Valorization charges as a method for financing urban public works : the example of Bogota, Colombia

Valorization Charges as a Method for Financing Urban Public Works: The Example of Bogota, Colombia SWP-254 World Bank Staff Working Paper No. 254 March 1977 FILE COPY This paper is prepared for staff use. The views expressed are those of the pU B LIB ithors and not necessarily those of the World Bank. r epared by: William A. Doebele (Consultant) PUB with Orville F. Grimes, Jr. (Staff) D N OT H 3881.5 velopment Policy Staff ..W57 __W675 REMOVE ~- W67 no.254   This paper is for staff use. The views expressed are those of the authors and not necessarily those of the Bank. WORLD BANK Staff Working Paper No. 254 March 1977 VALORIZATION CHARGES AS A METHOD FOR FINANCING URBAN PUBLIC WORKS: THE EXAMPLE OF BOGOTA, COLOMBIA This paper provides an analysis of the financing of street improvements, sewer extensions, and other urban services by valorization charges, a system of taxation by which the cost of these services is allocated to neighboring properties in proportion to the benefits conferred. The experience of Bogota, Colombia illustrates the impacts of this system as a means of facilitating the provision of urban public services in developing countries at a rate more nearly in line with urban growth. On the basis of the Bogota experience, it is possible to suggest that valorization could constitute an improvement in the fiscal capability of many major cities to install infrastructure benefiting families at all income levels. Particularly if they could be tied to a reasonably well organized training program and effective technical supervision during the early years of operation, valorization programs do not appear to be beyond the administrative capacity of many developing countries. Prepared by: William A. Doebele (consultant) with Orville F. Grimes, Jr. (staff) Development Policy Staff   1. INTRODUCTION AND SUMMARY 1.1 This report analyzes conceptually and empirically the financing of urban improvements by valorization charges in Bogota, Colombia, and assesses the usefulness of this system as a means of reducing capitalization requirements for the instal- lation, expansion, and improvement of urban public facilities in developing countries. 1/ Such charges attempt to make urban services in cities of developing countries largely self-financing, thus reducing the burden on the system of general municipal taxation. By diminishing these financial constraints, valoriza- tion has the potential for these cities to facilitate the provi- sion of urban infrastructure at a rate and rhythm commensurate with their rapid growth. 1.2 The concept of valorization, or the assessment of neighboring property owners for urban improvements, is not new. Indeed, it has been traced at least as far as the Roman Digest of the sixth century, and variations of the concept appear in the legislation of most European countries and in the United States in the form of "special assessments." Colombia, however, 1/ The research for this report was carried out on a mission to Colombia undertaken as part of research project No. 670-98 ("Urban Land Taxation and Control") and led by Mr. Grimes. Mr. Doebele subsequently drafted the body of the present paper, which Mr. Grimes has revised for the present publi- cation. The authors benefited from comments on an earlier draft by Johannes Linn.  has perhaps used valorization techniques more extensively than any other developing country. 1/ 1.3 Within Colombia, attention has been focused on Bogota for several reasons. Not the least of these is its rapid phvsical and demcor-Epb ic expansion, particularly since the 1950s, and resulting demands for new services. Furthermore, as a national capital with a special administrative status, Bogota has had to plan its growth so as to internalize and reconcile the interplay among various levels of government. Its experience in doing so with valorization extends back more than five decades, a record which should prove instructive to other countries considering the adoption of valorization techniques as public policy. 1.4 Valorization is a system of taxation by which the cost of public works is allocated to neighboring properties in proportion to the benefits conferred. Since valorization depends on the availab:ility of liquid assets to pay charges, its application is less straightforward in lower income residential areas, in which higher property values resulting from public works may have to be realized through sale of the house in order to pay the charge. On the other hand, a variant of valorization, which encourages prepayment of 30% of costs, eliminates overhead charges, spreads installment payments at 1/ For a recent review of the experience of other Latin American countries with betterment collection, the interested reader is referred to Jorqe Macon and Jose Merino Manon, Financing Urban and Rural Development through Betterment Levies: The Latin American Experience (New York: Praeqer, 1977).  -3- low interest, and limits total charges collectible has been a considerable success in paving local streets. 1.5 The chief strength of valorization in Bogota is the flexibility shown in determining the distribution of benefits from projects under widely varying conditions, so as to arrive at a reasonably equitable distribution of charges. Adminis- trative problems have often been a major disadvantage of valorization and other forms of special assessment. The Bogota experience suggests that it is possible to maintain a highly flexible approach, yet one which is basically equitable, to deal with the central problem of distribution of the valori- zation charge. The methodology of calculation, using a "Basic Coefficient" and "Coefficient of Form" modified by any other coefficients judged appropriate, seems both adequate and capable of evolution into new forms to meet new conditions without loss of continuity with other projects. The chief problem in Bogota/ has been the lack of a vigorous policy of collection, though enforcement procedures and standards appear to have recently been tightened. At least with respect to allocation of the charges, the administrative costs of the system do not seem to be excessively high. 1. 6 Valorization apnears to have important financial benefits to the municipalitv, in addition to some positive allocative and distributive effects. TWhile it has not: signi- 'icantly altered income disparities and has probably increased the nrivate return to land development in the cities affected,  - 4 - it seems also to have been instrumental in enlarging the choice and accessibility of lower income groups to jobs and to resi- dential building sites. 1.7 This study concludes that the system, while not without problems, could constitute an improvement in the fiscal capability of many major cities to install infrastructure at a rate more nearly commensurate with the pace of urban growth. Such programs are in fact more feasible politically and administratively under conditions of rapid urban growth, rising land values, and infla- tion, conditions prevailing in most major cities of developing nations. The operation of a valorization program does not appear to be beyond the administrative capacity of many developing countries, particularly if it could be tied to a reasonably well organized training program and effective technical supervision during the early years of operation. In this regard the Colombian system, with its detailed descriptions prepared for each project and long experience in administration, is a particularly suitable one for study.  2. VALORIZATION IN BOGOTA: THE SETTING A. Antecedents 2.1 While some historians have traced references to the idea of valorization in Colombia to as early as 1599, valoriza- tion in its modern form began with Article 3 of National Law 25 of 1921, which defined the concept as: "...a contribution from the real Droperty which is benefitted by the execution of works of local public interest, such as the cleaning and channelization of rivers, the construction of dikes to prevent flooding, the drying of lakes, swamps and wet lands, irrigation and other analogous purposes, which contribution is to be exclusively designated to meet the costs required by said works." 1/ In 1936, these powers were conferred on the City Council of Bogota. National Law 1 of 1943 extended the scope of valori- zation from the flood and irrigation areas defined in 1921 to all "works of public service," and removed the limitation of recovery of charges to the costs incurred. The national legis- lation of 1943 also gave municipalities almost complete freedom to determine the methods by which benefits should be calculated, and the form of administration they wished to use. A year later the highest judicial body on administrative matters made the important rulinq that valorization charges could be assessed and 1/ In this report all translations of legislation and other documents are by the senior author unless otherwise noted. Laws and decrees pertaining to valorization are provided in Rafael Soto Pinedo, La Contribution Nacional de Valorizacion (Bogota: Editorial Temis, 1974).  - 6 - collected as soon as a plan and budget for a project had been approved and before actual work had begun -- thus (in theory at least) reducing substantially the amount of municipal interim financing necessary. 1, 2.2 A national Legislative Decree of 1966 (No. 1604) enlarged the definition to "... works of public interest which are executed by the Nation. the Department, the Special District of Bogotl, the Municipalities, or any public entity, and which benefit real property..." (Article 1; underlining added). The apparent meaning of this decree is that in Colombia any public works project can be made the subject of valorization, so long as it results in an increase in property values. In 1966 legislative action also provided that if a project subject to a valorization tax in an urban area was in fact paid for out of national or provincial funds, the municipality (even though it had paid none of the costs) could still collect the tax, if it acted within two years and had the consent of the other level of government concerned. In such a case the collections would go into the revolving fund or pool used tco finance all valori- zation projects in that city. 2/ 1/ See Carolina Botero, "The Valorization Tax in Colombia as Applied at the Municipal Level" (mimeo, Department of City and Regional Plannina, Harvard University, May 1975), pp. 6-7. 2/ This facet of the 1966 decree was pointed out by Botero, op. cit., p. 7.  -7- 2.3 Bogota began using the system for street construction in 1944, and in 1958 passed a comprehensive municipal "acuerdo" (ordinance) defining the procedures in great detail (Acuerdo 41/1958). Although this is still the basic legislation for the citv's operations, subsequent ordinances have enlarged its scope and introduced new procedures. In 1959, for example, Acuerdo 85 extended a variant of valorization to the redevelopment of deteriorated parts of the city, although in fact this power has not yet been used, the category instead having been converted into a channel by which other budgetary funds have been mobilized for the construction of an office complex for the city adminis- tration. In Acuerdo 6/1963, presumably in response to criticism that road improvements were primarily benefitting the wqell-to-do portions of the city, a Local Paving procedure was created to help low income neighborhoods improve their streets. It sought to retain some of the advantages of valorization to the public treasury while easing the burden on property owners by providing for payments for as long as 36 monthly installments at 3% annual interest. 1/ As figures presented later in this report indicate, this has become a popular program. 1/ Currently the payment period is 30 months. See J. Linn, "Valorization in Bogota, Colombia: Organization and Financing" (draft, World Bank; August 1976), p. 27.  -8- 2.4 The coverage of valorization was further extended when in 1968, the establishment of parking facilities ("Estacionamientos," Acuerdo 46/1968)and green spaces ("Zonas Verdes y Comunales," Acuerdo 45/1968) was :included. In 1973, a small program in zones of historical interest was begun, and a so-called "Plan Muelas," to construct loading bays and widen streets in congested parts of the city, has also been added (see also paragraph 2.16). 2.5 Given the broad 1966 definition of valorization as including "...works of public interest ... which benefit real property ..." (see paragraph 2.2), it is possible that Bogota can in the future extend valorization further than the seven areas just mentioned (streets, local paving, redevelopment, parking facilities, green and communal spaces, zones of historical interest and loading bays). Indeed, such expansion is contemplated in the enabling legislation (Acuerdo 19/1972) of the agency chiefly responsible for undertaking valorization-financed improvements in Bogota, the Institute of Urban Development (IDU). B. The Institute of Urban Development (IDU) 2.6 Valorization was traditionally administered by a depart- ment within the BogotA Special District. 1/ In 1972, however, it 1/ As is often the case with capital cities, Bogota' enjoys a special status. Known technically as the "Distrito Especial de Bogotd," it has many of the powers of a "departamento" (or province). Its present boundaries include almost all of the urbanized area, with room to spare, except at a few points at its northern and southern extremities where urban growth has overspilled its limits to a minor degree.  - 9 - became necessary to create a special institution to carry out the complex "Avenida de los Cerros" project, which involved not only the construction of a long, wide highway in the mountains behind the city, but also the building of a series of public service installations, especially in its southern portions, whose aim was to benefit low income groups. Acuerdo 19 of 1972 therefore created a semi-autonomous Institute of UrbeLn Develop- ment (IDU) to execute this and other major public works. 1/ Simultaneously, the entire operation of the Valorization Department was shifted to the IDU. 2/ C. Valorization for Public Services not Directly Under the Jurisdiction of the Bogota Special District 1. Water and Sewer Authority ("Empresa de Acueducto y Alcantarillado de Bogota," or EAAB) 2.7 Water and sewer facilities in Bogota (except for some site-related installations whose construction is the responsi- bility of the land developer) are planned and constructed by a separate enterprise known as the EAAB. It is a "pub:Lic entity" with powers to impose valorization, and has its own Department of Valorization for distributing the cost of major trunk lines to 1/ Due to the public controversy surrounding it, the Avenida de los Cerros project was terminated after the change in national administrations in 1974. 2/ Administrative details of the IDU are covered in Linn, op. cit., pp. 3-6 and 8-10.  - 10 - adjacent property owners. However, all actual collections of valorization charges are done by the District (now the IDU), for which it receives a 20% commission. The major undertaking of the EAAB now is the execution of a three stage Sewerage Master Plan, further discussed in paragraph 2.13 below. 2. Ministry of Public Works 2.8 While most of the Ministry's road building activities are outside urban areas, it has constructed one very important work in Bogota--a divided highway from the international airport to almost the center of the city. As in the case of the EAAB, the so-called Autopista Eldorado was planned and executed by the Ministry, but collections have been turned over to the IDU on 20% commission. For reasons which will be discussed later, the collection experience has been a poor one. As far as is known, the Ministry is not planning anv more major works in the city which would involve valorization. D. The Record of Experience 2.9 Until the administration of Mayor Virgilio Barco (1967-1971), valorization was not a major program in the BogotI Special District. With the acceleration of civic orojects which marked his administration, however, expenditures  - 11 - on valorization projects in the single year 1968 were greater than in the entire eight years of 1959-1966. From 1967 to 1973, nearly five times as much money was expended for valorization projects as in the preceeding eight years (Table 1): Table 1: VALORIZATION EXPENDITURES, 1959-1973 (thousands of current pesos) Year Expenditure Year Expenditure 1959 19,781 1967 172,559 1960 19,777 1968 259,540 1961 26,570 1969 216,583 1962 18,342 1970 150,229 1963 28,726 1971 141,761 19E4 42,277 1972 143,195 a/ 1965 40,750 1973 122,296 a/ 1966 55,253 Total, 1959-1966 251,476 Total, 1967-1973 1,206,163 a/ Excludes IDU elemen.t for comparability. Sources: Bogota, Department of Valorization, Summary Budget 1959-1971; Summary Budget 1965-1972; 1973 Budget. 2.10 The Local Pavircja procedure of 1963, which began operations the following year, is in effect a special simpli- fied version of valorization, applying only to smaller roads and streets. 1/ Unlike larger projects, there is no 20% 1/ i.e., those with a maximum support capacity of 8,000 to 24,000 pounds. For further information on the Local Pavin- program, see Linn, op. cit., pp. 25-27.  - 12 - charge for administration in the budget of the Local Paving Fund, and no provision for participation of property owners. Furthermore, charges are distributed pro rata on the basis of meters of frontage of each property, instead of the more elaborate methods used for regular valorization, and are limited to 5% of the assessed value of the property. 2.11 Although Local Paving started as a very small program (only 4.9% of total expenditures in its first year), by 1973 it amounted to a fifth of the total and 40% of the outlay for streets and avenues, the oldest and most important of the valorization programs. If the comparison between these two programs is made in terms of investment in actual works in 1973--which would seem a truer measure of their physical impacts on the city--the relative importance of Local Paving is even greater. Looking at all six programs active in 1973 in this way, the following picture emerges (Table 2):  - 13 - Table 2: EXPENDITURES ON VALORIZATION PROGRAMS, 1973 (thousands of pesos) Of which: Expenditure Devoted to Land Total Expenditure and Construction Investment Amount Percent Amount Percent Streets and Avenues 61,792 50.5 35,695 40.6 Local Paving 24,496 20.0 24,109 27.4 Redevelopment 26,394 21.6 20,094 22.9 Parking Facilities 7,767 6.4 7,752 8.9 Green Spaces 1,700 1.4 -- -- Historic Zones 147 0.1 147 0.2 122,296 100.0 87,797 100.0 Sources: Bogotl, Department of Valorization, 1973 Budget; Project Investments 1968-1973. Table 2 suggests that the Streets and Avenues and Green Space programs are heavily burdened with debt service, resulting in budgetary charges disproportionate to their physical impact. Thus, Streets and Avenues comprise half of the total 1973 expenditure for these six categories, while representing on:Ly about 40% of investment in land or construction. Liquidity and collection issues in valorization will be taken up later in more detail. 2.12 As will be noted from Table 2, the other applications of valorization are relatively small. UTse of the system for the redevelopment of blighted areas "'Redesarrollo") did not appear in budgets until 1967, and as of 1974 involved only one project-- the construction of the District Administrative Center. The parking program begun in 1969 has also been small and confined  14 - to some badly needed facilities in the congested central part of the city. Green spaces and historical zones have also been, as the table suggests, of minor importance. 2.13 Sewers, however, are another matter. The Sewerage Master Plan calls for a massive investment in increased capacity, maintenance, and renovation in the metropolitan area, divided into North, Central and South Sectors. Valorization-financed works for the North Sector alone will amount to more than 153 million pesos (over US$4 million), which is greater than the 1973 expenditures for all other projects. 2.14 Conceptually there is no reason why valorization could not be applied to any public project which adds substantial values to adjoining land. Green or open spaces, often judged to create "spill-over" values which are very hard to estimate, have been financed in this way, although not yet on a scale sufficiently large to judge the system's effectiveness. 2.15 In terms of the physical layout and functioning of Bogota, the impact of valorization has been quite substantial. Virtually all or tne impor_anc LUie durifiy 1i7/-9J-73 were constructed through its application, and now it also covers construction costs under the Sewerage Master Plan. It is difficult to imagine this scale of public works being achieved through conventional methods of public finance in a country at Colombia's stage of development.  - 15 - E. The Valorization Process 1. Selection of Projects 2.16 All projects for the improvement of streets and avenues are part of the "Plan Vial," a comprehensive plan prepared by the District Planning Office which also includes the parking facilities program. As mentioned earlier, sewers are also being constructed subject to their own comprehensive Master Plan. It is not known whether a similar long range plan has been made for green spaces, but this is, in any case, a minor program. Valorization for "redevelopment" has been limited to the construction of the Administrative Center, and comprehen- sive planning for the older parts of the city has yet to be undertaken. However, an approach through valorization financing has been made to the most serious problem of the older city-- traffic--by the so-called "Plan Muelas." A "muela" is a parking bay or cut into the sidewalk so that vehicles which must stop (such as buses and delivery vehicles) mav do so without blocking the flow of other traffic. In narrow streets, this can become an important factor in the effective functioning, and hence the property values, of older commercial-administrative areas. 2.17 The major valorization projects which are not subject to comprehensive planning are Local Paving. These are initiated by the Junta de Accion Comunal, working in collaboration with the Department of Communal Action of the IDU, in the various  - 16 - low income neighborhoods (barrios). The barrio Junta will ordinarily raise from its citizens 30% of the projected cost. When this occurs, IDU must immediately carry out the work, and the remaining 70% will be paid on 36 monthly installments. (See also paragraph 2.3 above). 2.18 All projects must be approved by the District Planning Board and the Board of Directors of IDU. Projects involving the seven programs of the IDU must be authorized by the District (City) Council, except for street works of under 200 meters in length, which may be authorized by the Mayor. Projects by the EAAB are, of course, authorized by its own board. 2. Technical Preparation 2.19 After approval is secured, the staff of IDU prepares an "Explanatory Memorandum of Provisional Liquidation" for each project. This document presents in a concise and complete form virtually all of the essential information about a project: its scope and objectives, its budgeted cost, and the system by which the cost is to be distributed by valorization. The standard sections of- these reports are worth noting: a. Authorization: A brief summary of the legal basis of the project--the date and number of the Resolution of the Board on Valorization (Board of Directors of the IDU) and of the authorizing ordinance (acuerdo) of the District Council.  - 17 - b. Zone of Influence: Normally, the area over which valorization charges are to be spread is established by the District Council (presumably on the recommendation of the Board of Valorization) in the same ordinance that authorizes the project. A frequent formulation is "all properties located within two parallel lines 500 meters on each side of the axis of the street." However, in some cases different zones are outlined by which the Council will, in effect, draw a specific map. c. Representatives of the Property Owners: The law of valorization requires that the property owners in the Zone of Influence must be permitted to elect a "Representative" who oversees all the critical points of the administrative process, receives owners' complaints, and theoretically acts as a "watchdog" of the entire procedure. Representatives are paid for these efforts at a rate of 0.15% of total project cost. The formalities of the election of the principal representative and his alternate, from a list of candidates prepared by IDU, are quite detailed, and this section of the Memorandum certifies that they have been complied with and gives the names of the persons elected. d. Actual State of the Street: This section describes in a few sentences the conditions which make improvemen-t necessary. e. Description of the Work: This lays out the speci- fications of the project.  - 18 - f. Budget: This states the provisional budget, broken into (a) the cost of the land; (b) the cost of con- struction; (c) the cost of administration, always 20% of (a) plus (b); (d) contingency, generally 5-10% of (a) plus (b); (e) interest, equal to (a) plus (b) during the time period estimated to execute the project, presumably evaluated at the rate of interest at which money is currently available; and (f) the honoraria to the representative and alternate (0.15% of (a) to (e) for each). All of these together are the provisional cost which must be liquidated through valorization. g. '"Factorizacion": This section of the Memorandum deals with the distribution of valorization charges ir the project area, which are calculated by a set of coefficients more fully described in Section 4 below. When a final coefficient has been assigned to every parcel, the provisional budget described above is allocated among the parcels as outlined in Section 4. h. Statis-tical Analysis: The Memorandum concludes with a rather elaborate statistical analysis of the valorization charges, grouped by type of charge per square meter, average charge, lowest and highest charges, and so on. Together these give quite a detailed picture of the general pattern of charges for the project.  - 19 - F. Problems in Design and Application 2.20 Colombian legislation and practice on valorization, like similar programs elsewhere, have had to come to grips with two issues: Should the valorization charge be viewed as recovering a benefit otherwise bestowed gratuitously on the landowner by the State (that is, as recovery of betterment), or primarily as a way of increasing municipal solvency by recouping the actual cost of public investment? How is the principle of valorization to be applied in low income neighborhoods? Since the benefits from public investment attach to the land, not the owner, land values will increase in low-income areas as well as in others. In practice, however, low-income owners may find it difficult to pay a valorization charge, since they have little savings and no means of converting the increased land value into cash except by sale of their home. 2.21 Betterment Recapture or Cost Recovery. As mentioned, the original national legislation of 1921 limited valorization to the cost of the works undertaken. This limitation weLs removed  - 20 - in 1943, and Bogot's comprehensive Acuerdo 41/1958 provided that not only should valorization consider the cost of the works, but also the increase in values stemming from the project. Two evaluations were called for, one before construction and one after, to determine the total contribution which could be imposed. 1/ This amount could, however, be reduced to collect only the sums actual:Ly expended on the works, including necessary complementary works, indemnizations, and administrative costs. Under these provisions the Board of Directors of the Valorization Department apparently could choose which philosophy of recovery to favor for each project. 2.22 In practice, however, no case has been found in which full recovery was attempted during the eight years in which this provision was in effect. In national Legislative Decree 1604/ 1966, total charges were limited to direct project costs plus a maximum of 30% for administrative overhead, and in no case more than the benefits received. A quite similar formula in a Bogotg decree of 1967 states that collections should be limited to direct costs plus 20% administration, plus a "prudent percentage" for contingencies, all "within the limits 1/ Acuerdo 41/1958, A:rticles 74 and 75.  - 21 - of the beinefit that it (the project) produces to the real property which is to be obligated." 1/ 2.23 While the requirement of before and after evaluations-- obviously a difficult task--is still on the books, the actual practice since 1967 has been to assume that the cost of the project is always within the benefit produced, and to get on Uith the more immediate task of a reasonable allocation of the costs themselves. In this the city officials appear to feel with Rhoads and Bird that in a growing city "the forecast of benefit exceeding tax will be true in almost all cases." 2/ Informal and fragmentary studies by the IDU section charged with land acquisition support the thesis that charges seldom exceed project benefits, especially if inflation provides an additional cushion. Benefits are not ignored; indeed, calcu- lations of benefits--often detailed and elaborate--are the basis on which the project costs are allocated among individual properties. It is clearly, however, much less difficult to use a broad conception of benefits as a means of assigning relative costs than it is to specify a distribution of benefits conferred as the basis for a 100% recovery of them. Thus, valorization 1/ Acuerdo 5/1967, Article 1. 2/ W. G. Rhoads and R. M. Bird, "The Valorization Tax in Colombia: An Example for Other Developing Countries?," Land Economics 43 (November 1967), p. 406.  2 22 recovery is based primarily on the cost of the works, not on recovering all the benefits the project may be expected to confer. 2.24 Adjustment for Low Income Property Owners. The basic 1958 Acuerdo provided that the V7alorization Board would be able to exclude from payment the costs of physical construction (streets, sidewalks, utility networks, and so on) when the project was "in sectors of low economic level." 1/ On th1e other hand, the costs of land for the project, indemnities for damages caused, the 20% for administration, and the 0.15% for representatives could not be excluded. The 1967 statute did not change this provision, but it added another significant concept: "In each activity the Board of Valorization will determine the method that is to be followed for the distribution of the contributions of valori- zation, in accordance with the characteristics of the undertaking and the types of benefit, to the end that the resulting contributions are adjusted to the legal norms and principles of equity that control the contributions of valorization..." 2/ 2.25 In short, the Board would now seem to have two strings to its bow: (a) It may exclude from payment of a major share of project costs "sectors of low economic level;" and (b) In 1/ Article 76, Secticn 2. 2/ Acuerdo 5/1967, Article 2.  - 23 - determining the method of assigning the original contribu- tions, it may use a formula which lessens burdens on any low income areas which may be involved. By the older orocedure all properties must be burdened, and then an exclusion made. By the newer one, low-income characteristics are considered in assigning the original burden. While the two procedures can produce the same result, the second is administratively more efficient and is now being frequently used, as will be further discussed in Section 4 below. 2.26 The next three sections of this report will deal with three central technical problems of valorization: establishing the values of land and buildings, distributing valorizat.ion charges, and collecting the charges. Subsequent sections will assess the overall impacts of valorization and provide an appraisal of its effectiveness.  - 24 - 3. TECHNICAL PROBLEMS: ESTABLISHING ACQUISITION PRICES FOR LAND AND BUILDINGS A. Creation of Price "Contours" for Land 3.1 IDU uses the general Colombian system of costing land and building acquisiltions. Adopted by national law in 1939 and amplified by subsequent legislation, this system is in essence analogous to determining contours of elevation by topographic surveys. That is,as a first step the market prices of a certain number of "prototypical" points are established. The points are then connected by "isovalue" lines, building up a set of price "contours" which may then be read like a topographic map. Originally designed to determine cadastral values for tax purposes, the price contour system has been used in Bogota since 1972-73 and in other Colombian cities since 1963 as the basis for establishing individual property values prior to public acquisition. 3.2 Three steps are taken to determine the market values of the key points. First, interviews are conducted with property owners in the area to be affected. Up to 20 inter- views may be taken to determine a single point. Whether the results are adjusted for variations in the respondents' presumed knowledge of local conditions is not known. However, information useful for such a judgment, including experience in the real estate profession (if any) and length of residence, is collected  - 25 - during the interview. Second, evidence of sales is obtained from two real estate associations which keep records of sales of property by their members. Finally, a sample of market transactions in the area during the preceding 18 months is taken from sales investigations recorded by Notary Offices. Despite the likely presence of some straw sales or unde!rreporting, these figures do give another sense of the value range. 3.3 Next, by comparing specific bits of data with averages and calculating sampling errors, a confidence level is established for the estimate of key points. Data which lie outside the confi- dence interval are weighted accordingly. If further sampling is not required, the estimate becomes the officially adopted "price" for that location, and one of the key points for drawing the contours. When enough key points are established, the points and actual contours are placed on a detailed map of the area concerned. The values of all other properties are then read by linear interpolation, as one would do with an ordinary topographic map. 3.4 Since the prices for the key points and contours are for "prototypical" lots, the square meter prices of ot;her lots, even those on the same contour, are adjusted for differences in street access, frontage, shape, and other physical characteristics. For even further precision, different sizes of rfprototypical" lots are emploved for residential, commercial, and industrial property.  - 26 - B. The Value of Construction 3.5 While land costs are being determined by the methods described above, a simultaneous field study is made of construc- tion. In areas in which building is conventional, this process is relatively simple, and can be more or less limited to buildings which are actually expected to be acquired. A fairly elaborate point system has been developed for evaluating every element of construction, with values assigned to different construction materials and techniques, weighted in each case by whether their condition is "grood," "regular," or "poor." Overall age is also added as another modifying coefficient. 3.6 For non-conventional construction in low-income areas the procedure is more complex. In this case, a different set of construction categories is used, more suited to informal and auto-construction. Instead of three categories of quality (each with preassigned values) as with conventional construction, this form has four categories, only some of which are weighted in advance. Point-weightings for the remainder are left up to the field investigator to assign. 3.7 Furthermore, while the value of conventional construction is considered to be more or less uniform, it is felt that the value of informal construction tends to vary more among barrios. To convert the point totals decided upon to unit assessed values, therefore, separate tables are compiled for each affected barrio.  - 27 - Moreover, the assessed values established are adjusted for dwelling age and size. As one illustration, values are adjusted in certain neighborhoods for houses over four years old and larger than 100 square meters in size. 3.8 As has been noted, great efforts have been made to make the evaluation system comprehensive and fair. But while very careful in what it does consider, the system also at times appears to have insufficiently considered other elements of importance in low-income areas. The requirement that title to property be demonstrated before the whole system of compensation applies, however much it facilitates the acquisition process, can negate much of the sophistication of the above valuation procedure, since the system under which a great proportion of the low income areas of Bogota is urbanized frequently does not deliver clear title. 1/ Title can also be acquired by ten years' adverse possession under Colombian law, but this is clearly also difficult for low income families to document. A related problem is that renters, who often comprise over 25% of the population of 1/ For a detailed analysis of the so-called "pirate neighbor- hoods" ("barrios piratas") system in Bogota', see Georges Vernez, "BogotA's Pirate Settlements: An Opportunity for Metropolitan Development" (Ph.D. dissertation, University of California, 1973), and William A. Doebele, "The Private Market and Low Income Urbanization in Developing Countries: The 'Pirate' Subdivisions of Bogota," American Journal of Comparative Law (forthcoming).  - 28 - low income barrios, receive no compensation under the above system even though they may experience rent increases in a new location after incurring the cost of moving. Even owners who can establish sufficient claim of title, and receive compensation as outlined above, sustain moving expenses and all of the psycho- logical and financial hazards of relocation in a tight market. C. Preparing an Actual Offer 3.9 Once the values of land and construction are determined, a comprehensive worksheet is prepared for each property to be acquired. Proper title is verified; the worksheet calls for registration information for current and previous owners. A formal "Valuation Memorandum" is prepared, which is then trans- mitted to the appropriate office of IDU to begin negotiations. According to the officials interviewed, these offers are generally accepted by landowners so that compulsory acquisition procedures seldom need to be resorted to. Owners who question the value are informed of the procedures just outlined, and generally conclude that it would be very difficult for them to construct an opposing case as convincing as the evidence compiled by these methods. 3.10 The valuation system described above is similar in its essentials to those successfully used in many European countries for these purposes. Certainly its "cross-referencing" of infor- mation and attempts to record in writing and objectify all relevant  - 29 - data are attractive in themselves and some check, at least, on subjectivity. It is clear from the description given that there are still many points at which individual judgments are critical, as they indeed must be in a system of assessment. Nonetheless, the system overall would seem superior to practices (such as those of the US) that rely more heavily on the personal judg- ments of assessors, and produce more litigation since it is normally possible for those who have the resources to find assessors of equal qualifications and differing judgmenits.  - 30 - 4. TECHNICAL PROBLEMS: DETERMINATION OF THE ZONE OF INFLUENCE AND ALLOCATION OF VALORIZATION CHARGES A. Introduction 4.1 One of the most important technical problems in valorization is to apportion the total cost of the project in an equitable manner, siD that no resident will pay more than he receives in benefits. The distribution of benefits from a public improvement raises a host of theoretical and pragmatic issues, which the system of va:Lorization must somehow solve through bureaucratic and publicly defensible procedures. In Bogota the administration of valorization has approached these problems with a set of doctrines, defined partially by law and partially by tradition, but at the same time has shown great flexibility in applying the doctrines to specific cases. 4.2 The basic Bogota ordinance of 1958 provided one method to allocate valorization charges, which was to create a series of equally wide parallel zones along the project, and to assign a prescribed benefit to each. 1/ After a second adjustment to take account of variations in lot size and frontage, charges could be allocated. 4.3 To this rather inflexible method a 1960 statute added a second system. 2/ Its provisions call for a detailed study of each property, including especially the following ten factors: 1/ Acuerdo 41/1958. 2/ Acuerdo 70/1960.-  * 31 - (a) Lot size (b) Frontage in relation to area (c) Distance from the project to be undertaken, considered a measure of access (d) "The economic conditions of the road on which the property is located, such as destination, demand for transportation, accessibility, etc., in cases where these circumstances determine a greater or less attraction of the betterment emanating from the project" 1/ (e) Lot shape (f) Topography (g) Natural accidents, such as flows of water, swampiness, and "sources of unhealthiness" (h) Proximity of the property to sectors of low economic and social level (i) Inputs which the owner has voluntarily contributed to the project (j) The change in economic activity on the property brought about by the project. 1/ A literal translation. This apparently means that some properties, due to preexisting economic development, profit more than others.  - 32 - 4.4 Moreover, although this list would seem to be fairly exhaustive, an ordinance adopted in the late 1960s, at the beginning of the surge of valorization projects in Bogotl, stated that the Valorization Department was not bound by either of the above two methods, but could proceed "in accordance with the characteristics of the project and the nature of the benefits, with the end that the resulting charges are adjusted to legal norms and principles of equity." 1/ As will be discussed below, the concept of equal parallel zones is now almost never applied, although the zone of influence is freauently identified as running parallel to the project. The ten factors, however, have had a great impact on the way administrators think about valorization benefits, even though all ten are not applied in each case, and many new ones have been added. B. The Basic Coefficient 4.5 In the previous section, which dealt with establishing the price of land and buildings for acquisition, it was pointed out how the values of certain "prototypical" lots were first determined, and then other values from them. Somewhat analogously, as a rough first approximation at distributing total project costs 1/ Acuerdo 5/1967, Article 2.  - 33 - according to benefits, a "Basic Coefficient" is establishiedfor an "ideal lot" measuring 10 meters of frontage by 25 meters of depth. The estimated rise in the value of properties affected by the project is the measure of benefits used in calculation of the Basic Coefficient. Such value increases are first estimated for properties that will receive the greatest, and the smallest, benefits. They are expressed as pure numbers, frequently in the range of 3-30 to 100-12C0, though varying greatly from project to project. Generally, but not always, the distance of the parcel from the project determines its position in the range of benefits. A curve is then constructed between the highest and lowest figure (of which more below) so that values can be given to all properties receiving intermediate increases. While it "tastes" of a money value, the Basic Coefficient is carefully identified as a pure number, not pesos, for ease of understandina of its use in combination with other coefficients to give a relative weight to every parcel within the zone of influence. C. The Coefficient of Form 4.6 To adjust for any lot not exactly 10 x 25 meters, a second coefficient is invariably necessary. This Coefficient of Form, developed in a standard formula by the Geographic Institute of Colombia, deals not only with this problem but also with the effect of expropriating only a portion of a lot rather than the lot itself.  - 34 - 4.7 Conventional assessment practices of most countries hold that of two lots equal in size or external dimensions, the one with less street frontage will have the lower value. Thus, in the adjustment formula for lot shape .144 -.404 I = 3.9174 (F + 1) p , (1) where I is an "Index of Form," F is frontage, and P is the perimeter of the lot, the index, which typically varies between 0.9 and 1.3, is lower the narrower and the longer the parcel. Guidelines for use by nonspecialized personnel make the formula relatively easy to apply. 4.8 Expropriating a portion of a lot raises a question of equity common in almost all countries. If a part is taken, the owner receives compensation at market value for what is lost, plus a large increment in value on the remainder as a result of the project. The owner who loses an entire parcel, however, gets only the compensation which, if awarded at current use value, contains no share of the future increments. Valorization increases the equality of treatment in these situations by applying a further adjustment with the formula I* = 2Ib - Ia (2) where the adjusted Coefficient of Form I* is twice the post- project Coefficient less the original Coefficient. For lots  - 35 - not "cut" by the project,therefore, Ib and Ia will be equal, and consequently no adjustment of the original coefficient of form is made. D. Other Coefficients 4.9 Beyond these two standard coefficients, any number of other variables which might affect benefits may be added, simDly by expressing their presumed effects by coefficients varving around 1.0. Topography is probably the most common, but land use, socioeconomic composition of the neighborhood, and others beyond even the influences on value enumerated in paragraph 4.3 are frequently employed. As will be seen below, the possibili- ties are limited only by the circumstances of the project and the imagination of the administrator. E. The Valorization Charge 4.10 All of the other variables of benefit are then multiplied together with the Basic Coefficient to arrive at a value known as the Final Coefficient. Expressed on a per square meter basis, the Final Coefficient is an index of the estimated benefit received by the parcel in question relative to benefits on every other parcel in the zone influenced by the project. When multiplied by the number of square meters in each parcel, the Final Coefficient gives what is called that parcel's "Area Virtual." As can be seen from the above, the "Area Virtual" (although expressed in square meters) is not an area at all, but a measure of benefit received by the parcel as a whole, ranked in scale relative to all others.  - 36 - 4.11 All of the Areas Virtuales are then summed to produce a figure whose magnitude may or may not be similar to that of the total budget of the project. Suppose that the former figure is 12.5 million square meters, while the project budget is 10 million pesos. A conversion factor of 0.8 pesos per square meter is then applied to the Area Virtual for each lot, to determine a valorization charge which (a) is in proper weight relative to charges on other parcels, and still (b) yields the amount needed to cover the budgeted costs of the project. F. The Coefficient System in Practice: Selected Illustrations 4.12 The administrative discretion that characterizes valorization in practice results in a diversity of application that makes classification of projects into any completely satisfactory categories almost impossible. Nevertheless, based on their principal similarities the 26 projects studied have been grouped into seven categories, though the categories are quite rough and almost every project has features of groupings other than its own. At the same time, the supposedly formidable administrative complexities of valorization are commonly given as the reason why the system has not found greater acceptance in other developing countries. For this reason, there is greater need than usual to describe in some detail how individual projects are carried out, so that a reasonable understanding may be achieved of its effectiveness in practice.  - 3? 1. Standard Formulas 4.13 One project begun in the 1960s (Carrera 68) illustrates the use of standard declining value curves and formulas to calcu- late the Basic Coefficient and Coefficient of Form, though with modifications to suit the project circumstances. Only 800 meters long, this street upgrading project was neverthe:Less a vital link in improving a major circumferential highway, and would affect areas previously lacking good road access and infrastructure. The project was approved in 1966 by the District Council, which at the same time set the zone of influence as "all properties located within two parallel lines 500 meters on each side of the axis of the road." Preparation of the Explanatory Memorandum and the final distri- bution of valorization charges took a little more than two years. The provisional budget was nearly 19 million pesos, of which about 4.6 million represented the value of land and buildings to be acquired and an additional 9.7 million was actual project cost:  - 38 - Table 3 Carrera 68, Autopista Medellin to 500 Meters Southwest of Calle 68: Provisional Budget (pesos) Value of land and construction to be acquired 4,840,845 Value of works 9,685,232 Subtotal 14,526,077 Administrative costs a/ 2,905,215 Contingencies b/ 726,304 Interest at 6% for 10 months 726,304 Subtotal 18,883,900 Honoraria to representatives c/ 56,652 Total Project Cost 18,940,552 a/ 20% of 14,526,077 b/ 5% of 14,526,077 c/ 0.15% of 18,883,900, times two representatives Source: Technical Section, Department of Valorization 4.14 Since this project had a zone of influence 500 meters on each side of the road, it was decided to make the Basic Coefficient a single curve, with its highest point at the edge of the road and lowest point at the 500 meter parallel line. These values per square meter were determined to be 100 Pesos and 3 Pesos respectively. However, strict application of the standard formula 500 x = k tog y - log 30 , where k is a constant, x is the  - 39 - distance from the property to the edge of the road, and y is the Basic Coefficient, would have yielded a reasonable range of values for the Basic Coefficient of 100 to 30, which did not correspond exactly to the anticipated range of variation in benefits. Accordingly, a slightly modified formula, to represent a Basic Coefficient range of 75 to 5 instead of 100 to 30, was used instead. 1/ Though in this case distance was probably determined using a straight line perpendicular to the project, the current, more pragmatic approach favors road distances. No distinction was felt necessary in this case between properties having access to a street perpendicular to the project and those with access to a street parallel to the project, since the benefits were judged to be sufficiently similar. 4.15 Two other modifications were then made to the basic curve. Because of their "evident and disproportionate" benefit from being on a new four-lane arterial highway, all direct 1/ The formula decided upon was y = 30 [10 x 2/k2 - 251, where the constant k is 675 and x' is 500 - x, where! x again refers to distance. Though most applicable to the middle range of properties (since it gives values over 100 pesos per square meter for very close-in parcels), t:he modified formula has also been applied in a number of other cases.  - 40 - frontage parcels were increased in value by 35%. In addition, to account for the fact that an existing subdivision along part of the project already had urban services and had ceded land for construction, the project lots from this subdivision were given a 25% lower coefficient. 4.16 For a lot in this subdivision whose Basic Coefficient is 75.00, these modifications resulted in an increase of 10% -- 35% for having direct frontage minus 25% for being in the sub- division. The Basic Coefficient would then be 75.00 + 7.50, or 82.50, which, when multiplied by the Coefficient of Form (1.23 in this instance), gives a Final Coefficient of 101.48, per square meter. 1/ When multiplied by the area of each property (to produce the 'Area Virtual") and summed, these calculations yielded a total of 25,966,359 square meters, compared with the total provisional budget of the project of 18,940,552 pesos. Distributing this project cost over the sum of benefits yields a conversion factor of 0.7294265582 pesos per square meter, which is applied to each "Area Virtual" to obtain the specific peso valorization charge to be levied on each property. The size distribution of charges for this project is given in Appendix Table 1. 1/ Since topography and other features in the zone of influence were uniform, no other coefficients were calculated for this project.  4.17 A substantially similar methodology was used in five other projects studied, including further improvements on Carrera 68. In one instance (Av. Caracas), a coefficient of topography, from 1.00 to 0.00, was used to adjust for land little benefited because of slopes. To account for very large parcels which could not be legally subdivided without ceding (for roads, public spaces, and so on) 30 to 35% of their area, in another instance Coefficients of Form of 0.70 and 0.65 were applied. 2. Matrix of Points 4.18 In a number of projects the calculation of the Basic Coefficient has been by a quite different and somewhat more complex system, reminiscent of the contouring method used to establish land values for acquisition. Like the preceding case of Carrera 68, improvement of Calle 100, an end-link of the same circumferential highway, was inscribed as a priority undertaking in Bogota's Master Plan of Roads (Plan Vial). The zone of influence, which had been delineated boundary by boundary by the District Council, was divided into a western and eastern segment, running not parallel but perpendicular to the project itself. A series of points was then laid out, with declining or increasing values according to the presumed desirability of the location after completion of the project. The range of values was auite broad, from a low, at the edge of one zone  - 42 - segment, of 1.5 to a high of 270. Other values formed a matrix of points across the entire zone. Since more than 20 points were designated, together with the gradients which connected them, the value for any parcel was easily obtained by inter- polation. As in this system no single formula is used, however, heavy reliance is placed on the discretion of its creators for accurate estimation of values. 4.19 Calculation of the final valorization charges for this project illustrates that as long as the coefficients accurately represent the relationships among the factors influencing benefits in the project area, the absolute values attached to these factors are of no other consequence. In this case the sum of the "Areas Virtuales" was 39,551,262, or more than twice the budget of 17,618,746 pesos. This was no problem, however, since the conversion factor was simply set at their ratio (0.44546608), and all values then would be properly adjusted in the final conversion. The same is true when the sum of ascribed benefits is less than the budget, as was the case for a smaller project (Carrera 11) that used a variant of the matrix method. All properties in that project were adjusted upward by 1.374950509 to arrive at the final valorization charge.  4 43 3. Multiple Factor Analysis 4.20 The two largest projects undertaken to date in Bogota-- the Eldorado Highway to the airport, by the Ministry of Public Works, and Carrera 7, one of the main links of the high and middle income north to the central city and itself a stimulant to high-rise residential and office construction--also had technically the most sophisticated methods of allocating valorization. Equally complex are the Master Plans for Sewers (North and Central), which involve charges to literally hundreds of thousands of properties. 4.21 In the lengthy Explanatory Memorandum for the Eldorado project, prepared by a local consulting firm, five possible methods of allocating benefits are considered but rejected as too simplistic for a project of this size. Use of the value of each parcel before and after the project as the proportion of the burden it should share is deemed impractical because (a) it requires waiting for completion of the project before collecting any charges, resulting in a lack of funds for con- struction; (b) the criteria of the first and second valuations are impossible to hold constant; (c) an accurate second valuation may mean a long wait, since the extent of value incre&ses from the project is not known until land use changes and new con- struction have fully exploited the changed conditions; and (d) changes in ownership are likely to occur which will make  - 44 - collection difficult. Calculations based on frontage may be satisfactory for sidewalks, paving, secondary sewers, and the like, where the impacts are local and the costs small, but not for a major highway affecting a large surrounding area. Since the highway runs from the central city to rural zones, encom- passing wide variations in impacts per square meter, methods based on area or mixed frontage and area are also rejected. Finally, for much the same reason, the parallel zone approach and its variations are regarded as unsuitable. 4.22 The method ultimately adopted was to determine with great care the benefits on a series of "base lots" (much as was done in the Calle 100 example described in paragraph 4.18), and then extrapolate from them to the remaining parcels. This could have been done either by direct assignation of a value repre- senting all of the benefits received by the base lot (as for Calle 100), or by a separate analysis of many coefficients for each of the base lots. Project managers chose the latter approach, identifying the relevant coefficients as: a. Location: An adjustment was made for differences in the impact on properties along the highway where it enters the central part of the city, and those near the airport, where at the time of the Memorandum there was little development. This coefficient runs from 100 to 30, representing  the judgment that actual increases in values at the two locations were 200 and 60 Pesos respec- tively. b. Distance: While the location coefficient adjusts for position along the length of the highway, "distance" takes account of position, perpendicular to the axis. Various non-linear relationships were experimented with before arriving at a curve which adequately depicted the pattern of declining benefits with distance. c. Size and Shape: An equation similar to the standard formula for the Coefficient of Form, but with lot size as well as shape, expresses the effects of different frontage and area combinations on values: -0.1359 0.1364 -0.7692 Y = 2.9622 A (F+l) P Jr (3) where Y is percent of the frontage price, A is area in square meters, and F and P are frontage and perimeter in meters, as in equation (1). In addition,another adjustm.ent is made for large parcels which will have to cede 35-40% of their land to the District before being urbanized.  - 46 - d. Access: This factor takes into account that some of the properties will have to use indirect routes to enter the limited-access Eldorado Highway. A coefficient of 1.20 applies to those on the high- way, running to 0.50 for those whose access is more difficult than their physical proximity would suggest. Parcels which are swampy are also adjusted under this Coefficient. e. Disturbance from Air Traffic: A complicated set of specifications for the landing and approach zones to the airport, permitting an adjustment from 1.00 (for those outside the zones entirely) to 0.20 for those most severely affected. 4.23 The coefficients are then multiplied together to obtain a first estimate of the value attributable to each parcel, which is then subjected to a final adjustment to ensure that each lot stands in reasonable relation to its neighbors, and in particular that the factors of distance and access appear to be equitable. 1/ 1/ It appears that values for each parcel were also calculated directly, using the factors just described, rather than only valuing base lots and extrapolating from them to the rest.  - 47 - 4.24 A similar method was used by the same consultirLa firm in calculating charges on the Carrera 7 project, as of 1974 the most costly (with a total budget of nearly 50 million pesos) and, from the standpoint of urban development, the most impor- tant single project undertaken by valorization in Bogota. Since this major arterial runs from the heart of the new office and commercial center for 36 blocks through a variety of land uses, topographv, street patterns and lot configurations, the apportionment of charges posed the most difficult problems of any valorization project. 4.25 With an undertaking which would radically transform many middle and low income residential neighborhoods and improve access to the city center, the expected benefits might well be high in relation to costs. Such was indeed the case for Carrera 7. Direct benefits to properties within the zone of influence, which extended over 5 kilometers with a width between 140 and 600 meters, were estimated to be 300 to 1,000 Pesos per square meter, as against individual tax burdens of 60 to 150 Pesos. While properties not facincl the Carrera were expected to increase an average of 30 Pesos per square meter over some 2,200,000 square meters, those with direct frontage would go up in value as much as 60 to 70%. The grand total of increment turned out to be three times  - 48 - the project cost. 1/ This fact, however, did not remove the difficult task of an equitable allocation of the valorization burden. 4.26 Charges were determined by the "base lot" method modified by the application of several other coefficients. After 36 "key" points along the route were selected, a lot, neither the highest nor the lowest valued, was taken as the base lot and assigned a coefficient of 100. One of three "location" coefficients, from 1.20 to 0.70, was applied to each base lot depending on land use in relation to the central city and established residential and commercial districts. Other factors such as distance and topography were also applied. A special "improvements" factor of 1.20 to 2.00 was added where a parcel contained a building of 10 or more stories, on the premise that the improved street adds more value to land employed in this fashion than land which is not. Moreover, it was predicted (quite accurately) that many of the unoccupied lots affected by the project would be used in a similar manner, for high rise apartments and offices. Parcels of the depth required for such buildings would gain enormously in value. However, rigid application of the standard formula for the 1/ Despite the large overall capital gain, representatives of the property owners objected to the inclusion of street lighting and traffic lights in the project, but were over- ruled by the Board of Valorization.  - 49 - Coefficient of Form, because of its tendency to give low square meter values to lots with small frontage and large depth, would have allowed much of the gain to go uncaptured. Regularity of lot size with neighbors was consequently given priority over application of the standard formula in adjusting for shape. Results for the base lots were then further modified fcllowing discussions with representatives of the property owners. 4.27 An undertaking of even greater scope is the ten-year construction program being carried out under the SeweraLge Master Plan and financed through valorization. In fact., this project was judged to be beyond the size feasible for identi- fication of a single zone of influence and levies on all property owners at once. Under the law of valorization, once c charge is levied the public body has an obligation to complete the work as prograrmed. Since the Master Plan extends for ten years, it would be intolerably constricting to give ae multitude of owners the right to immediate improvements. The solution was to divide the Plan into three sectors (North, Central, and South), and to create up to 100 or more districts in eeLch sector,so that properties are charged according to localized, not global, costs and benefits. For the most part, the districts established have been sufficiently homogeneous as to land use, socio-economic composition, densitv, and the like that the coefficient for a single influence, while varving between  - 50 districts, can be the same for all properties within a district. 4.28 For the North Sector, the Explanatory Memorandum for which was issued in March 1969, five major factors have been identified. A land use coefficient ranges from 1.00 for high density residential properties to 0.70 for institutional, industrial, and recreational land, based on uses permitted by the Planning Department. To incorporate the notion that valuable parcels receiuve a greater proportional increase in price from sewers than less valuable ones, a land price dif- ferential factor is 1.00 if the pre-project land value is 110 Pesos or more, and as low as 0.40 if it is 10 Pesos or less. Another factor, for existing sewer services, accords the greatest gain to land which has been urbanized, but lacks sewers now (1.00). Next come those not urbanized and needing additional private investment before being able to benefit (0.90). Those already urbanized and with sewers gain least (0.80). A fourth coefficient, given far greater weight than the others, adjusts for neighborhood income levels, in five categories from 10.00 for high income districts to 6.00 for working class areas. If a district is mixed in economic levels,-an intermediate value is taken from the above and applied to all parcels. Finally, flooded areas receive special benefits from having drainage, while other parcels may actually be damaged by having these works installed, and  - 51 - a fifth coefficient (1.20 to 0.40) was therefore appliedl. When multiplied together times the total area of each district, these five factors yielded the district "Area Virtual," which was adjusted by a conversion factor to the actual budget needed to carry out the North Sector program of works. 4.29 Calculations for the Central Sector, completed in 1973, continued the innovation of determining the "Areas Virtuales" directly from a set of descriptive coefficients, but tended in other ways to return to more traditional concepts. Since in the city center land prices vary much more widely (from an estimated 30 to 10,000 Pesos per square meter) than in the North, a land value parameter ranging from 10.00 to 38.33 was made the "basic" factor. The effect of distance from the project was incorporated using a conventional curve for Basic Coefficients, though here this factor was a modifying coefficient rather than the "basic" one. 4. Adjustment for Degree of Urbanization 4.30 When the same project influences both properties already urbanized and those still in large unsubdivided acreage, an equitable allocation of charges would need to take account of the greater impact likely from an improvement on rawv land than on land already subdivided and at least partially serviced. Consideration must also be given to the fact that investment is necessary to put the unsubdivided acreage into marketable condi-  - 52 - tion. Several projects, including two additional improvements to Carrera 68 (see also paragraph 4.13), divide the zone of influence into two according to whether services were already available, and make a further adjustment for the specific services lacking in each zone. 4.31 Basic Coefficients in the unurbanized zones of these projects have typically been calculated using a standard curve of declining value with distance from the road. The valorization resulting from road accessibility alone is modified by a set of factors (nine in one case) denoting the value of services that will be available in the road bed. It is then assumed that the value of access to services declines at the same rate with distance as road access benefits, which adds an equal percentage gain, sometimes as much as 50%, to the original values. In the already urbanized sector the Basic Coefficient is calculated in a similar manner, and another set of special factors accounts for the additional services which the project will give to certain neighborhoods not yet fully serviced. 5. Adjustment for Inflation 4.32 A special problem arises in connection with the construction of loading bays along streets in the older parts of the city. Under a 1948 statute 1/an owner wishing to parti- cipate in this program may cede a part of his land necessary 1/ Acuerdo 44.  - 53 - for the street widening. The value of his contribution will then be carried on the books of the city until the project is undertaken by valorization, at which time the value will be offset against the valorization charge. If the project takes several years to accomplish, as is often the case, the ceding owner will receive no credit for the effects of increases in land values, while his neighbors, whose land is purchased at the time of construction, will receive compensation in current prices. 4.33 In two cases which illustrate this point (Carrera 9 and Calle 13), a sliding scale from 1.00 to 0.70 was established. In effect, this gave no adjustment (1.00) to owners receiving all payment in cash at the time of the project, and as much as a 30 to 40% reduction in valorization charges for owners who had made cessions, with intermediate values for those making some payments in cash. 6. Ability to Pay 4.34 Several recent projects exemplify the more intensive efforts since the early 1970s to take account of socioeconomic factors reflecting the capacity of a neighborhood to pay for valorization. In earlier projects these factors were handled as one of many Trodifying coefficients to the basic charges. The more recent tendency is to determine, with judgments derived in part from field investigations, if the charges are  - 54 - within the owners' ability to pay. In cases where charges would amount to more than a. maximum percent of income per month, the method of payment (not the charge itself) is adjusted. 4.35 One project applying this method (Av. Boyacg) 1/ affects at least 17 neighborhoods of varying socioeconomic characteristics. Using the multiple factor approach, charges were first calculated. Neighborhood conditions were then studied in greater depth, with questionnaires administered in the field to a representative sample of owners in the zone of influence. In one segment of the project studied, a 0.5% sample of low and middle income owners revealed that the difference between income and consumption expenditure (which was taken to be the capacity for monthly payment) was 588 Pesos per month in the low income area and 1,577 Pesos per month in the middle income neighborhoods. Since nearly 70% of the charges would be less than 3,936 Pesos, and over 86% were less than one-tenth of the average market value of land and buildings in the zone, the burden was felt to be easily accommodated provided installment payments were arranged on a sliding scale according to the size of the charge. 1/ Information is taken from a staff memorandum, dated November 23, 1973.  - 55 - 7. Special Cases 4.36 As will be clear from the above discussion, valoriza- tion in Bogota has resisted standardization of methodology. A number of projects do not even fall into the somewhat forced categories outlined in this section, but still contain elements that warrant a brief summary. 4.37 Calculation of charges for a four level underground parking garage (Plazoleta del Rosario) in the central area of the city was based on a wide zone of influence, but with extremely large benefits to nearby parcels and a very rapidly dropping level of benefits with distance. Moreover, this project recognized that in practice, the desired level of collections might well be less than that to which the city is theoretically entitled. The authorities, it was argued, ought to leave a "margin of benefit" for the owner, to reimburse him for the stress of paying for improvements in the public interest without having to sell his property to realize the higher value. Estimates of value increase should contain "a margin of security to avoid errors about the value appreciation and injustice to the owners." 1/ 1/ Informe Inicial sobre la Distribucion de la Contribucion de Valorizacion: Plazoleta del Rosario (Bogota, Valorizacion y Asesorias Ltda., 1969), p.10.  - 56 - 4.38 However, the recommended policy of recovering not more than 80% of full benefits would not have provided enough resources to finance the project. In order to carry it out, a parallel recommendation had to be made that the city absorb both the normal 20% cost of administration charges and the contingency, or a total of 30% of the estimated project budget. Actual distribution among parcels was highly differentiated, and was done essentially on a parcel by parcel basis without an elaborate system of coefficients. 4.39 As is indeed inevitable whenever two improved streets intersect, many valorization projects in Bogota have overlapping zones of influence. Double and triple burdens on one owner are possible. The general rule for dealing with this problem is to grant--upon application, but automatically--extensions in payment periods, without interest. However, this policy has resulted in collection problems, which will be discussed in the next section. 4.40 One project (Calle 68) illustrates another possibility, of assuming that when (double valorization occurs,the level of benefits changes. The zone of influence of this project was divided into two, on the basis that one of the sub-zones would soon be subject to valorization from another project. The Basic Coefficient of the first, unaffected sub-zone runs from 75 to 5. The range of the second is 25 to 1, to account for the benefits forthcoming from the neighboring project.  - 57 - 5. TECHNICAL PROBLEMS: COLLECTION OF VALORIZAT:ION CHARGES A. The Collection Experience 5.1 Problems in collection have been encountered by virtually all countries that have attempted to recapture betterment, even those such as Great Britain with highly developed administrative systems. Though the record of recovery of valorization-type charges in Colombia is superior to that of most other Latin American countries, collection of charges has indeed been one of the most difficult aspects of valorization. 5.2 As of December 1973, 65% of payments on all valoriza- tion projects undertaken by the District government remained to be collected. Considering the life span of each project it is found that, on average, about four-fifths of project funds are recovered by the end of the five year statutory payment period, with payments virtually complete (96.2%) after the twelfth year. 1/ These figures, however, mask a considerable variation in the performance of individual projects, stemming from the nature and incidence of benefits, the collection arrangements in force, and the duration of the project. 1/ Linn, op. cit., p. 49.  B. Provisions for Payment 5.3 Two legal provisions govern the method of payment. First, since 1958 the repayment period has varied according to the taxpayer's income. 1/ Normally, the charge must be paid within six months, but if, within 20 days of being notified of the tax, a person obtains a certification of his taxable income from the Ministry of Finance, he may claim a longer period. If annual income is less than 50,000 Pesos (roughly $2,000 in mid- 1974, when field research was conducted), he must pay on the following terms: If the charge is: Payment must be made within: 10% or less of annual income 6 months 20% 12 months 30% 18 months 90% 54 months More than 90% 60 months If income is more than 50,000 Pesos, two-month extensions (up to 20 months) will be granted on a scale which in effect requires that no more than 60% of annual income need be paid. However, 1/ Article 117, Acuerdo 41/1958.  - 59 - even if the charge exceeds 100% of annual income it still must be paid within the 20-month extension period. Clearly, for a middle income family earning, say, 55,000 Pesos per year, 60% could pose a considerable burden. 5.4 On projects underway by 1974 the average charge seldom exceeded 10,000 Pesos (US$400), and one-half or more were 3,000 Pesos (US$120) or less. For an unskilled worker earning about 1,000 Pesos per month (US$40) these sums are not small. However, if his charge were, say, 3,600 Pesos, he would have 18 months to pay, for a charge of 200 Pesos per month or 20% of income. Such calculations demonstrate why adjustments need to be made for the socioeconomic characteristics of lower income neighborhoods, and whv the Board of Valorization is able to grant relief in specific hardship cases. While these arrangements make payments easier particularly in conditions of rapid inflation, the granting of installments, which carry no interest charges, is a considerable loss from the stand- point of IDU. 5.5 The second legal provision which affects payments established a system of discounts. 1/ This system arose from the notion, prevalent in the earlv davs of valorization (1944 to the early 1960s), that it was not feasible to impose charges prior to the completion of a project. The resulting liquidity problem was handled by creating a Revolving Fund ("Fcndo 1/ Article 119, Acuerdo 41/1958.  -560 - Rotatorio de Valorizacion`) whnich paid contractors in bondsc instead of cash. Whe contLractors objectead to the lac` 3f a market in such bonds, however, the Valorizatioc- Depar=nment agreed to accept the bonds as payment fo-, 7. oization charges. 5.6 Bonds were the usual metL.hcdi of -.pav-,ent to contractors until 1965, when their use be&-al to be phased outs Since the last issue was made in 1968 'at 6% interest for 20 years, the same terms as the 15 previous issues made between 1946 and 1968) alternative financing from international and other sources has been available. The bonds have scld on the Bogotd stock market at a fixed discount of 9% moreoveer, if an indi- vidual pays his valorization charge within three :.months, he. may use the bonds at face value and, in addtionp reieave a 10% premium. 1/ If he pays in three to six mont-hs the special discount is lost, but bonds will still be redeemed at face value. After six monthsd urless installmtent p:.lv' lwe,es nave been granted, payments must be in cash, and i nterest at 1.5% per month begins to run. If installments are nerm4tted, eaECh installment of up to l,000 Pesos may be paid in bonds (again-. resulting in a 9% saving) if done within the month due, other- wise only in cash plus accrued interest. When an individual is in delay on six installments d the installment pri-ilege is cancelled, and the entire smr, is due, at an onterest of 1% per 1/ Thus, to pay a 1000 peso charge, 900 pesos in bonds could be purchased at the 9% discount (cost: 81S-. Pace value becomes 990 with the 10% premi-um, requiring an addlitional payment of 10 pesos. In th:Ls example the 1000 peso charge -is paid off at a cost of 829, a 17.1% discount.  - 61 - month until paid. The loss from these discounts and premiums has been far from insignificant: about 6.5% of collections in 1972 and 6.2% in 1973. This cost should drop further, however, as the number of bonds outstanding continues to decline. C. Buildup of Arrears 5.7 As noted above, interest on defaults is 1.5% per month for six months on overdue installments, after which it becomes 1% per month on the entire balance. While not compounded, it clearly mounts rapidly, making repayment more difficult. A 1970 statute attempted to give amnesty on interest for lthose who would pay the balance due. 1/ However, since at the time commercial interest rates were higher than the IDU charge, borrowing to take advantage of the amnesty offer was not profitable. This provision has had only a minimal effect on actual collections. From experience distilled so far, it appears that property owners tend to pay valorization whaen they have money, with or without interest forgiveness, but not to borrow to do so. Monthly interest charges,designed to provide an incentive to pay promptly, have instead probably had the opposite effect. 5.8 The extent to which 16 valorization projects were in arrears at the end of five years, and the distribution 'by source of these deficits, are given in Appendix Table 2. WThether 1/ Acuerdo 1/1970.  - 62 - attributable to the accumulation of interest or simply to the psychology of debt, it is seen from the table that collection problems appear to become more difficult after a lapse of time, irrespective of whether private or public owners are involved. Nor does collection performiance seem related to the size of the project. 5.9 When an owner is clearly in arrears, the administra- tive unit charged with enforcing payment can have the claim entered as an encumbrance on the property, and eventually sell the property at auction in settlement oF the charge. This process can be done in ten months if there is no opposition, but it can occupy two to ten years if there is. Very few auctions have actually taken place, however, since owners almost always pay when their land is judicially attached. 1/ D. Financial Relations with Government and Charitable Orcganizations 5.10 Under Colombian law there are few exemptions from the valorization charge. Only Church property as defined by the Concordat of 1888 and "properties of public use" are exempted. In practice, however, administrative difficulties arise in attempting to collect on property owned by a national Ministry, the Church, charitable foundations, or units of the municipal government itself. 5.11 Taken together, these amounts in 1973 represented 35% of outstanding obligations. Over half the amounts due from official entities were owed by the national government 1/ An additional incentive to nay is the fact that in Colombia, land cannot be transferred without a "'Paz y Salvo" document obtainable from local taxina authorities, which includes a certification that no valorization charges are outstanding.  - 63 - and by the Beneficiencia de Cundinamarca, a public charitable foundation which provides a number of public facilities and social services in BogotZ, and whose heavy investment in land has made it one of the largest owners of peripheral land in the city. Another 18% was owed by the city itself. 1/ In a few cases, IDU has been able to cancel a debt on one project by having a public body donate land for another, but this is, of course, dependent upon favorable accidents of location. For non-exempt Church properties, the procedure has been for IDU to notify the District Council of charges levied on Church holdings, after which the Council passes an appropriation to cover them. This system does not seem to have been consiLstently carried out in practice, and Church property appears as a signi- ficant portion of the "private"deficit in valorization, though with great variation from project to project. 5.12 Another source of difficulty is the failure of the District government in recent years to deliver annual subven- tions from general tax receipts to IDU at the rate it is committed to do so. For 1974, only 15% of the amount owed from general funds appears actually to have been paid to IDU. 2/ 1/ Unpublished report, IDU, September 1973. 2/ See Linrn, op. cit., p. 61.  - 64 - E. Special Administrative Problems 5.13 As mentioned earlier, many properties in the city will be subject to two, three, or even four valorization charges from different projects. In such cases installments run consecutively, the second beginninc when the first expires. 1/ Consequently, if an individual is eligible for five years of installments and liable for three projects, payments may continue over 15 years, all without interest or adjustment for inflation. Arrears on at least four of the 16 projects listed in Appendix Table 2 are lower than indicated because the existence of simultaneous projects means that some payments have been deferred. 5.14 Collection problems also arise from the large number of properties, generally near the edges of zones of influence, with quite small charges. On these it is clearly uneconomic to devote much effort to collecting overdue payments. Never- theless, it is difficult to conclude that collections have varied in any significant way with the size of the levy. In recent years, in fact, the record most closely exemplifying the kind of sustained, self-supporting financial performance hoped for in valorization projects has been turned in by the Local Paving Fund, which specializes not in widening major thorough- fares but in small, local improvements benefiting low and middle 1/ Article 124, Acuerdo 41/1958.  - 5 - income neighborhoods. Evidently the greater attention to resi- dents' capacity to pay, plus the close association between valorization charges and benefits in such projects, have helped increase the acceptability of this program. 5.15 In a process as complicated as valorization, errors will inevitably be made: in the size of parcels, in application of the correct coefficients, in simple mathematical and clerical slips. The system of Representatives does apparently catch many of these, and the total quantity is in any case small: 0.1% of total collections in 1972 and 0.8% in 1973. Correction of these errors does take time, however, and many months may pass before an individual can be properly billed. 5.16 Other problems occur when land is being occupied by squatters. In such cases,sanctions on owners for non-payment of valorization would have little effect, since they may not care if the land is taken for taxes or not. Similarly, in the more common case of extralegal or "pirate" subdivisions where occupants have "contracts to buy" but not actual title, the owners will be reluctant to pay because possession has effect- ively passed permanently out of their hands. The IDU in these instances suspends attempts to collect until the title can be resolved, and it has a special section to deal with such matters. Delays can be considerable, however, and collections meanwhile fall further behind schedule.  _ CE: _ 6. IMPACTS OF VALORIZATION A. Access to Employment and Housing 6.1 In considering the effects of valorization on urban growth and incomes, one should look at changes in the overall distribution of services it has helped to produce as well as the impact of individual projects. Systems of municipal finance in the large cities of most developing countries are normally hard-pressed to meet the costs of rapid urban infrastructure expansion. Valorization in Bogotaj, to the extent that it has put most arterial streets, the sewer system, and local street paving on a more or less self-financing basis, has probably permitted the city to install these services at a more rapid rate than would otherwise have been possible. This, in turn, has had consequences for all income groups. 6.2 Valorization projects have unquestionably enhanced the value of Bogota central city real estate. Without them, traffic in the old central area would undoubtedly have reached the saturation point many years ago, and the surge of investment taking place in the early 1970s might not have occurred. High income groups have also benefited from projects like Carrera 7 and the recent improvements to Calle 100, which have greatly increased the accessibility of their living areas in the north end of the city to employment in the center.  - 67 - 6.3 In spite of these benefits, however, the rich have traditionally been the strongest opponents of valorization. 1/ Presumably they have felt that their influence was sufficiently strong to obtain good infrastructure in their neighborhoods, and that if services were financed from general municipal revenues their relative burden would be less than under valorization. The persistence of this phenomenon tends to support the suggestion of Rhoads and Bird that this in itself is evidence that valori- zation is more progressive than other forms of municipal taxation. 6.4 Middle income groups, who also live in the north and northwest, have also been afforded better access by valorization, presumably at a faster rate than would otherwise have occurred. The half-dozen new projects to improve Carrera 68 stand especially to benefit the most recent band of middle income subdivisions. 6.5 In addition to the Local Paving program (a significant improvement given Bogota's rainy climate), low income neighbor- hoods have also benefited from valorization projects. Two early projects involving Carrera 10 directly improved bus transportation to low income areas to the south of the central city, as did the two projects (February 1968 and February 1973) for the improve- ment of Carrera 14. 1/ See, for example, Rhoads and Bird (1967), p. 409.  6.6 Aside from individlaal projects, ho'rI r, e 7 . -apersuas `ve if indirect, case can be rad2e for valorI1 za-- as an -id-i 'tot .e poor through its cor.tribution to an adeaua;_ a reri_ sc reet - system that has permitted a re latively Satte-nd pe. e--n of Urnci zation and ease of movement oy -pub- 1 ra1soratl on Q ith a highly competitive bus systm,. o,,;_a-,la -:ey ungrthad ued fares, lower income fa milsi-s have a relatively larce c-noc-l e o- housing locations and wide accessibility to jobs at low cost, with reasonable speecd and conven-lence. Ir a smra l sur-vey o f three new neighborhoods, for example, 62-'6% of those suarveyed investigated at least one other neiahbocrhood `efore mtakinca a f-inal choice of residence. I/ Possibly even miore .1important a the transportation system permits emnployment access -from -moxre remote locations and hence higher residential lot sizes: 150-170 square meters is not uncommon in the "pirat-e neighbcor_hoo6dS. Higher lot. size facilitates the creation< of rental acccK-=cda tions without taking away space needed b- the fami.,ly, n Bogot Z; about one-fifth of the income of persons ir the 1i r ate nec5J-g borhoods is from rent of this type. 2/ Small commercial and 11/ Survey of 115 heads of household in three Iiratev barrios ,n Bogota, conducted under the supervision of William A. Doebele, April 1974. 2/ G. Vernez, "Bogota's Pirate Settlements: An Opportunity for Metropolitan Deve:Lopment" (Ph.D. dissertation, University of California, 1973), p. 104. In three barrios studied, rental income ran from 18e5% to 48.8% of the total ir various income ranges. Some 46% of all Bogota families liv-e i "pirate" neighborhoods. While they cannot be afforded by the very lowest income households, and would not be chosen by the upper middle class and above, these neighborhoods are the overwhelminglv popular form of housing for the upper-lower and lower-middle segments of the population  - 69 - craft activities in the house are probably also easier to under- take on larger lots; in a survey of one such neighborhood, 19% of all occupied lots had activities of this kind. 1/ Vegetable gardens in the back yard and other such activities are also facilitated. To the extent that the arterial street system of Bogot