: Lessons from the Reform Experience,” OECD Economic Department Working Paper No. 374.
Ebrill, Liam, MichaelKeen, Jean-PaulBodin and VictoriaSummers, 2001, The Modern VAT, International Monetary Fund, Washington, D.C.
EU-Commission/Economic Policy Committee, 2006, “The Impact of Ageing on Public Expenditure: Projections for the EU-25 Member States on Pensions, Health Care, Long-Term Care, Education and Employment Transfers (2004–50), European Economy Report and Studies No. 1.
Feldstein, Martin, 1995, “The Effects of Marginal tax Rates on Taxable Income: A Panel Study of the 1986 Tax Reform Act,” Journal of Political Economy, vol. 103, no. 3.
Gordon, Roger H. and James R.Hines, 2002, “International Taxation,” in Handbook of Public Economics, vol. 4 (edited by Alan J.Auerbach and MartinFeldstein), Elsevier Science B.V.
Hall, Robert E., and AlvinRabushka, 1985 and 1995 (2. edition), The Flat Tax, Hoover Institution Press, Stanford, California.
Heckman, James, LanceLochner and ChristopherTaber, 2002, “Tax Policy and Human Capital formation,” American Economic Review, vol. 88, no. 2.
Hemmings, Philip and EdwardWhitehouse, 2006, “Assessing the 2005 Czech Proposals for Pension Reform,” OECD Economics Department Working Paper no. 496.
Hurnik, Jaromír Hurník and DavidNavrátil, 2005, “Labor-Market Performance and Macroeconomic Policy: Time-Varying NAIRU in the Czech Republic (in English),” Czech Journal of Economics and Finance (Finance a uver, ISSN: 0015-1920), Charles University Prague, Faculty of Social Sciences, vol. 55(1-2), pages 25-40, January.
IMF, 2004, Czech Republic: Selected Issues, IMF Country Report no. 04/265.
IMF, 2005, Czech Republic: Selected Issues, IMF Country Report no. 05/275.
IMF, 2007, Czech Republic: Selected Issues, IMF Country Report no. 07/85.
IMF, 2008, Czech Republic: Staff Report.
Juradja, Stepan, and DanielMunich, 2002, “Understanding Czech Long-Term Unemployment,” William Davidson Working Paper No. 498.
Keen, Michael, YitaeKim, and RicardoVarsano, 2006, “The “Flat Tax(es): Principles and Evidence,” forthcoming in International Tax and Public Finance.
Moore, David, 2005, “Slovakia’s 2004 Tax and Welfare Reforms”, IMF Working Paper No. 05/133.
McKinsey Global Institute, 2003, New Horizons: Multinational Company Investment in Developing Economies (October).
Ministry of Finance of the Czech Republic, 2007a, Fiscal Outlook April 2007.
Ministry of Finance of the Czech Republic, 2007b, Convergence Program.
Mullins, Peter, 2006, “Key Considerations in Contemplating a Negative Income Tax,” IMFmimeo.
Nickel, Stephen, 1997, “Unemployment and Labor Market Rigidities: Europe versus North America,” Journal of Economic Perspectives, vol. 11, no. 3.
OECD, 2007a, Economic Outlook May 2007, Paris, France.
OECD2007b, Employment Outlook 2007, Paris, France.
OECD, 2006, OECD Economic Surveys: Czech Republic, Paris, France.
Schiff, Jerald, PhilippeEgoume-Bossogo, MihoIhara, TetsuyaKonuki, and KorneliaKrajnyak, 2006, “Labor Market Performance in Transition: The Experience of Central and Eastern European Countries,” International Monetary Fund, Occasional Paper No. 248.
Schneider, Ondrej, 2004a, “Who Pays Taxes and Who Gets Benefits in the Czech republic”, Institute of Economic Studies, UK FSV – IES Working Paper No. 68.
Schneider, Ondrej, 2004b, “Protecting people While Enhancing Work Incentives: The Role of the Welfare System,” in Czech Republic: Ageing and Employment Policies, OECD, Paris.
Sørensen, Peter Birch, 1999, “Optimal Tax Progressivity in Imperfect Labour Markets”, Labour Economics, vol. 6.
Tirpak, Marcel, 2007, “Migration in EU8 countries”, mimeo, Warzaw Regional Office of the International Monetary Fund.
Tuladhar, Anita and DennisBotman, 2008), “Tax and Pension Reform in the Czech Republic—Implications for Growth and Debt Sustainability”, Art. IV Selected Issues Paper, International Monetary Fund.
Personal Income Tax: (i) Introduction of a 15 percent flat tax based on a super gross wage base that includes health and social contributions of 35 percent for the employer and 12.5 percent for the employee (calculated on the existing base, the new tax rate is 23.1 percent)—replacing a progressive tax rate scale of 12, 19, 25 and 32 percent; (ii) abolition of joint taxation for married couples; (iii) abolition of the minimum tax base for small enterprises and capping their annual cost allowance; (iv) unified final withholding rate on dividends and interest of 15 percent; (v) enhanced tax credits (like in the current system, only the child credit is refundable):
|2008 tax system|
|-- for taxpayer||7,200||24,840|
|-- unemployed partner|
|-- disability (1stlevel, partial)||1,500||2,520|
|-- disability (2ndlevel, full)||3,000||5,040|
|-- disability (3rdlevel, hard)||9,600||16,140|
|Tax bonus per child||6,000||10,680|
Corporate income tax: (i) phased reduction of corporate tax rate: from 24 percent in 2007 to 21 percent in 2008, 20 percent in 2009 and 19 percent by 2010; (ii) limited base broadening measures, e.g. stricter limits for deduction of financial expenses.
VAT: Increase in lower rate from 5 percent to 9 percent.
Property tax: (i) exemption of farm land from real estate tax at the discretion of the municipalities; (ii) setting of coefficients for real estate tax on buildings at the discretion of municipalities; (iii) reduced inheritance and gifts tax.
Environmental tax: Introducing taxes on electricity, coal, solid fuels and natural gas.
Excise: Increase in the excise tax on cigarettes and tobacco.
Tax administration: Raising the turnover threshold for book-keeping requirements; cancellation of the required use of cash registers with fiscal memory.
Social contributions: ceiling at four times the nominal average wage for employees and self-employed (previously there was no cap for employees whereas the cap for self-employed was about two times the minimum wage).
Sickness insurance: (i) one year postponement of the law on sickness insurance, which decreases employer contributions while shifting the responsibility of half the payments for the first 14 days of sickness to the employer; (ii) cancellation of benefits for the first 3 days of sickness; (iii) reduction in sickness benefits (mostly for short term sickness, i.e. less than one week).
Casualty insurance: Postponement of law on casualty insurance supplementing insurance payments with social benefits for injured workers.
Health insurance: (i) introduction of a nominal, flat co-payment fee for doctor’s visits, emergency visits and hospital stays; (ii) freezing of budget contributions for state-dependents till 2009; (iii) an independent committee of experts is being formed to evaluate the systemic health care reform proposals of the different political parties to be implemented in 2010.
Social benefits: (i) deindexation of social benefits (except pensions) from the level of subsistence minimum and minimum living standards; (ii) the level of subsistence minimum and minimum living standards to be decided at the government’s discretion; (iii) the level of parental and child allowance decided in absolute nominal amounts with changes requiring a new Act; (iv) standard child benefit now capped at 2.4 times the minimum living standard (MLS) against previously 4 times the MLS (Table A2); (v) the additional child benefit (“social extra allowance”) limit reduced from 2.2 times the MLS to 2.0 times the MLS;43 (vi) reductions in the maternity and parental benefits, the birth allowance, the foster care benefits and the funeral allowance; (vii) cancellation of the teaching aid allowance; (viii) enhanced targeting of disability (from two to three grades); (ix) stricter rules for receiving unemployment benefits (reasons for termination of employment and denial of activation); and (x) enhanced childcare facilities.
For analytical purposes, this paper also adds the changes made in January 2007 to the household accommodation allowance (which basically takes accommodation expenses out of the MLS calculation, while introducing a new allowance for accommodation expenses, depending on income, type of accommodation, geographical location and family size).
The pre-2007 accommodation allowance was calculated as: Minimum living standard for household costs*(1-net income/(MLS_family*1.6)
The new generic formula for the accommodation allowance is: Cost of accommodation (or the norm cost) – 0.3*net family income
|Current regime||New regime|
|Family income||0-1.5 MLS||1.5-2.4 MLS||2.4-4.0 MLS||0-2.4 MLS|
Pensions: No legislative changes are currently in process. Discussions under the first phase of reforms have focused on (i) increasing the statutory retirement age further (beyond the currently legislated 63 years which is being phased in by 2013); (ii) tightening of eligibility criteria for pension benefits (from 25 to 35 years of occupation); and (iii) enhancing incentives for late retirement. Under the second phase of reform, proposals include (iv) second pillar reform to separate the assets of clients and fund sponsors; (v) prefunding measures to create a separate account for saving future pension surpluses; and (vi) incentives for supplementary pensions. Third phase of reforms will include models for opting out of the public pension scheme, etc.
Wage bill: (i) freeze wages of judges and MPs at 2007 level; (ii) limit nominal payroll increase in the public sector to 1.5 percent annually during 2008-2010; (iii) reduce number of civil servants by 3 percent annually until 2010.
Senior economist with the IMF’s European Department. I am grateful for comments received from Subhash Thakur, Michael Keen, Sonia Munoz and seminar participants at a European Department seminar as well as a joint conference (IMF/CNB/Ministry of Finance) held in Prague, November 2007, in particular Stepan Jurajda. All remaining errors are mine.
The Ministry of Finance (2007a). The Ministry also finds that, in order to meet the EU requirement for government debt not to exceed 60 percent of GDP at any point in time, an improvement in the government balance of about 4 percent of GDP is necessary. This could be interpreted as a minimum long-term requirement for improvement in the government balance.
Throughout the paper we compare, whenever possible, with three other new EU member countries and regional competitors (Poland, Hungary and Slovakia), two established EU members with widely differing tax and social systems (Germany and the United Kingdom) as well as the EU-15 average.
The current unemployment rate is substantially below recent national estimates of the rate of structural unemployment (the NAIRU) at about 7-7.5 percent (Benes and N’Diaye, 2004; Hurník and Navrátil, 2005)—a level broadly in line with the euro-zone average, but somewhat above levels in for instance the United Kingdom, the Netherlands, Denmark and Sweden (OECD, 2007a).
As noted in IMF (2004), the thrust of the restructuring process happened somewhat later in the Czech Republic —in the late 1990s—than in many of its neighbors.
The minimum wage, for instance, was raised twice during 2006, bringing it to about 40 percent of the average wage (against 20 percent in the late 1990s).
It is envisaged that the flat rate will be reduced to 12.5 percent in 2009, but these plans—and their possible financing—are uncertain and hence not considered in the following.
Throughout this paper we follow the standard methodology of analyzing effective tax rates, i.e. including personal income tax, social security contributions and the effects from generally available benefits such as the child benefits and the accommodation allowance (see for instance OECD, Taxing Wages 2005/06). The norm cost for rented apartments outside of Prague is assumed to reflect actual housing costs. The norm cost is assumed to increase by about 20 percent from 2007 to 2008 (estimates of the Ministry of Social Affairs).
In particular, there was a substantial scaling up of the accommodation allowance effective January 1, 2007, where the allowance was changed from an minimum housing costs principle to one based on actual costs (subject to a limit). The change was triggered by the elimination of rent controls from the same date and other administrative increases in the cost of housing. Other new benefits were also introduced in 2007, such as the new system for minimum living standard transfers and additional accommodation allowances.
The scaling up of tax credits also influences the various social transfers through the effect on the take home pay and hence the means testing. An exception is the non-wastable child tax credit, which is excluded from the income measure used for means testing.
In the current system, the income tax would be positive for earnings exceeding CZK 60,000 per year (for a single wage earner without children). In the new system, the “no-tax” threshold is raised to about CZK 120,000.
The accommodation allowance alone adds about 26 percentage points to the effective marginal tax rate at low incomes (31 percentage points in Prague) and the additional child benefit typically 20-40 percentage points (depending on the number of adults and children in the family).
For instance, a single mother with two children going from the minimum wage of about CZK 100,000 to an annual wage of CZK 200,000, will, in 2008, receive only about CZK 22,000 in additional take home pay—the rest, CZK 78,000, is taken away as taxes and, in particular, reduced benefits. In the 2006 system, she would keep about CZK 47,000 of the increased gross earnings as take home pay and leave about CZK 53,000 in taxes and reduced benefits.
A primary earner with an unemployed spouse, for instance, will see a reduction in the effective tax take from 99 to 50 percent if moving from unemployment benefits (based on a previous job at the average wage) to a minimum wage job. If, instead, she moved into an average wage job, the effective tax take would increase from 64 to 67 percent.
Recent data on migrant flows shows that emigration is far less of an issue for the Czech Republic than in Poland, Slovakia or the Baltic countries (Tirpak, 2007).
Joint taxation is normally found to enhance the labor supply – hours --of the primary earner (through lower marginal tax rates in a progressive system), while discouraging the labor supply of the second earner through both higher marginal rate than if he/she was taxed individually and lower average tax rates for the household (see for instance Caliendo and others (2007) for a study of the effects of joint taxation in Germany).
The need for the investment tax credit and the “double-deduction” of R&D spending, however, should be reconsidered as part of the rate cutting strategy—not only to broaden the tax base but particularly to reduce the distortions associated with such incentives.
Implying that a corporate tax rate at, say 20 percent, would bring in corporate tax revenue of 4 percent of GDP (0.2*20).
Bearing in mind that it is the headline tax rate which is the relevant rate to consider in shaping incentives to shift paper profits in and out of the Czech Republic by transfer pricing or financial arrangements.
Technically, the AETR is defined as the proportion of the lifetime pre-tax profit of some investment that is taken in tax. The METR is defined as the difference between before and after tax returns on a project that the investor finds just worthwhile. Both the AETR and the METR may vary across different assets and by methods of finance.
The AETR’s and METR’s calculated here are based on the same methodology and assumptions as applied by Devereux and others (2003), for a range of OECD countries (see also www.ifs.org.uk). Personal income taxation is not considered in these calculations (assuming the marginal investor is a de facto tax exempt entity). If personal income tax were added to the calculation, the Czech Republic would have a lower overall rate than Hungary. Special tax preferences and investment tax credits are not considered.
The evidence suggest that investment decisions do respond to tax considerations. Conducting a meta-survey of 25 empirical studies, de Mooij and Vederveen (2001) find that the mean of the various estimates of the (semi-) elasticity of inward FDI to the host country tax rate is –3.3: that is, a 1 percentage point reduction of the host country tax rate raises FDI in that country by 3.3 percent. Gordon and Hines (2002), and Devereux and Griffith (2003), among many others, provide additional evidence of the significance of taxation. It is also widely recognized, however, that other factors—like education of the labor force, infrastructure, red tape, and, not least, good governance and macroeconomic and political stability—are at least as, or maybe even more, decisive for investors than are tax incentives (see for instance the McKinsey Company, 2003).
The corporate income tax system might also be enhanced by further tightening of the rules for transfer pricing and thin capitalization, but experience from other countries suggest that these are more likely base protecting, rather than base expanding, measures.
After January 1, 2008, only residential property being provided as part of social policy will continue to enjoy the low VAT rate. The Ministry of Finance has prepared an amendment to the Value Added Tax Law that would make it possible to apply the lower 9 percent VAT rate to, among others, the transfer of apartments of up to 120 m2 and family houses of up to 350 m2 (floor area).
Assuming no substitution across spending items or over time.
The continued indexation for retirees does not matter as much for participation since the stock of retired individuals are unlikely to react strongly to changes in replacement rates (whereas the flow into retirement will depend on the indexation system).
To the extent the changes overall help to reduce long-term unemployment and raise productivity among low-skilled workers, there could also be benign dynamic distributional effects for these groups – however, as discussed above, it is not evident that such effects would be substantial.
The government will introduce measures to cut prices for some drugs in order to compensate for the increase in the low VAT rate and the prescription fee.
In theory, the composition of the labor tax wedge on social security contributions paid by the employer; social security contributions paid by the employee; personal income tax; and tax paid on goods and services, does not matter for labor supply and demand (the so-called invariance of incidence proposition—see e.g. Nickell, 1997). However, in practice, the composition can have a substantial influence, for instance if the minimum wage is binding for large groups (Bassanini and Duval, 2006), or if social and unemployment benefits are indexed to gross, rather than net wage developments (OECD, 2007b). Ultimately, the final incidence of any tax is determined by the wage-formation process and relative bargaining strengths of firms and the various groups inside and outside of the labor market.
A technical solution would need to be found to ensure that benefits, like earnings, are taxed on a “grossed-up” basis.
For a recent international comparison and discussion of active labor market measures, see OECD (2007b). See also Annett (2007), which provides empirical evidence on the effects on labor market performance (employment and wage growth) of changes to fiscal, product market and labor market policies in four successful labor market reformers in Europe.
More analysis is also needed to understand the nature and causes of the low youth participation, in particular the role of the education system.
Although the two measures could yield similar results, their relative effectiveness depends on the supply and demand conditions on the labor market, in particular specific rigidities such as binding minimum wages, structural unemployment etc. The former strategy has especially been adhered to in continental European countries, whereas the latter has mostly been used by Anglo-Saxon countries (OECD, 2007b). Reducing social security contributions at the low end of the earnings scale should always happen through the employer contribution so as to ensure that labor costs are reduced, also for those working at the minimum wage.
The earned income tax credit, in particular, has also proven to be administratively non-trivial in both the United Kingdom and the United States, hence the technical issues associated with its implementation should not be underestimated.
On the corporate side, the special regimes and investment tax credits should be scaled back and the regimes for transfer pricing and thin capitalization tightened.
The Ministry of Finance estimates that about 28 percent of all taxable goods and services are currently subject to the low rate. The increase from 5 to 9 percent in 2008 will yield additional revenue of about 0.6 percent of GDP per year, while further increasing the rate to 19 percent would add an extra 1.3 percent of GDP to the annual revenue gain.
Although a cap on social security contributions is not uncommon among OECD countries, a majority of countries levying social security contributions does not have a cap (OECD tax database). And, although the direct revenue loss from implementing the cap is relatively small (0.12 percent of GDP according to Ministry of Finance estimates), there might be more substantial consequences down the road from instituting such a practice (in particular, aligning the social security contributions of the employed and the self-employed without a cap could potentially result in more significant revenue improvements).
Capital income is taxed at the margin by 24 percent corporate tax plus (1-0.24)*15 percent on dividends, i.e. a total of 35.4 percent (to be reduced gradually to 31.2 percent as the CIT rate is reduced to 19 percent). Without caps, employment income is taxed at the margin by (1+0.35)*15 percent per plus 12.5 percent social security and health contribution, a total of 32.8 percent.
The additional child benefit allowance (“The social extra allowance”)
Current formula: Benefit = child’s basic personal requirement * (1 – (net family income/(MLS*2.2))) If family net income is lower than the MLS_family (Minimum Living Standard), the MLS_family is used in the formula instead of family net income.
Proposed formula (source: Public Finance Reform 2007-2010, Ministry of Finance): Benefit = child’s basic personal requirement * (1 – (net family income/(MLS*2.0))) If family net income is lower than the MLS_family (Minimum Living Standard), the MLS_family is used in the formula instead of family net income.