Financing Transport Infrastructure in Developing Country ... - CiteSeerX

Financing Transport Infrastructure in Developing Country Cities: Evaluation of and Lessons from the Nascent Use of Impact Fees in Santiago de Chile Paper to be presented at the 82nd Annual Meeting of the Transportation Research Board (TRB) Washington, DC, January 2003 Christopher Zegras Massachusetts Institute of Technology Department of Urban Studies and Planning Cambridge, MA, USA Tel: 617 784 1775 Fax: 617 258 8081 Email: [email protected]

ABSTRACT This paper explores the potentials for and limitations to the use of impact fees to finance urban transport infrastructure in developing country cities, drawing from the specific case of Santiago de Chile. The paper first assesses the current state of urban transport infrastructure financing and some of the inherent complications. The paper then overviews the growth of impact fee use as an infrastructure-financing tool in the United States and presents principles for their “appropriate” use, based on the U.S. experience. The paper then summarizes the nascent use of transport impact fees in Santiago, where they are currently being used to help finance transport infrastructure demands resulting from rapid suburbanization there. After providing a background to the current urban transportation infrastructure finance system in Chile, the paper assesses the use of impact fees in Santiago, based on the principles of use developed in the U.S. Based on this preliminary assessment, the paper recommends that, if transport impact fees are to be used effectively in Chile then proper over-arching legal guidance for their use must be established, a uniform approach to their application should be employed, the relationship of impact fees to other user fees and other forms of development exactions must be clarified, the question of who actually bears the ultimate burden of impact fee costs should be answered, and, finally, the effects of transport impact fees on other public policy goals should be more clearly understood. The paper concludes with lessons from the Santiago experience.

7,032 words; 2 Tables.

TRB 2003 Annual Meeting CD-ROM

Paper revised from original submittal.

INTRODUCTION The challenges of transportation infrastructure development in a rapidly growing city are well known. Travel demand generally increases with growth in population and per capita income, expansion in infrastructure capacity generally does not keep pace with demand, and the ubiquitous urban transportation externalities (i.e., congestion, air pollution) result. Among the many barriers – such as environmental and community impact concerns – to transport infrastructure expansion in urban areas, finding adequate sources of finance continues to figure prominently in both the developing and the industrialized worlds. The finance challenge is exacerbated by: the multiple institutions usually involved in urban transport infrastructure development and maintenance, the range of direct and indirect user fees employed, and the distortions in investment signals that typically result. Theory and practice suggest that well-justified and well-designed impact fees can play an important role in supporting the efficient financing and delivery of urban infrastructure and also contribute to more efficient patterns of urban growth. To date, however, most experiences with impact fee deployment have been confined to the industrialized world, particularly the urban areas of North America. Impact fee use in developing country cities remains much less common. Nonetheless, it is precisely in developing world cities – facing rapid urbanization, rapid urban outgrowth and chronic infrastructure shortages – where impact fees could play an invaluable role. An earlier paper (1) documented the rise of the use of roadway and environmental impact fees in Santiago de Chile in the 1990s. This paper extends on that work, situating the Santiago experience with transport impact fees within the broader context of urban transport infrastructure finance; describing advances in the application of transport impact fees in Santiago; assessing the Santiago experience according to principles for impact fee deployment drawn from experiences in the United States; making recommendations, based on that assessment, for improvements in Santiago’s impact fee use; and, drawing lessons from the Santiago case. THE CONTEXT: URBAN TRANSPORT INFRASTRUCTURE FINANCE The use of transportation impact fees must be assessed within the overall context of their use – the urban transport infrastructure financing system. In an “ideal world” the urban transport finance system would have the following characteristics: fuel prices covering the resource costs (i.e., the border price) and the environmental costs of carbon dioxide emissions (which are directly proportional to fuel consumption); road maintenance and congestion costs charged directly through highly differentiated tolls; local environmental costs charged through emission fees; and any redistribution objectives pursued through non-distorting lump sum taxes (2). Such a system would not only send accurate signals to system users to ensure “efficient” system use, it would also provide a sustainable financing source. For example, Mohring and Harwitz demonstrated in 1962 that the revenues generated from efficient congestion charges will exactly cover the costs of providing the infrastructure, if the road provider optimizes road capacity (and also is not subject to economies or diseconomies to scale) (see, for example, (3)). Unfortunately, the “real world” of urban transport infrastructure finance falls far from the ideal. Few accurate, direct user charges exist. Instead, users pay for road space through a variety of indirect mechanisms, particularly fuel taxes and vehicle license fees, as well as via real estate and other taxes. Furthermore, since fuel consumption is relatively inelastic to price, governments often use fuel taxes as an important and buoyant general revenue source. In the developing world, where vehicle ownership rests largely with the wealthier classes, governments also sometimes use vehicle ownership fees and fuel taxes for general income redistribution. The picture is further complicated by the fact that the infrastructure supplying “agents” are multiple and fractured – often with responsibility for construction separated from that for maintenance and management, and with each of these areas of responsibility falling to different levels of government (national, regional, and/or local). Thus, in the “real world,” formally established, transparent urban transport infrastructure “budgets” – with explicit fees recognized clearly by users as prices – rarely exist, leaving different levels of government scrambling for resources from whatever sources at their disposal. IMPACT FEES One of the tools in this clouded urban transport finance system is the impact fee. Impact fees fall under the broader set of tools known as exactions, a term referring to a requirement that a real estate developer contribute to certain “public” infrastructure needs arising due to a particular project. “In-kind” exactions require the developer to physically provide the infrastructure, through, for example, the provision of local roads and street lighting. Once developed, the responsibility for the operations and maintenance of the infrastructure typically



passes to authorities. If development occurs with rational design standards, the use of in-kind exactions can be an economically efficient way of delivering roadway infrastructure – requiring a developer to build roads and incorporate the related costs into the selling price of the real estate project will force the developer to make a trade-off between the total road investment and road access (which effects the value of the remaining land for sale) (4). Exactions, Impact Fees & Value Capture In contrast to in-kind exactions, financial exactions, generally referred to as impact fees, charge developers for the infrastructure costs that their developments produce. Authorities typically charge impact fees to offset costs external (i.e., due to “off-site” trunk road infrastructure) to a particular real estate project. Impact fees have long been considered a way to “efficiently” provide for urban infrastructure by making developers face the infrastructure costs imposed by new residents – a “speculation” Brueckner (5) formalizes in an economic model. Still, the one-time levy implied in an impact fee cannot guarantee efficient use of supplied infrastructure, so that a combination of user charges (i.e., congestion fees) and impact fees is more theoretically justifiable (4). Absent such an approach, impact fees alone might provide a second best, if they accurately represent how costs differ among development types and locations, particularly “if the development types can be categorized to accurately reflect future consumption” (6). Impact fees are related to the concept of valorization (often referred to as “value capture”), except that they basically work in reverse and hinge, ultimately, on different precepts. Valorization (through, for example, the use of betterment taxes) aims to recapture from property owners a portion of the increased property values that accrue due to public investments in infrastructure (or other government interventions, such as changes in regulations). Impact fees, on the other hand, aim to charge property owners for the direct impacts their projects will have on infrastructure. Impact fees are most typically used to finance infrastructure for new developments in high growth areas, while value capture tools are most often used to recapture part of the increased property values from new infrastructure in already occupied areas. Impact Fees – Precedents Rapidly growing local communities in the southern and western parts of the United States pioneered the use of impact fees during the 1970s, and by the mid-1980s they were being used to finance a wide variety of facilities, including police and fire stations, sewers, schools, etc. (7). By the mid-1980s some 60 percent of localities in the U.S. were using impact fees and exactions, with the growth in use fueled by: environmentalism, rising nogrowth/slow-growth citizen activism, a general public revolt against new taxes, reductions in federal aid for infrastructure, concerns about infrastructure shortfalls, and fiscal impact analysis as a guide for land use decision making (6). Most communities originally adopted impact fees on a “trial-and-error” basis that often ended up producing legal battles (8). Today in the United States, many state and/or local laws and regulations have formally enabled impact fees use. In the developing world, various forms of exactions exist, although their use is not well documented. In Latin American countries, for example, regulations commonly require real estate subdivisions to provide land for utilities, roads, parks, etc.; non-compliance, however, is often the norm (9). Regarding valorization tools, Jakarta Indonesia has had a betterment tax in place since 1972, aimed at recovering a portion of the government costs of providing new and improved infrastructure. Reportedly, the city has only sporadically used the tax to finance road projects (10). In Latin America, proponents have advocated the use of valorization since the 1920s (11) and nearly all countries in the region have national legislation allowing the public sector to capture the increase in land value resulting from public investment or changes in public regulations (9). Perhaps the best-documented use of value capture in the region is the contribución de valorización in Colombia. In use since at least the 1960s, the annual proceeds from valorización in Bogotá, measured as a percentage of all local public expenditures, have varied widely (from 0.5% to 16%) (9). The Colombian valorización approximates an impact fee, with: the infrastructure costs estimated; a zone of influence around the project defined; and, costs allocated to specific properties, typically using a coefficient system reflecting the project’s effect on the property (12). Again, however, the valorización assumes an increase in the value of abutting land, whereas an impact fee, technically, is charged based on the need that the abutting land imposes on infrastructure. A Framework for Assessment In response to the historically ad-hoc approach to impact fee deployment in the U.S., Lillydahl et al. (8) proposed guiding principles for model legislation (an “enabling act”). These nine principles, outlined in Table 1, offer a framework to assess impact fee use. While each of the principles is important, we detail the fifth principle, which we have called “Demonstrated Responsibility,” here. This principle aims to ensure that (8):



1.

2.

the need for new facilities indeed arises due to the new development being charged, not from existing infrastructure deficiencies – which requires a general planning process, including appropriate facility standards and a capital improvement plan with scheduled improvements; and the costs of any additional facilities required are then fairly apportioned to new developments – which requires knowing: the cost and financing approach of existing facilities; new developments’ contribution to infrastructure costs through other fees; if new developments should receive credit for financing facilities that have previously been provided without charges; a time-price differential to account for fee payments by different developments at different times.

Transport Impact Fees An early example of impact fee use for transport infrastructure comes from Broward County, Florida (USA) (13). Using impact fees since 1977, Broward County first adopted a roadway impact fee system (the Traffic Review and Impact Planning System or TRIPS) in 1983. TRIPS uses a simulation model for each real estate development, assessing impact fees on all development types, with the ultimate amount depending on the type, location, and size of the development. In 1984, the County adopted formal policies and a special accounting system for road impact fee spending and credits. The concept of credits links closely to the overall context of the urban transport infrastructure finance system, and is closely tied to Principle 5, “Demonstrated Responsibility” (see Table 1). An example roadway impact fee formula, as presented by Nicholas & Nelson (14), helps illustrate the point: Impact Fee = (ADTi * TLi)/2 * C – Creditsi Cap Where: ADT = TLi = Cap = C= Creditsi =

[1]

Average daily trip ends for land use i (based on accepted local trip generation parameters). Average trip lengths for land use i (estimated by authorities). Capacity of lane at planned level of service standard. Cost of right-of-way acquisition and construction per km of road lane. Discounted, present value of the stream of road user revenues used to finance road capital costs for each use, i.

Equation [1] calculates an impact fee for an individual land use (i.e., a housing unit), by charging the unit according to its “fair share” of trips generated (attributing the cost per lane of road to the individual land use by assigning it responsibility for one-half of the total trip ends generated) and the subsequent road capacity required to maintain a minimum roadway level of service. The formula also gives the individual land use credit for any relevant payments over time that would, ostensibly, go towards roadway infrastructure construction – thereby avoiding “double charging” a development. Without a clear understanding of the transport finance system (the specific revenues raised and spent for construction/maintenance by different levels of government) the credit level cannot be accurately estimated. As an example, Nicholas & Nelson (14) calculate the average estimated gasoline consumption and subsequent share of relevant gas taxes accruing to the local government; they also calculate the residential property tax payment and estimate the share local government uses annually for road construction (based on historic records). Both revenue streams are discounted over the facility’s assumed 25-year life-span. Variations on equation [1] can be found underlying impact fee ordinances and subsequent impact fee schedules across communities in the United States. Although impact fees have historically been employed to finance infrastructure in areas of new development (i.e., in suburban areas), examples of broader city-wide applications also exist. The City of Orlando, Florida (USA) offers a notable recent proposed variation, looking to modify a city-wide transportation impact fee program to allow for differential fees based on a project’s location in the metro area (see (15)). The approach starts with the calculation of an average vehicle distance traveled (ADT) per development unit type (i.e., single family home, commercial, industrial, etc.) for the entire region. Then, using the region-wide ADT, analysts calculate a ratio of ADT per development unit type for specific traffic analysis zones (TAZ). They then use this ratio to calculate corresponding correction coefficients to adjust the transport impact fee for a development unit type in a given TAZ. The city has not yet implemented the proposed modifications due to issues relating to roadway connectivity features in different neighborhood types as well as concerns regarding the potential fiscal impacts of the changes (personal communication with Pedro Leon, Transportation Impact Fee Administrator, Planning and Development Department, City of Orlando).



TRANSPORT IMPACT FEES IN SANTIAGO In Chile, the use of “in-kind” exactions dates to at least the military government (1973-1990), with private developers responsible for providing the land for, and building and paving the streets within, their real estate projects. More recently, these “in-kind” exaction requirements have been formalized and expanded through regulations (such as Urban Impact Study requirements and Roadway and Urban Transport System Impact Studies; see, for example, (16)). At the same time, explicit financial exactions, or impact fees, have also come onto the scene, beginning in the early 1990s when – in the face of rapid land development pressures and the lack of alternative financing sources – two Municipalities in the Santiago Metropolitan Area (SMA) turned to an ad hoc impact fee scheme to charge real estate developers for major road connections to the rest of the urban area. More recently, authorities in Chile have turned to impact fee financing in response to massive real estate development plans in a rapidly suburbanizing area – Chacabuco Province – in the northern SMA (see (1)). Attempts are now being made to formalize the use of impact fees in Chile. Transport Finance System & Institutional Context Before assessing the current use of impact fees in the SMA, we first consider the institutional and financial context of their use. The SMA is located in the Metropolitan Region (RM), one of the 13 Regions that make up Chile’s national political landscape. Historically, for transportation and other planning purposes, authorities defined Greater Santiago as consisting of 34 Municipalities, with an urbanized area of roughly 60,000 hectares. More recently, at least 3 other Municipalities comprising Chacabuco Province (immediately to the North of Greater Santiago) have been effectively added to the Greater Santiago planning area. Chacabuco consists of over 200,000 hectares, of which 19,000 can be urbanized according to current regulations. Municipal Government. Municipal governments in Chile are locally-elected and each has a limited degree of autonomy in terms of raising revenue, building road infrastructure, and zoning local land uses. Municipal Governments levy local property taxes and charge annual vehicle registration fees (among many other fees not directly relevant to urban transport infrastructure finance) – the National Government sets the schedules for these fees. Municipal governments cannot issue bonds. As of 1994, 70% of all properties in Chile were exempt from property tax payments (17), primarily as a means to alleviate the tax burden on lower income property owners. Re-appraisal of property values has been sporadic, at best (18). Fifty percent of vehicle registration fees and 40% of property tax revenues remain with the Municipal government, with the rest passed on to the National Government for redistribution via transfers back to the Municipal governments (the degree of local government income disparity indicates the need for income redistribution: the richest Municipalities in the RM have 8 times more income per capita than the poorest even after these transfers; see (19)). Ignoring transfers, property taxes comprise the largest single source of municipal revenues, followed by vehicle registration fees – in 1997, at an aggregate level, Municipal governments collected US$560 million (76% of their total revenues) from property taxes and US$103 million (14% of their total revenues) from vehicle registration fees (20). Regional and National Governments. The Regional Government of the RM serves essentially as a deconcentrated arm of the National Government, with virtually no independent revenue sources, and a limited degree of responsibility for infrastructure construction and maintenance, typically carried out through regional secretaries of the National Ministries. Regional Governments depend almost entirely on revenue transfers from the National Government. The National Government, accounting for 95% of all public revenues raised in the country (20), plays a major role in local and regional finance and investments, not only through the transfers of funds, but also through investments by its relevant Ministries. Transport Finance. The author knows of no methodical assessment of Chile’s transportation finance system – transportation plans are generally not clearly linked to explicit financing plans. Nonetheless, several direct charges and one indirect user charge exist. The first direct fee is the National Government imposed excise tax (in addition to the national value added tax) on gasoline and diesel, which exists explicitly to finance roadway construction and maintenance. The second direct fee comes from the previously mentioned vehicle license fees charged annually by the Municipalities. The National Government’s infrastructure concessions program will introduce an additional direct fee – the tolls to be charged for roads built and operated by the private sector. Currently the SMA has one road infrastructure concession project under construction, with several more in advanced planning stages.



In 1994 the amount of revenues collected via the two direct roadway user fees in the 34 Municipalities of Greater Santiago equaled roughly US$290 million – 12% from vehicle registration fees and 88% from dedicated fuel taxes. This total updates an earlier estimate (in 21) to account for the fact that 50 percent of vehicle registration fees accrue directly to the Central Government for income redistribution (thus not intended to be user fees by authorities). Ostensibly, these revenues should go to financing roadway construction, maintenance and management, although whether the system “pays its way” remains unclear. Some degree of cross-subsidy implicitly exists, using road user revenues to finance the city’s urban heavy rail (Metro) capital infrastructure (the Metro covers its operating costs through revenues). Considering the transport system’s “external costs” (such as the estimated US$165 to $520 million annual air pollution costs; see (22)), then system users likely do not fully pay their way. Property tax revenues can be considered an indirect user fee. Some amount of property value derives from the access benefits of transport (i.e., “valorization”; although, some property values may also be adversely affected by related transportation externalities, such as air and noise pollution). So, we should consider that a portion of property tax payments charge for the access provided by the transport system. Assuming, for purposes of estimation, that property tax proceeds accrue roughly in proportion to GDP (in 1997 the RM accounted for 47% of Chile’s GDP (23)), then the RM collected $257 million of the US$560 million in property taxes collected nationwide. Since only 40% of property tax revenues stay with the Municipality, we can estimate that the RM’s Municipalities receive some US$100 million in annual property tax revenues. What share of those revenues might fairly be characterized as transport system user fees cannot be easily determined. Current Transport Impact Fee Use in Santiago Based on the general context outlined above, we now examine the use of impact fees in the SMA. As mentioned above, two rapidly-growing Municipalities in the SMA first turned to a form of impact fee to finance roadway infrastructure in the early 1990s. The Municipalities embraced impact fees because they faced strong real estate development pressures, but had limited resources to invest in the necessary concurrent roadway connections to the rest of the city – national law forbids that Municipalities issue bonds or otherwise go into debt and these relatively sparsely populated Municipalities had limited existing tax bases. Reportedly, revenues from this mechanism contributed to nearly all of the Municipalities’ major roadway development through the 1990s (for more detail, see (1)). These two early cases offered a precedent for financing transportation infrastructure in another part of the RM, Chacabuco Province, just north of the consolidated SMA. Historically an agricultural area, Chacabuco became the focus of intense real estate speculation throughout the early 1990s, but development faced at least one barrier – a lack of adequate transport infrastructure to link the various real estate projects together and to the rest of the urban network. As the areas of focus involved multiple Municipalities, planning responsibility fell to the National Government. By the late 1990s, the Ministry of Public Works, Transportation and Telecommunications (MOPTT) developed plans – using the transportation model EMME/2 – for a roadway network considered necessary to satisfy the estimated peak period transportation demand resulting from 14 proposed real estate projects (with a total of 40,000 households by 2010) in Chacabuco, accounting for each real estate projects’ size, location, socio-economic characteristics and subsequent travel demand. The identified roadway infrastructure included the 21 km Northeast Radial – to be developed under the government’s infrastructure concession program – and 41 additional kms of roadways, plus several major interchange upgrades. The non-concessioned infrastructure will cost an estimated US$81 million; the real estate projects will also contribute US$25 million to the concessionaire of the Northeast Radial, bringing the total amount to be financed through impact fees to US$106 million. Authorities determined each real estate project’s contribution towards that total cost based on peak demand generated – measured in vehicles per hour. Using each project’s peak demand, authorities developed a relative index of infrastructure “consumption” and used this index to allocate total infrastructure cost to individual real estate projects. The government agreed to not charge developers the cost share incurred due to the travel demand from low income housing. During negotiations, the government also ultimately agreed to assume direct financing responsibility of 39% of total infrastructure costs. Assessment of the Santiago Experience Owing to their relatively recent arrival to the infrastructure financing process in Chile, transport impact fees have not been subjected to any systematic evaluation or legal challenge there. This section provides a brief assessment of impact fee use to date in the SMA, based on the nine principles outlined in Table 1. 1. Guidance. Currently no official guidance for impact fee use exists in Chile. Absent any explicit legislative or regulatory justification, authorities essentially invented impact fee use “on-the-fly,” negotiating the



fees on almost a case-by-case basis with real estate developers, without clearly demonstrating their relationship to existing infrastructure financing sources. 2. Demonstrated Need. Authorities have deployed impact fees in Chacabuco Province in specific response to the need for transportation infrastructure arising due to rapid suburbanization. They have developed the plans for this infrastructure in accordance with accepted transportation planning principles in Chile. 3. Links to Exactions. The first agreement on impact fee use for Chacabuco Province (24) does not clearly differentiate between transport impact fees and exactions. On the one hand, the agreement specifies that the impact fees to finance the structural road infrastructure are separate from the local roadway infrastructure that each real estate project must provide. On the other hand, the agreement mixes exactions and impact fees, giving real estate developers the option to make their contributions by providing the land for relevant infrastructure (with the value of the land – determined through the law governing expropriations – counting towards that developer’s required overall contribution). An additional complication relates to the fact that transportation authorities have developed transport impact fees in virtual isolation from environmental impact fees (another form of exaction) being levied by environmental authorities on many of the same real estate projects due to transport pollution production (see (1)). 4. Rational Connection. Through the transport planning process for Chacabuco, authorities have shown the nexus between future real estate projects and the infrastructure requirements needed to support that growth. 5. Demonstrated Responsibility. Without the planned real estate projects, Chacabuco’s new structural road infrastructure would likely not be required. Complications inherent in modeling the system, however, make it difficult to firmly justify this assertion. For example, certain parts of the existing road network will be adversely impacted by the developments in Chacabuco, however it is not straightforward to analyze what would happen to the existing network without Chacabuco (i.e., where would those 40,000 households otherwise locate and what would be the subsequent network effects?). Furthermore, the current methodology for allocating costs (based on estimated network “consumption”) may not be the most appropriate approach. This may introduce a particular challenge when future developments in the area arise or when the currently planned developments add housing units beyond those considered in the existing plans (for an alternative approach, see Equation [1]). In addition, authorities have not paid adequate attention to the overall transport finance context; in other words, they have not considered other user charges that might arise in the future nor have they taken into account credits for infrastructure contributions via other users’ fees. 6. Clear Benefits. Authorities have clearly established the link between the expenditure of fees (development of infrastructure) and benefits to the individual projects. 7. Minimal Housing Cost Impact. Lillydahl et al. (8) suggest several ways to levy fees to minimize housing cost effects (see Table 1). In the Chacabuco case, the upfront payment of impact fees suggests that the fees will be passed on to homebuyers; but, the true effects of impact fees on housing prices requires more detailed analysis. The effects depend on the competitiveness of the regional housing market (and whether substitutes to the land types offered in Chacabuco exist), the predictability of impact fee levying, quality of subsequent services, buyers’ sensitivity to price changes, etc. (25). 8. Uniformity. Authorities have uniformly applied impact fees for each real estate project under the current agreement. Whether that uniformity will continue with future projects in Chacabuco (or other parts of the SMA) and whether the impact fees will be extended to non-residential uses remains to be seen. 9. Mitigate Adverse Impacts. The costs related to the transport demand from real estate projects’ low income housing units have been exempted from impact fees. This closely allies with the government’s continuing policy priority of expanding low cost housing supply. By introducing the fees in a rapidly suburbanizing area, the government may also positively affect its ostensible policy of controlling urban expansion, although the true impacts on urban and regional form and efficiency require more study. The effect of impact fee use on government attempts to introduce congestion pricing remains to be seen. Table 2 presents a summarized judgment of impact fee use in the SMA to date.



Recommendations for Improvements Being relatively ad hoc, impact fee deployment in the SMA seems only partially defensible when measured according to principles for guiding impact fee use that were developed based on U.S. experiences. A comparison to those principles (see Table 2) suggests several recommendations to improve impact fee use in Chile. First, authorities need to establish over-arching legal guidance on the use of transportation impact fees in Chile. Such legal guidance will also help ensure a uniform approach in future applications. Second, transportation impact fees’ relationship to related instruments, particularly environmental exactions, needs to be clarified. Third, transportation impact fees need to be more clearly understood within the broader transport system finance context, to clarify the difference between impact fees and other user charges and the potential need for impact fee credits. The current plans for the government to subsidize 39% of the relevant infrastructure costs in Chacabuco may be considered a de facto credit, however this subsidy resulted from negotiations with real estate developers, not an explicit consideration of other user fees. Fourth, the ultimate incidence of transport impact fees in the SMA case needs to be better understood – who will ultimately bear the burden (i.e., the owners of undeveloped land, new or existing residents, landowners) of these fees? Fifth, the extension of impact fee use to non-residential land uses should be considered. Finally, authorities need to assess the effects of transport impact fees on other public policy goals and their potential integration with other relevant instruments. For example, if authorities ultimately formalize impact fee use across the SMA as an efficient form of financing roadway infrastructure, they might consider a scheduled fee system, with the impact fee varying according to the development zone and type and anticipated impact on the city-wide transport network (see (15)). LESSONS FOR OTHER DEVELOPING COUNTRIES As discussed above, Altshuler and Gómez-Ibáñez (6) identified several factors leading to the rise of impact fee use in the United States. Of these, concerns regarding infrastructure shortfalls and the relative reduction in financing available likely combine as the greatest single factor contributing to the arrival of financial exactions in Chile. Without bond issuing authority or other means of debt financing, two SMA Municipalities turned to impact fees to finance roadway infrastructure in the early 1990s. More recently, the national government also turned to impact fees to finance large-scale regional road infrastructure, ostensibly because it felt that private real estate projects should pay “their own way.” The Chacabuco case marks the first steps towards formalizing impact fee use, although it is still too early to comment on the outcome. Many cities in the developing world face challenges similar to the SMA: rapidly increasing travel demand and ongoing urban expansion, linked to demographic changes, economic growth, motorization, etc. For financing transport infrastructure improvements in already built-up areas, some developing countries already have experience with instruments related to impact fees (i.e., valorization). For financing infrastructure expansion in direct response to new real estate projects, impact fees offer a promising mechanism since welldesigned fees not only promise to supply financial resources but also to serve as a growth management tool (by ensuring adequate capital facilities exist). The progress to date in Santiago – combined with general lessons drawn from the considerable U.S. experience – provides some indication of impact fees’ potential applicability in other developing countries. Among the most important lessons that may be drawn from the Santiago case: 1.

2.

Administrative capacity and growth management controls. This is perhaps the most important prevailing condition necessary for the use of impact fees in any city (2, 26). In Santiago’s rapidly growing Chacabuco Province, authorities have been somewhat successful in regulating real estate project development. In response to massive land speculation in the Province during the 1990s, the government enacted new land use plans and development regulations for the area. A relatively effective building permit approval process and recently revised transport impact study requirements (16) helped authorities ensure some management of the development pace and form. Since the area under development crossed local government boundaries, the national government essentially had jurisdiction; not only does the national government have the strongest technical capacity for transportation planning in Chile, but its participation may have defused any potential intra-municipal competition that might have subverted impact fee use in the area. Furthermore, the national government likely served as a stronger counterpart in negotiations with real estate developers. Government concerns with “efficiency” and “fairness”. Confronting rapid urban expansion and rapid growth in transportation demand in the face of many other competing demands for scarce public resources, Chilean authorities have consistently shown a predisposition towards self-financing of transportation infrastructure and services, when viable and equitable (i.e., the infrastructure concessions program, self-



3.

4.

financing Metro operations, and pending plans for implementing road congestion pricing). As such, Chilean authorities’ claim to be unwilling to subsidize new real estate developments with roadway infrastructure is not surprising. Such a perspective is necessary to move forward impact fee use. Unintended and unanticipated consequences. Chilean authorities have attempted to reduce impact fees’ potential negative effects on affordable housing provision by exempting these properties from impact fee payments. Authorities have, however, paid less attention to other potential effects of impact fee deployment, such as the potential risk of displacing growth to other parts of the metropolitan area and the relation of impact fees to other user charges. The public finance context. For the sake of clarity to developers, potential homebuyers, other public authorities and the public at-large, authorities need to deploy impact fees within a clear and transparent public finance framework. This would help to justify the fees, to understand their relationship to other charges, and to calculate any credits (see equation 1). Unfortunately, clear rules of public finance and clear transportation system budgets tend to be the exception in both developing and industrialized world cities.

The Chilean experience, similar to the U.S.’s early experience, suggests that impact fee use can precede explicit legal authorization (i.e., “enabling” legislation). However, also similar to the U.S. experience, Chile’s impact fee system may wind up in the court system. If other countries also opt to undertake ad-hoc impact fee application they would do well to consider the general lessons from the Santiago case and proceed following the framework for analysis presented here (Table 1). Finally, one should keep in mind the limitations of the SMA example. As the capital of an uppermiddle-income developing country (with an estimated GDP per capita of approximately US$5,000 in 2000), Santiago does not necessarily represent a “typical” developing country city. Many cities may not have the same administrative capacity evidenced in Chile, such as well-established travel forecasting skills and transportation planning procedures, and a basic degree of growth management capabilities. Without the latter, additional land owners may simply “free-ride” on infrastructure financed by developers paying impact fees (even in the Chacabuco case, this risk exists). For cities with unclear land titling, impact fees will face considerable barriers. However, in such areas, particularly parts of cities with large “squatter” settlements, impact fees are inappropriate since strong equity arguments can be made against charging them for basic access connections. For most developing countries, impact fees hold the most promise for cities with concentrated areas of largescale new real estate developments, where few alternative development sites exist (either due to actual space limitations or effective growth management capabilities). Impact fees do not offer a panacea to the transport finance and growth management challenges developing country cities face. But the Santiago case suggests their potential. To further understand impact fees’ role in the developing world, future research and applications might explore the potential to structure impact fees to promote “transport-efficient development” (by allowing for reduced fees for real estate developments that will have a demonstrable reduction in travel demand) and/or to finance public transport infrastructure (i.e., dedicated busways). In the end, however, we cannot expect impact fees to solve such challenges as housing provision and employment needs (26) – issues on the forefront of the developing world’s urban challenges. CONCLUSIONS Impact fees have been used for nearly 30 years in the United States to finance various types of urban infrastructure. Drawing from the U.S. precedent, this paper has evaluated the growing use of impact fees for financing roadway infrastructure in Santiago de Chile. This evaluation leads to several suggestions to improve Santiago’s current system, including: providing overarching guidance; developing more explicit links to other forms of exactions; establishing a clearer relationship to the overall urban transport financing system; undertaking theoretical and empirical work to assess the ultimate impacts on, for example, housing affordability; and, ensuring that the approach will be uniform in time across the city. While in need of refinements, Santiago’s use of impact fees offers clues to their potential in other developing country cities. At a minimum, impact fee use requires government technical capacity and a predisposition to such instruments, the availability of growth management controls, a clear understanding of the related public finance system, and care to ensure the avoidance of negative consequences on other public policy goals.



ACKNOWLEDGEMENTS This paper builds partly on research originally conducted with support from the Lincoln Institute of Land Policy. The author wrote this paper while a Research Associate at MIT’s Laboratory for Energy and the Environment. He would like to thank Raul Barrientos, with the Coordinación General de Transporte Santiago, for generously providing information and comments. REFERENCES 1. Zegras, C. Mecanismos Financieros Aplicados Al Desarrollo Urbano Como Consecuencia De Sus Impactos En Los Sistemas De Transporte: El Caso De Santiago De Chile. Tranvía: Revista de Transporte, No. 13, 17 October, 2001 (http://www.revistatranvia.cl). 2. World Bank. Cities on the Move: A World Bank Urban Transport Strategy Review. Washington, DC, draft of 17 October 2001. 3. Small, K. Urban Transportation Economics. Harwood Academic Publishers, Chur, 1992. 4. Gillen, David. Efficient use and Provision of Transportation Infrastructure with Imperfect Pricing: Second Best Rules. The Full Costs and Benefits of Transportation: Contributions to Theory, Method and Measurement (D.L Greene, D.W. Jones, M.A. Delucchi, Eds.). Springer, Berlin, 1997. 5. Brueckner, J. K. Infrastructure Financing and Urban Development: The Economics of Impact Fees. Journal of Public Economics, 66, 1997, pp. 383-407. 6. Altshuler, A. & J.A. Gómez- Ibáñez. Regulation for Revenue: The Political Economy of Land Use Exactions. The Brookings Institution and the Lincoln Institute of Land Policy, Washington, DC & Cambridge, MA, 1993. 7. Nelson, A.C. Development Impact Fees: Introduction. Journal of the American Planning Association, Winter, Vol. 54, No. 1, 1988, pp. 3-6. 8. Lillydahl, J., A.C. Nelson, T. Ramis, A. Rivasplata, S. Schell. The Need for a Standard State Impact Fee Enabling Act. Journal of the American Planning Association, Winter, Vol. 54, No. 1, 1988, pp. 7-17. 9. Smolka, M. and D. Amborski. Value Capture for Urban Development: An Inter-American Comparison. Lincoln Institute of Land Policy, Working Paper, Cambridge, MA, 2000. 10. Cervero, R. and B. Susantono. Rent Capitalization and Transportation Infrastructure Development in Jakarta. Review of Urban and Regional Development Studies, Vol. 11, No. 1, March, 1999, pp. 11-23. 11. Smolka, M. and F. Furtado. Lessons from the Latin American Experience with Value Capture. Land Lines. Newsletter of the Lincoln Institute of Land Policy, Cambridge, MA, July 2001. 12. Bahl, R.W. and J.F. Linn. Urban Public Finance in Developing Countries. Oxford University Press, New York, 1992. 13. Auerhahn, E. Implementing an Impact Fee System: Ten Years of Experience in Broward County, Florida. Journal of the American Planning Association, Winter, Vol. 54, No. 1, 1988, pp. 67-70. 14. Nicholas, J.C. and A.C. Nelson. Determining the Appropriate Development Impact Fee Using the Rational Nexus Test. Journal of the American Planning Association, Winter, Vol. 54, No. 1, 1988, pp. 56-66. 15. City of Orlando Applicability of Vehicle Miles of Travel to Transportation Planning. Transportation Planning Bureau, Orlando, Florida, 1998. 16. Secretaría Regional Ministerial de Transportes y Telecomunicaciones – Región Metropolitana. Manual de Procedimientos y Metodología de los Estudios de Impacto sobre el Sistema de Transporte Urbano EISTU Región Metropolitana, Santiago, 2001 (http://www.mtt.cl/rm/web/sivu/index.htm). 17. Asociación Chilena de Municipalidades (ACHM). Manual No. 4: Finanzas Municipales. Santiago, 1995. 18. Youngman, J. & J. Malme (1994) An International Survey of Taxes on Land and Buildings, Kluwer Law and Taxation Publishers, Deventer, The Netherlands. 19. Subsecretaría de Desarrollo Regional y Administrativo. Democracia Regional y Local. Ministerio del Interior, Santiago, July, 2000. 20. Barra, P. & M. Jorratt. An Analysis of the Chilean Tax System. Division de Estudios, Servicio de Impuestos Internos, Santiago, October, 1998. 21. Zegras, C. and T. Litman. An Analysis of the Full Costs and Impacts of Transportation in Santiago de Chile. International Institute for Energy Conservation – Latin America, Santiago, 1997.. 22. Zegras, C. and T. Litman. Cost Estimates of Transport Air Pollution in Santiago, Chile. Transportation Research Record 1587. TRB, National Research Council, Washington, D.C., 1997, pp. 106-112. 23. Banco Central de Chile. Anuario de Cuentas Nacionales de Chile. Santiago, May, 1999. 24. Secretaría Regional Ministerial de Obras Públicas - Región Metropolitana. Acuerdo Marco que Establece las Bases y Principios sobre Mitigación de Impacto Vial en la Provincia de Chacabuco, de la Región Metropolitana, Santiago, 7 September, 2001. 25. Singell, L. & J. Lillydahl. An Empirical Examination of the Effect of Impact Fees on the Housing Market. Land Economics, Vol. 66, No. 1 February, 1990, pp. 82-92.



26. Nicholas, J.C., A.C. Nelson, and J.C. Juergensmeyer. A Practitioner’s Guide to Development Impact Fees. Planners Press, American Planning Association, Chicago, 1991.



List of Tables and Figures Table 1: Nine Principles for Guiding Impact Fee Deployment Table 2: Fulfillment of the Principles in the Chacabuco Case



TABLE 1 Nine Principles for Guiding Impact Fee Deployment Principle Explanation 1. Guidance

Providing clear guidance on the type of facilities eligible for impact fee financing and the conditions under which impact fees can justifiably be used as a supplement to existing financing sources. 2. Demonstrated Showing the need for impact fees by situating the revenue expenditure within the context Need of a capital improvement plan, itself directly linked to a comprehensive community-wide development plan. 3. Links to Clearly showing how impact fees relate to “in-kind” exactions (i.e., local street Exactions requirements). 4. Rational Demonstrating a “rational nexus” between the real estate project and the infrastructure Connection needs to serve it. 5. Demonstrated Showing that the need for new infrastructure results from the new development, not from Responsibility existing deficiencies. This requires an appropriate planning process and fair cost apportioning (see text for details) 6. Clear Benefits Clearly establishing the connection between expenditure of fees and benefits: 1) reasonable expectation that contributing projects will use the facilities; 2) facilities must be proximate and available in reasonable time. 7. Minimal Housing Designing the fees and payment schemes to minimize the ultimate impacts on housing Cost Impact affordability, i.e. through: avoiding exacting fee at permitting stage, delaying levy until project is occupied; allowing payment over time, at subsidized interest rates 8. Uniformity Assessing the fees on each development in a similar way. 9. Mitigate Adverse Considering the effects of the fees on other policy priorities (i.e., affordable housing, Effects industrial development). Source: Derived from (8).



TABLE 2 Fulfillment of the Principles in the Chacabuco Case Principle 1. Guidance 2. Demonstrated Need 3. Links to Exactions 4. Rational Connection 5. Demonstrated Responsibility 6. Clear Benefits 7. Minimal Housing Cost Impact 8. Uniformity 9. Mitigate Adverse Effects


Fulfilled? No Yes Partially Yes Partially Yes Uncertain Uncertain Partially
