3. Policy recommendations
Chapter 3 offers guidance on how to improve and reform the current system of local government finance in Israel. It includes a set of 13 policy recommendations designed to improve the efficiency and equity of the system. The first set of eight policy recommendations are designed to improve the existing Arnona system. Priority is given to changes that are actionable and could be adopted and implemented relatively quickly. The final five policy recommendations represent a blueprint for a more substantial reform of the system of local government finance in Israel. The centre piece of this reform is the establishment of a value-based system of local property taxation applied to both residential and non-residential property.
On the basis of the review and analysis of the system of local government finance in Israel presented in the first two chapters of this report, chapter 3 suggests a number of policy recommendations. On the whole, the recommendations are motivated by a desire to improve local government finance in Israel along a number of important dimensions:
Increasing economic efficiency by reducing distortions created by the Arnona system.
Increasing the horizontal and vertical equity of the Arnona both within and among local governments.
Simplifying and increasing the transparency of the system of local government finance.
Enhancing the effectiveness of the public services provided by local governments.
Providing local governments with increased autonomy over their fiscal affairs while also making progress on national goals of improving education, reducing poverty, increasing social welfare, and enhancing environmental quality throughout Israel.
The 13 OECD policy recommendations that follow are divided into two distinct sections. The first eight policy recommendations are designed to improve the existing Arnona system. Priority is given to changes in the Arnona system that are actionable and could be adopted and implemented relatively quickly. Once they are implemented, they should begin to produce results in terms of the criteria of an effective system of local public finance.
The final five policy recommendations represent a blueprint for reforming more radically the system of local government finance in Israel. The centre piece of this reform is the establishment of a value-based system of local property taxation applied to both residential and non-residential property.
Our first policy recommendation calls for the reduction of non-residential Arnona rates as a way of reducing the existing disparity between residential and non-residential rates. As the reduction of Arnona rates will reduce the revenue available to local governments, the second and third policy recommendations address the sources of replacement revenues. The second recommendation suggests new sources of local government own-source revenue and the third recommendation proposes increases in central government grant to finance delegated services. The next five recommendations address specific changes to the Arnona that address explicitly some of the shortcomings of the current system.
Policy recommendation 1: Reduce the large disparity between non-residential and residential Arnona rates by reducing non-residential rates relative residential rates
As illustrated by Figure 1.17, compared to most OECD countries, Israel relies very heavily on the property tax to finance its local governments. In addition, as shown in Table 1.4, average non-residential Arnona rate are in many cases two or three times higher than average Arnona rates on residential property. While it is appropriate that the Arnona play a central role in local government finance, its high rates, especially on non-residential property, distort economic development decisions and exacerbate problems of horizontal and vertical inequity. Although hard data are not available, it appears that the Arnona revenue raised on non-residential property is substantially in excess of the cost of public services delivered to those properties. The potential to collect a large fiscal surplus from non-residential property appears to be influencing the economic development strategies of local governments. Some governments are constructing large commercial-industrial parks and designating large portions of available land for exclusive non-residential uses. In cases where available land for commercial-industrial development exceeds potential demand, land may remain vacant for long periods. The residential-non-residential rate differential may also serve to discourage some local governments from developing land for new residential development.
The government should reduce the ratio of non-residential rates to residential rates by mandating reductions in non-residential Arnona rates. As the reduction of Arnona rates will reduce the tax revenue available to local governments, this policy recommendation is only feasible if it is combined with policies to replace the lost Arnona revenues. Policy recommendation 2 and 3 propose alternative sources of own revenue and increases in government grants to local governments as replacements for reduced Arnona revenues from non-residential property.
Policy recommendation 2: Develop alternative sources of local government revenue, through taxation, user charges and income from assets
According to OECD Revenue Statistics, more than 80% of local government tax revenue in Israel comes from the property tax, a percentage that is nearly twice the OECD average (Figure 1.16). One way to help finance a reduction in non-residential Arnona rates would be for local governments (with permission from the central government) to enact other local taxes to diversify their portfolio of taxes. Although they are unlikely to generate a substantial amount of revenue in most jurisdictions, it is appropriate for local governments to levy tourist taxes, such as taxes on hotel accommodations or short-term apartment rentals, e.g. Airbnb, or taxes on restaurant meals (Box 3.1).
Tourism taxes, fees and charges, often referred to as occupancy, bed night, room, accommodation or transient visitor taxes, are quite common in the OECD and European Union. Tourism taxes focus on tourism or accommodation providers (hotel, B&B, campsites, vacation rentals, etc.). They are typically levied on short-term stays in paid accommodation.
According to OECD Tourism Trends and Polices (2014[1]), there has been a general increase in the number and scope of tourism-related taxes, fees, charges over the last 10 to 15 years. In 2017, tourism taxes were levied by more than two-thirds of EU Member States, in about one-half of the states in the United States, Canada, Mexico and Switzerland. European Union and OECD countries not levying a tourism tax include the UK, Ireland, Denmark, Sweden, Finland, Estonia, Latvia, Luxembourg, Iceland, Chile and Israel.
Greece introduced an Overnight Stay Tax in January 2018 to help reduce public debt. Japan introduced a "sayonara tax," in January 2019 paid by international visitors as they leave the country, which will be used to enhance Japan's tourism infrastructure before the 2020 Summer Olympics in Tokyo. Some big Japanese cites (Tokyo, Kyoto and Osaka) had already set up lodging taxes. New Zealand introduced an International Visitor Conservation and Tourism Levy (IVL) on 1 July. Collected through visa fees and via the new Electronic Travel Authority, the IVL will support conservation and the development of tourism infrastructure. Scotland has recently introduced legislation allowing municipal governments to levy a tourist tax – this follows on from an extensive consultation exercise in late 2018/early 2019. Additional consultation is underway and the legislation is not expected until 2020-21.
The overwhelming majority of taxes specific to hotels and accommodation facilities are administered at the subnational and primarily the municipal level, with only few countries identifying examples at the national level such as Greece, Malta. Norway decided against a national tourism tax in 2017 but has allowed municipalities to introduce local schemes should they wish. The tourist tax receipts can be also shared between different bodies including cities, counties, the national tourist board, and the Red Cross (Croatia) or regional and municipal tourist bodies (Spain).
In general, taxes may be levied on a number of bases, such as “per person, per night”, “per room, per night”, or as a proportion of the room cost. Tax rates applied also vary widely. They can be set at national level but rates may vary from one city to another, when municipal discretion over the rates is applied. There can be differentiated rates depending on the location (for example between historical city centre vs suburbs e.g. Amsterdam), quality ratings and cost of accommodation, the period of the year (high versus low seasons), or the payer (foreigners/external versus local residents e.g. in Amsterdam); exemptions or reduced rates might also exist for young people or business travellers (e.g. Berlin and Hamburg).
Tax collection mechanisms also vary across countries. Tourism taxes may be included in the price paid in advance for accommodation, or may be payable in person; they may be paid by the accommodation owner directly, or through intermediary platforms like AirBnB.
The use of tourist tax revenues varies from one country to another, or even from one city to another. They can be dedicated or hypothecated to support the tourism sector (local tourist boards, tourism promotion activities, construction and maintenance of tourism infrastructure and amenities, workforce training in the tourism sector), to finance environmental protection and conservation (e.g. the Great Barrier Reef Marine Park Environmental Management Charge [EMC]), historical and cultural preservation and restoration, and improve urban infrastructure, including public transportation.
Although generally small (tourism taxes typically represent a small proportion of the overall cost of accommodation), tourism taxes can play an important role in complementing local government revenues. In addition, tourists have typically a greater willingness to pay than residents (because of preferences and higher income), and taxation can price in any negative externalities. In countries where tourism price competitiveness is already very high, the impact is, in general, negligible on foreigners.
Sources: European Commission (2017[1]), The Impact of Taxes on the Competitiveness of European Tourism; Government of Scotland/UK (2018[2]), Transient Visitor Taxes in Scotland: Supporting a National Discussion - A Scottish Government Discussion Document; Ollivaud, P. and P. Haxton (2018[3]), “Making the most of tourism in Indonesia to promote sustainable regional development”, Working Paper of the Economic Department; OECD (2014[4]), “Taxation and tourism”, https://doi.org/10.1787/tour-2014-6-en; The European Tourism Association (ETOA).
The government could also consider giving local governments the possibility to raise taxes that are related to the provision of local public services. In OECD countries, most common local own-source taxes are taxes related to the provision of local public services such as waste collection tax, cleaning tax, street lighting tax, etc.
Taxes related to local vehicular traffic, such as parking taxes or taxes on ride-sharing service, would also be appropriate. These taxes have a strong link with local transportation infrastructure provided by local governments, in the maintenance of local roads. Therefore, if the revenue from these taxes is used to finance roads and other transportation services, it can match payment for services to the benefit from the service.
Local governments should also consider implementing license taxes related to specific local activities, such as advertising, gambling, entertainment, personal services, markets, or real estate transactions. This type of tax is used by local governments in a number of OECD countries. Tax on advertising in particular is widely used at the local level both in developed and developing countries. Although expected revenues may be limited, such tax may be seen as desirable with many respects. First, it is easy to implement locally. It is unlikely subject to local tax competition as the location of advertisements within the local space is mainly linked with commercial requirements. It is poorly subject to business cycle. It is also directly linked to the control of local public (and private) space which is mainly under the responsibility of local governments. As a result, the setting of the tax rate could be left to local governments without large efficiency costs. In the same vein, some countries (France for example) levy taxes or tariffs on local facilities of network industries (power stations, dams, windmills telecom antennas and repartitors....). These taxes /tariffs are often set by the central government and the proceeds are shared with local governments (it is not a piggy-tax). Finally, a local surcharge on the tax on real estate transactions could be also considered: while such tax may provide additional revenues to municipalities (despite its sensitivity to business cycles), it could be seen as a way to introduce a tax based on (real) property values at the local level, and therefore represent a step further in the perspective of the Arnona reform.
Figure 1.12 shows that compared with subnational government in OECD countries, local governments in Israel raise very little revenue from user charges: user charges and fees accounted for 4.5% of local government revenue in 2016 well below the OECD average of 14.9% in 2016. Although the potential revenue that could be raised from user charges and fees may be limited, the government should explore the revenue potential of establishing new fees, related to the provision of local public services, or consider raising rates on existing fees. Although provisions should be made for residents with low incomes, taxes or user fees on specific municipal services such as waste collection, the use of recreational facilities, or street lighting, may be appropriate.
Finally, local governments in Israel derived 1.3% of their total revenue from their physical and financial assets. These revenues come mainly from the rent and sales of local assets (land and buildings) as well as from dividends paid by water companies. It is less than in the OECD where on average, subnational governments raised 2% of their revenues from these sources in 2016. Although many municipalities in Israel have relatively few assets, at least some local governments could explore the possibility of raising more revenue related to their ownership of property assets. There are major obstacles to this development related to the fact that many municipalities in Israel have relatively few own assets, which limits the potential of this option.
Policy recommendation 3: Reform the central government fiscal transfers in support of the major delegated functions-education and social welfare
If, as proposed in recommendation 1, non-residential Arnona rates are reduced, the grants for the delegated functions of education and social welfare would need to be increased to compensate for the lost Arnona revenues, some of which is used by local governments to finance public education and social welfare activities carried out by local governments.
Without increases in ministry grants, spending per student and per social welfare client would fall substantially in many local governments. If alternative sources of local government revenue are implemented (see recommendation 2), then the portion of the ministry grants for education and social welfare that were intended to replace Arnona revenues (as recommended above) could be reduced.
The matching requirements attached to the grants for education and social services should be reconsidered to better reflect wealth differentials across municipalities. Israel could establish variable matching rates based on municipal tax capacity instead of the current “fixed rate” system.
The analysis presented in this report support the conclusion of the OECD’s (2018[5]) Economic Survey of Israel that “disadvantaged schools need much greater funding”. More generally, it suggests that the current allocation of grants do not adequately compensate for the unequal distribution of Arnona revenue. While the ministry grants for delegated functions have succeeded in reducing inequalities among local governments, there remain substantial differences across local authorities in their capacity to provide the public services for which they are responsible. With this in mind, it is important that the allocation of education and social welfare grants be reformed so that a larger portion of the total grant amount goes towards municipalities in the lowest socio-economic clusters.
Larger ministry grants for education and social welfare will require additional government revenues. These revenues will need to come from reductions in other government programs, from raising rates on existing taxes, or for the establishment of new taxes. It is well beyond the scope of this report to recommend a specific finance mechanism. It is however worth pointing out that businesses will be the primary beneficiaries of the reduction in non-residential Arnona rates. It may thus be appropriate to consider financing larger government grants by implementing increases in businesses taxes.
Policy recommendation 4: Standardise the classification of types of property across the country and reduce the number of sub-classification categories
Arnona rates vary by the type (and use) of property. While the government has defined 13 major “classes” of property, each local authority can (with approval of Ministries of Finance and Interior) establish its own set of subclasses within the major classes established by the central government. There are currently thousands of different sub-classes, with large differences in classification systems across local governments. The myriad of subclasses contributes to economic inefficiency, and to horizontal and vertical inequities in the Arnona. The existence of these subclasses of property increases the complexity of the Arnona system and reduces its transparency. The government should take steps to reduce the number of subclasses and standardize the system of classification across all local governments. Although a standardized classification system would have advantages in terms of increased efficiency and equity, and lower administrative costs, it would come the cost of further reducing the fiscal autonomy of local governments.
In effect, allowing local governments to have control over sub-classification was a backdoor way of providing them some additional control over local rates. It would be preferable to standardize the classification system but give local governments some additional latitude in setting their own Arnona rates.
Policy recommendation 5: The central government should establish and enforce a uniform system for measuring taxable area for purposes of implementing the Arnona (pending more comprehensive reform –see recommendations 9 and 10)
The Arnona is an area-based system, with tax liabilities depending on the square meters of property used. Local governments however have been free to determine their own method for calculating areas. Some use internal dimensions, while others use external dimensions. There is also no consistent treatment of various real estate features such as common areas, parking area, and porches. Evidence suggests that different systems for calculating area can result in differences of up to 30% in calculated area.
It is difficult to find an economic justification for allowing local discretion in the methodology used to measure the size of the tax base. A national uniform method should be adopted by the central government. To enhance the transparency of the Arnona system, the government should also consider making publicly available the size (in square meters) of each property (possibly on the website of the Ministry of Interior). In some countries using value-based property tax systems, the assessed value of each property is public information, often available on the websites of local jurisdictions.
While the standardization of the method used to calculate taxable area might be seen as another reduction in local fiscal autonomy, as argued in policy recommendation 4, local governments can be compensated by allowing them more discretion in the setting of Arnona tax rates.
Policy recommendation 6: Addressing the problems created by the current system of Arnona exemptions and discounts
Exemptions provided to residential Arnona taxpayers
Because it is extremely difficult to make an area-based property tax progressive, there is a broad consensus in Israel that the residential Arnona is regressive. In an attempt to reduce the burden of the property tax on certain categories of individuals, such as the elderly, persons with mental or physical disabilities, students, soldiers performing compulsory national service, and families with low incomes, the Ministry of the Interior establishes criteria for eligibility for Arnona discounts. Although local governments retain some flexibility in determining the percentage level of discounts, they are required to provide Arnona discounts to eligible taxpayers.
While the granting of exemptions, reliefs and discounts to needy individuals reduces economic hardships created by the Arnona and improve the vertical equity of the tax, it also has unintended serious side-effects.
First, by reducing Arnona revenue from discount-eligible households, local governments have an incentive to discourage potential new residents who may be eligible for Arnona discounts. This discouragement may take the form of restrictive housing or land use policies. The granting of discounts increases the probability that the costs of providing services to these new residents will exceed the residential Arnona revenue that they would generate. Second, by reducing the Arnona revenue associated with residents, the current discount policy may also encourage local governments to aggressively target non-residential development. Third, the current system of local government financed discounts also increases fiscal disparities among local governments and places especially heavy fiscal burdens on economically-disadvantaged communities that have high concentrations of residents who are eligible for Arnona discounts. Finally, it results in a reduction of the property tax base and a net loss of revenues for local governments equal to approximately one quarter of the total revenue from the residential Arnona. The full cost of the discounts decided by the State (central government or Knesset) is borne by local governments in the form of reduced Arnona revenues.
In particular, the current system of exemptions creates substantial fiscal problems for some local jurisdictions, especially those with low levels of Arnona revenue and high numbers of discount-eligible taxpayers. There is little question that the requirement that local governments must self-finance government mandated discounts and exemptions contributes to the fiscal disparities that exist among local governments in Israel. The data in Table 1.7 show that discounts reduce Arnona revenues by an average of 36% in local governments in the five lowest socio-economic clusters, but by an average of only 15% in local governments in the top five clusters.
There are different options to address the fiscal problems of local governments created by the current system of discounts and exemptions. In order to define the most appropriate option to reform the current system, and thus reduce fiscal disparities and secure fiscal balances at both national and local levels, it is recommended to carry out further in-depth research, impact assessment analysis as well as to conduct consultation with key national and local stakeholders.
In the short run, it is recommended to better standardise Arnona discounts and exemptions and establish strict eligibility criteria for the receipt of Arnona discounts. In addition, in order to promote more consistency within the system of local government financing and coordination across the central government concerning local fiscal affairs, it is recommended that the regulation of the system of Arnona discounts and exemptions becomes the joint responsibility of the Ministry of Interior and the Ministry of Finance, as is the case in other matters related to local government budgets. Clause 12(b) of the Arrangement Law would need to be revised to require the approval of both ministries. The involvement of other ministries in the discussion of reforming the system of Arnona discounts could also be envisaged. Finally, a consultation platform or forum could be established with representatives of the different associations of local governments (Federation of Local Authorities, Federation of Regional Councils, etc.) to foster vertical coordination across levels of government on local finance matters.
Exemptions and rebates provided to non-residential Arnona taxpayers
There are numerous exemptions and rebates provided to non-residential Arnona taxpayers. They raise no equity issues, but they substantially reduce the tax base. By reducing Arnona revenue they put fiscal pressure on local government budgets, and due to the unequal spatial distribution of exempt property, they increase fiscal disparities among local authorities. The Arnona equalisation fund established in 2017 should hopefully reduce these disparities in the future. Any exemption granted to non-residential property in effect provides a subsidy to the users of that property. The exemptions are often justified on the grounds that the use of an exempt property, perhaps for a government office, by a hospital, or a house of worship, provides a benefit to the community that more than compensates for both the lost Arnona revenue and the cost of public services that benefit that the exempt property. These issues have not been systematically studied. A full review of the benefits and costs of exemptions and rebates granted to non-residential property should be undertaken by the government.
Policy recommendation 7: Allow some degree of rate setting by local governments. Extra revenues could be used to increase spending on delegated or discretionary functions
Local governments in Israel have very little control over local tax rates. 95% of the tax revenue raised by local governments in Israel is collected using rates set by the central government and any changes to these rates must be approved by the Ministries of Finance and Interior. The degree of local tax autonomy is de facto quite limited. It is the view of the OECD that local tax autonomy should be increased, by allowing local governments some degree of freedom to set their own Arnona rates. As a general guideline, subnational governments need own-source revenues, on which they have some leeway over tax rates and bases because this contributes to accountability and efficiency of local public service provision, as underlined in one of the 10 guidelines of the recent OECD Handbook for Policy makers to make decentralisation work (OECD, 2019[6]). Tax autonomy encourages local residents to evaluate costs and benefits of local service provision and benchmark local government performance against itself as well as with neighbouring jurisdictions. This also facilitates yardstick competition which encourages local politicians to maximize the welfare of local residents instead of promoting their own self-interested goals.
As suggested in policy recommendations 4, 5, and 6, increased local government discretion in setting rates could be combined with various structural reforms of the Arnona system that would somewhat reduce various aspects of local autonomy that now exist.
Several options could be pursued in granting local governments more tax rate autonomy. The first option would be to set bands with a minimum and maximum allowable rate for both residential and non-residential property, and let local governments set their rates anywhere between the minimum and maximum rates. It is possible, perhaps probable, that with this type of system most local governments will set their residential rates close to the minimum of the residential band, and their non-residential rate close to the maximum of the non-residential band.
A second option, consistent with our recommendation 1, involves the linking of residential and non-residential rates. This kind of linked system has long been used in a number of OECD countries including Germany and France. The general idea is to constrain local governments’ ability to change the balance between residential and non-residential rates by linking changes in one set of rates to changes in the other. A third option sets an additional constraint, requiring that local rates aren’t too far above or too far below the average tax rate set by either a subset of local governments or all local governments. This kind of system can be complicated. Box 3.2 provides a brief description of the French system of linking residential and business tax rates.
This type of system is complex, but it can also be made rather flexible. For example, exemptions to the linkage rules could be introduced in order to incentivize local governments to merge, or to reward “strong” municipalities, which operate efficiently (see policy recommendation 12).
A set of rules aims to prevent local assemblies from approving substantially different tax rates for their residents and for their businesses, and from setting rates that are too different from the average tax rate of neighbouring local governments. These tax rate linkages apply to three local taxes on residents or property owners – the taxe d’habitation, which is a resident tax levied on all residents, whether they are property owners or not, the property tax on undeveloped land (taxe foncière sur les propriétés non bâties), the property tax on developed land (taxe sur les propriétés bâties), and to the property tax on businesses (Cotisation foncière des entreprises).
The rates of each of the three residential taxes cannot exceed 2.5 times the corresponding average rates for the previous year in the municipalities of the département or if the latter is higher, the average rate in all French municipalities in the previous year. Also, the rate of the Cotisation Foncière des entreprises cannot exceed 2 times the average rate in all the municipalities in the previous year.
A second set of rules aims at assuring that property tax changes faced by residents or property owners are proportionate to property tax changes on businesses. In the simplest case, the local assembly chooses to change each of the four rates by the same percentage relative to the previous year (proportional variation). If the local assembly chooses to change the rates in a non-proportional way, two additional constraints are set. First, the rate of change of the property tax on undeveloped land must increase or decrease less than the rate of change on residential property. Second, the property tax rate on businesses cannot increase faster (or decrease less) than the tax rate change on residents, (or if the later, the weighted rate of the taxes on residents and on developed and undeveloped property). In addition, special constraints are put upon the co-variation of rates for municipalities within communities.
Source: French Ministry of Interior (2018[7]), Guide pratique sur le vote des taux des impôts directs locaux et taxes assimilées, https://www.collectivites-locales.gouv.fr/files/files/finances_locales/fiscalite_locale/fiches_fdl/guide_votedestaux.pdf
Policy recommendation 8: The government should actively assist local governments in increasing their Arnona collection rate by providing training, technical assistance, and capital grants for the modernisation of computer systems
Raising Arnona collection rates is a key challenge for many local governments. The statistics (see Table 1.7) show the seriousness of the problem. In 2016, the collection rate in the bottom five socio-economic clusters was below 75%, and in the bottom two clusters below 50%. The exact reasons for the low collection rates are not known to the OECD. In fact, it appears that the government has almost no data on the administrative and collections costs associated with the Arnona. It is possible that collection rates are high in some local governments because those local governments have the resources to devote to tax administration, while other jurisdictions lack both the expertise and the capital infrastructure needed to effectively and efficiently collect the Arnona.
It would be useful to know whether there exist substantial scale economies in the administration and collection of the Arnona, whether regional collaboration in Arnona administration may be cost effective, and whether the use of private collection companies reduces collection costs. Knowledge of the relationship between collection rates – measured as Arnona revenue as a percentage of net Arnona charges – and spending on tax administration would also help in the design of policies to improve collection rates.
The government should actively assist local governments in increasing their Arnona collection rate by providing training, technical assistance, and capital grants for the modernization of computer systems. The Israel Tax office could enter into voluntary agreements with municipalities with low collection rates. Under those agreements, the Tax Office could directly assist local governments in their efforts to raise Arnona collection rates.
The next two recommendations propose that the Arnona be reformed by converting it from an area-based to a market value-based property tax system. The reform process should start with the non-residential Arnona (recommendation 9) and then be followed by the implementation of a value-based residential Arnona (recommendation 10). Recommendation 11 calls for the expansion and reform of the Balance Grant. The explicit purpose of the Balance Grant is to reduce the fiscal disparities that exist among local governments in Israel. The current allocation formulas can be reformed so as to better target central government resources to the most disadvantaged local governments. The final two recommendations suggest steps to enhance cooperation among local governments and propose research on the quality of public service delivery by local governments.
Policy recommendation 9: Establish a value-based system of property taxation for all non-residential property
As detailed in Chapter 2 of this report, there are a number of advantages to a value-based property tax system. Note that Box 2.2 (Chapter 2) highlights a set of guiding principles for property tax reform, which were developed by the OECD. These principles include a statement that whenever possible property taxes should be based on the market value of property.
For reasons outlined below, it will be relatively easy to establish a value-based non-residential Arnona. Hence, the OECD recommends that the government implement a value-based non-residential Arnona prior to the conversion of the residential Arnona. During the transition to a value-based property tax system, the government should split the residential and the non-residential Arnona into two separate taxes.
A value-based non-residential property tax system is not a radical reform. Until the 1960’s, Israel utilised a value-based system of property taxation. Properties values are also part of the present Israeli tax system. Both the local Improvement Levy, and the central government’s Betterment tax are based on changes in the market value of property. Moreover, two recent academic studies present evidence of direct or indirect links between Arnona tax collections and property values (Horne and Felsenstein, 2010[8]; Portnov, McCluskey and Deddis, 2001[9]). Comprehensive market value data on non-residential property is not currently available in Israel. However, a great deal of data needed to calculate market values are already available in the form of information that businesses must use to comply with existing taxes, such as the corporation business tax and transactions taxes. Information on construction costs of non-residential properties and on the rental value of property can all be used to help developing estimates of market values. The fact that the number of non-residential parcels is so much smaller than the number of residential parcels, will make the implementation of a value-based non-residential property tax much easier to complete. Also, special discounts for “needy” taxpayers are not needed for the taxation of non-residential property, although exemptions for very small businesses might be considered.
During transition period from an area-based to a value-based tax system, large changes in tax liability would certainly need to be phased in over a number of years. In the transition to a value-based business property tax system, particular attention must be paid to the interactions of the new tax with the rest of the tax system. Property taxation must be aligned with other forms of capital taxation (including the betterment tax) and with user fees. Coordination with the income tax system is important, especially if it includes the taxation of (imputed or actual) rents.
Finally, any tax reform may have important implication for fiscal disparities across local governments. Reform should always be accompanied with appropriate changes to existing government grant programs so as to prevent unintended increases in fiscal disparities.
Policy recommendation 10: Establish a value-based system of residential property taxation
Although the Arnona, as an area-based property tax system, has many of the positive attributes of any property tax system, it compares poorly to a value-based system in terms of both vertical and horizontal equity. Area, even when adjusted by other factors, is a quite imperfect proxy for value, and provides a base that is very imperfectly linked to the ability-to-pay principle. A value-based system with a single-residential rate applied in each local government will result in considerably more transparency about the operation of the local property tax, especially if the assessed values of each property are publicly available.
The most common criticism of a value-based system compared to an area-based system is that it is much more complex, and costly to administer. While that was undoubtedly true in the past, recently developed techniques for property assessment combined with new technologies have greatly reduced the costs involved in determining the value of residential property (see Boxes 2.3 and 2.4). Property assessors argue that the two most important factors in determining the value of property are location and size. As long as both of these are known, it is not terribly difficult or expensive to develop a reasonably accurate assessment valuation model.
That being said, the establishment of a value-based residential property tax system will involve a considerable investment in the training of property tax assessors and an investment in the technology that is the foundation of a modern assessment office. The process will undoubtedly be considerably more complex, time consuming, and costly than establishing a value-based system for non-residential property taxation (policy recommendation 9). For this reason, the government should only begin implementing a value-based property tax system for residential property after it has successfully converted to a value-based system for non-residential property. In the short-run, the government could initiate a study of contemporary property assessment techniques and analyse which methodology and technology will best fit the Israeli situation. The government should set out a strategy to move to a value-based system overtime.
Policy recommendation 11: Consider increasing the balance grant and simplifying the allocation formula
The design of the balance grant formula is consistent with the fundamentals of revenue equalisation. According to OECD guidelines, equalisation mechanisms, if properly designed, enhance fiscal equity (citizens are treated equally by the public sector regardless of their place of residence) and fiscal efficiency (by discouraging fiscally induced migration - mobility of factors in response to differential net fiscal benefits across the nation). Care therefore must be exercised in the design of such program that they do not inadvertently contribute perverse incentives that may lead to a misallocation of resources and thereby retarding the growth of the regional and national economies (OECD, 2019[6]).
As shown in Tables 1.10 and 2.6, balance grants contribute to lowering the fiscal disparities among local authorities. Their ability to reduce the existing fiscal disparities among local governments (see Table 2.5) could be improved by increasing the overall amount of the grant and by simplifying the highly complex allocation formula. Improvements in the formula could be made on both the tax capacity side (“normative revenue”) and on the fiscal needs side (“normative expense”). On the revenue side, disentangling the tax capacity component, which aims at lowering the disparities in potential tax revenue, and the incentives given to local authorities to collect revenue might be useful. Reforming the Arnona and allowing local governments to use alternative sources of local tax revenue will also require changes in the definition of normative revenues. On the expenditure side, it would be helpful to make an explicit delineation between the normative fiscal needs deriving from the government standards, and additional expenditures based on local preferences. Formula improvements which simplify the allocation formula would be very helpful in enhancing the transparency of the government’s transfer system.
Policy recommendation 12: Enhance cooperation among local governments by providing incentives for the merger of small neighbouring local governments, the joint provision of some public services, or the sharing of tax bases
While many policy initiatives for improving Israel’s system of local government public finance involve reforming taxes and other revenue instruments, and instituting reforms of the system of intergovernmental finance, improvements can also come through the reorganisation of local governments or through the development of mechanisms to enhance cooperation among local governments.
Although the merging of local governments is political challenging, there are circumstances, such as a merger of two small adjacent jurisdictions, where mergers would allow for the exploitations of scale economies. In these cases, the government should take steps to encourage and incentivise these mergers.
More generally, the government should continue its efforts to encourage cooperation among neighbouring local governments. Continued support of “regional clusters” (described in Chapter 1) should be continued. Assigning responsibility for the delivery of specific services directly to the clusters should be encouraged. Cross jurisdiction cooperation on specific projects and the sharing of technical expertise among local government officials may both be relatively low-cost ways of improving the delivery of public services.
In many OECD countries, large metropolitan areas are composed of a central city surrounded by a large number of independent local governments. The Tel Aviv metropolitan area (Gush Dan) is one example. All too frequently, local governments compete with each other to attract businesses and high-income residents, while attempting to exclude residents and land uses that are considered undesirable. The actions of individual governments, while perhaps serving the short-run interest of their residents, often have significant negative consequences for the metropolitan area as a whole, leading to patterns of inefficient development and growing fiscal disparities. Examples of policies to address these issues come from the U.S. and France. The U.S. experience (described in Box 3.3) focuses on the sharing of non-residential tax base among all local governments within a single metropolitan area.
The French experience involves the creation of interjurisdictional cooperative bodies, (described in Box 1.1) that have proved to be successful. The French experience with regional cooperation is particularly relevant to Israel because both countries are unitary countries sharing a solid tradition of centralization; metropolitan areas in both countries contain a large number of local jurisdictions; and local governments in both countries rely heavily on a local business tax (the non-resident Arnona in Israel). Also relevant is the fact that the French reformed their “taxe professionnelle” (the local business tax) over a ten-years period employing both a territorial reform which incentivizes a taxe professionnelle-sharing within new forms of inter-municipal bodies (“Communautés à fiscalité unique”), and a robust reform of local taxation (ten years later), which included the removal of the taxe professionnelle and the establishment of new forms of local taxation on businesses.
The seven-county metropolitan area that includes the Twin Cities of Minneapolis and St. Paul contains 182 general-purpose local governments. In the late 1960s concerns developed that competition among local governments for non-residential tax base was resulting in growing fiscal disparities within the metropolitan area, and that this competition was leading to inefficient and environmentally harmful patterns of regional economic growth. In 1971, the Minnesota Legislature responded to these concerns by establishing a metropolitan area tax base sharing program, formally known as the Fiscal Disparities program.
Under the legislation, all local governments in the metropolitan area must contribute 40% of the growth in commercial, industrial, and public utility property tax base since 1971 into an area-wide shared pool. By 2016, the shared pool of area-wide tax base was worth $373 million and was equal to one third of the total non-residential tax base. All general-purpose local governments in the metropolitan area share in the distribution of revenue from the shared tax base. The property tax revenue raised each year by taxing the shared base at an area-wide rate is allocated among local governments using a formula that distributes per capita amounts in a manner inversely related to the size of each local government’s per capita property tax base. The property tax base is defined as the market value of both residential and non-residential property. Thus, local governments with a below-average per capita tax base, receive an above-average allocation from the shared base. In 2016, 99 out of 182 local governments in the Twin Cities metropolitan area where net recipients from the shared tax base, meaning that their per capita distribution from the tax base exceeded the per capita amount they contributed to the shared base because of growth in their non-residential tax base.
Data indicate that since the program went into effect in 1975, it has reduced fiscal disparities within the metropolitan area. The evidence on whether the tax-base sharing program has contributed to a more efficient pattern of metropolitan are economic growth is somewhat less clear. For an early economic assessment of the program see Reschovsky (1980[10]).
Policy recommendation 13: Support research to measure the quality of public service provision of local governments
Local governments play a key role in Israeli society. They provide Israeli residents with a wide range of public services. It appears, however, that relatively little research has been conducted on the efficiency and effectiveness of local governments. In order to assess whether public resources are being spent wisely, it is important to assess the quality and quantity of public service delivery. The government should encourage and support a robust research agenda that focuses on assessing public service quality and evaluating the performance of local governments in delivery essential public services.
References
[1] European Commission (2017), The Impact of Taxes on the Competitiveness of European Tourism.
[7] French Ministry of Interior (2018), Guide pratique sur le vote des taux des impôts directs locaux et taxes assimilées, https://www.collectivites-locales.gouv.fr/files/files/finances_locales/fiscalite_locale/fiches_fdl/guide_votedestaux.pdf.
[2] Government of Scotland/UK (2018), Transient Visitor Taxes in Scotland: Supporting a National Discussion - A Scottish Government Discussion Document.
[8] Horne, R. and D. Felsenstein (2010), “Is property assessment really essential for taxation? Evaluating the performance of an ‘Alternative Assessment’ method”, Land Use Policy, Vol. 27/4, pp. 1181-1189, https://doi.org/10.1016/j.landusepol.2010.03.008.
[6] OECD (2019), Making Decentralisation Work: A Handbook for Policy-Makers, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/g2g9faa7-en.
[5] OECD (2018), OECD Economic Surveys: Israel 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_surveys-isr-2018-en.
[4] OECD (2014), “Taxation and tourism”, in OECD Tourism Trends and Policies 2014, OECD Publishing, Paris, https://doi.org/10.1787/tour-2014-6-en.
[3] Ollivaud, P. and P. Haxton (2018), “Making the most of tourism in Indonesia to promote sustainable regional development”, Working Paper of the Economic Department.
[9] Portnov, B., W. McCluskey and W. Deddis (2001), “Property taxation in Israel: A non ad valorem approach”, Land Use Policy, Vol. 18/4, pp. 351-364, https://doi.org/10.1016/s0264-8377(01)00029-1.
[10] Reschovsky, A. (1980), “An evaluation of metropolitan area tax base sharing”, National Tax Journal, Vol. 33/1, pp. 55-66.