Moral Hazard in property tax administration: a comparative analysis of the Czech and Slovak Republics.

Bryson, Phillip J ; Cornia, Gary C

INTRODUCTION

As the transition in central and east Europe proceeds, fiscal decentralisation is expected to enhance the efficiency of public services as government is brought closer to the people. The ideas supporting this view of decentralisation have a long history in the public finance, public choice, and public management literatures (see, eg, Tiebout, 1956; Bish and Kirk, 1974). Decentralisation makes the decisions of public servants more transparent, permits citizens to participate at low cost, and most importantly, participate more effectively (Oates, 1998).

Still, many approach decentralisation with caution, citing the inability especially of small, poorly endowed governments to provide services. Critics claim that, as provided service benefits spill over into areas beyond those doing the funding, local governments will respond in their narrowly perceived self-interest by reducing (in a welfare sense, under-providing) those services.

Successful decentralisation requires effective performance of a whole set of complex tasks. As the World Bank (1996, p. 110) notes, `the state has to move from doing many things badly to doing fewer core tasks well'. The reassignment of personnel and responsibilities, often with a reduction in services, is not always easy for citizens to understand and support (Amsden et al., 1994).

Fiscal decentralisation, the public sector counterpart of privatisation, also requires that local governments have access to an independent and autonomous source of tax revenue (Bird et al., 1995). Local activity increases the visibility of choices made by elected and appointed public officials, and accountability increases pari passu with visibility (Litvak et al., 1998). The literature suggests that the property tax embodies the positive characteristics required of a local tax, because it is immobile, stable, potentially neutral, and (as a direct tax) visible to taxpayers (Musgrave, 1993; Oates, 1996).

As is true of decentralisation generally, the design and implementation of an effective property tax is challenging. Since it is direct and visible, it makes citizens and officials less comfortable than do indirect taxes such as the VAT (Youngman and Malme, 1994). Both citizens and officials usually prefer the use of excise taxes and fees on a variety of transactions (Shleifer and Vishny, 1998), since the property tax too often generates political problems for local officials (Paugam, 1999).

There are also administrative and practical problems with property tax implementation (Bahl and Linn, 1992). Examples of the latter include the difficulty of establishing market-oriented property values in the absence of a functioning real estate market (Bertaud and Renaud, 1994) and the uneven distribution of the property tax base (Netzer, 1966). As a consequence of political and practical issues, the property tax remains critically under-used (Dunn and Wetzel, 2000). (1) Especially in the transition countries, this can be an impediment to successful decentralisation and undermine the welfare gains expected from that process.

In this article, we compare the implementation and administration of the property tax in the transitional Czech and Slovak republics. In indicating that access to independently generated funds can help make fiscal autonomy possible, we do not argue that one can gauge fiscal autonomy by measuring property tax receipts. There are other independent sources of revenue for subnational governments such as the local user fees collected for public services and the proceeds from privatisation. The property tax, however, still offers largely untapped potential (Bird, 1999). The effectiveness of collective efforts cannot, of course, provide an actual measure of local autonomy. However, the fiscal effort applied, as we will see in the Slovak case, is an important element in the attempt to develop fiscal independence.

There are significant differences in the importance and growth of the property tax between these two countries. In Slovakia, the property tax is increasing in importance at a rate that will cause associated revenues to double approximately every 6 years. Conversely, in the Czech Republic the property tax as a local revenue source is declining in relative importance. These differing trends are interesting, because the two countries have uniquely common roots extending back to the early 20th century, including a 40-year period when they shared a united government. Further, the commonality of the two fiscal systems in the Czechoslovak federation continued through the Velvet Revolution of 1989, and even beyond the 1993 `Velvet Divorce' that created the new Slovak and Czech republics.

It would be easy to assume that the difference in the fiscal importance of the property tax is due to a difference in the tax base and tax rates. However, it is notable that the two republics have retained identical property tax laws, identical area-based valuation methods, and even identical rates that have remained unchanged since the Velvet Divorce in 1993 (Bryson and Cornia, 2001a, b).

Why, then, do we observe the divergent rates of growth in Czech and Slovak property tax revenues? At least part of the answer to this question can be found in basic tax administration, which we assume, along with Bird and Slack (1991), Hubbard et al. (1993), and Shah (1994), makes an important difference in how successful a tax will be. In this article, we focus on three policy decisions that have had significant fiscal consequences for the Czech and Slovak property taxes. First, which level of government, central or subnational, should have responsibility for the policy design of the property tax? Second, which level of government should have administrative responsibility? Third, what level of funding should the centre provide to subnational governments or municipalities?

Both countries centralised property tax policy issues, but the Czechs centralised property tax administration and the Slovaks decentralised it. With respect to the third issue, the level of funds transferred, the two republics have also performed differently: The Czech central government has been far more generous with respect to the aggregate of funds transferred to local governments.

These sister republics, with their common origins, provide a rare opportunity for a study of fiscal decentralisation. They permit comparisons for which other things actually are equal, and we can `control' for some administrative and policy decisions made after the Velvet Divorce in 1993 that appear to have had a direct effect on the property tax. The examination of these issues should add insight into more general problems of fiscal decentralisation, since the issues involved have important implications for transition processes across the region and in other parts of the developed and developing world.

Having briefly revisited in this introduction some of the issues pertaining to fiscal decentralisation and the need for an independent revenue source for the achievement of local self-government, the paper will proceed as follows. We review in the next section the formation of local governments and the history of fiscal federalism in the two republics since the beginning of the transition era. In the subsequent section, the level of government responsible for property tax policy and administration from the perspective of principal-agent relations is discussed. In doing so, we will take into consideration the moral hazard problems such relations imply for property tax systems. In the context of both national and local budgets of the twin republics, municipal revenues derived from the property tax and other fiscal transfers are addressed in the penultimate section. The last section briefly summarises our conclusions.

MUNICIPALITIES IN THE CZECH/SLOVAK TRANSITION

During Czechoslovakia's central planning period, decisions bearing on local public services were made exclusively by the central government. Regional and local governments did not disappear, but their role was certainly not an independent one; they merely administered and facilitated the federation's policies (an activity termed `state administration.') Under central planning, local autonomy was also compromised by the very institutions of intergovernmental finance. Funding for police, public utilities, fire protection, and education--not to mention housing, food, and medical care--was provided by the centre and financed by indirect enterprise taxes.

After 1989, Czechoslovakia began to re-establish local governments charged with the delivery of public services. Since, in the central planning era, hyper-centralisation had run counter to Czech and Slovak traditions and preferences, the initial decentralisation activities were bold (Jabes and Vintar, 1996). The new order established in 1990 permitted the formation of new local governments, which soon numbered 2,781 in Slovakia and 6,234 in the Czech region. Although neither region initially developed a genuine, intermediate level of government between the centre and the municipalities, the Czechs recently established regional governments and the Slovaks have proposed to do so.

Most of the municipalities in the two republics have fewer than 500 inhabitants, many of them less than 100. In the smaller towns and villages, there is obvious concern about whether they have sufficient size to provide an acceptable level of services or raise funds to finance such services. As Table 1 indicates, Slovakia and the Czech Republic have significantly more municipalities per 10,000 citizens than their regional neighbours. Local autonomy is intensely important to Czechs and Slovaks.

Contemporaneous with the re-establishment of local governments was the assignment of responsibilities to them from the centre. In the early days of the transition, local governments were assigned fire, police, sanitation, water, recreation, and cultural functions. They were also given responsibility for public records, building codes, regulation of the environment, and pre-high school education. As functions were assigned to the local governments of both republics, a series of earmarked and conditional grants were also established. They were allocated on a per capita basis to fund social services, education, and public safety.

In both republics, such grants have been unreliable and have come with strings attached. Oliveira and Martinez-Vazquez (2001) indicate that in the Czech Republic there are `numerous constraints and conditions placed on the transfers' to local governments, which they fear may be `too restraining to local autonomy; They find the current grants system `a rather unstable source of revenue' which `has damaged the ability of local governments to plan and budget their expenditures in an effective manner' (p. 56). The local governments in both republics have been unhappy with the amount and uncertainty of the grants.

In a significant decision, perhaps as a substitute for the revenues essential to the performance of their tasks, local governments also received as a transfer from the centre many of the public housing units and small, state-owned businesses located within their jurisdictions. The transfer was a mixed blessing. Although both housing and public businesses could be viewed as potential revenue sources, both were badly in need of funding for operations and maintenance investments. Local governments responded to the transfer of these facilities by either selling them outright or going through a non-market process of privatisation. Unfortunately, privatisation often created even more problems. The properties sold were the ones requiring the least amount of repair. The unsold properties were the most difficult to maintain and operate. Consequently, local governments have been left with a number of non-performing assets. Moreover, the receipts collected from such sales were generally used to fund the annual operating expenses of those municipalities involved, but they were insufficient to fund anything beyond that.

TAX POLICY, TAX ADMINISTRATION, AND MORAL HAZARD

As noted above, a federalist country adopting a tax must decide which level of government will design tax policy, which will administer the tax, and which will collect the revenues. Often, the same level of government performs all three functions, just as the federal income tax is handled in the United States. Of course it need not happen this way. In Taiwan, the VAT is largely administered at the provincial level, policies governing its use are set by the national government, and the revenue is split between the two government levels. In the United States, state governments establish property tax policy and local governments administer the property tax, but the majority of the revenue goes to school districts, which have no policy or administrative role in the process.

Sound arguments can be made either for combining policy and administration at one level of government, or for separating them. There are also arguments for centralising and for decentralising the two functions. Table 2 illustrates the potential combinations of policy and administrative tradeoffs facing tax design.

There are arguments for placing most of the property tax's policy and administrative activities in the first cell of Table 2. In developing and transitional countries, for example, centralisation of policy and administration usually assures more abundant fiscal and administrative resources (World Bank, 2001). Centralisation would increase access to policy resources, including technical expertise, computers and information, more sophisticated technologies, international advisors. It would also enable an overall harmonisation of tax policies.

Another reason for policy centralisation is the key issue of establishing policy safeguards to assure property tax uniformity, both between tax jurisdictions (harmonization) and within them (uniformity). A centralised set of policies would describe the types of property to tax, how valuations are to be made, the level and frequency of property assessment, the tax rate applied, the use of the funds, and the required degree of transparency (Kelly, 2000).

The need to create harmonisation and uniformity safeguards is an issue for every tax, especially the property tax. For all other taxes the taxable value, or tax base, is established in an arm's-length, market transaction. However, this is not possible for the property tax in transition economies, since property is sold or exchanged so infrequently that individual parcels with market-determined values cannot be seen as representative. Thus, a tax administrator must estimate a market and a taxable value for each parcel, and there is a strong chance that political or other less relevant issues can influence the process of estimation. Even in developed countries the lack of property tax uniformity can be a serious problem (Sirmans et al., 1995).

The challenges to the property tax can be even more fundamental than inept estimation. Unfettered local governments acting in a self-interested manner may decide to leave certain types of property out of the mix while including other types, or they may tax some types of property very heavily and reduce the burden on other types. Similar arguments can be made with respect to equity and stability. There exists a strong potential for a race to the bottom in terms of what kinds of property to include in the tax base, and a race to undervalue some properties. Centralised policy design could prevent many of these potential abuses, so there is a very legitimate case for centralisation.

Considering the major administrative functions of a property tax, there is also much to be said for centralisation. Administration must include a process to discover taxable properties and to inventory and organise property-specific information, a system to estimate the taxable value of the property, billing and tax collection systems, and an audit process to ensure tax compliance and tax equity (Eckert, 1990). As is the case for policy design, almost all of the administrative functions benefit from the additional resources expected to accompany a centralised approach. Table 3 lists a variety of administrative functions and describes the potential advantages accruing to central government performance.

The case for decentralised administration begins with the observation that property values estimation is accomplished more effectively at the local level. In the case of an ad valorem tax, local appraisers would have a feel for the economic nuances of the local market. Even in cases where technology is used for the appraisal process, accuracy would improve if appraisers were to review results and make modifications according to appropriate subjective criteria. The same logic can be used with respect to discovery and audit, both of which are also advantageously performed at the local level.

It would be of interest to know whether the seemingly omnipresent problem of corruption is more serious under local or central government management in a transitional country. If the property tax were a more significant part of the fiscal mix, it would assume an importance that would more likely tempt corruption than under the current situation in which the property tax is nominal. On balance, opportunities for corruption would seem to be more likely a problem under local management than when the property tax is centrally administered. A form of duplicity that may or may not be less egregious is the problem of local governments offering tax advantages to private entities to influence their location decisions, a part of the `race to the bottom' phenomenon (Musgrave, 1997).

One must recognise that local administration can be expected to fall short in a variety of administrative and technical areas. It does, nevertheless, have the important advantage that the actors involved are informed about local economic, political, and fiscal conditions (Dillinger, 1992), which are especially important for the limited functions of discovery and audit.

Once again, centralisation may render less strident some of the political opposition to the property tax. Finance Ministry officials in the Czech Republic often assert that local government personnel in that country want nothing to do with property tax administration, since centralisation insulates them from political opposition. Coming from the old central planning environment, this is a direct and very visible tax (Bird, 1993). California's Proposition 13 reminds us of the hostility many feel toward this form of taxation even in the United States.

Moral hazard

Arguments for centralisation almost assume away policy or administrative conflicts between central and sub-national governments, viewing the incentives of the different governmental levels as essentially identical. In the real world, of course, conflicting motivations are common. We generally view these as principal-agent conflicts, which gives us a fruitful way of thinking about the issues involved, even if potential limitations do not encourage formal modelling. In tax administration, as elsewhere, agents may well be expected to pursue their own interests rather than those of other interested parties.

Thinking in those terms about property tax administration, who is the agent and who is the principal? Variations on these assumptions have been used fruitfully in the literature. Bale and Dale (1998), Pratt and Zeckhauser (1985), and Moe (1990) describe the principal-agent relationships, for example, between senior bureaucrats and subordinate bureaucrats, and also between citizens, politicians, and governmental levels. For most purposes, we would consider the local government the principal, since it is closer to the citizens and is to some extent their proxy in sharing the desire for sufficient revenues to provide good schools, community infrastructure, etc. The less proximate Czech central government, engaged to collect the tax for the subnational governments, becomes the agent. (2)

The principal (local government) thus receives tax revenue as a result of the agent's (central government's) actions. For its efforts the agent receives none of the revenues collected. In some instances, there will be a collection fee for this service, but in the Czech case there is none. This relationship creates an obvious opportunity for a moral hazard problem to arise. Why would there be moral hazard? It is because the agent understands the opportunity costs (scarce manpower, computer and information systems, and even political capital) required for property tax collection. Nor is the agent prepared to ignore obvious considerations of self-interest.

Subnational governments can only hope their agent will maximise the collection effort. They may well have a sense of the revenue potential of the property tax, but are not in a position to monitor the use of available resources in the central agent's collection effort. Thus, the yield is very likely to correspond to the agent's efforts--it will be less than it could be. This realistically describes the situation in the Czech Republic. Local governments there have not shared property tax data for at least 50 years, and there is no evidence of any local officials monitoring the Czech state administration's tax collection activities.

The fact that the principal will receive less property tax revenue may encourage the agent to offer greater revenue transfers from other taxes or revenue sources, so that the local governments are financially no worse off. Since this holds for the Czech Republic, one cannot view the fiscal situation too dimly. For the time being, because of the current problems of assessment and valuation in property tax administration this outcome is not troubling. In the long run, however, as these problems are overcome, the property tax should become a more significant part of the mix of tax instruments. As we have noted, a pressing concern with the source of revenues available to the municipalities and regions is whether such funds come with mandates on how transferred funds are to be used. In the Czech Republic, the central government does not attempt to control the decision about how transferred property tax revenues are to be used. At the end of the day, however, when transferred funds are exhausted, and all mandated tasks are not fully funded, anyone desiring to satisfy local preferences must still wonder where to find funding. In the Slovak Republic, funding local government is no less problematic, perhaps it is more so, but at least the municipalities may apply whatever efforts they choose to enhance revenue collection. Fortunately, in both republics there is no question about the `ownership' of property tax revenues. They are conceded by their respective central governments to belong to the municipalities. The only issue here is the administrative effort put forth to collect the revenues.

A brief description of the administrative processes in both republics should help to clarify our assumptions with respect to moral hazard. As noted, both countries have identical (area-based) valuation laws and identical rates. They both use self-reported returns supplied by the property owners (Bryson and Cornia, 2001a, b), which presents the obvious need for discovery and audit. It is in these two functions where moral hazard is most likely. In the Czech Republic, the central Ministry of Finance receives the property tax returns, enters the data in an electronic inventory system and remits to local governments the property tax revenue based on the situs of the property owner. `No efforts are made to find under-reported values or non-reported parcels.

In contrast, self-reported returns in Slovakia are submitted to local finance departments, which attempt to verify (audit) reported data and also match reported data with other records (discovery). In both countries, the existing cadastral records were outdated and inaccurate following the socialist period, so discovery and audit would naturally require more effort. In the larger cities of Slovakia, that is, Bratislava and Kosice, local finance offices are further decentralised into municipal districts with substantial fiscal autonomy, thus enhancing the decentralisation effort.

The potential for policy and administrative success and the inherent potential for moral hazard problems are described in Table 4. Clearly, there is a role for central government, especially in the development of policy guidelines. Nevertheless, centralisation of administration creates the potential for incentive incompatibilities and the eventual failure of the property tax. Even attractive, smart policy and administrative designs cannot assure that a tax will succeed in the face of inept incentives (Tanzi, 1991).

Fiscal transfers

An additional potential for moral hazard problems emerges when the central government transfers revenue to local governments. Fiscal redistribution is appropriate when (a) the distribution of local resources is skewed, and (b) when subsidies promote expenditures particularly valued by the centre (Musgrave, 1961). In considering (a), it is obvious that both republics allowed the creation or recreation of many very small local governments that would not be able to function without fiscal assistance from the centre. However, if the redistribution process provides central funds that merely offset revenues the municipalities could reasonably have raised locally, we encounter an additional twist on the moral hazard problems described above. The transferred revenues carry no dangerous political transparencies for local officials. With a more transparent tax like the property tax, local officials would run the risk of opposition to the tax and become the target of any political repercussions accompanying its implementation. This is a particularly significant problem in transition countries where, for the first time, taxes have become visible and direct, and have the potential to impact local elections.

When such transfers occur, the municipalities generally abdicate fiscal power to the central government; they relinquish a certain degree of self-determination for the convenience of revenue provision. They cede to the centre the responsibility and political pressure for revenue collection, and are willing simply to accept whatever funds the centre will provide. Oliveira and Martinez-Vazquez (2001) have also suspected Czech transfers of having `led to negative incentives for revenue mobilisation at the local level' when the central government `was perceived as reducing the level of discretionary transfers at any time that local governments increased their own revenues' (p. 56). The consequence of these actions is to diminish the importance and effect of the property tax, and to diminish the potential for local fiscal autonomy.

The real question is which government, Czech or Slovak, has been most successful in terms of local finance, and why? As we have noted, the Slovak Republic has administered its property tax locally; the Czech Republic has done so centrally. The Czechs have transferred considerable revenue to the local governments, while the Slovaks have not. The Czech Republic, having chosen a centralised approach to property tax administration, transfers substantial amounts of funds from the national to the local governments, but the Slovaks do not.

FINANCING MUNICIPALITIES

Any successes enjoyed in the general process of economic transformation will be reflected in public sector data. The macro situation and economic growth in the economy will impact both national and sub-national budgets. After taking account of price inflation, both the Czech and Slovak Republics have experienced growth permitting expansion of state and local budgets. Budget data are presented in Table 5.

Under the central-planning regime and beginning around 1984, local budgets in the Czechoslovak federation experienced a gradual, long-term reduction in subsidies from the central government. Since 1969, subsidies had constituted nearly 60% of total local receipts. The opening to market economics and democracy did not reverse this decline. Although there was a brief expansion of transfers from 1990 to 1992, the decline continued. After the mid-1990s, subsidies represented no more than approximately 25 percent of the total receipts of Czech and Slovak municipalities (Pekova, 1996).

Table 5 reports the financial situation basically from the end of central planning through the year 2000. We observe from the data that Slovakia's municipal budgets are substantially more modest than those of the Czech Republic. From 1993 on, after the federal neighbours parted ways, Czech municipal budgets were more than twice as large as those of Slovakia, even without taking into consideration the greater value of the Czech crown. The per capita public services expenditures for Czech citizens have been more than three times those of their Slovak counterparts (Bryson and Cornia, 2000). Local budgets in Slovakia comprise only about 13 to 14.5 percent of the total national budget. In the Czech Republic, local budgets range annually from one fourth to just over one-third of the national budget.

A comparison of the grants from the respective central governments to the municipalities reveals the reason for the relative poverty of the Slovak local governments. The much smaller national grants in Slovakia started at 1.5 billion SK in 1993; they ranged from a low of 1.1 billion SK in 1994 to 2.77 in 2000. The grants transferred by the Czech central government to the municipalities ranged from just over 27 billion CK in 1993, to over 40 billion CK in the later years. The 59.5 billion CK in 1996 was a one time aberration. It is instructive that in the years just prior to Slovak independence (1991 and 1992), the government in Prague provided grants of 7.9 billion and 2.4 billion crowns, respectively, for the neighbouring Slovaks. Independence also separated the Slovak Republic from the Czech central budget.

It seems apparent that budget transfers from the centre give the Czech local governments considerably more options than their counterparts in Slovakia enjoy. It is less obvious that, because so much of Czech municipal funding comes from the state, there is much less budget autonomy. By contrast, although national independence resulted in the impoverishment of Slovak municipalities, bringing a hard and tightly binding budget constraint, it also brought a large measure of budget autonomy from the centre. By forcing Slovak municipalities to be more responsible for their own revenues, the Slovak Republic made it essential for the municipalities to increase the property tax yield, which should be a primary source of local public revenues in any case.

In contrast, the Czech Republic has other priorities for the provision of municipal funds. It has not utilised the property tax as an important source of independent revenues. More generous provision of fiscal grants and transfers to Czech municipalities has obviated the need to develop the property tax as a more independent revenue source; at the same time, it has not provided the basis for genuine autonomy for local government.

Table 6 shows the relative importance of the property tax and fiscal transfers. Although the table offers only eight data points, these are worthy of note. They suggest that the property tax has never been an important source of revenue for local government in the Czech Republic. Under the current system there is only modest reason to expect it will ever change. In contrast, the relative importance of the Slovak real estate tax is highlighted by its increasing trend. The Czech property tax as a share of local budgets may even be going in the opposite direction. The trends are less distinct with respect to fiscal transfers from the centre; there has been a decline in the relative importance of transfers from the Czech central government and an increase in those from the Slovak central government. The most notable distinction is the significant difference between the level of the financial transfers to the Czech and Slovak local governments.

Local governments in both republics face increasing revenue needs for current operations, not to mention the backlog of needed investments, which has been building over several years. Up to the present, local governments in both countries have been able to sell off assets in the form of the public housing and commercial facilities transferred to them by central government at the outset of the transition. This has been a significant revenue source, but with the completion of privatisation, it is drying up. This demonstrates the principle that one-time revenue sources should not be used to fund current and ongoing expenditures. Even with this source of funds, there have been ongoing revenue shortfalls in the municipalities. Without additional revenue, the problem can only become more serious. In recent years, the national governments of both republics have been confronted by their own fiscal crises. The response has been to reduce revenues to the municipalities, while transferring additional, centrally managed services to the local governments. The localities are always concerned about unfunded federal mandates, although it is only fair to indicate that the Czechs have been less guilty of that offense than most other transition countries. In the Czech Republic, even when funding is made available for centrally mandated expenditures, state influence over the use of central funds can be assumed, which impairs further development of local autonomy.

Until recently, Slovakia struggled with a serious political liability in the form of the Mechiar coalition, which put the development of self-government at risk by pursuing a process of recentralisation. Given the prevailing legacies from the planning era and the constraints of the Mechiar era, the Slovak Republic has done about as well with fiscal decentralisation as could have been expected. And it was the Mechiar government's withholding of revenues that placed the Slovak municipalities on the path of a serious search for improved property tax revenue yields. In Slovakia, if the property tax continues to develop in a healthy fashion into the new political era, a transfer of additional services to the local governments could be a part of the acquisition of autonomy and local control, both vital elements of decentralisation.

CONCLUSIONS

The Czech and Slovak Republics have openly committed to political decentralisation. Whether or not they have continued to pursue the independence of their municipalities, they began their history beyond the velvet revolution by allowing the emergence or re-emergence of a large number of municipalities. The record on decentralisation is not as strong with respect to fiscal reorganisation. In our view, the attempt to decentralise created opportunities for principal-agent conflicts. In at least one regard, these conflicts have been reduced, because the Slovaks have assigned the responsibility for tax administration to local government. Thus, whether or not as a coherent strategy, the Slovak Republic has adopted some noteworthy policies. In spite of the low levels of transfers from central government to the local governments of Slovakia, there have been some very positive outcomes in that country's recent fiscal history. The assignment of property tax collection to local governments, specifically the discovery and audit functions, has proven beneficial. In an historic irony, the Mechiar era's financial neglect of local governments also had its positive aspect, viz., necessitating the municipalities to pursue their own funds as vigorously as possible. We conclude that the Slovak Republic is on a long-run path that will lead to stronger local government because local governments have at least limited access to an autonomous revenue source and have had to accept responsibility for their own fiscal situation.

Although the Czech Republic has been more generous with its municipalities and may for the foreseeable future have more affluent local governments than in Slovakia, those Czech subnational governments will not likely enjoy the same political independence. Many questions remain regarding the fiscal transformation of the two sister republics. Generalising from the Czech and Slovak experiences, however, the assignment of property tax administration to the local level does in fact lay the foundations of a stronger, more independent fiscal system.

Table 1: Number of sub-national government units selected transition
economies

Country Type of government Number Country population

Albania Municipality 356 3,400,000
Bulgaria Municipality 255 8,900,000
Czech Municipality 6234 10,300,000
Hungary Municipality 3148 10,300,000
Poland Municipality 2459 38,400,000
Romania Municipality 2948 22,700,000
Russia Municipality 2000 149,000,000
Slovakia Municipality 2781 5,300,000
Ukraine Municipality 619 52,100,000

Country Municipalities per 10,000

Albania 1.05
Bulgaria 0.29
Czech 6.05
Hungary 3.06
Poland 0.64
Romania 1.30
Russia 0.13
Slovakia 5.25
Ukraine 0.12

Source: Calculated from Bird, Ebel, and Wallich. See also respective
statistical yearbooks.

Table 2

 Level of policy development

 Central Sub-national
 government government

Level of tax administration
 Central government 1 2
 Sub-national government 3 4

Table 3: Centralisation vs decentralisation of property tax
administrative tasks

Administrative Appropriate Explanation of outcome
function administrative
 level

Property discovery Sub-national Local officials have better
 knowledge of land and
 improvements to land. Better
 knowledge of ownership and use of
 land and improvements.

Valuation skills Central There are few valuation experts
 at either level of government.
 Simple valuation techniques
 minimise the problem. Local
 officials are more easily
 influenced by political pressure
 to keep values low.

Technical support Central Central administration will have
 access to better technical
 support and international
 experts. It will also have better
 access to information technology,
 GIS, and modelling techniques.

Resources Central Central administration will have
 access to more financial
 resources, including resources
 from international agencies.

Scale economies Central To the extent that economies
 exist the central government
 will be in the best position
 to exploit them.

Administrative Central Central government has greater
innovation access to international experts
 and opinion leaders.

Appeal Central Central administration is less
 likely to be influenced by
 politics during appeals.

Audit Sub-national Local units will be more
 motivated to audit performance
 and compliance with the property
 tax.

Table 4: Moral hazard: property tax policy and administration

 Policy Development

 Central government

Administration Complete centralisation benefits from
 Central resources, technical support,
 government and policy experience. There is little
 incentive to improve administrative
 outcomes. The result is a lagging
 effort on the part of central
 administrators. Discovery and audit
 are especially underdeveloped.
 LikeLihood of moral hazard
 problems is high.

 Sub-national Centralisation of policy facilitates
 government development of policy. Local
 governments benefit directly in
 the outcomes of the administrative
 processes. Revenues collected can
 be used by the local government,
 The potential for moral hazard
 problems is low.

 Policy Development

 Sub-national government

Administration Sub-national Governments have substantial
 Central input on policy issues, but there is less
 government chance to monitor the behaviour of the
 administrator of the property tax.
 The potential for moral hazard problems is
 moderate.

 Sub-national Complete decentralisation
 government diminishes the uniformity between taxing
 jurisdictions, creates potential for tax wars,
 and has uneven resources assigned to policy.
 To overcome potential policy problems
 administrators rely on strong administrative
 responses that only exacerbate the problems
 of the tax. Likelihood of moral hazard
 problems is high.

Table 5: National and local budgets: Czech and Slovak Republics

 1993 1994 1995 1996

Slovak Republic
 Local budgets * 19.5 19.0 21.0 21.5
 Local as % of national budget 13.0 13.7 12.9 13.2
 State grants to local budgets * 1.5 1.1 1.2 1.3
 Total local revenues * 21.0 20.1 22.2 22.9
 Local budget expenditures * 19.3 19.1 18.9 21.9
 Real estate tax * 1.6 1.8 1.7 2.1

Czech Republic
 Local budgets * 91.1 111.0 129.1 161.7
 Local as % of national budget 25.5 28.4 29.3 33.5
State grants to local budgets * 27.0 29.3 33.4 59.4
 Total local revenues * 101 111 129 162
 Local budget expenditures * 90.1 112.1 132.3 171.1
 Real estate tax * 3.0 3.8 3.8 4.0

 1997 1998 1999 2000

Slovak Republic
 Local budgets * 23.5 22.7 21.3 24.8
 Local as % of national budget 13.4 12.7 11.9 11.1
 State grants to local budgets * 1.8 2.1 2.2 2.8
 Total local revenues * 26.7 25.9 24.2 27.4
 Local budget expenditures * 25.3 25.8 23.9 26.5
 Real estate tax * 2.6 2.4 2.7 2.9

Czech Republic
 Local budgets * 145.3 157.2 187.7 181.8
 Local as % of national budget 30.3 30.8 30.9 32.6
State grants to local budgets * 35.9 37.4 41.4 46.1
 Total local revenues * 147 162 188 181
 Local budget expenditures * 150.5 158.0 173.0 190.0
 Real estate tax * 3.9 4.1 4.2 4.4

* Billions SK, CK.

Slovak State Grants in 1991: 7.961, 1992: 2385.

Source: Federal Ministry of Finance, Czech and Slovak Republics,
and own calculations.

Table 6: Local property taxes and transfers from central
governments: Czech and Slovak Republics

Year Property tax as % of local budget

 Czech republic Slovak republic

1993 2.99 7.62
1994 3.43 8.91
1995 2.95 7.79
1996 2.48 9.34
1997 2.68 9.78
1998 2.53 9.27
1999 2.26 11.24
2000 2.44 10.47

Year Fiscal transfers as % of local budget

 Czech republic Slovak republic

1993 29.67 7.69
1994 26.35 5.79
1995 25.77 5.71
1996 36.75 6.05
1997 24.68 7.66
1998 23.79 9.16
1999 22.07 10.52
2000 25.33 11.17

Source: Federal Ministry of Finance, Czech and Slovak
Republics, and own calculations.

(1) There is also evidence that the property tax can be underused in developed countries like the United States (Hovey, 1996).

(2) The principal-agent framework is simplified here; since we do not explicitly include the role of property owners, citizens, and those who benefit from public expenditures. In a democracy, the citizen is, of course, the principal, but considering the local governments in most situations to share the citizen's incentives highlights the core issues and significant relationships.

In a federal system, the local government will have trouble monitoring the agent central government, which is problematic because of their conflicting goals. Agents are not always unfettered in their behaviour. In some instances, national and sub-national governments will monitor each other to the extent possible, with both attempting to promote the interests of the principal, the citizen.

REFERENCES

Amsden, A, Kochanowicz, J and Taylor, L. 1994: The market meets its match. Restructuring the economies of eastern Europe. Harvard University Press: Cambridge, MA.

Bahl, R and Linn, J. 1992: Urban public finance in developing countries. Oxford University Press: New York.

Bale, M and Dale, T. 1998: Public sector reform in New Zealand and its relevance to developing countries. The World Bank Research Observer 13(1): 103-121.

Bertaud, A and Renaud, A. 1994: Cities without land markets. World Bank: Washington, DC.

Bird, RM, Ebel, RD and Wallich, CI. (eds). 1995: Decentralization of the socialist state: Intergovernmental finance in transition economies. World Bank: Washington, DC.

Bird, RM and Slack, E. 1991: Financing local governments in OECD countries: The role of local taxes and user charges. In: Jeffrey, O and Panella, G (eds). Local Government: An International Perspective North-Holland: Amsterdam. pp. 83-97.

Bird, RM. 1999: Rethinking tax assignment: The need for better subnational taxes. Draft paper, Fiscal Affairs Department, International Monetary Fund: Washington, DC.

Bird, RM. 1993: Threading the fiscal labyrinth: Some issues in fiscal decentralisation. National Tax Journal, 46(2), 215-216.

Bish, RL and Kirk, RJ. 1974: Economic principles and urban economics. Prentice-Hall, Inc.: Englewood Cliffs, NJ.

Bryson, PJ and Cornia, GC. 2000: Fiscal decentralisation in economic transformation: The Czech and Slovak cases. Europe-Asia Studies 52(3): 507-522.

Bryson, PJ and Cornia, GC. 2001a: Taxes on real property in the Czech Republic. In: Malme, JH and Youngman, JM (eds). The Development of Property Taxation in Economies in Transition: Case Studies from Central and Eastern Europe. The World Bank: Washington, DC. pp. 39-50.

Bryson, PJ and Cornia, GC. 2001b: Land and building taxes in the Republic of Slovakia. In: Malme, JH and Youngman, JM (eds). The Development of Property Taxation in Economies in Transition: Case Studies from Central and Eastern Europe. The World Bank: Washington, DC. pp. 51-66.

Dillinger, WA. 1992: Urban property taxation in developing countries. World Bank: Washington DC.

Dunn, J and Wetzel, D. 2000: Fiscal Decentralization in Former Socialist Economies: Progress and Prospects. World Bank: Washington, DC.

Eckert, JK (ed). 1990: Property appraisal and assessment administration. International Association of Assessing Officers. Chicago, Ill.

Hovey, H. 1996: The property tax in the 21st century. The Finance Project: Working paper Series, Washington, DC.

Hubbard, RG, Giovannini, A and Selmrod, J. (eds). 1993: Studies in international taxation. University of Chicago Press: Chicago.

Institute of Revenues Rating and Valuation. 1997: Principles for local government finance. London.

Jabes, J and Vintar, M. 1996: In: Public Administration in Transition. Proceedings from the Third Annual Conference, Bled, Slovenia, March 23-25, 1995, NISPAcee, University of Ljubljana: Bratislava. pp. 159-168.

Kelly, R. 2000: Designing a property tax reform strategy for sub-Saharan Africa: An analytical framework applied to Kenya. Public Budgeting and Finance 20(3): 36-51.

Litvak, JI, Bird, RM and Ahmad, J. 1998: Rethinking decentralisation in developing countries. World Bank: Washington, DC.

Moe, TM. 1990: Political institutions: The neglected side of the story. Journal of Law, Economics, and Organizations 6(1): 213-253.

Musgrave, RA. 1961: Approaches to a fiscal theory of political federalism. In: Public Finance: Needs Sources and Utilization. National Bureau of Economic Research, Princeton University Press: New York, and Princeton. pp. 97-122.

Musgrave, RA. 1993: Who should tax, where, and what? In: Charles EM Jr (ed). Tax Assignment in Federal Countries. Centre for Research on Federal Financial Relations, Australian National University: Canberra. pp. 2-19.

Musgrave, RA. 1997: Devolution, grants, and fiscal competition. The Journal of Economic Perspectives 11 (4): 65-72.

Netzer, D. 1966, The economics of the property tax. The Brookings Institution: Washington, DC.

Oates, WE. 1996: Taxation in a federal system: The tax-assignment problem. Public Economics Review 1(1): 35-60.

Oates, WE. 1998: Introduction. In: Oates, WE (ed). The Economics of Fiscal Federalism and Local Finance. Edward Elgar Publishing, Inc.: North Hampton, MA. pp. xiii-xix.

Oliveira, JdoC and Martinez-Vazquez, J. 2001: Czech Republic: Intergovernmental fiscal relations in the transition. Washington DC: World Bank.

Paugam, A. 1999: Ad Valorem property taxation and transition economies. ECSIN, Infrastructure Unit Europe and Central Asia Region, World Bank, Working paper No. 9.

Pekova, 3. 1996: Obce a dotace. Obec & Finance 1(1): 28-31.

Popov, V. 2000: Shock therapy versus gradualism: The end of the debate (Explaining the magnitude of transformational recession). Comparative Economic Studies 42(1): 1-57.

Pratt, J and Zeckhauser, R (eds). 1985: Principals and agents: The structure of business. Harvard Business School Press: Boston.

Shah, A. 1994: The reform of intergovernmental fiscal relations in developing and emerging market economies. Washington, DC: The World Bank.

Shleifer, A and Vishny, RW. 1998: The grabbing hand: Government pathologies and their cures. Harvard University Press: Cambridge, MA.

Sirmans, GS, Diskin, BA and Friday, HS. 1995: Vertical inequity in the taxation of real property. National Tax Journal 49(1): 71-84.

Tanzi, V. 1991: Public finance in developing countries. Aldershot, England: Edward Elger Publishing Limited.

Tiebout, CM. 1956: A pure theory of local expenditures. Journal of Political Economy 54(5): 416-424.

Youngman, JM and Malme, JH. 1994: An international survey of taxes on land and buildings. Kluwer Law and Taxation Publishers. Deventer, Netherlands.

World Bank, From plan to market. World development report, 1996. Oxford University Press: Oxford.

PHILLIP J BRYSON (1) & GARY C CORNIA (2)

(1) 616 TNRB, Marriott School of Management, Brigham Young University, Provo, UT, 84602, USA. E-mail: phil_bryson@byu.edu;

(2) Romney Institute of Public Management, Marriott School Brigham Young University, Provo, UT, USA. E-mail: gary_cornia@byu.edu