Evaluating the performance of Development Charges in financing municipal infrastructure investment. Discussion Paper. Second Draft 23 March 2009

Transcription

1 Evaluating the performance of Development Charges in financing municipal infrastructure investment Discussion Paper Second Draft 23 March 2009 Prepared by David Savage for the World Bank

2 Contents Executive Summary Introduction and Background The Role of Development Charges Scope Revenue instruments Payment mechanisms Impact Development Charges in South Africa The current legal framework Administrative practices Charging practices Payment procedures The Revenue Performance of Development Charges Data constraints Revenue trends Revenue adequacy Enhancing the Performance of Development Charges in South Africa Enhanced policy and legislation by national government Improved operational practices by municipalities Conclusion REFERENCES

3 Executive Summary Municipalities currently significantly under-recover revenues from development charges, amounting to anything between R487 million and 4,7 billion per year. This is increasingly problematic as the revenue foregone limits the ability of municipalities to investment in the expansion of infrastructure that supports economic growth and poverty reduction. By failing to adequately charge for and collect this revenue source municipalities are: a) Transferring benefits to private developers through permitting them to maximise their profits at the cost of ratepayers b) Failing to tap into a ready source of infrastructure finance, drawn from the property finance industry and home-owner creditworthiness, rather than their own balance sheets. c) Implicitly reallocating resources away from other priorities, such as pro-poor expenditures, particularly over time as they will inevitably be forced to expand infrastructure networks at a later date to cope with congestion d) Imposing a cost on the economy, in terms of both forcing an inappropriate allocation of capital away from the property sector in their areas where they do not provide infrastructure, and through imposing economic costs associated with network congestion until they actually do invest This is fundamentally inequitable and inefficient. Two interventions are appropriate to address these problems. Firstly, improved national regulation and monitoring, secondly improved municipal administration. An immediate next step would be for interested municipalities to initiate discussions on common approaches to both interventions. 2

4 1. Introduction and Background South African municipalities face significant challenges in mobilising finance to meet their infrastructure investment requirement. A number of analyses have pointed to infrastructure investment needs emanating from the need to expand access to services, to expand infrastructure that supports economic activity, and to maintain, rehabilitate or replace existing infrastructure that is nearing the end of its design life. Two factors have begun to compound the effects of this situation. Firstly, construction costs have risen dramatically in the last few years, resulting in municipalities being able to construct fewer assets for the same amount of revenues. Secondly, municipalities now face a significant accumulation of investment needs from both previous years and the present. Government policy has responded to this situation in two ways. Firstly, grants targeted at infrastructure investment have increased fairly substantially, and now make up the largest single source of capital revenues for all categories of municipality. Secondly, the policy framework for municipal borrowing has been substantially reformed, to improve the long term capacity of creditworthy municipalities to borrow on a sustainable basis. Both of these efforts, while important, face their own constraints. It is neither likely nor desirable that national grant funding will sustain its current share of municipal capital revenues, particularly in larger urban municipalities. The limited growth in municipal borrowing appears to be at least partially a response to increased national transfers, and partially a function of weaker than expected expenditure capacity within municipalities. In this context, there appears to be a need for municipalities to look beyond traditional sources of revenues, drawn from national government and mainstream local taxes and user chargers respectively. Development charges represent one such additional and alternative financing mechanism for municipal infrastructure investment. They tap into a new source of revenue, improving the equity of the system of infrastructure financing, and draw resources from a different base, essentially through the private property financing system (such as home loans) as associated infrastructure investments directly affect the market value of underlying land assets. While these charges have long been used in some municipalities, they are largely under-researched and their role in the current municipal environment not well understood. The evidence suggests that they are under-utilised, which introduces a fundamental distortion to land markets in South Africa, either through pricing property developments below their real costs, or allowing developers to capture the benefits of under-charging without passing this on to property owners and thus ensuring that investment returns in the industry are higher than elsewhere. The aim of this paper is to make a contribution to the understanding of development charges in the South African environment. It will describe, analyse and evaluate the theory as well as legal, institutional and fiscal practices associated with development charges, and recommend objectives, principles, approaches and specific issues to be resolved in refining the current policy framework for these charges. 2. The Role of Development Charges Development charges are commonly used levies that are imposed on developers of new or existing properties, usually at the point that a property is subdivided or when a development or building permit is issued (in other words, in the course of an effective change in land use rights). The primary purpose of a development charge is to contribute to the cost of additional municipal infrastructure arising from the more intensive development associated with these land use rights. 3

5 For example, a new residential development on previously undeveloped land requires new water, sanitation, roads and electrical infrastructure in the form of new connections and feeder infrastructure, and absorbs the residual capacity of existing bulk infrastructure such as existing water treatment plants. Similarly, a re-development of a single story house as a multi-story townhouse complex requires improved infrastructure capacity to serve the additional units, and places additional strain on existing bulk supply and roads in the vicinity. Development charges are thus used to ensure that new developments contribute their fair share of the costs associated with connecting them to infrastructure networks. However, determining the scope, instruments and size of this fair share is conceptually and administratively complex. Conceptually, the scope of Development Charges requires careful definition both spatially and temporally. Administratively, calculating, imposing, collecting and expending associated revenues needs to ensure revenue adequacy for municipalities, equity considerations for contributors and the overall efficiency of the land market. 2.1 Scope In general development charges reflect all or some of the following four components, which are borne differentially over both space and time: a) A charge to recover all or part of the directly apportionable costs of connecting a development to each municipal infrastructure network. This component of the charge is often based on estimates of actual costs for additional infrastructure. It is a generally well accepted charge related to the direct additional cost of a new development or change in land use. These can be known as connection charges, capital cost recovery charges, or capital contribution fees, that any new development must pay in order to connect to water, sanitation or electrical networks, and some cases also for roads and highways (such as a new intersection). b) Shared direct costs of additional infrastructure associated with the development but that will benefit other users as well. For example, a road nearby may require upgrading for increased traffic volumes, additional capacity may be required at the water treatment plant, or a new bulk pipeline might be required. These costs are also levied based on estimates of the actual cost of additional infrastructure, but are spread between different developers in the same area, and often only collected over time (as each development may take place at different times) 1 c) Contributions towards sunk or historical costs of previously-installed infrastructure that will be partly used up by the development. For example, a new development may utilise the remaining capacity of a bulk pipeline or water treatment works that has been installed previously. This infrastructure benefits the new development, but has been paid for and maintained by existing residents and businesses. A new development might thus be required to contribute a share of the cost of this infrastructure, typically valued at the portion of the replacement cost of the infrastructure utilised by the development. 1 These shared direct costs have some similarities with general infrastructure improvements in a local area, such as those that might be financed through a Business Improvement District (BID) initiative with an associated surcharge on property taxes. In both circumstances a number of users benefit directly from an investment and thus can be expected to make an equitable contribution to the cost of infrastructure upgrades. However, the difference is that BIDs investments do not typically have to accommodate additional uses over time, and thus spread the financing burden temporally. 4

6 d) Cost associated with any externalities associated with the development, such as increased traffic volumes, the degradation of the natural environment or the long term operating cost implications for the municipality of development in a particular location. These costs are the most difficult to quantify, although environmental economics has made considerable progress in costing the impacts of development on the environment. This cost component is also most commonly seen as a tool that can incentivize spatial restructuring, through raising the costs of peripheral urban development. Diagram 1: Components of Development Charges All but the last dimension of development charges relate to contributions towards infrastructure that is either already installed or will shortly be installed. The cost of externalities, however, does necessarily relate directly to infrastructure installation costs. Including externalities does raise the question of whether positive externalities should also be seen as a component (or legitimate area) for development charges. For example, public investments that raise the value of a nearby property (such as a new park, upgrading sewerage or water capacity, or locating a railway or bus station close to a property) can have a significant positive effect on its value. In some cities, beneficiation charges are levied to allow taxpayers in general to share in the windfall benefits that accrue to specific property owners through these publicly funded investments. In addition, the concept of externalities is often used to justify the use of development charges as instruments for spatial restructuring, something that is of particular importance in South African cities with their legacy of apartheid spatial development patterns. These two issues are not considered in this report, as improving the administration of the current system of development charges is arguably of primary importance at this point, and other local financing mechanisms, such as Tax Increment Financing and Rates surcharges are arguably better instruments to effect local value capture and spatial restructuring. 5

7 2.2 Revenue instruments Development charges can be levied through a wide range of different instruments, including development application fees, connection charges, development levies and development taxes. The instruments and their importance vary widely between local areas, based on their intent and implementation arrangements. There are typically two major policy choices associated with the selection of an appropriate instrument: a) A general or specific instrument: An initial choice is whether the costs of new developments should be recovered up front, in the form of a development charge, or over time through user fees and property rates. This debate is often clouded by the desire of municipalities to attract development, which boosts their revenue over the long term. Positive long term revenue impacts are of course an important consideration for any municipality, and there is some evidence that development charges can be inefficient 2. However, this assumes that the costs of developments are included into tax rates and utility prices in a manner that ensures their equitable distribution across all taxpayers or service users over time. In general this is not the case. Future property tax revenues and user fees from new developments cover ongoing operating costs of services, as well as the depreciation of these assets. In other words they fund the future replacement of assets rather than the costs of their initial installation or upgrading. A failure to levy appropriate charges up front thus more closely represents an investment incentive or subsidy, and at the very least should be transparently treated as such. In fact, this results in current consumers being required to subsidise new development both by building capacity for new development and by financing those costs over time. b) A tax or charge: A subsequent choice is the extent to which the charge is treated as a straightforward tax on developments, or a fee associated with a specific and quantifiable set of infrastructure services. This largely depends on which components of the charge are pursued, as discussed in the previous section. In general, a development charge begins to assume the characteristics of a tax as costs become harder to quantify. In other words, as more components are added to the charge (from directly apportionable costs to externalities) the charge evolves from a user fee into a tax. Confusion over the status of development charges often leads to a proliferation of revenue instruments that attempt to cover one or more of these cost components. Clearly, the choice of instrument(s) has important implications for the administrative burden associated with levying the charges. More practically, it can have important impacts of the revenue productivity of development charges. For example, a tax based on the value of the development is both buoyant and easy to calculate, whereas a host of smaller charges that attempt to reflect actual costs require complex administrative systems to manage, and require regular updating of the underlying cost data to remain productive over time. However, taxes are often subject to far greater public scrutiny as to their equity effects than fees and charges, which can also limit their productivity over the longer term. 2.3 Payment mechanisms There are a variety of mechanisms that can be used to effect payment of development charges. These should not be confused with the charge itself. However, payment mechanisms have typically provided an important avenue for municipalities to innovate in the provision of infrastructure. 2 See, for example, Giesecke, Dixon and Rimmer, 2005, who consider the macro-economic effects of alternative methods of financing urban infrastructure in Australia 6

8 In general, payment can be made in cash (either for each service on an individual basis, or collectively and then apportioned by the municipality to specific services), or can be made in the form of assets of a comparable value. Cash payments are typically used for small developments (such as the rezoning of a small land parcel). Asset transfers are typically associated with larger developments. In this case, a developer will construct infrastructure and then donate this to the municipality. In some areas, land dedications (for a new community park, servitude or road) may also be used. Various innovations can result from the use of asset transfer mechanisms for the payment of development charges: a) Most directly, the use of asset transfers provides municipalities with a tool to circumvent internal procurement procedures and reduce inflation risks. Municipalities can obtain assets without going through their own complex budget approval and procurement procedures. This is attractive to both municipal service delivery departments and developers, who can ensure that development charges directly benefit their projects. In addition, asset transfers can provide an inflation hedge for municipalities, as the construction sector inflation risks can be transferred to developers along with the liability for constructing assets. b) Some asset transfers are subject to over-sizing requirements, where a developer will build infrastructure that exceeds immediate needs, and recovers costs from the municipality over time, or directly from future developments. This allows a municipality to install infrastructure in advance of need, without having to directly finance it off its own balance sheet c) Asset transfers can also be associated with operating partnerships. In many instances developers are required to operate and maintain assets for a year or more as part of the commissioning process. In other instances longer term PPP arrangements or management contracts are formed as part of the asset financing agreement. 2.4 Impact Development charges can have both an economic and spatial impact. From an economic perspective, the charges are intended to ensure that there is an equitable distribution of the costs of new developments between new developments and existing ratepayers. This is based on the benefit principle, which calls for an equitable distribution of costs based on the benefits each party receives from the new development and associated infrastructure. The challenge facing policy makers is to reconcile the incidence of the costs of development charges (the cost incidence) with the incidence of the benefits of associated infrastructure (the benefit incidence): a) Benefit incidence: The benefits derived from new infrastructure by ratepayers in general and new residents differ. New residents receive a direct benefit of being able to use infrastructure on a daily basis, whereas ratepayers in general receive far more diffuse benefits, arising from the integrity of infrastructure networks (the ability to occasionally use a road, for example), and the expansion of the revenue base of the city so that future tax burdens are shared among a greater number of taxpayers. b) Cost incidence: It is largely accepted that the cost burden of development charges largely falls on the ultimate owners of new properties, rather than the developer. However, this is subject to two potential influences. Firstly, the price of new properties is set by the market so, if market conditions are not conducive, the developer may land up paying a portion of the charges in the form of lower profits. Secondly, if development charges are set at a level that does not recover the cost of associated 7

9 infrastructure expenditure then ratepayers in general will bear a disproportionate share of the costs. However, as the price of a new property is set by the market and is likely to include the assumed value of infrastructure services, this cost saving is unlikely to be passed the ultimate property owner and will be retained by the developer. Diagram 2: Components of the market price of land and associated charging instruments Reconciling the cost and benefit incidence in practice is a complex exercise. Developers and new property owners typically complain that this results in double-charging, through a specific development contribution and then again through property rates 3. Confusion over the financing role of property rates is at issues here, as these rates do not cover the initial cost of installing infrastructure, but rather the costs associated with its operation, maintenance, rehabilitation and eventual replacement. From a spatial perspective, development charges can served to reduce urban sprawl through raising the costs of peripheral developments to account for their full costs, such as extending bulk supply networks or installing new infrastructure. However, there are no examples of development charges effectively addressing spatial policy objectives such as this, largely because of the associated administrative complexity. In essence, the charging mechanism would need to differentiate the costs of infrastructure across space (in other words use marginal cost pricing, rather than average or historic costs), and this has proved too complex for local administrative systems to manage and maintain effectively. 3. Development Charges in South Africa In South Africa, municipalities have levied development charges in various forms for many years. In the former Cape Province, Betterment Fees were in use from at least 1935, based on the percentage of the value of any improvement that was made to a land holding. The fees were typically levied at rates of around 2% of the value of the improvement The current legal framework The former provincial administrations revised this approach in the 1980 s, settling on a formula that sought to determine the impact of a new development on the use of new and existing assets. Developer contributions were calculated on the basis of installed asset values, including the value of loans on those 3 In some instances there are allegations of triple charging, when sectional title levies or homeowners association assessments are included as well, such as may be the case in gates communities. 4 Oakenfull, et al (1997) provide a comprehensive review of the history of Development Charges in South Africa that is not repeated here. 8

10 assets, as well as additional infrastructure requirements directly apportionable to that development. This approach has a number of new features: a) Charges were levied at the point of a chance in land use or development rights, rather than after the improvement (construction) of a property. This improved the ability of municipalities to collect revenues, through making the payment of development charges a condition for the approval of any change in development rights. In most instances this means that a developer has to sign an Acknowledgment of Debt (or Notarial Deed in KZN), which must be cleared before that associated property can be further sub-divided or transferred. b) Charges were enabled and regulated by provincial land use planning ordinance, rather than seen as a general revenue instrument for municipalities. The ordinances have generally been interpreted to restrict the scope of the charges to public expenditures directly arising from the change in development rights, whether these are current or historical. c) Municipalities began to increasingly permit the payment of charges in kind, particularly for larger developments, through negotiating services agreements that required developers to install or upgrade infrastructure and transfer this over to the municipality. Extract from Section 42 of the Land Use Planning Ordinance, 1985 (Cape Province):- (1) When the Administrator or a council grants authorisation, exemption or an application or adjudicates upon an appeal under this Ordinance, he may do so subject to such conditions as he may think fit, having regard to- (a) the community needs and public expenditure which in his or its opinion may arise from the authorisation... [or] the public expenditure incurred in the past which in his or its opinion facilitates the said authorisation, and (b) the various rates and levies paid in the past or to be paid in the future by the owner of the land concerned, include conditions in relation to the cession of land or the payment of money which is directly related to requirements resulting from the said authorisation, exemption, application or appeal in respect of the provision of necessary services or amenities to the land concerned. The restriction of the scope of the charges has resulted in revenue pressure on municipalities, as they have found it difficult to levy charges for shared direct costs and other indirect costs. To address this constraint municipalities have either pursued debt agreements with developers or have expanded the number of revenue instruments to raise revenues beyond what is possible through connection charges. For example, some municipalities have raised development application fees considerably beyond the cost of processing them. This has often resulted in intensive litigation by developers seeking to avoid what they perceive as excessive charging. Underlying this problem is the question of whether development charges are a tax or user fee. The judgement in Johannesburg City Council v Victteren Towers (Pty) Ltd (1975(4)SA334(W)) alludes to the charge being essentially a tax, yet this is contradicted by: a) The empowering provincial ordinances that refer to the payment of money which is directly related to the development authorisation, suggesting that the charges are currently framed to more closely represent a user fee to recover costs of infrastructure. This is borne out by the methodological precision that some municipalities attempt to achieve in determining and apportioning costs; 9

11 b) The Municipal Fiscal Powers and Functions Act, which requires any municipal tax or surcharge to be approved by the Minister of Finance, subsequent to a formal process. Should these charges indeed be a tax then they may be subject to legal challenge on procedural grounds; and c) The findings of the KZN Town Planning Appeals Board, in ethekwini Municipality v Dunstan and Bekker (Appeal no. 363) on the issue of whether the municipality had the authority to impose a development charge based on the leasable area of the property. The Board found that the municipality had no statutory authorisation to levy a charge on this basis, nor to compel the developers to enter into a debt agreement. An additional legal problem is exposed by the KZN Town Planning Appeals Board, in ethekwini Municipality v Dunstan and Bekker (Appeal no. 363). In this case the municipality had, among other things, required the registration of a road servitude over private land as a condition of development approval. This is essentially a land endowment, as a form of asset based development charge. The finding notes that this amounts to expropriation without due compensation, and set the condition aside. From a policy perspective this is a disturbing result, as developers must always set aside roads, parks school sites and rights of way as part of the development process. It points to a potentially significant weakness in the current legal framework. A final legal issue is raised by Johannesburg City Council v Victteren Towers (Pty) Ltd (1975(4)SA334(W)) relating to the retrospective application of development levies following a change in the ordinance. Although the Applicants case was granted, it did raise the issue of whether any Acknowledgement of Debt associated with a Development Charge can be enforced, if this acknowledgement was given before any physical change to a property has occurred. At issue here is the point at which a charge may be levied or collected, as an application for a change in development rights does not necessarily result in development occurring, and may in any event be refused or overturned. It is clear from this brief review that significant legal contestation occurs between municipalities and developers over the levying of development charges. These matters are adjudicated by town planning appeals boards at provincial level. Developers have a considerable incentive to pursue this process in order to maximise their profitability, given the considerable sums of money that are involved. Adjudication is made more complex by the existence of parallel national legislation guiding spatial development, in the form of the Development Facilitation Act. This Act requires municipalities to provide bulk services associated with development applications, but is silent on who must pay for this, or how this payment may be effected 5. This requires municipalities to specifically ensure that special conditions regarding developer contributions are attached to individual decisions of the Tribunals established in terms of the Act. 3.2 Administrative practices Two different procedures for administering the levies appear to have emerged in different municipalities. In ethekwini, the Town Planning Department manages the levies in general, while individual infrastructure departments manage their own connection fees. In Cape Town, the charges are managed by Engineering Departments, who levy the charge during their review of development applications submitted to the Planning Department. 5 The proposed Land Use Management Bill is similarly silent on these issues, although it appears that the Bill is no longer at an advanced stage of preparation. 10

12 In Cape Town, the Engineering Department remains responsible for the collection of the charges, which is paid by the Revenue Office into specific asset funds established for each affected service by local area of the municipality. The result is a proliferation of fund accounts, as a number of funds for water infrastructure contributions may be established across the municipal area (typically a number of wards), in parallel to those for other infrastructure services. In Cape Town, for example, there are 16 fund accounts across the municipal area. 3.3 Charging practices There are three issues that arise from the current approach to the levying of development charges: a) The scope and coverage of development charges Municipalities currently interpret the legal framework narrowly, in order to project development charges as a fee that is directly related to identifiable infrastructure costs associated with new developments. This approach is intended to reduce the risk of litigation by developers, who might claim they are being charged multiple times for the same infrastructure or endowments. However, this results in a tendency to ignore shared costs and externalities, whose costs are borne by ratepayers in general. Subsidised low-income housing developments are formally exempted from development charges. The rationale here is that national infrastructure grants, particularly the Municipal Infrastructure Grant, are provided to fund the cost of this infrastructure, with the housing subsidy funding the costs of internal infrastructure. Although conceptually correct, in practice this has led to two problems: Firstly, no municipality currently attempts to directly offset these infrastructure costs against MIG contributions, and so it remains unclear as to whether adequate MIG funding is being provided relative to housing subsidy allocations. Secondly, institutional and gap housing projects are seen as normal developments and required to pay development charges. This introduces inequity into the system. Current proposals by the National Department of Housing to withdraw funding for internal infrastructure from the housing subsidy may further exacerbate these uncertainties. Overall, the lack of clarity on the legal scope of development charges creates significant administrative challenges for municipalities, and has resulted in them foregoing revenues in an effort to reduce litigation. b) Cost calculation methods Many municipalities also appointed consultants to determine calculation methods and cost averages for each infrastructure service. The majority of these remain in place even today, despite significant changes in municipal boundaries. This means that calculation methods can vary within a single municipality (such as is the case in Cape Town) In Cape Town, this generally resulted in the development of two formulae, for Greenfield and Brownfield developments respectively to reflect the different costs they impose on the municipality. Other cities, such as ethekwini, deploy individual formulae for each services. These generally cover only the direct costs of new infrastructure requirements. The need to regularly re-establish (rebase) unit costs has proved to be the primary weakness in these formulae. In theory this should be done every five years, with standard cost escalations applied in intervening years. In practice, however, municipalities have found it difficult to allocate funds to undertake complex surveys of input costs and so escalations have been applied for longer period. This has led to some divergence between actual and estimated unit costs. The result is that municipalities must subsidise the cost differential, often through delaying investment or maintenance over time. 11

13 c) Alternative charging instruments As a result of these uncertainties, and due to the rapid pace of urban development, cities such as Cape Town have begun to experiment with alternative charging mechanisms. This has been based on a simple calculation of average costs per unit for a set of typical land uses in Greenfield developments (or in other words an average rate for each square metre of gross leasable area, or GLA). ethekwini municipality, in addition to connection fees, has experimented with the introduction of two additional development charges, namely, increased application fees, based on practices in Hibiscus Coast Municipality, and a general development tax (from 1 June 2004) that focussed on two rapidly growing areas of the city, amounting to R10,000 per residential unit and / or R15,000 per 100m2 of gross leasable area of commercial development. Alternative instruments have run into practical and legal difficulties. In practice, application fees are seldom high enough to cover infrastructure costs, while the introduction of a formal tax is almost certainly procedurally illegal in terms of the Municipal Fiscal Powers and Functions Act, Payment procedures The linkage between land development authorisations and the levying of development charges has provided a powerful credit control mechanism. Developers are required to sign an Acknowledgement of Debt as a condition for the approval of a development application. The charge is then collected at a later date, which will vary by the nature of the approval granted. The failure to make a payment will result in any further transfer of property ownership or development approval being automatically blocked. In Cape Town, for example, payment must be made prior to the transfer of property in the case of subdivisions, and prior to building plan approval in the case of the development of other properties (such as flats). Smaller developments in inner city areas typically pay the charge in cash. These funds are allocated to service specific fund accounts until such time as sufficient funds are available for investment in respective services in that local area. Developers are also able to make asset contributions to the municipality in lieu of cash payments. This requires that a back-to-back Services Agreement is signed between the municipality and the developer. Municipalities will typically use the services of a professional consulting engineer to cost and monitor the implementation of these agreements. Services Agreements are typically used in large scale, Greenfield developments on the periphery of municipal areas. This mechanism is preferred by municipalities as it removes the need for them to design, procure and construct assets on their own. In addition, it allows municipalities to transfer the risk of construction cost inflation to the developer after the agreement is signed, which is not possible with cash payments. However it does leave open some scope for abuse, so requires careful oversight. The type and extent of infrastructure services must be carefully specified, and the cost differences between the Acknowledgment of Debt and the Services Agreement must be subjected to an exit audit. In addition, the value of assets transferred to municipalities through these agreements is not always reflected in municipal budgets and financial statements, leading to a weakening of oversight and potential long term asset management difficulties. Although current accounting practices do require full disclosure of asset transfers, this is not always done. In Cape Town, approximately 40% of development charges are paid through Services Agreements, though these tend to be from a few large developments on the periphery of the city. A far larger number of smaller payments are made in cash. 12

14 4. The Revenue Performance of Development Charges 4.1 Data constraints Data on municipal revenues from development charges is captured under Public Contributions and Donations on municipal budgets. Generally Recognised Municipal Accounting Practices (GRAP) requires that municipalities recognise income from both cash and assets under this heading. Unfortunately, the quality of available data is of a particularly poor quality. It is not directly comparable across years or between municipalities due to the variable implementation of accounting practice, with some municipalities only recently capturing the value of asset transfers, and some having yet to do so. Still other municipalities do not record any revenue from development charges in some or all years for which data is available. This may be a result of income being captured under other revenue heads. Any analysis of Development Charges must therefore be treated with some circumspection. Anecdotal evidence suggests that development charge revenues are considerably greater than what is reported. In Cape Town, for example, it is estimated that development charges provide approximately five times the resources of general capital budget allocations for road infrastructure investment, to the point where some road infrastructure that is not strictly related to a development may be funded through this mechanism 4.2 Revenue trends Setting aside these severe data weaknesses for the purpose of a basic analysis reveals that few municipalities in South African report any income from development charges. An average of 29 municipalities, out of 284, reported any annual income at all from this source, between 2004/05 and 2006/07. Table 1: Revenue from development charges by municipal sub-category (2004/05 to 2006/07) Outcome Outcome Outcome Outcome Medium-term estimates 2003/ / / / / / /10 Average Category A Public Contributions % contribution 6.1% 1.3% 1.3% 1.8% 3.0% 2.5% 3.4% 2.8% % growth -72.4% 27.5% 78.9% 163.0% -15.3% 4.1% 31.0% Category B Public Contributions % contribution 0.6% 3.2% 2.9% 1.0% 1.7% 2.0% 1.5% 1.8% % growth 505.0% 18.3% -61.3% 277.3% 15.1% -44.5% 118.3% Category C Public Contributions % contribution 0.0% 0.3% 0.1% 0.0% 0.1% 0.1% 0.7% 0.2% % growth -14.9% % -50.0% 900.0% Secondary Cities Public Contributions % contribution 0.6% 5.6% 3.9% 1.5% 1.0% 0.9% 1.3% 2.1% % growth % -15.8% -51.7% 29.1% -17.5% -5.2% 159.5% Total Public Contributions % contribution 3.5% 1.9% 1.7% 1.4% 2.1% 2.0% 2.3% 2.1% % growth -33.2% 21.5% -4.8% 192% -6.2% -10.8% 26.5% Source: LGBER,

15 Municipalities reported collections of R286.7 million in Development Charges in 2006/07. This constitutes only 1.4% of the total capital revenues of municipalities in 2006/07. This constitutes part of a long term decline in revenues, on both nominal and proportional terms, as shown in Chart 1 below, which is expected to continue into the medium term, despite some optimistic budgeting by municipalities. Chart 1: % contribution to capital revenue by municipal category Source: LGBER, 2008 Until recently, district municipalities reported no income from development charges, but this has begun to change. Regardless, around 60% of the revenue from development charges is collected in the six major metropolitan municipalities. This rises to 67% of total revenues when secondary cities are included. Metropolitan municipalities and large towns have been the primary beneficiaries of this growth. Metropolitan municipalities increased revenues from Development Charges at an annual average rate of 31%, far exceeding the 23.7% average annual growth in their capital budgets. Growth in revenue from Development Charges in large towns largely matched that of total capital revenues over the period, but has been extremely volatile. Secondary cities, small towns and rural municipalities have all experienced negative or slower growth in development charges, either through nominal declines in revenue or as a decline in its contribution to total capital revenues. 14

16 Chart 2: Nominal revenue from Development Charges by municipal category Source: LGBER, 2008 Development charges are the single most volatile source of capital revenue for municipalities, as shown in Chart 3, below. This volatility is partially a function of the small size of these charges (which encourages large shifts in this indicator from relatively small revenue changes, and may be induced by variable municipal reporting, as discussed above), as well as speculative budgeting by municipalities for the 2007/08 financial year. However, volatility does inhibit capital investment planning and may reflect poor management practices by municipalities. Chart 3: Annual growth in capital revenue by source (2003/04 to 2009/10) Source: LGBER, 2008 Finally, there is a wide variation across municipalities in the effective rates at with development charges are levied. Table 2 calculates the effective rate at which development charges are levied by municipalities, based on the revenue received as a function of the value of all buildings completed in a municipalities between 2003 and Average rates charged by municipalities have been volatile across years, averaging 15

17 2.1% over the period. Individual municipalities have also experienced significant volatility in rates charges between years. Table 2: Rates of municipal investment and development charges Average Cape Town 0.7% 0.8% 0.7% 0.6% 0.6% 0.7% Drakenstein Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% George Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Mossel Bay Municipality 0.0% 0.0% 2.5% 0.4% 0.8% 0.7% Stellenbosch Municipality 0.8% 0.0% 0.0% 2.5% 4.9% 1.6% Buffalo City Municipality 0.3% 9.5% 2.4% 3.3% 0.0% 3.1% Nelson Mandela Metro 15.7% 0.9% 1.2% 0.6% 3.3% 4.3% //Khara Hais Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Sol Plaatje Municipality 0.0% 1.2% 0.0% 0.0% 0.3% 0.3% Dihlabeng Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Maluti-a-Phofung Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Mangaung Local Municipality 2.4% 3.0% 0.0% 0.0% 1.8% 1.4% City of umhlathuze 0.0% 57.1% 17.4% 29.0% 63.2% 33.3% Ethekwini Municipality - 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% Hibiscus Coast Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Msunduzi Municipality - 0.0% 0.0% 4.4% 0.0% 0.3% 0.9% Madibeng Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Rustenburg Local Municipality 0.0% 1.3% 0.0% 0.0% 0.2% 0.3% City of Johannesburg 1.1% 1.4% 1.3% 2.3% 4.5% 2.1% City of Tshwane 12.4% 0.0% 0.0% 0.0% 2.5% 3.0% Ekurhuleni Metropolitan Municipality 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% Emfuleni Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Kungwini Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Midvaal Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Mogale City Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Emalahleni Local Municipality - 0.0% 31.8% 18.6% 0.9% 0.6% 10.4% Mbombela Local Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Greater Tzaneen Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Polokwane Municipality 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% AVERAGE 1.1% 3.7% 1.7% 1.4% 2.9% 2.1% Source: LGBER 2008 and Stats SA The relatively small contribution and volatility of development charges to total capital revenues, and its uneven distribution between municipalities raise concerns beyond the accounting treatment of these revenues: the apparent volatility in rates suggests that municipal policies around development charges are poorly developed, while the associated swings in actual revenues must undermine associated planning and expenditure management processes. In some instances municipalities may be reducing (or failing to implement) development charges in order to attract investment. In other instances, a lack of capacity to negotiate these charges may result in the financing burden of associated infrastructure being transferred to local ratepayers in general. 16

18 4.3 Revenue adequacy Revenue trends contribute to a more general problem associated with the current performance of development charges: the adequacy of revenues generated relative to associated municipal investment needs. Cart 3 demonstrates that revenue generated from development plans has fallen as a percentage of the value of buildings completed and building plans passed, which are themselves a proxy of municipal investment needs. Chart 3: Development charges as % of building plans passed and completed Source: LGBER, 2008, StatsSA Two methods can be used to calculate an appropriate level of revenues from development charges, based on (a) revenue potential, based on the average effective charging rate of municipalities, as extracted from Table 2, above; or (b) expenditure needs, based on estimates of infrastructure investments required from municipalities to support high-income, residential, commercial and industrial development, drawn from the Municipal Infrastructure Investment Framework model. Table 3: Estimates of Revenue Adequacy R' Avg Actual revenues Expenditure Needs Shortfall - Needs Revenue Potential Shortfall - Potential Both methods indicate a substantial and growing shortfall in revenues currently being generated by municipalities, relative to both revenue potential and revenue needs, as shown in Table 3 above. This shortfall ranges averages R482 million per year based on the revenue potential method, and R4,7 billion a year based on the expenditure needs method. 17

19 The implication of this under-recovery of development charges is that a significant subsidy is being provided to the property development sector. This subsidy is paid for in two forms: either as a direct transfer from ratepayers in general, who must pick up the costs of required infrastructure investment, or by the economy as a whole, due to the deadweight costs imposed by network congestion that arises from delayed investment in additional infrastructure by municipalities. As noted earlier, this subsidy does not necessarily translate into lower property prices, as the benefit is largely captured by developers themselves in terms of higher profit margins. 5. Enhancing the Performance of Development Charges in South Africa This document has reviewed the conceptual and legal foundations of development charges in South Africa, and assessed their performance based on available data. It has shown that municipalities currently significantly under-charge and under-recover the costs associated with private investment in the built environment. Significant additional revenue is currently being foregone by municipalities, which could amount to anything between R500 million to R4.7 billion per year. This has two negative effects: Firstly, it transfers the financing burden for required infrastructure to ratepayers in general, which in turn may limit the pace at which municipalities can finance overall urban infrastructure needs. Secondly, it imposes deadweight costs on local economies, associated with congested infrastructure systems as municipalities delay additional investment. Addressing the current under-performance of the system of development charges will require two related sets of interventions at national and municipal levels respectively: 5.1 Enhanced policy and legislation by national government The legal framework that regulates development charges is currently under-specified and overly prescriptive. Municipalities in the seaboard provinces assume their powers to levy the charges in terms of a small portion of provincial land use ordinances. In the inland provinces the legal specification is more developed. But in both instances, the current system leads to extensive litigation as developers seek to minimise the applicable charges and maximise their profits. The jurisdiction of service tribunals or boards over these charges, often without the requisite understanding of the financial implications of their decisions, further complicates this situation. In addition, the legal framework is too restrictive in focussing on direct connection costs as this ignores the cost associated with new developments absorbing marginal network capacity and creates complexities in calculating charges. The prescriptions and vagueness of the legal framework have resulted in municipalities adopting highly restrictive interpretations of their powers to levy development charges, and in some cases may have discouraged smaller municipalities from imposing them at all. Some municipalities have evolved additional administrative procedures, such as the Acknowledgement of Debt in Cape Town, to avoid legal challenges. This outcome is both inequitable and inefficient and suggests that a legal remedy is required to re-regulate the system of development charges to clarify the scope of charges that may be levied, reduce incentives for litigation, and facilitate administration by municipalities. In addition, considerable improvements are required to the data that is available on the performance of development charges at a national level. A spot audit of actual revenues in both cash and kind may be required to establish the extent to which under-reporting of revenues has occurred. More detailed guidance on appropriate accounting practices may be required as well, including a method to reconcile reported revenues with underlying costs in the case of low cost housing developments and services 18