Urban property tax reform : the case of the Philippines' real property tax administration project

INU-16 THE WORLD BANK POLICY PLANNING AND RESEARCH STAFF Infrastructure and Urban Development Department Report INU 16 Urban Property 'ITax Reform The Case of the Philippines' Real Property Tax Administration Project by William Dillinger May 1988 Discussion Paper This is a document published informally by the World Bank. The views and interpretations herein are those of the author and should not be attributed to the World Bank, to its affiliated organizations, or to any ndividual actng on their behalf.  Copyright 1988 The World Bark 1818 H Street, N.W. All Rights Reserved First Printing May 1988 This is a document published informally by the Wcrld Bank. In order that the information contained in it can be presented with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The World Bank does not accept responsibility for the views expressed herein, which are those of the author and should not be attributed to the World Bank or to its aFfiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of material, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, city, area, or of its authorities, or concerning the delimitations of its boundaries or national affiliation. The principal author is William Dillinger from the Infrastructure and Urban Development Department of the World Bank. Comments and suggestions by Per Ljung, Francois Pechon, and Yoshine Uchimura at the World Bank, and G. Peterson of the Urban Institute are gratefully ackiowledged. Any errors remain the responsibility of the author.  The World Bank Urban Property Tax Reform The Case of the Philippines' Real Property Tax Administration Project Discussion Paper   PREFACE The property tax is the most widely used municipal revenue source in the developing world. In almost all LDC's with large urban populatiVs, some form of property tax is assigned to the support of local governments.- The performance o:E the property tax is, however, generally disappointing. Its contribution to total municipal revenues is small (generally under 25 percent) and its burden tends to fall with a high degree of arbitrariness. This is the result of problems in both policy and administration. Policy decisions--low tax rates, broad exemptions, infrequent adjustments for inflation--reduce the statutory level of the property tax. Poor administration--incomplete tax rolls, haphazard valuations, low collection efficiency--reduce the proportion of the statutory base that is effectively taxed, and introduce arbitrariness into the tax's incidence. Efforts to improve the performance of the property tax need to address both types of problems. While policy actions offer the prospect of quick revenue increases, taken alone they exaggerate the inequities in the tax's incidence. A rise in the tax rate, for example, places the burden of the increase on those few irLdividuals whose properties are on the rolls and are accurately valued, and from whom taxes are actually collected. Improvements in administration offer the prospect of increasing yields by improving equity. Improvements in the comprehensiveness of tax rolls, the objectivity of valuations, and the efficiency of collections increase revenues by increasing the burden on those who presently underpay. Taken alone, however, such administrative improvements may not be worthwhile. If tax rates remain low, the absolute level of property taxes may be trivial. Under these conditions, the property tax, although equitably administered, may not produce enough revenue to be worth collecting. The case study that follows is the first of two proposed studies of major property tax reforms in developing countries. (The report on the second case, Brazil, will be issued separately.) The two studies, financed by the UNDP (INT/86/006), will be the basis for a guidelines paper on urban property tax reform, intended for use by LDC officials and staff of international development agencies. 1/ Among the 50 largest LDC metropolitan areas, only those in Iraq, Viet Nam and China (which is presently considering a property tax) do not receive revenues from an urban property tax.  - iii - The cases are intended to provide two kinds of lessons. The first concerns property tax systems and ptocedures. Both case study countries introduced low-cost systems of property discovery and valuation, which could serve as models for other countries facing similar problems. The second concerns project design and implementation. Both cases were nationwide in scope. As such, they addressed one of the traditional constraints on large-scale property tax reform: the decentralized nature the administrative aspects of property taxation. In packaging technical assistance for implementation on a large scale, the two cases provide lessons for governments and international development agencies that are attempting reforms of a similar scope.  - iv - SUMMARY The Philippines Real Property Tax Administration Project (RPTA) was an effort by the Government of the Philippines to improve local property tax administration on a nationwide scale. As a major program with a demonstrated track record, it merits the attention of other countries seeking to improve the performance of local taxes. RPTA had a simple objective: to increase the coverage and improve the accuracy of property assessments. To do so, it proposed to change the system used to identify and value property. Prior to RPTA, the system relied upon voluntary declarations by property owners. RPTA replaced this with a system based upon comprehensive maps and field-verified property information records. Technical Lessons From a technical standpoint, RPTA illustrates the successful application of low cost mapping and assessment techniques. In a typical jurisdiction, RPTA began with only two sources of property information: long- outdated base maps (prepared at the time the property was originally transferred from the public domain) and the incomplete and inaccurate declarations of current property-tax payers. From this starting point, RPTA was able to produce comprehensive parcellary maps, and individual, objectively derived, property assessments, at an average cost of about US$2.50 per land parcel. This was made possible by the adoption of minimum standards for mapping and assessment. In preparing the parcellary maps, no attempt was made to demarcate legal boundaries or to adjudicate title. Mapping standards were instead adopted to the specific requirements of a fiscal cadastre: to account for all land in the jurisdiction, to delineate parcels sufficiently to permit subsequent identification, and to obtain the square footage data needed for assessment. Assessments themselves were derived using a simple mass appraisal technique, in which a small number of measureable property characteristics are used as the basis for calculating relative value. By adopting this approach, RPTA minimized the use of scarce and expensive technical personnel. In most jurisdictions, the program was implemented by the local assessor's staff, recent (and otherwise unemployed) engineering graduates, and casual labor. Program Design and Implementation Lessons As a model for the design and implementation of property tax reform programs, RPTA's experience yields several lessons:  Reforms should be based on a comprehensive view of the problem Although RPTA successfully increased the number of properties on the tax rolls and the accuracy of their valuations, its ultimate impact on tax revenues was negligible. On average, RPTA projects increased local property tax revenues by only 1.1 percent. The program's impact was negligible because its scope was too narrow. While reforming the system of property discovery, RPTA did not address other problems in the tax's administration. Problems in collection enforcement, in particular, were neglected. As a result, in many jurisdictions RPTA-induced increases in assessments were more than offset by decreases in collection efficiency. RPTA also failed to address specific policies which constrain the yield of the property tax. Centrally-imposed restrictions on tax rates and assessment ratios, for example, limit the maximum statutory rate on urban residential land to 0.75 percent in provinces, and 0.9 percent in cities. Effective levels of property taxation are even lower. Centrally-decreed postponements in the introduction of new pricing factors have reduced the effective rate of taxation on urban residential land2 to less than 0.20 percent. This experience suggests that property tax reform, in order to achieve its intended revenue impact, must address both the entire sequence of steps involved in property tax administration (discovery, valuation, and collection) and the specific policies (rates, assessment ratios, and inflation adjustments) which constrain its yield and buoyancy. Administrative reform should aim at permanent procedural change, particularly in maintenance systems RPTA is a successful example of a means to achieve sustainable improvements in property tax administration. The program fundamentally changes the way property (and property characteristics) are discovered for purposes of taxation. By permanently altering these procedures, RPTA increases the liklihood that its impact will endure. RPTA's use of local assessors to implement the program also improves the prospects for sustainability. By using local assessors, RPTA ensures that the people responsible for using the new procedures will, in fact, be capable of doing so. RPTA's impact will, however, only be take effect during general revaluations. The new system of discovery does not change the way interim 2/ Measured as a percent of current market value in July, 1987. These figures include a 1 percent surcharge on the local property tax earmarked for local expenditures on primary education.  - Vi - changes in property characteristics are brought to the attention of the assessor and incorporated in the tax rolls between general revaluations. Because general revaluations are often postponed, RPTA's neglect of these maintenance procedures is a serious flaw in its design. RPTA's experience thus has two ifmplications for the design of sustainable property tax reforms: (1) the reform should aim to achieve permanent procedural changes; and (2) a focus of such changes should be maintenance. Detailed design should preceed implementation RPTA's initial years of operation were plagued by administrative delays. Project completions during the program's initial 3.5 years averaged 70 per year, against a target of 160. These problems, and the solutions eventually devised by RPTA staff provide several general lessons. The importance of bureaucratic "fit": Although local assessors report to the Ministry of Finance, RPTA was initially administered by the Ministry of Local Government. Assessors' reluctance to take direction from a rival ministry slowed implementation. This problem was resolved by transferred the program to the Ministry of Finance. The need for adequate and timely financing: RPTA's initial financing frequently proved to be too little, too late. Levels of funding were fixed without regard to the size of each jurisdiction. Half of the funding for each project was witheld until project completion. As a result, implementation was often halted until alternative sources of funds were obtained. RPTA's financing arrangement has since been changed: the amount of funding is now based upon the number of properties to be surveyed and funds are disbursed as claims are submitted. The benefits of "packaged" technical support: Central technical support was sparse during R]PTA's initial years of operation. Training programs did not focus on the specifics of project implementation. In the field, project leaders were left to devise thieir own procedures and operating manuals. This resulted in much duplication of effort and delay. Plans for the next phase of RPTA call for a "package" approach to delivering technical support. Program staff are now preparing standardized procedures manuals, forms, and training materials which will guide project leaders through the RPTA process. This approach is expected to speed project implementation considerably. The need for selectivity in the choice of participating jurisdictions: RPTA's initial lelays are also attributable to the absence, in some localities, of adequate base conditions or insufficient adminstrative support by the local government. Jurisdictions were enrolled that had too few taxable properties to justify the cost of project mobilization; misunderstandings arose over local obligations to provide in-kind support  - vii - (office space, clerical personnel) and counterpart funds. The next phase of RPTA proposes to addrer this problem by establishing specific criteria for inclusion in the program- . While the specific solutions adopted by RPTA will not apply to all situations, one broad conclusion is clear: careful and detailed program design should preceed program implementation. 3/ Jurisdictions must have a minimum of 4000 taxable properties; base maps (or a basis on which to prepare them), and a permanently-appointed assessor in place, and must demonstrate that office space, equipment, and support personnel, as well as counterpart funding, will be available.  - viii - URBAN PROPERTY TAX REFORM THE CASE OF THE PHILIPPINES' REAL PROPERTY TAX ADMINISTRLATION PROJECT T'able of Contents Page No. I. * INTRODUCTION ......................................... ....... 1 A. Institutions .....................* .. .................... 1 B. Local Government Finance ........... .. .. .. .. . .. ............. 1 Revenue Sources ......... ................... . .. ....... ....... . 1 Revenue Performance ....... .......................... ... . .. * . 2 II. PROPERTY TAXATION IN THE PHILIPPINES ............... .......... 3 A. Legal Framework . ......... . ...... . *............ . ........ 3 Definition of the Tax Base ..................... . ....... 3 Method of Discovery ................................... 3 Valuation and Assessment ...... ........................ 4 Assessment Ratios and Tax Rates ...................... . . 5 Billing and Collections ....... ...... .... . .. ... 5 B. The Real Property Tax Administration Project ............. 6 RPTA's Diagnosis of the Problem .............. ...... 6 RPTA Procedures . .. ...... ......... .. .. . ................ 7 Preparation of Base Map ......................... 7 Prefield Tie-up ................................ 0............ 9 Field Work ....................... ........................ 10 III. * EVALUATION OF PROJECT ........................... 10 A. Achieving the Coverage Target ............... *.......... 10 Organization ................. .......11 FPinancing ............ ............11 Central Management . ....... . .. ............ *. ......... 13 B. Achieving the Assessmnent Target ... ......... 15 C. Measures Needed to Achieve Sustained Yields ..........**. 15 Improving Assessment Techniques .......................... 16 Institutionalizing Maintenance............................. 17 Tax Rates and Other Policy Variables. 18 Improving Collection Efficiency ......................... 19 D. The Question of Cost-Effectiveness ....................... 21  - ix - FORMS Form 1: Declaration of Real Property ............... o .............. 23 Form 2: Pre-Tax Mapping Control Roll ..** .......... ................ 24 Form 3: Certificate of Title ...................................... 25 Form 4: Assessment Roll ......................I...... ............. 27 Form 5: Real Property Tax Order of Payment ..... ................... 28 Form 6: Real Property Tax Record .................................. 29 Form 7: Abstract of Real Property Tax Receipts ... ................. 30 TABLES Table 1: Sources of Local Government Revenue .... .................. 2 Table 2: Tax Rate Ranges ....... .................................... 5 Table 3: RPTA Project Completions ................................. 11 Table 4: Sources of Change in Assessed Values RPTA Phase II ....... 16 Table 5: Frequency Distribution of Projects by Percentage Change in Collections . .......................... ....... 20 Table 6: RPTA Project Cost Estimate (Cost Per Land Parcel) . ....... 21  URBAN PROPERTY TAX REFORM THE CASE OF THE PHILIPPINES' REAL PROPERTY TAX ADMINISTRATION PROJECT I. INTRODUCTION This report describes the Philippines Real and Property Tax Administration Project (RPTA), a program of the Philippine government, aimed at improving the administration of local property taxes on a nationwide scale. A. Instituti,Dns The Philippines is a lower-middle income country (GDP per capita US$680, 1985) with a population of 54.5 million (1985). Roughly 40 percent of the population live in urban areas. Of these, one-third (5 million) live in metropolitan Manila. The government of the Philippines is organized as a unitary state. The national territory is divided into 75 provinces and 61 chartered cities. The provinces are further divided into municipalities (totalling 1550), each containing a small town (poblacion) and surrounding rural areas. Both municipalities and cities are divided into barangays or districts. The term "local government" used in this report appLies to provinces, municipalities and cities. During the period under study (1978-1983) the Philippines was ruled under martial law. Provincial governors were appointed directly by the president. Political leadership at the municipal and city level, however, continued to be selected through local popular elections. B. Local Governmenl: Finance Revenue Sources. The central goveirnment maintains a high degree of control over the instruments of local finance in the Philippines. The allocation of revenue instruments is made under the national Local Government Code and Real Property Tax Codle. The property tax is one of two broad categories of tax bases assigned to local government. In 1985, property taxes accounted for 17 percent of total local receipts (Table 1). The second revenue base assigned to local government consists of various forms of business taxation. The Local Government Code assigns different forms of business taxation to the different types of local government. To the provinces, the Code assigns taxes on specific kinds of businesses, defined by industry (printing and publishing, sand and gravel extraction) or mode of distribution (peddlers tax, truck delivery tax). To the municipalities, the code assigns a single tax on businesses engaged in manufacturing and distribution of goods. Cities, having the legal attributes of both provinces and municipalities, are authorized to impose the taxes  - 2 - assigned to both these levels of government. In aggregate, business taxes contribute about 14 percent of local government revenues. Table 1: SOURCES OF LOCAL GOVERNMENT REVENUE (Millions of Pesos, 1985) Percent of Total Taxes 2696.0 31 Real Property 1492.3 17 Business, Other Taxes 1203.7 14 Fee and Charges 2217.8 26 Transfers 3735.5 43 Formula Transfers 3205.2 37 Other Transfers 530.3 6 Total 8649.3 100 Source: MOF In addition to taxes, the Local Government Code authorizes local governments to impose a variety of fees for the use of public facilities (markets, slaughterhouses, roads, and bridges) and to operate revenue- generating enterprises. Though no single fee or charge constitutes a major revenue source, the category as a whole accounts for about one-quarter of total local revenues. Locally administered taxes and charges are supplemented by intergovernmental transfers, which as a group account for 43 percent of local revenues. The largest of these, the BIR, is derived from a 20 percent share of central government tax revenues (calculated as a percentage of collections three years prior) and is distributed among local governments by formula. In aggregate, these revenue sources (plus minor contributions from borrowing) support a level of local expenditures equal to about US$8.25 per capita (in 1985), or roughly 10 percent of public sector expenditure. Revenue Performance. The yields of local revenues, were relatively stable during the late 1970s and early 1980s, remaining virtually constant in real per capita terms between 1978 and 1983. A sudden increase in inflation in 1984, and continued inflation in 1985 seriously eroded revenue yields in real terms, however, with real per capita revenues falling 32 percent over the two year period. Property tax yields proved especially vulnerable to inflation, increasing by 20 percent in nominal terms, but declining by 31 percent in real terms. As a result, property tax yields now (1985) average only US$1.47 per capita, or 0.24 percent of GDP. This ranks the Philippines slightly above average for developing countries (0.20 percent of GDP, based on  -3 the IMF's most recent survey, covering 55 countries), but well below the level prevailing in Philippines in earlier years (0.32 percent in 1983). II. PROPERTY TAXATION IN THE PHILIPPINES A. Legal Framework The legal framework for property taxation in the Philippines is established in Presidential Decree (PD 464) the Real Property Tax Code. Issued in 1974, the decree builds on earlier legislation governing local property taxation dating back to the Administrative Code of 1916. Though the terms of Code are not always observed in practice, the law provides the framework for property tax administration. Definition of the Tax Base Under PD 464, the base of the property tax is defined as "all real property, including land, buildings, machinery and other improvements attached or affixed to real property." Both urban and rural land, improved or unimproved is subject to property taxation. The decree allows five classes of exemptions: (a) government-owned property (including property owned by local governments and government-owned corporations so exempt by charter); (b) property owned by charitable and religious institutions used exclusively for religious or charitable purposes; (c) low value property, defined as properties owned by one person in a given city or municipality having aggregate value of P1000 or less; (d) certain agriculturaL properties (including land acquired from the public domain and used for dairying or livestock production, for five years following acquisition); and (e) machinery of a pioneer and preferred industry, as certified by the Board of Investment, for a period of three years from the date of initial operation. Method of Discovery Under PD 464, discovery of property is based on owner declaration. The decree (as amended) requires all persons, natural or juridical, to file a sworn statement of real property ownership with the local assessor, containing a description of the property sufficient to enable the assessor to identify it in the field. Sworn statements must be filecd every three years (corresponding to the general revaluation cycle provided for in law) and whenever an individual acquires land or makes construction improvements. The decree also enjoins the Registrar of Deeds in each jurisdiction to furnish the assessor with an abstract of his register (with description of property, name of present owner, and date of acquisition) within two years of the original  - 4 - enactment of the decree (1974), and to provide the assessor with copies of all subsequent contracts conveying, leasing or mortgaging the property. Valuation and Assessment PD 464 defines the basis for valuation as the "current and fair market value", defined as the "highest price .... which the property will buy if offered for sale in an open market." The decree provides that the "Ministry of Finance shall be primarily responsible for the efficient and effective administration of the property tax." To discharge this responsibility, MOF is charged with "developing a uniform assessment system, adopting measures to ensure maximum collection efficiency, and formulating policies and procedures for improving technical skills." The task of assessing property for tax purposes is assigned to provincial, municipal, and city assessors. Local government have surprisingly little administrative control over these officials. Although assessors are paid from the local government budget, responsibility for hiring, transferring, and disciplining local assessors rests with the MOF's Office of Local Government Finance. Detailed regulations for the valuation of property are issued by MOF. The regulations essentially prescribe a mass appraisal approach, where standardized unit costs of land and improvements are first calculated on the basis of a sample of market transactions. Under the MOF regulations, these unit cost tables (or "schedules of values", as termed in the Philippines) are to be prepared prior to each general revaluation. Data on the physical characteristics each taxable property are then to be gathered, either by field inspection, or from tax declarations. Individual properties are then to be valued by applying the standardized unit costs to the physical characteristics data for each property. To permit price increases and changes in property characteristics to be reflected in the tax rolls, PD 464 provides for general revaluations of properties, nationwide, every three years. As provided in law, the revision demands preparation of a new schedule of values, and reinspection of properties for purposes of verifying ownership and the physical characteristics to be used in appraisal. The President, however, has power to postpone the date at which revised assessments become effective. The revision currently in force dates from 1981; the revision scheduled to become effective in 1984 has been repeatedly postponed by executive decree, and is now (April 1987) scheduled to become effective July 1, 1987. Between general revaluations, PD 464 permits certain changes in property characteristics to be incorporated immediately onto the tax rolls. These include the creation of new parcels through subdivision, changes in ownership, and major new construction. PD 464 provides a variety of measures to aid assessors in discovering these changes. Surveyors (private or public) are required to provide the assessor with copies of approved subdivision plans within 30 days of receipt from the Bureau of Lands. When property is sold in cases of sale of property, the registrar of deeds is required to provide the assessor with a copy of the deed of sale; and both the purchaser and the seller are required to separately notify the assessors office, within 60 days. In the case of new construction, the official issuing the building permit (generally the municipal or city engineer) is required to notify the  - 5 - assessor, as is the owner (in the form of a tax declaration). No adjustments in schedules of values are, however, permitted between general revaluations. Assessment Ratios and Tax Rates The Real Property Tax Code fixes assessment ratios ranging from 15 percent to 80 percent. (These are percentages by which property valuations are to be multipled in order to determine assessed value.) Actual tax liability is calculated by multiplying assessed values by the nominal tax rate. The Code specifies a range of tax rates which local government may apply. As shown in Table 2, maximum local tax rates range from 1 percent in municipalities (taking into account the maximum provincial rate) to 2 percent in cities. A mandatory 1 percent additional levy is mandated in all jurisdictions, with revenues earmarked for education. Table 2: TAX RATE RNiGES /a Minimum Maximum Municipality .25% .5% Province .25% .5% Combined rate .5% 1.0% city .5% 2.0% /a As percent of assessed capital value. Rate excludes 1 percent Special Education Fund surcharge. Billing and Collections Legal liability for the property tax rests with property itself. The determination of legal ownership is not a precondition to imposing the tax. For administrative purposes, the owner (rather than occupant) is the target of collection effort. The primary source of information on ownership is the tax declaration submitted by claimants to property. By law, no inquiry need be made into the validity of each claim,, and assessors have the right to accept, for tax purposes, more than one claim per property. Where a property exists for which iio declaration has been made the assessor has the authority to declare the property in the name of the owner (if known to the assessor) or against an unknown owner. Where property is being assessed for the first time, the assessor is required to notify the owner, by having the notice of assessment delivered directly, or through registered mail. Where the owner does not reside on the property, the assessor may deliver the notification of assessment to the occupant, or mail  - 6 - it to the owner's last known address. Inability to deliver the notice of assessment to the owner does not relieve the property of tax liability. Collection of property taxes is the responsibility of the local government treasurer. Like the local assessor, the treasurer is paid from local funds, but is subject to the administrative control of the MOF. By law, property taxes become due on the first day of January of each year. The treasurer is not legally required to produce and deliver tax bills provided a general notice of payment dates has been posted in local government buildings, in a prominent place in each barangay, and has been published in a newspaper or announced by a public crier. Payment to the treasurer may be made in full by March 31 (in which case it is subject to a discount); or paid in quarterly installments. Failure to pay on schedule subjects the taxpayer to a penalty payment of 2 percent per month, on the outstanding amount. The tax code provides three additional enforcement mechanisms, by which a treasurer may enforce payment of property taxes: (1) distraint of personal property (subject to certain limitations); (2) sale of the property itself at public auction; or (3) collection through the courts through civil action. In addition, outstanding arrears accumulate as liens on title, reducing the property's value at time of sale, or as collateral for mortgage. B. The Real Property Tax Administration Project RPTA's Diagnosis of the Problem The property tax system established under PD 464 was, in principle, workable. Most of its provisions are common in countries using capital or sales value as the definition of property value. As designed, however, it depended on a high degree of voluntary compliance on the part of taxpayers, and efficient information-sharing among various agencies of government. This dependence was not justified by prevailing practices. Of particular concern to Government were problems in the system of discovery. Reliance on owner declarations was believed to be the source of several problems. First, it was believed that a significant number of land parcels were undeclared, and therefore entirely missing from the tax rolls. Second, on parcels which were declared, property characteristics were believed to be understated. Thus valuations based on owner-declarations of parcel size and building characteristics were significantly too low. In addition, parcels with disputed ownership were often separately declared by each of the disputants. While multiple claims inflated the identified tax base, subsequent disputes undermined collection performance. The Philippine Real Property Tax Administration Project was designed around this understanding of the problem. The projects' approach was to fundamentally change the system of discovery, from one based on owner- declaration to one based on government-executed inventory. Through this reform, the project aimed to compile in each jurisdiction a comprehensive  - 7 - inventory of all land parcels, incorporating missing parcels and eliminating duplicate claims, and an accurate measurement of all land and building characteristics, to be used in valuation. Pilot projects were begun under a USAID-assisted Provincial Development Action Program in 1972. RPTA was formally organized as a free- standing project in 1978. Since then, the project has been implemented in two phases. The first comprises the initial 43 months of implementation, during which the project was administered by the Ministry of Local Government and Community Development. In May of 1982, the program was shifted to the Ministry of Finance. The remaining months of project operations, through May 1983, are termed Phase II. Following official closing of the project, work began on revisions to the program. These el-forts, now in the field-testing stage, are termed Phase III, in this report. RPTA Procedures The RPTA project is typically executed by regular staff of the local assessors office, supplemented by contractual labor hired for the purpose with RPTA funds. Project inputs also include surveying equipment, vehicles, and training materials. While specific project implementation. varies between the projects and phases, the RPTA procedures follow a common sequence of four major steps: (1) preparation of base maps, (2) prefield "tieup"; (3) field work, and (4) post- field completion. Preparation of Base Map. Prior to tax mapping, the assessor's property record system typically consists of stacks of sworn statements, or tax declarations, (Form 1), which have been submitted over several years by claimants to property. These may have been supplemented by declarations containing the name and address of the property owner (if known), the address of the property (barangay, plus street address) and a description of the property's boundaries either in the form of a technical description (compass bearing and length of each boundary) or as a listing of the names of the owners of all adjoining properties. The tax declaration may also contain one or both of two identifying numbers: the title number, referring to the number assigned by the National Land l'itle and Deeds Registration Authority (NLTDRA) to the property's title document, or the cadastral number, assigned by the national Bureau of Lands (BOL) to the properl:y at the time it was originally surveyed or (if the product of subsequent subdivision), the number assigned by BOL at the time of subdivision approval. The tax declarations (as shown in Form 1) also contain space for the owner to declare the characteristics of the property, in sufficient detail as to permit the assessor to appraise the property without visual inspection. (These include, for example, class and area of land, floor area, and construction material of buildings.) Prior to going to the field, the RPTA team will attempt to link as many tax declarations as it can to parcels drawn on a base cadastral map. The team's first prefield task is therefore the preparation of the base map.  -8- The source information used in preparing the base map is obtained from the Bureau of Lands (BOL). BOL's principal mandate is to survey all lands in the public domain that have been declared alienable and disposable, and to initiate the process leading to issuance of title. To accomplish this, BOL contracts private surveying firms to carry out cadastral surveys in designated jurisdictions. This process consists, in essence, of (1) obtaining copies of surveys of any titled parcels in the jurisdiction, and identifying them, on a base map, and (2) surveying, mapping and monumenting the boundaries of all other claims (public and private) to land within the jurisdiction. This process yields a cadastral map of the jurisdiction; showing parcel boundaries (as yet without legal significance). Claimants are then invited to apply to BOL for issuance of patents; a process leading to the granting of title. As a permanent means of identification, BOL assigns each parcel a permanent cadastral code number, according to a nationwide numbering system. Many of BOL's cadastral maps are very old. The agency's cadastral mapping program was initiated during the early years of the American occupation and many of the cadastral maps date from before the Second World War. As BOL is not charged with updating these maps reflect subsequent subdivision (or monitoring the process of adjudication) the maps remain essentially unchanged from the time of the initial survey. BOL is, however, required to review new subdivision plans and privately-sponsored surveys, and to assign new cadastral numbers to these parcels. As a result BOL's regional offices are a repository of maps, from different years, at different scales, and varying degrees of accuracy. The task of preparing the RPTA base map consists of consolidating, on a uniform scale, the various maps and subdivision plans on file at the BOL. This task is accomplished by cartographers hired for the purpose under the RPTA program. The product of this task is an as yet incomplete base map, showing such parcel boundaries as are known, and the survey number assigned to each parcel. Prefield Tie-up. Having prepared the base map, the second step is to link as many tax declarations as possible to parcels on the base map. This process of "tieing up" relies largely on survey numbers listed on both the tax declarations and the base maps. Identifying data from each tax declaration are listed on the pre tax mapping control roll (Form 2). Where a parcel on the base map and one (or more) tax declarations have been matched, a check is placed against the parcel on the control roll and on the base map, signifying that the location of the parcel has been identified and that the tax mappers in the field need only verify the information (concerning ownership dimensions, title, and improvements). This process leaves (1) some parcels identified on the map with no corresponding tax declarations; (2) some parcels on the map with more than one corresponding tax declaration; and (3) some tax declarations on the control roll with no corresponding parcel on the map.  -9 - Field Work. The objective of field work is to "tie up" the remaining parcels and tax declarations and to verify irLformation on properties "tied up" in the prefield stage. Field work is generally undertaken by teams of four, consisting of a tax mapping leader, a tax mapping aide (responsible for sketching parcel boundaries on the base map), a data gatherer (who acts as custodian of the tax declarations and is responsible for verif'ying and updating the data on identified parcels and gathering appraisal and billing information on newly discovered parcels) and the barangay captain, who acts as an intermediary between the team and the local residents, and provides the team with information on property ownership. In the field, the team will meet with the owner (or occupant) of each parcel, walk the boundaries of the parcel, sketch the parcel on the base map, anid prepare a revised or new tax declaration (depending upon whether or not the parcel was tied-up in the prefield stage). No effort is made in the field to adjudicate or verify title, or to confirm the boundaries claimed by the occupant. As noted earlier, an owner's claim is considered sufficieni: to fix liability. Where claims overlap, the team sketches the boundaries of both claims, allowing both owners to be held liable for taxes on the overlapping area.- Where a tax declaration is found to be obsolete (i.e., where an original claimant has sold the property or has died), the tax declaration is cancelled. Where no parcel can be found to match a tax declaration, the declaration is also cancelled. In addition to measuring parcel boundaries and confirming the name of the claimant, the tax mapping team, while! in the field, prepares a tax declaration for the property, containing new or revised data on land area and use, floor area and construction materials:, type and age of improvements. (Tax declaration forms were generally used as appraisal records during t e first two phases of RPTA. A special purpose appraisal record, permitting a more elaborate appraisal, will be introduced under Phase III). Field operations are expected to take 6 months with each tax mapping team completing 20 parcels per day. 4/ In principle, the legal standing of a claim could be verified using title records on file at the Registrar of Deeds, provided a title has been issued. Title documents (Form 3) contain detailed technical descriptions of property boundaries and are indexed by BOI. survey number so as to permit titled to be found, even where no tax declaration exists. Even where BOL has completed a cadastral survey, however, titles may not have been registered. The titling process initiated by BOL can be held up at any of several stages in the administration or judicial process. Claimants themselves may prefer to arrest the! process at the point of which the Court (or the BOL) has instructed the NLTDRA (or the Registrar of Deeds) to issue title. Halting the process at this stage enables the owner to enjoy de facto tenure security, without the liability for charges and fees that would be 'Levied if the title document were actually picked up.  - 10 - Post Field Post field tasks consist of: (a) preparation of a final copy of the base map with each parcel assigned a permanent Property Identification Number (PIN) linking it to a new or revised tax declaration; (b) preparation of final copies of tax declarations, containing new or revised data on each property, and its PIN, linking it to the parcel map; (c) preparation of an assessment role (Form 4), listing all the properties, by PIN, with abstracts of property information (owner's name, assessed value of land and improvements) to be used by the assessor as an index to the tax declaration files and by the treasurer as a control for tax collection records; and (d) preparation of the assessor's portions of the Real Tax Order of Payment. The RPTOP (Form 5) is the form used to notify the taxpayer of his new assessment. For each taxpayer, the assessor completes the tax portion, showing the PIN and assessed value of each property claimed by the taxpayer; the form is then passed to the treasurer, who calculates the amount of tax due on each property, and then notifies the presumtive owner. Owners are allowed 60 days to contest the assessment by appeal to the local Bureau of Assessment Appeals. Failing that, the assessment is binding until revised by the assessor. A permanent record for each property, listing current owner, assessed value, and tax payments received is maintained by the Treasurer (Form 6). An abstract of property tax receipts is maintained, permitting monthly reports of revenues to be produced (Form 7). III. EVALUATION OF PROJECT In terms of its specific objectives, RPTA was a near-success. The program's original objective was to introduce the map-based system of discovery in 576 municipalities and 240 city districts over a five year period. At project completion, the RPTA had reached 70 percent of its coverage target. A. Achieving the Coverage Target Between initiation of RPTA in May 1978, and formal completion in May 1983, 571 projects were completed. Of these, 91 were projects taken over from the earlier PDAP technical assistance project, and completed under RPTA. RPTA in turn, began an additional 46 projects, which were still underway at the time of formal project completion. The rate of project completion varied between the two phases of the program's administration. During the 3 1/2 years of the first phase of RPTA, 244 projects were completed; an average of 70 per year. During the project's second phase (1 1/2 years) completions averaged 220 per year (Table 3). While this partly reflects the larger number  - 11 - of ongoing projects taken over in Phase III., it also reflects improvement in RPTA project design. RPTA's difficulty in achieving the original coverage targets in traceable to initial organization problems, in appropriate financing mechanisms and problems in central administration. Table 3: RPTA PROJECT COMPLETIONS Taken over Started and Total Taken over but Phase and finished finished finished not finished Phase 1 91 153 244 142 May 78 - Dec. 81 Phase 2: 142 185 327 Jan. 82 - May 83 Total Phases 1 233 338 571 142 and 2 Organization The program began in the wrong ministry. Phase I of RPTA was placed under the control of the Ministry of Local Government and Community Development (MLGCD). This ministry is responsible for monitoring local government compliance with national regulations, and was the lead ministry in the provincial development ad[ministration program (PDAP) which gave rise to RPTA. As discussed earlier, property tax assessment and collections are the responsiblility of local government assessors and treasurers. These officials are under the direct executive supervision not of MLGCD, but of the Ministry of Finance. Part of the delay in implementing Phase I of RPTA was due to the reluctance of local assessors to take direction from a rival ministry, particularly where the instructions issued by MLGCD (with respect to map specifications, numbering systems, etc.) conflicted with those issued by MOF. This problem was addresssed by shifting responsibility for RPTA from MLGCD to MOF, effective January 1982. Financing A second source of RPTA delay has been the structure of project financing. RPTA demands three kinds of expenditures by local government, over and above regular salaries and overhead of the local assessors office: (1) costs of contractual and casual labor, (principally involved in map making ad surveying); (2) costs of vehicles and equipment; and (3) incremental travel i  - 12 - costs, including per diems for regular staff in the field. These costs are incurred throughout project implementation, well before any project-related revenue increases become available. Under Phase I, RPTA financing was provided in two stages: A P30,000 initial grant disbursed at the time the local government was accepted into the program, and a second grant of P30,000 disbursed when the local government's project was certified as completed. (An additional P10,000 was to be disbursed upon completion of an undefined collection program.) Funding was intended to cover incremental costs incurred by regular staff. (Vehicles and equipment were to be provided directly by the central RPTA office). The funds provided under this arrangement proved to be too little, too late. Too late, because half the program funds were released only after the project was finished, thus requiring local governments to finance expenditures in the interim from their own resources. Too little, because the total amount of grant money equalled only a small fraction of the cost of implementing RPTA. (Incremental costs of the RPTA program averaged roughly P250,000 per project, over four times the amount of RPTA financial assistance.) As a result, RPTA projects were often halted, or slowed, as local governments ran out of money to meet program costs. During the time of MLGCD's tenure over RPTA, the Ministry of Finance was operating a separate financing plan for tax mapping. This plan, capitalized by a P50 million grant from the national government, and structured as a revolving fund, was authorized to finance up to 100 percent of total costs of tax mapping (including salaries of the regular staff of the assessor's office, to the extent they were involved in tax mapping). Funds from this source were to be disbursed: (1) 50 percent upon approval of a tax mapping implementation plan, with funds to be used for purchase of equipment; (2) 30 percent upon delivery of the equipment; and (3) the remaining 20 percent upon completion of 75 percent of the implementation plan. (MOF funds carried 0 percent interest, and 3-5 year repayment periods). A study of both tax mapping programs found (as of March, 1980) that MOF's revolving fund was much more heavily used than MLGCD's RPTA financing; the former accounting for 35 percent of total tax mapping expenditures, as opposed to 3 percent contribution by MLGCD. At the same time, self-financing by local governments was still the largest source of funds for tax mapping, accounting for 62 percent of the total. RPTA's financing system was changed under Phase II, to address these problems. The method for calculating the amount allocated to each project was changed from a flat P60,000 per project to a cost-sharing arrangement, with MOF financing 65 percent of project costs. The timing of disbursement was changed from the two-stage payment of Phase I to an ongoing cost-reimbursement process, with funds disbursed as work proceeded. A 1983 evaluation by Sycip, Gorres, Velayo and Company (SGVC) still found the timing of RPTA financing to be a source of delay, however. Under the new financing plan, MOF was to reimburse local governments for qualified expenditure, after making a request for funds to USAID, and receiving the funds from the Agency. This process, according to SGVC typically took three  - 13 - to five months. In the interim local governaent were forced to finance work from their own resources, halting projects when funds ran out. A new problem, moreover, had arisen with the use of fixed-share cost- reimbursement as a basis for calculating the amount of assistance given to each local government. Local governments had responded to the open-ended incentives implicit in the formula by increasing costs unnecessarily. The current plan for a third phase of RPTA calls for further changes in RPTA's financing plan, to address both these problems. To address the need for cost controls, the amount to be financed will be based on the number of parcels to be tax mapped. The amount of financing, per parcel, will be based on historical experience, under Phase I and 'EI. (These cost estimates cover contractual and casual labor, vehicles, equipment and supplies, and incremental travel cost of reguLar staff). The share of estimated costs to be financed will range from 50 percent to 70 percent under AID-assisted projects; and will be 90 percent under IBRD-assisted projects. To address the problem of delayed reimbursements, the project incorporates a special account, into which an amount equal to 90 days expenditures are deposited. Reimbursement to local governments will be made directly from this account, and funds will be regularly replenished by the lenders as work proceeds. Central Management In addition to initial organization and financing problems, problems in execution of central management functions Fprevented RPTA from achieving its entire coverage target. Central management of RPTA has three groups of functions: (a) Screening applicants for inclusion in the program; (b) Developing procedures and training lccal staff in their use; and (c) processing procurement and disbursement applications and monitoring and evaluating execution. Much of RPTA's initial problems arose in attempting to carry out these functions in too many jurisdictions with too few experienced staff. At the time of the first external evaluaticn of RPTA Phase I (Syracuse University, May 1980) the numbier of central administrative staff assigned to the program totalled 37. This compliment consisted, however, of eleven inexperienced middle-level staff coordinating requests and monitoring implementation; eleven trainers; one lawyer advising on procedures, and thirteen messengers and file clerks. When the program was transferred to the Ministry of Finance, this staf-f was transferred with it. Within Phase II, RPTA benefitted from existing experienced management in MOF in the person of the director of the Office of Local Government Finance (OLGF) and her immediate assistants. These staff continued to hold responsibilities for other OLGF functions, while attempting ta manage the program. As a consequence, the shortage of experienced, more senior staff persisted.  - 14 - The Syracuse evaluation recommended, as one means of addressing the evidence, that the number of projects be reduced. This recommendation was not implemented in Phase II. Screening procedures, aimed at weeding out potential problem projects were, however, introduced in Phase II. These envisioned an application process in which local governments would first seek approval from their respective provincial assessors, prior to applying to central management for inclusion in the project. Criteria to be used by provincial assessors were never specified, however. As a result, projects continued to be passed, uncritically, on to central management. Steps have been taken under Phase III to deal with this problem. First, the number of projects to be processed in a given year has been reduced. Phase III calls for tax mapping 143 projects over 2 1/2 years; about one-quarter the rate targetted for Phase II. Explicit criteria intended to screen out potential problem projects have been introduced. These require, inter alia, that jurisdictions applying to the program must: (1) already have BOL base maps, or be willing to pay for base map preparation separately from the RPTA project; (2) have demonstrated capacity to supply counterpart funding, counterpart staff, and office space; and (3) have a minimum of 4000 properties (the minimum needed to justify the mobilization costs of the program). On the staffing side, the number of higher-level professional staff at headquarters will be increased to 15 (despite an overall drop in headquarters staff to 28). Central headquarters will consist of a project manager; a technical unit responsible for reviewing project proposals, orchestrating technical assistance, and monitoring implementation; and an administrative unit responsible for program budgeting, procurement, disbursement, and personnel management. A separate training and tax education unit at the central level would be responsible for developing training materials, but would not conduct courses itself. The role of project staff at the regional level would be expanded to include screening (according to clearly designed criteria), provision of training and technical assistance, and monitoring implementation. The format of training will also be improved under Phase III. The Syracuse evaluation of Phase I concluded that the training program was too strongly oriented toward large, short term off-the-job seminars. At the time of the Syracuse evaluation, RPTA's training program consisted of: (a) 2 training sessions on tax mapping per region, with 200 persons per session and a duration of one week; (b) periodic 3 day refresher courses, concentrating on field work, with smaller groups (20-30 persons); and (c) 3-day lectures on collections, in sessions of 60-100 persons. This formal training was supplemented by spot visits by MLGCD staff, averaging 2.5 visits per project per year. The Syracuse study recommended that training be reoriented to increase spot field visits, and reduce the number of large seminars. This  - 15 - recommendation was not implemented in Phase II. Phase III, however, incorporates an initial "capacity building" phase during which manuals and other training materials will be prepared, and training courses designed. As of this writing, products of this phase have not been completed, and could not be evaluated. B. Achieving the Assessment Target Completion of the RPTA program was expected to produce an average increase in assessments of 50 percent. The precise measurement of RPTA's assessment impact presents a problem of definition. As noted earlier, the program has three potential effects. (1) the discovery of undeclared land parcels, which increases total assessments; (2) the valuation of declared properties based on more accurate informiation, which again increases assessments; and (3) removal of invalid records which decreases total assessments. As shown in Table 4, decreases in assessment due to removal of invalid records has a significant impact on total assessed value, particularly in municipalities. On the average for municipalities under Phase II, the removal of invalid records reduced total assessments by 10.3 percent. If this negative effect is included in the calculation of RPTA's impact the average net effect of RPTA is an increase of 40 percent in municipalities; and only 13.1 percent in cities. A good case can be made for excluding the invalid parcel effect from the calculation of RPTA's impact. If invalid records are in effect uncollectible, their removal has no impact on total size of tax base that is effectively taxable. Recalculating RPTA's imapact on assessments, using only valid records as the denominator, yields the percentage increases shown in the bottom line of Table 4. The sources of growth in assessed value, as shown in Table 4, are surprising. Contrary to expectations, the major source of growth in assessments is not the discovery of new parcels, but rather the revaluation of existing ones. In cities, new discoveries produced only a 4.1 percent increase in total assessments, compared to a twelve percent increase due to revaluations. In municipalities, new discoveries increased total assessments by 9 percent; revaluations increased total assessments by 42 percent. This suggests that the basic diagnosis underlying RPTA is somewhat off-target; that reform in the system of discovery is needed not so much to bring missing parcels onto the rolls, but rather to improve the accuracy of data used in valuation. C. Measures Needed to Achieve Sustained Yields RPTA can be evaluated in terms of its specific objectives, i.e., the number of jurisdictions tax mapped, and its impact on assessments. By these criteria, the program's first two phases were a moderate success, with  - 16 - Table 4: SOURCES OF CHANGE IN ASSESSED VALUES RPTA - PHASE II Impact on Total Assessed Value Source of Change Cities Municipalities Total Invalid records removed -2.7 -10.3 -7.2 New land parcels discovered +4.1 +9.2 +6.9 Reappraisals of existing properties +11.7 +41.5 +28.0 Increase in total assessments +13.1 +40.4 +27.7 Increase in valid assessments +16.3 +56.5 +37.5 prospects for further improvement under the planned third phase. But the project also needs to be evaluated in terms of its ultimate impact on the level and sustainability of property tax revenues. By these criteria, the success of the project is less clear; there appears to be a need for expanding the scope of RPTA to address other areas of weakness in the tax's administration. More importantly, RPTA's efforts to improve administration must be accompanied by changes in property tax policy. It is centrally- mandated policies, as much as poor administration, which limit the yield and bouyancy of property taxes. Improving Assessment Techniques The scope of technical assistance provided under the project needs to be expanded in several respects. The first consists of administrative improvements in assessment techniques. As noted earlier, the Philippines uses a mass appraisal technique to value property. The system relies on two principal inputs: (1) data on the physical characteristics of each property to be valued, and (2) a schedule of values to convert the physical characteristics to estimates of market value. The RPTA program does improve the accuracy of data on the physical characteristics of property. During the field work phase, the RPTA team remeasures and verifies the physical data reported on each tax declaration. RPTA does not, however, address the second input to valuation, the schedules of values. By law, new schedules of values must be prepared by the local assessor prior to each general revaluation. MOF regulations specify the sources of information to be used in calculating the schedules. To calculate the unit value of land, the assessor may draw on (1) sales data provided by the Registrar of Deeds or by notaries public (both of whom are required by law to furnish the assessor with the information); (2) sworn statements (required  - 17 - under PD 464); or (3) opinions of real estate agents. Procedures by which the assessor divides sales data into subclasses of property, rejecting non- representative observations and then calculating the price per square meter of land in a given subclass are slpecified in guidelines issued by MOF. Schedules of values for buildings are to be calculated by estimating the cost of constructing buildings of various types, using cost information provided by lumberyards and ]hardware stores, and standard ratios of labor costs to materials costs. Unit values are then calculated by dividing total construction costs by floor area, for each of the various building types. The schedules of values produced by these procedures are consistently low. This is in part because the market data on which the calculations are based is suspect. In calculating land values, assessors cannot rely on official sources. The sales data collected by the Registrar of Deeds is unreliable, as sellers normally understate the value of transactions in order to minimize taxes on property transactions. The sworn statements filed by property owners are similarly untrustworthy. Assessors must therefore rely on the opinions of local real estate agents. In an environment where political pressure can be brought to bear against these agents, and where reliable official data is not available to buttress the assessor's position, a downward bias in schedules of values is difficult to a.void. Institutionalizing Maintenance Action is also required to improve the system for maintaining current records on the property tax base. The base of the property tax is subject to two major types of change: (a) changes in the physical characteristics of the base; e.g. the creation of new parcels through subdivision and the construction or demolition of buildings; and (b) changes in the unit cost of land and improvements. At present, two mechanisms are used to update property tax records: (1) periodic general revaluations, in which a new schedule of values (reflecting price changes) is calculated and each parcel record is reviewed (to discover changes in physical characteristics and ownership); and (2) interim revisions to individual property records, in which changes in physical characteristics and ownership are incorporated in the property records as they are discovered by the assessor. As noted earlier, PD 464 mandates a general revaluation every three years, but in practice, this requirement is not observed. At the same time, the system used to discover interim changes also functions poorly. It relies, at its peril, on other agencies of government and on the cooperation of taxpayers. To discover new subdivisions, it depends upon purchasers (or the original subdivider) to register the titleEi to each new parcel, and on the Registrar of Deeds to communicate this information to the assessor. New construction is not discovered unless the person commissioning the work seeks a building permit, and the agency issuing the permit informs the assessor.  - 18 - Despite the physical proximity of th5e assessor's office to the deeds registries and building permit agencies - this information pooling does not occur routinely. RPTA's impact on maintenance systems is limited. In shifting the system of discovery from one based on owner declarations to one based on maps, RPTA provides a geographical base for future general revaluations. It also simplifies the task of linking reports of changes in property characteristics to specific property records. It does not, however, affect government's willingness to conduct future general revaluations on schedule, nor does it offer improvements in the system of interim maintenance. It does not, finally, provide any mechanism for adjusting valuations to reflect general increases in prices. It is therefore likely that gaps in geographical coverage and obsolescence in building and ownership data will reemerge in RPTA jurisdictions, and that the existing problem of obsolete prices will persist. Tax Rates and Other Policy Variables Changes in tax rates and other policy variables will also be required to convert the property tax into a buoyant, high-yield source of revenue for local government. The nominal tax rates imposed in the Philippines are fairly typical of countries defining property value on a capital basis. The minimum rate (including the Special Education Fund surcharge) is 1.5 percent; the maximum is 2.5 percent in provinces and 3.0 percent in cities. The effective rate of taxation is, however, considerably lower. Legally mandated assessment ratios reduce the value of the base to which the tax rate is applied. Under PD 464, assessment ratios are fixed by use and value class. Residential land, for example, is to be assessed at 30 percent of market value. Residential improvements are to be assessed at 15 percent to 80 percent of market value, depending on value class. As a result, the maximum legally-permitted rate of tax on residential land is 0.75 percent in provinces and 0.9 percent in cities. The maximum legally-permitted rate on residential improvements can be as low as 0.38 percent in provinces and .45 percent in cities. When combined with the underestimates of market values imbedded in the schedules of values and the failure of maintenance systems to adjust for inflation, the effect is a level of property taxation in the Philippines that is approaches insignificance. In round terms, the maximum rate of taxation on land, measured as a percent of current (July, 1987) market value, is 0.15 percent in provinces and 0.18 percent in cities. The maximum tax rate on the lowest value class of residential improvements is 0.075 percent in provinces and 0.09 percent in cities. Authority to increase the effective rate of property taxation rests largely with the central government. It is central legislation that fixes ----------------------------------- 5/ The Registrar of Deeds maintains an office in every city and every provincial capital.  - 19 - maximum tax rates, and specifies assessment ratios. The central government also controls the date on which new schedules of values (and thus inflation adjustments) become effective. Because RPTA deals exclusively with administrative problems at the local level, it does not address these centrally imposed constraints on the yieldsi of the property tax. Because these constraints persist, RPTA's impact on the absolute level of property tax revenues in the Philippines is small. Improving Collection Efficiency Action is also required to improve collection efficiency. Phase I and II of the RPTA had no specific component to improve collection performance. One would expect, however, that tax mapping, and reappraisal under RPTA would increase collections. Collection efficiency (current revenues as percent of current: billings) should increase, due to the removal of invalid (and presumably uncollectible) property records from the tax rolls. Actual revenues should increase as the assessed value of the tax base increases. Analysis of RPTA's impact on collections is difficult. Aggregated data on collections is available only for projects completed under Phase I. Of these, only projects completed in 1980 and 1981 offer a valid basis for comparison owing to the distorting effect of the 1979 revision on earlier projects. In addition, data provided by local treasurers on growth in current demand (or "collectibles") is inconsistent wiith assessor's figures on growth in assessments. (This reflects in part, the practice of adding newly- discovered or reassessed properties onto the tax rolls as they occur. Where an RPTA project runs over several years, this practice produces an underestimate of the program's impact on carrent demand when only the last pre-RPTA year and the first post-RPTA year are compared). To the extent the data permits any conclusion, it appears that RPTA's expected impact on collection do not occur. In the 76 projects providing data, collection efficiency dropped in the first post RPTA year; from an average of 56 percent pre-RPTA, to an average of 50 percent post-RPTA. Total collections (on current account) increased by only 1.1 percent, despite an increase in valid assessments of 37.5 percent:; and an increase in collectibles (as reported by treasurers) of 13.6 percent. Data on distribution of projects by percent change in collections (Table 5) shows a clustering of projects in the lowest two brackets (net decrease, increase of less than 10 percent). Only 24 of the 76 projects exceeded a 10 percent increase in collections; only 13 on increase of over 30 percent.  - 20 - Table 5: FREQUENCY DISTRIBUTION OF PROJECTS BY PERCENTAGE CHANGE IN COLLECTIONS Change in Annual N of Project Percent of Collections Projects Projects Net Decrease 35 46 Increase of: 0 - 10% 17 22 10.1 - 30% 11 15 30.1 - 60% 12 16 Over 60% 17 1 Total 76 100 Problems in collection are not due to a lack of enforcement mechanisms. The Tax Code authorizes treasurers to enforce payment of delinquent accounts by: (1) distraining the personal property of the taxpayer; (b) seizing and selling the property itself; and (c) suing the taxpayer in the court of appropriate jurisdiction. The treasurer is not obliged to formally notify the delinquent taxpayer before pursuing any of these remedies6/. In addition, the Code effectively precludes a taxpayer from witholding tax payment pending resolution of any legal challenges to the assessment. Despite ample legislative support, enforcement actions against delinquent accounts are not routinely taken. To some degree, this is due to administrative problems in the office of the treasurer. In smaller jurisdictions, treasurers reportedly have difficulty identifying delinquent accounts and monitoring the progress of enforcement proceedings. (In one jurisdiction visited by the mission, individual payment records were kept in an undersized file cabinet. To find delinquent accounts, the treasurer had to sort through the entire file; a task undertaken annually). Phase III of RPTA will address this problem by introducing a new records management system with a separate procedure for processing delinquent accounts and a reporting system capable of monitoring collection efficiency. But the overriding factor underlying weak collection enforcement is the reticence of local treasurers and strength of local political opposition. Treasurers, particularly in smaller towns, find the more drastic forms of enforcement (distraint, seizure and sale) too politically conspicuous and the more restrained legal procedures too cumbersome and expensive. 6/ The Code requires only that the treasurer post of notice of delinquency in a "public and conspicuous place" in each barangay and publish said notice in a newspaper of general circulation once a week for three weeks.  - 21 - D. The Question of Cost-Effectiveness Is RPTA a good investment for the Philippines? If viewed as a financial decision, the answer is a qualified "yes." The costs of RPTA are difficult to determine. The official project costs of Phase I and II totalled US$7.2 milLion (about P54 million at the average exchange rate prevailing during the implementation period). The amount includes the cost of equipment, training and technical assistance procured by the project's central administration and the cost of grants to local governments under the Phase I and Phase II formulas. These grants covered part of local governments' expenditures on contract and casual labor; materials and supplies, and the salaries, overhead, and travel expenses of regular staff assigned to the project. The remaining part of project costs, paid from local goverrnments own resources, is not known. Data based on individual project compLetion reports show a total cost of "field operations" of P57.2 million. This figure is based on 488 of the 571 completed projects. Extrapolating from this sample suggests a total project cost of P66.9 million (US$8.9 million). The definition of "field operations" is, however, not supplied in the source documents. The only detailed figures on project costs are estimates prepared for RPTA Phase III. Costs of tax mapping and "records conversion" (meaning in this case, valuation) are estimated at P46.6 per parcel in prices of 1984. These costs cover contractual and casual labor, equipment, materials, and salaries, overhead, and travel costs of local personnel. The definition of "iparcel" however, varies among project documents, in some cases referring to land parcels; in others referring to "real property units" (RPU's); a unit of measure which counts improvements and machinery separately from the land on which they stand. Pending clarification from RPTA staff, the costs of the program can only roughly be estimated at P40 per land parcel, in prices of 1984 (Table 6). Table 6: ElPTA PROJECT CO3T ESTIMATE (Cost per Land Parcel) Source Prices of 1980 1984 AID project audit 15 31 RPTA project completion reports 16 33/a RPTA Phase III proposal 22 46 Estimate (IBRD) 407a /a includes cost of contract and casual labor, vehicles and equipment, and the salaries, travel, and overhead of reguLar staff assigned to the project.  - 22 - The financial rate of return on RPTA expenditures are, as noted earlier, negligible. The 1.1 percent increase in actual collections under Phase II is equal to roughly P0.30 per parcel. At this rate, a local government would have to wait 140 years to recover (without interest) the costs of the project. If pre-RPTA collection performance could, however, be sustained on post-RPTA assessments, the financial return would improve considerably. The 37 percent increase in valid assessments in Phase II represents an average increase of about P2000 per land parcel. At an average tax rate of 2.5 percent, this would yield an increase in tax liability of about P50 per land parcel. Assuming a sustained collection rate of 50 percent, actual tax revenues would increase by P25 per land parcel per year, an amount sufficient to cover the full cost of RPTA in slightly less than two years. A change in central government restrictions on assessment ratios, tax rates, and revaluation schedules would further shorten the payback period. If, for example, the planned 1984 general revision had been made effective on schedule, the payback period for a 1984 RPTA project would be about ten months. Financial returns are a poor guide to public policy in this case, however. While increased tax revenues generate benefits in the form of increased public services, they also represent a loss in benefits due to their impact on private sector consumption. This loss is at least equal to reduction in private utility resulting from the transfer of resources from the private to the public sector. To the extent the tax increase induces changes in the pattern of private consumption, the loss in benefits may be even greater. The size of the net benefit created by this transfer of resources from taxpayers to government is difficult to determine. From a policy perspective, the more manageable question is: given the decision to increase the transfer of resources from taxpayers to local government, is RPTA the best way to do it? Here again the answer is a qualified "yes." RPTA is designed to accomplish its impact on revenues by increasing the tax burden on those who presently underpay. As designed, the burden of RPTA-increased taxes should fall on property owners who presently fail to file tax declarations, or who underreport the characteristics of their property. By placing the burden on tax evaders, RPTA can, in theory, increase revenues by improving equity. This desirable condition exists, so far, only in theory. With its negligible effect on actual revenues, RPTA now succeeds only in shifting the point of evasion from assessments to collections. In addition, given existing policy constraints on tax yields, the revenue effect of even a comprehensive improvement in property tax administration would be very limited.  -23 - Form 1 PROI'ERTY INDEX NO . ................................... RPA Form No. I TAX DECLARATION NO. ,. ..... . ..... DECLARATION OF REAL PROPERTY (FILED UNDER PRESIDENTIAL DECREE NO. 464) (Address) D Administrator .................................... ......................... .................... (Addressl DESCRIPTION AND OTHER PARTICULARS OF PROPERTY c Location of Property ........................................... ............................ .....I........................... (Number and Street) (Barangay/Distnict) (Municipality City;Province) Certificate of Title No . ......................... Cadastral Lot No .Assessor's Lot No. c Boundaries: Block No. .u North: ..... South: East: . ..... West: . (State streets, ' )ts, or streams by which bounded or namrns of owners of adjoining lands) I (b) LAND (AGRICULTURAL/M4INERAL) OWNER'SDECLARATION ASSESSOR'S FINDINGS ! Kind | Area Value_ Kind Area Class Unit Values Market Value o - - -- -- -- -- -- -- - - - - ~~~~~~~~~~~~~~-7- - ---- r ------_ _-_-_-_-____--- ~'~---~ --~- - ~-~~ ----~ o ----___ __- __ _----I ___ _ ____ ___-