93689 CITY PLANNING LABS A CONCEPT FOR STRENGTHENING CITY PLANNING CAPACITY IN INDONESIA PREPARED BY THE CITY FORM LAB, SINGAPORE UNIVERSITY OF TECHNOLOGY AND DESIGN (SUTD) FOR WORLD BANK INDONESIA © 2013 The International Bank of Reconstruction and Development/ The World Bank East Asia and Pacific Region/East Asia Infrastructure Sector (EASIS) 1818 H St., NW Washington, DC 20433 USA All rights reserved This volume is a joint publication of the staff of the International Bank for Reconstruction and Development/The World Bank and the Australian Aid. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of the World Bank, the governments they represent or of Australian Aid. The World Bank does not guarantee the accuracy of the data included in this work. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. All queries should be addressed to the Task Team Leader, Thalyta Yuwono: The World Bank Jakarta Office Indonesia Stock Exchange Building Tower II, 12th Floor. Jalan Jenderal Sudirman Kav. 52-53, Jakarta 12190, Indonesia e-mail: tyuwono@worldbank.org. Disclaimer The views expressed in this publication are those of the authors and not necessarily those of the Australian Aid. CITY PLANNING LABS A CONCEPT FOR STRENGTHENING CITY PLANNING CAPACITY IN INDONESIA PREPARED BY THE CITY FORM LAB, SINGAPORE UNIVERSITY OF TECHNOLOGY AND DESIGN (SUTD) FOR WORLD BANK INDONESIA TABLE OF CONTENTS Acknowledgement................................................................................................................................ v Abbreviation and Acronyms ............................................................................................................. vi Executive Summary ............................................................................................................................ vii Introduction ............................................................................................................................................1 1.1 Background ..........................................................................................................................3 1.2 Rationale ...............................................................................................................................5 1.3 Objectives .............................................................................................................................6 1.4 Scope of Activities ...............................................................................................................7 Sector Module A: City Planning Labs and Spatial Growth Analytics ........................................9 2.1 Background ....................................................................................................................... 11 2.1.1 Context ...................................................................................................................... 11 2.1.2 Implementing P3N Technical Assistance through City Planning Labs ............. 11 2.2 Objectives .......................................................................................................................... 12 2.3 Scope of Activities ............................................................................................................ 13 2.3.1 Establishing the City Planning Lab ........................................................................ 14 2.3.2 Spatial Growth and Change Analytics ................................................................ 18 2.3.3 Planning Enforcement .............................................................................................. 21 2.4 Risks .................................................................................................................................... 24 2.5 Outputs ............................................................................................................................... 24 2.6 Team and Timeline ........................................................................................................... 26 Sector Module B: City Economic Competitiveness ....................................................................... 27 3.1 Background ....................................................................................................................... 29 3.2 Objectives .......................................................................................................................... 29 3.3 Scope of Activities ............................................................................................................ 30 3.4 Risks and Mitigation ......................................................................................................... 37 3.5 Outputs ............................................................................................................................... 37 3.6 Team ................................................................................................................................... 37 3.7 Resource Allocation and Timeline .................................................................................. 38 Annex Module B: Assessment of Data Environment ................................................................ 39 Sector Module C: Slum Analytics and Management Systems ................................................... 41 4.1 Background ....................................................................................................................... 43 4.2 Objectives .......................................................................................................................... 44 4.3 Scope of Activities ............................................................................................................ 44 4.4 Risks and Mitigation ......................................................................................................... 49 4.5 Outputs ............................................................................................................................... 49 i 4.6 Team ................................................................................................................................... 50 4.7 Timeline .............................................................................................................................. 50 Annex Module C: Data Collection ............................................................................................. 51 Sector Module D: Disaster and Climate Resilient Planning Analytics ...................................... 55 5.1 Background ....................................................................................................................... 57 5.2 Objectives .......................................................................................................................... 58 5.3 Scope of Activities ............................................................................................................ 60 5.4 Risks and Mitigation ......................................................................................................... 63 5.5 Outputs ............................................................................................................................... 63 5.6 Team ................................................................................................................................... 64 5.7 Timeline .............................................................................................................................. 64 Annex Module D: Data Collection ............................................................................................. 65 Sector Module E: Monitoring Land and Real Estate Markets ................................................... 67 6.1 Background ....................................................................................................................... 69 6.2 Objective ........................................................................................................................... 70 6.3 Scope of Activities ............................................................................................................ 70 6.4 Risks and Mitigation ......................................................................................................... 75 6.5 Outputs ............................................................................................................................... 75 6.6 Team ................................................................................................................................... 76 6.7 Timeline .............................................................................................................................. 76 References .......................................................................................................................................... 77 Annex 1: Demonstration Report of Spatial Growth Analytics Module ..................................... ix ii LIST OF TABLES Table 1. Characteristics of Pilot Cities .............................................................................................5 Table 2. Scope of Activities and Timeline .................................................................................... 30 Table 3. Example of Analysis: Ratings of Indonesian Cities on Economic Performance (2000 – 2010) .................................................................................................................................. 35 Table 4. Risks and Mitigation of Economic Competitiveness Module ...................................... 37 Table 5. Inventory of Data for Economic Competitiveness Module......................................... 39 Table 6. Inventory of Data for Disaster and Climate Resilient Module ................................. 65 Table 7. Example Dataset: Price Range of Flats Offered by Housing Development Board in Singapore (in Thousand SGD) .................................................................................................... 72 LIST OF FIGURES Figure 1. Screen Capture of a QGIS Open-source Data Platform Work Environment ...... 15 Figure 2. City Planning Lab Partnership Framework ................................................................. 16 Figure 3. City Planning Lab Staffing............................................................................................. 17 Figure 4. Example Analysis Output: Accessibility to Jobs ......................................................... 18 Figure 6. Methodology Illustration: Export Performance Tool.................................................. 35 Figure 7. Methodology Illustration: Value Chain Mapping Tool .............................................. 36 Figure 8. Methodology Illustration: Cost Structure Analysis Tool ............................................. 36 Figure 9 A,B. Example of Analysis: Sao Paulo HABISP Online Housing Information System ............................................................................................................................................................. 47 Figure 10 A,B. The InaSAFE Tool.................................................................................................... 59 Figure 11. Example Analysis: Distribution of Building Types in Singapore ........................... 71 iii iv ACKNOWLEDGEMENT The City Planning Labs provide a concept to build capacity for an integrated, evidence- based spatial planning and investment decision making to help cities in Indonesia to achieve sustainable and inclusive economic growth. This concept was prepared through a consultative process in Indonesia which included meetings with Central and Local Government authorities and site visits to pilot cities. The City Planning Lab Concept has been prepared by a core team led by Thalyta Yuwono (EASIS) in collaboration with The City Form Lab, Singapore University of Technology and Design. Andres Sevtsuk and Reza Amindarbari from The City Form Lab were the main authors of this concept. Inputs have been provided by Chandan Deuskar (EASIN), Renata Simatupang (EASIS), Connor Spreng (EASFP), and Pranav Kumar (FCDKP), under the guidance of Taimur Samad (EASIS) and Nathan Belete (Sector Manager, EASIS). Wilmar Salim and Ari Kuncoro, consultants, provided important contribution to the preparation of this concept. The team benefited from wide range of consultation with the Government of Indonesia: Ms. Hayu Parasati (National Planning Agency/Bappenas), Mr. Basuki Hadimuljono (Ministry of Public Works), Mr. Dadang Sumantri Mochtar (Ministry of Home Affairs), Mr. Dodi Sukmayadi Wiradisastra (Geospatial Information Agency/BIG); and also with the Mayors and local agencies in Surabaya, Denpasar, Balikpapan and Palembang. The team greatly appreciates technical contribution from various stakeholders who were consulted during the preparation of this concept. Finally, the team would like to acknowledge the generous support provided by Australian Aid. v ABBREVIATION AND ACRONYMS Bappeda Badan Perencanaan Daerah/ Local Planning Agency Bappenas Badan Perencanaan Nasional/ National Planning Agency BMKG Badan Meteorologi, Klimatologi, dan Geofisika/ National Agency for Meteorology, Climatology, and Geophysics BNPB Badan Nasional Penanggulangan Bencana/ National Agency for Disaster Management BIG Badan Informasi Geospasial/ National Agency of Geospatial Information BPBD Badan Penanggulangan Bencana Daerah/ Regional Agency for Disaster Management BPN Badan Pertanahan Nasional/ National Land Agency BPS Badan Pusat Statistik/ Statistics Indonesia C/K City (urban municipality) and Kabupaten (rural municipality) CCA Climate Change Adaptation CPL City Planning Labs DRR Disaster Risk Reduction GDP Gross Domestic Product GIS Geographic Information System GRDP Gross Regional Domestic Product IDR Indonesian Rupiah IFC International Financial Corporation KPI Key Performance Indicators KPPOD Komite Pemantauan Pelaksanaan Otonomi Daerah/ Regional Autonomy Watch LMA Land Market Assessment MOF Ministry of Finance MoU Memorandum of Understanding MP3EI Masterplan Percepatan dan Perluasan Pembangunan Ekonomi Indonesia/ Masterplan for Acceleration and Expnsion of Indonesian Economic Development MPW Ministry of Public Works MUDP Metropolitan and Urban Development Program, or P3N NGO Non-governmental Organization P3N Program Pembangunan Perkotaan Nasional, or MUDP RPJMN Rencana Pembangunan Jangka Menengah Nasional/ National Midterm Development Plan RT Rukun Tetangga RW Rukun Warga SGD Singapore Dollar SME Small and Medium Enterprises SNDB Subnational Doing Business Report SNG Subnational Government vi EXECUTIVE SUMMARY The cities that emerge from Indonesia’s rapid urbanization will be key determinants of the country’s overall economic development and competitiveness, as well as their inclusiveness and environmental sustainability. However, without strategically planned investments, policy interventions, and institutional capacity, mismanaged urbanization could become an obstacle to sustainable growth. Indonesia has been no exception to the rapid urbanization experienced in many East Asian countries. With average annual urbanization rate estimated at 4.2% between 1993 and 2007, Indonesia is urbanizing faster than its Asian counterparts. This has made Indonesia one of the most urbanized countries in Asia, with an urban population share of 51% in 2011. Projections of urbanization suggest that this figure will increase to 68 % by 2025. However, Indonesia has yet to achieve the economic returns to urbanization that other countries have achieved. For every additional 1% that the country urbanizes, it achieves just 2% of additional GDP growth, whereas other countries in the region achieve 6-10% GDP growth per 1% of urbanization. Under the Metropolitan and Urban Development Program (MUDP/P3N), currently under preparation, the World Bank is engaging directly with large cities through investments in transformative infrastructure. The Bank has initiated direct engagements with local governments, targeting large and medium cities and metropolitan areas with populations over 500,000 to prepare and facilitate investments in transformative infrastructure. In addition to investment support, a key component P3N is building technical and institutional capacity in cities and metropolitan authorities, which will take the form of City Planning Labs. The City Planning Lab (CPL) is envisioned as the driver of improved integrated and evidence-based spatial, development and investment planning. The City Planning Labs core module will be initially implemented in four cities: Surabaya, Palembang, Denpasar and Balikpapan, with two additional modules in each city. In the short term, the CPL will (i) provide “just in time”, demand driven data and analysis that can feed into immediate decisions, and (ii) streamline ongoing urban management functions, such as building permitting and tax-related functions. In the medium term, it will provide cost-effective analytics to cities that can feed into planning and investment decisions, reducing the expense involved in contracting consultants during each planning cycle. In the long term, the CPL will build local technical capacity, by gathering expertise from Indonesia and international sources to work closely with local staff. Over time, external involvement will diminish as local capacity strengthens. vii The proposed activities of the CPL will be conducted in modular fashion, each pertaining to a different sector. The proposed sector modules are: A. Instituting the City Planning Lab & Spatial Growth Analytics (Core Module) B. City Economic Competitiveness C. Slum Analytics and Management Systems D. Climate and Risk Resilience Planning Systems E. Monitoring Land and Real Estate Markets While the details of the activities will differ, they will all take a common approach, which will involve (i) data gathering; (ii) inputting new and existing data into an integrated cross-sectoral data platform; (iii) using data in ongoing urban management functions; (iv) analyzing the data; and (v) working with city leaders to help them use the insights from data analysis in planning and decision-making. viii INTRODUCTION 1 2 1.1 BACKGROUND The cities that emerge from Indonesia’s rapid urbanization will be key determinants of the country’s overall economic development and competitiveness, as well as their inclusiveness and environmental sustainability. There is reason to be cautiously optimistic about Indonesia’s urban future. However, without strategically planned investments, policy interventions, and institutional capacity, mismanaged urbanization could become an obstacle to sustainable growth. Indonesia has been no exception to the rapid urbanization experienced in many East Asian countries. With average annual urbanization rate estimated at 4.2% between 1993 and 2007, Indonesia is urbanizing faster than its Asian counterparts, such as China (3.8%), India (3.1%) and Thailand (2.8%). This has made Indonesia one of the most urbanized countries in Asia, with an urban population share of 51% in 2011. Projections of urbanization suggest that this figure will increase to 68 % by 2025. These statistics tell a powerful story of structural transition in Indonesian society, from predominantly rural and agricultural society into more urban, manufacture and service based economy. However, Indonesia has yet to achieve the economic returns to urbanization that other countries have achieved. For every additional 1% that the country urbanizes, it achieves just 2% of additional GDP growth, whereas other countries in the region achieve 6-10% GDP growth per 1% of urbanization. Under the Metropolitan and Urban Development Program (MUDP/P3N), currently under preparation, the World Bank is engaging directly with large cities through investments in transformative infrastructure. The Bank has initiated direct engagements with local governments, targeting large and medium cities and metropolitan areas with populations over 500,000 to prepare and facilitate investments in transformative infrastructure. In addition to investment support, a key component P3N is building technical and institutional capacity in cities and metropolitan authorities, which will take the form of City Planning Labs. The City Planning Labs core module will be initially implemented in four cities: Surabaya, Palembang, Denpasar and Balikpapan, with two additional modules in each city. CITIES: Surabaya: Surabaya is the second largest city in Indonesia, and the capital of East Java Province. The city has become one of the main ports of Java, which connects the western to eastern part of Indonesia. Surabaya comprises of 31 kecamatan (sub district), with total area of 326.81 Km2. The city is the core of Gerbangkertosusila metropolitan (Gresik, Bangkalan, Mojokerto, Surabaya, Sidoarjo and Lamongan), with estimated total metro population of 9.1 million people. In 2010, the population of Surabaya was 2.76 million people, with population density of 8,462 people per Km2. The average population growth rate from 2000 to 2010 is 3 0.63% annually, and average size of household is 3.6 people per household. Its unemployment rate in 2010 was 10%, which was higher than the national average. The economy of Surabaya is dominated by hotel, trade and restaurant sector (43%), followed by manufacture (22%) and transport and communication (10%). GRDP per capita in current price for 2010 was IDR 64,279,710 (USD 6,766), which was significantly higher than Indonesia’s GDP per capita (USD 2,850). The city has experience relatively constant economic growth in the last five years. In 2010, the economic growth was 7.1%, which was higher than national growth rate of 6.1%. The economy is expected to continue to grow, albeit at slightly lower rate, since Surabaya is struggling to create jobs for the existing work force and immigrants that come into the city. Palembang: Palembang is the capital of South Sumatera province. The city comprises 16 kecamatan (sub-district), with a total area of 400.61 Km2. Palembang borders Kabupaten Banyu Asin to the east, west and north, and Muara Enim to the south. The topography of Palembang is mostly flat lowlands, located at 8 meter above sea level. There are four rivers passing through the city: Musi (the largest), Komering, Ogan, and Keramasan, with total of 108 tributaries. In 2011, the population of Palembang was 1,481,814 people, with an average annual growth rate of 1.76% over the last decade. The population density is 3,698 people per Km2. The unemployment rate of Palembang in 2011 was 10%, and most of the population works in tertiary sector. The economy of Palembang is dominated by manufacture sector (43.8%), trade, hotel and restaurant sector (17%), followed by service sector (12.8%). GRDP per capita in current price for 2010 was IDR 32.6 million (USD 3,430), which is higher than the national GDP per capita (USD 2,850). The economy grew at 7.4% in 2010, and 10.8% in 2011. Oil refinery and fertilizer are the most prominent industries of the city. As with other oil related economies, Palembang is also susceptible to energy price fluctuation and was deeply affected by 2008 global economy crisis. Denpasar: Denpasar is the capital of Bali province, making it an important hub to other tourism sites in Bali island. The city comprises 4 kecamatan (sub-district), with total area of 127.98 Km2. The city had reclaimed land of 380 Ha, or 2.27% of its total area. Denpasar is bordered by Kabupaten Badung to the west and north, and Kabupaten Gianyar to the east, with Badung Strait to the south. The topography of Denpasar is mostly sloping to the south, between 0 – 75 meter above sea level. Denpasar has 10 Km of coastline, which is prone to abrasion. The city also makes an effort to maintain the 10 rivers that pass through the city through community participation in keeping the rivers clean. In 2010, the population of Denpasar was 788,589 people, with a population density of 6,171people per Km2. 31% of the population lives in Kecamatan Denpasar Selatan (South Denpasar), 29% lives in Denpasar Barat (West Denpasar), while North and East Denpasar house 22% and 15.5% of total population, respectively. The unemployment rate of Denpasar in 2011 was 6%, and 79.8% of the population works in tertiary sector. 4 The economy of Denpasar is dominated by trade, hotel and restaurant sector (37.4%), followed by finance sector (14%) and transport and communication (12.8%). GRDP per capita in current price for 2010 was IDR 15.85 million (USD 1,668), which is lower than national GDP per capita (USD 2,850). The economy grew rapidly at 16.2% in 2010, and has always been growing above 13% annually over the last 4 years. As the economy relies heavily on tourism, it is susceptible to global economic downturn and security issues. Table 1. Characteristics of Pilot Cities City Province Area Population Population GRDP per (Km2) (2010) Density capita (people/ Km2) (2010, USD) Surabaya East Java 327 2,765,908 8,462 6,766 Palembang South Sumatera 401 1,481,814 3,698 3,430 Denpasar Bali 128 788,589 6,171 1,668 Balikpapan East Kalimantan 503 557,579 1,108 4,721 Source: BPS, 2011 Balikpapan: Balikpapan is the second largest city in East Kalimantan province, which gains its economic importance as the oil refinery and base operation for multinational mining service companies. The city comprises of 5 kecamatan (sub districts), with a total area of 503.3 Km2. Balikpapan is bordered by Kabupaten Kutai to the north, with Makassar Strait to the south and east side, and Kabupaten Penajam Paser Utara to the west. 85% of Balikpapan’s area is hilly, while flat planes are mostly located along the coast. Due to its topography, the land is prone to erosion. To avoid landslide, the government of Balikpapan plan to limit development to only 48% of its area, leaving 52% as green space (Spatial Plan 2012-2032). In 2010, the population of Balikpapan was 557,579 people, with a population density of 1,108 people per Km2 and average annual population growth of 2.1% in the last five years. Most of Balikpapan’s population is in the productive age group (15-64 years old), where the workforce constituted of 46.5% of population. The economy of Balikpapan is dominated by manufacture sector (51%), followed by trade, hotel and restaurant (16%) and construction (15%). GRDP per capita in current price for 2010 was IDR 44,850,051 (USD 4,721), which was significantly higher than Indonesia’s GDP per capita (USD 2,850). As a refinery and mining services city, Balikpapan’s economy is susceptible to global energy prices. Economic growth has fluctuated heavily during the last five years, with growth at 12.4% in 2008, followed by 1.7% in 2009, due to global oil crisis. The economy has bounced back, with 5.19% growth rate in 2010 and 9.7% (preliminary figure) in 2011. 1.2 RATIONALE The City Planning Lab (CPL) is envisioned as the driver of improved integrated and evidence-based spatial, development and investment planning. Local governments in Indonesia understand the importance of improved data and technical analysis for strategic, evidence-based, integrated planning and decision-making. In the 5 attempt to address this need, technical assistance to cities usually takes the form of isolated studies which, while they may be helpful in the short term, often do not systematically increase cities’ technical capacity, or improve urban management on an ongoing basis. Instead, in order to make technical assistance under P3N more sustainable, it will be anchored in a dedicated facility in each partner city, called the City Planning Lab. The CPLs aim to establish technical capacity at the municipal level to provide reliable analytic support to a city’s planning, policy and infrastructure decisions, and to enable access to leading technical assistance in urban management, analytics and planning systems. The focus of the facilities will be to build up technical and institutional capacity in city planning and regulatory agencies to produce reliable and up-to-date data about the cities, well-informed plans, effective public investments, and to support the enforcement of development regulations. The facilities will operate by offering a menu of technical engagements for immediate as well as long-term projects on a demand-driven basis. CPLs will seek strong support and cooperation from the City Government with the aim of becoming technically and materially self-sustainable within two to three years. By acting as a single ‘nerve center’ or focal point for analytical work across a range of sectors, touching on spatial growth, land use, land markets, slums, economic competitiveness, and climate and risk resilience, the CPL will help to habituate city leaders to thinking about urban management in an integrated, holistic way, allowing them to meet a range of needs through select but strategic interventions. As described in detail under the ‘core’ module, the CPL will facilitate coordination through various agencies, with the Directorate General of Spatial Planning, Ministry of Public Works (MPW) at the center of the technical engagement at the national level, and Bappenas playing an important coordinating and advisory role, and donor support from the World Bank. At the local level, the CPL will have dedicated staff from various local government agencies, as well as external experts with long-term commitments to working with the Lab. It will also establish working relationships with academic and research institutions. It will conduct technical studies in modular form to respond to immediate needs, while also serving as the venue for the transfer of technical knowledge and the building of local capacity in the longer term. 1.3 OBJECTIVES In the short term, the CPL will (i) provide “just in time”, demand driven data and analysis that can feed into immediate decisions, and (ii) streamline ongoing urban management functions, such as building permitting and tax-related functions. In the medium term, it will provide cost-effective analytics to cities that can feed into planning and investment decisions, reducing the expense involved in contracting consultants during each planning cycle. In the long term, the CPL will build local technical capacity, by gathering expertise from Indonesia and international sources to work closely with local staff. Over time, external involvement will diminish as local capacity strengthens. 6 MPW will facilitate an ongoing objective alongside those mentioned above will be to work with MPW to demonstrate and disseminate the value of this approach more broadly to local governments throughout the country. 1.4 SCOPE OF ACTIVITIES The proposed activities of the CPL will be conducted in modular fashion, each pertaining to a different sector. The proposed sector modules are: F. Instituting the City Planning Lab & Spatial Growth Analytics (core module) G. City Economic Competitiveness H. Slum Analytics and Management Systems I. Climate and Risk Resilience Planning Systems J. Monitoring Land and Real Estate Markets While the details of the activities will differ, they will all take a common approach, which will involve (i) data gathering; (ii) inputting new and existing data into an integrated cross-sectoral data platform; (iii) using data in ongoing urban management functions; (iv) analyzing the data; and (v) working with city leaders to help them use the insights from data analysis in planning and decision-making. In addition, the Ministry of Public Works will lead an overarching component involving three key activities: (i) preparation of guidelines for establishing CPL, (ii) a capacity building program beyond the core cities, and (iii) dissemination activities. The detailed outputs, budget, timeline and potential risks for each module are discussed separately. A summary is provided below: A. City Planning Labs & Spatial Growth Analytics (core module) Four cities (Surabaya, Palembang, Denpasar, and Balikpapan) Major outputs:  Geospatial database  Support to detail planning process  Pilot of a new permitting decision support platform.  Report on spatial accessibility of urban services  Report on urban expansion trends, 2000-2010  Report on land value impacts of infrastructure  Report on infrastructure demands over 10 years B. City Economic Competitiveness Two cities Major outputs:  City economic competitiveness review  City economic planning and decision support capacity building 7  2-4 Workshops for public-private dialogues  City economic competitiveness dashboard C. Slum Analytics and Management Systems Two cities Major outputs:  Slum Information Database, incorporating all collected data  Survey materials  Report outlining slum management strategies  Planning of pilot implementing programs for selected sites  Report outlining (a) the process of slum formation, as observed through case studies; and (b) recommendations for strategies for preventing slum growth in specified areas D. Climate and Risk Resilience Planning Systems Two cities Major outputs:  Data inputs on disaster risk into city’s geospatial database  Customization of the InaSAFE software tool based on user needs  Report outlining the drivers of disaster and climate risk to core sectors and areas/neighborhoods, with risk-sensitive micro zoning maps, and recommendations for resilient land use and infrastructure investment planning E. Monitoring Land and Real Estate Markets Two Cities Major outputs:  Cadastral real-estate database, showing each land parcel with its associated buildings, occupants’ demographics, accessibility characteristics and valuation estimates.  Land and property market assessment report  Housing segmentation study report  Impact analysis report, documenting the observed real estate value impacts of selected infrastructure investment projects  Real Estate Financing Analysis  Hedonic pricing analysis, explaining variations in land and real estate values based on the spatial attributes and accessibility 8 SECTOR MODULE A: CITY PLANNING LABS AND SPATIAL GROWTH ANALYTICS 9 10 2.1 BACKGROUND 2.1.1 CONTEXT As Indonesia urbanizes, the forms of its metropolitan areas will have profound and long- lasting socio-economic and environmental consequences. Present urban expansion can, on the one hand, foster economic growth, offer better opportunities to citizens and improve regional and international connectivity. On the other hand, rapid urban expansion also brings about important challenges, such poor integration of complementary land uses, exhaustion of urban resources and social inequality. In order to overcome such challenges and harness the opportunities, Indonesian cities need a capacity to analyze the current growth trends, understand their underlying forces and forecast their future consequences. At present, a number of medium and large-scale cities in Indonesia, where a large share of urban growth is occurring, lack the analytic capacity to examine how and much they are growing, what factors drive the growth and change, where and what types of public infrastructure investments are needed, how well past policies and investments have performed, and how future plans can be informed by current development trends. This leads to uncoordinated planning and enforcement efforts, inefficient use of scarce resources, and poor returns on infrastructure investments. The lack of basic urban information systems impedes the necessary information sharing across different city departments, making decision coordination and planning enforcement difficult to achieve. Without reliable information and analytics, scarce public resources cannot be effectively allocated and policies cannot be effectively designed nor enforced to address key urbanization issues. In order to support efficient, sustainable and equitable urban growth in the next decade, it is critical for Indonesia’s cities to invest into new information, analytic and regulatory systems of urban planning and development. 2.1.2 IMPLEMENTING P3N TECHNICAL ASSISTANCE THROUGH CITY PLANNING LABS As part of the Metropolitan and Urban Development Program (P3N), the World Bank aims to support the Government of Indonesia in establishing City Planning Labs (CPLs) in medium and large-scale cities, starting with four pilot cities in 2013 – Denpasar, Palembang, Surabaya and Balikpapan1. The CPLs will house a number of planning support activities for a wide range of urban problems that are divided into several modules. The central focus of the labs is to provide reliable Urban Spatial Growth Analytics and to upgrade information management for Regulatory Enforcement Systems. The focus of the Spatial Growth Analytics Module will be on analysis that provides a clear practical benefit to cities, which can serve as inputs into decision-making around policies and investments, and which can eventually be carried on by the cities independently. Provisioning the right amount of land and utility systems for future housing needs, for instance, can reduce the development of slums and save costly 1 Denpasar (metro population 1.8 million), Palembang (metro population 1.6 million), Balikpapan (population 0.6 million) and Surabaya (metro population 5.6 million) have been selected due to their their existing planning efforts, their fair results in coordinating planning efforts with the central government, and their keen interest in the initiative. 11 legal land readjustments later. Positioning key infrastructure, such as new roads, in places that generate the greatest and the most equitable benefits to landowners, can lead to a rise in land values and rental incomes that greatly exceed the initial investment. Paralleling analytic work, the technical facility will also assist city planning enforcement agencies to transition into a transparent, electronic permitting and enforcement workflows. A great deal of planning and development regulation today is paper based and fragmented between different approval processes, making it difficult to have a holistic overview of developments that are being approved. A number of spatial planning agencies have voiced that the present enforcement system also fails to integrate critical information between enforcement and planning groups in local governments, hampering their capacity to carry out approved plans. These two activities are described as part of the core CPL concept note below. Additional CPL activity modules are described in separate concept notes as follows: A) Land and Real Estate Market Monitoring Module; B) City Economic Competitiveness Analytics Module; C) Slum Analytics and Management Systems Module; D) Climate and Risk Resilience Planning Systems Module. Human resources, technical infrastructure and data management systems will be hosted by a single CPL facility in each city and shared by the activities of all analytic modules. Section three of this note describes three related steps of the proposed CPL implementation process: i. Developing City Planning Labs as institutionalized municipal platforms for spatial analysis, integrated and evidence-based spatial development and investment planning. ii. Implementing core urban spatial growth analytics (as well as other analytic modules described in separate concept notes) and using the outputs in planning activities. iii. Establishing an effective data exchange system between spatial planning and enforcement agencies for an improved and automated planning enforcement framework for core urban land use and construction permitting functions. 2.2 OBJECTIVES The primary objective of setting up the support facilities at municipal governments is to establish technical capacity to measure, analyze and respond to urban development pressures in an evidence-based and timely manner. By supporting evidence-based decision making, capacity building in urban analytics and more seamless information sharing across city departments, we expect the CPLs to lead to substantial cost savings in spatial management and enforcement, plans that are aligned with the city’s aspirations, more effective enforcement of planning goals, as well as greater multiplier effects on infrastructure investments in the medium and long run. The initial core activity of the facility is urban growth analysis, the objectives of which include to: 12  Project future growth, based on existing trends, and forecast the future demand for land uses and amenities.  Help integrate projected demographic and economic changes into Masterplans and Detail Plans.  Communicate information relating to future spatial plans over web-based maps to other related agencies to strengthen regulatory enforcement.  Foresee infrastructure requirements from current trends and help avoid supply shortages by proposing possible planning responses.  Conduct spatial cost-benefit analyses of public investment decisions.  Evaluate the social and environmental impacts of public investments.  Assess the equality of public investment distribution across all demographic and income groups.  Provide accurate and reliable geospatial data to private sector developers and individual stakeholders. The upgrading of regulatory enforcement systems module of the facility aims to improve information sharing and information capture between planning and regulating arms of the local government in order to develop a more effective and transparent decision chain for carrying out the city’s planning intentions. The objectives of the proposed regulatory technical assistance are to:  Analyze the present paper-based regulatory processes for building permits and change-of-use permits in local spatial planning offices.  Develop a comprehensive action plan to upgrade the present permitting system to computerized databases that allow permitting officers to instantly access approved planning information about parcels under question via a simple web interface.  Implement a pilot data capture system for building permits and change-of-use permits that will record each approved permit in a database and automatically update the city’s GIS parcel and building map layers with accurate information.  Display Masterplan and Detailed plan information to landowners publicly over a web-based map server, without requiring personal consultations to find out the allowable buildable volumes on site.  Evaluate the effectiveness of the above pilot schemes with respect to more effective regulation and adjust the system implementation accordingly. The CPLs will additionally offer the World Bank and other donor organizations a valuable platform for predicting and tracking the impacts of transformative infrastructure investments in Indonesian cities. 2.3 SCOPE OF ACTIVITIES Addressing the goals and challenges discussed above, the three steps to implement the CPL activities outlined in this note are: i. Establishing City Planning Labs: This involves providing assistance to the city on institutional setup, data collection, software and hardware and human capacity. 13 ii. Implementing Spatial Growth and Change Analytics: This involves developing the preliminary analytical work on spatial growth monitoring, and proposing activities for future phases, to be conducted by the City Planning Lab with external assistance. iii. Improving Planning Enforcement Systems: This involves assisting the cities’ spatial regulatory agencies to implement computerized information and permitting systems that are synchronized with spatial information with other city agencies. 2.3.1 ESTABLISHING THE CITY PLANNING LAB 2.3.1.1 Software and Data Platform To fulfill their primary goal of assembling, maintaining and distributing large geospatial databases, the City Planning Labs need a data platform that satisfies four fundamental requirements. The platform should:  Allow the data to be stored and management in a well-organized way  Allow the data to be shared across different departments or with members of the public over internet browsers  Enable all data management operations to be performed from a local networked computer  Enable the end-users to interact with the datasets, by querying their attributes, overlaying different data layers, using simple base-maps to situate the information, and sharing personal information layers on published maps. The capacity to operate basic spatial functions (e.g. spatial search, measurement or proximity search, attribute table joining etc.), would be desirable additional functions for the end users, though not a first-order priority. Combined, these basic requirements necessitate setting up a GIS map server platform. There is a considerable list of open source and proprietary GIS server technologies. Proprietary technologies include ArcGIS Server, ArcGIS Online and MapInfo Spatial Server, while open source options include GeoServer, GeoNode, and PostGIS. The World Bank’s Platform for Urban Management and Analysis (PUMA), currently under development, is also a potential open source option for the City Planning Labs. Based on the vital and desired functionalities, cost and budget limits, and the platform’s flexibility for scaling up, a few options will be introduced to the Lab. While setting up the data platform should be tackled at the outset of the lab, its maintenance and potential expansion – given the envisioned collaboration with a larger number of government departments – will continue throughout later phases. It is possible, for instance, to start off with a proprietary off-the-shelf system that requires little setup time, such as ArcGIS Online, while the staff are technically trained to set up a more long-term open-source system. Apart from the platform for geographic data, general software (e.g. text editors) and operating system, the lab requires two types of desktop software tools for assembling data and conducting analysis: 14  Spreadsheet software with basic statistical analysis capabilities (e.g. Microsoft Office Excel, Access; Open Office Calc, Base)  GIS desktop software (e.g. ArcGIS, MapInfo, QGIS) These desktop tools are available both as proprietary and open source, with different functionality. The potential options will be introduced to the lab, based on the required capacities. Figure 1. Screen Capture of a QGIS Open-source Data Platform Work Environment 2.3.1.2 Institutional Arrangements Organizational location: A few different options are available in terms of situating the City Planning Lab within the existing local government. The exact institutional setup would be tailored to the preferences of the local governments. An effective institutional model would be to have the Lab located within Bappeda, who would provide the physical space and some of the basic investments in setting up the Lab. It is recommended that both 15 Bappeda, as well as the Department of Spatial Planning, would provide two full time staff members to work as part of the Lab team. In order to ensure coordination across agencies, it is recommended that the Lab be advised by an Advisory Committee convened by the Mayor, with members from Bappeda, Spatial Planning, Public Works, Revenue, BPS, BPN and other planning related agencies or city departments. The committee may also include representatives from neighboring jurisdictions or regional governments, in order to ensure coordination across the whole metropolitan area. In addition to coordinating between city agencies and departments, the advisory committee will liaise between CPL and the Ministry of Public Works in order to inform the national level spatial planning by local analysis, data and plans. CPL in each pilot city will also assist the local governments by informing their planning enforcement systems of national plans. This committee would likely meet once every month or two in order to set the strategic direction for the work of the Lab. It is not recommended that the advisory committee intervene with the daily operations of the lab, which could be done more efficiently by the CPL staff. Figure 2. City Planning Lab Partnership Framework Partnerships: The Lab would establish institutional partnerships with external entities in order to facilitate knowledge exchange. For example, there may be MoUs signed with Indonesian universities to foster collaborative projects between students and the Lab, internships or part-time positions for students who may work at the Lab for short periods, or research projects conducted by universities that complement Lab activities. MoUs may also be signed with agencies at other levels of government, including data sharing agreements with BPS or BPN. In addition, consultants would be hired to work closely on specific analytical areas on a project basis. During the first two years of the implementation phase, outside consultants and partner organizations will be required to collaborate closely and transfer knowledge and skills to the CPLs. The World Bank team would play an ongoing advisory role, which would phase out over time. In this way the Lab would gradually become technically proficient and self-sufficient to support all 16 necessary spatial analytic support function for the cities, and form the means by which the city interacts with key partners in urban planning and management. City Figure 3. Planning City Planning Lab Lab Staffing Director Technical Staff Civil Servants Admin. Assistant (GIS, databases, web) (Bappeda, Spatial Planning) Staffing: In the first phase, the Lab is recommended to have 6-7 full time staff. This would include the following: i. Director: responsible for managing the daily activities of the Lab. Ideally the Director would be an individual with a Master’s or higher degree in urban planning or a related field, and approximately ten years of experience in urban planning in Indonesia, who is familiar both with the kinds of analytical tools and approaches that the Lab will use, as well as with the functioning of local governments in Indonesia, and has experience in starting up new institutions or ventures. This individual would most likely be hired from outside the government. ii. Representative of Bappeda: responsible for coordinating with Bappeda’s spatial planning activities. This would be someone with at least 3 years of experience in the local government, who is familiar with the operating procedures of Bappeda. He or she would be assigned to work with the Lab full time. iii. Representative of Department of Spatial Planning: responsible for coordinating with Department of Spatial Planning activities. This would be someone with at least 3 years of experience in the local government, who is familiar with the operating procedures of the department. He or she would be assigned to work with the Lab full time. iv. 2-3 technical staff: responsible for data gathering, managing databases, and using software tools to perform the analysis. These individuals would need to be highly proficient in ArcGIS and AutoCAD. They should have some background in urban planning, policy, geography, architecture or other relevant field. At least one of these should have experience in setting up and managing data servers. These individuals would most likely be hired from outside the government. v. An administrative assistant. These individuals would be involved in the functioning of the Lab full time from its establishment onwards. In subsequent phases, more staff may be added as necessary. 17 Equipment and Space: In its first phase, it is recommended that the Lab be situated in a space of approximately 40 sq. m., with a desk and a computer station for each full-time staff as well as one additional work station for visiting consultants, and a small meeting area. The equipment necessary would include a computer for each work station, a laser color printer / scanner, a 36-in color plotter, a large-format scanner, and a 46-in flat- screen display for presentations. Figure 4. Example Analysis Output: Accessibility to Jobs Source: City Form Lab Note: Accessibility to jobs within a 10 minutes walking range from each building in Cambridge and Sommervile, MA, USA 2.3.2 SPATIAL GROWTH AND CHANGE ANALYTICS 2.3.2.1 Analytics A core objective of the SP Module is to provide spatial analyses and evidence-based decision support to different city agencies and outside constituencies. The CPL will play an important role here. The spatial information gathered and analyzed by the CPL should enable the city to keep track of the growth and changes in its overall development, to monitor its land and real-estate markets, and to forecast and monitor the impacts of its planning interventions. The analytics performed by the CPL will be used as a basis for the city’s Masterplanning and detailed planning efforts, for setting the priorities and predicting the impacts of public financing and infrastructure investments, and for making reliable spatial information available to various planning and enforcement decision 18 makers (i.e. building permitting office) on a continuous basis. The Bank’s staff and outside technical experts will work closely with the local CPL teams over the first two years to transfer technical knowledge and to build up the skills needed to perform the information management and analytics autonomously. For the first year of operations, the CPL aims to achieve the following analytic outcomes: Phase 1 (Months 1 to 6): - Creating interactive geospatial databases: Existing spatial datasets are uploaded to an online map server for interactive viewing by different city departments. The interactive viewing should be web browser-based, not require any additional software from end users. This will allow stakeholders to overlay different spatial data (e.g. current built- out areas and the existing Masterplan) and to query simple attributes about the map elements by clicking on them (e.g. click on a parcel to see its area, ID, etc.). The exact list of existing datasets to be uploaded will be decided together with the city planning agencies based on availability (e.g. high-resolution satellite image, street centerlines, building footprints, parcels, schools, hospitals etc.). - Urban growth analysis: The growth of the metropolitan area and its corresponding population from year 2000 to 2010 will be obtained (from the World Bank’s ongoing East Asia and Pacific Urban Flagship activity) and used to analyze the spatial extent and rate of the city’s growth in the past decade. The previous decade’s expansion areas will be overlaid with current building and land-use data in order to analyze how much land was consumed by different land-use categories. This analysis, combined with regional economic and demographic forecasts, will subsequently be used as a reference to develop likely estimates for growth in the current decade, from 2010 to 2020. Phase 2 (Months 7 to 12): - Accessibility analysis: Existing spatial information on public facilities and resources (e.g. drinking water sources; drainage points; schools; hospitals; markets; transit stops) will be used to estimate accessibility to these resources in different parts of the city. This analysis should illustrate underserved areas and provide an empirical basis for future public investments. - Support to planning: As the planning agencies (Bappeda) of the participating cities engage in developing detailed plans (1:5,000 scale) from their current Masterplans (1:25,000 scale), the CPL will help develop the supporting spatial analysis required to achieve the goals of detailed plans. Palembang planners indicated that they need to develop 16 detailed plans for the different parts of the city, indicating the allowable land-uses, building heights, building coverage, infrastructure changes and buildable areas in different parts of the city. CPL analyses will help choose the areas in need for public investments (i.e. new roads, transit stops, schools, flood protection, etc.); for determining the likely economic growth poles in the city; and for forecasting the needs for different land-uses at the detailed plan scale during the next five years. The planning agency (Bappeda) can integrate these inputs to detailed plans and associated legal development regulations. 19 - Impact analysis: CPL will additionally develop impact analyses for ongoing public investment projects, such as choosing the exact location for the second bridge in Palembang, for locating sanitation and water facilities in Denpasar, collaborating with Public Works in choosing the placement of a new toll road, etc. on a per need basis. Phase 3 (Months 13 to 18): - Projections: More accurate and up-to-date spatial data will allow the CPL to start developing more accurate forecasts for near-term and long-term projections on land use requirements, housing needs, transportation demand, infrastructure needs etc. Analytics outlining such needs will help the cities prepare for potential problems (i.e. housing shortages, congestion) before they occur in the future. CPL staff will a long term (20 year) forecast for the city’s growth and start analyzing planning and policy responses needed to accommodate the projected growth. 2.3.2.2 Data Spatial and development plans cannot achieve their envisioned goals without accurate projections of supply and demand for housing, infrastructure and services, and forecasts for broader socio-economic and environmental situations to which planners must respond. Private sector developers and individuals can also make better decisions and contribute to the progress of the city if they have access to accurate data on how the city is growing and changing, and potential risks and bottlenecks. One of the primary objectives of the City Planning Lab in the four pilot cities is piecing together a comprehensive geospatial database from both the existing data and new data sources. A large body of data currently exists in local and national agencies; however, the absence of a well-structured collaborative information system has obstructed the flow of appropriate information among the government departments and the public. A considerable amount of data has not yet been digitized, prohibiting the data from being shared or used for computer- based analysis. The City Planning Lab aims to fill this gap by assembling existing data through a close collaboration with municipal agencies, and initiating collection mechanisms for new datasets. The Lab will develop an online platform to which government departments can contribute data they collect or record. The contributors, in return, will have access to more comprehensive and linked datasets, benefiting their own operations. Apart from continuous updating of the database, another important task for database maintenance is the verification of the accuracy of data. Accuracy verification will be a continuous task for CPL staff. The first phase of data collection will involve identifying existing datasets, obtaining data from multiple departments, and integrating the data to standardized formats. This process will involve a significant digitization effort – e.g. generating GIS maps with useful attribute tables from the current paper maps showing allowable building regulations. Some early data collection activities will require external support. In the second phase, the Lab will start to build databases by joining different datasets together – e.g. adding land 20 values, land uses and establishment locations to the building dataset. The third phase will be mostly dedicated to field surveys for filling in the missing data and collecting new data. The use of government accounting and registries (e.g. data recorded for permitting or land and real estate transaction taxation) constitute an important future source of reliable and up-to-date data. Piloting the collection of such data is discussed further below. 2.3.2.3 Communicating Planning Goals Beyond performing analytic work to support ongoing planning, development and regulation work in the city, the CPL also aims to gather, document and visualize the planning goals that form the basis of Masterplans, Detail Plans and other spatial development initiatives. CPL’s analytic work is impactful only if it is well aligned with the city’s goals and initiatives. Yet such goals are often unclear and dispersed among multiple agencies. CPL could provide a venue that collects and visualizes the different initiatives and goals graphically in order to help disseminate the ideas across departments and to the general public. This can be done through web-maps, info-graphics and printed publications that are shared across the city’s departments. 2.3.3 PLANNING ENFORCEMENT 2.3.3.1 Restructuring the Planning Enforcement Procedures: Any government accounting and registry procedure naturally leaves a trace of data behind, which could be effectively used if a proper structure for the flow of data is developed. The structure of the existing planning enforcement systems in Indonesian cities, however, does not allow for the effective and efficient utilization of these registry and accounting records. Planning enforcement procedures are still paper-based and the lack of a standardized national addressing system makes it difficult to integrate them with other spatial databases. This concept note proposes restructuring three planning enforcement procedures – building permitting, change-of-use permitting and the communication of zoning regulations – as a pilot initiative in the four cities. CPLs will develop a detailed assessment of the current enforcement mechanisms at the local spatial planning agencies and propose comprehensive improvements to digitize and streamline development-permitting processes. CPLs will carry out a pilot implementation of a data capture system in building permitting and change-of-use application procedures that will demonstrate an integrated information flow for keeping a city’s geospatial building and land-use data up to date. Building and Change-of-Use Permitting (Phases 1 and 2): Building and change-of-use permits are potentially the best source of data for keeping a city’s spatial database up to date, as such permits capture changes in all legal development activities. In order to actively harness these data, the permit issuance procedures in Indonesian cities need to be restructured. 21 During the first two phases, the CPL will perform a detailed assessment of the present permitting procedures in the Department of Spatial Planning and develop a plan for updating the processes to digital standards. The new procedures will allow each permit to be recorded in a digital database, which can be referenced via parcel and address indicators and geographic coordinates to other existing databases (e.g. parcel, building and business location databases). This is expected to produce two important benefits. First, linking permitting with existing geospatial data will allow permitting officers to instantaneously retrieve approved planning information about the permit sites under question, eliminating an information gap between planning goals and enforcement. Second, a continuous updating of building and use data based on permitting procedures will also significantly lower the on-ground or aerial surveys required in the future for data updating. Phase 1 (Months 1 to 6) deliverables: - Assessment of new building and change-of-use permitting. CPL will document and evaluate the current procedures for issuing new building permits and change-of-use permits at the local spatial planning agencies, producing a report of the current workflows and potential opportunities for improvement. The report will also outline the success rate of the current planning enforcement system, overlaying legal spatial plans with issued permits on the ground. Phase 2 (Months 7 to 12) deliverables: - Recommendations and activity plan for a new permitting decision support platform. CPL will produce a report outlining recommendations for a new, digital permitting decision support platform that will allow permitting and enforcement agents to seamlessly access cross-linked information about planning regulations for parcels, buildings, zones in the city. The report will also outline a proposal for making general planning and zoning regulations accessible to land-owners and developers via an online portal. Zoning Regulations and Spatial Plans (Phase 3): Zoning regulations and spatial plans are currently not fully shared with the public, which has imposed an unnecessary work load on the Department of Spatial Planning, who communicates this information on a case-by-case basis to interested property owners. Prior to applying for any building permit, property owners are required to submit an inquiry about allowable coverage, height, use, and setbacks for each property. A planning officer retrieves this information from paper-based documents and communicates back to the requestor in written form. Such zoning information, which is publicly available in most developed countries, can also be made publicly available in Indonesia. Integrating Registry and Accounting Records into the cities Spatial Database (Phase 3) As discussed above, one of the primary objectives of The City Planning Labs is to piece together and maintain a comprehensive geospatial database. In addition to readily available data, additional spatial information harnessed from the government accounting and registry documents offer important potential sources for expanding the datasets and 22 keeping them up-to-date. Building permits, for instance, can be used for updating the building dataset in real time and in the most accurate and cost-efficient way. This requires proper digital and spatially referenced registries that could be linked with the CPL databases. Transaction data from local tax services or notary offices could allow land and real estate value datasets to be updated each time a transaction is made. Such mechanisms are common in developed countries’ planning systems. The City Planning Labs will not only be collectors of data, but they will also provide participating government departments (local and national) with integrated and updated geospatial databases, built upon the data provided by individual agencies themselves. Bappeda, for instance, will benefit significantly from registry and permit data from the Department of Spatial Planning (Dinas Tata Kota), which can be used for preparing the detail plan of sub-districts. As the permit information is currently not transferred to Bappeda in a ready-to-use manner (it is not digital nor spatially referenced), Bappeda instead uses open-source satellite images to update its building datasets, leading to outdated and inaccurate information. Phase 3 (Months 12 to 18) deliverables: - Planning enforcement portal. By the end the third phase the CPL, in collaboration with Bappeda and Department of Spatial Planning, will prepare and publish currently available maps of zoning regulations and spatial plans to the general public on designated websites. Since the information is legal and explicitly stated, this upgrade is expected to relieve an unnecessary burden of private consultancies. - Pilot program for permitting decision support platform. CPL will implement a pilot program for permitting decision support that will test an integrated digital workflow for permitting officers. The platform should allow permitting officers to instantaneously retrieve approved planning information about the permit sites under question, providing a more integrated planning and reinforcement workflow. Each issued permit should automatically update building and land-use data in the city’s building and parcel databases. This pilot program will be implemented on two permitting procedures in each city: new building permits and change-of-use permits. The pilot program seeks to understand the existing data flows, and the required procedures for integrating and maintaining a real-time database between different city departments. After evaluating the first phase pilots, CPL aims to scale such efforts up in the second phase. 2.3.3.2 Digitizing Historic Data While keeping track of the registry and accounting records offers an up-to-date capture of the existing condition of a city, it is not sufficient for understanding the current trends and forecasting their future changes. This requires several datasets of spatial conditions over time. These snapshots can be collected gradually over time. The existing planning enforcement systems have been collecting valuable data, although many of them are not digital or in an appropriate format for analytical purposes. During the first and second phases of the project, the City Planning Lab will collaborate with municipal government departments and 23 agencies to first map available historic data, and to then digitize and integrate these historic data with current conditions. 2.3.3.3 Planning Tools The availability of close partnerships with outside institutions (e.g. CPLs in other cities, the World Bank, outside consultants) also offers a unique opportunity to collect and document information about planning tools and implementation mechanisms in other successful cases. Such tools may include zoning regulations, incentive systems, building guidelines etc., which could potentially be implemented in the city as part of planning initiatives. CPL can help disseminate knowledge about such tools to different stakeholders in online and print publications. Keeping an up-to-date overview of planning goals and their related implementation tools will help CPL ensure that the analytics performed are aligned with the city’s needs. 2.4 RISKS Potential risks include the following: i. Difficulty in transferring skills in a sustainable manner: Local governments in Indonesia often lack the technical expertise necessary to perform the kinds of analytical work proposed for the Lab. For this reason, much of the work in the early phases will be done by external consultants. There is a risk that knowledge will not be sufficiently transferred to the local government counterparts involved in the Lab. In order to address this risk, the Lab will involve local officials as key team members from the beginning, and will be overseen by the mayor or a local government agency. Any external consultants will be required to work closely with the local officials in the Lab. Every technical assistance activity will have the dual objective of producing the analytical output itself while simultaneously training local staff to perform such analysis. This will ensure sustainability of skills in the Lab. ii. Lack of coordination with other agencies: There is a risk that while the local staff directly involved in the Lab will adopt new analytical approaches, the overall urban planning and management systems will carry on with business as usual. This risk will be most effectively mitigated if there is a high-level champion for the Lab, ideally the mayor, the head of Bappeda, or a board consisting of heads of various departments (see section on institutional arrangements), to ensure the proliferation of analytical approaches and operating procedures developed in Lab throughout the rest of the government. 2.5 OUTPUTS PHASE 1 At the end of Phase 1, the Lab should have a physical space, with hardware and software equipment with full-time staff set up. The outputs of the phase 1 analytical activities will be as follows: 24 - An interactive online geospatial database, featuring datasets that are already available for the lab, will be ready for use for different city departments on web- browsers. - A short report will outline the 2000-2010 urban expansion increase in the given city and the likely growth scenario for the current decade based on urban extent and population data from World Bank’s ongoing East Asia and Pacific Urban Flagship activity. - A report assessing the current procedures for issuing new building permits and change- of-use permits at the local spatial planning agencies, describing the current workflows and potential opportunities for improvement. PHASE 2 - A spatial accessibility databases will be analyzed at the individual building level, illustrating how easily households in different parts of the city can access critical urban resources – drinking water sources; drainage points; schools; hospitals; markets; transit stops. The results will be described in a short report and in graphic material (e.g. paper-based and online maps) that can be shared with various city departments. - Based on collaboration with the detailed planning team in the respective city, CPL staff will support the development of the detailed plans with spatial analyses. CPL analyses can help choose the areas in need for public investments (i.e. new roads, transit stops, schools, flood protection, etc.); for determining the likely economic growth poles in the city; and for forecasting the needs for different land-uses at the detailed plan scale during the next five years. These analyses will be determined on a per-need basis and documented in written and online reports with supporting geospatial evidence. - CPL will produce a report and supporting geospatial data, outlining the likely land- value impacts of ongoing public investment projects, such as the addition of a second bridge in Palembang, for locating sanitation and water facilities in Denpasar, choosing the placement of a new toll road, etc. - Recommendations and activity plan for a new permitting decision support platform. CPL will produce a report outlining recommendations for a new, digital permitting decision support platform that will allow permitting and enforcement agents to seamlessly access cross-linked information about planning regulations for parcels, buildings, zones in the city. PHASE 3 - CPL will produce a report, which analyzes the directions and magnitudes of the effects that different land improvement strategies have on land and real-estate values in the respective cities. The report, based on hedonic price models, will indicate how access to critical infrastructure (roads, water, transit) and land-use linkages (commerce, jobs, parks) affect land prices and real estate sales. 25 - Forecasts will be prepared in the form of a written report and supporting graphic material to describe land use requirements, housing needs, transportation demand and infrastructure needs for the next 10 years in the city. - CPL will compose a report and hold a workshop with various city planning related departments to describe the results of the digital data integration and capture pilot program through planning enforcement mechanisms. The report will outline the successes and shortcoming of the pilot program and make concrete recommendations for the expansions and automation of the data capture system in the future. - CPL, in collaboration with Bappeda and Department of Spatial Planning, will prepare and publish currently available maps of zoning regulations and spatial plans to the general public on designated websites. - CPL will implement a pilot program for permitting decision support that will test an integrated digital workflow for permitting officers. The platform should allow permitting officers to instantaneously retrieve approved planning information about the permit sites under question, providing a more integrated planning and reinforcement workflow. Each issued permit should automatically update building and land-use data in the city’s building and parcel databases. This pilot program will be implemented on two permitting procedures in each city: new building permits and change-of-use permits. 2.6 TEAM AND TIMELINE This module will be carried out in three phases of six months each. In addition to full time staff members listed above, additional expertise required for providing consultation to the spatial growth analytics module will include: i. IT/GIS Server Specialist ii. Urban Economist iii. Urban and Regional Planner 26 SECTOR MODULE B: CITY ECONOMIC COMPETITIVENESS 27 28 3.1 BACKGROUND Indonesia’s cities remain a challenge – and a major opportunity. The Indonesian economy has performed strongly over the past decade. The country has also been rapidly urbanizing, but has been unable to fully capture the productivity benefits from agglomeration. Improved capabilities at the municipal/urban level are the key to unlocking the potential for improved economic competitiveness of Indonesian cities, since i. The size and diversity of Indonesia calls for customized strategies for sustained economic growth. This is true particularly in the current political environment where risk aversion among decision makers at the national level ahead of the 2014 elections means that reforms at the city-level have a greater chance for success. ii. Cities can influence the business environment considerably, thanks to decentralization and partial devolution of regulatory authority. According to the Subnational DB study (2012), obtaining a construction permit in the city of Bandung, for example, takes on average 44 days, while in the capital Jakarta, less than 150 Km away, the same procedure takes on average 158 days, more than 3 times as long. To start a business in the city of Palangkaraya, 27 days are needed for the official procedures, while the same steps in Jakarta take almost twice as long, 45 days. These variations indicate that cities have the ability to improve the regulatory environment independent of reforms (or lack thereof) at the national level. iii. City land use and infrastructure planning can directly and significantly impact the cost of doing business especially rental cost, ease and cost of brownfield expansion of businesses and access to last mile infrastructure (road, power, broadband, etc) Examples of notable success (and failure) of international cities can provide guidance as to what can be done to enhance city competitiveness and make best use of comparative advantages. Locally appropriate policies are needed to provide the simple, transparent, and supportive operating environment that businesses need to succeed and grow. 3.2 OBJECTIVES The objective of the City Economic Competitiveness Module of the Metropolitan and Urban Development Program (P3N) is to enable client cities to actively guide and foster their municipal and regional economic development through superior planning and decision making and deep consultation with private sector stakeholders. Success indicators include the number and quality of jobs available in the city (and surrounding areas, as applicable), the improvement of productivity in targeted sectors, and the inclusiveness of economic growth (target measure TBD) within the city. The module’s components will be achieved by helping cities (i) get smarter about understanding trends in their regional economy, and the impacts that public policy, investment and planning decisions have on their economic prospects; and (ii) work with the private sector to leverage this newfound understanding to initiate a series of local reforms/policies/investments. Activities will focus on building institutional capacity to 29 significantly strengthen the targeted cities’ competitiveness planning and analysis capability, as part of the City Planning Labs (CPL), as well as on market sensing and working closely with the private sector to ensure that the cities’ efforts on planning and analysis are well targeted. Since the achievement of this component’s objectives ultimately depends on the private sector’s response to the cities’ improved planning, the involvement of and dialogue with the private sector from the start is critical. 3.3 SCOPE OF ACTIVITIES In keeping with the other CPL modules, the activities in this module will be conducted in three phases of six months each. The module comprises four components, which will span all three phases, though at varying intensity. The module is designed on the basis on World Bank’s past experience with such programs, latest literature on the subject, and our own initial assessment of the cities. An overview of the key products that will be produced under the four components is in the table below. i. City Economic Competitiveness Review: city-level economic review and comparison across cities, sectoral studies and in-depth analysis. ii. Capacity building: building city economic planning and decision support capacity at a local level iii. Consultative workshops and Public-Private Dialogue (PPD): 2-4 workshops for public- private dialogues; from early consultations to institutionalized, regular dialogue and fine-tuning of policy initiatives. iv. City initiatives and dashboard: from considering external case studies, to finalizing decisions on policy initiatives and implementation, including joint action plan with private sector and dashboard to communicate initiatives and monitor progress. Table 2. Scope of Activities and Timeline Phase 1: Phase 2: Phase 3: 0-6 months 7-12 months 12-18 months City economic City Economic Competitiveness Sectoral studies, In-depth analysis competitiveness Review of pilot cities including possibly based on review comparative analysis additional data Capacity Integrating city economic data Building analytical Transferring lead of building with the core spatial planning capacity at CPL; map activities to CPL platform; map economic analysis economic analysis to to core spatial data platform decisions at city level Consultative Early consultations with private Decision on key Institutionalizing workshops / sector; formulating top sectors and policy dialogue; decision on PPD hypotheses on constraints and options action plan for policy policy levers initiative City initiatives External case studies Policy options Policy finalization and and dashboard decision, incl. action plan 30 Component 1: City Economic Competitiveness Review Local government intervention to boost competitiveness should start with a clear understanding of the market and the main spatial and sectoral drivers of city economic growth. This report aims to do an in-depth study of the city’s economy. Description: The City Economic Competitiveness Review attempts to answer the following questions – (a) what is the state of the local city economy in terms of GRDP, GRDP per capita (as a proxy for productivity), GRDP mix, total jobs, average wages, and exports and their growth over last 10 - 20 years; (b) how does the economic performance of the city compare to peers and what are city’s strengths and weaknesses including education and skill level of workforce, land pricing, regulatory environment; (c) what are the spatial and sectoral drivers of city’s economic performance and competitiveness; (d) which are the sectors where the city can be considered or has potential to be nationally and globally competitive; (e) for the top sectors, what are the market failures and major barriers to growth across regulation/policy, skills, infrastructure (including land), technology and access to finance? Methodology: To answer the different questions stated above, the work on the report will consist of both quantitative and qualitative studies. It will build on existing work and on expertise by the World Bank and development partners. It will also involve collaboration with stakeholders from different government agencies, private sector, universities, and industry associations.  Trends in city’s economic performance will be based on data from statistical agency (BPS) and data available with local government agencies. For example, it is instructive to contrast the rate of growth in jobs, productivity and exports of a given city with its peers in Indonesia and potentially from outside Indonesia to determine a city’s strengths and weaknesses and their key drivers. Based on data from BPS, we have compared the growth trajectories of top Indonesian cities in Figure 1. Between 2000 and 2010, Palembang’s real GRDP grew 3.3 percent per annum (behind most cities) due to only 2.1 percent per annum growth in productivity (using the proxy of real GRDP/capita). Denpasar’s productivity growth was even lower at 1.7 percent per annum between 2000 and 2010. Denpasar’s GRDP growth is more respectable at 6 percent per annum driven by 4.2 percent per year growth in population.  Sector specific analysis: To analyze drivers of economic performance, sector specific data will be used. In particular, we would like to identify the traded and resource based industries2 that are true source of competitiveness. World Bank has developed 2 Michael Porter's work on competitiveness and clusters (e.g., US Cluster mapping project) talks about 3 types of industries -  Local industries - customers are local e.g., construction, retail  Traded industries - goods and services traded across regions/nations due to higher productivity achieved in the cluster/city (e.g., Auto components)  Resource based industries - raw materials and other resources are local (e.g., Oil and gas) 31 tools and surveys which can be used to drive this analysis. Some of the major tools, which we are likely to use for this analysis will include Sector feasibility checklist, Enterprise Survey, Value chain mapping, Porter’s five forces, Market trends, and Cost Analysis (same as cost-structure benchmarking). Figure 2, 3 and 4 are illustrative slides on methodology used for sector analysis.  Leveraging Sub-national Doing Business (SNDB) database and insights, which was earlier work done by World Bank and IFC with the pilot cities  Interviews with mayors, planners on top economic priorities, initiatives underway, challenges faced, city administration organization structure, decision rights, KPIs and incentives  Interviews with industry associations, major private sector players, investors, banking analysts, and central bank  Literature and case study survey of work done by World Bank, donors, partner institutions, academics, Indonesian think tanks  Study of government masterplans including MP3EI (Masterplan for Acceleration of Regional Economic Development), RPJMN and city plans  Additional, targeted data collection for in-depth and sectoral studies, if needed and as agreed to with all stakeholders Timeline and phases: The work on the City Economic Competitiveness Review will span across all three phases. However, the first version of the report, based on existing data and key informant interviews only, will be completed during Phase 1, in order to support the dialogue and generation of policy options during the subsequent phases. The analysis will then be deepened and focused on agreed-upon sectors during phases 2 and 3. This may include additional data collection in the cities, if it turns out that such data is needed to complete the analytical work. Component 2: Capacity building Description: This part of the module links CPL outputs to the planning process, budget and resource allocation, land use, governance, and private sector decision support. The completion of the City Economic Competitiveness Review should translate into an ongoing capability and become an input into all the planning related to the cities’ economic development. The new analysis and insights from the review should be linked to action on the ground by (a) agreement on immediate initiatives and (b) permanent linking of new data and analyses to ongoing decision making by the city government and other players. Competitiveness may come initially by exploiting resource advantage (resource based industries), and may come sustainably by developing traded industries. One way to analyze these local versus non-local industry concentration in a given city or cluster is to look at industry share of total output of the cluster and compare it to national averages. Palembang’s share of rubber and palm oil industry output to total Palembang GRDP would be way above the national average across clusters – making them Palembang’s competitive sectors. 32 This can be achieved by capacity building at Bappeda and other relevant institutions through on-the-job personnel training, setting up appropriate decision analysis tools, and institutional changes to ensure application of results. Methodology:  Map all controllable decisions at a city level, current data and facts used for those decisions, current information gaps, and specific lab outputs that will bridge this gap. This step will use process maps/ decision rights and other tools  One-time intervention to agree on immediate initiatives for private sector development followed by other institutional interventions that improve decision making to support private sector development on an going basis. This is essentially transforming the way Bappeda works and will be typically achieved through internal decision making workshops involving all relevant stakeholders. Timeline and phases: The work on capacity building in this module will be especially closely coordinated with the activities of other modules, which also have capacity building as a key component across the three phases. It should be noted that the goal is to transfer the lead of the city economic competitiveness work to the CPL during Phase 3. Support of the work and of the CPL can continue, but the aim is to have the city government fully in the driver seat by that time. Component 3: Consultative workshops and Public-Private Dialogue (PPD) Description: The consultation with the private sector is critical throughout, since the success of this module is particularly dependent on the private sector’s reaction to the policy initiatives (and, later on, to the associated infrastructure investments). This module envisions, in addition to ongoing, informal consultations, 3-4 workshops that will involve all key stakeholders, including government agencies, private sector players of companies of different sizes (large, medium and small), representation from all major sectors, industry associations and other experts and academics. The workshops should serve the purpose of both information sharing (from the City Economic Competitiveness Review) and getting feedback. Timeline and phases: Proposed workshops and touch points (including timeline) include:  During Phase 1: initial work shop to discuss project setup, objective and vision sharing, collecting top hypothesis on challenges faced and solutions (in terms of controllable local decisions)  During Phase 2: work shop to discuss initial results from the City Economic Competitiveness Review, including comparison to peers; seeking input and agreement on sector prioritization. In addition, methodology and approach for remaining work can be shared and feedback sought  During Phase 3: After the expanded version of the City Economic Competitiveness Review has been finalized, this workshop is the major event that discusses results, 33 implications, city’s plans going forward and gets feedback from players. Subsequent formal periodic PPD program will be established. Component 4: City initiatives and dashboard Description: This component represents in effect the culmination and combination of the first three components. The output of this component puts the insights and recommendations into practice in the form of policy decisions and other initiatives that are agreed upon and an implementation plan to execute the initiatives. The proposed dashboard is a tool to achieve these objectives by bringing the critical information together, while the content of the actual policy decisions and initiatives will of course be customized to each city. Eventually, it is expected that the City Planning Lab will help prioritize and structure the investment projects for the city. The dashboard will have three sections:  city overall economic indicators;  city top sectors indicators;  city economic competitiveness initiatives scorecard While the final policy formulation itself is an output that will be developed during Phase 3, this component offers the opportunity to demonstrate what the City Economic Competitiveness module can do from the very beginning. During Phase 1, case studies of successful cities can be discussed with stakeholders to illustrate what might be possible and to motivate the stakeholders’ further engagement. During phases 2 and 3, the development, discussion and finalization of policy options are a culmination of the work done as part of the analysis, capacity building and dialogue. The dashboard itself will be generated starting during Phase 3 and periodically by CPL thereafter. There will be formal setting (perhaps a quarterly steering committee review) where the Mayor of the city will review the dashboard with all relevant stakeholders to take stock of progress and make important decisions. This could be combined with review of other modules of the P3N Methodology: For city wide economic indicators and sector specific indicators, we will leverage Bank’s deep experience in helping cities and creating Mayor’s dashboards along with deep consultation with all stakeholders. For the third aspect on initiatives, World Bank has significant experience in design and delivery of project monitoring and evaluation (M&E) framework. Typically, the framework includes metrics at input, output, outcome and impact level. Timeline and phases: While some discussion of case studies will take part in Phase 1, the work of this component will start in earnest during Phase 2 and intensify during Phase 3. 34 Table 3. Example of Analysis: Ratings of Indonesian Cities on Economic Performance (2000 – 2010) Rank Real GRDP 2010 Population GRDP/capita Real GRDP Population GDRGP/cap Jobs 2010 (IDR Tril) 2010 (Mil) 2010 growth growth growth (Mil) (IDR Mil) (% p.a) (% p.a) (% p.a) 1 Jakarta (JKT) 396 JKT 9.6 JKT 41 BDG 16.1 BDG 8.1 MDN 9.3 JKT 4.3 2 Surabaya (SBY) 86 SBY 2.8 SBY 32 MDN 9.1 BTM 8.0 BDG 7.4 SBY 1.2 3 Medan (MDN) 36 BDG 2.4 BTM 30 SBY 6.5 DPS 4.2 SBY 5.9 BDG 1.0 4 Bandung (BDG) 32 MDN 2.1 MDN 17 DPS 6.0 BGR 2.4 JKT 4.2 MDN 0.8 5 Batam (BTM) 28 SMG 1.6 SMG 14 BTM 5.8 JKT 1.4 MKS 4.1 SMG 0.7 6 Semarang (SMG) 21 PLB 1.5 BDG 13 JKT 5.7 PLB 1.1 SMG 3.3 PLB 0.5 7 Palembang (PLB) 18 MKS 1.3 MKS 12 BGR 5.6 MKS 0.9 BGR 3.1 MKS 0.4 8 Makassar (MKS) 16 BGR 1.0 PLB 11 MKS 5.1 SBY 0.5 PLB 2.1 BGR 0.4 9 Denpasar (DPS) 6 BTM 0.9 DPS 7 PLB 3.3 MDN -0.2 DPS 1.7 BTM 0.4 10 Bogor (BGR) 5 DPS 0.8 BGR 5 SMG 1.2 SMG -2.0 BTM -2.0 DPS 0.4 An analysis of Indonesian cities reveals wide variation in economic performance. A more nuanced story of Indonesian growth over the last decade (2000 – 2010) - Palembang real GRDP grew 3.3% p.a (lower than national average) driven one third by population growth and two third by productivity growth - Denpasar real GRDP grew 6%, but most of it came from population growth (4.2% p.a) with productivity growth declining (1.7% p.a) Source: Team’s analysis from BPS data (various years) Figure 5. Methodology Illustration: Export Performance Tool 35 Figure 6. Methodology Illustration: Value Chain Mapping Tool Figure 7. Methodology Illustration: Cost Structure Analysis Tool 36 3.4 RISKS AND MITIGATION The project has a few key risks which have been highlighted in the table below, along with mitigation measures. Table 4. Risks and Mitigation of Economic Competitiveness Module Key Risks Potential Mitigation The data for comprehensive Creative combination of different data sources will analysis does not exist (i.e. quality seek to quantify any remaining data gaps and the of analysis is limited due to data resulting uncertainty in the analyses gaps) The data exists but is not delivered, Getting a client owner, possibly at MoF or MPW, to due to coordination failures act as influential coordinator between agencies The data is of poor quality Data quality tests underway and design of project contingent on agreement on data quality Delays or failure to adequately Availability of CPL physical space and staff staff the CPL with right talent commitment to be used as an engagement criteria with clients Project monitoring and governance Bank will insist on formal institutional arrangements to risk (e.g. Mayor’s dashboard is not ensure project monitoring discipline used in practice) 3.5 OUTPUTS As indicated above, the following outputs will be produced as part of the City Economic Competitiveness Module: i. City economic competitiveness review ii. City economic planning and decision support capacity building iii. 2-4 Workshops for public-private dialogues iv. City economic competitiveness initiative roll out and monitoring dashboard 3.6 TEAM In addition to the regular staff of the City Planning Lab, the expected composition of the technical assistance team specific to this activity is as follows: (i) Senior Economist, team leader, (ii) Finance and PSD Specialist, (iii) Economist, (iv) Competitive Industries Practice Specialist 37 3.7 RESOURCE ALLOCATION AND TIMELINE Significant emphasis is being put on design and implementation rather than data analysis. Approximately two-third of cost is concentrated for travel and on the ground work, while a third is for pre and post mission desk based analytics. Implementation of the module emphasizes using local expertise complemented with the World Bank’s international knowledge and experience. The module envisages 8 week-long missions to the two cities. To maintain the momentum and continue to build the needed relationships, as well as to provide capacity building on the competitiveness module to the city planners in the City Planning Lab, a local consultant will be hired to be the person on the ground in both cities. 38 ANNEX MODULE B: ASSESSMENT OF DATA ENVIRONMENT It must be noted that some of the non-government databases in the table below have solid historical data but weak prospects of periodic updates in future (e.g. KPPOD). Table 5. Inventory of Data for Economic Competitiveness Module Area Description Coverage Period Source General/ City/Kabupaten in Figures; All cities and 1990 – 2010 BPS background published annually kabupaten, nationwide information General/ Susenas (Household Survey) All cities and 1976 – 2010 BPS background kabupaten, nationwide information Manufacturing Statistik Industri (census of Nationwide. Data 1990 – 2010 BPS sector large and medium available at city/ manufacturing, 20+ kabupaten level employees) Manufacturing Survey of micro (1-4 Nationwide. Data 2010, 2012 BPS sector employees) and small (5 – 19 available at provincial employees) manufacturing level companies Manufacturing Number of small industry Some cities/kabupaten Varies City/Kabupaten sector establishments and employees in Figures (BPS) Service Survey on manufacturing and Nationwide 2006 BPS non-manufacturing firms in Economic Census Business climate SNG Doing Business assessment 14 cities (2010) 2010, 2012 WB - IFC on metrics such as days, cost, 20 cities (2012) number of procedures to obtain licenses and permits Business climate Regulatory environment for 90 C/K (2001) 2001 – 2005, KPPOD doing business; economic 134 C/K (2002) 2007, 2011 governance 200 C/K (2003) 214 – 245 C/K (2004- 2011) Business climate C/K rankings and Autonomy C/K in East Java, 2009 – 2012 Jawa Pos Award on economic Central Java, DI Institute Pro- development, public service, Yogyakarta, South Autonomy (JPIP) and local political performance Sulawesi Labor Sakernas (Labor Survey): Nationwide 1976 – 2010 BPS survey of employment status, field of work Banking Credit trend lines by sector Select cities Bank Indonesia and by size of firm Regional Offices Local economy GRDP by economic sector Nationwide. Data 1990 – 2010 BPS available at C/K level Other Enterprise Survey Varies WB Transportation Survey Manufacturing Survey SME Survey (underway) 39 40 SECTOR MODULE C: SLUM ANALYTICS AND MANAGEMENT SYSTEMS 41 42 4.1 BACKGROUND Like many rapidly urbanizing countries, Indonesia has seen the growth of informal settlements in many of its cities. The Ministry of Public Work estimates that a quarter of the urban population (roughly 25 million people) lives in slums and informal settlements. While the growth of slums is an indicator of the economic draw of urban areas, it is also a sign of inefficient land and housing markets, and unequal access to urban services. Addressing existing slums is critical to alleviating urban inequality, while prevention of future slum growth and protection of land rights is essential to attracting investment to cities. Some Indonesian cities have taken innovative and progressive approaches to slum upgrading, and through policies and small investments have managed to upgrade slums into viable neighborhoods for poor urban communities. In-situ upgrading of slums is not always possible, however, since they are often located on risk-prone or contested land. Most cities are forced to try to address the issue of slums in the absence of vital information. Cities often have no systematic way to answer basic questions about slums, such as: i. What are the primary causes of slum formation in the city? ii. How do slum dwellers make location choices? iii. How have recent government policies or actions (e.g. housing policies, infrastructure and service provision, slum upgrading, formalization, land sales, etc.) affected slum residents and the formation of new slums? iv. How does a slum household’s intention to invest in or otherwise upgrade their dwelling correlate with other factors, such as tenure security, income, duration of residence, etc.? v. What determines prices/ rents in slums, e.g. tenure security, distance from various amenities, etc.? vi. How can slum areas be classified in terms of their origins, characteristics, or expectations of future growth, in order to devise the most appropriate government responses? In order to answer these and other questions, slum analytics and management systems will be a key strategic activity of the City Planning Lab (CPL). The slum information database produced as part of this work will be an important input into decisions on investments in basic infrastructure and services, helping devise appropriate interventions and target them to areas of greatest need. It will also help cities devise more effective slum policies and regulations. Performing this activity as part of the broader City Planning Lab initiative will take advantage of several synergies, as the findings will both feed into and benefit from the analytical work in parallel modules. The findings on distribution of slums will support the work on spatial growth analytics, which in turn will help provide spatial context to the growth of slums. It will also be a strong indicator of where the demand for affordable 43 land is likely to be highest, which will add value to the land market analytics module. The slum analytics and land market analytics together can help the city identify and proactivity respond to high demand for land before new slums emerge. The module on disaster and climate risk resilience will help identify vulnerable slums. Not only will these synergies provide efficiency through shared data and analysis, they will also ensure that the analysis done by the CPL as a whole puts special emphasis on the most disadvantaged populations. 4.2 OBJECTIVES The main objective of this module is to assist partner cities in improving the management of slum areas, using detailed information and mapping of slum areas and vacant lands. Technical assistance provided under the module will consist of three components with specific objective as follows: Component 1: Slum mapping and information systems: The objective of this component is to help the city develop and maintain a geo-referenced database of slums, using satellite imagery and other data sources to provide an overview of the slum situation, as well as a survey in selected areas recording attributes such as legal status, year of construction, quality of construction, disaster risk, price/rent, access to urban services, access to transportation networks, etc. This database would be managed and maintained within the City Planning Lab, and would allow slum-related policies to be informed by an empirical understanding of the needs of slum communities. Component 2: Slum management framework: The objective of this component is to use the analysis emerging from the City Planning Lab’s database developed in component 1 to formulate a medium-term program for a citywide strategy and investment program targeting existing slum areas. This includes identifying slum areas that are suitable for in-situ upgrading, and those which are vulnerable to disaster risk and where resettlement may be required. This slum management framework would outline strategies for community participation, institutional capacity building, and investments. Component 3: Managing new slum growth: The objective of this component is to work with city leaders to develop strategies to prevent growth of new slums in areas identified as vulnerable to disaster risk or planned for public use. 4.3 SCOPE OF ACTIVITIES In keeping with the other CPL modules, the activities in this module will be conducted in three phases of six months each. Component 1 will be completed at the end of phase 1. Components 2 and 3 will begin at the start of phase 2 and will carry over into phase 3 (see timeline below). The duration of the complete module is 18 months. 44 Component 1: Slum Mapping and Information Systems This component will be divided into three stages, as follows: 1A: Creation of a basic Slum Information Database This stage will involve the creation of a basic version of the GIS database on slums (a part of the CPL’s broader database) populating it with all information on slums that can be derived from satellite imagery and field visits. The goal will be to develop a preliminary picture of the situation in the city with regard to slums. The tasks for this component in this stage would include: i. Gathering existing spatial data on the location of slums in the city from government and other sources; ii. Converting the above data into standard, non-proprietary formats (e.g. shape files, KML files), digitizing paper maps where necessary, and recording the associated metadata; iii. Using satellite imagery (e.g. Google Earth) to identify possible slum areas in the city; iv. Conducting field visits to these areas for verification (with photographic documentation of slum areas during visits); v. Using data collected from all the preceding steps to create a GIS map of slums in the city. Each slum area will be represented as a polygon depicting the boundaries of the slum area, and associated with a table of attributes reflecting all the available data for each slum; and vi. Uploading all unclassified data to the local government website in the form of PDFs and GIS files, as well as to an open source web-based mapping service (e.g. OpenStreetMap). 1B: Additional secondary data collection This stage will involve gathering additional secondary data that can help develop a more complete picture of the characteristics of slums and slum households, and inputting this data into the Slum Information Database. Tasks during this phase will include: i. Gathering and inputting into the database, in the standard format, all available data on demographic and socio-economic characteristics of slum and non-slum households, citywide land ownership, zoning, disaster risk, transportation infrastructure and routes, utilities (water, sanitation, drainage), and other relevant data; ii. Adding all unclassified data collected during this phase to the open source web-based mapping service used in component 1A. 1C: Primary data collection and analysis This stage will involve conducting household surveys to collect primary data in selected slum areas, to further develop the database and contribute to a more robust understanding of the slum areas. Tasks during this phase will include: 45 i. Designing a methodology and questionnaire for a household survey, covering a significant number of random households in selected, high-priority slum areas. (See Annex section for an indicative list of the kinds of household attributes that may be included.) ii. Gathering feedback on the survey design and site selection from local government counterparts and external stakeholders and refining it accordingly; iii. Conducting the household survey, while also recording constraints faced while conducting the survey (e.g. households inaccessible, households declining to respond, gender/age bias in respondents, etc.); iv. Adding all information to the Slum Information Database, with metadata; v. Analyzing the data obtained from all phases, through regression analysis and other means, in order to answer the questions listed in the ‘Rationale’ section above and others. Component 2: Slum Management Framework This component aims to formulate a medium-term program for a citywide strategy and investment program targeting existing slum areas. Activities in this component would include: i. Identifying slum areas that are suitable for in-situ upgrading, and those which are vulnerable to disaster risk and where resettlement may be required. ii. Creating the typology of existing slum areas based on its characteristics such as type of dwelling, dwellers, tenure status, land ownership, etc., identified and analyzed in Component 1 above. iii. Developing strategies based on the typology of area, which includes site analysis, building and urban design, land management, financial assessment, and temporary shelter. This would utilize the phase 1 outputs of the other CPL modules. iv. Outlining strategies for community participation, institutional capacity building, and investments for pilot sites, selected based on discussion with the city government. v. Developing a program of future activities to implement the selected strategies, in coordination with related government agencies. An example of output produced by similar activity is HABISP, a sophisticated housing information system in City of Sao Paulo, Brazil. The system features maps with data on slums and other low income housing. 46 Figure 8 A,B. Example of Analysis: Sao Paulo HABISP Online Housing Information System 47 Component 3: Managing New Slum Growth This component aims to develop strategies to prevent the formation of new slums in areas of high risk or those reserved for public use. Activities in this component will include: i. Together with the team working on the disaster risk module of the CPL, identifying all currently vacant land in the city that is: (a) prone to disaster risk (using existing data); or (b) identified in the current city spatial plan as usable for public purposes, documenting the ownership status of all such land (private or public, and if public, which agency), and inputting this data into the Slum Information Database described above; ii. Developing an understanding of the process by which land is encroached by slums in the city, using case studies or other means; iii. On the basis of this understanding, making recommendations for strategies to safeguard the land identified in task (i) above. These recommendations should address: a) reforms to regulations; b) reforms to enforcement procedures; c) capacity building of relevant institutions; d) reforms to the process of land use planning; e) public awareness strategies; and f) community-based prevention and participation strategies. iv. Working with local government agencies to help them implement the recommended strategies. Workshops In order to ensure that the technical assistance activity is useful to the city at every stage, the team will conduct workshops in order to share the results of the work done so far, as well as to receive guidance from government leaders on future directions. i. A kick-off workshop will be held in order to discuss the work ahead and establish working procedures; ii. An interim workshop will be held in month 7, to share the findings and recommendations of all activities completed up to that point, including all of component 1, and to develop plans for the collaboration with the counterparts for the remaining duration. iii. A wrap-up workshop will be held at the completion of all activities, to discuss plans for government agencies and/or donors to carry on the work, and reflect on lessons learned. 48 4.4 RISKS AND MITIGATION The primary risk with this activity is that the information from the database built as part of the first component and the recommendations that emerge from the second and third components will not be mainstreamed into either the day-to-day decision making with regard to municipal actions affecting slums, or into the long-term visioning and planning for the city. The team will address this potential risk by working closely with the staff of various city agencies during the various activities, under the City Planning Lab framework, as well as periodically consulting with city leaders through workshops, in order to ensure that the data collected and the recommendations for slum policy are relevant to the city’s needs. 4.5 OUTPUTS The following outputs are expected from the technical assistance: Component 1: Slum Mapping and Information Systems i. All data gathered during all three stages, with metadata, transferred to the Bank team and to the relevant government agency as digital files in a standard format, and uploaded to an existing online mapping service; ii. A report describing: (a) hosting options; and (b) future enhancements to the database. iii. All materials associated with survey, including: a) completed questionnaires (may be in original language of survey, may be scanned hard copies); b) a spreadsheet displaying the data collected; and c) a report briefly describing methodology and constraints, and summarizing the findings. Component 2: Slum Management Framework i. A report outlining typology of existing slum areas in the city and strategies for its management. ii. Selection of slum area sites as pilot projects for implementing strategies and the programming of activities for implementation. Component 3: Managing New Slum Growth i. Layers in the GIS database mapping vacant land, hazard-prone vacant land, and vacant land zoned for public use, with all available associated data, including ownership; ii. A report outlining: 49 a) the process of slum formation, as observed through case studies; and b) recommendations for strategies for preventing slum growth in disaster-prone or strategic areas, as described earlier. 4.6 TEAM In addition to the regular staff of the City Planning Lab, the expected composition of the technical assistance team specific to this activity is as follows: i. Urban Planner as Team Leader ii. Social/Low-income Housing Specialist iii. Economist iv. Community Development Specialist v. GIS Specialist vi. Urban Design Specialist vii. Governance/Institutional Specialist 4.7 TIMELINE This module will be carried out in three phases of six months each. 50 ANNEX MODULE C: DATA COLLECTION The following is an indicative list of the kind of information gathered from primary and secondary sources and input into the slum information database3: Slum attributes: 1. Name 2. Location – jurisdiction 3. Location – description (urban core or fringe area) 4. Year of establishment 5. Area of slum (sq. meters) 6. Land use surrounding slum (residential/ commercial/ industrial/ other) 7. Physical location of slum (along road/ along railway tracks/ riverside etc.) 8. Legal status of slum (provide details) 9. Ownership of land 10. Estimated population 11. Estimated Number of households 12. Primary source of water 13. Primary sanitation facility 14. Primary means of garbage disposal 15. Connectivity to citywide water supply system (fully/ partially/ not connected) 16. Connectivity to citywide storm-water drainage system 17. Connectivity to citywide water sewerage system 18. Flooding risk (no flooding/ floods 15 days a year/ 15-30 days/ more than 30 days) 19. Frequency of garbage disposal 20. Frequency of clearance of open drains 21. Condition of access road to slum (paved/ unpaved, motorable/ unmotorable) 22. Condition of internal roads 23. Distance from nearest motorable road 24. Street light availability 25. Distance to nearest pre-primary, primary, high school 26. Distance to nearest primary health care facility, public hospital, maternity center 27. Availability of communal facilities (meeting halls/training center/ night shelter, etc.) 28. Active presence of NGOs Household Attributes: 1. Name of slum 2. Address (house number, street) 3. Existence of formal street addressing 4. Number of family members 5. Number of school-age children 3Adapted from “Formats and Guidelines for Survey and Preparation of Slum, Household and Livelihood Profiles of Cities/Towns”, Government of India Ministry of Housing and Urban Poverty Alleviation, National Buildings Organization. 51 6. Number of disabled people 7. Land tenure status 8. Type of structure (permanent/ temporary) 9. Construction material used in floor 10. Construction material used in roof 11. Source of light 12. Source of cooking fuel 13. Source of drinking water 14. If piped water, duration of availability during the day 15. If outside source of water, distance from dwelling 16. Existence of toilet facility 17. Bathroom facility 18. Condition of road in front of house 19. Vehicle ownership (none/ bicycle/ motorcycle/ care/ truck) 20. Number of years in current dwelling 21. Migrated from (urban/ rural) 22. Reason for migration 23. Migration type (seasonal/ permanent) 24. Number of earning adult family members 25. Number of earning non-adult family members 26. Number of non-family adult members (specify if renter) 27. Average monthly income of household 28. Average monthly expenditure of household 29. Debt outstanding as on date of survey 30. Educational qualifications/ training of adult members 31. Employment status (self-employed/ salaried/ casual labor/ others) 32. Place of work (within/ outside slum) 33. Length of daily commute 34. Mode of daily commute 35. Monthly earning 36. Source of income 37. Income-generating activity within dwelling unit (home-based industry/ commerce) 38. If unemployed, main reason for unemployment 39. Acquisition of dwelling unit (self-built/ bought/ rented) 40. Price / rent of dwelling unit 41. Income from renting space in dwelling unit 42. Intention to invest/ upgrade dwelling 43. Intention to move away 44. Major constraints to formal housing Business attributes 1. Type of business 2. Average monthly/ annual earnings 3. Seasonal/ regular 4. Number of household members employed 5. Number of hired employees 6. Resource needs of business (water/ power) 7. Waste produced by business 52 8. Means of waste disposal 9. Spatial needs of business (must be easily accessible to public, e.g. in a market place/ space needed for production or processing/ other) 10. Intention to expand (none/ more employees/ more space) 53 54 SECTOR MODULE D: DISASTER AND CLIMATE RESILIENT PLANNING ANALYTICS 55 56 5.1 BACKGROUND Indonesia’s rapidly growing urban population is particularly vulnerable to natural disasters. More than 110 million people in roughly 60 cities, mostly located in coastal areas are exposed to hazards including tsunamis, earthquakes, flooding and impacts of climate change. With nearly 70 percent of Indonesia’s population expected to live in urban areas by 2025, coupled with the increasing wealth of the population, Indonesian cities are increasingly vulnerable to both large-scale and persistent natural hazard events. The limited capacity of urban centers to absorb new residents because of lack of fundamental infrastructure investments has also resulted in the creation of many unplanned settlements. Inadequate zoning and lax enforcement led to the occupation of many hazard-prone locations. The Ministry of Public Work estimates that a quarter of the urban population (roughly 25 million people) lives in slums and informal settlements. Indonesia’s unique geological setting and the complexity of its population settlements has generally led to more disasters causing greater damage (loss of life, economic impacts etc). Although hazardous natural events cannot be prevented, the severity of their consequences can be minimized or even avoided through disaster and climate sensitive urban development coupled with better community preparedness and enhanced coping capacity to achieve greater city/urban resilience. Climate change and variability in the near and long term can only increase the level of risk. In addition to higher intensity meteorological events such as floods and droughts, the climate also influences food production patterns and outputs, creating additional uncertainty in the event of a disaster that further exacerbates its impact. While there is growing awareness of the need to address the impact of climate variability and change, more accurate identification of vulnerability and evidence-based response and adaptation measures must be developed. Cities also often lack the fiscal capacity to initiate programs that require sophisticated technical expertise and dedicated investment. In preparation for addressing topic, World Bank team has engaged in disaster and climate risk reviews in six cities. The purpose was to take stock of the baseline information on climate and disaster risks and identify critical gaps in addressing the cities’ risk sensitive planning and investment needs thereby setting the priorities for this proposed module on disaster and climate resilient planning analytics. The current urban planning practice in Indonesia still consider hazards and risks from disaster and climate change only as constraining parameters in the selection of sites suitable for development. Where the risks originate and how existing growth trends and investment will impact or be impacted by the pattern of disasters have not been thoroughly analyzed during the planning process. As part of the objective of Metropolitan and Urban Development Program (P3N) to establish technical capacity to measure, analyze and respond to urban development pressures in evidence-based and timely manner, a Disaster and Climate Resilient Planning Module is needed to address the following challenges: 57  The gap in high-resolution hazard and exposure information required for a detailed city-level planning;  The absence of policy instruments and practical guidelines to introduce disaster and climate resilient practices into detailed spatial plans and city level infrastructure investments decisions.  The lack of customized geospatial analytical tools to conduct risk analysis that can easily integrate various data sources and facilitate the implementation of risk sensitive planning; Addressing these issues will be a key strategic activity of the City Planning Lab (CPL) to take advantage of several synergies, as the findings will both feed into and benefit from the analytical work in parallel modules. The Spatial Growth Analytics and Slum Analytics and Management Systems modules provide data that when combined with the climate and disaster risk analytics can provide valuable insights into the largest potential threats to the city’s future growth. 5.2 OBJECTIVES The primary objective of the Disaster and Climate Resilient Planning Module is to provide essential risk information and analytics to measure, analyze and identify options to address urban development pressures from disaster and climate related hazards. The overarching objective of the work across the three components will be building the capacity of local agencies in undertaking thorough and integrated but practical analysis incorporating disaster and climate risk management options into city investment program. The methods and approaches used during these activities may be adopted and continued by local agencies beyond the timeframe of this engagement and become standard practice in the city’s approach to resilient urban management inclusive of land use and infrastructure planning. The specific goals that will be fulfilled include: Component 1: Filling risk information and data gaps This component will compile or develop baseline hazard and asset exposure data as essential inputs for climate and disaster risk analyses that inform planning and investment decisions. Priority areas identified in the climate and disaster risk review through area- focused and risk-based approach as needing higher resolution data will be addressed developed by combining several potential information sources. Expert sources at technical agencies or universities, as well as participatory methods to engage the community and civil society are critical to developing robust hazard and exposure data. There will be an element focused on improved data sharing and management. In coordination with the core CPL, this will include both the platform software that can integrate with other P3N activities and the policies that can be established for city agencies in line with the guidelines set out in the National One Map Initiative of the Geospatial Information (BIG). Component 2: Establishing capacity to carry out detailed land use planning and infrastructure investment screening This component will specifically build the capacity of targeted cities to implement the three options for disaster and climate risk management through translating the preventive, 58 avoidance, and adaptive approaches into practical targeted investment under the slum management and urban growth modules of the City Planning Lab of the Metropolitan and Urban Development Program (P3N). This is to enable the use of risk information to support resilience in key sectoral operations such as land-use zoning and infrastructure planning. In coordination with regional divisions of the Ministry Public Works’ DG Spatial Planning, a pilot of detailed risk-sensitive spatial planning will be carried out to showcase evidence-based planning enforcement/action instruments. Component 3: Developing tool for practical Climate and Disaster Risk Analysis This component will enable the integration of risk information into the City Planning Labs data platforms and analytical capabilities based on the guidelines developed in Component 2. The risk data can be accessed by analytical modules to support different planning functions within the city government (e.g., zoning, infrastructure, community actions/development). The current Indonesia Scenario Assessment for Emergencies (InaSAFE) which is still focused on contingency planning application will be expanded with additional analytical modules. InaSAFE tool supports better disaster risk reduction decision-making by providing a simple yet rigorous approach to analyzing the likely effects of future disaster events or climate change scenario. This component will use the baseline risk information generated in the first component as the data stream. Figure 9 A,B. The InaSAFE Tool 59 5.3 SCOPE OF ACTIVITIES The activities in this module will be conducted over a span of 18 months split into three month increments for initial planning purposes. Component 1: Baseline Risk Information and Participatory Mapping This component will be divided into three sub components, as follows: 1A. Baseline information on hazards This sub-component will involve the creation of a basic version of the GIS database, and populating it with all information on key hazards. National-level agencies such as BNPB, Badan Geologi, and BMKG as well as Universities are producing highly technical, scientific information on hazards and risk. However, with improved coordination and capacity development, cities can take better advantage of existing information and be aware of gaps and need to invest in better data to support local level resilience activities. New hazard information needs will be identified during phase 1 of the risk review. For example, if a city is planning micro drainage investments, ideally there needs to be a detailed flood hazard model to develop risk-sensitive design standards as well as the necessary micro zoning in the surrounding areas. The tasks for this component in this phase would include: 60 i. Gather existing spatial data on hazards affecting the city from government and other sources identified in the risk review; Convert the above data into standard, non-proprietary formats (e.g. shapefiles, KML files), digitizing paper maps if necessary, and recording the associated metadata; ii. Develop new scenario or probabilistic hazard data based on the specific needs defined in scoping phase; iii. Confirm that all hazard data is formatted in InaSAFE- compatible files; iv. Determine data sustainability issues including methods for updating, guidelines for licensing and official usage. This activity would be coordinated with the core CPL components. 1B. Participatory mapping to develop baseline administrative boundaries, public asset inventory of critical infrastructure, and past hazard event or hazard prone data. This sub-component will involve gathering information to create a GIS enabled database of Kelurahan and RT-RW (ward/neighborhood) boundaries, public assets including critical infrastructure, and detailed GIS data of past hazard events. The methodology will follow BIG’s draft Standard Operating Procedures developed under the Participatory One Map Initiative (POMI). Tasks during this phase will include: i. Evaluate the resolution of freely available satellite imagery through OpenStreetMap platform; ii. Establish working group of technical stakeholders for the participatory mapping, provide training for group to learn OpenStreetMap tools and platform; iii. Gather existing spatial data on RW-RT, public assets from government and other sources, perform basic validation and/or conversion into standardized GIS format; iv. Organize community workshop with OpenStreetMap training to gathering information on critical infrastructure for baseline data; v. Collect data on past hazards to develop maps of hazard prone areas at the RW-RT level; vi. Conduct quality assurance and validation of each data set. 1C. Institutional data management and sharing It is necessary to establish good protocols for data sharing between government stakeholders and with the broader community of civil society, private sector. i. Workshop with key stakeholders to review existing data sharing process and to present options for implementing; ii. Customization of data sharing agreements based on workshop feedback. Component 2: Resilient Land Use Planning and Infrastructure Investment Guidelines The risk-based and area focused land use planning component aims to: i) identify and mitigate the root cause of disaster risks embedded in existing land development practices through regulated use of land in hazard-prone areas and building codes, ii) promote controlled urban growth without generating new risks, ‘building back better’ through 61 rebuilding and upgrading infrastructure using hazard-resistant construction in accordance with a comprehensive plan. In close coordination with an operationalized risk-based land use planning mechanism supported by the Ministry Public Work’s Directorate General of Spatial Planning, cities can be supported in make detailed spatial plan as the basis for locational decision of investments that have the primary purpose of risk reductions such as urban drainage and flood control or screening mechanisms that would introduce resilience criteria in infrastructure design, construction and economic development more broadly. The activities under this component will include: i. Dissemination of the detailed risk-sensitive spatial planning principles and guidelines and their practical implications to city operations and urban management; ii. Visioning exercises and mentoring to catalyze a holistic analytical-based planning process informed by the data developed in Component 1 in which disaster and climate risk management serve as the norm for balancing city’s growth and community resilience (i.e., green development) for key investment in city services such as utility, transport/mobility and natural landscape and water management; iii. Conduct participatory planning workshop targeting selected high-risk areas in the cities to present sectoral implications and options for rezoning, redevelopment and adaptive investment.. This exercise will reinforce the use of detailed spatial planning process as action instrument to ‘enforce’ the spatial plan; iv. Develop risk-sensitive planning and investment guidelines through the translation of the vulnerability and site planning spatial analysis into detailed zoning map and its descriptive land use designation and restriction. Component 3: Climate and Disaster Risk Analysis Tools Using the hazard and asset exposure data collected in Component 1, it will be possible to conduct a baseline risk analysis for the city. It is important that capacity be built within the CPL to easily use the results and conduct various secondary analyses related to planning and urban management. This project will leverage the InaSAFE tool which supports better disaster risk reduction decision-making by providing a simple yet rigorous approach to analyzing the likely effects of future disaster events or climate change scenarios. Under this component, this tool will be adapted and applied in support of analytics for various risk sensitive land use and infrastructure investment planning. The activities under this component will include: i. Expand user need assessment based on priorities identified in the risk review for spatial analysis of disaster and climate risk impacts to support various city level sectoral and area-based planning; ii. Design of user-friendly GIS functionality within the software architecture of InaSAFE that is compatible with the spatial data infrastructure of the CPL; iii. Develop demonstration version testing of InaSAFE in the CPL to show results of baseline analysis; iv. Customize modular tools to support integration of risk analytics into detailed spatial planning and infrastructure investment screening as defined by the guidelines in Component 2; and v. Training and integration of the tool into the CPL’s core functions. 62 Workshops In order to ensure that the technical assistance activity is useful to the city at every stage, the team will conduct overview workshops in order to share the results of the work done so far, as well as to receive guidance from government leaders on future directions. The workshops are also important opportunities to foster partnerships between the local government leaders, CPL, Universities, and Civil Society groups. These meetings will be in addition to the participatory activities embedded within individual components. i. A kick-off workshop will be held in order to discuss the work ahead and establish working procedures including government leadership for the participatory mapping exercises of Component 1; ii. An interim workshop will be held in month 7, to share the findings and recommendations of all activities completed up to that point, including all of component 1 and 2, and to develop plans for the collaboration with the counterparts for the remaining duration. iii. A wrap-up workshop will be held at the completion of all activities, to discuss plans for government agencies and/or donors to carry on the work, and reflect on lessons learned. 5.4 RISKS AND MITIGATION The primary risk with this activity is that the information from the database built as part of the first component, the analytics from the second and the recommendations that emerge from third components will not be mainstreamed into either the day-to-day decision making with regard to municipal actions building disaster and climate resilience, or into the long-term visioning and planning for the city. The team will address this potential risk by working closely with the staff of various city agencies during the various activities, under the City Planning Lab framework, as well as periodically consulting with city leaders through workshops, in order to ensure that the disaster and climate risk data collected and resilient planning guidelines are relevant to the city’s needs. 5.5 OUTPUTS The following outputs are expected from the technical assistance: Component 1: Baseline Risk Information and Participatory Mapping i. All data gathered during all three stages, with metadata, transferred to the Bank team and to the relevant government agency as (a) digital files in a standard format, 63 (b) uploaded to an existing online mapping service, (c) selection of hard copy maps for key data sets professionally designed for display in city offices; ii. Partnership agreements, data sharing between key data providers and the CPL. Establish an extended network of technical experts to provide future advisory services on climate and disaster risk assessments. iii. A report describing: (a) hazard modeling methodology, including resolution and limits of use for output hazard maps data, (b) strategy for the maintenance and/or updating of the data, (c) summary of the data sharing and management workshop; iv. Materials associated with the participatory mapping exercise, including: a) Field survey templates; b) Extracted and GIS files; c) Customized training materials ; and d) A report briefly describing survey and quality assurance methodology, roster of trained volunteers and city staff, and summarizing the findings. Component 2: Resilient Land Use Planning and Infrastructure Investment Guidelines A report outlining: a) the drivers of disaster and climate risk to core sectors and areas/neighborhoods; b) risk-sensitive micro zoning maps; and c) recommendations and practical roadmap for implementing the resilient landuse and infrastructure investment guidelines. Component 3: Climate and Disaster Risk Analysis Tools i. A report outlining the users’ needs assessment findings and design criteria for the customization of the InaSAFE tool. ii. Users’ manual, support documentation, and detailed training materials for InaSAFE. iii. Fully deployed and bug tested installation of InaSAFE software on CPL servers. 5.6 TEAM In addition to the regular staff of the City Planning Lab, the expected composition of the technical assistance team specific to this activity is as follows: i. Disaster Risk Management Specialist as Team Leader ii. Climate and Natural Disaster Hazard Specialist x 2 iii. Community Mapping Specialist x 2 iv. GIS Specialist 5.7 TIMELINE This module will be carried out in three phases of six months each. 64 ANNEX MODULE D: DATA COLLECTION The following is an indicative list of the kind of information gathered from primary and secondary sources during the scoping phase: Table 6. Inventory of Data for Disaster and Climate Resilient Module Types of Data Source Physical condition Geography Cipta Karya of the city Topography Cipta Karya Geology Cipta Karya Social Population BPS Total population by sex BPS Total population by age BPS Density BPS Growth and projection BPS Urban population in coastal cities BPS Economy PDRB BPS Dominant economy activity BPS Income BPS Budget and subsidy (For DRR and CCA) Bappeda Hazard Type of hazard BPBD History BPBD Intensity BPBD Level of hazard BPBD Damage (loss) BPBD Vulnerability Main infrastructure Cipta Karya Population welfare Bappeda Vulnerability projection Cipta Karya/ Bappeda/ Related Agency Risk Type of risk BPBD Level of risk BPBD Agriculture/food security Agriculture Agency Forestry Forestry Agency Water shortage Cipta Karya Biodiversity Forestry Agency Planning Spatial planning Cipta Karya Midterm Development Planning Bappeda Long Term Development Planning Bappeda Climate Indicator Type(rainfall, temperature, La Nina, El Nino, etc) BMKG/Agriculture Agency History BMKG/Agriculture Agency Trend and projection BMKG/Agriculture Agency Intensity BMKG/Agriculture Agency Sea Level Rise Level of sea level rise Bappeda/Related Agency Trend and projection Bappeda/Related Agency Mitigation Program (type) Bappeda/Related Agency Level of Mitigation Bappeda/Related Agency 65 Types of Data Source Mainstreaming to other planning document Bappeda/Related Agency Adaptation Program (type) Bappeda/Related Agency Level of adaptation Bappeda/Related Agency Mainstreaming to other planning document Bappeda/Related Agency Community education Bappeda Community preparedness Bappeda Institution Government agency (collaboration) Bappeda, BPBD Local NGO Bappeda, BPBD National NGO Bappeda, BPBD International NGO Bappeda, BPBD Local University/research center Bappeda, BPBD Related research document Bappeda, University 66 SECTOR MODULE E: MONITORING LAND AND REAL ESTATE MARKETS 67 68 6.1 BACKGROUND As in many rapidly urbanizing economies, the net worth of new constructions in the real estate market of Indonesia constitutes one of the largest sectors of annual investment and contributors to the GDP. The share of the construction sector in the GDP was 10.4 percent in 2012 – approximately a 90% increase from a 5.5-percent share in 2000 (Bank Indonesia, 2013). Real estate and construction sectors are among the major drivers of economic growth, along with transportation, communication and finance sectors. Growth in the construction sector, for instance, has outpaced the total annual GDP growth between 2000 and 2012 been by a factor of 1.45 times on average (Bank Indonesia, 2013). Between 2007 and 2011, 36 to 43 percent of all annual construction occurred in real estate products (Suraji, Pribadi & Ismono, 2012). Although an accurate assessment of the total value of the real estate market in Indonesia is not available due to lack of reliable data, a rudimentary estimation based on The Wealth of Nations Dataset of the World Bank (2005) suggests that urban land and structures that occupy it constitute approximately 20 percent of Indonesia’s $4.36 trillion total wealth.4 Given its durability, this large bundle of assets forms an important part of long-term national assets in Indonesia. Given the considerable importance of the land and real estate market for the economy, Indonesia’s cities would benefit from ensuring their efficient functioning. At present, however, most medium and large-scale municipal governments in Indonesia lack the institutional capacity to monitor the performance of their land and real estate market, or assess the impact of their policies and regulatory decisions on this market. The cities are unable to forecast the rapidly increasing demand for residential, commercial and industrial land for Masterplans and land use plans, and supply consequently does not meet demand. Resilient economic growth in Indonesia cannot be achieved without informed land and real estate policies that guarantee the availability of affordable space in demanded locations for living, working as well as recreation. As land and real estate markets provide space for all economic activities, they naturally impact various sectors of the economy. An inadequate provision of residential land, for instance, may inflate housing prices everywhere and trigger an increase in informal settlements, which in turn reduce the population’s spending capacity for transportation and other vital expenditures. In the absence of an understating of how real estate and land markets function, rapidly urbanizing cities of Indonesia expand on natural resources and agricultural land even before all existing urban land and infill development sites are exploited. Furthermore, lacking reliable data and analytics on land and real estate markets, Indonesian cities are unable to foresee and prevent abrupt fluctuations and bubbles in these markets. Lacking an empirical understanding of metropolitan growth, argues David E. Dowall (1995), leads to a “blind flight” for local governments and a failure to effectively deal with rapid population change and land development. In order to address these limitations and necessities, this concept note proposes to integrate a Land and Real Estate Market Monitoring Module to the planned activities for the P3N City Planning Lab facilities in two pilot cities in Indonesia. It discusses the needs 4 The total wealth estimate includes human and natural resources. 69 and objectives for necessary land and real estate market monitoring and lay out the proposed activities. This module is envisioned to be closely integrated with CPL’s core Spatial Growth Analytics module, since the way that physical expansion and internal restructuring take place, is largely determined by and reflected in the land and real estate market. Cities grow or change internally when the supply side of the market responds to changes in demand for different land uses by a) converting peripheral non-urban land to urban use, b) changing land-use designations internally, c) adding infill development and densifying, or d) changing occupancy levels on existing land uses. All these scenarios are monitored in the proposed Spatial Growth Analytics module. The present module is also connected to the Slum Analytics and Management Systems module, which collects and analyzes data on the informal housing and commercial land uses. 6.2 OBJECTIVE The primary objective of CPL’s land and real estate market monitoring module is to enhance the resilience and efficiency of these markets through supporting evidenced- based planning, policy and investment decisions. This includes projecting future demand for land and different types of real estate products – residential, commercial, industrial, institutional – and integrating the projection into Masterplans, detailed plans and regulatory development policies. Reliable information about the projected demand will allow municipal planning agencies to make sure that enough affordable land and real estate will be supplied in desired locations, and that the agricultural to urban land use conversion along with the destruction of natural resources is not over exploited. The module will also help cities foresee and prevent potential bubbles and sudden fluctuations in the land and real estate market. This requires an understanding of how the land and real estate market performs in the context of the broader capital market, and how the real estate space and asset markets are related. The module will help local governments of the pilot cities foresee the impact of shifts in other sectors of the economy on the real estate market, and conversely predict shifts in other sectors of the economy triggered by real estate and land market changes. The module will also inform the local finance and tax agencies of the current and projected state of the land and real estate market, and of investment/revenue opportunities, which can help improve the efficiency of their mortgage plans and taxation systems respectively. CPL staff, together with infrastructure and transportation departments, will explore value capture taxation systems as potential ways of unlocking financing for much needed infrastructure improvements. 6.3 SCOPE OF ACTIVITIES Addressing the objectives above, the activities for the Land and Real Estate Market Monitoring module in the two pilot cities are proposed as follows: 70 Phase 1: Compiling the land and real estate database (Months 1-6): In line with the core module, the CPL will first assemble all real-estate and land market related datasets that already exist in different government departments, integrate them and host them for cross-departmental viewing on an online map server. Existing datasets that are important for monitoring the land and real estate market include:  Cadaster: land parcel dataset containing ownership, occupancy, use – by sub-type e.g. single family, multifamily and mixed residential – coverage, FAR, and assessed value data. Additional attributes, such as size, frontage and distance to nearby amenities can be calculated for each land parcel by CPL staff.  Buildings: building footprints containing ownership, occupancy, use (by built area and sub-types), past sales transactions and assessed values. A reliable building dataset that distinguishes building types is needed for evaluating the total supply of different types of real estate products on the market. Overlaying the building dataset with the cadaster will also provide the total supply of land that is available for development within the currently urbanized extent of each city. Figure 10. Example Analysis: Distribution of Building Types in Singapore 71 Table 7. Example Dataset: Price Range of Flats Offered by Housing Development Board in Singapore (in Thousand SGD) 2 Rooms 3 Rooms 4 Rooms 5 Rooms Town Selling Selling Selling Selling price Selling Selling Selling Selling price price less price less price price less price price less AHG/SHG5 AHG/SHG AHG/SHG AHG/SHG Bukit - - 137 – 189 107 – 159 217 – 298 207 – 288 274 – 386 274 – 386 Panjang Choa Chu - - 146 – 172 116 – 142 229 - 284 219 – 274 295 – 364 295 – 364 Kang Punggol 85 - 111 25 - 51 150 – 242 120 – 212 257 – 390 247 – 380 335 – 484 335 – 484 Sembawang 92 - 116 32 - 56 158 – 191 128 – 161 255 – 310 245 – 300 - - Sengkang 83 - 112 23 - 52 134 – 220 104 – 190 255 – 370 215 – 360 283 – 456 283 – 456 Yishun - - 156 – 193 126 – 163 234 – 296 224 – 286 277 – 381 277 – 381 Source: Singapore Housing Development Board  Road Network: the road network is essential for performing accessibility measurements for each building and parcel. Location, or more accurately the accessibility of a location, is the main indicator of land and real estate value.  Public Transit networks: In addition to road-level accessibility, land values also depend on available transit options. All forms of public transit (e.g. bus, minibus, regional lines) can impact land and real estate values.  Points of Interest: accessibility to amenities and businesses is also known to impact land and real estate values. Proximity to commercial destinations and other desirable venues or establishments, such as parks, hospitals, or museums – can be measured on the available road and transit networks in different parts of the city.  Census and Household Survey Data: these datasets are essential for estimating the demand side of the market.  Rents and prices: Available sales and rental prices for different property types (e.g. housing, retail) and subtypes (e.g. 1-Bedroom Apartment) will be collected from the main brokerage firms in each city. If possible, then each observed transaction should also indicate how long the unit was on the market and illustrate other general characteristics of the larger building complex the unit is part of. This data may be available at address level resolution, at a zone or street-level resolution.  Upcoming developments: Information should also be collected about all real estate development projects that are currently under construction or otherwise planned to be completed. Approximate type and size of each development should be listed and an approximate date of delivery recorded. This will allow the CPL staff to account for 5AHG: The Additional Central Provident Fund/CPF Grant, given to eligible first-timer families who are applying to buy a 2-room or bigger flat who are able to meet the eligibility conditions. SHG: The Special CPF Housing Grant, given to eligible first-timer families who are applying to buy a 2- room, 3-room or 4-room flat in a non-mature estate and who are able to meet the eligibility conditions. 72 future supply additions in estimating the needs for land and real estate products in five-year and twenty-year master plans.  Defining Analytic Zones: The first step for conducting property market analysis is to divide the cities into value zones based on their location, accessibility, uses, assessed values, and morphological properties – e.g. plot size, frontage, floor area ratio. The type of land and buildings in each zone should be as homogenous as possible. The census data and household survey data will be associated to each zone.  Supply and Demand Estimation: The data collection efforts in Phase One will be concluded by developing a supply and demand estimation for different real estate products in each zone, providing a fundamental basis for both real estate analysis and spatial planning. The supply of land includes non-developed lands that are available for development, as well as lands that can be converted to other uses or densities – e.g. conversion of single family housing to multifamily housing. The analysis will yield an estimated supply and demand overview for different real estate products in different parts of the city based on assessed values. The estimation of the demand side of the market, however, will rely on census data and household surveys (e.g. household size, and household income), as well as financing options. The latter will require collecting data from local banks and mortgage brokers. Although the assessed values may be significantly lower than the real values, they can provide an indication of spatial shifts in the market. The results will be later compared to and synthesized with the land and property market surveys in Phase Two. Phase 2: Surveys (Months 6-12): In the second phase of the project, CPL will carry out two surveys with local real estate brokers to compile a database of observed real estate market transactions and to develop an understanding of the segmentation of households by housing market access.  Land and Property Market Assessment: CPL will carry out a land and property market assessment survey using the methodology outlined in Dowall’s 1995 Land Market Assessment (LMA) and a simplified update from 2010. The survey involves interviewing experienced land brokers in each city to determine the prices for prototypical land parcels in different parts of the city. These property values are expected to differ from official assessment estimations, which often undervalue properties for tax reasons. One additional improvement to Dowall’s original LMA strategy is to exploit more easily accessed satellite imagery in the categorization of housing stock. Bertaud (2008) outlines this approach and draws attention to the importance of incorporating considerations for transportation infrastructure and urban growth patterns in LMA. CPL staff will be trained to carry out the survey periodically in the future and to use the results of the surveys as the basis for evidence based policy recommendations in the land and housing sectors, and to also enable infrastructure projects to be developed and financed in a more integrated manner from the outset.  Housing Market Segmentation Study: CPL will perform a household survey that assesses the mechanisms through which people access housing in different income groups. The study is expected to yield important information about actual housing demand and supply for different unit types. Disaggregating housing demand into market segments 73 (based on income or on other criteria) is an important step in understanding how the housing market functions as a whole, and in identifying the distribution and trends of demand across segments. Different segments of the population access and combine the basic inputs into housing (Land, Finance, Materials and Labor) using a range of different methods. Analyzing these different variables and streams of supply, as well as the bottlenecks they face, is a crucial step in formulating more precise and targeted Government housing programs. These two survey activities will follow the approach outlined in the recent World Bank document, “Land and Property Market Assessment - Housing Market Segmentation Study: Existing Tools and Survey Strategy”. Phase 3: Land and Real Estate Analytics (Months 12-18): The analytics proposed below will form the basis for real estate and spatial planning in the pilot cities. In addition to being a platform for conducting land and real estate analytics, CPL will assist the pilot cities in incorporating the analysis into their real estate and spatial planning.  Accessibility and Land/Real Estate Value analysis: Along with the accessibility analysis in the core module, CPL will analyze the impact of accessibility to surrounding land uses and amenities on land and real estate values. This analysis will extend in the final phase to a full hedonic pricing model that takes into account all major determinants of land and real estate value.  Impact analysis – before-after comparison: CPL will evaluate the impact of key infrastructure investments on land prices – e.g. a new road or a public hospital – by comparing historic land value data before and after development. Controlling for other factors that can affect land values (e.g. city-wide shifts, inflation), the before and after comparisons offers a useful methodology for evaluating the multiplier effects of public infrastructure, which can be used for supporting investment decisions in the future.  Hedonic Pricing Model: The full hedonic land price model is an extension of two pervious analyses – impact analysis and accessibility analysis described above. When a sufficient amount of spatial information and land / real-estate market data have been collected, CPL will be able to develop spatial hedonic pricing models to analyze variations in land and real estate values. Initially, the analysis could focus on explaining the direction and magnitude of infrastructure and service amenities on land values. How do new roads, sanitation facilities, transit systems, plot sizes and demographic characteristics impact land values? How far in space do such effects reach (e.g. how far can a parcel be from a paved road to have a value impact)? Such analyses should become regular activities at the CPL, accompanying all significant public investment projects and planning initiatives. Hedonic land value analyses can also form a basis for potential value capture regulations in the future.  Projections: Using the hedonic model and examining the current trends in the land and real estate markets, CPL staff will develop evidence-based forecasts for near-term and long-term changes in land and real estate values that are likely to result from 74 foreseen developments. Additionally, using hedonic pricing models, CPL staff will work across municipal departments to investigate the financial feasibility of a pilot value capture taxation program around a planned infrastructure investment project. 6.4 RISKS AND MITIGATION The primary risk concerning the activities proposed above involve the reliability of the gathered land and real estate market data and the validity of the estimations that result from these data. In order to address this risk, we have proposed to collect the data from various sources, which will allow CPL staff to cross check the results. In Phase One, various datasets are collected from existing sources, including the assessed land and real estate values from DG Tax and BPN. In Phase Two, similar information is collected through personal surveys with experienced local real estate brokers. Even though the surveys can only cover limited parts of the city, consistent offsets and interpolations can be used to adjust all officially assessed data accordingly. 6.5 OUTPUTS The following outputs are expected from the land and real estate market module: A cadastral real-estate database. All spatial data gathered in the first phase of the project will be compiled into an online geodatabase, showing each land parcel with its associated buildings, occupants’ demographics, accessibility characteristics and valuation estimates. A land and property market assessment report. The report will present the findings on demand and supply for different real estate products (e.g. residential, commercial, industrial land) and provide the basis for evidence based policy recommendations in the land and real estate sectors. A housing segmentation study report. The report will outline the demand and supply for different types of housing units and outline discrepancies between availability and need. The segmentation study will enable policy makers to detect which demand categories (e.g. low income residents) are most burdened by inefficiencies in the market and point to solutions that can be used to address these inefficiencies. Impact Analysis Report. The report documents the observed real estate value impacts of selected infrastructure investment projects undertaken by the city. The exact choice of projects (e.g. road construction, bridge or a facility) will be made together with local planning agencies on a per-need basis. The results of the study are expected to inform what multiplier effects future investments could have and how the benefits are spatial distributed. Real Estate Financing Analysis. The study will outline existing options and conditions that are available for short term, medium term and long term real estate financing, outlining potential shortcomings and improvement opportunities for desirable financing options. 75 Hedonic Pricing Model Results. A report describing the controlled multivariate analysis results of land and real estate values in the respective cities. Hedonic price models explain variations in land and real estate values based on the spatial attributes and accessibility conditions of buildings and land parcels. These results can be used to estimate the likely market effects of future plans and infrastructure investments, forming the foundations of a sound real estate market policy. 6.6 TEAM In addition to full time staff members listed above, additional expertise required for providing consultation to this module will include: i. Urban Economist ii. Housing and Real Estate Planner iii. Market Analyst Specialist 6.7 TIMELINE This module will be carried out in three phases of six months each. 76 REFERENCES Bank Indonesia. Economic and Financial Data for Indonesia: http://www.bi.go.id/sdds/series/NA/index_NA.asp , Accessed on April 8, 2013 Bertaud, A. 2008. Spatial Tools to Analyze the Impact of Land Markets on Affordability and Urban Spatial Structures. Presentation at the World Bank, Washington, DC, February 28th. Dowall, D. 2010. Literature Review and Proposed Methodological Approach, Land Markets in Latin American and Caribbean Cities. Inter-American Development Bank: Washington, DC. Dowall, D.E., 1995. The Land Market Assessment: A New Tool for Urban Management. Washington, D.C.: Published for the Urban Management Programme by the World Bank. Suraji, A, Pribadi SK., Ismono, (2012). The Indonesia Construction Sector. The Proceeding of the Asia Construct Conference 18th, Singapore The World Bank, 2005. The Wealth of Nations Dataset: http://data.worldbank.org/sites/default/files/total_and_per_capita_wealth_of_natio ns.xls , Accessed on April 8, 2013 The World Bank, 2013. Land and Property Market Assessment - Housing Market Segmentation Study: Existing Tools and Survey Strategy 77 78 ANNEX 1: DEMONSTRATION REPORT OF SPATIAL GROWTH ANALYTICS MODULE ix x DEMONSTRATION REPORT SPATIAL GROWTH ANALYTICS TECHNICAL SUPPORT FACILITY FOR NATIONAL URBAN DEVELOPMENT PROGRAM IN INDONESIA DEMONSTRATION OF ANALYTICS SPATIAL GROWTH ANALYTICS Contact Authors: Andres Sevtsuk SUTD City Form Lab PI, City Form Lab 20 Dover Drive Singapore University of Technology and Design 138682 Singapore Reza Amindarbari +65-63036600 Researcher, City Form Lab cityformlab@mit.edu Singapore University of Technology and Design Date: July 14, 2013 Collaborators: Taimur Samad Senior Urban Economist World Bank-Indonesia Wilmar Salim University of Bandung Chandan Deuskar Urban Development Analyst World Bank Renata Simatupang Economist World Bank-Indonesia 2 SECTIONS 1 INTRODUCTION 2 GEOSPATIAL DATA 3 SPATIAL GROWTH AND CHANGE ANALYTICS 4 PLANNING DECISIONS SUPPORT 3 INTRODUCTION 4 1. INTRODUCTION A key engagement of the World Bank under the National Urban Development Program (P3N) is technical and institutional capacity building for supporting evidence-based spatial and development planning in Indonesia. This activity will take place through new permanent technical assistance facilities – City Planning Labs (CPLs) – that are proposed within Bappeda in four participating cities – Denpasar, Surabaya, Palembang and Balikpapan. Proposed CPL teams, supervised by a director, include civil servants from Bappeda, Dinas Tata Kota and other related city departments, and technical staff with backgrounds in urban planning, geographic information systems (GIS), and spatial analysis hired from outside the local government. In order to build the essential capabilities at CPLs, teams of international World Bank consultants will collaborate closely with CPLs in the first eighteen months of the project. CPLs’ primary goal is to collect geospatial data, and to conduct analyses that can inform spatial planning decisions in the abovementioned cities. CPLs’ analytic activities will be structured around a core module and four supplementary modules. The core module aims to establish the CPL facilities, institutional structures and data platforms, and to train CPL staff to implement basic urban growth and change analysis in each respective city. The optional modules include: (1) city economic competitiveness, (2) slum analytics and management systems, (3) climate and risk resilience planning systems, (4) and monitoring land and real estate markets. A separate World Bank concept note describes the analytic activities proposed for each module. The purpose of the present report is to Figure 1. Palembang is a city of 1.7M inhabitants in demonstrate the implementation and specification of the analyses proposed in Sumatra. Due to the presence of frequent earthquakes the core urban growth analysis module. The issues and opportunities that and the lack of investment in commercial real estate, the city center remains horizontally dense but Indonesian cities in general, and the four pilot cities in particular, face due to vertically low rise. rapid urban growth are diverse and locally specific. The list of analytic activities needed to understand and address all of them with appropriate planning tools and policies is too exhaustive to be captured here. This report focuses on a few basic, but useful spatial analysis techniques that are proposed as part of the core CPL module in each city. 5 GEOSPATIAL DATA The report first discusses essential data requirements for the proposed analytics and the needed geospatial data platform for sharing geographic information among government agencies, stakeholders and the general public. The first section of the report focuses on data preparation, appropriate units of analysis, required attributes for each dataset, and potential sources for the corresponding data in Indonesia. The second section concentrates on the analytic activities. The analytic activities we discuss and demonstrate include spatial growth and change analysis, spatial accessibility analysis, impact analysis, and spatial-statistical models. Technical implementation details and required geo-processing tools are discussed through a series of examples using data from other cities and countries. The third and final part of the report focuses on potential planning applications of the proposed analysis techniques. Developing an evidence-based planning decision support system that relies on empirical and up-to-date data and utilizes powerful spatial analysis tools could make a significant contribution towards turning the rapid urbanization in Indonesian cities into a vast opportunity for economic growth, equitable resource distribution and access to human development opportunities. The data and analysis techniques described in this report are, however, not only applicable to Indonesian cities – they could also benefit a number of other rapidly urbanizing countries in South East Asia and beyond. The City Planning Labs project in Indonesia will test the implementation of geospatial data systems and analysis techniques in four medium-scale cities, which will offer a unique opportunity to learn and improve from the experience. Another report will be compiled at the end of a 12 or 18 month engagement with the CPLs in Indonesia, describing how the implementation unrolled and what could be done better next time. 6 2.1. GEOSPATIAL DATA The first responsibility of CPLs is to collect and disseminate geospatial data required for supporting planning decisions in cities. A large amount of data that CPLs require for their analytical activities already exists in different agencies and departments at the local and national level in Indonesia. We provide a table at the end of this section showing all the datasets that are recommended to be collected at each CPL, where we also indicate whether and where we have witnessed the availability of such data. Not all of these data will be available in each of the four cities. During the first year, the table can be used as a wish-list for geospatial data for the core CPL module. CPLs would benefit from establishing a sustainable long-term collaboration with agencies who generate or harvest geospatial data in Indonesia, in order to have access to the most up-to-date information and to also share the data they compile with other agencies. Much of the existing data, however, is not digitized or attributed for GIS. Construction and change-of-use permits, land use maps, cadastral records, Masterplans and survey records can be paper- based, making referencing and data sharing difficult. CAD drawings, where they exist, often come without a spatial reference for geographic location. Increasingly, Indonesian cities have started collecting these data digitally and some have access to remote-sensing information, such as medium or high- resolution aerial or satellite imagery, LiDAR scans, or satellite stereo imagery. A number of cities have gone through initial efforts to digitize building footprints, street networks and land-use classification zones, albeit with Figure 2. High-resolution satellite image, London, Oxford Circus. relatively limited attribute information describing the characteristics of the geometric elements. In the first project phase, CPL staff will need to collect, organize and standardize a broad range of existing datasets available in each city. As existing data are often partial, CPLs will also need to obtain some data, such as remotely sensed topographic data through specialized service providers. Ground surveys should be prepared for building use and condition maps or additional socio-economic data gathering. CPLs should develop routine processes for implementing annual ground surveys for accuracy 7 verification and for completing or updating existing datasets. The World Bank consultants in this project will propose standards and guidelines for such data collection during the first implementation phase. 2.1.1. Units of Analysis All geospatial data is associated with geometric spatial units (e.g. address points, building or parcel polygons, street centerlines), which are geographically referenced. Each individual spatial unit contains certain attributes about the environmental feature it symbolizes (e.g. building floor area, census block population, parcel value etc.). The spatial units for available data typically depend on both the geometric elements observed in the built environment (streets, buildings, parcels), the ease with which the data can be physically surveyed on ground (census blocks can be typically surveyed in less then a day), as well as the graphic representation constraints (e.g. sloping land is usually symbolized with horizontal elevation lines that are not witnessed in reality). For privacy and security reasons, or technical limitations, agencies that collect data do not always release spatial information at the same spatial resolution or with the same units of analysis that data was originally assembled. Statistic Indonesia (BPS), for instance, collects all census data at a census block level (often comparable to a city block), but only releases data at the village, sub-district and higher levels. Original spatial units of data, as disseminated by different data collection agencies, do not necessary match the needs for urban growth analyses proposed in the core module. An important part of CPL’s activities will be dedicated to compiling and deriving new datasets, using geo-processing techniques for aggregating or disaggregating raw data to desired spatial units. Figure 3. Different spatial units of analysis and their respective For building level analysis, for example, land use data or economic census data attributes should be disaggregated from original tract levels to individual building levels (discussed below). To classify spatial data in this report we focus on the spatial unit rather than the content of attribute fields. Datasets with different spatial units may have 8 similar attributes – parcels, buildings, census blocks and villages can all contain information about the residential population, the built area, or the number of business establishments they accommodate (Figure 3). The following list of data is seen as most essential for CPLs to collect, in order to carry out the analytics proposed in the core module. For each dataset, desired attributes, required raw data, and potential data sources are explained where possible. 2.1.1.1. Buildings and address points Individual buildings are among the finest spatial unit of analysis commonly used in cities. A substantial part of human activities takes place in buildings; the people that buildings accommodate and the activities they engage in, generate the origins and destinations of most pedestrian and vehicular movement on city streets. Buildings are typically represented as polygon features (building footprints). Polygon features illustrate the realistic geometry of the actual building footprints and allow valuable post-processing analysis that may not be Figure 4. Building footprints and their centroids, Harvard included in the original attributes (building area or perimeter calculations, 3d Square, Cambridge MA. height extrusions, volume calculations). However, buildings can also be represented as point features, placed at either the centroid of the actual footprint or at one or more exterior entrance locations of the building. Point representation of buildings can be useful for analyzing accessibilities between buildings on the network of city streets that requires discrete locations (Figure 4). Polygons can easily be converted to points, but not vice versa. Centroid points, may fail however, to capture the accurate relationship between a building and its street(s) or the visual sightlines available between buildings. It is therefore preferable to represent buildings with realistic polygons. Address points, discussed hereafter, can be placed at entrances to accurately capture the relationship between buildings and streets. The geometry of buildings (footprints or 3D volumes) are typically obtained from satellite imagery or aerial scans (e.g. LiDAR, stereo imagery). A high-resolution geo- referenced satellite image can be used as a base for drawing building 9 polygons with vector-based lines in drafting software like AutoCAD, DraftSight or Rhinoceros. An effort should be made to distinguish two autonomously owned or used buildings into two separate polygons whenever possible. This can be challenging to do if buildings on satellite imagery appear to share walls or are otherwise densely spaced, but a careful distinction of individual buildings according to addresses can greatly benefit later analysis. A physical ground check is usually required to cover areas that are poorly visible in the satellite image (e.g. cloudy or obstructed by trees) and to ensure that the drawings match reality. Generating address point features typically requires a ground survey, using GPS tools for recording the spatial position of each observed address location on the ground (Figure 5). If accurate street network data is available (e.g. from Navteq, Tom Tom) and the total number of addresses on each street known, then address point locations can also be interpolated in GIS (with some spatial error). Tools for performing such addressing are offered in ESRI’s ArcGIS. Managing standardized addressing data is usually a national level activity; the system should be consistent throughout the country. Efforts towards a national address database appear to be under way at the Indonesian Figure 5. Parcels and address points, Los Angeles. Geospatial Information Agency (BIG). Desired building and address point attributes for proposed CPL analytics include: Volume, total floor area and footprint area Building volume and floor area indicate the amount of available space for human activities. Although floor area is a more accurate indicator of space for living, working, studying, it is not always easily obtained and requires detailed surveys or registry records. Building volume, however, can be directly computed from the basic geometry information (footprint area and building height). Approximate building floor area can be found by diving the building height by a typical floor height (e.g. 3 meters) and multiplying it by the area of the footprint. Typical floor-to-floor height for residential buildings is 2.8 meters 10 for office building 3.5 meters, for institutional buildings 4 meters and for industrial buildings 5 meters. Address Street addressing is the practice of assigning unique names to spatial features, typically buildings, plots and business locations, using a consistent hierarchical system. A street addressing system contains several components that are consistent across all individual units. The most typical components of an address are street segment name, street type (road, drive, highway etc.) plot or building number, unit number (if applicable), and an area ID such as ZIP or postal code. The purpose of using such a hierarchical naming system is to allow users to locate an address even when they do not have access to GIS data. Street addressing is vital for locating facilities and infrastructure (businesses, hospitals, schools etc.), and delivery services (e.g. postal or emergency services) in an urban setting. Developing a standard street addressing system as a common platform among all public and private agencies is also crucial for urban information management. It allows for systematically storing surveyed data at the highest possible resolution (household or business level). It also allows for generating a great deal of spatial data from registry recodes that contain address attributes and keep them continuously updated with little Figure 7. RT-RW is currently the smallest addressing unit in Indonesia. Each RT contains several households. Source:John effort and cost. Taylor. Developing a street addressing systems is often a national level effort, but conducted at a local level, where CPLs can play a significant role. There are a number of examples of such efforts in developing countries, including in a series of Sub-Saharan countries in Africa, in collaboration with the World Bank (see Farvacque-Vitkovic et al 2005). While in the long-run, public and private agencies may update their address information using a standard addressing system, CPLs can also help assemble data from registry records using available street information. This requires bringing presently available addresses into a uniform format. ArcGIS geocoding tools, and Python or VB string functions allow for matching the 11 … existing addresses that come in different formats: “20 Dover Dr.” to “20 Dover 33 Manufacturing Drive” or “20 dover drive.” 42 Wholesale Trade 44 Retail Trade Businesses establishments and Employment 441 Motor Vehicle and Part Dealers 4411 Automobile Dealers Businesses activities often take place in buildings; building, thus, offer a natural …. unit of representation for the distribution of businesses in a city. Business 4412 Other Motor Vehicle Dealers establishment and employment data at an individual building level should 44121 Recreational Vehicle Dealers ideally include the total number of businesses and employees classified by 441210 Recreational Vehicle Dealers 44122 Motorcycles, Boat, and Other Motor Vehicle Dealers different industry categories (e.g. retail establishments or services), as shown 441222 Boat Dealers in figure 7. Raw business location data typically comes at address or 441228 Motorcycle, ATV and all other Motor Dealers geographic coordinate (point) level, as explained further below. Aggregating 45 Retail Trade such point data to building footprints, however, provides convenient units of 48 Transportation and Warehousing analyses and allows buildings to be used as inputs in multiple types of spatial … analyses. Business establishment and employment data are usually available in two Figure 7: NAICS business classification; business different forms. They may be available in aggregate census tract level (or establishments should be grouped and categorized based on other statistical boundaries), indicating the total number of businesses within standard systems such as North American Indusial Classification Systems (NAICS), or Standard Industrial each aggregated area. Storing detailed business classification information is Classification Systems (SIC). Depending on the analytical task not typically feasible in this case. Second, every business can be shown as an that is conducted, the classification depth would vary; e.g. in individual unit, represented by points with attribute information (see NAICS the six-digit level is the most detailed classification, 2.1.1.3.business locations). When business establishments are shown at the however, the first two digits are enough to distinguish retail trade business establishments. building level, the business attributes should be summarized, showing the sum total of all establishments that occupy each building. BPS collects data on medium and large business establishments with more than 20 employees, which constitutes a small percentage of all businesses in Indonesia. BPS also surveys small samples of all business every year, which cannot be disaggregated lower than city scale. CPLs may need to conduct ground surveys to collect more comprehensive data on business establishments. 12 Numbers of residents / households Population is the key determinant of demand for a city’s resources. Detailed data on spatial distribution of population – and demographic sub-groups – allows for efficient estimates for a city’s resource needs. Population and demographic data are not commonly disseminated at the building level, but aggregated to census block or tract levels. In Indonesia, BPS conducts household surveys and provides census data at the village (Desa or Kelurahan) level. If individual buildings’ type (e.g. residential, commercial etc.), and floor area or volume are known, then population values from higher-level spatial units can be disaggregated to the building level with reasonable accuracy. The total number of residents in a census block can thereby be allocated between residential structures, weighing the allocations by the size of each building (Figure 8). Building type and subtype Building type describes the types of activities that take place in the building (Figure 9). Reliable assessment of the real estate market (asset and use) is not feasible without building type and subtype information. Building types or subtypes do not necessary share the same market. Commercial and residential spaces belong to separate markets and separate demand segments. To determine the supply side of each market, it is essential to keep track of building stock by type. Figure 8: Disaggregating population information from census tract level to buildings. The population of the census tract is Building subtypes (e.g. housing) can also have separate markets (Figure 10). allocated only among buildings that contain residential uses, The demand for large landed houses is composed of a different socio- weighing the allocation by the building volumes. Naturally some spatial error is generated in the process, but storing economic group of buyers and renters than the demand for small studios or population estimates at an individual building level is useful public housing units. for a number of high-resolution analyses. The two main sources for building type data are zoning maps, which usually do not contain subtype information, and ground surveys. Developing and maintaining an accurate building type database can be very labor intensive, but the pay-offs are also high since building level data allows for many useful analyses about city’s real estate market. 13 Once a reliable building type and subtype database is assembled through ground surveys, it is practical to maintain and update it via building permit, modification permit, demolition permit and change-of-use permit databases. If a new permit is issued, the finished building occupancy permit can automatically signal to the building database managers that a new building has been added to the stock. The building type database can then verify the data and add the new building to the repository. A similar procedure can follow other types of building permits. 2.1.1.2. Parcels Commercial Buildings (10,912) Institutional Buildings (12,401) Parcel geometry records land ownership borders. The geometry of parcel Industrial Buildings (7,048) borders is often provided by national land agencies (e.g. BPN in Indonesia). In Residential Buildings (58,566) Indonesia, Tax Directorate General also prepares parcel polygon datasets, for its own land value assessment purposes. The two parcel databases currently Figure 9: Building stock by type in Singapore; markets for different 1 building types are to a large extent independent of each other. remain separate, but BPN appears to be working on a joint database . Attributes that parcels should contain include: Assessed value and transaction history Parcel is an intuitive unit for land market related analyses, as transactions and value assessments are conducted at the parcel level. In Indonesia, Tax Directorate General keeps track of land transactions, and assesses land values. Zoning Parcel is the appropriate unit for containing zoning attributes, such as land use, plot ratio, height limit and setbacks, as building permits are issued for specific parcels. Zoning information presented in masterplans and detailed plans, which are prepared by Bappeda in each city in Indonesia, typically specify zoning regulations for each parcel. Public Housing (916,842 units) Condominiums (200,000 units) Landed Houses (70, 000 units) 1 Source: personal communication with BPN Figure 10: Housing segmentation (subtypes of residential building categories) in Singapore. 14 Building properties Most physical building structure or land improvement data can be also aggregated to the parcel level: e.g. total floor area, total building volume, address, total number of businesses. Frontage In addition to typical geometrical properties (perimeter and area), it is useful for parcel datasets to contain the length of street frontage: the length of parcel perimeter that is directly connected to a street (Figure 11). Frontage is an important determinant for land value, and essential for developing hedonic pricing models for land. Parcel type Parcels can also optionally be classified based on the number of streets that a parcel is directly connected to (Figure 12). A “middle parcel” has access to one Figure 11: Parcel street frontage street, but a “corner parcel” can have access to 2, 3 or 4 streets Similar to frontage, parcel type, as defined above, is an important determinant of land value, and useful for developing hedonic pricing models. 2.1.1.3. Business locations As mentioned above, raw business establishments and employment data is best stored at an individual business establishment level, represented as points (Figure 13). Representing separate business establishments with separate point features is the most robust and useful way of storing the business establishments’ data. Points can always be aggregated or joined to other larger units (e.g. buildings or parcels) if needed. The attribute information of business locations should typically indicate: - The legal name of the business Figure 12. Parcel type, indicating the number of streets that a - The name of a parent company (if applicable) parcel has direct access to. 15 - Detailed industry classification code (e.g. NAICS) at as detailed level as available (e.g. 6 digits), see Figure 7. - Year established at the present location - Number of employees - Longitude and latitude coordinates - Address - ZIP code - Town, Region CPLs may conduct ground surveys to collect business location data as BPS collects data only on medium and large business establishments with more than 20 employees. In the longer run, accurate business location data can be collected from DG Tax records that should account for all business locations for income tax and sales tax reasons. A good tax system can produce ample spatial data annually, at almost no extra cost. Figure 13. Pedestrian Network and business locations, Bugis, 2.1.1.4. Transportation network Singapore. Source: City Form Lab. Each business establishment point contains a set of attributes describing the business. The movement of goods and people in cities takes place through three layers of networks: vehicular roads, pedestrian paths, and public transit networks. The latter is often used together with the pedestrian network, and forms a multi-modal network. Analyses that help us understand how resources and facilities are accessible to users through the mentioned networks require data. Most cities collect and prepare road centerline datasets (Figure 14) and sometimes public transit network datasets, but often overlook the pedestrian network. Road centerlines are the most commonly used GIS data for accessibility analyses, not only for vehicular movement, but also for pedestrian movement. A large part of pedestrian flow takes place along streets. However, street centerlines do not capture pedestrian routes that are not along roads (e.g. through unoccupied parcels in informal settlements). Purely vehicular routes, such as toll roads, do not have sidewalks. It is, thus, recommended that the CPLs prepare geospatial datasets of street centerlines, public transit lines, as well as sidewalks and other pedestrian paths (Figure 13). A great deal of a 16 city’s circulation in Indonesia occurs on foot. Beyond accessibility analyses, sidewalk databases will be also useful for sidewalk improvement plans. Street network centerline data may be available in Indonesian cities via third- party data providers, such as NAVTEQ. Potential attributes for road network datasets include: - Width or number of lanes - Type (e.g. paved or unpaved) - Street name - Road classification - Traffic directionality Desired attributes for pedestrian network datasets include: - Width - Type (e.g. indoor, outdoor, outdoor but sheltered) Potential attributes for public transport network datasets include: - List of buses using the segment - Average time consumed on the segment Figure 14: Street centerlines, Los Angeles, CA, and their - Frequency (of bus or train on the segment) attributes: drive direction, road type, road surface type, and road - Start and end stations of the segment segment length. 2.1.1.5. Administrative boundaries Administrative boundaries are abstract extents that define the spatial authority of governance of communities in a hierarchical order from national level to smallest groupings in neighborhoods e.g. RT or RW in Indonesian cities). Administrative boundaries are common spatial units for storing socio- economic data. 17 Kota: BPS provides census data at the village level, as well as higher aggregation City levels such as city, district, or province. Micro data in Indonesian cities is typically collected by the head of village at the RT level (Figure 15). The CPLs should digitize and distribute the following set of administrative boundaries: - Regency (Kabupaten) or City (Kota) - Sub-district (Kecamatan) - Neighborhood/village (Desa or Kelurahan) - Block (RT/RW) Kecamatan: Each administrative polygon should carry a unique identifier name. Lower level District polygons should also indicate the names or IDs of the higher level polygons they are part of. The polygon shapefiles can be likely obtained from the local BPS office up to the neighborhood level. Mapping the RT-RW boundaries could require collaboration with the heads of villages. Such mapping has been previously implemented in Solo. 2.1.1.6. Other spatial data The datasets discussed above constitutes only the most essential data that can be used in the core module of CPLs. Much of the data is also directly Kelurahan or Desa: Village useful for other optional CPL modules. The list of spatial data that cities collect or already have can be very exhaustive. Many of those data can be associated to one or several spatial units mentioned above; e.g. energy consumption can be associated to buildings and parcels, crime rates to any administrative boundaries. However, there are some other data that require their own spatial units: for example, data on water infrastructure or Wi-Fi hotspot. A list of spatial data that CPLs are recommended to collect is provided in the table below. RT-RW: Smallest addressing unit Figure 15: Administrative subdivisions in Indonesia. Source: John Taylor. 18 blank Not available but desired for proposed analytics ! ! ! ! ! ! ! ! ! Available ! ! ! " Available in some of the participating cities ! ! ! ! ! ! ! !   ! ! ! ! ! ! ! ! ! ! ! ! ! DG Other DG !! INDONESIA URBAN DATA BPN BPS Bappeda Spatial Potential BIG Notes Tax Planning Sources !! DATA 1! IM AGERY !! High-resolution satellite image !! !! " " !! " !! Aerial photography !! !! " " !! " !! Digital Elevation Model (DEM) of urban topography !! !! !! !! !! World !! Overall urban extent (built-up area in the metro area) !! !! !! !! !! " Bank 2! PLA NNING REGUL ATIONS !! !! !! !! !! !! Zoning plans !! !! ! ! !! !! Floor area ratios (gross plot ratios), as specified in regulations !! !! ! ! !! !! Land use as shown in city master plan !! !! ! ! !! !! Building heights limit, as specified in regulations !! !! ! ! !! 19 3! STA TISTIC AL BOUNDARIES !! !! !! !! !! !! Provincial administrative boundaries !! ! !! !! !! !! Municipal administrative boundaries !! ! !! !! !! !! District administrative boundaries !! ! !! !! !! !! Sub-district administrative boundaries !! ! !! !! !! !! Zip code areas !! !! !! !! !! !! Census tracts (Village) !! ! !! !! !! 4! DEM OGRA PHIC CHARAC TERISTIC S !! !! !! !! !! BPS data are aggregated at village !! Residential population (census) !! !! !! !! ! level BPS data are aggregated at village !! Residential population by sex !! !! !! !! ! level BPS data are aggregated at village !! Residential population by age group !! !! !! !! ! level BPS data are aggregated at village !! Residential population by residence type !! !! !! !! ! level BPS data are aggregated at village !! Household surveys: household income !! !! !! !! ! level BPS data are aggregated at village !! Household surveys: family size !! !! !! !! ! level 20 5! URBAN FORM !! !! !! !! !! !! Observed floor area ratios !! !! !! !! !! !! Observed land use !! !!     !! World !! Officially recognized informal settlements !! !! " " !! Bank !! Building floor areas !! !! "! !! !! !! City blocks !! !! " " !! !! Parcel boundaries, ownership ! !! " " ! !! Building footprints !! !! " " !! !! Building heights !! !! !! !! !! !! Building ages !! !! !! !! !! !! Building uses (e.g. residential, commercial, etc.) !! !!     !! Building types (e.g. walk-up, condo, row-house, kampong, !! informal)* !! !! !! !! !! !! Building addresses/ZIP Codes !! !! !! !! !! 21 6! INF RA STRUCT URE !! !! !! !! !! Transportation infrastructure: roads by category, # lanes, !! !! !! " " !! direction, setbacks, centerlines, intersections, traffic lights, toll gates. !! Transportation infrastructure: rail * !! !! " " !! !! Transportation infrastructure: bus lines and stations* !! !! " " !! !! Transportation infrastructure: bicycle routes !! !! " " !! !! Transportation infrastructure: pedestrian sidewalks, crossings* !! !! " " !! !! Service infrastructure : water, sewage, and drainage* !! !! " " !! !! Service infrastructure :electricity lines, substations !! !! " " !! !! Drinking water supply networks* !! !! " " !! !! Potable water source locations (e.g. wells)* !! !! " " !! 7! ECO NOMIC C HARACT ERISTICS !! !! !! !! !! !! Municipal/district expenditure by economic categories* !! " !! !! !! !! Province expenditure by economic categories* !! " !! !! !! !! Municipal/district revenue by sources* !! " !! !! !! !! Province revenue by sources* !! " !! !! !! 22 8! EST ABL ISHM ENTS !! !! !! !! !! !! Firm distribution (Individual establishment locations)* !! !! !! !! !! !! Job distribution (jobs per area/ type)* !! !! !! !! !! !! Points of interest (museums, institutions)* !! !! " " !! !! Public institutions (hospitals, police stations, libraries etc.)* !! !! " " !! !! Public institutions: schools (with levels, and no. students)* !! !! " " !! !! Public institutions: hospitals (with specialties, and capacities)* !! !! " " !! !! Public institutions: others (detailed)* !! !! " " !! 9! NAT URAL HAZARD !! !! !! !! !! World !! Seismic hazard zones !! !!     !! Bank World !! Flood zones !! !! " " !! Bank World !! Land topography (points / topolines) !! !! " " !! Bank 23 10! LAN D AND REAL EST ATE M ARKET !! !! !! !! !! !! Land cadaster: parcels (tax, transactions, etc.)* ! !! !! !! ! Developers DG Tax estimates do not often indicate !! Land prices* !! !! !!   ! & Brokers real market value Developers DG Tax estimates do not often indicate !! Housing prices* !! !! !!   ! & Brokers real market value Developers DG Tax estimates do not often indicate !! Commercial real estate prices* !! !! !! !! ! & Brokers real market value Developers DG Tax estimates do not often indicate !! Housing Rents* !! !! !! !! ! & Brokers real market value Developers DG Tax estimates do not often indicate !! Commercial rents* !! !! !! !! ! & Brokers real market value Developers DG Tax estimates do not often indicate !! Land rents* !! !! !! !! ! & Brokers real market value Developers !! Housing tenure (vacant, owner-occupied, rental occupied) !! !! !! !!   & Brokers Developers !! Residential unit sizes/ no. of rooms !! !! !! !! !! & Brokers !! Issued building permits within the past x period* !! !! !! ! !! !! Issued demolition permits within the past x period* !! !! !! ! !! Projects currently in construction Developers !! (residential/commercial/industrial/office/other) !! !! !! ! !! & Brokers Sales Prices / Rate of Sales at each new development currently Developers !! on the market !! !! !! !! !! & Brokers List of all permits required for new developments by land use !! type and typical durations. !! !! !! ! !! 24 2.2.DATA PLATFORM To fulfill their primary goal of assembling, maintaining and distributing large geospatial databases, the City Planning Labs need a digital geospatial data platform that satisfies five fundamental requirements. The platform should: ! Allow data to be efficiently and conveniently stored and managed ! Allow data to be shared across different city departments or with members of the public over Internet browsers ! Enable all data management operations to be performed from a local networked computer ! Enable users to download data layers that they have security clearance to access ! Enable the end-users to interact with the datasets on a web-browser, by querying their attributes, overlaying different data layers, using simple base- maps to situate the information, and overlaying personal information layers on published maps. Figure 16. New York City OpenData. More than 1500 spatial datasets are public available through this geospatial platform, The capacity to operate basic spatial functions (e.g. spatial search, some updated daily. Software: OpenGeo Suite Enterprise measurement or proximity search, overlay function etc.) would be desirable Edition. additional functions for the end users, though not a first-order priority. There is a considerable list of open source and proprietary GIS server technologies available for managing spatial data. ArcGIS Server, ArcGIS Online, and MapInfo Spatial Server are among the most commonly used proprietary options. On the other hand, GeoServer, OpenGeo Suite or GeoNode are some examples of widely used open source web-based geospatial content management platforms. The World Bank’s Platform for Urban Management and Analysis (PUMA), currently under development, is also a potential open source option for the City Planning Labs. Apart from the initial cost difference, fast setup time, and off-the-shelf availability of functions 25 are the main advantages of proprietary platforms. On the other hand, open source platforms allow for a higher level of customization. Some Indonesian cities and agencies have already developed their own platforms (Figure 17) using both proprietary and open source options. For example for their permitting systems, Surabaya has utilized ArcGIS Online, while Balikpapan has developed its open source platform. At the national level, BIG, has developed an integrated open-source/proprietary (ArcGIS online) data-sharing platform. In their decision for data platform options, CPLs’ should consider platforms the respective cities are currently using. In addition to geospatial content management software, each CPL requires server space for storing its geospatial content. CPLs can host their geospatial data on local servers or use networked enterprise storage systems, such as cloud storage on Amazon Web Services (AWS). While storing data on a cloud can be more expensive than storage on a local server, outsourcing could also Figure 17. Solo Kita Kota, the interactive web-based map for the city of Surakarta (Solo). Software: Google Maps API. provide other benefits. Cloud storage systems are typically more stable than local hosts, particularly resilient to power shutdowns and human errors, keeping data constantly online. Using cloud storage also shifts the maintenance burden from CPLs to the service provider. Professional data storage systems offer qualified technical assistants with service contracts. Geospatial content management software can be used to set up different levels of security access to different data users. It is possible that CPLs use a different platforms in short and long run. ArcGIS online in combination with Amazon cloud storage, for instance, could be an option for short term, as it is easy and fast to set up with low initial cost, while CPLs develop their customized open-source permanent platforms. Figure 18. MIT GeoWeb data sharing and visualization platform. 26 Figure 19. One of the data platforms of the City of Cambridge – Cambridge City Viewer – through which a large amount of geospatial data including buildings, parcels, paved surfaces, sidewalks, street centerlines, trees, and infrastructure systems is published. Figure 20. Census block level demographic data of Cambridge, Boston and Somerville, from Census Bureau platform. 27 SPATIAL GROWTH AND CHANGE ANALYTICS 28 3.1. SPATIAL GROWTH AND CHANGE Monitoring trends in spatial data is a very useful analytical technique that can provide valuable information for planners. Effective planning decisions are built upon short and long-term projections of the spatial distribution of resources and demand. These projections are often made by extrapolating current trends in spatial data. Forecasting the spatial distribution of resources (e.g. housing, jobs, agricultural land) and demand (represented, for instance, by population and income) relies on an understanding of current and past trends. While projected spatial values are crucial inputs for more complex statistical models, a simple comparison of projected values with benchmarks from other cities or existing conditions can provide a ground for evidence-based decision-making. Based on projected values, planners can decide whether intervening actions should be taken to strengthen, slow down or reverse current trends. Monitoring the shifts in demographic data, for instance, can reveal a significant Figure 21. Indonesian cities are facing rapid growth with an increase in the number of households with young children in peripheral areas average annual urbanization rate of 4.2% between 1993 and of a city, which can increase the demand for schools, hospitals, or food 2007. resources. Evidence of such a growing demand allows planners to decide if new facilities are needed to support this trend or if policies are required slow it down. Understanding the rate of population growth should form the basis for land-use planning. Although monitoring trends in spatial data does not explain their underlying causes, and thus cannot suggest what interventions will affect them, it can at least frame the issues that planners should focus on. 29 3.1.1. Mapping Change in Spatial Values A convenient way of monitoring, storing, and representing change in spatial properties is to include change values – either absolute change or its rate – in the attributes of each spatial feature (Figure 26). For example, census blocks can contain attributes that represent the change in their building stock, particular demographic group, employment, land use coverage, energy consumption or per capita area of green space in a given period of time. 3.1.2. Spatiotemporal data If every spatial feature contains start and end time data, we can visualize snapshots of different points of time, using only one dataset (Figure 22). Spatiotemporal databases are useful when spatial units themselves change over time, not merely the value of their attributes. Changes in building stock or business establishments can be best stored by spatiotemporal databases, as the spatial features change over time – some buildings are demolished and some are added to the building stock over time. 3.1.3. The expansion of urban extent The most fundamental aspect of growth and change in cities is how much and where urban areas are emerging. The expansion in the boundary of cities represents a shift in land use coverage. The main source of land for a city’s growth is usually agricultural or forest land. Given the significant role of agriculture sector in Indonesia’s economy, annual loss of agricultural land can significantly reduce its cities’ food security. Figure 22. Spatiotemporal datasets. Each feature have start and end time, allowing for monitoring change in spatial features: for example in their size, location or existence. 30 Overlaying snapshots of the city’s extent over time can capture trends in 2 physical expansion. A future extent can be projected based on the past . However, distinguishing the urban land coverage from non-urban land extent of a metropolitan area is not a trivial task, and depends on the definition of “urban land”. Enhancements in remote-sensing technologies and availability of high-resolution satellite imagery have made accurate geo-referenced maps of urban extent available for many cities. Extent boundaries for most medium to large cities in Asia can be obtained upon request from Annemarie Schneider at the University of Wisconsin. By overlaying different snapshots of urban extent over time can not only capture the growth rate, but also its character. We can distinguish between leapfrog growth, edge growth or infill developments just by overlaying the extent maps of different times (Figure 23). Projecting the future extent of a city also requires information about buildable area around it. Geographical constraints such as water bodies, steep lands, or Figure 23: Jakarta 2000-2010: urban expansion, protected forests pose a barrier to growth and should be accounted for in categorized by expansion type (from 1,158km2 to 1,520km2) growth projections. The availability of space not only affects the rate of possible growth but also its character. Cities that are constrained by Edge growth 262 Sq.Km Leapfrog growth 73 Sq.Km geographic features, such as water bodies or steeply sloped land, grow very Infill growth 24 Sq.Km differently from those with no barriers around them. The former, for instance, leave no room for leapfrog development, set serious limits on sprawl, and tend to develop at higher densities (e.g. Hong Kong). The latter allow for lower density development and fragmented growth (e.g. Los Angeles). Note that The growth and change mapping we have discussed so far, examined trends in a single variable over time. Analytical techniques that are presented in the - following, Accessibility Analysis, Impact analysis, and spatial statistical models, can be used to study change among multiple sets of spatial data. 2 A more accurate projection of boundary requires controlling for determinants such as population, and economic performance of the city. This requires regression models, which are explained later in this report. 31 Figure 24. Population of census blocks in Cambridge, 1990. Figure 26. Population growth between 1990 and 2000 in Cambridge, Blocks represented in white experienced the lowest change (growth or decline) in the population among all census blocks of Cambridge. Figure 25. Population of census blocks in Cambridge, 2000. 32 3.2.ACCESSIBILITY ANALYSIS Understanding how accessible the resources of a city are to people is key to planning new infrastructures. Accessibility analyses investigate how locations of one group of phenomena (origins) are related to another group of phenomena (destinations). These spatial relations can be described in various ways. Accessibility can be assessed along transportation networks, or along idealized continuous space that simplifies constraints to movement (Figure 27). It can also be described geometrically (based on distance) or topologically (e.g. based on number of turns on a network, or number of steps in a topological grid). In this report we focus on examples of accessibility analyses that are conducted over urban transportation networks. Accessibility measures such as Gravity and Reach are computed at a fine resolution for individual buildings or address points over a network of city streets. Accessibility analysis helps us identify underserved areas, for instance, areas with low pedestrian accessibility to schools (Figure 31). Comparing accessibility values of individual origins across the city, or comparing their values to benchmarks from other cities, allows us to detect areas with problematic accessibility values. Accessibility analyses also provide inputs to other analytics proposed in this report. They are key determinants of land and real property values, land use patterns, and business clustering. Accessibly values can be used in hedonic pricing models for land and real properties. Potential impacts of new network infrastructure, such as bridges, roads, bus routes, or sidewalks, can be analyzed using before and after accessibility values. 3.2.1.Accessibility Measures Among accessibility metrics, Reach and Gravity metrics are simple to specify and can be interpreted most intuitively. Reach and Gravity metrics can be Figure 27. Accessibility types. Euclidean distance, computed over real transportation networks, and be implemented for various network distance, number of turns (topological), number spatial units, including buildings, address points, parcels or KT zones. of steps in a grid space (topological) 33 Computing accessibility values over network and at building or address point level can capture a detailed image of proximity to a city’s resources at an individual household level. Reach The Reach metric counts the number of resources that can be reached from an origin within a given search radius over a network of paths (Figures 28, 31, 32). For individual buildings, for instance, it can tell us how many jobs, schools or wells are available in a 10 minutes walking radius around it. The metric doesn’t capture the variation in distance to different reachable resources; it simply counts all destinations within the given radius. If we compute accessibility to retail spaces, an establishment located 600 meter away from the origin is treated the same as one that is only 50 meter away from the origin. This drawback is addressed by Gravity metric. The advantage of the Reach metric is that is intuitive to understand and communicate to multiple stakeholders – everyone can understand what it means to have two schools within 10 minutes walking radius as opposed to none. Figure 28. The Reach metric is a network analysis r measure that captures the number of destinations that The reach centrality, R [i] , of a building i, in a street network G describes the can be reached around a place within a given travel number of other buildings in G that are reachable from i at a shortest path radius. Reach can be specified to summarize accessibility to any kind of destination – people of a certain type, distance of at most r. It is defined as follows: buildings, firms, transit stations etc.– and the travel radius can be specified for different travel modes, such as r R [i]= ! !∈! !{! }:!! [! ,! ]!!! !W[j] walking, driving, biking or public transit. where d[i,j] is the shortest path distance between nodes i and j in G and W[j] is the weight of destination node j. Figure 32 illustrates the implementation of Reach to jobs in Cambridge and Somerville, MA within a 600m walking radius along the available street network. 34 Gravity Similar to Reach, the Gravity metric counts the number of resources that can Less Gravity be reached from an origin within a search radius over a network, but additionally accounts for their distance from the origin (Figure 30). A building with a few shops located next door will get a higher accessibility value to commercial establishments than a building with a large number of retail establishments that are located far away. The attraction of destinations does not drop linearly when their distance from the origin increases, but at an exponential rate, and it varies for different modes of transport. The inverse exponent of distance is often used instead of simple inverse distance for weighing destinations, and the distance decay rate is controlled by a coefficient for each mode of transport. Gravity of point i, in graph G, can be specified as: r ! [! ] Gravity[i] = ! !∈! !{! }:!! [! ,! ]!!! ! !.![!,!] ! More Gravity r where Gravity[i] is the gravity index at point i in network G within search radius r, W[j] is the weight of destination j, d[i,j] is the shortest distance between i and j, and b is the exponent for adjusting the effect of distance decay. These accessibility measures can be specified in the Urban Network Analysis Toolbox in ArcGIS. A more exclusive help document is available with the toolbox to explain the specifications in detail. In order to run the toolbox, ArcGIS 10 and the network Analyst extension are required. Figure 30. Illustration of the Gravity metric. 35 Figure 31. Underserved areas; the areas in orange don’t have access to public schools within a 1200 meter network radius. Overlaying the underserved areas and census data shows that approximately 10,000 people don’t have walking access to public schools 36 Figure 32. Reach to jobs located within 600 meter network radius (approximately 10 minutes walk) in Cambridge, MA. 37 3.3.SPATIAL-STATISTICAL MODELS Unlike accessibility and growth analytics, spatial-statistical models can examine relationships between more than two spatial values. Growth and change analytics each capture over-time change in only one property of a spatial unit. Spatial-statistical models, however, can examine the relationship of one spatial value to a number of other variables, and are thereby better suited for projecting changes under more realistic multivariable scenarios. Statistical models can be developed to examine the relationship of land prices to various determinants including accessibility (e.g. to bus stops, retail establishments etc.), frontage, area, or parcel type. Having such explanatory models, one can then predict how the value of each individual parcel is likely to change when, for example, a new bus stop or road is constructed, controlling for covariates. 3.3.1.Regression Analysis In statistics, regression analysis examines whether and how a dependent variable is related to one or more independent variables. The results of a regression function generate coefficients for the effects that each of the independent variables has on the dependent variable and an indication of whether and how significant these effects are. The models also tell us how much of the total variation in the dependent variable is explained by variations in the given independent variables. Those coefficients that are found to be significant can be used to predict future changes under similar conditions. Hedonic pricing models, which form one type of regression models, are widely used for projecting land or real estate value. In these models, the selling price of a real property (e.g. a housing unit) is predicted based on a linear function of the characteristics of the unit – age, size, number of rooms, structural quality, accessibility, ownership structure, lot size etc. 38 An accurate specification of regression models requires the use of specialized software like SAS, Stata or SPSS and necessitates a clear understanding of concepts and assumptions that regression analysis is grounded on. It is recommended that these models be specified by only staff who have had proper training in regression analysis and understand their foundations thoroughly. Simple multiple regressions and bivariate scatter plots can also be specified in MS Excel, using the analysis toolbox. 3.3.2.Trend Estimation and Autoregressive Models Trend estimation is a form of regression analysis, where time is the only linear predictor of the dependable variable. Trend estimation examines a correlation between the outcome values and time at which they took place. Trend estimation is suitable for projecting the long-term trend in variables whose key determinants are not fully known but a pattern in their values can be identified over time. We may not know, for instance, what variables can predict the increase of travelers to the city center, but a trend with a significant yearly time coefficient may be used to observed project the number based on past observations. Even if the data oscillated up and down, a trend regression can help us determine whether a significant long-term increase or decrease is present (Figure 33). The value of variables sometimes follows a cyclical pattern over time, where variables at one observation period are dependent on values during the previous period. Energy consumption in a census block, for instance, may Figure 33. Trend estimation of growth in total residential follow a cyclical pattern, following the winter-summer cycle in the floor area in China environment. Cyclical patterns in data may be independent of the overall long- term trend. While there may be a winter-summer cycle in the energy consumption at the household level, the long-term trend may be insignificant (the total annual energy consumption not changing), even when the energy consumption at a particular period (spring) may exhibit a cyclical decrease. Autoregressive models are used to predict over-time changes in variables with cyclical patterns, where independent variables include the value of the 39 Median Price of Houses Sold in the US dependent variable in the previous measurement period, as well as a linear time predictor that may or may not be significant for the long-term trend (Figure 34). More than one time lag variable can be used to capture longer Million M2 cyclical effect and the linear time variable can be squared to capture nonlinear effects. Both linear trend analysis and autoregressive analysis can be applied to a number of important planning problems in cities. Trend analysis can capture the long term pattern in key urban growth indicators – annual rural to urban land conversion, increase in residents or jobs, city GDP change, growth in transit ridership, etc. Cyclical trend analysis can capture predicted land and real estate values, seasonal changes in resource consumption or cyclical patterns in construction permit applications. 3.3.3.Spatial Regression Analysis Figure 34. The cyclical patter in the median price of houses sold There are various techniques for carrying out regression analysis, but common in the US can be explained by an autoregressive model where assumptions underlie most ordinary least squares (OLS) regression the predictors of the median price of houses are the pervious techniques. One of the underlying assumptions is that the dependent variable observed median prices. Source: Economagic, reproduced in City Form Lab. on the left-hand side of the regression equation can interact with independent variables on the right-hand side, but separate observations of the dependent variable are independent of each other. The price of land may depend on several independent factors, such as lot size, location and building height, but it should not depend on the price of land of the neighboring parcel. In reality this assumption may not hold; land values can depend on neighboring land values around them. This independent distribution of the dependent variable assumption of OLS regressions is relaxed in spatial lag and error type models. Spatial autocorrelation models allow either the dependent variable to depend on adjacent dependent variables or the error terms of adjacent observations to be correlated. The former case can be modeled by the “spatial lag models,” and the latter by the “spatial error models”(Anselin, 1988). 40 Spatial regressions can be specified in GeoDa or GeoDa Space software packages that are freely distributed. Many phenomena in spatial analyses exhibit spatial autocorrelation which such models capture. If spatial autocorrelation is present then OLS autoregressive models yield a better explanation to the variations in observed data (Figures 35, 36 & 37). Figures 35 and 36 provide an example of assed land value distribution in Cambridge, MA. A spatial lag regression is specified in GeoDa to predict how the per-square-foot value of land depends on four independent variables: plot ration, parcel size, access to streets and distance from the nearest subway station. In Figure 35 (top) an OLS model is specified with all four variables, but without allowing for autocorrelation between neighboring parcels. Below, a spatial lag model is specified, which adds spatial autocorrelation in the dependent variable (price per square foot) to the estimation (“W_LV_PSF” in the model). The high z-values suggest that land values in Cambridge are indeed strongly correlated with neighbors – beautiful improvements in the neighbors’ yard can significantly increase surrounding land values. Figure 36 plots the bivariate effect of proximity to subway stations, showing how land values decrease as the distance to the nearest subway station increases, controlling for other variables. Figure 35. Hedonic pricing model for Land value in Cambridge:. The dependent variable is the assessed price of land is US$ per square foot, and predictors are plot ratio, land area, parcel type (nr. Of streets directly accessed to from parcel), and distance to subway station. Unlike the spatial lag model (bottom), the ordinary least square regression model (top) does not account for the spatial autocorrelation among the land values of neighboring parcels. 41 70 60 50 40 30 Land value (US$/Sq.Ft) 20 10 0 0 500 1000 1500 2000 2500 3000 3500 4000 Distance to Subway Station (m) Figure 36. The relationship between the land value (US$/Sq.Ft) and distance to subway station predicted by the spatial lag model. All other predictors of land value are kept constant. Figure 37. Containing land value (US$/Sq.Ft.), land area and aggregated building data (such as total floor area), the parcel dataset of Cambridge, MA, provides the basis for the hedonic pricing model for land. Each parcels’ distance to subway stations, is computed using the network analyst of ArcGIS, which requires street network and transit station locations datasets. 42 3.4.IMPACT ANALYSIS Effective decisions in policy and spatial planning need to be evaluated before their implementation. Such evaluations require an understanding of the potential impacts of the decisions. Comparing the probable impacts of a series of alternative decisions to the existing conditions allows planners to establish a concrete base for informed decision-making. Impact analysis includes two broad groups of analytics. The first group includes analytics and statistical models that can predict the impact of a proposed policy (e.g. zoning) or spatial intervention (e.g. sidewalk improvement) on an outcome variable, such as land value, accessibility, crime rate or employment. The second type of analysis keeps track of changes in question variables (e.g. land value, or pedestrian movement) before, during and after an intervention. The latter provides useful empirical evidence upon which future planning decisions can be made. Accessibility analyses and regressions can be used as inputs to impact analysis, since they can capture changes in a variable when spatial conditions change. To analyze the impact of a new bridge or bus route, for instance, accessibility analyses can be used to measure the change in accessibility values – how much the citizens’ accessibility to jobs changes when a new bridge is built over the river. Utilizing a hedonic pricing model allows us to assess the impact of the same bridge on land values across the city. The output of accessibility analysis, which computes the changes in accessibility values, can be used in the hedonic pricing model for land, in which accessibility values form key independent variables. If other variables are kept constant, the reflected land value change reveals the pure impact of the new bridge on land values. Figure 38 illustrates the impact of how a hypothetical highway that cuts through the Geylang neighborhood in Singapore, on the accessibility of buildings to businesses. Comparing the gravity index from buildings to businesses before and after the proposed highway in a local 1km walking 43 range shows a 56% decline, on average, in accessibility to business establishments. The impact of change in accessibility on other key variables such as land value can be then analyzed by a spatial statistical model (Figure 39). Figures 39 and 40 below model the potential impact of a new subway station on land values in Cambridge, MA, using the hedonic pricing model for land in Cambridge, based on the present land values (See Figure 35). The example in Figure 39 demonstrates the estimated difference in land values before and after the proposed subway station. Using coefficient estimated in the spatial lag model of Figure 35, the total hypothetical change in land values that could result from adding the new subway stop is around $20 million. The distribution of the new per square foot prices is shown in Figure 40. 44 Gravity Percentage index change Figure 38. Comparison of the accessibility values before (left) and after a highway cut through Geylang, Singapore shows a significant drop in the local gravity to business establishments (center). Such comparisons in accessibility values can be used as input to regression models for predicting the other impacts of a spatial intervention (see Figure 39 and 40). The percentage change in accessibility to businesses as a result of the proposed highway is shown on the right. 45 Figure 39. Analyzing the impact of a proposed subway station in north Cambridge on the value (US$/Sq.Ft) of lands within 10-minute walk around the proposed station, using the hedonic pricing model for land developed in the previous section (see Figure 35). The figure shows a comparison between the present values (left) and the predicted values (right). ` 46 Figure 40. As a result of the new subway, land values increase 8% in dollars per square foot on average, which is approximately $20,000,000 in total for all parcels located within 600 meters from the proposed subway station. 47 PLANNING DECISIONS SUPPORT 48 4.PLANNING DECISION SUPPORT The ultimate goal of CPL data collection and spatial growth analyses is to support cities in their planning decisions with concrete evidence. Informing planning decisions by measurable evidence does not always require complex analytics; planning evidence can sometimes be directly extracted form raw spatial data. In the previous sections we discussed geospatial data, a number of analytical activities and their potential applications for urban planning. The way these analytical techniques can inform planning decisions can be summarized as follows: 1) By describing qualities of space in measureable terms, analysis of spatial data makes it possible to compare existing condition to certain benchmarks, and to thereby inform planners of present challenges. Spatial data and analytics help cities identify problems and frame questions they should focus on. Benchmarks can be chosen to meet a city’s goals and ideals based on national or international examples, or based on more complex underlying causes of the observed patterns. For example, by looking at rental payments as a share of household income in census tracts and comparing that to desired ratios, one can directly assess whether some household income groups are paying too large a portion of their monthly income on housing. In other cases, the assessment may require several analytical steps. Accessibility analysis, for instance, can inform planners whether access to key infrastructures or facilities is underserved in certain area, and if so, informs planners where such areas are located. 2) In addition to identifying existing challenges, spatial data and analytics can be used to identify forthcoming challenges. Trend estimation analyses can describe probable forthcoming issues by comparing the predicted value of a variable to its desired 49 benchmark value. Keeping track of trends in the number of multi- family building permits that are annually issued, and trends in demographic groups that form typical occupants for such units, planners can predict whether the city is headed toward shortages or oversupply in the market for multi-family housing. 3) Geospatial data and analytics can inform planners of underlying interactions and correlations between spatial variables. Representing spatial qualities with numeric data allows us to utilize statistical regression analysis to explain relationship between such variables. Spatial-statistical models can be used to identify the determinants of observed socio-economic variables. Statistical models can outline spatial conditions that require change in order to improve socio-economic indicators. If a model shows a strong negative correlation between the existence of commercial establishments that face directly to streets and crime rates on these streets, planner may use this evidence for deciding where to allocate commercial space in zoning plans. A statistical model that analyzes previous sidewalk improvement outcomes may reveal a positive correlation between sidewalk quality and business revenue along sidewalks. One interpretation of this correlation is that sidewalk improvement can be an effective tool for increasing business revenue in areas where proper sidewalks do not exist but other preconditions for commerce are in place. The correlation coefficient of the model can be used to estimate how much business owners could benefit from such public investment, and whether they could be involved in financing sidewalks through taxation. 4) Impact analyses allow planners to assess different future investment or improvement scenarios based on a key outcome variable. A comparison of different alternatives can allow one to identify the most impactful scenario. 50 The spatial analysis techniques that form the focus of the CPLs’ core module include a) spatio-temporal change mapping, b) accessibility assessment, c) trend analysis, d) spatial regression analysis and e) impact analysis. Rather than elaborating on any one application of these techniques at greater length, we have tried to provide a brief overview of the nature and utility of each technique, pointing towards applications for various urban planning and management tasks. The exact application focus of the techniques in the four participating CPL cities – Surabaya, Denpasar, Palembang and Balikpapan – should be identified together with the local government representatives and CPL staff. The analysis should be chosen to address the most important spatial analysis and planning questions specific to each city. 5.Refrences Anselin, L. (1998). Exploratory spatial data analysis in a geocomputational environment. In P. Longley, S. Brooks, B. Macmillan and R. McDonnell (Eds.), GeoComputation, a Primer, 77-94. New York: Wiley. Farvacque-Vitkovic, C., Godin, L., Leroux, H., Verdet, F., & Chavez, R. (2005). Street Addressing and the Management of Cities. The World Bank, Washington, D.C. Schneider, A, Friedl, M. A., & Potere, D. (2009). A new map of global urban extent from MODIS satellite data. Environmental Research Letters, 4(4), 44003. Schneider, A., Friedl, M. A., & Potere, D. (2010). Mapping global urban areas using MODIS 500-m data: New methods and datasets based on “urban ecoregions”. Remote Sensing of Environment, 114(8), 1733–1746. 51